Scandinavian Personal Names in Norfolk: A Survey Based on Medieval Records and Place-Names
 9185352268, 9789185352265 [PDF]

  • 0 0 0
  • Gefällt Ihnen dieses papier und der download? Sie können Ihre eigene PDF-Datei in wenigen Minuten kostenlos online veröffentlichen! Anmelden
Datei wird geladen, bitte warten...
Zitiervorschau

ACTA ACADEMIAE REGIAE GUSTAVI ADOLPHI

62

ACTA ACADEMIAE REGIAE GUSTAV! ADOLPHI LXII

Scandinavian Personal Names in Norfolk A Survey Based on Medieval Records and Place-N ames BY

JOHNINSLEY

UPPSALA 1994 Distributor ALMQVIST & WIKSELL INTERNATIONAL STOCKHOLM, SWEDEN

The publication of this book has been made possible by a generous financial grant from the Swedish Council for Research in the Humanities and Social Sciences.

Abstract lnsley, John, 1994: Scandinavian Personal Names in No,folk. A Survey Based on Medieval Records and Place-Names. Acta Academiae Regiae Gustavi Adolphi 62. xliii + 455 pp. Uppsala. ISBN 91-85352-26-8. The present work is a survey of the Scandinavian personal names found in medieval records and as the first elements of place-names in Norfolk. The period covered extends from the tenth to the mid-thirteenth century. A wide range of published and unpublished sources has been excerpted. A brief introduction is followed by a comprehensive bibliography and by a corpus of name-forms. Each entry in the corpus is accompanied by a full commentary with bibliographical information and comparative material from elsewhere in England. The material is an important primary source for any discussion of the Scandinavian settlement of East Anglia, and it allows comparisons to be made with the other areas of Scandinavian settlement in England.

ISSN 0065-0897 ISBN 91-85352-26-8 © John Insley and the Royal Gustavus Adolphus Academy 1994

Printed in Sweden Textgruppen i Uppsala AB, 1994

Table of Contents

Acknowledgements ...............................................................................

vii

Bibliography and Abbreviations ... .. .............. ......... ....... ..... .... ... ....... .. ...

IX

Introduction ..................................................... ...................................... xxxiii Arrangement of the Material.................................................................

xlii

The Material..........................................................................................

1

Appendix...............................................................................................

447

Acknow ledgements

This book has been a long time in the making, and I have been the recipient of many kindnesses along the way. I hope that my placing of the completed text at the disposal of scholarship will go some way towards repaying the debts I have incurred over the years. It is based on my University of Nottingham Ph.D. thesis of 1980, and it is fitting that I should first thank my former supervisor, Professor Kenneth Cameron, for his support and encouragement in the years when this work was being written. He has proved a true friend, and he has shared his incomparable knowledge of the toponymy of the Danelaw and its source materials with me unstintingly. With characteristic generosity, he placed the personal name collections of the late Sir Frank Stenton, left to him by Lady Stenton, and the Norfolk collections ofthe late O.K. Schram, then deposited with the English Place-Name Society in Nottingham, at my disposal. I am also indebted to the late Dr Olof von Feilitzen of the Royal Library in Stockholm, from whose publications I leamt much, and who untiringly helped me with my queries about personal nomenclature when I was a young and inexperienced research student in the early seventies. For copies of their publications, I owe thanks to the late Cecily Clark, Dr Gillian Fellows-Jensen, Mr Peter McClure, Mr Richard McKinley, Dr Bo Selten, Dr Karl Inge Sandred and the late Professor Erik Tengstrand. The discussions I had with Cecily Clark and Peter McClure in the latter stages of the completion of the original thesis were especially fruitful, and I owe both of them a good deal. My particular thanks are due to Dr Sandred. Our first contact was at Nottingham -kel/, -kil bad become widespread in Scandinavia, though it should be noted that the syncopated forms, though less common as a whole than the full forms, 24 are quite well attested in East Anglia. The strength of the full form -ketill in East Anglia is more akin to the situation in Normandy than to that in the territory of the Five Boroughs and the North, where the syncopated -kel/, -kil predominates (cf. below, s. nn. Arnkell, Äskell, Porkell, Ulfkell). A further factor revealed by the material collected in the present corpus is that Scandinavian personal names often occur in areas without any large scale place-name evidence for Scandinavian settlement. This is especially true of the area around Lynn, where there were large numbers of persons of extremely modest means hearing Scandinavian personal names. However, the strength of this evidence may well be due to the fäet that this area has extensive charter material from the cartularies of West Dereham Abbey and Castle Acre Priory, as well as lists of tenants' names in the Ely surveys of 1222 and 1251, evidence which is lacking in less well documented areas. Nevertheless, the evidence itself is too strong to be accidental, and it raises important questions about the nature of settlement patterns, name-giving and social status. The comparative material shows that Norfolk shared certain features with Suffolk in its Scandinavian personal nomenclature, e.g., the strength of Anglicization, the survival of the conservative -ketill, etc., though the material for Norfolk is larger and more varied. However, there are differences. Thus ON Porm6ör, ODan Thormöth is more frequent in Suffolk, and there are many names which occur in one county, but not in the other. For example, ON Sand-, SQndulfr, ODan, OSwed (run.) Pormar and ON Prymr, ODan Thrym

the seignorial character of the 'Grirnston hybrids' is underlined by reference to Norman place-names of the type Toutainville, 'l>orsteinn's villa', in which a Scandinavian personal name is compounded with the Romance habitative element -villa. See also R.H.C. Davis, 'East Anglia and the Danelaw', TRHS 5th Series 5 (1955), pp. 23-39. 23 For a recent account of this question, see G. Fellows-Jensen, 'Of Danes-and Thanes-and Domesday Book', in: People and Places in Northern Europe 500-1600: Essays in Honour of Peter Hayes Sawyer, ed. I. Wood and N. Lund (Woodbridge 1991), pp. 107-121, at pp. 112-120. 24 An exception to this is provided by ON Porkell etc., the syncopated forms of which also predominate in East Anglia, though this might be explained as reflecting the fame of the Viking chieftain l>urkil the Tall, who was fora time Cnut's earl of East Anglia.

xxxviii

have been noted in Suffolk, but are absent in Norfolk, 25 while ODan Böndi, which is extremely frequent in Norfolk, is rare in Suffolk. In terms of Scandinavian dialect distribution, the Norfolk material in the present corpus is of a heterogeneous nature. Thus the characteristically Danish names Böndi and Töki are frequent, but so are such typically West Scandinavian names as Steingrimr and Sumarliöi. The bulk of the material in the present work is from records of the period 1086-1251. Material from the Old English period is sparse, this of course being a consequence of the fäet that little Old English charter material from East Anglia is available. It is true that there are same eleventh-century wills, such as those of lElfric Modercape (Sawyer, no. 1490; ASWills, no. 28) and Thurketel Heyng (Sawyer, no. 1528; ASWills, no. 25), which contain relevant Norfolk material, but even these are only extant in cartulary copies from the late thirteenth century. The number of Scandinavian personal names among the names of Norwich and Thetford moneyers of the period from the reign of Edgar to the Norman Conquest is insignificant. After 1250, the number of Scandinavian baptismal names found in Norfolk records declines rapidly. From the second quarter of the thirteenth century onwards, Old English and Scandinavian personal names occur as hereditary surnames in Norfolk. Such surnames are well represented in later records, but no systematic attempt has been made here to collect the Norfolk examples of surnames of this type deriving from Scandinavian personal names. 26 In discussing the phonology of the name forms, I have preferred to use

25 ON Sand-, S{mdulfr: SandolfWest c. 1230 (c. 1270) BuryS, fo. 117a, Sandulph' de Monte ibid., fo. 118a, Adamfilius Sandolf, Sandolf de ecclesia, Hugo filius Sandolf, Sandolf le Heiward, Ricardusfilius Sandolfibid., fo. 118b, Sando/fibid., fo. 119a (tenants at Palgrave); ODan, OSwed (run.) l>ormar: Thurmero, abl. (commended to the abbot of Bury St Edmunds, held at Vledana) TRE DB, fo. 374b (PNDB, p. 395); ON l>rymr, ODan Thrym: Threm (dat., grantee of thirty acres of assart at the wood calledLeinoenhal at Pakenham), predictus Threm 1156--1180 (e. 13th) Bury, no. 164. 26 Despite its weaknesses, Reaney's surname dictionary (DBS) is still the starting point for any investigation of the patronymic surnames with Old English and Scandinavian personal names as base forms. In the present volume, reference has been made to the second edition of this work, dating from 1976. This has been recently superseded by the third edition (P.H. Reaney, A Dictionary of English Surnames, 3rd edition with corrections and additions by R.M. Wilson [London and New York 1991]). In this new edition, the 'British' in the title ofthe first two editions has been replaced by 'English', and it contains 4,000 more names than the edition of 1976 with a good deal of material from the Middle English period. Reference to this edition (DBS 3) should be made for additional comparative material for the following entries in the present work: AScand *Anki (Appendix) (DBS3, s.n. Anke; the form Johannes filius Anke 1251 Ely Reg, fo. 274b [Hartest Sf] is wrongly dated to 1277); ON Auöunn (DBS3, s.n. Oden); ON Qzurr (DBS 3 , s.n. Asser; the entry is too uncritical, since forms standing for Welsh Asser and possibly for OE JEschere are also grouped here); ON B6fi (DBS3, s.n. Bovey); ON Bolli (DBS3, s.n. Boll; OE Bolla is given as an alternative here); OScand *Dottr (DBS3, s.n. Datt; OE *Datt is given as an alternative); ODan

xxxix

traditional methods of description. I have also largely followed traditional views with regard to the vexed question of Anglo-Norman influence in postConquest records. However, it should be noted that the whole concept of the 'Anglo-Norman scribe' was called into question by the late Cecily Clark. 27 She quite rightly pointed out that our post-Conquest personal names occur in Latin documents, and underlined the anachronism of ascribing French origins to fourteenth-century English scribes. 28 It is obvious that after the Conquest the scribal conventions of post-Carolingian Neustria were adopted in Latin documents produced in England. Among these conventions were such Latinized forms as Radulfus for OFr Ra(o)ul or Willelmus for OFr (Northem) Williaume. One might also think that initial T(h)- for [8] in names in ON Por-, ODan Pur- belongs to this category of post-Conquest convention. On the other hand, the French influence on post-Conquest Latin documents established by Zachrisson and von Feilitzen cannot be easily contradicted. It is true, as the recent calendar of documents drawn up by Pelteret29 has shown, that the native scribal tradition in the production of vemacular records lasted well into the twelfth century, and it is also the case that French was largely a school language in England in the Middle English period. 30 Nevertheless however, we can reckon with French speakers in official positions in England until the loss

Feggi (DBS3, s.n. Fegg); ON Gaddr (DBS3, s.n. Gadd, where the sumame is derived from the substantive ON gaddr m. 'a goad'); ON Inga fem./lngi masc. (DBS3, s.n. Ing); ON Kol(l)sveinn (DBS 3, s.n. Colswain); AScand *Liulf (DBS3, s.n. Lyolf, where the etymon is wrongly given as 'OG Leutwulf, OFr Lyulf); ON Lurkr (DBS 3 , s.n. Lurk); ON Ragnhildr fem. (DBS 3, s.n. Ragnell, where the etymon is wrongly given as 'OG Raginhild (f)'); ON Saxi (DBS3, s.n. Sax); ODan Skuli (DBS3, s. n School); ON Spakr (DBS 3 , s.n. Spack); AScand I>ör (DBS3, s.n. Thor); ON l>ora fem. (DBS3, s.n. Thor); ON l>orir (DBS3, s.n. Thury [for examples of OEScand Puri]); OScand *l>rykki (DBS3, s.n. Trick, where the etymon is given as OFr trique 'trick, deception'. This may be true of the surnames contained in the forms William Tryk' 1332 and Roger Trix 1642, but it can be rejected in the case of the Y orkshire form of 1277, Trikke del Nortgate, which clearly represents the Scandinavian *Prykki). 27 C. Clark, 'Towards a Reassessment of "Anglo-Norman Influence on English Place-Names" ', in: Language Contact in the British Istes: Proceedings of the Eighth International Symposium on Language Contact in Europe, Douglas, Isle of Man, 1988, ed. P. Sture Ureland and George Broderick, Linguistische Arbeiten 238 (Tiibingen 1991), pp. 275-293; Clark, 'Onomastics', in: The Cambridge History of the English Language, Il: 1066-1476, ed. Blake, pp. 548-550; C. Clark, 'The myth of "the Anglo-Norman scribe" ', in: History of Englishes: New Methods and Interpretations in Historical Linguistics, ed. M. Rissanen, 0. Ihalainen, T. Nevalainen and I. Taavitsainen, Topics in English Linguistics 10 (Berlin and New York 1992), pp. 117-129; C. Clark, 'Domesday Book-a Great Red-herring: Thoughts on some Late-Eleventh-Century Orthographies', in: England in the Eleventh Century, ed. C. Hicks, Harlaxton Medieval Studies, Il (Stamford, 1992),pp. 317-331. 28 Clark, 'Onomastics' (as in n. 27), pp. 548-549. 29 D.A.E. Pelteret, Catalogue of English post-Conquest Vernacular Documents (Woodbridge 1990). 30 See Clark, 'Onomastics' (as in n. 27), p. 548, and the works cited there.

xl

of Normandy in 1204, and it should also not be forgotten that many of the Latin documents forming our sources, such as Pipe Rolls and Curia Regis Rolls, were products of the central administration. Perhaps same features which have been hitherto described as 'Anglo-Norman' are pehaps better described as belonging to the system of Latin orthography introduced after the Conquest. This alone would make a subject for a systematic investigation far beyond the scope of the present work. My aim has been merely to present a corpus of Scandinavian personal names found in medieval Norfolk with an accompanying commentary and comparative material as a contribution to the elucidation of the nature of the Scandinavian settlement of East Anglia. We are concemed with English and Continental scribal traditions and with the phonology of early Middle English, but lurking behind this is the Scandinavian settlement of the late ninth and early tenth century, for which the material isa source of major importance.

xli

Arrangement of the Material

Place-names and field names containing the personal name in question are grouped sub a). Personal name forms recorded in independent use are grouped sub b). Where necessary, the forms have been further subdivided within these groups. An attempt has been made to localize the persons listed under b). This applies to the participants in a document both when the participant himself has a Scandinavian name and when he or she has an ancestor or a relative with a Scandinavian name. The name of the place in which these participants held land or with which they were involved in pleas or fines etc. is given in brackets after the source. In the case of witnesses, the place indicated is that with which the document in question is concemed. The Latin case is indicated where necessary. For uninflected forms, the case is bracketed. Where forms appear without any indication of case occurring in the form itself or without any indication in the text, these forms are in the nominative. The case is given both when the Scandinavian name is that of the actual participant in a record and when it is that of the father, brother, wife, husband, etc. of the participant. The case indicated always relates to the name of the participant, and not to that of the relative. For example, in a form of the type Ricardo filio Turchil, abl. (witn.), the case and the designation (witn.) refer of course to Ricardus, and not to his father Turchil, whose name is here an uninflected genitive. Forms taken from printed texts have been rendered as they occur in these texts, except in the case ofvolumes published by the Record Commissioners (RC) and the earlier Pipe Roll Society volumes (PRS 1-24), whose abbreviated record type has been silently extended. Manuscript abbreviations have been silently extended where they are clear, but where they are ambiguous, the extension is bracketed. Letters which have been supplied to MS. forms or coin legends are inserted between pointed brackets, as inA(n)ger for MS. [Godwinus filius Brexi] Ager, (below, s. n. Åki, p. 3). Capitals have been standardized throughout for the initial letters of proper names, even when these occur in the text in lower case. Manuscript material has been cited according to folio and side (a = recto, b = dorso ). Where more than one charter occurs on a side of a folio, the charters are numbered in sequence from the top of the folio. For example, fo. la(i) signifies the first complete charter from the top on folio la. Documents which are continued from a previous folio or side and documents which cover a complete side are cited according to folio and side. For example, fo. 1b denotes either a document which is continuous throughout folio lb or one which isa continuation of a record begun on folio la. This method has been followed in all the Norfolk cartulary material excerpted in the present work, except in the case of Langley, where much of the material has been merely cited according to folio and side. Surveys like ElyA, which are continuous records, and the unpublished material collected by Sir Frank Stenton and now in the possession of Professor Kenneth Cameron have been cited according to folio and side. In the case of frequently cited cartularies and records like ElyA, the manuscript date of the main hand of the text is given in the bibliography. The manuscript dates of en-

xlii

tries in hands of a different date to the hand of the main body of a record of this kind are supplied with the individual name entries. The manuscript date is also supplied with the individual name entries in the case of composite works like Rams or the Bury St Edmunds material collected in Bury and BuryS. For original post-Conquest records, merely the date is given, but for pre-Conquest records, whether contemporary or in later copies, the date of the record itself and the manuscript date are given. Entries from the Anglo-Saxon Chronicle are cited with the date given by the text itself and not with the corrected date, which is sometimes as much as three years at variance with that given by the text. Phonetic transcripts and phonemic transcripts are rendered in the usual fashion (enclosure between [] for phonetic transcription, enclosure between // for phonemic transcription).

xliii

A

ODan Abbi, OSwed Abbe a) (i) Abbecroft c. 1235 (m. 14th) Rams, i, p. 413 (f. n. in Brancaster).

ODan Abbi, OSwed Abbe is probably a hypocoristic form of ODan, OSwed Äbiorn 1 (cf. DgP, col. l; SMP, col. 1). SMP, col. 1, also includes the possibility of Abbe being a hypocoristic form of Abraham. This may be the case in Swedish examples from the late medieval period, but not in the above example from the Ramsey Cartulary, for, if we are concemed with ODan Abbi, OSwed Abbe here, the name would have been taken into English long before the large scale medieval borrowing of biblical names in Scandinavia, and would, therefore, most likely be a hypocoristic form of ODan, OSwed Äbiorn. Altematively, however, Abbecroft may contain OE Abba. For Land Y examples see SPNLY, p. 1.

ODan Aggi, OSwed Agge a) (i) Aggescroft Edw3 Rental (f. n. in South Creake) (OKS). b) (i) De Rogero filio Hagge 1165-1166 P, p. 34 (Nf or St).

ODan Aggi, OSwed Agge is a hypocoristic form of names in Agh- (SMP, col. 19). 2 Aggi is quite common in ODan (DgP, cols. 10--12; for the examples from Skåne see also SMP, cols. 18-19). The name occurs as the first element of three Danish place-names in -thorp (DgP, cols. 11-12; DS XIV, p. 128; DS XVI, p. 49). Aggi with the secondary genitive in -s may occur in a further two Danish -thorp names, though in these two cases the related name *Aggir (gen. -is) is equally possible (DS XII, pp. 62, 200). Aggi is also found as the first element of one Danish place-name in -böth (DS Il, p. 55). Only a few instances of OSwed Agge are recorded (SMP, col. 19). 1 Modeer, SvP, p. 40, in a list of hypocoristic names with consonantal gemination, has: "Abbe: Abiorn, Arnbiorn m. fl. möjligheter". 2 SMP, col. 19, comments: "Fda. Aggi, hypokorism sannolikt till namn med f. l. Agh- (ev. fda. Aghwith). För de östg. beläggen bör närmast Aghmund komma i åtanke." In OSwed the name is limited to Östergötland. Modeer, SvP, p. 40, takes OSwed Agge to be a hypocoristic form of Aghmund, Algot etc.

1

Åki, Åki, Åke SPNLY, pp. 1-2, collects examples of Aggi from L and Y under the heading Aggi, Agi and comments (p. 2): "It is not possible to determine whether the pers. ns. in L and Y derive from Aggi or Agi so all forms have been assembled under one heading." Agi, that is ODan, OSwed Aghi, 3 occurs in the place-name Authorpe (two examples) L and in the form Henricum f Aghi 1202 L, SPNLY, p. 1, but the other forms given by SPNLY, pp. 1-2, under the heading Aggi, Agi, e.g. Agge Bighe 1189-1203 (13th) L, Aggo, abl., 11451152 (1. 12th) Y, clearly represent ODan Aggi, OSwed Agge. 4 For L examples see also von Feilitzen, Notes, p. 53, where derivation from ODan Aggi, OSwed Agge is given. In the Fens forms representing ODan Aggi, OSwed Agge are still found in the early 13th century, cf. Agge filius /ohannis, Salem(und) filius Agge,5 Agge Dromund' ,6 Johannes filius Agg' 1222 ElyA, fo. 152a, Agge filius Iohannis ibid., fo. 152b, Salemu(n)d' filius Agge, 5 Agge Dromund6 ibid., fo. 153a (Tydd St Giles C). A late example is Reginaldo filio Agge, abl., c. 1250-1260 Swafjham' s Register, fo. 231 (Spalding L) (FMS). The initial H- in the form Hagge is inorganic and the result of AN influence (see PNDB, § 138; SPNLY, § 140.i).

ON Åki, ODan Åki, OSwed Åke a) (i) Akieslond c. 1220 CAcre, fo. 22b(iii) (f. n. in Little Dunham). (ii) Okesdik c. 1265 Wals, fo. 31b(ii) (f. n. in Great Walsingham). b) (i)Aky (monk of St Benet ofHolme) (gen.) (witn.) 1101-1116 Holme, no. 119 (Little Melton),Aky prior,Aky priore, abl. (witn.) 1127-1134 (1. 13th) Holme, no. 127 (Tibenham),Aky prior (witn.) 1127-1134 (14th) Holme, no. 130 (Rollesby), Aky tunc temporis prior (witn.) 1128-1129 Holme, no. 126 (grant ofthe office of seneschal ofthe Abbey of St Benet of Holme and the lands pertaining thereto), Aky tunc prior (witn.) 11341140 Holme, no. 131 (Tibenham), terram scilicet illam quam Wido nepos Aky prioris tempore regis Henrici primi tenuit in eadem uilla 1168-1173 Bod!NfCh, no. 602 Aghi would seem to be derived from *AgreiRR (see NK VII, p. 208). The forms Aggi prepositus and Aki preposito 1202 L cited by SPNLY, p. 1, refer to the same person and here Aki is merely a scribal error for Aggi (cf. von Feilitzen, Notes, p. 53, n. 3). Similarly Agk, which appears in documents cited by SPNLY, p. 2 (RA, vii, nos. 2037, 2040) in the form Roberti filii Agk, gen., c. 1245-1255 L, occurs in another document (RA, vii, no. 2034) as Agge in the form Roberto f Agge c. 1219 L, SPNLY, p. 2. Apart from these cases of scribal error and the instances of Aghi mentioned above, all the forms cited by SPNLY, pp. 1-2, under the heading Aggi, Agi have spellings in -gg- and, clearly, represent ODan Aggi, OSwed Agge. 5 MS. Salem fil Agge, Salemud' fil' Agge. Most likely the nasalizing stroke has been missed over the u of Salemud' and, therefore, here we are concemed with ON SQlmundr, ODan, OSwed Salmund. 6 The byname Dromund represents ON Dr6mundr which occurs in Dromonby NRY where it seems to have replaced an earlier Dragmall (see PN NRY, p. 168). The byname Dr6mundr derives from ON dr6mundr '!arge warship'. 3

4

2

Åki, Åki, Åke (printed, SSMEA, Appendix I, no. 25) 7 (Yarmouth). -Achi prepositus et Turchillus f(rater) eius (witn.) c. 1130 CAcre, fo. 103b(i) (grant by Alanus de Swofuam (Swaffuam Nf) to the monks of Acre). -Godefrido filio Achi, abl. (witn.) c. 1135-1140 CAcre, fo. 16b (Massingham). - De terra Achi, unum uicum in Gememua que appellatur terra Achi c. 1150-1155 CAcre, fo. 65b(i) (Yarmouth). -terram que fuitAky de Waltone 1157 (e. 14th) NorwCPCh, no. 35 (East Walton). - pro Achi 1157-1158 PR, p. 130 (Nfor Sf, prob. Nf). -Petrusfilius Achi 1157-1163 Lewes, fo. 282a (Etuna and Walpole), Petrus filius Ach' (witn.) c. 1160 Lewes, fo. 276b(ii) (West Walton), Petra filio Achi, abl. (witn.) ibid., fo. 277a (West Walton and Walsoken), c. 1170 Lewes, fo. 285b(ii) (Nf Marshland), c. 1175 Lewes, fo. 234a(i) (Heacham). -de catallis Aki Coc 1176--1177 P, p. 137 (Nfor St). -Aki prepositus 1180-1181 P, p. 89, 1181-1182 P, p. 70 (Nfor St). - quas Alicia filia Aki tenuit in Est Tudeham 1197 FF (PRS, xx, no. 108) (East Tuddenham). - Thurketil' filius Aky 1199 CRR 17, 18 8 (CurR, ii, p. 22) (Wiggenhall). - Carta Brixi Aunger et Aky fratris eiusdem filii Elfrici Anger,9 Brixi Anger et Aki frater eius fil ii Elrici Anger de Linna c. 1200 Der, fo. 236a(iii), Brixi et Aki prenominati ibid., fo. 236b, de omnibus terris et seruiciis quas Brigsi Anger et Aki fratris sui nobis dederunt, Brixi Anger et Aki frater eius homines mei ibid., fo. 236b(i) (Carta Haconis de Herlham), de terris que Brixi et Aki nobis dederunt, Brixi et Aki fil ii Elurici Anger de Linne, predictorum donatorum Brexi et Haki, predicti Brexi et Aki ibid., fo. 237a(i), Aki, Brexi pater meus et Aki patruus meus filii Elfrici Anger de Lenn', predictorum donatorum Brexi et Aki c. 1210-1220 ibid., fo. 237a(ii) (Godwinus filius Brexi A(n)ger de Lenna issues this charter) (Clenchwarton). - lx acras terre in Hillingt' cum pertinenciis que fuerunt Aki aui mei c. 1200-1210Der, fo. ll0a(i) (CartaRadulfi filii Galfridi de Dunham) (Hillington). -Achius filius Ailirici 1201 (13th) Rams, iii, p. 294 (Walsoken). -Alanus f Ake 1209 P, p. 72 (Nf). -SimonemfiliumAchi 1209 AssRoll 558 (Pleas, iv, no. 4294) (Stiffkey). -Gaufridusfilius Achi 1209 AssRoll 558 (Pleas, iv, no. 4303) (claim in Flitcham). - Galfridus f Aki 1209 P, p. 34, 1210 P, p. 54, 1211 P, p. 22, 1212 P, p. 176 (Nf). terram Hugonis filii Aki, Hugone filio Aki, abl. (witn.) c. 1210-1215 Lewes, fo. 226b, Hugone filio Aki, abl. (witn.) ibid., fo. 233b(i) (Heacham), fo. 234b(ii) (Hillington), c. 1220 Lewes, fo. 236b(i) (Letton and Wood Rising). -Ake 1222 ElyA, fo. 125b (Northwold). -tofte Aky filii Iuonis 1237-1251 Holme, fo. 106a(iii) (Ludham). (ii) inter Willelmumfilium Oki. et Godiuam uxorem suam petentes. et Gocelinum Brais. et Willelmum fratrem suum tenentes, jus et hereditatem. prefati Oki et Godiue uxoris sue 1198 FF (PRS NS, xxvii, no. 12) (Wiggenhall). -Radulfo Oki, dat., 1199 CRR 51 (Cur, i, p. 93) (Burnham). -terramquefuitquondamPetrifilii Oki c. 1200-1220 CAcre, fo. 79b(i) (Emneth). - Ellich filius Oki, Elvredus filius Oki, Radulfus filius Oki 1222 ElyA, fo. 157b (Emneth). - Godefridusfilius Oki, Mainer filius Oki, Petrusfilius Oki, Laurenciusfilius Oki 1222 ElyA, fo. 171b (West Walton). Wlfward'.filius Oki 1222 ElyA, fo 209a (Thorp in Shipdham). -Alanus Oky 1234 Binham, fo. 80a (Thursford). Rogerus Oky 1239 Binham, fo. 81a (Thursford). (iii) Eluiua Ackes 1203 AssRoll 817 (NthRS, v, no. 832) (Haveringland). - Gaufridus f Acke 1203 AssRoll 817 (NthRS, v, no. 836) (Flitcham). -Anselmum filium Acke 1205 CRR 34 (Cur, iii, p. 270) (Appleton). -tota terra que fuitAcke 1209 FF (PRS NS, xxxii, Also Holme, no. 302. These two rolls are duplicates. 9 The byname Anger represents either ME anger 'grief' < ON angr or the personal name Anger which is an AN form of either ON Äsgeirr etc. or West Frankish Ansegarius, Ansger. For further discussion of Anger see below, pp. 39-40, 45--46. 7

8

3

Åki, Åki, Åke no. 210) (Burnham). -iuxtadomum VincenciiAcke 1214-1229 Holme, fo. 82b(v) (Potter Heigham). - Vincencio filio Acche de Fritton', dat., predictus Vincencius filius Acche 1214-1229 Holme, fo. 85b(ii) (Fritton). -Acke 1222 ElyA, fo. 126a (Northwold).

An EScand name, Äki corresponds to OHG Annicho (Brpndum-Nielsen, § 268.2), and is a diminutive formed with the -ki suffix from a base belonging to the IE root *an-, which Pokomy, IEW, p. 36, glosses as 'Bezeichnung för männlichen oder weiblichen Ahnen'. The Danish runic form qaki (DR, no. 288) shows that Ä- in Äki develops from original *An-. 10 ODan Äki is common and largely eastem Danish (DgP, cols. 18-20), and would seem to have originated in the Danish islands, where it occurs in no less than 12 place-names in -thorp (Hald, Vore Stednavne, p. 134; for examples see: Hjorth Pedersen, Bebygg. på -by, p. 17; DS IX, p. 243; DS XIII, p. 104). In addition, it occurs in two -thorp names in Halland (SOH I, pp. 108, 171). ODan Äki is also found compounded with three Danish names in -by (Hjorth Pedersen, Bebygg. på -by, p. 17). 11 In OSwed the name occurs in runic inscriptions and in medieval and 16th century records (SRÖg, no. 183; SRSö, no. 238; SRSmå, no. 16; SRVg, no. 217; SRU, nos. 570, 946, 1016, 1158; SMP, cols. 26-35; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 151, s.n. Åke). OSwed Äke has also been noted as the first element of Swedish place-names in -thorp (see Hjorth Pedersen, Bebygg. på -by, p. 17, for references). ON Aki was borrowed into Norway from EScand at the end of the 13th century (see Lind, cols. 8-9). A pre-Conquest English example of ON Aki, ODan Äki, OSwed Äke is Aki (son of Toki) 1052-1056 (1. 11th) KCD, no. 805 12 (Teddington and Alstone Gl). In DB the name occurs in Bd, Brk, C, Ess, Hrt, L, Mx, Nth, Sf, St, W, Wa and Wo (PNDB, p. 142), and it is also recorded in the Liber Vitae of Thomey Abbey C (SPNT, p. 142). Äki forms the first element of Oakthorpe Lei (PN Db, p. 650-651), and of the Sf Akethorp 1207 FF (PRS NS, xxxii, no. 471). For Land Y examples of the name see SPNLY, pp. 3-5, and Fellows Jensen, Lincolnshire Tenants, pp. 87, 92. A late example of the name from Nt is contained in the form Willelmus Genus Oki m. 13th (14th) RuffordCh, no. 19b 10 Hjorth Pedersen, Bebygg. på -by, p. 17, would derive OEScand Äki, OHG Anihho (= Annicho) from a root contained in ON ai and OHG ano. However, OHG ano m. 'grandfather, ancestor' (for which see AHDW, col. 532) contains the IE root *an- (Porkomy, IEW, p. 36), while ON ai m. 'great-grandfather' belongs to the IE root *a1Jo-s (Porkomy, IEW, p. 89). SinceÄki contains the root *an-, it is not to be linked with ON ai. For runic qaki and the original nasalization see DR, col. 628, Anm. 1, where it is remarked that initial qa- may be explained as reflecting a convergence of earlier and later 'scribal' forms (qki - aki). 11 For futher discussion of ODan Äki etc. see M. Kristensen, Gamle Navne: Spredte lagttagelser og Overvejelser (Copenhagen 1938), pp. 22-36.

12

4

For literature discussing this charter see Sawyer, no. 1408.

Åki, Åki, Åke

(Nottingham). Early 13th century examples of Äki from the Fenland are: Frepesent Oki 1222 ElyA, fo. 92a (Ely C), Robertus filius Aki, Elricus filius Aki ibid., fo. 150b (Leverington C), Oki ad pontem ibid., fo. 154a, Petrus filius Oki ibid., fo. 155b (Elm C). The name is not common in Sf, but in addition to the DB examples and the above-mentioned lost place-name Akethorp, the following examples have been noted: Achy 1087-1098 (c. 1175-1200) Bury, p. 44 (Litlecerce St); unam acram terre que fuit Alani filii Oki e. 13th (e. 13th) Leiston, no. 86 (Theberton); Robert Oki, Willelmus Oki c. 1230 (c. 1270) BuryS, fo. 117b (Palgrave). For further references to the name in England see DBS, p. 255, s.n. Okey, Oakey, NPN, p. 2, and ZEN, p. 12. 13 In Normandy the name occurs as the first element of one place-name (Adigard. pp. 72, 375). On the phonological aspects of the forms the following is to be noted: (i) /kl is represented by k and ch. OScand k and OE velar c are often represented in ME sources byk before e and i (and y with the value of i). 14 Though rare in OE and OFr orthography, k was not unknown (PNDB, § 118; SPNLY, § 125.i). In early AN orthography ch is the normal form for /kl before e and i (and y with the value of i), while c is normal before all consonants except w (the OE combination cw- is represented by qu-) and velar vowels. See PNDB, §§ 113-120, fora full discussion of the AN spellings of /kl. (ii) OScand a in stressed syllables is rendered by the corresponding OE a in Scand loan-words in English and takes its development in ME (see Brunner, Abrill, § 17; Luick, § 383.1; on the special case of the element As- see s.n. 13 Björkman, ZEN, p. 12, comments: "Auch Aketon, Acketon Testa de Nevill (Northumberl., Norf., Suff.) z. B. 300b, 386b, Acheby Yorksh. D. B. Akethorp Suff. R. Hundr., Achewelle Norf. R. Hundr. könnten hierher gehören." As noted above, Äki is found as first element of the !ost Sf Akethorp. It also occurs in DB Achebi NRY (SNPLY, p. 4). The Sf Acton is clear enough, cf. the forms to Acantune,fram Acyntune c. 1000 (e. 11th) ASWills, no. 15, which prove that the base is OE Ac(c)antun, 'Ac(c)a's tiin'. For the Nb example Mawer, PN NbDu, p. 2, takes the first element to be either OE Aca or the substantive äc 'oak'. On Björkman's suggestion, Mawer comments: "Björkman (Z. E. N., p. 12) suggests 0. N. Aki, but this seems unnecessary in face of the well- established English name." No Nf example occurs in the Testa de Nevill (Books of Fees) (see Fees, iii, Index Nominum, s. nn. Acketon, Aketon, Aketun etc.). The Nf forms medietatem molendini de Aketune e. 13th Der, fo. 34b(iv), de molendino de Aketon' ibid., fo. 34b(v), molendinum de Aketun' ibid., fo. 35a, in campa de Aketon' ibid., fo. 47b refer toa place now !ost in the hundred of Clackclose. We may be concemed here with ODan Äki as first element, but OE Ac(c)a and the substantive äc 'oak' are also possible. The interconsonantal e in the spelling Aketune etc. can be an AN inorganic vowel (on inorganic vowels between two consonants see PNDB, § 52.3). The common place-name Acton < OE äctun shows shortening of OE ä before two consonants in early ME (see Luick, § 352). Acton < äctun names also occur with AN inorganic vowels cf. Aketon 1211, 1213, 1316 (Acton Mx; PN Mx, p. 81), Acheton 1086 (Acton Reynold Sa; DEPN, s. n. Acton !.). It should be noted that Clackclose hundred does not lack signs of Scandinavian influence, e.g. Milneholm 1247-1248 Der, fo. 34a(iii), (f. n. in Wereham) (OE myten+ ON holmr), Derebouthe close 1562 Terr (f.n. in Shouldham) (OKS) (OE deor + ODan böth), but it does not lie in the main hybrid-tun area which is in east and central Nf. Achewelle may also have OEäc or OEAc(c)a as first element. There is no need to suggest ODanÄki. 14 y with the value of i isa common scribal device, see SPNLY, § 22, for full details.

5

Åki, Äki, Äke

Äskell in the present work). Beginning well before the middle of the 12th century, ä is rounded to Q in Southern and Midland dialects of ME. Spellings in o for ä first occur in manuscripts of the early 12th century and they become frequent in southern manuscripts in the course of the 12th century. In the 13th century they become general, reaching the northern Midlands by about 1300 (for the ä/Q boundary in ME see Kristensson, SMED, pp. 30-38, 283, Map 17). The appearance of forms in Oki at the end of the 12th century and the beginning of the 13th century in Nf is a manifestation of the development of ä > i). Note that in the Ely Survey of 1222 forms in Aki still occur, though the Oki forms predominate. For examples of Oki in the early 13th century in L see SPNLY, p. 4. On the rounding of ä to Q see Brunner, Abrill, § 11.4; Luick, § 369; Mosse, § 27 .2. (iii) Forms in Acke call for special comment. In some cases we may be merely concerned with a scribal variant of the forms Aki, Achi. The forms Gaufridus f Acke 1203 AssRoll and Gaufridus filius Achi 1209 AssRoll both relate to the same parish and are in all likelihood references to the same person. There is also a possibility that Galfridusf Aki 1209-1212 P might belong here. Björkman comments on the relationship of Acke to Äki: "Auch Acke Pedes Fin. (Surt. Soc.) S. 129, 130 dtirfte fernzuhalten sein" (ZEN, p. 12, n. 1). However, there are L forms which show Acke, Accha as an AScand short form of ON Äskell, i.e. de Accha Muding 1. Hy2 DCh, p. 391, Gunnildafilia Acke Mudding de Salfletby e. 13th FP, p. 118 (no. 96) = Aschel Mudding 1. 12th FP, p. 118 (no. 95). Stenton in a note to no. 96 says: As the Acke Mudding of this charter is certainly identical with the Aschel Mudding of the previous document it follows that Acke was a recognised short form of Askell in the Northem Danelaw. This example, and another, Ernaldus filius Acke, de Ernaldo f Akke (= Ernald' f Asce) 1202 LAss, are examined by von Feilitzen, Notes, pp. 64-65. He takes them to be short forms of Äskell, but remarks (p. 65): "In other eontexts ME Ac(c)he, Acke may go back to OE Acca (Redin 81)." In OE sources OE Acca does not seem to be found after the early 10th century (see Redin, p. 81; Ström, pp. 58, 59, for examples; Redin dates his latest example, contained in the form Eadwold Accan sunu, to 905). 15 The name does not seem to have completely died out in ME, cf. (quam dimidiam bo)uatam Acca tenuit c. 1210 (orig.) RA, vi, no. 1891 (Tetford L). The form Acka found in the witness list of an original charter of 858 (BCS, no. 496, Latin with an OE (Kentish) summary) is merely a scribal form. 15 This is known from the A and D texts of the ASC. The correct 6rciarson, fslenzk fornrit, IV, Rekjavik, 1939, p. 50) and P6rketils (gen.) in a verse attributed to Kormakr Qgmundarson (Vatnsdrela Saga, Hallfreöar Saga, Kormaks Saga, ed. E. 61. Sveinsson, fslenzk fornrit, VIII, Rekjavik, 1939, p. 268). 38 Elsewhere in OWScand, the following examples of the full formare found: arnktil (ON Arnke(ti)ll) in an 11th century Norwegian runic inscription (NiyR, iii, p. 163 (no. 230)), Halkretil Karlson (ON Hallke(ti)ll) in a Norwegian document of 1523 (LindS, col. 394), Rafnketlsstadir39 (ON Hrafnke(ti)ll),

37 The examples given by SMP, col. 150, s. n. Ar(n)kil are from the Liber daticus Lundensis and the Necrologium Lundense and thus belong to ODan as do the two Skåne place-names which contain the name as first element and which are cited by SMP, col. 151. 38 The piece of verse from Eyrbyggja Saga can be dated to 981 (see Eyrbyggja Saga, ed. Sveinsson and Mröarson, pp. xxxiv, 40--50). The grammatical and metrical forms in the verse ascribed to Kormakr point to composition in the 10th century, see E.O.G. Turville-Petre, Scaldic Poetry (Oxford 1976), p. 47. Turville-Petre also examines in detail the general question ofthe authenticity and the accuracy of the transmission of scaldic verse (ibid., pp. lxvi-lxxiv). See also the same author's The Origins of Old Icelandic Literature (Oxford 1953), pp. 44--45. 39 This place-name appears in the form Hrafnkelsstaöir in other documents (Lind, col. 566). Noreen, Ais!Gramm, §§ 274.2, Anm., 359.2, takes -ketl to have been formed by analogy with the dative -katle.

20

Arnkell, Arnketil, Ar(n)kil, !Er(n)kil

an lcelandic place-name form of 1397, and roskitil (ON Hrosske(ti)ll) in a 10th century runic inscription from the Isle of Man (NiyR, iii, p. 164, n. 1). In OSwed several examples of the full formare known from the 11th century runic inscriptions, e.g. brunkitil (SRU, no. 371) (OSwed (run.) Brunkretill), stinktil (SRSö, no. 70) (OSwed (run.) Streinkretill), Porkatil (SRSö, no. 229), Purkitils, gen. (SRSö, no. 113), Purktil, acc. (SRSö, no. 194), Purktil (SRU, no. 344) (OSwed (run.) Porkretill), ulfkitil (SRU, no. 633) (OSwed (run.) Ulfkretill), ujJkitil, acc. (SRVg, no. 102) (OSwed (run.) (t)ökretill < Auökretill). In Runic Danish there are only contracted forms in -krel, -kil (see DR, Ordog navneforråd, s. nn. Alfkil, Arnkil, Grimkil, Porkil, PyjJkil, Ulfkil , /E,skil, 0kil). Elsewhere in ODan DgP has the forms Magnus J- Arketil 1226-1241 (undated), Thrugillus diet. Arnkietelss: 1269, Thrugilli Arnkretelss: 1310 (DgP, col. 60) and Porcetel 1076-1080 coin (DgP, col. 1389). In view of the strong English element among Danish moneyers in the 11 th century, it might well be that the above mentioned Porcetel came from the Danelaw. Among the names of Scandinavian pilgrims from the Verbriiderungsbuch of the Swabian abbey of Reichenau, we have the following: Arnchetil 1. 11 th12th, fo. 151 (Reichenau, pp. 9, 16) (Arnkell etc.); Ascethle, Ascleit 1. llth12th, fo. 151 (Reichenau, pp. 9, 17), Askelet4° e. 12th, fo. 159 (Reichenau, pp. 12, 17), Ascetle 1. llth-12th, fo. 161 (Reichenau, pp. 13. 17) (Äskell etc.); Durchgetil e. 11th, fo. 8 (Reichenau, pp. 6, 7, 22) (Porkell etc.). The c. 670 Scand names in the Reichenau list are from the 11th and 12th centuries.41 It is noteworthy that in names in -ketill, -kell, -kil the contracted forms predominate here (see Reichenau, pp. 17, 18, 19, 21, 22, s. nn. Äskell, Eskill, Grimkell, Heimkell, Steinkell, Ulfkell, Porkell). The OScand evidence suggests that full forms in -ketill existed side by side with the more numerous contracted forms in -kell, -kil until the 11 th century in OSwed and ON. In ODan the contracted forms seem to have supplanted the full forms at an earlier date. The appearance of full forms in later medieval documents is difficult to explain. One should note that fem. names in -katla

°

4 Finnur Jonsson comments (Reichenau, p. 17): "Navneformeme Ascelet (f. Asketel ??) osv. er ret mrerkelige, men reprresenterer mäske en form Åsketl, en mellenform mellem Asketill og

Askell. 41 The names on fo. 8 belong to the beginning ofthe 11th century (Reichenau, p. 6). The names on fo. 151 seem to be later than the close of the 11th century (see Reichenau, p. 9, n. 2 and Reichenau Tavle I, opposite p. 16). The group of 13 lcelandic names on fo. 159 belongs to the second half of the 12th century (Reichenau, pp. 6, 12), but it is in a different hand to the other names on the same folio (Reichenau, p. 12, n.2). These datings are based on the handwriting. Forms found elsewhere in the Reichenau list have been dated in the present work on grounds of their onomastic context, that is, the names belonging to the same folio and same hand have been taken as a group and dated according to phonological criteria.

21

Arnkell, Arnketil, Ar(n)kil, lEr(n)kil are largely OWScand,42 e.g. Arnkatla, Hallkatla, Oddkatla, Porkatla (see Lind, s.n. -katta). ODan has only runic askatla (DR, no. 120) of DR Per. 2.2 (Viking Period; inscription type is Efter-Jelling), and there are a few late medieval examples of ODan Thorkel fem. (DgP, cols. 1388-1389; since the forms of this name, which DgP, col. 1389, takes as cognate with ON Porkatla, end in -kilt, -kylda, -ckel, -kel, it is in any case not relevant to the discussion). It is unlikely that medieval Scandinavian forms in -ketill, -kcetill were the result of association with fem. names in -katta. It might be that association with the simplex name Ketill, Ketil, Kreti! is the reason for the occasional appearance in later medieval sources of -ketill, -kt:etill instead of the normal -kelt, -kil. In England Arnkell/Arnketil/Arkil is known in both full and contracted forms: NPN, p. 8; ZEN, p. 14; Björkman, Festermen, pp. 5, 7; PNDB, p. 163; SPNT, p. 142; DBS, p. 11, s.n. Arkell; Ekwall, SelectedPapers, p. 96; SPNLY, pp. 14-16, 353; Smart, Moneyers, p. 228; von Feilitzen, SCBI, xi.i, p. 12; Smart, Cnut's York Moneyers, p. 224; SCBI, xxi, pp. xlvi, xlvii, xlviii, xlix; DEPN, p. 12, s. nn. Arkengarthdale, Arkle Beck, Arkle Town, Arkleby, Arksey, Arkston; PN La, pp. 168,254; PN NRY, pp. 253, 260,295,332; PN Nth, p. 291; PN Ess, pp. 516-517; PN Wa, p. xxi; PN Cu, pp. 310, 325, 506; PN Db, p. 762; PN WRY, i, p. 24; PN WRY, ii, p. 231; PN WRY, v, p. 99; PNWRY, vi, pp. 52,156; PNWe, i,p. 60; PNWe,ii,pp. 34,160; SSNY,p. 152; Insley, JEPN, x (1977-1978), p. 48; Insley, Nomina, iii (1979), p. 55. Note also the following examples: /Ernketel 978-1016 (e. 14th) KCD, no. 971 (Hickling and Kinoulton Nt and Lockington ERY ;43 Arcil 1038-1040 SCBI, xviii, nos. 506, 507 (moneyer, Stamford L); Archetel 1087-1098 (c. 1175-1200) Bury, p. 38 (Honington Si); Rodberto filio ArkiL-abl. (witn.) Hy2 DCh, no. 354, Roberto filio Arkilli, abl. (witn.) 1. Hy2 DCh, no. 350 (Eakring Nt); Rogero filio Arkil de Clawrda, abl. (witn.) c. 1200 DCh, no. 421 (Mattersey Nt; Clawrda is Clayworth Nt); quendam liberum hominem Arkeld, Arkild 1201 AssRoll 1171 (Pleas, ii, no. 619) (Co); Gillebertusfilius Arki/44 1202 CRR 28 (Pleas, ii, no. 758) (Papworth Everard and St Agnes C), Gileberti filii Archil, gen., 1212 CRR 54 (Cur, vi, p, 197) (C); Radulfusfilius Arkilli 1208 CRR48 (Cur, v, p. 258) (Y); Willelmus filius Arkil 1212 CRR 56 (Cur, vi, p. 330) (Headingley WRY).

The relative infrequency of the full form in -ketel, -chetel etc. in English records is noteworthy. In DB the full form occurs TRE only in He, while the contracted form in -chel, -chil(lus) is found in DB TRE in Ch, He, L, Lei, Nt, Sf, St, Wa and Y (PNDB, p. 163). In Y SPNLY, pp. 14--16, has 74 forms, of

42 For fem. names in -katla, which are almost entirely Icelandic in OWScand, see Lind, s. nn. Arnkatla, Askatla, H allkatla, Oddkatla, *Solkatla, Porkatla. 43 Thisk'rnketel also appears asArnketel 1019 (m. 14th) Rams, iii, p. 167 (obituary of Amketel and his wife Wlfron). Amketel's wife bore the OE name Wulfrun and their son was k'öelstiin, abbot of Ramsey from 1020 or 1021 to 1043. DBS, p. 11, s. n. Arnkell wrongly ascribes this Arnketel to Nf. For KCD no. 971 see Sawyer, no. 1493. 44 Corrected from Arketill'.

22

Arnkell, Arnketil, Ar(n)kil, JEr(n)kil

which only two, both from before the Conquest, represent the full form. 45 In England the name is well attested in Y but is rather infrequent elsewhere. In Normandy the name occurs in its full form in one place-name (Adigard, p. 379), and is represented in independent use by the undated form Architellus (Adigard, p. 275, n. 10). In Normandy personal names in OScand -ketill, -kell usually appear in the full form, e.g. Adigard, pp. 287-294, lists 34 persons for the period before 1066 who bore the name Aske(ti)ll in its full form and only two persons who bore this name in its contracted form. Adigard, pp. 322-326, has 13 persons from the period before 1066 who bore the OScand name Porke(ti)ll in its full form. The contracted form Porkell is known in Normandy only from a place-name (Adigard, p. 429). Forms in -chil(l), -kil(/) show the OEScand development -hetill > -hell> -kilt in which ;e is raised to i when it precedes palatal / (Il or l + consonant) and follows certain consonants, especially g and k (Brf,indum-Nielsen, § 146.2; Hald, Sprogstof, p. 151; Modeer, SvP, p, 54). In ODan this development seems to have been quite early because forms in -kil predominate in Danish runic inscriptions. 46 Loss of interconsonantal -n- in Archetel, Arkil etc. may be a 1. OE development (cf. Campbell, § 477.5), but it may be also Scand, for it is known in both OEScand (Brf,indum-Nielsen, § 372; Modeer, SvP, p. 54) and OWScand (Noreen, AislGramm, § 291.9). PNDB, § 77, takes the loss ofinterconsonantal-nto be the result of OE or OScand sound development, but adds that it might have been favoured by the AN tendency to avoid heavy consonant groups. From the evidence of OE and OScand forms, the loss of interconsonantal -nin dithematic personal names in OScand Arn- in England is an OScand or an OE development, but after 1066 it would certainly have been accentuated by the AN tendency to reduce heavy consonant groups. The spelling -ch- for /kl is normal in early AN orthography belore e and i (PNDB, § 113).

45 Arkitel minister (witn.) 958 (14th) YCh; Arcetel (festerman of .t'Elfric in Y) c. 1050 (c. 1075) YCh (also Björkman, Festermen, p. 5). 46 Forms in -kel occur in two runic coin inscriptions, i. e. krimkel (ODan Grimkil) 1065-1075 DR, M0. 45c, s ....kel 1065-1075 DR, M0. 91 (both Lund). In view of the English element among 11th century Danish moneyers, it is possible that these two had come to Denmark from the Danelaw, and, if this is the case, the -kel forms would not be unusual. Runic Purkal (DR, no. 132) probably represents Purkad, the older form of Purkil (DR, col. 674, s. n. -kil, Anm. 1), though there isa possibility that the second element -kal represents -karl with assimilation of -r- (see DR, col. 728, s. n. Porkil, Anm. 1). DgP, col. 1389, is sceptical about the idea that Purkal isa compound in -karl and favours derivation of the second element from -ka?l.

23

Åsa, Åsa

ON Åsa, ODan, OSwed Äsa fem. b) (i) sorori sue Ose, dat. (Edmundus auunculus Ricardi filii Rogeri), Ose mater sua (Ricardus filius Rogeri), predicta Osa 1209 AssRoll 558 (Pleas, iv, no. 4368) (Saxlingham and Field Dalling).

Asa is known throughout Scandinavia (Lind, cols. 60---62; LindS, cols. 46--53; DgP, cols. 64, 1640; SMP, cols. 176--177), and isa short form of such fem. names as AsbjQrg, Asdis, Asgerör, Ashildr, Aslaug etc. (NK VII, p. 61). For English examples of the name see NPN, pp. 9-10, PNDB, p. 165, SPNT, p. 142, and SPNLY, p. 18. Initial Os- represents Anglicization in which OE Ös- has replaced the corresponding OScand As-. The substitution of OE Ös- for ON As- is common in East Anglia (see below, s. nn. Asgautr, Asgeirr, Askel[).

ON Åsbhi_>rn, ODan Esbiorn, OSwed IEsbiorn a) (ii) Osebernescrundellond c. 1220---1230 Wals, fo. 22a (f. n. in Little Walsingham). Osbernesheuedlond c. 1230 Der, fo. 40b(ii) (f. n. in Boughton). - Osbernysdole 1503 /pm (f. n. in Wiggenhall St Mary the Virgin) (OKS). b) (i) Esbern Wml (1074-1077 ?) SCBI, xi.ii, no. 25 (moneyer, Thetford), (ii) Osbern 997-1003 SCBI, ii, no. 864, SCBI, vii, nos. 1231, 1232, 1003-1009 SCBI, vii, nos. 1233, 1234, Osberen 1003-1009 SCBI, vii, no. 1230, Osber 991-997 SCBI, vii, no. 1229, Osbrn 997-1003 SCBI, vii, no. 1235 (moneyer, Thetford). - Osbern (witn.) 1043-1044 (l. 13th) ASWills, no. 31 (Nf). - Osbernus TRE DB, fo. 185 (Shipden). - Osbernus teinnus TRE DB, fo. 267b (Hudeston in Bunwell), Obernus teinnus TRE DB, fo. 267b (Raveningham). - Osbernus de Daneuala c. 1090---1100 CAcre, fo. lb (francigenus of Hugh de Wanci, gave tithes to Castle Acre Priory). - Osbernus et Radulfus filii eius (Osbem and Ralph were the sons ofHugh de Wanci)47 c. 1090---1100 CAcre, fo. 1b (charter of William Il de W arenne in which the grant of Hugh de W anci of the churches of Deepden and Barsham and the tithes of Hugh' s francigeni were confirmed to Castle Acre Priory). -Osbernus de Gloecestre c. 1090---1100 CAcre, fo. la(ii), duas solidatas terre quas Osbern de Gloecestre tenebat 1146-1147 CAcre, fo. 3b(i) (Castle Acre). 48 -de hornagio Osbernifilii Wlfrici? 12th (m. 14th) Rams, i, p. 100, no. 404 (? Upwell). - Osberni Fabri, gen.,? 12th (m. 14th) Rams, i, p. 100, no. 405 (Welney). - terram Oseberni archidiaconi 1100---1101 (l. 12th - e. 13th) NorwCPCh, no. 112, terram Osberni Archdiaconi 1100---1101 (e. 14th) NorwCP, p. 34 (Lakenham). Osbern de Byntre (abl.) (witn.) 1101-1107 Binham, fo. 2a (grant by Peter de Valoignes to Binham Priory; Byntre is Bintree). -Osberno Presbitero, abl. (witn.) 1114-1130 (m.

47 The Norman family of de Wancy came from Wanchy (Seine-Infärieure) (see DBS, p. 342, s. n. Wansey). 48 Osbern(us) de Gloecestre also occurs as Osberto de Gloecestre, abl. (witn.) c. 1090---1100 CAcre, fo. la(ii). Gloecestre is Gloucester. This confusion of Osbern with Osbert (< OE Ösbeorht or OSax Ösber(h)t) is common. See below for further examples.

24

ÅsbiQrn, Esbiorn, lEsbiorn 14th) Rams, i, p. 144, no. 70 (Outwell). - Osberno de Redham, abl. (witn.) 1127-1134 HarlCh 44 E 19 (printed, Holme, Appendix E) (Heigham next Norwich), Osbernus de Redham (witn.) 1134--1140 Holme, no. 131 (Tibenham), ibid., no. 137 (lease ofland in South Erpingham H and in Tunstead H pertaining to the Abbey' s desmesne of Scottow to Wyther Tumel), ibid., no. 139 (Hellesdon), Osbernus miles (witn.) 1134--1140 BodlNfCh, no. 604 (printed, Holme, no. 135) (Little Melton), Osbernus de Redeham (witn.) 1146-1149 BodlNfCh, no. 609 (printed, Holme, no. 156) (South Walsham and Panxworth),49 Osbernus pater suus, pater suus Osbernus (Osbem was the father of Stephanus de Redham) 1175-1186 Holme, no. 203 (Reedham, Scottow, Lamas, Coldham, Sco Ruston and Norton Subcourse), Osbernus antecessor suus (Stephanus de Redham) 1237-1251 Holme, fo. 116a(ii) (Scottow and Lamas; Osbernus is said to haveheld in the time of Henry Il and is, therefore, the Osbernus de Red(e)ham mentioned in the previous entries). - Transcriptum Osberni presbiteri, Osberno presbitero, dat., 1134-1140 Holme, no. 138 (Paston), Osbernus presbiterprob. 1153-1168 BodlNfCh, no. 608 (printed, Holme, Appendix B) (Thwaite Hall and Paston),50 Osbernus de Tweyt (presbiter) (witn.) 1153-1168 Holme, no. 181 (Thume; Tweyt is Thwaite Hall in South Erpingham H). - Transcriptum Osberni de Thurgertune, 51 Osberno de Thurgerton', dat., Osberno presbitero, dat., 1146-1149 Holme, no. 147 (Thurgarton), 52 Osbernus de Turgerton' (witn.) 1175-1186 Holme, no. 203 (Reedham, Scottow, Lamas, Coldham, Sco Ruston and Norton Subcourse), Osbernus decanus de Turgertona (witn.) 11751186 Holme, no. 207 (Antingham). -Terram etiam que fuit Osberni Blund' 53 de Hindringham 1146 - c. 1170 (15th) NorwCPCh, no. 130 (Hindringham). - Osbernus armiger Morelli 54 (witn.)? m. 12th (1302) NorwCPCh, no. 259 (12) (Filby). - Osbernus c. 1150 (1302) NorwCPCh, no. 259 (2) (Norwich). 55 -terra que fuit Osberni presbiteri in Thima 1153-1168 Holme, no. 181 (Thume). - Rogero filio Osberni, abl. (witn.) c. 1155 CAcre, fo. 106b(i) (Tuddenham). - Osbernus Altus 1161-1177 (13th) Rams, iii, p. 286 (Wimbotsham and Downham Market). - Osbernus (witn.) 1175-1186 BodlNfCh, no. 610 (printed, Holme, no. 225) (Yarmouth). - Osbernus medicus (witn.) 1175-1186 Holme, no. 236 (North Walsham). - Willelmofilio Osberni, abl. (witn.) c. 1180 Binham, fo. 20a(i) (Hindringham). - terram Osbern c. 1200--1220 Bromh, fo. 52a(iii) (Somerton). -Osbernusfilius Botild' et Henricusfrater suus 1209 AssRoll 558 (Pleas, iv, no. 4145) (Nf). - Aliz filia Osberni de Scothowe 1214--1229 Holme, fo. 8 la(v) (Swanton Abbot; Scothowe is Scottow). - Hosebern Kailie (abl.) (witn.) c. 1230 CAcre, fo. 47b(ii) (South Creake). 56 - terram Philippi filii Osebern, terram eiusdem Philippifilii Hosebern c. 1230 CAcre, fo. 48a(i) (South Creake).

Wrongly dated by the editorofHolme to 1141-1149. The document is probably to be dated to 1153-1168, but the entry conceming Osbernus presbiterrelates to the period 1135-1140. 51 Osbem of Thurgarton is doubtless the same person as Osberti presbiteri, gen., Osberto presbitero de Turgertone, dat., 1146-1149 (wrongly dated by the editor of Holme to 1141-1149) Holme, no. 146 (land between Runton and Felbrigg), Osbertus de Turgertona decanus (witn.) 1175-1186 Holme, no. 206 (South Walsham, Irstead, Swanton Abbot, Haddon in Hoveton and the marsh of W ebbecros ). 52 Wrongly dated by the editor of Holme to 1141-1149. 53 Var: Hosberni Blund (1. 13th). 54 Morell de Merlay. 55 Var: Osbertus c. 1150 (e. 14th) NorwCP, p. 64. 56 The family of Cailli came originally from Cailly (Seine-Inferieure) in Normandy (DBS, p. 62, s.n. Caley). The name Osbert was used by both the English and Norman branches of the 49

50

25

ÅsbiQrn, Esbiorn, lEsbiorn AsbiQrn is very common in OWScand (Lind, cols. 63-65; LinvdS, cols. ~3-64). In OEScand the name is usually found in its i-mutated form Esbiorn, &sbiorn (DR, col. 746; DgP, cols. 260-265; SRÖl, no. 56; SRÖg, no. 115; SRSö, nos. 205,213,347, 733; SRSmå, no. 45; SRU, nos. 394, 726; SRVg, nos. 95, 182, 157, 178, 184; Lundgren-Brate, pp. 311-312). Late medieval instances of forms in JEs-, Es- in Norway are loans from OEScand (NK VII, p. 65; SPNLY, p. 18). Runic Danish forms in qs- and as- are also known (DR, col. 746), and these represent the unmutated form. The unmutated form is fairly well represented in OSwed (SRÖg, nos. 22, 47, 153; SRU, nos. 40, 186, 258, 432,586,627, 1028, 1096, 1106; SRSö, no. 266; SRVg, nos. 77, 101; SMP, cols. 177-178). Of the above Nf forms, Esbern represents the OEScand variant with i-mutation. Some difficulty arises with forms in Os-. SPNLY, § 11.iii, takes Os- to represent either the substitution of OE Ös- for OScand As- or a Scand development from *ansu (by u-mutation) to Qs- and then (by nasalization) to Os- or possibly, in the case of such names as Osbern, Osfyrö and Osmund, rounding under the influence of a following labial consonant (b, f, m). As references SPNLY, § 11.iii, cites Br0ndum-Nielsen, §§ 85 Anm 3, 154.2 and DR, cols. 115-117, 695 Anm 1). The references in Br0ndum-Nielsen refer, however, to the substantive, PrScand *ansuR m. 'god', in its function as a rune name (see Br0ndumNielsen, I, Ordregister, s. v. äs). Here there does appear to be a development *ansuR > i)ss > os(s) (Br0ndum-Nielsen, § 154.2). The references in DR refer to the forms usbiaur (DR, no. 80), usbiarnaR, gen. (DR, no. 81). The editors of DR transcribe the two forms into Modem Danish as "Osbj0m(?)" (DR, no. 80) and "Osbj0ms(?)" (DR, no. 81). DR, col. 695, Anm. 1, lists several theories about these forms, but favours as most probable the suggestion of Marius Kristensen that they are u-mutated forms of Äsbiorn with u for initial o. DR, col. 695 Anm 1, mentions that Os- is normally a sign of English influence but believes that it can also be native to Scandinavia, citing runic ausualti (SRÖg, no. 224) as a certain example and runic ausburn (SRÖg, no. 61) as a probable example. DR, col. 695 Anm. 1, believes it is unlikely that the usbiaur, usbiarnaR forms derive from an OScand Husbiorn or a Danish cognate of OSwed (Z)sbyr/)rn (< 0st-). SRÖg, no. 224, transcribes runic ausualti into Modem Swedish as Osvalde, with runic au representing u-mutated nasalized q-, while SRÖg, no. 61, takes runic ausburn to represent either OSwed 0s(t)br/Jrn or a form of Äsbiorn with family (Farrer, Honors and Knights' Fees, iii, pp. 382-384). Cf. Osbertus de Cailleo (witn.) 1118-1138 CAcre, fo. 3a(ii). Probably the formHosebern is here a case of confusion with Osbert. Farrer, Honors and Knights' Fees, iii, pp. 383-384, gives references toa certain Osbert de Cailly (de Kaylly), who appeared in documents of the period 1244-1287, and who held in Nf and Sf.

26

ÅsbiQrn, Esbiorn, lEsbiorn u-mutation. A further example is SRÖg, no. 226, where runic uslakr is transcribed into Modern Swedish as Oslak, the name being taken as a variant of OSwed Äsläk, with initial runic u- representing a mutated form of nasalized q-. In all three cases, however, u-mutation seems very unlikely. None of the second elements of the names (runic -ualti for OSwed -valde; runic -burn for OSwed-b()rn, cf. SRSmå, no. 45, where runic esburn, acc. is presumed to represent an OSwed &sb()rn; runic -lakr for OSwed (run.) -läkR) could have caused u-mutation of the first element As- and the -u- of the PrScand personal name elementJsu- < *Ansu- was probably syncopated before u-mutation took place (see Noreen, AislGramm, § 80.1). In view of these considerations, it must be presumed that runic us- represents the name element Ös-, which is the Ingvaeonic cognate of OScand As- and which would have been borrowed into OEScand from OE or OFris or OSax. In OSwed there are several cases of runic os- for OScand As- (examples of Äsbiorn with initial runic os- are: SRSö, no. 266; SRU, nos. 186,586; SRVg, nos. 77, 101), but here the initial runic ois merely an orthographic device for the representation of nasalized q- (SRVg, pp. xliv, li). In ODan Osbern is known from several old annals, from a document of 1157-1182, which only exists in copy form, and among the names of 11th century moneyers (DgP, col. 260). Hald, Om Personnavnene, p. 184, takes the 11 th century Danish moneyers' names Osgod, Osgut, Osbrn and Osward to be OScand names in As- from the Danelaw. This is certainly true of Osgod, Osgut < ODan Äsgot, -gut, and of Osbrn < Osbern, where OE Ös- has replaced the corresponding OScand As-, but in the case of Osward the etymon may be OE Ösweard rather than the rare Scand Asvarör (see below, s.n. Asvarör). There is no need to postulate, as SPNLY, § 11.iii, does, the possibility that the English forms Osbern, Osfyrö and Osmund have initial 0- through Scand rounding of A- under the influence of the following labial consonant. For Osbern, one can postulate Anglicization of ON AsbiQrn etc. or an OSax origin (for the reasons why OSax origin is unlikely see below). For Osmund, OE, OSax (formally possible but unlikely) or Anglicized OScand origin can be postulated (see below, s.n. Asmundr), while Osfyrö represents OE Ösfriö and not the rare OScand Asfriör. Anglicization of OScand names in As- through the substitution of the corresponding OE Ös- is very common in East Anglia (see below, s. nn. Asgautr, Asgeirr, Asket[), but in the case of Osbern the issue is complicated by the existence of OSax Ösbern (Schlaug I, p. 135; Schlaug Il, p. 139). In Normandy Adigard, pp. 275-287, has 55 persons bearing the name Osbern from the period between 968 and the late 11th century. Note also the following northem French examples: Depositio Osberni, monachi et sacerdotis S. Crucis 9th ObitSens, p. 247 (La Croix-Saint-Leufroy (Eure)); Osbernus, monachus et sacerdos Sancti Germani Parisiacensis 11th ObitSens, p. 278 27

Åsbi'i_>rn, Esbiorn, !Esbiorn (Saint-Germain-des-Pres (Paris)); Osbernus 1. 1lth-e. 12th Longnon, ii, p. 62 (Baugency (Loiret)). The 9th century example, which is from an area which later belonged to the Duchy of Normandy, clearly represents OSax Ösbern. It may be the result of Saxon settlements Åsrpör), Ösforö

A further possibility is that the forms in -bern in Norman and post-Conquest English sources reflect the influence of the Continental Germanic (OHG and OSax) name element -bern. 69

ON *Åsfr00r (> Åsr00r), AScand ÖsforO a) (i) Osfordismere c. 1180 Wymondham, fo. 95b(ii), Osfordesmere, ibid., fo. 95b(iii) (f. n. in Wicklewood). b) (i) Osfordus TRE DB, fo. 168 (S. Pickenham). -/.fiberhomo Osfort TRE DB, fo. 186b (held in Alby, sub Heroldo).

ON Asn,,ör < *Asjr()ör70 occurs in the Icelandic Landnamab6k as Asroör, AsRaudr, the name of one of the first settlers in Iceland, and as qsrupr in a Manx runic inscription (Lind, col. 85). In England the name is found among the names of pre-Conquest moneyers in the forms Asferd, Asferö, Asferth,71 Asforö (Coinage of Edgar, p. 188; Smart, Moneyers, p. 228; Smart, Cnut's York Moneyers, pp. 224--225; SCBI, xiv, nos. 1545-1548, 2165; SCBI, xviii, nos. 928-929). The mint of the examples from Edgar's reign is unknown, but the other examples come from the mints of York (Asferö, Asferth, Asferd), Lincoln (Asferö, Asforö) and London (Asferö) and belong to the 11th century. DB has Asford (TRE Nt, Y; 1086 L), Asfort (TRE, 1086 L), Asuert (1086 L), and the forms Asford and Asforth occur in medieval sources from L (PNDB, p. 165; SPNLY, p. 20). The name also forms the first element ofthe place-names Asfordby Lei and Asserby L72 (DEPN, pp. 13-14, 17, s. nn.; SPNLY, p. 20). DB Aseret (St) might contain ON Asr()Ör, but ODan (run.) Äsriipr is also possible (PNDB, p. 169). The Nf forms Osford, Osfort show the replacement of ON As- by OE Ös-. The second element -Jord, fort represents AScand -forö < ON -jr()Ör (PNDB, § 14). Final post-consonantal t for OE [8], OScand [ö] is an AN spelling with

69 For examples of such names from English sources see PNDB, pp. 141,253, s. nn. Ailbernus, Fridebern, Frithu-, Frethubern. 7 For loss of interconsonantal/ see Noreen, Ais!Gramm, § 291.4. 71 Hald, Om Personnavnene, p. 187, takes Asferth, Asferd, which is found among the 11th century Danish moneyers' names, to be a hybrid formed from a Scand Ä.sfrior and OE Osfero. It is more likely, however, that the name is an OScand Ä.sfrior, Ä.sjr(ftor, whose second element has been replaced by OE -fero, though the existence of OE Osfero probably influenced the process of Anglicization. The hybrid form of the name makes it clear that it was brought to Denmark from the Danelaw. The runic Danish name asfarjJ, borne by a moneyer in Lund in the period 1065-75 (DR, Mp. 13a--e, 14) is of similar origin (cf. DR, col. 632). 72 An early form of Asserby, Esfordebi 1202 SPNLY, p. 20, has initial Es- by analogy with such names as Esbern, Esger, which have initial Es- < Äs- as a result of OEScand secondary i-mutation.

°

33

*ÅsfrJ;ör (> ÅsrJ;ör), Ösforö the phonetic value [t] and d is an inverted spelling for [t] (PNDB, § 107; SPNLY, § 116). Forms such as Osfyrö 979-985, Osferö 979-991, Osfeö 991-997 Smart, Moneyers, p. 244 (moneyer, Thetford), Osfert 1086 DB, fo. 185 (Barningham Norwood) and Osferö (gen.) 1176-1177 P, p. 138 (Nf or Sf, prob. Nt) represent OE Ösfriö. 73 In Normandy and post-Conquest England the forms An(s)fred(us) and An(s)frid(us) are common (Adigard, pp. 348-356; NPN, p. 11; PNDB, p. 161; SPNLY, p. 19). NPN, p. 11, takes Norman Ansfrid to have a base of OScand origin. PNDB, p. 161, regards Ansfrid as an OG name, but elsewhere (PNDB, p. 164) takes Ansfred to be a Norman form of ODan Äsfrith, in which ODan Äs- had been replaced by Frankish Ans-. SPNLY, p. 20, comments: It is uncertain whether the majority of the L and Y forms in -frid do in fäet represent Scand AsfrilJr, but since the possibility cannot be excluded they have been included here. It is perhaps more likely that they derive from ContGerm Ansfred, Ansfrid or OE Eanfrid (Forssner 28), although the An(s)- spellings could represent adaptation of the Scand name by AN scribes to forms to which they were more accustomed.

OE Eanfriö, though known in the 7th century from the ruling houses ofBemicia and of the Hwicce (Ström, p. 166), is rare and can be ruled out as a possible etymon for the examples of Anfred(us), -frid(us) found in English sources of the ME period. 74 ODan Äsfrith is also rare and is only known in the Viking period from Frankish sources (DgP, col. 65). However, FrankishAnsfred, -frid is well attested in northem France. In the 9th century it is known from the Paris region and the Thimerais and in the 10th century from the Drouais (Adigard, p. 181). In Normandy the name An(s)fred(us), -frid(us) is borne by 33 persons before 1066 and it is the first element of 10 place-names (Adigard, pp. 348-356, 437-438). Adigard, pp. 182-183, takes the appearance of the OScand name in Denmark and the Danelaw and the presence of An(s)fred(us), -frid(us) in close proxirnity to OScand names in Normandy together with the rarity of the Frankish name in place-names in the rest of northem France to be

73 OE jera is the result of a development -fria > -frea > jera (PNDB, § 14). In the second elements of compounds [i] can become [~] owing to reduction of stress (Campbell, § 371). Forms in -frea are known from the 9th century, e.g. Herefrea, Sefrea, Sigefrea 867-870 (contemp.) Harmer, SEHD, no. 7. Sometimes forms in -fria and -frea occur in the same document, e.g. Beornfria, Sefrea 843---863 (contemp.) Harmer, SEHD, no. 4. The form with metathesis, jera, is known in the 9th and 10th centuries, e.g. Beorrifera 855-870 SCBI, ii, no. 422, Osferaes, gen. 899-924 (10th) Harmer, SEHD, no. 18. The form Osferd Hlytte 911 (c. 1000) ASC (B), 911 (m. 11 th) ASC (C) contains a fully Anglicized form of OScand Äsfriar, *Asfrr/)ar. The spelling -fyra in Osfyra shows the development-fria > jira (see PNDB, § 14), with subsequent rounding of i> y before r (see Luick, § 285.2). 74 Apart from its appearance as the first element of the field name eariferjJes hlau 956 PN 0, p. 485, OE Eanfria is not attested after the 8th century.

34

Åsgautr, Åsgot, Åsgut

arguments for the Scand origin of the Norman forms. Adigard, p. 183, allows for the possibility that the OScand and the Frankish name had fallen together and become indistinguishable in Normandy. The arguments in favour of an OScand origin for Norman An(s)fred(us), jrid(us) are unconvincing. 75 The OScand name is very rare. The English form Asforö seems to reflect an OWScand *Äsfrpör and Asferö is probably merely a variant in which the second element has been replaced by its OE equivalent. An OScand Äsfriör is only known from ODan and there it is rare. The Frankish name Ansfred, -frid is, however, adequately attested in the pre-Scand period in northem France and the validity of this evidence is not affected by the apparent rarity of Ans/rid as the first element of northem French place-names. Since no full survey of the place-names of the region between the Channel and the Loire has been made, it is difficult to make definitive statements about the distribution of particular personal names in the place-names of this area. In view of this evidence, the East Anglian forms of An(s)fred(us), -frid(us) are not included in the present work since they are here regarded as having West Frankish Ans/rid etc. as etymon. Norman An(s)fred(us), -frid(us) does not occur very frequently in East Anglian sources of the ME period and is more common in Sf than in Nf where it is extremely rare.

ON Åsgautr, ODan, OSwed Åsgot, Åsgut a) (ii) Osegodes renneles c. 1190 Holme, no. 266 (f. n., prob. in Hoveton). 76 - Osgoteshil c. 1200 Wals, fo. 87a(i) (f. n. in Weyboume). - Osegoteshag' 1227-1236 NorwCPCh, no. 193, Osegoteshage 1227-1236 (e. 14th) NorwCP, p. 110, Osegateshag' 1246 (1. 13th) NorwCPCh, no. 212, Hosgoteshawe 1273 (e. 14th) NorwCPCh, no. 310 (f. n. in Cressingham). - Hosgoteshirne c. 1235 Der, fo. 151 b(i) (f. n. in Tilney). Osgotesdole 1237-1251 Holme, fo. 113a(i)(f. n. in Honing). -Hosegotesslede c. 1275 Coxford Il, fo. 46b (f. n. in Tattersett). - Hosegoddescroft 1. 13th AddCh 59839, Osgotescroft 1290 AddCh 59826 (f. n. in Briningham). - Osegotishawe 13661pm (f. n. in Mulbarton). b) (i) Rogerifilii Asgoti, gen., c. 1170-1180 Wymondham, fo. 32b(ii) (Happisburgh).

75 A document of 1035-1037 which is known from a 14th-century copy contains the forms Asfredivilla, Aschitillus rufus and Aschetillus de Coltun (Adigard, pp. 83, n. 95, 182 and 182, n. 34, 245, n. 6, 290, 437). This might be a case of scribal error or it might be that the forms Aschitillus and Aschetillus were forms from the Danelaw which had kept initial OScand As- and that the initial As- in the place-name form Asfredivilla was written through analogy with these personal name forms. For the Norman form Hasgerus see below, p. 40. 76 From the evidence provided by the witness list, the editor of Holme dates this charter to some date defore 1196.

35

Åsgautr, Äsgot, Äsgut (ii) Osgote. Clape. (witn.) 1043-1045 (1. 13th) ASWills, no. 31 (Nf). 77 - Osgotus TRE DB, fo. 152 (Kelling). - Osgotus 1086 DB, fo. 202 (Gooderstone). - Osgod presbitero de Binham, abl. (witn.), Osgoto presbitero de Gunthorp, abl. (witn.) 1101-1107 Binham, fo. lb (grant by Peter de V aloignes to Binham Priory; Gunthorp is Gunthorpe, Holt H). -Osgotf Alfrici de Len 1193 P, p. 24, Osegot78 f Aluric' de Len 1194 P, p. 59, Osgotusf Alurici 1195 P, p. 70, 1196 P, p. 130, 1197 P, p. 230 (Lynn). -Godardumfilium Alueue. et Rogerum et Willelmum. et Johannem et Osegod filios Godardi, Osegod (nom.), Osegod (dat.) 1198 FF (PRS NS, xxvii, no. 20) (West Winch). -domum Osegot 1198 FF (PRS NS, xxvii, no. 36) (Hardwick). -Adamf Osgot 1198 P, p. 93, Adamf Osgoti 1199 P, p. 273, 1200 P, p. 137 (Nf). - Edwardus f Osgoti 1199 P, p. 285, 1200 P, p. 144 (Nf or Sf, prob. Nf). - Willelmum Osgot' 1202 FF (PRS NS, xxvii, no. 410) (Grouele). -Rois' filiam Osegod' 1209 CRR 49 (Pleas, iv, no. 3737) (Nf). - Walterus f Osgoti 1209 P, p. 34 (Nf). - deAliciafilia Willelmi Osegot c. 1210-1220 BodlNJCh, no. 22 (Birlingham). iii) Ansgot (gen.) TRE DB, fo. 185b (one fiber homo in Trowse Newton was commended to Ans got TRE). - terra Ansgoti camerarii 1123-1133 Wymondham, fo. 70b(i) (Happisburgh),Ansgoto camerario, abl. (witn.) c. 1135 Binham, fo. 90a(ii) (Edgefield), Ansegod Camerarium 1166 RBE, i, p. 398 (Nf), terram Ansgodi camerarii 1182 Wymondham, fo. 10a, terra que fuit Ansgoti camerarii 1193-1199 Wymondham, fo. 20a, terra Ansgoti camerarii 1230 Wymondham, fo. 22a (Happisburgh), de feodo Rogeri filii Ansgoti 1176-1189 Wymondham, fo. 18b, 1193-1199 Wymondham, fo. 19b, ex dono Rogeri filii Ansgoti 1182 Wymondham, fo. 10a (Wymondham and Happisburgh), Rogero filio Ansgoti, abl. (witn.) c. 1185 Wymondham, fos. 37a(ii), 38a (Nf), in manu Rogeri filii Ansgoti camerarii et Anger nepotis eius c. 1185-1190 Wymondham, fo. 43a (Wicklewood), Rogerus filius Angoth c. 1200 Bromh, fo. 39a(iii) (Rogerus filius Angoth grants land which Gocellinus de Bromhol' held of his fee to Bromholm Priory), ex dono Rogeri filii Ansgoti 1230 Wymondham, fo. 22b (Wymondham and Happisburgh), CartaRadulfi AnS3ot de Hapesburg c. 1180 Wymondham, fo. 32b(ii), vnam acram terre quam Radulfus filius Ansgoti legauit ecclesie de Habesburc. ad obitum suum 11931199 Wymondham, fo. 20a, Rogerus filius Radulfi filii Agnat de Hapesburg, Rogerus filius Radulfi Angot de Hapesburg c. 1220 Bromh, fo. 40a(iii) (Happisburgh). -Alured' filioAnggoti, abl. (witn.) c. 1155 CAcre, fo. 65a(i) (Reymerston). -Ansgot 1157-1158 PR, p. 130 (Nf or Sf, prob. Nf). -Anggot (abl.) (witn.) c. 1160 CAcre, fo. 63a(iii) (Pudding Norton). -Hangot 1166-1170 RBE, ii, p. cclxviii, no. 2 (Castle Rising). -Ansgoto presbitero de Walpol, Radulpho filio Angoti, abl. (wits.) c. 1170 Lewes, fo. 186b(i) (terram que fuit Elfere de Bertona), Ansgoto presbitero, Radulfo filio Ansgoti, abl. (wits.) c. l l70Lewes, fo. 285b(ii) (NfMarshland). -Alueredo filioAnsgot, abl. (witn.) c. 1175 Lewes, fo. 234a(i) (Heacham). -Ade filio Ansgoti, dat. 1175-1186 BodlNfCh, no. 610 (printed, Holme, no. 225), Adam filius Angoti de Gernemutha 1180-1181 P, p. 89, 1181-1182 P, p. 70 (Yarmouth), Adamf Ansgoti79 1201 P, p. 133, 1202 P,80 p. 108, 1210 P, p. 45, 1211 P,81 p. 17, Adamf Angoti 1203 P, p. 239, 1204 P, p. 236, 1205 P,

77 For the career of Os got Clapa see ASWrits, p. 569. The byname Clapa probably represents an OScand *klapi, a weak form of OWScand klapr 'a coarse and rough fellow' (see Tengvik, pp. 303-304, for full details). 78 OsgotCR. 79 Angot' CR. 80 Ansegot CR. 81 Angoti CR.

36

Åsgautr, Åsgot, Åsgut p. 245, 1206 P, p. 25, 1207 P, p. 169, 1208 P, p. 4, 1209 P, p.41 (Nf). -de catallisAngoti de Schuldham 1176--1177 P, p. 138 (Shouldham). - Carta Angoti de Kangham, Angot de Kangham c. 1185 Wymondham, fo. 45a(iv ), Robertum filiumAngod, Radulfum filium Angod 1200 FF (PRS NS, xxvii, no. 235) (Congham), Radulfus filius Angot uersus Robertum fratrem suum 1200 CRR 19 (Pleas, i, no. 2707), Carta Radulfi filii Angoti de Kangh', Radulfus filius Angoti de Kangh' c. 1200 Wymondham, fo. 50a(i), Radulfus filius Angoti ibid., fo. 54a(i), Carta Galfridi Angot de Kangham, Galfridus filius Radulfi filii Angoti de Kangham c. 1225 Wymondham fo 50a(iii) (North and South Wootton; Kangh', Kangham is Congham). -Matildam Angot 1198 FF (PRS NS, xxvii, no. 187) (Wretton). -Angotus Seinurel de Marie, Ansgotus Seinurel c. 1200-1210 Wymondham, fo. 59a(iii), Hubertusfilius Angot de Morlai 1227 JMN, p. 209 (Morley). -Johannes/ Ansgoti de Mortein 1204 P, p. 246 (Tilney, de feodo de Sculham; Mortein is Mortain (La Manche) in Normandy). - Eilric' de Raueningham uersus Henricum de Brom' de placito seruiciorum per Willelmumfilium Angod' 1205 CRR 40 (Pleas, iii, no. 1690) (Raueningham is Raveningham, Clavering H). - Radulfum filium Angod 1209 CRR 49 (Pleas, iv, no. 3980) (Nf). - Joscelinus filius Angot de Rolueston' 1209 AssRoll 558 (Pleas, iv, no. 4141) (surety in a case of novel disseisin conceming free tenements in Starston Nf, Mendham Sf and Wellingham Nf; Rolueston' can represent Rolleston Lei, Stor Rollestone W). - Petrus Angoti 1209 P, p. 33 (Nf).

OWScand Äsgautr was not common in Norway until after 1300 and was always rare in Iceland, though a few examples are known from the Landnamab6k and from late 15th century diplomas (Lind, cols. 66--68; LindS, cols. 6467). Janzen, NK VII, p. 104, takes names in -gautr in OWScand to be largely borrowed from Sweden, though he adds that in some cases they could have been formed in the parts of south-eastem Norway bordering on Götaland. In OSwed Äsgot, Äsgut is quite common in runic inscriptions and in medieval records (SRÖl, nos. 9, 23, 47; SRÖg, nos. 60, 155; SRSö, nos. 97, 122, 123,296,323,336; SRSmå, no. 126; SRVg, nos. 66, 74, 102, 103, 140; SRU, nos. 52, 84, 124, 181, 504, 610, 800; SMP, cols. 180--181). Some runic examples have initial os-, that is, runic o- for nasalized q- (SRÖl, no. 23; SRSö, nos. 122,123,296; SRVg, nos. 66, 74, 102, 103). The form esku- (SRVg, no. 140) may showvmodificationv of initial as- through analogy with such i-mutated forms as JEsbiorn and JEskil, but this is uncertain since the initial e-rune is not clear and there is also the possibility of error on the part of the runemaster (SRVg, no. 140, p. 270). Cf. also the runic form jsk11,tr (SRSö, no. 323). Runic Danish has askutr, nom. (DR, nos. 291, 392) and, showing original nasalization, qskautr, nom. (DR, no. 296), qsgutr, nom. (DR, no. 135) and qskaus, gen. (DR, no. 202). ODan Äsgot, Äsgut is also not infrequent in medieval Danish records (DgP, cols. 66--68, 1640). In England ON Äsgautr, OEScand Äsgot, Äsgut is well attested in independent use and as the first element of place-names and field names (NPN, pp. 14-16, 106; ZEN, p. 16; PNDB, pp. 165-166; Tengvik, p. 171; SPNT, p. 144; ELPN, p. 80; DBS, p. 257, s.n. Osgood etc.; SPNLY, pp. 20--22; Smart, Moneyers, pp. 229, 235, 245, 254; von Feilitzen, SCBI, xi.i, p. 12; Smart, Cnut's 37

Åsgautr, Åsgot, Åsgut York Moneyers, p. 225; von Feilitzen, PNWinton, p. 149; DEPN, p. 352, s. nn. Osgathorpe, Osgodby, Osgoodby; Lindkvist, pp. 138-139; PN NRY, pp. 104, 190, 332; PN Nth, p. 291; PN BRY, pp. 38, 261-262; PN Hrt, pp. xx, 254,255; PN Nt, p. 291; PN WRY, i, p. 211; PN WRY, ii, pp. 1, 78, 79; PN WRY, iii, p. 58; PN WRY, v, p. 92; PN WRY, vii, p. 300; SSNY, pp. 34, 101102). Cf. also the following: Osgostunstrete Edw3 AD (C 2546) (lost, in Wherstead Sf); 82 Osgote mine mey Eadulfes sune, dat., 942-951 (1. 13th) ASWills, no. 1 (Great Barton, Rougham, and Pakenham Sf; the will is that ofBishop l>eodred ofLondon); Osgot 963-992 (m. 12th) ASChR, no. 40 (Castor Nth), Osgode on Badingtune, abl., ibid. (Peterborough; Badingtune is Bainton Nth); Osgod Clape 83 1033-1035 (MS. lost) ASWrits, no. 53 (St Paul's, London), Osgod Clapa (acc.) 1044-1051, prob. 1044-1046 (e. 12th) ASWrits, no. 77 (grant at Chalkhill Mx to Westminster Abbey); Osgod (bis), Osgo/5 (acc.) 1043-1044 (12th) ASChR, no. 98 (Os god received a life lease of landat Adderbury O from the Bishop of Winchester in retum for landat Wroxhall Wt); Osgot Sveyn (witn.) 1043-1045 (1. 13th) ASWills, no. 31 (Sf); Osgote, 84 dat., 1044/5-1047, possibly 1046-1047 (1. 13th) ASWrits, no. 14 (Pakenham Sf); OsgodapudHeailea (witn.) 1052-1056 (1. 11th) KCD, no. 805 (Alstone and Teddington Gl; Heailea is unidentified); Osgot 1086 DB, fo. 104b (Colchester Ess); Osgot 1087-1098 (c. 1175-1200) Bury, p. 35, Osgot JEdrici filius ibid., p. 36 (Timworth Sf), Osgotibid., p. 37 (Troston Sf); Osgod e. 12th LVD, fös. 14b, 22b, 51b; Angod' (acc.) e. Hy2 CAcre, fo. 68b(i) (Mendham Sf); Osgod (acc.) e. Hy2 CAcre, fo. 68b(ii) (also BS, no. 355) (soke of Mendham Sf); Ansgot de Mendham, Robertus filius Ansgoti (wits.) Hy2 CAcre, fo. 68b(iii) (Mendham Sf); Galfridum filium Ansgoti Hy2 Alvingham, fo. 59b (? Cockerington L) (FMS), Galfridum filium Osgoti ibid., fo. 60b (Cockerington L) (FMS); Aceri filii Osgot, gen., 1163-1165 Alvingham, fo. 59a (Alvingham L) (FMS); Osgot filius Harald, Lecelina que fuit uxor Osgot' c. 1175 Spalding, fo. 249a (Sibsey L) (FMS); Osgotf Bald' 1190 P, p. 97, 1191 P, p. 39, 1192 P, p. 184, 1193 P, p. 18, Osgotusf Bald' 1194 P, p. 51 (Nfor Sf, prob. Sf); Henricumfilium Osegod 1199 CRR 17 (Pleas, i, no. 2255) (C); unum toftum quod fuit Osgoti c. 1200 D and C Linc, Dii 85/2, no. 32 (Normanby on the Hill L) (FMS); terram que fuitRobertifilii Osgoti c. 1200 DCh, no. 355 (Eakring Nt); liberum maritagium ipsius Christiane ex dono Osegot patris ejus, Willelmusfilius Asegod 1203 CRR 26 (Cur, ii, p. 228) (Trumpington C); totam terram que fuit Osegod' del Sabelun 1205 FF (PRS NS, xxxii, no. 447) (Rendham Sf); Osegod de Rameseia (acc.) 1208 CRR 49 (Pleas, iv, no. 3672) (Ramsey Hu); Thomamfilium Osegot et Aliciam uxorem ejus 1211 CRR 54 (Cur, vi, p. 128), Thomamfilium Osegod' 1212 CRR 52 (Cur, vi, p. 180), 1212 CRR 54 (Cur, vi, p. 233), Thome filio Osegot, dat., 1212 CRR 54 (Cur, vi, p. 214) (Aldwych Mx); Willelmus filius Osegod 1222 ElyA, fo. 152a (Tydd St Giles C), Matill' Osgod ibid., fo. 155a (Elm C); terram Thome filii Osgoti c. 1240 Alvingham, fo. 66b (Cockerington L) (FMS).

Forms with initial As- are rare, while AScand forms, in which Scand As- has been replaced by the corresponding OE Ös-, are quite frequent. Forms in Ans-,

82 83 84

38

The document in which Osgostunstrete appears is incorrectly assigned to Nf in AD. For Osgod Clapa see above, n. 77. For a discussion of the identity of this Osgot see ASWrits, p. 567.

Åsgautr, Åsgot, Åsgut

An- call for particular attention. Occasionally such forms alternate with spellings in As- and Os- which clearly refer to the same person, e.g. the forms Rogeri filii Asgoti, gen., c. 1170--1180 Wymondham, fo. 32b(ii), Rogerifilii Ansgoti, gen., 1176-1189 Wymondham, fo. 18b, 1182 Wymondham, fo. 10a, 1193-1199 Wymondham, fo. 19b, 1230 Wymondham, fo. 22b, Rogeri filii Ansgoti camerarii, gen., c. 1185-1190 Wymondham, fo. 43a, Rogero filio Ansgoti, abl. (witn.) c. 1185 Wymondham, fös. 37a(ii), 38a and Rogerus filius Angoth c. 1200 Bromh, fo. 39a(iii) all refer to the same person. The same also appears to be true of the L forms Galfridum filium Ansgoti Hy2 Alvingham, fo. 59b (FMS) and Galfridum filium Osgoti ibid., fo. 60b (FMS) as well as of the Nf forms Adamfilius Angoti de Gernemutha 1180--1182 P, Adamf Angoti 1203-1209 P, Adamf Ansgoti 1201, 1202, 1210, 1211 P andAdamf Osgot(i) 1198-1200 P. SPNLY, p. 32, commenting on similar spelling variations in the first element of Aske(ti)ll in Land Y, remarks: This variation would seem to indicate that the ultimate identity of the various forms was generally recognised by the English and AN scribes so that none of the forms could have been mere fossilised remains.

The forms in Ans- and An- are Norman in origin (SPNLY, § 11.iv,v). In the above mentioned examples of variation, Norman Ans-, An- (< Frankish Ans-) has replaced OScand As-, or, in the case of the examples from the Alvingham Cartulary and the Pipe Rolls, OE Ös-. Ekwall, DEPN, p. 352, s.n. Osgathorpe etc., commented: OScand As- comes from Ans-; this form explains early spellings such as Ansgotesbi.

There is, however, no trace of original nasalization in any of the Scand names in Äs- (< *Ansu-) found in OE sources and there is no need to regard initial Ans- in such spellings as Ansgotesbi as reflecting anything other than Norman influence. For the Norman forms in An(s)- see below. The Norman names in Ans-, An- for which an OScand origin is possible can be grouped as follows: i) Names which are clearly of OScand origin. These are: Anschetillus, Anschitillus etc. (Adigard, pp. 287-293) < OScand Asketill; Anschillus (Adigard, p. 294) < ON Askell, OEScand *Äskil; Anslek, Anslech (Adigard, pp. 294-295) < OScand Asleikr. In these names OScand As- has been replaced by the corresponding Frankish Ans-. ii) The forms An(s)ger and An(s)got. These may represent ON Asgeirr, OEScand Äsger and ON Asgautr, OEScand Äsgot, -gut, in which OScand Ashas been replaced by the Frankish Ans-. The question is, however, complicated by the existence of Frankish Ans(e)gaud and Ansegarius/Ansger. A 9th century bishop of A vranches bore the name Ansegaud and further examples from this period are known from the Drouais and the Paris region (Adigard, p. 39

Åsgautr, Åsgot, Åsgut

184). Ansegarius/Ansger is found in the 9th century in Champagne, around Paris and in what was to become Normandy and occurs as the first element of the place-names Angerville and Angervilliers (Seine-et-Oise) (Adigard, p. 187). The form Hasgerus c. 1015 (Adigard, pp. 188, 359) could represent ON Äsgeirr, OEScand Äsger and be the name of a recent arrival from the Danelaw or Scandinavia or it could be the result of scribal error. Adigard, p. 188, comments: En ce qui conceme la forme et la filiation, signalons, d'une part, le cas de Hasgerus [vers 1015], qui, par l'absence de nasalisation de l'element initial, s'oppose nettement au nom franc Ansger[i]us, de l'autre, celui d'Ansger, pere d'Anskil, dont l'origine nordique parait corroboree par le nom de son fils.

In view of the fäet that Frankish and OScand names are often found in the same family in Normandy, the case of Ansger father of Anskil must be handled with care. The name of the son cannot be taken as clear evidence that the father' s name is of OScand origin, though in view of the tendency towards Frankish names in Normandy it would have perhaps been unusual for a man with a Frankish name to give his son the name Anschillus < OEScand Äskil which is not one ofthe OScand names commonly found in Normandy. 85 Adigard, p. 186, regards it as more reasonable to postulate an OScand rather than a Frankish origin for NormanAn(s)got, hut (ibid., p. 189) takes the view that prudence is necessary in dealing with Norman An(s)ger, in view of the presence of Frankish Ansegarius/Ansger in pre-Scand Normandy and the neighbouring French provinces. A further possibility suggested by Adigard, pp. 186, 189, is that in both cases the Frankish and the OScand names have fallen together in Normandy. This is plausible in view of the known 9th century examples of Frankish Ans(e)gaud and Ansegarius/Ansger in pre-Scand Normandy and the adjoining parts of northern France. It is out of the question that Norman Ans- is merely a scribal device for the representation of nasalized OScand Äs-. If this were the case, forms with initial Ans- might be expected to occur occasionally among Norman and postConquest English examples of OScand ÄsbiQrn, Äsmundr and Äsulfr, hut these do not occur (cf. PNDB, p. 164, n. 1). The Scand personal nomenclature of Normandy was influenced by the Frankish personal nomenclature of northern France and by the AScand personal nomenclature of the Danelaw whence many of the Scand settlers in Normandy had come. Hence OScand ÄsbiQrn, Äsmundr and Äsulfr always occur in Normandy with initial Ingvaeonic Ös85 Adigard, p. 294, has only one further pre-1066 exarnple of Anschillus and the n~e is absent from Norman place-narnes. The full form Anschetillus, Anschitillus etc. < OScand Asketill is one of the most common OScand personal narnes in Normandy (Adigard, pp. 83-84, 287-293, 382-

383).

40

Åsgautr, Åsgot, Åsgut which reflects AScand Ösbeorn, OE or AScand Ösmund and AScand Ösulf. 86 It is noteworthy that the Continental Germanic names Ansemund and Ansulf have had no influence here. In the Carolingian and post-Carolingian periods Ansemund and Ansulf are characteristic of the formerly Gothic areas of southem Gaul and the Spanish March (Modet, p. 39; Bergh, pp. 15-16; Kremer, pp. 58-59). Ansemund is also frequent in the Spanish Christian kingdoms in the 9th and 10th centuries and is here clearly of Visigothic origin (Piel-Kremer, p. 76; cf. also Kremer, p. 59). Ansemund and Ansulf do not occur in records from pre-Scand N ormandy or its environs and a West Frankish *Ans(e)bern is not on record. It seems likely that Ansemund, Ansulf and *Ans(e )bern are not to be counted among the Frankish names which the Scand settlers would have met in Normandy. On the other hand, West Frankish Ans(e)gaud and Ansegarius/Ansger are well evidenced in pre-Scand Normandy and the neighbouring parts of Carolingian Neustria and they were ready at hand to replace or modify OScand Ä.sgautr and Ä.sgeirr. The replacement of OScand Ä.s- by Frankish Ans- in such OScand names as Ä.sketill and Ä.sleikr which had no West Frankish counterpart is to be explained as the result of analogy with other Norman and Frankish names in Ans-. For Norman An(s)fred(us), -frid(us), which is of West Frankish origin, see above, s.n. *Asjrf()ör. Cf. also Norman Answit < ODan (run.) Äswipr, OSwed Äsvidh or Frankish Answid. 87 The development Ans- > An- in the forms Angot and Anger is the result of the 1 lth-century OFr loss of s before voiced consonants (PNDB, § 112). The loss of s before voiceless consonants in OFr did not become general until the 13th century, but there are indications that the process must have begun earlier (PNDB, § 112; SPNLY, § 11). This is proved by the appearance of Anketillus, Anketinus etc. in 12th-century records. 88 The second element, ON -gautr, OEScand -got, -gut appears as -god or

86 See above, p. 31. With reference to Osulf, the existence of OE Öswulf should be borne in mind, though in the Danelaw Osulf is probably often an Anglicizied form of ON Asulfr, ODan, OSwedÄsulf(cf. NPN, pp. 21-22; PNDB, p. 341; SPNLY, p. 35; and below s.n. Ä.sulfr). For the isolated Norman example of Osgot see above, n. 67. 87 Adigard, p. 191, took the view that the fäet that the form Answit is not Latinized would seem to speak for a Scand origin. However, a further Norman example occurs c. 11()() in the Latinized form Answidus (Adigard, p. 191 and n. 14), so it is unsafe to make any deductions from a single form lacking the Latin case ending. Frankish Answid is known from 9th and 11th century Lorraine but seems to have been very limited in its distribution (Adigard, p. 191), and it may well have been unfamiliar to the Norman scribe who, probably for this reason, left out the Latin case ending. Both the Frankish and the Scand name are rare but perhaps it is safer to take the Frankish name to be the etymon here. 88 For examples see SPNLY, pp. 25-32, and below, pp. 50---52. Anchitil occurs in 1086 in DB in L (SPNLY, p. 25).

41

Åsgautr, Åsgot, Åsgut -got in English sources. In OEScand unstressed au was monophthongized to o (Brpndum-Nielsen, § 108; Noreen, AschwedGramm, § 81.2b; PNDB, § 37). H. Andersen, NoB, xxiv (1936), p. 88, regarded this as having taken place in ODan at the latest by the beginning of the 10th century and probably earlier. ODan has two runic examples of Äsgautr/Äsgot in which the diphthongized form -gautr is preserved. The first of these, qskaus, gen. (DR, no. 202) belongs to DR Per. 2.2, and is an inscription of the "Fpr-Jelling" type. The "FprJelling" type of runic inscription belongs to a period around 900 (DR, col. 1023). DR, no. 202, commemorates one ElefR, who was the son of the abovementioned Äsgautr. This Äsgautr probably lived, therefore, in the third quarter of the 9th century, and, thus, the survival of diphthongized -gautr in DR, no. 202, would not be unusual. More difficult to explain is the qskautr of DR, no. 296, for this inscription, from Skåne, though also belonging to DR Per. 2.2, is of the "Efter-Jelling" type, which belongs to the period between c. 1000 and c. 1050 (DR, col. 1024). This qskautr set up the rune stone in memory of his brother ~rra, who was hemjxegi, i.e. follower, of Toki Gormsson, a Danish chieftain who had fought in Sweden at the beginning of the 11th century. 89 The form of the second element, runic -kautr (for OScand -gautr) may be the result of Swedish influence, for diphthongized -kautr for OScand -gautr appears occasionally in Swedish runic inscriptions, e.g. askautr (SRSö, no. 97), oskautrR (SRSö, no. 296), askaut, acc. (SRSö, no. 336; SRU, nos. 52, 124) for OSwed (run.) Äsgautr, ]Jorkautr (SRSö, no. 111), ]Jurkaut, acc (SRU, no. 646) for OSwed (run.) Porgautr, and uikautrr (SRSö, no. 285), uikaut, acc. (SRU, no. 706) for OSwed (run.) Vfgautr. On the other hand, it may be theoretically possible that we are concemed with an East Danish form in DR. no. 296. OScand au survives in runic inscriptions of the post-Viking period on Bornholm, though it should be noted that the runic inscriptions of Bornholm show strong Swedish influence (see DR, cols. 785-786). It is, however, clear that the monophthongization of unstressed au > o took place earliest and most completely in ODan, and, therefore, -god, -got as the second element of OScand personal names in England should be taken to represent ODan -got. In Asgout, a form found among the names of Cnut's York moneyers (Smart, Cnut's York moneyers, p. 225), the original diphthong is preserved in the second element. The final -d in the form -god is difficult to explain. The form -god for ODan -got is known in England prior to 1066, and in these pre-Conquest forms it is unlikely that final-dis an AN inverted spelling for [t].90 The influ-

89

90

See L. Musset, Introduction ii la runologie, pp. 428-429. In pre-Conquest forms which are only preserved in post-Conquest copies the possibility of

inverted spellings cannot be excluded. Such inverted spellings are, however, out of the question in original pre-Conquest documents and in pre-Conquest moneyers' names.

42

Åsgeirr, Esger, Åsger, !Esger ence of the Continental name element -gaudus, -godus, which is found before the Conquest in such forms as Hergod < West Frankish Harigaudus, the name of one of the Confessor' s Oxford moneyers (SCBI, xviii, no. 1152), and Winegod < West Frankish Winegaudus, 91 which occurs as the name of a moneyer of Cnut from Warminster (SCBI, xv, no. 4212), possibly contributed to the appearance of forms in -god for ODan -got. Björkman, ESt, li (1917-1918), p. 176, comments: Der häufige nordische name Porgautr, -gotr wurde in Frankreich (zunächst in der Normandie) und England in anschluB an die vielen aus dem Kontinentalgermanischen stammenden namen, in denen -god mit -got wechselte, in Purgod, Turgod umgewandelt. Besonders nahe lag es, bei Latinisierung aus dem namen ein T(h)urgodus zu machen.

Björkman's remarks are misleading as regards Normandy. The normal Norman form for ON Porgautr etc. is Turgotus, though one example of Toresgaudi,92 gen., which clearly has West Frankish -gaudus as second element, is known (Adigard, pp. 158, 320). It might be the case, as was suggested by Zachrisson, ESt, lii (1918-1919), pp. 199, 200, that the West Frankish name element -god was identified with the OE adj. göd 'good' in England, and was then substituted for ODan -got in Osgot to give Osgod. OE Ösgeat, which is cognate with ON Äsgautr, OEScand Äsgot, -gut, appears in Nf in the following forms: terram Ossiat Squen c. 1190 Lewes, fo. 226b(ii) (Heacham); terram Osiat Jilii Ailwardi c. 1220 Wals, fo. 48a(i) (Holkham).

ON Åsgeirr, ODan Esger, OSwed Åsger, !Esger a) (i)Asgarsthorp 1279 Ipm,Asgersthorp 1315 CI,Asgeresthorp 1371 Ipm,Asgasthorp, Asgazthorp 1314 Ipm, 93 Asgatesthorp 1317 HMC VarColl, vii, p. 154, no. 4,Algerthorp 1328 Banco (Algarsthorpe in Great Melton, Humbleyard H). -Asgereswong e. 13th Binham, fo. 102b (f. n. in Great Snoring). -Asgereshil, Asgereshil' 1202 FF (PRS NS, xxvii, no. 400) (f. n. in Carleton'). -Asgeristi c. 1220 Der, fös. 55b(i), 60b(ii) (f. n. in Barton Bendish). - Askereswong c. 1230--1240 Der, fo. 41a, Asgeriswong ibid., fo. 44b(iv), Asgereswong ibid., fo. 46a(i) (f. n. in Boughton). (ii) Eskeresthorp c. 1050 (1. 13th) ASWills, no. 33 (Algarsthorpe in Great Melton, Humbleyard H). (iv) Angerestoft Hy3 NorwCP Il (f. n. in Wicklewood) (OKS). b) (i) Colo liber homo Asgeri Stalre TRE DB, fo. 149b (Shelfanger). -Asgarus. I. fiber homo TRE DB, fo. 248 (Skeyton). -Nicholausfilius Asgari (witn.) 1153-1168 (1. 13th)

91 92

93

For West Frankish names in -gaudus, -godus see Longnon, i, pp. 316--317; Mansion, p. 22. This form appears in a document of c. 1050 contained in the cartulary of St Peter of Chartres. For the first five forms see also Lindkvist, p. 23.

43

Åsgeirr, Esger, Äsger, LEsger Holme, no. 128 (Burwood and Ludham), Transcriptum Nicholai filij Asgari, Nicholao nepoti meo filio Asgari, dat.,1153-1168 Holme, no. 181 (Thume; the charter isa grant by Abbot William Il of Holme to his nephew Nicholas son of Asgar), Walterus. Hugo. Nicholaus. filii Asgeri c. 1163-1166 BodlNfCh, no. 605 (printed, Holme, no. 164) (Hasardes holm near Potter Heigham). - Transcriptum Henrici filij Asgari, Henrico filioAsgarij, dat., 1153-1168 Holme, no. 175 (Homing). -RadulfumfiliumAsegar' 1202 FF (PRS NS, xxvii, no. 330) (Long Stratton and Torp, i.e. prob. Momingthorpe). - acram unam quam tenuit Asger e. 13th HMC VarColl, vii, p. 155, no. 8 (Bacton). terram Willelmi Asger' 1237-1251 Holme, fos. 92a(iii), 108b(iv) (Ashby in Flegg). (ii) Hugoni Coco filio videlicet Esgarij, dat., 1175-1186 Holme, no. 213 (Hoveton). (iii) Philippo filio Osgerii, abl. (witn.) c. 1170 CAcre, fo. 26b (grant by Willelmusfilius Gurwandi to the monks of Castle Acre of landat Becsti in Kempston and at Brictmerestoft in Heringeshae). (iv) Ansger. I. fiber homo Goduini TRE DB, fo. 175 (Surlingham). - quam tenuit Ansgerus. subAngero Stalra TRE DB, fo. 248 (Ryston). -Ansgerus 1165-1166 P, p. 29, 1166--1167 P, p. 29 (North Witton). - Godefridus nepos Angeri 1165-1166 P, p. 30 (Nf). -Ailm(ar) nepos Angeri 1165-1166 P, p. 30, Ailm(ar) nepos Ansgeri 1166--1167 P, p. 30, Ailmer nepos Ansgeri 1167-1168 P, p. 20, Ailm(aru)s nepos Ansgerii 11681169 P, p. 98, Ailmerus nepos Ansgeri 1169-1170 P, p. 6, 1170---1171 P, p. 6,Ailmarus nepos Ansgeri 1171-1172 P, p. 27, 1172-1173 P, p. 122, 1174--1175 P, p. 113, 11751176 P, p. 62, Ailmarus nepos Ansgerii 1173-1174 P, p. 41 (Nf). -Rod(bertus)filius Ans(geri) de Wimundham 1166--1170 RBE, p. cclxxiii, no. 25 (Wymondham). -Ansger 1167-1168 P, p. 32, 1168-1169 P, p. 104, 1169-1170 P, p. 9 (Nfor Sf). -Ailmarus filius Ansgeri 1174--1175 P, p. 119, 1175-1176 P, p. 66, 1176--1177 P, p. 128, 11771178 P,94 p. 23, Ailmarus filius Angeri 1178-1179 P, p. 4, 1179-1180 P, p. 17 (Nf or Sf). - in manu Rogeri filii Ansgoti camerarii et Anger nepotis eius, Angerus c. 11851190 Wymondham, fo. 43a (Wicklewood). -Angerus cocus c. 1190---1200 Binham, fo. Sa (printed, SSMEA, Appendix Il, no. 3, p. 268). -Angerus de Middelton' 1202 P, p. 113, 1203 P, p. 242 (Middleton), 1214 CRR 58 (Cur, vii, p. 198) (case concerning land in Walton and Wiggenhall), Anger' de Middeltona (acc.), prefato Angero, dat., predictus Angerus 1219 FF, no. 37 (Middleton). - Carta Anger de Risinges, Angerus de Risinges c. 1205 Co:iford Il, fo. 39a(iv) (Houghton near West Rudham; Risinges is probably Castle Rising). -Angerus de Stowe (surety) 1209 AssRoll 558 (Pleas, iv, no. 4378) (case of novel disseisin conceming a free tenement in Wood Dalling). - W alterus f Angeri 1209 P, p. 33 (Nf). -Robertusfilius Angeri 1214 CRR 58 (Cur, vii, p. 196), 1214 CRR 112 (Cur, vii, p. 261), Roberto filio Angeri, abl. (witn.) c. 1215-1225 Wymondham, fos. 40b(ii), 43a(ii) (Wicklewood). -Angerus Spalle 1219 CRR 71 (Cur, viii, p. 54) (Nf). - terras Angeri prepositi c. 1230---1250 Wals, fo. 23a(i) (Little Walsingham). - mesuagium Angeri Gris c. 1235 Der, fo. 220a(i) (Wiggenhall). - dominum Angerum abbatem c. 1235 Der, fo. 20b(iii) (Dereham and Wereham), Angero abbate ecclesie Sancte Marie de D., dat., ibid., fo. 150b (Tilney), dominum Angerum abbatem de D. ibid., fo. 162b(i) (Tilney), Angerus dei paciencia abbas de D. ibid., fos. 196b(i), 207b (Wiggenhall), 227b(i) (parish of St Peter ofLynn, Saddlebow), Angerum abbatem de D. 1245 Der, fo. 238a(i) (North and South Clenchwarton, Saddlebow, Wiggenhall, Lynn), 1250Der, fo. 184b (Tilney). -cruftamRogerifilii Angeri coci c. 1250 Wals, fo. 13a(iv) (Krimplingescroft on the royal road between Little Walsingham and Wighton), homag94

44

Almarus filius Ansgeri CR.

Åsgeirr, Esger, Åsger, JEsger ium Rogeri filii Angeri coci ibid., fo. 16a(ii), Rogerus filius Angeri coci ibid., fo. 16b (Little Walsingham).

ON Asgeirr is known both from Norway, where it is fairly common, and from Iceland (Lind, cols. 68--69; LindS, cols. 67-70). Runic Danish has qskaiR, nom. or acc. (DR, no. 65), qski, acc. (DR, no. 96), askis, gen. (DR, no. 91), and, with OEScand i-mutation, Lat. Esgervs, nom. (DR, no. 224), eskiR, acc. (DR, no. 400), isgi, acc. (DR, no. 135). The unmutated form with initial Äsoccurs only once in medieval Danish sources, but the i-mutated variant, which appears in the Latinized form Esgerus and as the short form Esgi, is very common (DgP, cols. 265-270; NK VII, pp. 192, 208). The unmutated form is found in Swedish runic inscriptions (SRVg, no. 154; SRU, nos. 350,473,592, 698, 759, 775, 844, 845, 1177). Some of these forms have initial runic o for nasalized 4 (SRVg, no. 154; SRU, nos. 473, 845). In medieval Swedish sources both the unmutated form with initial Äs- and the i-mutated variant JEsger are found, though the former is not common (Lundgren-Brate, pp. 18, 312; SvP, p. 26; SMP, cols. 179-180). The short form !Esge is also known from medieval Swedish sources (SvP, p. 26). In England the name appears as Asger, Asgar, Esger, Esgar and is fairly frequent (NPN, pp. 13-14; Lindkvist, p. 23; ZEN, pp. 15, 32; DEPN, p. 14, s.n. Asgarby; PNDB, pp. 166-167; ELPN, p. 75; PN Cu, p. 55; SPNLY, pp. 22-24). The Liber Vitae of Thomey has the form As3rer 11th (SPNT, p. 142). Forms with initial Es- represent the OEScand variant with i-mutation (PNDB, p. 164; SPNLY, § 11.ii). The place-name form Eskeresthorp c. 1050 (1. 13th) contains this OEScand variant with i-mutation as first element, though in subsequent forms of this place-name this is replaced by the unmutated form with initial As- (see above, p. 43, for forms). 95 In this connection it should be remarked that in DB the name of the AScand notable Esgar the Staller is spelt variously with initial As-, Es-, Ans- and An-, with initial !Es- occuring in IE (C) and ExonDB (PNDB, pp. 166-167). Fora discussion of similar variation in the spelling of OScand As- in certain examples of Asgautr and Asketill found in English sources see above, pp. 38-39. The form Osgerii, gen., may be an Anglicized form of the OScand name in which OE Ös- has replaced OScand As- or it may represent OE Ösgär. A similar case might be provided by the first element of the place-name Osgathorpe Ho WRY, though here one or two early spellings suggest rather Ösgot < Asgautr as etymon (PN WRY, i, p. 211). Forms with initialAn(s)- are the result ofNorman influence. In some cases 95 The subsequent development of Asgersthorp etc. to the modern Algarsthorpe is a result of the substitution of ME Algar for Asger. Fora discussion of the etymology of Algar see above, pp. 1112, 13. The forms Asgasthorp, Asgazthorp, Asgatesthorp show the influence of ON Äsgautr etc.

45

Åsi, Åsi, Åse they merely reflect the replacement of OScand Äs- by Frankish Ans-, but in other cases the form An(s)ger can represent the Frankish personal name Ansger. For a discussion of the forms Ansger, Anger and the associated problems see above, pp. 39-41). Many of the bearers of the name An(s)ger in English sources must have been Normans or men of Norman descent, though as the different forms of the name of Esgar the Staller mentioned above show, the Normanization of the Scand personal name by Norman scribes was not unknown. Anger sometimes occurs as a byname in English sources of the ME period. Here the Norman personal name form Anger might be possible but ME anger 'grief' < ON angr is equally likely (see s.n. Aki, n. 9). For this reason the entries above only relate to Anger in its function as a personal name and not as a byname. Fora discussion of the second element -gar, -ger etc. see s.n. Alfgeirr and n. 22. Sometimes in English sources there is altemation between -gar and -ger forms in the name of the same person, e.g. terram Willelmi filii Asgari, Willelmo filio Asgeri, abl. (witn.) e. Hy3 AD(L 3097) (Somercotes L) (FMS).

ON Åsi, ODan Åsi, OSwed Åse a) (i) Asecroft c. 1240 Bromh, fo. 6la(i) (f. n. in Wood Norton). -Aselund 1256--1268 Holme, fo. 13la(ii) (f. n. in Ludham). b) (i) De catallis Rogerifilii Ase 1167-1168 P, p. 30 (Nf; ?Eynsford H). (ii) Willelmus filius Osi 1222 ElyA, fo. 164b (Terrington).

Asi, a short form of names in As- (NK VII, p. 209), is not common. In OWScand the name was borne by the father of one of the original settlers in Iceland (Lind, cols. 72-73), and in OEScand a few examples of the name occur in medieval documents from Denmark and Sweden (DgP, col. 68; SMP, col. 178). Further examples are known from Swedish runic inscriptions (SRSö, no. 7; SRU, nos. 17, 846, 1069) and ODan Äsi forms the first element of the Danish place-name Aaserup (DS XI, p. 103).96 An ODan &si, Esi, a short form of names in &s-, Es- (< Äs- by i-mutation), is also found (DgP, col. 271; DR, no. 73; DS XIV, p. 257).97 Both Asi and &si, Esi are known from English sources, though neither is common (NPN, p. 16; ZEN, p. 16; Lindkvist, p. 170; PNDB, p. 167; SPNLY,

96 It may also be the first element of the place-name Aastrup in Jutland, though here the ODan substantive as m. 'ridge' is also possible (DS IV, p. 181). " " 97 For further Danish place-names which might contain /Esi, Esi see DS XVI, p. 115, and the references cited there.

46

Åsi, Åsi, Åse p. 24; von Feilitzen, JEPN, iv (1971-1972), p. 67). ON Ä.si forms the first element of Easedale We (PN We, i, p. 199), anda field name Asecroft c. 1210 is found in Nt (PN Nt, p. 278). A certain reserve is necessary in dealing with Ä.si and !Esi, Esi as the first elements of place-names and field names in England, since it is equally possible that these names contain ON Ä.sa, OEScand Äsa and ODan !Esa, Esa, the feminine counterparts of Ä.si and !Esi, Esi (cf. Lindkvist, p. 170). There is nothing in the forms of the Nf field names given above or in the forms of the L and Y place-names and field names given by SPNLY, p. 24, which allows a definite decision in favour of one or the other of these two altematives. The same is equally true of the forms from Nt and We. Commenting on problems of this nature, Stenton remarked (TRHS 4th Series, xxv (1943), p. 12, n. 1): It may be noted that in regard to Scandinavian, as to English names, it is often hard to distinguish between masculine and feminine forms. There is nothing, for example, in Touetun, the Domesday spelling of Towton in Y orkshire, to show whether the first element is the masculine Toft or the feminine T6fa. Ambiguities like this are frequent.

Von Feilitzen, JEPN, iv (1971-1972), p. 67 took&se, which oc~urs !n an OE inscription from York, to represent either the masculine ODan &si, Esi or the feminine !Esa, Esa. Reaney, DBS, p. 13, s.n. Ashcroft, Ascroft, has the form Margar' de Asecroft 1275 RH (Nt) with the etymology 'Dweller at a croft with an ash-tree.' There is, however, no evidence to link the medieval field name Asecroft with the modem sumames Ashcroft and Ascroft. The field name form Asecroft calls for particular discussion. OE [f] is represented by the spellings sh, ssh, sch, se, ss, s, and, in exceptional cases, ch, in ME (Luick, §§ 691,692; Jordan, § 181; Kristensson, SMED, pp. 38,202; Selten, pp. 143-144). The spellings in sand ss are AN in origin and are the result of the AN substitution of [s] for OE [f] (PNDB, § 125; Selten, p. 144). In DB the spelling se for OE [f] also stands for the AN substitution of [s] for [f], and is nota survival of the traditional OE spelling (PNDB, § 125). The same is true of examples from the ME period. The AN spelling s representing the replacement of OE [f] by [s] is rare and does not usually appear in intervocalic positions. 98 In this connection, it is irrelevant whether the second of the two vowels is an etymological vowel or an AN inorganic vowel. In intervocalic positions the AN replacement of OE [f] by [s] is usually represented by the spelling ss, e.g. Assemanus liber homo TRE DB, fo. 436b (Higham, Samford H, St) for OE &scmann (PNDB, p.

98 Cf. however, the forms Rogeri Ayseman, gen., Rogeri filii Aiseman de Snetesham, gen., c. 1210 Der, fo. 127b(i). The etymon of Ayseman, Aiseman is OE k:scmann. The forms Ays-, Aisfor OE/Esc- are AFr in origin, cf. Ais(s)eford(a) 1086 DB PN D, p. 547 (see also EPN I, p. 4, s. v. trsc).

47

Åskell, Åsketill, Eskil, Åskel, lEskil 182), Adel' filiam Tusseman c. 1155 CAcre, fo. 50a(i) (Creake) for OE *Tuscmann. The normal ME spelling of a compound of OE cesc and OE croft is represented by the form Asshecroft 1337 PN Gl, ii, p. 247 (Ashcroft Fm Gl). In view of these considerations, it is more likely that the first element of Asecroft is OScand Äsi, masc., or OScandÄ.sa, fem., rather than the OE tree name cesc. The form Osi has been Anglicized through the replacement of OScand Asby OE Ös-. It parallels exactly the fem. form Osa, Ose, given above, s.n. Äsa.

ON Åskell, Åsketill, ODan Eskil, OSwed Åskel, iEskil a) (i) Ascheteleshirne c. 1190-1205 HarlCh 45 F 15 (printed, SSMEA, Appendix I, no. 2, where it is dated 1. 12th) (f. n. in Hunworth). -Asketelholm 13241pm (f. n. in Sutton). -Askitels 1573-1582 OKS (tenement in Gimingham). (ii) Askillescro[t c. 1250 Der, fös. 142b, 153b(ii), 155a, Askilescroft Edwl Rental (OKS) (Sandred, OUA, 1972, p. 44) (f. n. in Tilney). (iv) Ankillescrof c. 1260 Binham, fö. 12b(iii) (f. n. in Binham). (v) Astineswong, Astingeswong 14th Stake (f. n. in Crimplesham) (OKS). b) (i) manredam Askitelis 1067-1075 Holme, no. 58 (South Walsham). - Asketelus 1100-1135 (13th) Rams, iii, p. 267 (Ringstead and Holme next the Sea). -Asketillo de Redham et Willelmo filio eius, abl. (witn.) 1101-1125 Holme, no. 121 (grant in Filby; Redham is Reedham). -Ascetel (priest of Walsingham) (abl.) (witn.) 1101-1107 Binham, fö. lb (grant by Peter de Valoignes to Binham Priory). - Willelmusfilius Asketilli (witn.) 1134-1140 Holme, no. 131 (Tibenham), 1134-1140 BodlNfCh, no. 604 (printed, Holme, no. 135) (Little Melton). -Aschetillo, abl. (witn.) c. 1140 CAcre, fö. 9a(i) (Suthcroft and Sidholm in Castle Acre). -De Aschetillo filio Godwini, Aschetillum filium Godwini c. 1155 CAcre, fo. 49b(iii), Aschetill' filio God(wini), abl., Aschetillum filium God(wini) clerici ibid., fö. 50a(i), Aschetill' filium God(wini) clerici ibid., fös. 50a(ii), 50a(iii) (Creake). -Aschetil 1161-1177 (13th) Rams, iii, p. 262,Asketelus ibid., p. 263 (Brancaster, Burnham Deepdale and Burnham). -Asketelus 1161-1177 (13th) Rams, iii, p. 268 (Ringstead and Holme next the Sea). - Ricardo filio Turolf 9 et Aschetillo 100 fratre eius. abl. (witn.) c. 1165 (15th) NorwCPCh, no. 124 (Lynn and Mintlyn). 101 -Aschetillus filius Roberti de Soun (witn.) c. 1170 CAcre, fo. 91a(i) (Gateley, Pentney, Massingham and land held by Rainaldus de Waltun). -Asketill' de Tikewell' 1174-1175 P, p. 123 (Titchwell). -Aschetillo coco, abl. (witn.) c. 1175-1180 CAcre, fö. 38b(i) (Congham). - de Asketello filio Sigar' 1176--1177 P, p. 130 (Nf or Sf). Radulfus de Karletunrode et Asketillus filius eius c. 1180-1185 Wymondham, fö. 25b(ii), Carta ipsius Asketilli ibid., fos. 26a, 26a(i), Radulfus de Karlet' Rod et Asketillus filius eius ibid., fö. 26a(i) (Stanfield; Karletunrode is Carleton Rode). -Asketel /oud c.

99 Var: Thurold (e. 14th); Thurolf (e. 14th). The form Thurold shows confusion with ON P6raldr etc. 100 Var: Astente/1 (e. 14th); Archetill (e. 14th). The form Archetill shows confusion with ON Arnke/1, ODan Arnketil etc. Perhaps the scribe misread the s of his source as rand so copied it. 101 On the dating of this document see above, n. 34.

48

Åskell, Åsketill, Eskil, Åskel, !Eskil 1190---1200 Binham, fo. 5a (printed, SSMEA, Appendix Il, no. 3, p. 269) (Binham). duas acras quas Aschetel tenuit iuxta bruarium de Stody c. 1190---1205 HarlCh 45 F 15 (printed, SSMEA, Appendix I, no. 2, where it is dated to the late 12th century) (Hunworth). - Galfridus f Asketill 1193 P, p. 26, Galfridus f Asketill' 1194 P, p. 60 (Nf or Sf, prob. Nf). - WillelmumfiliumAsketel 1198 FF (PRS NS, xxvii, no. 63) (Hackford). - Ulffilium Asketel 1198 FF (PRS NS, xxvii, no. 89) (Berham). - Godwinusfilius Asketelli 1198 AssRoll 559 (Pleas, ii, no. 28) (Launditch H), Goduinusf Asketill' 1199 P, p. 281, Godwinusf Asketill' 1200 P, p. 142 (Nf). -Herueiusf Asketill' 1199 P, p. 289, 1200 P, p. 145 (Nf or Sf). - Willelmus filius Asketill' 1199 CRR 17 (Pleas, i, no. 2356) (Nf). - Aschetello medico, abl. (witn.) c. 1200 Bromh, fo. 5la(ii) (Yarmouth). Aschetillus Carpentarius 1201 (13th) Rams, iii, p. 296 (Welles, i.e. Outwell). - Willelmumfilium Asketel 1202 FF (PRS NS, xxvii, no. 398) (Creake). -Asketillus Cole 1203 P, p. 247 (Nf or Sf, prob. Nf). -De Ricardo f Asketill' 1203 AssRoll 817 (NthRS, v, no. 780) (Nf), Ricardusf Asketil 1203 AssRoll 817 (NthRS, v, no. 834) (Gressenhall). Alexander Asketell' et Emma uxor ejus 1204 CRR 33 (Cur, iii, p. 217), Alexandra Asketell', dat., 1205 CRR 38 (Cur, iii, p. 229), mesuagium Alexandri Aschetel c. 1205 Wals, fo. 94b, mesuagium Alexandr' Asketel c. 1205-1210 Wals, fo. 94b(i), Alexander filius Asketelli de Gernemue 1209 AssRoll 558 (Pleas, iv, no. 4460) (Yarmouth; in Pleas, iv, no. 4460, Alexander son of Asketel of Y armouth was a surety in a plea of mort d' ancestor conceming 20 acres of land with appurtenances in Haddiscoe ). - mesuagium Ricardifilii Asketel c. 1215-1220 Wals, fo. 7la(i) (Syderstone). -terramRogerifilii Asketel c. 1225 Wals, fo. 71a(ii) (Syderstone). -Asketellum de Halle 1225 CRR 88 (Cur, xii, p. 194) (Weston). - de feodo Asketilli 1230 Wymondham, fo. 22b (Nf). 102 - terram Asketelli palmerii 1247 Langley, fo. l9a (Erodeswelle near Loddon). -Asketel, terr' Asketel, Willelmi Hasketel, gen.,? m. 13th (l. 13th -e. 14th) Wals, fo. 69a(i) (Bumham), 103 inter terras Augustini de Binham et heredum Willelmi Asketel de Burnham c. 1285 W als, fo. 51 b (Holkham; Burnham is Burnham). (ii) Eljledam uxorem Askilli 1153-1168 Holme, no. 190, 1175-1186 Holme, no. 257 (Potter Heigham). -Aschillus filius Brunilde c. 1180 CAcre, fo. 82a(i) (Wiggenhall). Alanus filius Askil, homagium Alani fil ii Askil, Alano filio Askil, abl., 1186-1211 ( 13th) BuryS, no. 139 (? Setchey). - homagia Askil Heir et Ailed c. 1205 Der, fo. 134a(ii), homagum (sic) Askil et 104 Heir et Ailed ibid., fo. 134b(i) (Tilney), Carta Askill' Heir de Tiln' de hornagio Ade Cholle de Len quod cum tenemento suo in Northlen, Asket Heir de Tiln' c. 1210 Der, fo. 243a(ii) (grant by Askil to Dereham of the homage of Adam Cholle of Lynn together with the tenement which Adam held of Askil in North Lynn and the annual dues which he rendered to Askil), CartaAskill' i (sic)filii Alberti de Tiln', Askillusfilius Alberti de Tilneia 105 c. 1225 Der, fo. 140a(ii)(Der, fo. 140a(ii)-fo. 140b,

102 Cf. de feodo Ansketilli 1182 Wymondham, fo. 10b (Nt), which refers to the same person as the present form de feodo Asketilli, and which is grouped below with the other forms in An(s)ketill(us), An(s)chetill(us) etc. 103 This document is in a hand which is different from that of the main body of the cartulary. 104 A mark under the Tironian sign here indicates that it should be deleted and the Askil Heir and notAskil et Heir is the correct reading. The byname Heir, Eyr is from ME eir, eyr, AFr heyr, OFr eir, heir 'heir' (see DBS, p. 17, s. n. Ayer). 105 In view of the fäet that Askil bears an OScand name and that the name of his sister, Eljleda (see below), is OE in origin (OE JEljJl;ed), it is clear that Askil's family was of native AScand origin. For this reason it is likely that their father's name, Albertus, is OE (Angl) Aldberht rather than OFr Albert (< Adalbert), though on förmal grounds the latter cannot be entirely excluded.

49

Åskell, Åsketill, Eskil, Äskel, /Eskil is a grant by Askil to Dereham consisting of land at Westneweland in Tilney, the homage of Bricius son of Roger son of Leuuinus of Wiggenhall 106 together with the tenement which he held of Askil in Islington and the homage of William Cholle in North Lynn together with the tenement which he held of Askil in North Lynn), Carta Askill' Eyr de Tiln' de hornagio Willelmi filii Ade Cholle de Lin', Askyl Heir de Tiln' ibid., fo. 164b(ii) (grant by Askil to Dereham of the homage of William son of Adam Cholle of Lynn together with the tenement which William held of Askil in North Lynn and the annual dues which he rendered to Askil, 107 terram Askil Heir c. 1225-1230 Der, fo. 18lb(i), Carta Willelmi filii Godefridi de hornagio et seruicio et redditibus Askillifilii Alberti de Tiln' et que legauit nobis, omnes terras quas Askillus filius Alberti de Tiln' dedit et legauit, predictus Askillus, Elfleda eiusdem soror Askilli c. 1230-1235 Der, fo. 149b(ii), Carta Elfleda (sic) filie Alberti de Tiln' de medietate tofte et crofe (sic) que Askill' fratris sui, medietatem totius terre tofte et crofte que fuerunt Askilli fratris mei c. 1230-1240 Der, fo. 149b(i), de terra que fuit Askilli fratris sui in Tiln' ibid., fo. 149b(iii), Askilli fratris mei, gen., ibid., fo. 150a, Askillus Eyr ibid., fo. 164b(iii), Askil frater meus, carte predicte Askill' fratris mei ibid., fo. 165a (Der, fos. 149b(iii) - 150a and 164b(iii) - 165a are charters of Elfleda filia Alberti de Tiln'), a terra que fuit Askil Heire c. 1235 Der, fo. 145b(i), in uilla de Tiln' in crofta que fuit quondamAskilli Heyr c. 1240 Der, fo. 158a(i), Carta Hawise filie Ricardi Wderoue de Tiln' de medietate latitudinis totius crofte que fuit Askill'i (sic), crofte que fuit Askilli auunculi mei ibid., fo. 158a(ii), Carta Hawise filie Ricardi Woderoue de Tiln' de homagiis et redditibus quas Askill' nobis dedit, Askil auunculus meus, carte predicti Askilli auunculi mei ibid., fo. 165a(i) (Tilney). (iii) Oschetel prepositus regis TRE DB, fo. 280 (Fomcett). - Osketel (liber homo) TRE, 1086 DB, fo. 273 (Moulton St Michael). -Reineriusfilius Osketeli 1161-1177 (13th) Rams, iii, p. 288 (Hilgay and Snorehill), Reinerius filius Osketelli c. 1180 CAcre, fo. 92b(i) (Hilgay). -Oschetel Cniht 1165-1166 P, p. 34 (Nf or Sf). -de Osketell' Muscher 1176--1177 P, p. 130 (Nf or Sf). - de Osketell' venatore 1176--1177 P, p. 130 (Nf), Oschetel venatore, abl. (witn.) c. 1180-1185 Binham, fo. 49b(i) (Bamey). -In Branteslund terram Bonde. et Oschetel 1176--1189 Wymondham, fo. 18b (Wymondham). -dimidiam acram que fuit Osketelli pelliparii c. 1180 Wymondham, fo. 95b(ii), dimidia acra que fuit Osketelli Hyot pellificis ibid., fo. 96a(i) (Wicklewood). - pro fuga Osketel 1197 P, p. 244, 1198 P, p. 88 (Nf or Sf, prob. Nf). - Willelmum et Osbertumfil' Osketel 1198 FF (PRS NS, xxvii, no. 34) (Karboistorp). - Osketel Wine 1199 P, p. 288 (Nf or Sf). - Osketel de aqua et Stephanus filius eius 1209 AssRoll 558 (Pleas, iv, no. 4130) (Bodney). -terram Rogerifilii Osketel c. 1230 Der, fo. 37a (Wereham). - Walterus Osketel c. 1230 (c. 1270) BuryS, fo. 127a (Tivetshall). - De Hosketel c. 1270 (m. 14th) Rams, i, p. 61, Hosketel ibid., p. 62 (Wimbotsham). (iv) Anschitillus 1086 DB, fo. 149b (Swannington). -Anschetellus prepositus 1086 DB, fo. 198 (Melton Constable). - Ansehetel filius Uspaci 1086 DB, fo. 279b (Bamingham Norwood). -Anschetil paruus 1165-1166 P, p. 25, 1166--1167 P, p. 25 (Lynn). - de Anschetillo et terra sua, Anschetill' (acc.) c. 1170 CAcre, fo. 102b(i) (Wellinga). 108 Anschetillum de Wenlinges c. 1170 CAcre, fo. 26a(iv) (? Kempston), terram que fuit

106 homagium Bricii filii Rogeri filii Leuuini de Wigenale. Bricius is 1. OE Brihtsige < Beorhtsige andLeuuinus is OELeofwine. 107 See the previous entry. 108 This charter occurs on the same folio as a group of charters relating to Wendling.

50

Åskell, Åsketill, Eskil, Åskel, lEskil Anschetilli de Wenthling c. 1175-1180 CAcre, fo. 28a(iv) (prob. South Acre). 109 - de catallis Anketilli de Derham 1176---1177 P, p. 138 (East or West Dereham). - Carta Anketilli de Starifeld c. 1180-1185 Wymondham, fo. 25a(ii) (Stanfield), ex dono Ansketilli de Stanfeld 1182 Wymondham, fo. 10b, ex dono Anketilli de Starifeld 1230 Wymondham, fo. 22b (Wymondham). - de feodo Ansketilli 1182 Wymondham, fo. 10b (Nf). 110 -Anchetill' (seruiens ofthe Prior) (abl.) (witn.) c. 1180-1185 CAcre, fo. 35a(i) (Weasenham and the Honor of Gressenhall), Anskitillo (seruiens of the Prior), abl. (witn.) c. 1185-1200 CAcre, fo. 39b(iii), Anketill' (seruiens of the Prior) (witn.) ibid., fo. 39b(v) (Rudham), Anketillo (seruiens of the Prior), abl. (witn.) ibid., fo. 80a(iii) (Lynn), Anketill' (seruiens of the Prior) (abl.) (witn.) ibid., fos. 65a(iii) (Norwich), 82b (Wiggenhall), Anket' (seruiens of the Prior) 111 (abl.) (witn.) ibid., fo. 82a(v) (Wiggenhall). -Anchetill' filius Alfredi c. 1180-1200 Lewes, fo. 287a(ii) (Nf Marshland). - de catallis Ansketill' Bulem 1182-1183 P, p. 16 (Nf or Sf). 112 - Rannulfus f Anketil 1191 P, p. 44,Rannulfusf Anketill' 1194 P, p. 189, Rannulfusf Ansketill' 1194 P, p. 54 (Nf). - Willelmusf Anketil 1191 P, p. 44, 1192 P, p. 189, (Willelmusf)Ansketill', 1193 P, p. 21, Willelmusf Ansketill l 194 P, p. 54 (Nf). -Ansketillus larchier 1192 P, p. 190, Anketill' le archier 1195 P, p. 68 (Nf). -Rogerusf Ansketilli 1194 P, p. 55, Rogerusf Anketill' 1195 P, p. 68, de Rogerof Anketill' 1196 CR (PRS NS, vii), p. 128, pro fugaRogerif Anketill' 1197 P, p. 246 (Nf or Sf, prob. Nf). -Anketil' 1199 CRR 18 (Pleas, i, p. 210, n. 4) (Lynn). -Auketil' (sic) (acc.) 1199 CRR 11 (CurR, i, p. 285) (Nf). -Willelmus filius Auketill' (sic) 1199 CRR 11 (CurR, i, p. 227), Willelmus filius Auketil (sic) 1199 CRR 11 (CurR, i, p. 302), Willelmumfilium Auketill' (sic) 113 et Thedam uxorem ejus 1199 CRR 11 (CurR, i, p. 412) (Nf), Willelmumfilium Anketilli 1200 CRR 16 (Cur, i, p. 220) (Lynn), 1200 CRR 24 (Cur, i, p. 321), de infirmitate Willelmifilii Anketilli 1200 CRR 24 (Cur, i, p. 350) (Nf), Willelmumfilium Anketill' et Thiedam uxorem suam, Willelmumfilium Anketill', Willelmo filio Anketill' et Thiede uxori sue, ss, whereas in England it represents the more conservative Atsur with the OFr spelling se for /ts/ (see above). In Scandinavia forms with spellings in s(s) merely stand for OEScand As(s)ur, but in England the spelling -ss- in the form Assori, gen., (above) and -s(s)- in the early forms of Azerley WRY (PN WRY, v, p. 199) is due to OFr usage (see Forssner, p. 39). PNDB, § 98, regards DB Assorin as reflecting AN reduction of ts > s or the ODan assimilation of ts > ss but also suggests that the possibility of native development should perhaps also be considered. Since the ODan assimilation of ts > ss ( > s) is only attested from the mid-llth century (see Br~ndum-Nielsen, § 337.3; DgP, col. 85; PNDB, § 98), 201 it is unlikely that the ODan form comes into question here. The Scand forms in sz (Aszur etc.) are the result of spelling uncertainty on the part of the scribe. In England different spelling forms for the name of the same person are common. The English form Azor contains o for unstressed u probably as a result of an OE development (see Campbell, §§ 373, 374). Note, however, that some of the examples of the name found in Scandinavia itself also contain o for unstressed u (Lind, col. 1264 ff.; SMP, col. 211 ff.). The use of e for unstressed u in the English forms Atser, Acer(us) and Ascer(us) reflects the 1. OE levelling of a, o and u in final and unstressed positions to e [.i] (Luick, § 440; Jordan, § 134; Brunner, AbriB, § 24; Mosse,§ 35; SPNLY, § 58).

199 No such forms occur in original OE documents or in other contemporary OE sources. The letter z with the value /ts/ is, however, known in OE from Biblical names such as Azarias or Baldazar and hence is sometimes also used in native words (Campbell, § 53). Also the OFr c for /ts/ is already found in I. OE as are the spellings dz and ds (Brunner, AeGramm, § 204, Anm. 7; Campbell, § 53). Such forms are, however, rare. 200 For the dates of the manuscripts of the charters in which these forms appear see the appropriate entries in Sawyer. The forms Azur and Adzur cited by Brunner, AeGrarnm, § 204, Anm. 7, do not appear in any original OE documents or in other contemporary OE material. 201 The s rune is also used to represent /ts/ in Danish runic inscriptions of DR Per. 2-3 (i.e. c. 750/800---c. 1150) (DR, cols. 972-973). DR, col. 973, gives the forms asur (DR, no. 131) and sasur (DR, no. 110) among the examples of the use of the s rune to represent /ts/ from DR Per. 2 (i. e. c. 750/800---c. 1050).

92

B

MEBagge b) (i) Willelmus Bagge 1165-1166 P, p. 30, 1166--1167 P, p. 30, 1167-1168 P, p. 19, 1168-1169 P, p. 97, 1169-1170 P, p. 6 (Nf).

OScand Baggi is an original byname 'pack, bundle, large unwieldy person' (DgP II, col. 29) and is known throughout Scandinavia both as a forename and as a byname (Lind, col. 108; LindBiN, col. 12; DgP, col. 90; DgP II, cols. 2629; SMP, cols. 221-222). Baggi seems to form the first element of several English place-names (PN NRY, pp. 145, 189, 332; PN ERY, p. 223; PN WRY, v, pp. 23,209; PN Nt, p. 131; PN Db, p. 220). SPNLY, p. 45, takes the OE substantive *bagga, apparently the name of a wild animal, and the OE personal name Bacga to be possible alternative etymons of the first element of the Y place-names and of the L place-name Bagholme. EPN I, p. 17, s. v. *bagga, takes Bagholme L to contain OE *bagga and regards it as the only case in which this element is compounded with a Scand place-name element. It is unlikely that OE Bacga occurs as the first element of place-names whose second element is of OScand origin, since monothematic personal names were probably rare in OE after the 9th century. The above Nf byname is probably a formation of parallel type to OScand Baggi, using ME bagge< ON baggi 'bag' as a nickname with the sense 'large, lumpish fellow'.

(?) ODan Bakki a) (i) Bachestorp 1086 DB, fo. 169b, Baketorp 1209 AssRoll 558 (Pleas, iv, no. 4323), Bakethorp 1254 Val (DEPN, p. 23), Bagetorp 1198 FF (PRS NS, xxvii, no. 107), Baggetorp' (p) 1206 CRR43A (Cur, iv,pp. 245, 302),Baggetorp' 1209FF(PRS NS,xxxii, no. 134) (Bagthorpe, Gallow H).

DEPN, p. 23, gives the first element of the Nf Bagthorpe as either ODan Bakki or OE Bacca. EPN II, p. 212, s. v. porp (ix) regards the etymon as being the ODan Bakki. For ODan Bakki see DgP, col. 91. If ODan Bakki is the etymon here, then the spellings in Bag(g)- would reflect AN interchange of k and g (see PNDB, § 119). However, the DB form Bachestorp has a strong genitive 93

Bersi, Brerse in -es, and this would perhaps suggest that the etymon is the ON byname Bak which is known from place-names in L and NRY. 1 In view of the rarity of monothematic personal names in OE sources after the end of the 9th century it is unlikely that OE Bacca would come into question as a possible first element for Bagthorpe. The DB form Bachestorp contains AN ch for OScand /k/ (see above, s.n. A.ki, phonology (i), for details and references).

ON, ODan Bersi, OSwed Brerse a) (i) Bersisty 1231-1253 (m. 14th) Rams, ii, p. 314 (f. n. in Ringstead). -Bersisholm 1234 (e. 14th) NorwCPCh, no. 357 (f. n. in Keswick).

This name was frequent in both Norway and lceland (in the forms Bersi, Bessi etc.) (Lind, cols. 132-133; LindS, cols. 163-168), but is rare in OEScand (DgP, col. 116; SMP, col. 533). The name is an original byname from an appellative belonging to lcel bessi 'he-bear', Norw besse 'large, heavy fellow', Swed biesse 'bear; large, heavy fellow, ruffian', Swed bjässe 'heavy fellow' (SRU, i, p. 108). The name is rare in England (NPN, p. 27; PNDB, p. 201; SPNLY, p. 53).

OScand *Biöil(I)-Åki a) (i) Bitlakebei 1086 DB, fas. 174b, 201, 272b, Bithlakebei ibid., fo. 217, Billokebi 1198 FF (PRS NS, xxvii, nos. 167,210), 1204 CRR 33 (Cur, iii, p. 182), 1209 AssRoll 558 (Pleas, iv, no. 4177), Billokesbi (p) 1202 FF (PRS NS, xxvii, no. 453) (Billockby, West F1egg H).

Ekwall suggests that the first element of Billockby might be the OScand personal name A.ki with ON biöill 'wooer' prefixed to it (DEPN, p. 43, s.n.). This would be perfectly acceptable as the etymon of the first element of Billockby. The prefixing of a substantive with the function of a byname to a personal name is well attested in OScand (NK VII, pp. 270, 275; Modeer, SvP, p, 97). Sandred thinks Ekwall' s suggestion is the best that has been put forward so far (Sandred 1987:313). The original form of Billockby would have been an OScand *Biöil(l)äkaby. The 1. 12th - e. 13th century forms in Billokebi show that the initial ä of the second component, the OEScand personal name Äki, was first rounded to Q before undergoing shortening (see Jordan, § 137). 1

Forthese place-names see PN NRY, p. 191; DEPN, pp. 30, 31, s. nn. Baston,Baxby; SPNLY,

p.45.

94

Bi~rn, Biorn The DB spelling t(h) for OScand intervocalic [ö] is the result of AN influence (see PNDB, § 108). The subsequent loss of intervocalic [ö] reflects an OFr development (see above, s.n. Auöi). For the OEScand personal Äki, which is quite frequent in medieval Nf, see above, s.n. Aki.

ON Birkibeinn (byname) b) (i) Willelmum Birkebein 1199 CRR 17 (Pleas, i, no. 2113) (Nf). - Elwinus Birkebeyn 1222 ElyA, fo. 166a (Terrington).

ON Birkibeinar was applied collectively to the followers of King Sverre of Norway (1184-1202) and in "MEBirkabeyn appears as the name of a mythical Danish king who was the father of Havelok in The Lay oj Havelok the Dane (Early Middle English Texts, ed. B. Dickins and R. M. Wilson, Cambridge 1951, p. 34). The Lay oj Havelok the Dane seems to have been composed 'towards the end of the 13th century' and the MS dates from the early 14th century, but the story was already known in the 12th century (Dickins and Wilson, op. cit., pp. 34-35). The appearance of the ON byname Birkibeinn in Nf sources of 1199 and 1222 is probably to be attributed to the popularity of some version of the Havelok story rather than to any direct contact with the Norse stories centred around King Sverre. The "ME dialect of The Lay oj Havelok the Dane has been traditionally regarded as that of northem L (Dickins and Wilson, op. cit., p. 34), but A. Mclntosh, 'The Language of the Extant Versions of Havelok the Dane', MJE, xlv (1976), pp. 36-49, though not disputing the evidence for the original version having been composed in L, takes the language of the extant text (Bodl. Laud Misc. 108) and that of the fragments preserved in CUL Add. 4407, Art. 19, to suggest that they were written by scribes from western Nf. The byname Birkibeinn is also known from later Nf sources, i.e. Thomas Byrkebein 1327 SR, Thomas Birkebeyn 1332 SR (Selten, Bahuvrihi Names, p. 45). Selten, (loc. cit.) defines the sense of the name as 'one who ties bark of birch round his legs'. For further comment upon this byname see NPN, p. 27; ZEN, p. 24; von Feilitzen, Notes, p. 66.

ON Bic;_>rn, ODan, OSwed Biorn a) (i) Bernestok c. 1180 Binham, fo. 20a(i) (f. n. in Hindringham). - Berneshoge 1200 Coxford, Bernishowe 1210 Coxford, Berneshoue 14th Terrier (f. n. in Barmer) (OKS). - Bernescrofte 1414 Ct (f. n. in Holme next the Sea) (OKS).

95

Bo An original byname, 'bear', Biorn, is well attested throughout Scandinavia in the Viking and medieval periods (Lind, cols. 143-147; DgP, cols. 124-128, 1649; SMP, cols. 318-346). The OEScand Latinized form Bero is probably the result of German influence (Modeer, SvP, pp. 59, 60) as is perhaps also the medieval Danish form Berno (cf. OHG Berno) (see DgP, col. 128). For the name in England see NPN, p. 26; PNDB, p. 200; SPNLY, pp. 5455; von Feilitzen, SCBI, xi.i, p. 12; Smart, Cnut's York Moneyers, p. 225. Usually it is impossible in English sources to distinguish between OScand Biorn etc. and OEBeorn (see PNDB, p. 200, SPNLY, p. 55). Some of the 11th century examples of Beorn in English sources represent the OScand name (NPN, p. 26; Redin, p. 4) but in ME both OE Beorn and OScand Biorn would result in a form Bern. 2 OE Beorn occurs as first element of the Nf place-name Barnbarn Broom and probably also of the lost Nf place-name Berneswrde 1086 DB, fo. 133.3 Note, however, that it is possible that the first element of the above Nf field names is the appellative OE bere-:ern, beren, bern 'barn, storehouse for barley and other grain'.

ODan, OSwed Bö a) (i) Botuna 1086 DB, fo. 261b, Boton 1239, 1286 FF (OKS), 1316, 1346, 1428 FA (OKS) (Booton, South Erpingham H).

OEScand Bö is derived from an earlier *Böi4 (Br111ndum-Nielsen, § 218). The name is an original byname with the sense 'inhabitant, resident' (NE VII, p. 204; Otterbjörk, Svenska förnamn, p. 79; SPNLY, p. 68) (cf. ON bua, OEScand böa 'to dwell'). The name is very common in ODan and OSwed (DgP, cols. 134-139; SMP, cols. 418--430). The OEScand Latinized forms Boecius, Boetius are the result of association with the name of the 6th century Latin philosopher Boethius (NK VII, p. 188; Modeer, SvP, p. 60). The OWScand counterpart of OEScand Bö is ON Bui which is rare (Lind, cols. 180-181; LindS, col. 192). It is a loan from OEScand which has been transformed in keeping with OWScand phonological development (NK VII, p. 133). 2 ON iq, OEScand io is represented by e in English sources (SPNLY, § 54). OE eo is monophthongized to /p/ in 1. OE and then to /e/ in ME (Luick, § 357). In stressed syllables OE eo would have been lengthened to eo in the 9th century before the homorganic consonant group rn (Luick, § 268). This eo would have become e in ME (through the intermediate stage /p:/) and would have been represented by the spelling e (see Jordan, § 84; PNDB, § 33). 3 OE worö is only rarely compounded with OScand personal names (EPN Il, p. 275). 4 The older form occurs in OSwed as runic bui (SRÖg, no. 81; SRSö, no. 148; SRSmå, no. 89; SRU, no. 476).

96

Bolli Ekwall, DEPN, p. 53, s.n., took the etymon of the first element of Booton Nf to be either OE Böta or OEScand Bö. EPN II, p. 198, s. v. tiln, gives the etymon of Booton Nf as ODan Bö and this would suit the forms better than OE Böta. Ekwall, DEPN, p. 52, s.n., also takes OEScand Bö to form the first element of Boothorpe Lei but Fellows Jensen, SSNEM, p. 104, regards the appellative OEScand bön. 'farm' as more likely here. Ekwall, DEPN, p. 56, s.n., also regarded OEScand Bö as the first element of Bowthorpe Nf but the forms here (Boethorp, Bowethorp 1086, Boytorp 1183, Bugetorp 1230) rather suggest that it contains OE boga or ON bogi 'a bow, an arch' in the sense 'a river bend'. This is also favoured by the site, Bowthorpe lying on a bend of the RiverYare. ON Bui, OEScand Bö occurs in independent use in Y TRE DB in the form Bu (PNDB, p. 211; SPNLY, p. 68).

ON Bofi, ODan Bövi, OSwed Böve b) (i) Bou. liber homo TRE DB, fo. 265b (Ringstead). -Adamfilius Boui 1165-1166 P, p. 30, 1166-1167 P, p. 30, 1167-1168 P, p. 19 (Nf). - Wlueuafilia Boui 1195 P, p. 75, 1197 P, p. 236, Wuluefafilia Boui 1196 CR (PRS NS, vii), p. 134 (Nf).

An original byname, 'clumsy fellow' (NK VII, p. 201), B6fi, Bövi etc. occurs throughout Scandinavia both as a personal name and as a byname (Lind, cols. 150-151; LindBiN, col. 34; DR, nos. 370, 390, 393, Mp. 30; DgP, cols. 156-157, 1652; DgP II, col. 114; SRÖg, nos. 103, 113; SRSö, no. 144; SRSmå, nos. 126, 153, 163; SRVg, no. 246; SMP, cols. 487-488). 5 The name is to be taken as largely OEScand (DR, col. 636). For English examples see NPN, p. 28; PNDB, pp. 204-205; SPNLY, p. 59.

ON, ODan (p. ns.) Bolli a) (i) Bollesdele c. 1210 Der, fo. 135b (f. n. in South Clenchwarton). b) (i) Osbertus Balle 12th - 13th (m. 14th) Rams, i, p. 100, no. 402 (Upwell). - Balle (gen.) c. 1140 CAcre, fo. 26a, terram Balle c. 1140 CAcre, fo. 33b(iii), 1146-1147 CAcre, fo. 3b(i) (Kempston). -Ailbryctus Balle de Wellis filius Goderici e. 13th Lewes, fo. 280a(ii), DeAilbricto Balle 1206-1207 (m. 14th) Rams, ii, p. 319 (Upwell and Welney).

An original byname, 'fat, round person', cf. ON bolli m. 'small, bulging vessel' (NK VII, p. 51; SPNLY, p. 59), Bolli is fairly well attested in OWScand 5 The Danish examples of the byname can, however, belong to e. ModDan bove 'rascal, scoundrel', MLG bove 'fellow, urchin' (DgP Il, col. 114).

97

Bondi, Böndi, Bundi, Bönde where it appears both as a personal name and as a byname (Lind, cols. 151152; LindS, cols. 172-173; LindBiN, col. 35). In Danish Bolli forms the first element oftwo place-names in -thorp (DS XIV, p. 128). Note also OE Bolla which is probably an original byname belonging to OE bolla m. 'bowl, cup' (and, therefore, cognate with the Scand Bolli) (PNDB, pp. 204--205). This OE name is formally possible as the etymon of the above Nf forms though perhaps in Bast Anglia the Scand etymon might be preferred. For further discussion of OScand Bolli in English sources see PN NRY, pp. 140,210; PN ERY, p. 261; PNDB, p. 205; SPNLY, p. 59.

ON Bondi, ODan Böndi, Bundi, OSwed Bönde a) (i)Bondesboydole 1214-1229 Holme, fo. 74a (f. n. inPotterHeigham). -Bundescroft c. 1225 Der, fo. 89b(i) (f. n. in Upwell). -Bondesbusch c. 1250 Langley, fo. 53a(ii) (f. n. in Hellesdon). -Bondeswong c. 1250 Wals, fo. 52b(iii), Bondeswongate 1270-1271 Wals, fo. 50b (f. n. in Holkham). -Bondeswong c. 1250 Wals, fo. 67a(v), Bondyswong 14th Wals, fo. 168b (f. n. in Great Snoring). -Bondeswode 14th Bromh, fo. 32b (f. n. in Hedenham). -Bondesmere 1423, 1435 Ct (f. n. in Flitcham) (OKS). Bondyshill 1425 Rental (f. n. in Gaywood) (OKS). (ii) Bundehaluaker, Bundehilhaluaker 1237-1251 Holme, fo. 106a(iii) (f. n. in Ludham). b) (i) Bondo liber homo TRE DB, fo. 121b (Crownthorpe). -Bundo TRE DB, fo. 184b (held one liber homo with 30 acres at Runton). - Bunde liber homo TRE DB, fo. 226 (West Walton). -/. liber homo Bondo TRE, 1086 DB, fo. 182b (Heckingham). -Bundo liber homo TRE DB, fo. 270b (Rushford). - Bundus. I. tenus TRE DB, fo. 258 (Great Walsingham), Bundus liber homo Heroldi TRE DB, fo. 260b (Erpingham), Bondus liber homo TRE DB, fo. 261 (Edgefield), Bonde liber homo TRE DB, fo. 197b (Gateley), Bondus TRE DB, fos. 233,237, 237b (East Raynham, Bodney, South Raynham), Bondus. liber homo TRE DB, fo. 237 (Burnham), Bundus TRE DB, fo. 237 (bis) (Langford, South Creake), Bundo TRE DB, fo. 238 (ter) (lslington, Middleton, West Bilney), Bundo. I. liber homo TRE DB, fo. 237b (bis) (Roughton, Beeston Regis), Bundo. liber homo TRE DB, fos. 238b, 239 (Gayton, Bodham), Bundo liber homo TRE DB, fo. 239 (Gateley). - Bundi Wml/Wm2 (1086-1089 ?) SCBI, xi.ii, no. 171 (moneyer, Thetford). - decima Bondi Gretsur c. 1100-1120 CAcre, fo. 2a (? Rudham). Bondus 1101-1107 (14th) Holme, Appendix A (Catfield), 6 Bonde pine et alterum Bondumfilium Offles ibid. (Repps). -Bondo 1108-1150 Wymondham, fo. 6lb(i) (in this document Lefricus de Crunkethorp granted three acres of arable which Bondo had held

6 Though this record was drawn up in the period 1101-1107, most of the material which it contains relates to the latter part of the 11 th century. The entry conceming Bondus of Catfield belongs to a period between Roger Bigod's acquisition of his Norfolk fee in the early post-Conquest period and 1086 (see Stenton, EHR, xxxvii (1922), pp. 226, 227 and n. 3) while that relating to Bonde pine and Bondus son of Offel belongs to the period immediately following the Norman Conquest (see Stenton, EHR, xxxvii (1922), pp. 227 and n. 6,233).

98

B6ndi, Böndi, Bundi, Bönde from him to Wymondham Priory; Crunkethorp is Crownthorpe). - terram que fuit Bunde aui sui etAde patris sui (Ricardus dePortan') c. 1115 (1395-1396) SSMEA, Appendix I, no. 56 (the soke ofNecton). -Bondus de Cressyngham (witn.) c. 1115 (13951396) SSMEA, Appendix I, no. 57 (the soke of Necton; Cressyngham is Cressingham). -Banda presbitero, abl. (witn.) c. 1115 (14th) CAcre, fo. lOla(i) (West Lexham). -terram Bondi 1118-1138 CAcre, fo. 3a(ii) (Barmer). -Banda; Torkilfilius eius (witn.) c. 1120 before 1122 (12th) France, p. 414, no. 1149 (Sporle and Mileham). - Ricardus filius Bondi (witn.) 1134--1140 Holme, no. 132 (Grengeuile ), Transcriptum Ricardi filii Bondonis, Ricardo filio Bondonis, dat., 1134--1140 Holme, no. 133 (Thwaite, Thurgarton), De Ricardo filio Bondonis 1134--1140 (l. 13th) Holme, no. 133 note (this entry almost certainly relates to Thwaite and Thurgarton since it appears in the MS. as a note in the margin of Holme, no. 133), Ricardus filius Bonde, transcriptum Ricardi filii Bonde 1134--1140 (l. 13th) Holme, no. 140 (Bessingham, Thurgarton, Thwaite), Richardus filius Bond' (witn.) 1140---1145 BodlNfCh, no. 607 (Grenesuilla), 7 Ricardus filius Bonde after 1153 BodlNfCh, no. 608 (printed, Holme, Appendix B) (Bessingham and Tibenham), 8 Ricardo filio Bondi, dat., Julianus consanguineus Ricardi filii Bundi c. 1155 HMC VarColl, vii, p. 176, no. 96 (North Erpingham H), Roberto filio Ricardifilii Bondi de Bashigham, dat., c. 1180 HMC VarColl, vii, p. 230, no. 325 (Sustead; Bashigham is Bessingham). - terra que fuit Bonde de Stalham c. 1140 Holme, no. 66 (Stalham). - toftum Bonde 1146 - c. 1150 BodlNfCh,9 no. 609 (printed, Holme, no. 156) (Panxworth). - Bondi (gen.) c. 1155 CAcre, fo. 49b(v), Banda presbitero, abl., ibid., fo. 50a(i), Bondum presbiterum et Lefstanum socium eius ibid., fos. 50a(i), 50a(ii), terram Bund' sacerdotis et Lefstani socii eius ibid., fo. 50a(iii), Bondum c. 1175 CAcre, fo. 42a(i) (Creake). - Erat Norwici puella octennis Agnes nomine cuius pater Banda, cognomento Hoc, et mater Gunnilda dicebatur? 1156 St William, p. 273 (Norwich). - Gaufrido filio Bund', abl. (witn.) c. 1160 Binham, fo. 55a(iii) (Bamey). Banda preposito monachorum, abl. (witn.) c. 1160 CAcre, fo. 16b(ii) (Massingham), Bundo preposito, abl. (witn.) ibid., fo. 20a(iii) (Little Massingham). - Bondus 11611177 (13th) Rams, iii, p. 268 (Ringstead and Holme next the Sea). - Banda Rond' 10 (abl.) (witn.) c. 1165 (e. 15th) NorwCPCh, no. 124 (Lynn and Mintlyn), 11 Bond' de Linna cognomento Rond c. 1185-1200 CAcre, fo. 80a(iii), Bond' cognomento Rond de Linna ibid., fo. 80a(iv) (Lynn). -Banda (Merciarius) 12 l l65-ll66P, p. 21,BondaMerciarius 1166--1167 P, p. 22 (Lynn). -Bonde Hund 1165-1166 P, p. 22, 1166--1167 P, p. 22 (Lynn). -deBondofilio Osward' 1174--1175 P, p. 123 (Nfor Sf). -Bonde de Colneia 1174--1175 P, p. 119, de Bonde de Colnea 1175-1176 P, p. 66 (Colney). -Boda Sellator 1174--1175 P, p. 121, 13 Boda se/latar 1175-1176 P, p. 67, 13 Banda sellator 1176--1177 P, p. 129, 1177-1178 P, p. 23 (Nfor Sf). -de catallisBonde Tixtore, de cat-

7 This document is printed as SSMEA, Appendix I, no. 26, where the form is incorrectly given as Richardus filius Bond, and as Holme, no. 161, where the form is incorrectly rendered as Richerusfilius Bondi. 8 This document is a list of alienations of lands of the Abbey of St Benet of Holme in the period 1101-1153 and was probably composed sometime between 1153 and 1168. The form Ricardus filius Bonde in this document belongs to the period 1134-1140. 9 Wrongly dated by the editor of Holme to 1141-1149. 10 Var: Bond' Rand, Bonde Rand (e. 14th). 11 For the u before nd, which was, in any case later than the period of the Scand settlements in England (see Fellows Jensen, NoB, lvii (1969), pp. 63-64). The AN inverted spelling would seem to be reasonably clear where Bond- and Bund- spellings altemate in the name of the same person. Altematively, the influence of the appellative in its 1. OE form bunda might be suggested.

ON Bosi, ODan Bösi, OSwed Böse a) (i) Lefrizfilius Bose tegnus regis TRE DB, fo. 228b (Roydon near Diss).

An original byname of uncertain meaning, 32 B6si, is known in both OEScand and OWScand as a personal name and byname, though it is much more frequent in the former (Lind, col. 157; LindS, cols. 173-174; LindBiN, col. 37; DgP, cols. 149, 1651; DgP Il, col. 113; DR, col. 637; XenLid, p. 104; SRSmå, pp. 45-46; SMP, cols. 445-446). In England the name is attested by DB in Brk (PNDB, p. 207). As PNDB, p. 207, points out, however, OE Bösa is formally possible as the etymon of the above Nf form and of the Ess form Boso (liber homo). 33

ON, ODan Broddi, OSwed Brudde a) (i)/. fiber homo Brode, Idem Brode TRE DB, fo. 164 (Rockland All Saints), Brodo TRE DB, fo. 277 (held, in company with Aluuinus, 44 acres of the king at Shipdharn).

Broddi is a hypocoristic side form of Broddr, an original byname based on OScand broddr 'spike' (NK VII, pp. 44, 59; Hald, Personnavne i Danmark: I. Oldtiden, p. 71). The name is well attested in Iceland and Sweden but is extremely rare in Denmark and is absent in medieval Norway (Lind, cols. 169-

31 Note also the field name Bonde(s)made C 1237, 1326, which PN C, p. 338, takes to belong to the appellative in the sense 'peasant proprietor'. 32 It might signify: 'a heavy, lumbering fellow' or 'a womanizer' (NK VII, p. 51; see also SRSmå, pp. 45--46). 33 PNDB, p. 207, does, however, group all three forms under ON B6si, ODan Bosi, OSwed Bose.

107

Broöir, Bröthir, Brödhir 170; SMP, cols. 503-506; DgP, col. 162). The strong form, ON Broddr, ODan Brod, OSwed Brud is known throughout Scandinavia, though it is not common in Danish and only represented by a single form from the Landnamab6k in Icelandic (Lind, cols. 170-171; LindS, col. 183; DgP, col. 162; SMP, cols. 500-503). In England OScand Broddi is recorded by DB in Bd as well as in Nf (PNDB, p. 208).

ON Broöir, ODan Bröthir, OSwed Brödhir a) (i) Brodercros 1086 DB, fös. 1 llb, 122, 169, 179, 183b, 191b, 233,237,257, 262b, Brodescros ibid., fo. 215b, Brodecros ibid., fo. 145b, Brodercos ibid., fo. 197b, Broöercros 1168 P, 1185 P, Brodercroshundredum 1190 P (Brothercross H; see EHN I, p. 66, for a full conspectus of forms). b) (i) de catallis Brother34 fullonis 1176-1177 P, p. 137 (Nf or Sf).

Based on the substantive br6öir 'brother', the personal name Br6öir is very common in Denmark and Sweden but is rare and late in Norway and only known in Icelandic from the name of a Viking of unknown nationality in Brennu-Njalssaga35 (Lind, col. 171; LindS, col. 183; DgP, cols. 163-165; 1652-1653; DR, nos. 90,118,268,275,343,370,383,398; SMP, cols. 494499; NK VII, pp. 40, 203; Kousgård S0rensen, Bebygg. på -sted, pp. 255-256; Hjorth Pedersen, Bebygg. på -by, p. 20). An OE Brööor occurs as the name of a 9th century Northumbrian moneyer in the form Broöer (PNDB p. 208, n. 7; von Feilitzen, SCBI, xi.i, p. 10). In 1. OE and ME, however, Broöor, Brother represents OEScand Bröthir (NPN, p. 30; ZEN, p. 27; PNDB, p. 208; DBS, p. 53, s.n. Brothers; SPNLY, p. 65). The 1. OE form Broöor shows Anglicization which was the result of association with the OE appellative brööor 'brother'. Post-Conquest examples of OEScand Bröthir from Sf are as follows: Brother (liber homo) 1066--1087 (1. 13th) Bury, no. 168 (Sf); Brother clericus (witn.) c. 1115 (c. 1175-1200) Bury, no. 174 (Santon Downham). The OEScand personal name is probably also the first element of the place-names Brothertoft L, Brotherton Sf, WRY and Brotherwick Nb, though the appellative OE brööor, ON br6öir or the OE personal name Brööor must also be considered as formally possible alternatives (cf. EPN I, p. 53, s.n. brööor; DEPN, p. 69, s. nn. Brothertoft, Brotherton, Brotherwick; PN WRY, iv, p. 45; PN WRY, vii, p. 50; SSNY, p. 126; PN NbDu, p. 32).

Bother CR. As NK VII, p. 40, points out the substantive 'brother' was also used as a personal name in West Germanic (see below for OE Brööor), and it might have been borrowed from there into OEScand. The absence of the name in Iceland suggests that it was not known in OWScand at the time of the settlement of Iceland and was only borrowed much later from OEScand. 34

35

108

*Brunsveinn, *Briinswein

When used as a byname in post-Conquest sources BrojJer, Brother etc. might occasionally have had a patronymic function, but probably in mast cases it stands for the appellative ME brojJer in the sense 'monk' (cf. ELPN, p. 141; SPNLY, p. 65).

ODan Briinkil, OSwed (run.) Briinkretill, Briinkil b) (i) Brunketel (abl.) (witn.) c. 1185 CAcre, fo. 16a(v) (Massingham), Brunketel (witn.) ibid., fo. 20b (Wariedecrundel in Massingham). -Brunketello de Tatersete, abl., Brunketellum de Tatersete c. 1220 CAcre, fo. 43a(i) (Tattersett).

In Danish Brunkil is not attested independently until the 16th century, though it occurs as the first element of a place-name in -thorp in Skåne (DgP, col. 167). In OSwed the name is known from the runic forms brunkitil (SRU, no. 371) and brunkil (SRU, no. 494). The name does not appear in OWScand. In England an example of Brunkil is recorded in L (von Feilitzen, Notes, p. 54; SPNLY, p. 67). It is not clear whether the Sx form Brunchild' (var. Brunkil) 1208 MemR belongs here or toan OE *Bruncild (von Feilitzen, Notes, loc. cit.). For the second element -ketel(l) see above, s.n. Arnkell.

OScand *Brunsveinn or AScand *Briinswein a) (i) Brunsueinescrof c. 1240 Der, fo. 235a(i) (f. n. in Clenchwarton). b) (i) Brusuan36 c. 1090--1100 CAcre, fo. lb (Nf). -Brunswen (acc.) 1091-1097 Lewes, fo. 12b (Walpole). -Brunsuen (abl.) (witn.) c. 1125-1135/36 CAcre, fo. 91b(iv) (Narford). - terram quam Brunswein de Wutton' tenuit in Wutton' c. 1190--1200 Wymondham, fo. 49b(i) (South Wootton). - Henricus filius Brunswein, Henricum filium Brunswein', Henrico filio Brunswein, dat., 1198 FF (PRS NS, xxvii, no. 23) (Wiggenhall), Henricum filium Brumsewein. et Rogerum. et Ricardum fratres suos 1198 FF (PRS NS, xxvii, no. 193) (Saddlebow), Ricardi filii Brunswain, gen., e. 13th (15th) CAcre, fo. 15lb(iii) (Wiggenhall), de hornagio Thomefilii Henricifilii Brunsuein de S., homagium Thome filii Henrici filii Brunsuen de Sadelbowe c. 1225 Der, fo. 199b(iii) (Wiggenhall St Peter), domum que fuit quondam Brunswen ibid., fo. 200a, domum que quondam Brunswein ibid., fo. 225b(i) (Saddlebow). -Ivonemfilium Brunsewein 1198 FF (PRS NS, xxvii, no. 121) (Tilney). -Briwiswein (sic) 37 de Wigehal' (acc.) 1199 CRR 11 (CurR, i, p. 243) (Wiggenhall), Briuiswein (sic) 38 1199 CRR 11 (CurR, i, p. 251)

The scribe has omitted the nasalizing stroke over the frrst u. Probably the editor of CurR has misinterpreted the minims of the form in the MS., and has, thus, transcribed it incorrectly. Probably the correct reading of the form is Brumswein. 38 This is probably a case of incorrect transcription similar to that described immediately above in n. 37. The correct reading of the form is probably Brunswein. 36 37

109

Brynhildr, Brynhild (Nt), Carta Alani Brunsuein de Wigehale, Alanus filius Brunsuein de Wigehale c. 1220 Der, fo. 163a(i) (Tilney), Carta Alani Brunsuein Blench de Wigehale, Alanus filius Brunsuein Blench de Wigehale ibid., fo. 18 la(i) (Tilney, Wiggenhall, Terrington; Wigehale is Wiggenhall). -Brunsuein (abl.) (witn.) e. 13th CAcre, fo. 80a(ii) (Lynn).

A personal name or byname *Brunsveinn is not recorded in Scandinavia itself, and the above Nf forms might represent an AScand compound formed from OE Brun- and OScand -sveinn. However, an OScand Brun-, corresponding to OE Brun-, occurs as the first element of ODan Brunkil, OSwed (run.) Brunkil, Brunkretill, and OScand -sveinn forms the second element of ON Bergsveinn, OSwedBirergsven. An OScand dithematic personal name *Brunsveinn, whose first element is an OScand Brun-, and whose second element is OScand -sveinn, can, therefore, be postulated without difficulty. A further possibility is that an OScand *Brunsveinn might have been an original byname, 'young man with brown hair or a brown complexion' (ON brunn 'brown' + ON sveinn m. 'young fellow, lad; free man in the service of another man; page'). Brunswein etc, seems to be a characteristically Nf name but cf. Brunswein (acc.) 1212 CRR 55 (Cur, vi, p. 249) (the document concerns Hanslope Bk). Forms in -swen, -suen show AN monophthongization of OScand ei > e (see Fellows Jensen, NoB, lvii (1969), pp. 69-70). In the second element of the form Brusuan, OScand -sveinn has been Anglicized as a result of association with the OE cognate of OScand sveinn, OE swan m. 'shepherd, swineherd; peasant; young man, warrior'.

ON Brynhildr, ODan, OSwed Brynhild fem. a) (i) apud Brunyld 1310 Ct (f. n. in Hindringham) (OKS). b) (i) Aschillus filius Brunilde c. 1180 CAcre, fo. 82a(i) (Wiggenhall).

This name appears to have come to Scandinavia from the Continent with the Frankish-Burgundian cycle of stories centring around Sigfrid and the treasure of the Nibelungen (NK VII, p. 67). 39 It goes back to OHG Brunihild, MHG Priinhilt. Leaving aside its appearance as the name of a literary figure, ON Brynhildr is quite frequent in Norway from the end of the 12th century onwards but is rare in lceland (Lind, cols. 175-177). OEScand Brynhild is extremely uncommon (DgP, col. 168; SMP, col. 510). A weak side form, Brynhilda, is also known from medieval Norway (Lind, cols. 174-175). In England ON Brynhildr, OEScand Brynhild has been taken to be the etymon of Brunil L VD and Brunild L VH and to be the first element of the field

39 For a discussion of the Sigfrid cycle in Scandinavia see J. De Vries, Altnordische Literaturgeschichte, i (Berlin 1964), pp. 87-92.

110

Brynhildr, Brynhild

names Brunildeshaga 1409 (Wa), Brunilde Cruch 1279 (C), Brundeshole, Brunildhole 1261 (Cu), Brunildeberge 1220-1250 (WRY) and Brunildesford 1360 (L) (NPN, p. 30; PN Wa, p. xxiii; PN C, pp. xx, 317; PN Cu, p. 505; PN WRY, vi, p. 237; SPNLY, p. 67). The last We minor name Rigg Brunild is taken by PN We, i, p. 99, to contain either ON Brynhildr or a ME Brunhild derived from the OG name. However, von Feilitzen, PNWinton, p. 152, took the form Brunild Prichelus 1148 (Ha) to belong to an unrecorded OE *Brunhild and for purposes of comparison cited the form Brunild l. 12th LVH, which Björkman (NPN, p. 30) had derived from the Scand Brynhildr. Ha lay in the heart of the old West Saxon kingdom, and, though OScand names are not unknown here in the ME period, von Feilitzen's derivation of the Ha Brunild from an unrecorded OE *Brunhild is plausible. OHG Brunihild need not be considered as a possible etymon for the English forms in Brunil(d)(e). Though it is known as the name of the famous 6th century Merovingian queen (who was, in fäet, of Visigothic origin), the name Brunihildis etc. does not seem to have been used in 10th and 11th century Normandy and the adjacent north French provinces.40 SPNLY, § 37, regards u in the Land WRY Brunilde- as an AN spelling for OScand y which implied no change of pronunciation. However, in Cu, Du, L, We and WRY, OScand Bryn- would have given rise to the ME spellings Brinand Bryn- (with y here being merely a scribal variant for i) on account of the unrounding of OE, OScand y to i in the ME dialects of these areas (see Jordan, § 41; SMED, pp. 116--120). The same is true in Nf, though occasionally the substitution of e for OE, OScand y might also be reckoned with here (see Ek, pp. 63-64; Selten, pp. 113-116). In C e, i and u for OE y are known, with e spellings slightly more numerous than i spellings and u spellings much less frequent than either e or i (PN C, p. xxv; Ek, pp. 62-63). In Wa and in Ha OE y generally kept the phonetic value /y/ until about 1400 and for this the AN spelling u would be quite normal (see Jordan, § 39.3; PN Wa, p. xxvi; von Feilitzen, PNWinton, pp. 223-224). 41 In view of these considerations it is thus possible that the forms Brunyld, Brunil(d)(e) found in English sources might

40 Morlet, p. 61, has 8th century examples from Alsace and 9th century examples from the Abbey of St Gallen in Switzerland and from the Rhineland. For the appearance of the name in Carolingian Fulda see Geuenich PN, pp. 57, 90, 145. Forthe use of the name in Spain in the 10th, 11th and early 12th centuries see Piel-Kremer, p. 105, s. n. Brunildi. 41 It should be noted, however, that i ande spellings for y appear occasionally in Ha and Wa in ME. For a discussion of the significance of such spellings in Ha see von Feilitzen, PNWinton, pp. 223-224, and the works cited there, and for those in Wa see PN Wa, p. xxvi. In view of the use of u for OE, OScand y in Wa in ME it is theoretically possible that the Wa f. n. Brunildeshaga 1409 could contain ON Brynhildr, OEScand Brynhild. This f. n. is, however, situated in Aston near Birmingham (PN Wa, p. xxiii), some distance from the Danelaw and its margins, and in view of this it is probably more likely that it contains the native OE *Brunhild.

111

Brynhildr, Brynhild

represent an unrecorded OE *Brunhild, though Anglicization of ON Brynhildr, OEScand Brynhild involving substitution of the OE name element Brun- for the etymologically unrelated OScand Bryn- might also be considered.

112

D

ON Dagr, ODan, OSwed Dagh b) (i) porta Edduardi hominis Oliueri Dages 1. 12th HarlCh 52 E 22 (printed, SSMEA Appendix I, no. 10) (? Edgefield). - Willelmo Dages, abl. (witn.) 13th HMC VarColl, vii, p. 178, no. 103 (Nf).

A name of considerable antiquity, OScand Dagr is frequent in medieval Norway and Sweden but is rare in ODan and fairly uncommon in Olcel (Lind, cols. 192-195; LindS, col. 196; DgP, cols. 181-183; SMP, cols. 559-562). The above Nf forms are paralleled by the Sf form IElfuine Dages 10871098 (c. 1175-1200) Bury, p. 39 (Elmswell Sf). Tengvik, p. 208, gives ON Dagr, OEScand Dagh(er) as the etymon of the Sf form Dages but also regards derivation from a short form of OE compounds in Dreg-, -dreg as a theoretical possibility here. This second suggestion of Tengvik is incorrect since the voiced palatal spirant 3 in OE Dreg-, -dreg was already vocalized to i in OE so that OE Dreg-, -dreg would have appeared as Dai-, -dai in e. ME (see Btilbring, § 505; Jordan, § 93; PNDB, § 132). Tengvik, p. 208, included the form Dages amongst a group of bynames formed from the genitive singular of personal names. Tengvik (pp. 207-208) regarded such bynames as patronymics in which OE sunu had been omitted. 1 In the case of Dages, however, there are good reasons for doubting whether this is the case. It is rather suspicious that OScand Dagr should appear three times in Bast Anglia in elliptical genitive formations and yet be apparently unrecorded in any other form there. 2 Perhaps it might well be that the form Dages has nothing to do with OScand Dagr and that it is, in fäet, a byname formed from an appellative. It might be suggested that it could be related to 1.

For further discussion of this type of byname see DBS, pp. xxxii-xxxiii. ON Dagr, ODan Dagh, OSwed Dagh(er) does not seem to have attained any degree of popularity in England. The sole example which I have been able to find in English sources is Dacher filius Eilaf e. 12th LYD, fo. 51b. Björkman, NPN, p. 31, regarded derivation of the form Dacher from ON Dagr, OEScand Dagh(er) as rather uncertain but, in fäet, there are no difficulties here. The spelling ch for OScand g ([y]) reflects the AN interchange of [y] and [kl (see PNDB, § 128; SPNLY, § 135). The ending -er probably represents the OScand nominative ending -r rather than an AN inorganic r. There are several other forms on L VD, fo. 5 lb which seem to have preserved the OScand nom. ending -r, i. e. Anander e. 12th (ON Qnundr, ODan, OSwedAnund), Grimer e. 12th (ON Grimr, ODan, OSwed Grim), Thorleuer e. 12th (ON Porleifr, OSwed Thorle/). 1

2

113

Danr, Dan ME (15th) dagge 'pointed thing or dagger' < OFr dague 'dagger' (MED, q. v.). Reaney, DBS, p. 93, s.n. Dagg took the ME byname Dag(g) (1275-) to be derived from OFr dague and regarded it as an ellipse for one who carried a dagger. Note also 1. ME (14th, 15th) dagges 'ornamental points or incisions on the edge of a garment' (MED, q. v.). This might have been applied elliptically to someone with a predilection for garments with such ornamental points or incisions or have been used as a metonymic for someone involved in the manufacture or sale of such garments.

ON Danr, ODan, OSwed Dan a) (i) Dannescroft c. 1235 Der, fo. 234a, Dannestoft ibid., fo. 234b (f. n. in Saddlebow).

An original byname 'Dane', this name is recorded throughout Scandinavia, though it is much more common in OSwed than elsewhere (Lind, col. 198; LindS, col. 197; DgP, cols. 184--185; SMP, cols. 563-572). The name does not, however, seem to have belonged to the stock of Scand names used in the Danelaw. It might also be suggested that the above Nf forms contain the OE personal name Dene. The spelling a for e and the doubling of n would have been AN features (see PNDB, §§ 8, 146). The form Dane for OE Dene with AN a for OE e is recorded in DB (PNDB, p. 223). Direct association with Lat danus, OFr daneis should also be considered here (PNDB, p. 49, n. 2).

OScand *Dottr a) (i) Dotescroft 1272-1288 NmwCP I (f. n. in Hindringham) (OKS).

A personal name Dat(us) occurs in DB in Bd, Bk, C, Ch, La and Sf and was regarded by von Feilitzen (PNDB, p. 226), following a suggestion of Björkman (NPN, p. 29), as being associated with either OE datt 'head of a boil', ModE dat 'a small lump, clot', ModE dialect dat 'a diminutive person or thing, a small lump' or with the etymologically identical ON dattr 'lazy, useless being', Norw dialect datt 'lazy, listless person'. 3 Von Feilitzen (PNDB, loc. cit.) believed that the local distribution of the DB forms on the whole favoured an OScand etymon. Note also that an ODan personal name *Dat has been suggested as the first element of the Danish place-names Daastrup and Daasroj (DgP, col. 201; DS VI, pp. 181-182).

3

For futher details of this word group see the references cited by PNDB, p. 226.

114

Du.fä, Diiva

ON Dufa, ODan, OSwed Döva (byname) b) (i) Normannus filius Duue 1161-ll 77(13th) Rams, iii, p. 286 (Wimbotsham and Downham Market). -Lefwinusfilius Duue, et Godemanfrater ejius 1161-ll 77(13th) Rams, iii, p. 288 (Hilgay and Snore Hall). -Godebaldfilius Duue 1165-1166 P, p. 26, 1166--1167 CR (1166--1167 P, p. 26 n.) (Lynn). -Th(ome)filio Duue, abl. (witn.) c. 1175 Wymondham, fo. 53b(i) (Gaywood). - de Willelmo f Duue 1212 P, p. 178 (Nf or Sf). -terram Thomefilii Duue de Lenn' c. 1215 Der, fo. 241a (Lynn). (ii) Radulfi Duue, gen., 1197 P, p. 246 (Nf). -Ricardus Duue 1201 CRR 25 (Pleas, iii, no. 427) (Nf). - mesuagium Thurstani Duue c. 1205 CAcre, fo. 102a(ii) (Kirtling near Bittering). -Alicia Duue 1237-1251 Holme, fo. 113a (Honing). (iii) Duua mater Stephani monachi nostri 1175-1186 Holme, no. 217, Duua uxor Rogeri Costardi ibid., no. 232 (Norwich).

A fem. name Duua is known from DB and von Feilitzen (PNDB, p. 227) took the etymon to be either an OE *Dufe, fem.(< OE *diife 'dove') or the OScand appellative dufa 'dove'. This appellative is used as a byname in both OWScand and OEScand (LindBiN, col. 66; DgP II, cols. 233-235; XenLid, p. 97). Curiously the byname seems to have been applied only to men in Scandinavia but there is a mythical ON fem. personal name Dufa which Lind, col. 204, compared with kel dufa 'to dip, to immerse' and OE dufan 'to *iahuR > *iäuR > i6r. 45 In OE personal names whose first element is formed from OE eoh the spirant [xl is normally lost (Ström, pp. 14--15). 46 For details of the dating of the hand see SPNT, pp. 129, 131-132. 47 Fora full conspectus offorms see Crawf, pp. 142-143. 48 See below, s. n. *Yri"c.

235

ivarr,lwar,lvar ON *Eyrikr, Ey-, however, seems to correspond either to ON ey f. 'island' < PrGerm *awiö- or to PrScand auja 'fortune, luck, (lucky) gift' (see above, s.n. Eymundr, and the references given there). The initial Au- in AurfkR would have been monophthongized to 0- in OSwed only in the post-Viking period (i.e. probably in the 12th century) (see Noreen, AschwedGramm, § 123) and hence it is impossible that the Thomey Eoric could represent OSwed @rfk. The most likely explanation is that the Thomey Eoric is the result of scribal error for Eiric and quite unconnected with the name of the Bast Anglian king which, as is shown above, is best explained as an Anglicized form of ON I6rekr. The variant spelling Eoric in the D MS. of the Anglo-Saxon Chronicle is also probably best explained as the result of scribal error. The D MS. 49 of the Anglo-Saxon Chronicle is written in several hands of which the oldest is hardly earlier than 1050.50 At such a late date it would be quite conceivable that a copyist could have mis-spelt the unfamiliar name form Eohric of his original text. It is significant that the correct form Eohric occurs in the earliest MS. of the Anglo-Saxon Chronicle, MS. A. No further examples of ON /6rekr have been noted in independent use in English sources. The name has been suggested as the first element of Y orfalls NRY (Yorcfal 1335), but the alternative explanation that the first element here is derived from the name of the city of York (cf. PN NRY, p. 93; SPNLY, pp. 157-158) is to be preferred.

ON ivarr, ODan Iwar, OSwed Ivar b) (i) lware diaconus (witn.) 1140-1153 Holme, no. 160 (Fibrigge, Fibrigg' in Norwich). -Yware (abl.) (witn.) c. 1170 CAcre, fo. 82b(iii) (Thorpland and Wiggenhall). de Ywara de Londele 1176-1177 P, p. 132 (Nf or Sf; Londele is unidentified). -Iwarum filium Osberti, lward' 1198 FF (PRS NS, xxvii, no. 104) (Martham). -Ywaro de Marram, abl. (witn.) c. 1200 CAcre, fo. 85b(iii) (Marham). - Ywarus Stiward', Ywar' f Wimarc', Ywar' f Sefled,Johannesf Ywar', Ywar' bedellus 1203 AssRoll 817 (NthRS, v, no. 802) (the document concems Martham). - totam tenementum cum pertinentiis suis quod fuitAgnetisfilie Ywar' in villa de Tutingtun' c. 1225 Bromh, fo. 49a(ii) (Tuttington).

The ultimate etymology of this name has been the subject of considerable debate. A. Noreen regarded fvarr as being ultimately identical with Yngvarr, the development having been conditioned by Vemer's Law according to which voiceless fricatives were voiced medially in PrGerm when the main stress did not fall on the immediately preceding syllable. This change depends on the 49

British Library Cotton Tiberius B iv.

so See ASE, p. 690.

236

ivarr, Iwar, Ivar original Indo-European system of variable accent and therefore must be dated to the early stage of PrGerm before the stress acquired a fixed position (usually on the root syllable). 51 Thus Noreen postulated a PrGerm *Inhu-harjaz with the main stress on the first element, which thus contained the voiceless fricative [x] side by side with a PrGerm *ln3u-harjaz with the main stress on the second element and, as a result, the voiced counterpart of [x], [y], in the first element. Noreen regarded the development of these forms as follows: PrGerm *Inhu-harjaz > PrScand *Ihu-harjaR > *Ihu-harR > fvarr; PrGerm *ln3uharjaz > PrScand *ln3u-harjaR > ON Yngvarr (with i > y through w-mutation), ODan Ingwar, OSwed Ingvar. Another, less likely, suggestion of Noreen postulates a development *Inn-guiRaR > *Inwarr > fvarr. 52 BrpndumNielsen took the development of fvarr to be: PrScand *Inu-gaiRaR (or InuharjaR) > *InwaRR > Iwar(r). 53 A further suggestion, made by J. Jonsson, 0. von Friesen and others, also separates fvarr from name formations in Ing(i)-, Yngv- and regards it as belonging toa PrScand *Iwa-harjaR, whose first component corresponds to OWScand yr, pl. ifar (< PrGerm *lwa-) 'yew tree, bow of yew wood' .54 ON fvarr is common throughout the medieval period in Norway but is rather less frequent in Iceland (Lind, cols. 660----663). ODan Iwar is represented by runic iuar, nom., in an inscription from the period 1150-120055 and it is extremely common in medieval Danish records from the 13th century onwards (DR, no. 74; DgP, cols. 709-716, 1683). In OSwed it is not clear whether runic iuar, nom., acc., represents OSwed Ivar or OSwed Ioarr, but OSwed Ivar is well represented in medieval and 16th century records in Sweden and Finland (SRÖg, no. 130; SRSö, no. 140; SRSmå, no. 80; SRU, no. 478; Lundgren-Brate, pp. 140-141; Modeer, Personnamn i Ktb, pp. 7, 50; Modeer, Östsmåländska bondenamn från 1500-talet, p. 63; Thors, pp. 6~5; Fredriksson, Västgötska dopnamn under 500 år, p. 183; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 124). ON Yngvarr appears as the name of several legendary personages in ON literary sources and occurs in the Landnamab6k and in Egils Saga SkallaGr{mssonar as the name of one of the early settlers in Iceland (Lind, cols. 1117-1118). The name was also borne by one of the sons of King Harald Fairhair of Norway, but it is then absent in OWScand until the early 14th century,

For Verner's Law see Krahe-Meid, I,§ 61. For Noreen's opinions about lvarr/Yngvarr see NK VII, p. 81, on which the above is based, and the bibliography given by NK VII, pp. 169---170, n. 349. 53 Brpndum-Nielsen, § 111.1. Brpndum-Nielsen takes the first element to correspond to the Gothic intensifying prefix in-. 54 See NK VII, p. 81, and the bibliography given by NK VII, pp. 169---170, n. 349. 55 For the dating of this inscription see DR, cols. 110, 843-845. 51

52

237

ivarr,Iwar,Ivar when it reappears as Ingwar, Inguar etc., being then quite well attested in late medieval ande. 16th century Norway hut rare in Icelandic documents of this period (Lind, loc. cit.; LindS, col. 816). This would suggest that it fell out of use in OWScand already in the latter part of the Viking period and that it was re-introduced into Norway in the 14th century from the OEScand area. In ODan Ingwar is extremely common and OSwed Ingvar is well attested from the Viking period onwards (DgP, cols. 639--646, 1682; SRÖg, nos. 30, 38,155; SRSö,nos.9, 105,107,108,131,173,179,254,277,281,320,335, 362; SRU, nos. 101, 111, 143, 147, 287, 307, 309, 310,363,439,478,513, 540, 644, 654, 661, 778, 837, 1032, 1068, 1143; SRVs, no. 19; LundgrenBrate, pp. 134-135; Modeer, Personnamn i Ktb, p. 7; Modeer, Östsmåländska bondenamn från 1500-talet, p. 63; Fredriksson, Västgötska dopnamn under 500 år, p. 183; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 124). Many of the runic examples relate to the Södermanland chieftain Ingvar who fell in company with many of his followers on an expedition against the Muslim territories east of the Caspian in the earlier part of the 11 th century (SRÖg, no. 155; SRSö, nos. 9, 105, 107, 108, 131, 173,179,277,281,320, 335; SRU, nos. 439,644,654,661, 778, 837, 1143; SRVs, no. 19).56 The first bearer of a name belonging to the group Yngvarr/fvarr to appear in English sources was the infamous Ivar the Boneless who was a son of Ragnarr Loöbr6k, the most famous Viking leader of the 9th century, and one of the leaders of the Danish Great Army which landed in East Anglia in autumn 865 and which proceeded to ravage England for many years after this. 57 The name oflvar the Boneless is rendered in English sources as follows: Jguuares, gen., 866, 878 (e. 11th) JEthelweard, pp. 35, 43, Ingware 870 (c. 1100) ASC (F), Iuuar 870 (e. 11th) JEthelweard, p. 36, Inwceres, gen., 878 (c. 900) ASC (A), lngwceres, gen., 878 (c. 1000) ASC (B), Inweres, gen., 878 (m. 11th) ASC (C), Jweres, gen., 878 (m. 11th) ASC (D), 878 (1121) ASC (E), Inwari, gen., 878 (11th) Asser, p. 43, Hinguari, gen., 878 (e. 12th) Annals of St Neots (Asser, p. 44). The earliest MS. of the Anglo-Saxon Chronicle, MS. A, 58 whose annals up to almost the end of the annal for 891 are written in a hand of c. 900,59 thus contains a form Inwceres, gen., which reflects a form of ON fvarr etc. in which the original nasalization of the initial f- is still apparent. 60 Forms with initial lng-, Hing- and lg- may show confusion of *[warr with ON

56 57

See E. Wessen, Historiska runinskrifter (Stockholm 1960), pp. 30-45. For the career of Ivar the Boneless see Smyth, Scandinavian Kings in the British Istes 850-

880. Corpus Christi College Cambridge MS. 173. See ASE, p. 689. 6 Cf. NPN, pp. 72-73, and also the Irish forms Imar 857 ff., Imber 12th (Marstrander, pp. 9, 67, 108). 58 59

°

238

ivarr,iwar,ivar Yngvarr, ODan Ingwar, OSwed Ingvar but they might also be merely the result of the replacement of the unfamiliar nasalized [- by the more familiar name element Ing- [itJg] on the part of English scribes. Initial Ig- in .tEthelweard's form Iguuares, gen., probably reflects an early example of the Continental spelling Igg- for [itJg] (cf. above, s.n. Ingiva/dr). Campbell, in his edition of .tEthelweard's Chronicle, interpreted this form differently, commenting upon it as follows: lEthelweard so changes (and generally improves) the forms of the names of Norse leaders seen in OEC, that it seems likely that he took some interest in the language of the invaders. For Inw:er he substitutes lguuar, which, since ig =lin the OE spelling of his time, well represents the ON form (cf. Olcel fvarr). (lEthelweard, p. lix) Campbell seems to have been unaware of the initial nasalization of the Scand name which is correctly rendered in the InwEeres, gen., of ASC (A) and hence his interpretation of the initial Ig- of .tEthelweard's form as merely reflecting Scand f- [i:] is to be rejected . .tEthelweard's form Iuuar is not examined by Campbell but it cannot be explained away as merely reflecting the later Scand fvarr in which nasalization of the initial I- has disappeared. The first two letters of the form Iuuar are still legible in a surviving fragment of the .tEthelweard MS., which was largely destroyed in the fire of the Cotton Library in 1731,61 while the last three letters are supplied by Campbell from the edition of .tEthelweard printed in 1596 by Henry Savile. 62 The surviving fragments of the .tEthelweard MS. date from the e. 11th century and Campbell considered that they may represent the first fair copy of the work. 63 It is thus reasonable to suppose that the form Iuuar probably stands for an original Jnuar (= InwEer in ASC (A)), the n of the original having been misread by the copyist as u. The form Iweres, gen., in the D and E MSS. of the Anglo-Saxon Chronicle64 is difficult to explain. The MSS. are both late, D being written in several hands, of which the oldest is hardly earlier than 1050 and E being written in one hand up to 1121.65 Björkman, NPN, p. 74, identified the form Iweres, gen., with the non-nasalized fvarr, but in view of the late date of the MSS. it might be that it represents a corruption of the original InwEeres etc. in the course of transmission. It is interesting to note that the name of Ivar the Boneless is recorded vari-

61 For the history of the MS. of ,Ethelweard's Chronicle and the various editions see Campbell, ,Ethelweard, pp. ix-xii. 62 H. Savile, RerumAnglicarum scriptores post Bedam praecipui (London 1596), fös. 472-483. 63 ,Ethelweard, p. xii. The fragments are now preserved in British Library Cotton Otho A x and Cotton Otho A xii. 64 The MSS. are: British Library Cotton Tiberius B iv (D) and Bodleian Library Oxford Laud MS. 636(E). 65 See ASE, pp. 690---691.

239

ivarr,lwar,lvar ously in medieval Danish annals and chronicles as Inguar, Ingvar, Hingvar, Inghvor, Ywar and Ywarum, acc. (DgP, cols. 639, 709). Saxo Grammaticus renders the name as lvarus and the Latin SkjQldunga Saga compiled at the end of the 16th century by the Icelander Arngrimur Jonsson from earlier texts, including two now lost,66 has the form lvarus cognomento Beinlaus, while an ON text contains the form lvarr enn Beinlausi (DgP, col. 709). OScand */varr is also represented in OE by the forms Inhwrer dux (witn.) 934 (contemp.) BCS, no. 702 (grant by King 1Ethelstan to his minister 1Elfwald of landat Derantune),67 Inwaer dux (witn.) 934 (14th) BCS, no. 703 (grant by King 1Ethelstan to St Peter's Church, York, of land in Amounderness La). 68 In DB ON fvarr, ODan iwar, OSwed Ivar is represented by a form Juuar, the name of a tenant who held at Alton and Denstone St TRE (PNDB, p. 300; DB, fo. 248b). ON Yngvarr, ODan Ingwar, OSwed Ingvar appears in DB TRE as Inguarus, Inguuarus, Inguuara, Ingarus (Ess), Inguare, lngara (lgwarus ICC) (C) and Ingeuuar (Hu) (PNDB, p. 298). J.H. Round, VCH Ess, i, p. 352, and n. 2, pointed out that these DB forms all relate to the same person. This is followed by von Feilitzen who, however, wrongly cited Round, loc. cit., as authority for the view that the Sf DB form Inuuari, gen., also belongs here (PNDB, p. 298, n. 5). In fäet, Round Kalli in ODan and OSwed see Brpndum-Nielsen, § 335.3, and Noreen, AschwedGramm, § 289.1). DR, col. 673, allows for the förmal possibility that runic kali might also reflect ODan Kalli, but points out that it is doubtful whether the assimilation of ODan Kar/i > Kalli had already taken place in the Runic Danish period (c. 800--1100). In OSwed Kali has been noted in runic inscriptions but is otherwise rare (SRSmå, nos. 11, 16; SRVg, nos. 22, 73; SRU, nos. 102,660, 708; Lundgren-Brate, p. 145; Franzen, Vikbolandet, p. 119). In England ON, ODan, OSwed Kali is also known as the first element of three L place-names in -porp (DEPN, p. 92, s.n. Cawthorpe; SPNLY, p. 160).3 There is also a Sf form Challi 1087-1098 (c. 1175-1200) Bury, p. 32 (Woolpit St). Von Feilitzen, Notes, p. 57, took this form to represent ODan Kalli, OSwed Kalle. It could, however, stand for ON, ODan, OSwed Kali with doubling of -/-. This would be most likely the result of AN orthographic practice, though it should be noted that the doubling of originally short consonants was a frequent feature in late OE MSS. whose phonetic significance is not clear (see PNDB, § 146). 3 Ekwall, however, believed that one of these three names, Little Cawthorpe (Calcewath W), which has an early form Carletorp 1205 Cur, may rather contain the personal name Kar/i (DEPN, loc. cit.).

244

Kani In Normandy ON, ODan, OSwed Kali forms the first element of two place-names in -toft (Adigard, pp. 115, 400--401). The byname Calle in the Nf form terram Hereberti Calle 1209 FF (PRS NS, xxxii, no. 217) (Scratby) does not belong here. It probably reflects ME calle, OFr cale 'a close fitting cap wom by women' used as a metonymic denoting 'a maker of cauls or coifs for the head', though ME calle 'sheepfold' used as a metonymic for 'shepherd' might also have to be considered (see DBS, p. 62, s. nn. Call, Caller).

ON (byname), OSwed Kani b) (i) Cane (abl.) (witn.) c. 1160---1170 Binham, fo. 173b (printed, SSMEA, Appendix I, no. 37) (grant by Adam filius Alueredi, grandson of Peter I de Valoignes and nephew of Roger de V aloignes, of properties in lngoldisthorpe to Binham Priory).

This name is to be etymologically connected with Modlcel, ModNorw kani 'a basin, a small wooden vessel, a type of boat', and probably ultimately had the sense 'boat', cf. the ON byname Batr < ON batr 'boat' (LindBiN, cols. 186187, s. nn. Kana-Barör, Kani). In OWScand Kani is attested as a byname in late medieval Norway, first appearing in a document of c. 1282 (LindBiN, cols. 186-187, s.n. Kani). In OSwed Kani appears as a personal name in two runic inscriptions and is also known from place-names (SRSö, no. 69; SRV g, no. 152). In addition to the above Nf example, the name has been noted in England as the name of a DB tenant who held TRE in Sx and Sr (PNDB, p. 213 and n. 2). It is also contained in the form Willelmus filius Cane c. 1213 Fees (Brk), though Reaney, DBS, pp. 63-64, s.n. Cane, regarded all these English examples as probably reflecting an OE *Cana. An OE *Cana has also been suggested as the first element of several English place-names and field names (DEPN, pp. 85-86, s. nn. Canfield, Canford, Canley, Cannings, Canwell, Canwick; PN Ing (2nd ed.), pp. 47-48; PN D, pp. 390, 399, 649, 682; PN Sr, pp. 230,353; PN Ess, pp. 148, 179-180, 472-473, 571; PN Wa, pp. 181,318; PN W, pp. 249-250, 418; PN Nt, pp. 104, 272; PN 0, pp. 308,477; PN Gl, iii, pp. 80, 142, 227; PN Gl, iv, p. 205; PN Brk, pp. 537, 920).4 The evidence usually cited in favour of the existence of an OE *Cana is that of the DB personal name forms from Sr and Sx (DEPN, p. 85, s.n. Canfield; PN Ing (2nd

4 The editors of PN Nt considered that the early forms ofGanabridge Wood Nt are too uncertain to allow any interpretation, but they noted the suggestion of Bruce Dickins that the first element is the personal name Cana as well as the opinion of Ekwall, that the base of the place-name is an OE gamen-brycg 'game-bridge', i. e. where such are held. (PN Nt, p. 104).

245

Kani ed.), p. 48; PN Ess, pp. 179-180; PN Gl, iii, p. 227; PN Brk, pp. 537, 920). 5 OE *Cana, however, is etymologically obscure and is not attested in pre-Conquest records. The DB and post-Conquest examples of Cana, Cane etc. in English sources are thus better explained by the Scand Kani which, though rare, is etymologically unproblematic. The distribution of these forms is no argument against a Scand origin since by the late OE period Scand personal names were also encountered outside the Danelaw. Von Feilitzen did not consider an OE *Cana as a possible etymon for the DB Cana, Cane etc. but commented on these forms: Perhaps as Redin suggests (pp. 87 f., 133), from ON Kani (byn), OSw (run) Kani. Connection with OEcanne 'can, cup', whichmay well have been usedas anickname, seems less likely in view ofthe persistent single n in the DB forms(§ 147). (PNDB, p. 213).

For several reasons, however, it is not possible to postulate the Scand name as the etymon of the place-names and field names mentioned above. In the first place, all these place-names and field names, with the exception of the doubtful Ganabridge Wood Nt (for which see above, n. 4), lie well outside the Danelaw. Secondly, none of these names hasa second element of Scand origin. Without exception these place-names and field names all have OE second elements, including one with the archaic suffix -ingas (Cannings W) which had long ceased to be a living place-name element at the time of the Scand settlements in England. OE *Cana is, however, open to question. It is somewhat curious that a personal name whose etymology is obscure and which is not independently attested in OE records should be postulated as the first element of such a relatively large number of English place-names. A possible alternative here would be the appellative OE canne f. 'can, cup'. A.H. Smith regarded OE canne as the base of Canwell St and Canford Do (EPN I, p. 80, s. v. canne), 6 names which were taken by Ekwall, DEPN, p. 85, s. nn., to contain an OE personal name *Cana. There are, however, cases in which the appellative canne is unlikely on topographical grounds, e.g. PN Wa, p. 181, takes the first element of Canley Wa to be the personal name *Cana rather than the appellative canne because Canley is situated on a well marked hill rather than in a hollow. It is noteworthy that in certain EPNS volumes the personal name is

5 A.H. Smith, referring to PNDB, p. 213, thought that the DB personal name 'may be of ON origin', though he seems to have had no doubts about the existence of an OE *Cana (PN Gl, iii, p. 227). Smith also incorrectly remarks that Cana occurs in DB in Sf (loc. cit.). Ekwall cited as a parallel to OE *Cana a ContGerm *Cano which he took to form the first element of the Flemish place-name Canegem (Caningahem 967) (DEPN, p. 85, s. n. Canfield; PN Ing (2nd ed.), p. 48). Mansion, pp. 27, 32, 34, took the first element of Canegem to be a personal name Cana, but he regarded this personal name as 'uiterst twijfelachtig' (ibid., p. 34). 6 Smith interpreted Canwell St as 'a well provided with a can' but took canne to have a topographical extension of meaning to 'a depression, a hollow, a deep valley' in the case of Canford Do (EPN I, loc. cit.).

246

Kani rendered as Canna (PN 0, p. 390) and Can(n)a (PN 0, pp. 399, 649, 682; PN W, p. 418). An OE *Canna is to be regarded as a weak masc. name derived from the appellative canne. PN W, p. 250, comments on the etymology of Cannings W, the first element of which is given by PN W, p. 418, as the personal name Can(n)a:

Caningas would seem to denote 'the people of Cana' (v. ingas and PN in -ing 69). The name is on record in DB in the forms Cana, Cane, Cano, Canus (Feilitzen 213), but the history of the name is obscure. Persistent single n prevents association with OE canne 'can, cup'. As is pointed out above, however, the OB forms represent the Scand personal name and byname Kani and they should be left out of any discussion of the etymology of Cannings W. It is true that the majority of the early forms of Cannings W given by PN W, pp. 249-250, have single n, but it is not clear that this prevents association with OE canne (or with a personal name *Canna). Simplification of double consonants can occur in OE in formative elements with geminates, and the graphical simplification of double consonants is an occasional orthographic usage in the OE period and is frequent after the Conquest as a result of the uncertainty of AN scribes as to the quantity of OE consonants (Campbell, §§ 66,457; PNDB, § 147; SPNLY, § 143).7 Probably many of the place-names and field names which have been taken to contain an OE *Cana contain the appellative OE canne, though in cases where this is topographically unlikely the etymologically related OE personal name *Canna might be suggested. The etymologically obscure personal name *Cana can thus be dispensed with. There is also a byname Kana, Kane, Chane, Cane of which the following forms have been published: Leofwine kana 11th L VH, Lewinus Chane 1066 LibWint (Ha); Herueus Cane 1177 P (St); HugoKane 1210 P (He); William le Cane 1332 SR (Sx) (Tengvik, pp. 377-378; von Feilitzen, PNWinton, p. 209 and n. 3; OBS, pp. 63-64, s.n. Cane). Tengvik (loc. cit.) derived the LVH and LibWint forms from the Scand name, remarking that the single n rules out derivation from OE canne which would otherwise have been a suitable nickname. Reaney (OBS, p. 64, s.n., took the view that "the nickname is ME, OFr cane 'cane, reed', used probably fora man tall and slender as a reed". Von Feilitzen (PNWinton, p. 209) considered that in the LVH and the LibWint the byname may reflect the OE appellative canne or, in view of the single n, possibly the Scand appellative kani to which the personal name and byname Kani belongs. He regarded Reaney's explanation to be unlikely in the case of the L VH and LibWint examples (ibid., p. 209, no. 3). 7 Fora possible I. OE reduction of long consonants in polysyllabic words see Karlström, NoB, xxii (1934), pp. 73 ff.

247

*Karboi The Scand personal name and byname Kani also forms the first element of Cana Barn NRY (Kanehou 1202 FF) (PN NRY, p. 185; SPNLY, p. 160). The spelling c for /k/ before back vowels as in Cana, Cane etc. is anormal feature ofME orthography (Jordan, § 178.1; Brunner, AbriB, § 38.1). Final -e for Scand -i in the Nf form Cane is the result of Anglicization (PNDB, §§ 45, 150; SPNLY, § 147).

AScand *Kärboi a) (i) Carboistorp 1086 DB, fös. 251, 274, Karboistorp 1198 FF (PRS NS, xxvii, no. 34) (lost, Clackclose H; O.K. Schram, Fenland Place-Names, p. 439, n. 4, identified this place with Shouldham Thorpe, but he did not state his reasons for doing so).

O.K. Schram, loc. cit., took the first element of this place-name to be an ODan *Karboie. An ODan *Kärboie would correspond in type to Kärtöki, a name which has been noted in Runic Swedish and in English sources (SRVg, no. 180; NPN, p. 78; below, s.n.). The first element of Kärtöki has the character of a byname and has been connected with the appellative ON karr m. 'curl, curls in the hair' (SRVg, p. 324). ON karr does not appear as an adjective but has been noted as the second element of various adjectival compounds, in some cases with the sense 'curly' (gullkarr, punnkarr (byname)), in others with the sense 'intense, quarrelsome' (afkarr, launkarr, svipkarr) (Kousgård S!1}rensen, Bebygg på -sted, p. 259; cf. also Kousgård S!1}rensen, NoB, lxii (1974), pp. 112-113). Liden considered that both these meanings were developments of an original meaning 'bent, curved, crooked' .8 An adjective *karr has been taken to form the second element of the Danish personal names Odinkar and Styrkar, and the extremely common personal name Kari also belongs to this word group (Kousgård S!1}rensen, loc. cit.; for Kari see also below, s.n.).9 The second element of Kärtöki is the extremely common OEScand personal name Töki. SRVg, p. 324, takes Kärtöki to denote 'the curly haired Töki', but it could also have the sense 'the vehement, the quarrelsome Töki'. In this context it is important to note that Kousgård S!1}rensen, Bebygg. på -sted, p. 259, casts doubt on the traditional interpretation of the personal name Käri as

8 lndogermanische Forschungen, xxx (1906), p. 341; Armenische Studien (Göteborgs högskolas årsskrift, xii, 1906), p. 111. 9 The derivation of -kar in ODan Othinkiir from an adjective *Karr has been questioned by H. Andersen, NoB, xxxii (1944), pp. 101-110. Cf. also Hald, Personnavne i Danmark: I. Oldtiden, p. 39. Kousgård Sfl)rensen, NoB, lxii (1974), pp. 108-116, interprets Odinkar not as a normal dithematic personal name but takes it to belong to an adjective ODan *öthomkiir, perhaps meaning 'inclincd to rage or madness'. Kousgård Sörensen (ibid., p. 112) also regards Styrkar as an originally adjectival formation with the sense 'the warlike one' or, strictly speaking, 'the quarrelsome'.

248

Kåri, Kari, Kare

an original byname denoting 'the curly haired one' and prefers to connect it with the meaning 'vehement, quarrelsome'. An ODan *Karboie would be a similar formation to Kartöki and would denote 'the curly haired Boie' or 'the vehement, the quarrelsome Boie'. ODan Boie, however, which first appears in Danish after the Viking period and which is mainly confined to Schleswig, is not native to ODan but is a loan from the south corresponding to OSax Böio, MLG Boyo, Boye (DgP, cols. 140-141). An ODan *Karboie can thus be ruled out on chronological grounds as the first element of the Nf Carboistorp, Karboistorp. This place-name contains as its first element an AScand formation *Karboi. The first element of this AScand *Karboi is clearly identical with that of the Runic Swedish Kärtöki, while the second element belongs to OE *boi, ME boi, boy, a strong side form of OE *boia, ME boie 'servant, underling, churl, boy' and the ancestor of ModE boy. 10 Cf. Aluuinus boi TRE DB, fo. 31 (Sr) and the following ME forms: Walterus Helleboie 1200 P, Roberti Litelboie (gen.) 1205 LeiRec, Richard Lateboy 1235 Ch, Agatha uxor Willelmi Godeboye 1269 Cl, Roger Jolifboie 1319 Pat (PNDB, p. 205; MED, s. v. boie (5)). In the ME byname forms ME boi(e), boy(e) is compounded with OE (Kt), ME (SE) hell 'hill', OE lyte!, ME titel 'little', OE !fet, ME late 'late', OE göd, ME göd(e) 'good' and ME (< OFr)jolif'gay, lively', respectively. An AScand *Kärboi fits into this pattem and can be taken to be a byname denoting a boy or servant with curly hair or a vehement or quarrelsome boy or servant. Cf. also the OE form Maneboia 963-992 (m. 12th) ASChR, no. 40 (Nth). This is, however, a formation of a different type, for, as is pointed out by Coinage of Edgar, p. 190, n. 4, it may be ON Mani or ODan Manni with the added byname boia. OE Manna is also a possible etymon for the first element.

ON Kari, ODan Käri, OSwed Käre a) (i) Karewelle c. 1190 Wals, fo. 8lb(i, ii) (f. n. in Clipstreet in Bale). - Karehowe e. 14thBinham, fo. 176a, Carhowe ibid., fos. 177a, 177b (f. n. in Toftrees). b) (i) terra que fuitHerberti Kari c. 1180--1190 BodlNjCh, no. 587 (Wramplingham). Walterus Cari 1199 P, p. 287, 1200 P, p. 144 (Nf or St).

ON Kari etc. is an original byname denoting either 'the curly haired one' or 'the vehement, the quarrelsome'. Fora discussion of the etymology see above, s.n. *Kärboi and the references given there. 11 Kari was frequent throughout

For ME boi(e) etc. see Coinage ofEdgar, pp. 189-191, and the literature cited there. For further discussion and references see Brl')ndurn-Nielsen, § 70.1, Anm. 2, and NK VII, pp. 51, 153-154, Il. 116. 10

11

249

Karl(i), Karl(e)

the Viking and medieval periods in both Norway and Iceland (Lind, cols. 675-676; LindS, cols. 542-547). In ODan and OSwed the name is found from the Viking period onwards but is relatively uncommon (DR, no. 287; DgP, cols. 724---725; Kousgård S!llrensen, Bebygg. på -sted, p. 259; SkOA XVI, p. 111; SRÖg, no. 81; SRSö, nos. 217,298; SRU, nos. 16, 37, 172 (?), 259,532, 683, 797, 866, 956, 1146; Lundgren-Brate, p. 147; Modeer, Östsmåländska bondenamn från 1500-talet, p. 66; Thors, p. 66). A notable Swedish runemaster of the period from about 1020 to the middle of the 11 th century was Asmundr Kärasun whose name appears in 20 runic inscriptions from Uppland and Gästrikland (KLNM, xiv, cols. 497- 498). Kari is also attested as a byname in medieval Norwegian, Danish and Swedish records (LindBiN, col. 188; DgP Il, cols. 540-542; Modeer, Personnamn i Ktb, pp. 69, 71). Note also the strong side-form ON Karr, OSwed (run.) Kärr, KärR, for which see below, s.n. In England ON Kari etc. appears in the form Carig as the name of a London moneyer of the period 1009-1016 and is recorded in DB as Cari, the name of a TRE tenant in Lei (Smart, Moneyers, p. 254; PNDB, p. 301). A further English example of the name is contained in the form Care Tokies suna (dat.) 1045-1046 (12th) ASChR, no. 105 (Holcombe Rogus D). 12 A form Kare has also been noted from an L charter of the time of Hy2 which is preserved in a copy of 1407 (SPNLY, p. 161). ON Kari etc. also occurs as the first element of place-names and field names in L and Y and of a field name in Db (DEPN, pp. 87, 88, 92, s. nn. Careby, Carthorpe, Caythorpe YE; SPNLY, p. 161; SSNY, p. 56; PN NRY, p. 226; PN ERY, p. 99; PN WRY, vi, p. 164; PN Db, p. 762).

ON, ODan Karl(i), OSwed Karl(e) b) (i) Alurici Carlessone terra 1138-1147 YCh, vii, no. 41 (Hockwold).

This name is an original byname belonging to the appellative ON, ODan, OSwed karl m. 'free man, fellow, peasant' (NK VII, p. 43; Hald, Personnavne i Danmark: 1. Oldtiden, p. 62; Otterbjörk, Svenskaförnamn, p. 106). The weak Karli is probably merely a hypocoristic development of the strong Karl (Hald, Personnavne i Danmark: 1. Oldtiden, p. 71). ON Karl, Karli is well attested in Norway but is rather infrequent in Iceland (Lind, cols. 676-677, 677-678; LindS, cols. 547-554). ODan Karl is well attested in medieval Danish records (DgP, cols. 725-729, 1683) and there

12

Fora full commentary to this charter see ASChR, pp. 447-449, and Sawyer, no. 1474.

250

Kårr,Karr,KarR area few examples of Karli in 15th century Danish records (DgP, cols. 729730). In OSwed Karl is well represented in runic inscriptions and is common in medieval and 16th century records in Sweden and Finland (SRÖg, nos. 44, 234, 235, 237; SRSö, nos. 205, 274, 286, 298; SRSmå, nos. 61, 71; SRVg, nos. 42, 112; SRU, nos. 36,306,311,317,321,384,689,885,901,904,906, 934,985, 1060, 1080, 1115; Lundgren-Brate, pp. 147-148; Modeer, Personnamn i Ktb, pp. 27, 31, 32, 101, 102, 119, 125, 128; Modeer, Östsmåländska bondenamn från 1500-talet, p. 65; Thors, pp. 66-67; Fredriksson, Västgötska dopnamn under 500 år, pp. 183, 185; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 127-128). The weak Karli, Karle is rare in Swedish sources, though it is already found in Runic Swedish (SRÖg, nos. 44, 201; SRVg, no. 137; Lundgren-Brate, pp. 148-149; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 128). In England several examples of OScand Karl and Karli occur in the 11 th century, but both these names are rare in later English records (NPN, pp. 7678; PNDB, pp. 301-302; SPNLY, pp. 161-162; Smart, Moneyers, p. 263; von Feilitzen, SCBI, xi.i, p. 12). Smart, Moneyers, p. 263, needlessly gives 'OG' Carl as a possible alternative etymon for Carla, the name of an Exeter moneyer of JEthelred II. This moneyer's name is correctly taken by von Feilitzen, SCBI, xi.i, p. 12, to represent the Scand Karli. A Sf example of OScand Karli is contained in the form Edmundum filium Carle 1205 CRR 39 (Cur, iv, p. 1) (Ipswich). The Lat form Carolus, Karolus, which occurs in East Anglia in the e. 13th century in the forms Carolus, Karolus 1208 Cur (Nf, St), Karolus filius Gerberge 1210 Cur (Nf), Carolus filius Willelmi 1212 Cur (Nf), does not belong to the Scand Karl, Karli but corresponds to OFr Charles, OHG Karl (DBS, p. 70, s.n. Charles).

ON Karr, OSwed (run.) Kärr, KärR a) (i) Karst', Karston', Kareston' 1198 FF (PRS NS, xxvii, nos. 77, 165, 166), Corestone 1254, Corston 1291, 1316 DEPN, p. 124, s.n. Coston (Coston, Forehoe H).

ON Karr, OSwed (run.) Kärr, Kärn isa strong side-form to ON Kari etc. (NK VII, p. 51; SRVg, p. 122). ON Karr is known in both Norway and Iceland as a personal name from the Viking period onwards and has been noted as a byname in Norway from the end of the 13th century onwards and possibly also in an Icelandic document from the earlier part of the 15th century (Lind, cols. 679-680; LindS, cols. 554-555; LindBiN, col. 190). In Sweden Kärr, Kärn has been noted in runic inscriptions from Södermanland, Västergötland and Uppland (SRSö, no. 128; SRVg, no. 73; SRU, nos. 643,644,654, 792). The 251

Kärtöki apparent absence of an ODan counterpart of ON Karr, OSwed (run.) Kärr, Kärn is noteworthy. 13 In England ON Karr etc. has been taken to form the first component of the name of the ancient dike Car Dike in L and Nth and has been noted as the first element of field names in Cu and Db (PN Nth, p. 5; PN Cu, p. 341; PN Db, p. 762). 14

OSwed (run.) Kärtöki a) (i) Kertokeswong 14th Terr (f. n. in Syderstone) (OKS).

For the etymology of this name and its appearance in Runic Swedish see above, s.n. *Kärboi. English examples of Kärtöki areas follows: Kartoca ure broöor 1016--1018/1019 (contemp.) Ker, p. 317, no. 247 (list of persons in confratemity with Christ Church Canterbury), Kartoca minister (witn.) 1024 (contemp.) KCD, no. 741 (grant by Cnut of landat Portisham Do). If Kärtöki does form the first element of the Nf field name Kertokeswong, then OScand ä in this personal name must have undergone 1. OE shortening and then subsequently the OFr interchange of a with e before r (and r + consonant). 15 Altematively, however, it might be suggested that the first element of this field name represents Scand *Kjarr-T6ki. Here the first component, which would have had a byname function, belongs to ON kjarr n. 'brushwood', ODan kiar n. 'bog, marsh', while the second is the OEScand personal name Töki. A Scand *Kjarr-T6ki would denote 'T6ki who dwells at the brushwood' or 'T6ki who dwells at the marsh'. The spelling e for the Scand diphthong ia (< e through a-fracture) in the Nf field name Kertokeswong reflects the normal development of this diphthong in English, cf. ME ker 'a bog, a marsh, esp. one overgrown with brushwood' < ON kjarr, ODan kiar. 16 Cf. also the form Thor Muntokes sune? 17 1. 11th LVD, fo. 51b. Björkman, NPN, p. 97, interpreted Muntoke as belonging toa Scand *Munn-T6ki 'T6ki

An ODan *Kärr is not recorded in DgP or in DgP Il. Ekwall, p. 86, s. n., includes the possibility that the first component of the name of the Car Dike may altematively be ON Kari etc. 15 For I. OE shortening of long vowels in stressed syllables before groups of two or more consonants see Luick, § 352; Jordan, § 23; Brunner, AbriB § 9.1; Mosse, § 21. For the OFr interchange of a with e before rand r + consonant as manifested in English sources see PNDB, § I; SPNLY, § 2. 16 For the development taken by the Scand diphthong ia in I. OE and ME see Luick, § 382.5, PNDB, § 44, SPNLY, § 54. 17 Björkman, NPN, p. 97, gives the form as Thor ve[ Siward Muntokes sune, but though vel Siward is interlined above the form Thor Muntokes sune there is no real proof that it belongs to this form and for this reason I have preferred to omit it. 13

14

252

Kaupmaör, Köpman

with the (big) mouth' and compared the OWScand byname Munnr < ON munnr m. 'mouth'. 18

ON Kaupmaör, ODan (byname), OSwed Köpman a) (i) Copemannescrofthill I. 13thLangley, fo. 45a (f. n. in Hales). -Copmansmerefurlong 1580 Ct (f. n. in Freethorpe) (OKS). b) (i) Copmannus Clokersuo 1146 Holme, no. 142 (Mancroft in Norwich). - Copmanno de Fonte, abl. (witn.) c. 1160 CAcre, fo. 87b(i) (grant by William of Wormegay to Castle Acre Priory of the church of Westbriggs and of rents from the tenure of William son of Reiner of Shouldham and from the mills of Bawsey and Hadduna). - Eustacius filius Copmanni 1176-1177 P, p. 133, 1177-1178 P, p. 24, 1179-1180 P, p. 18, 11801181 P, p. 84, Eustaciusfilius Copman' 1178-1179 P, p. 5 (Nf). -IEdwardumfilium Copman 1181-1182P, p. 72, 1182-1183 P,p. 12, 1183-1184P,p. 6, Tomas et/Edwardusfilius Copman 1181-1182P, p. 72, Tomas et/Edwardusfilii Copman 1182-1183 P, p. 12, 1183-1184 P, p. 6, 1184-1185 P, p. 34, 1185-1186 P, p. 61, 1186-1107 P, p. 53, 1192 P, p. 181, 1193 P, p. 16, Tomas et /Edward' filii Copman 1191 P, p. 36, Tomas et IEdwardusf Copmanni 1194 P, p. 49, 1195 P, p. 65, 1196 CR, p. 126, Tomas et Edwardusfilii Copmanni 1197 P, p. 227, Tomas et Ebrardusfilii Copman 1187-1188 P, p. 56, Tomas et Ebrardus filii Copmanni 1190 P, p. 94 (Norwich). - Copmannus auunculus ipsius Thome (i.e. Thomas faber) 1210 FF (PRS NS, xxxii, no. 249) (Holt). (ii) Eustaciusf Johannis Copman 1205 P, p. 251 (Nf).

This name belongs to the appellative ON kaupmaör, ODan, OSwed ki)pman m. 'merchant'. The personal name Kaupmaör, -mann occurs in Norway from the late 14th century onwards but is infrequent (Lind, cols. 682-683). There are also a few Norwegian examples of the use of ON kaupmaör as a byname, e.g. in the name of Biorn kaupmaör, a son of King Harald Fairhair (LindBiN, col. 192). ODan ki/Jpman is quite well attested as a byname in 15th century Danish records and there are a few examples of a personal name Ktj)pman in medieval Swedish sources (DgP Il, col. 644; Lundgren-Brate, p. 163). In England ON Kaupmaör etc. appears as the name of a 10th century moneyer in the form Copman (NPN, p. 86; Coinage ofEdgar, p. 192). In the postConquest period it is not common in England, but it appears as a personal name in L records from the end of the 12th century and the beginning of the 13th century and has been noted as a byname in a Y charter of c. 1160-1170 (13th) (SPNLY, pp. 163-164). It should be noted that it is formally possible that this Y byname and the Nf byname given above sub (ii) belong to the appellative kaupmaör rather than to the personal name. ON Kaupmaör etc. might also form the first element of certain Y field 18

For the OWScand byname Munnr see LindBiN, col. 262.

253

Ketilbert names though in all these cases the appellative is equally possible (PN WRY, vi, p. 138; SPNLY, pp. 163-164). For the use of o for OScand au see above, s.n. Auöhildr, and the references given there. It should be noted that the diphthong is preserved in the spelling ou in some of the Land Y forms given by SPNLY, pp. 163-164.

Scand-West Frankish hybrid Ketilbert b) (i) de catallis Ketelberti 1176---1177 P, p. 138 (Nf or Sf).

This name would appear to be a compound of OScand Ketil- and West Frankish -bert (cf. OHG -beraht, -ber(h)t etc.). Cf. also Porbert (below, s.n.). Forms in Chetelbert(us) occur in DB TRE in Hu, Nth, Wo and Y and there are several examples of the Latinized Ketelbertus, Chetelbertus in post-Conquest sources in Land Y (PNDB, pp. 302-303; SPNLY, pp. 170-171). Some of these forms, however, may stand for AScand Ketelbe(o)rn < ON KetilbiQrn etc. with confusion of the elements -bern and -bert caused by the occasional loss of final -n and -t in AN records (see PNDB, loc. cit.; SPNLY, § 86.iii; cf. also above s.n. AsbiQrn for examples of the interchange of Osbern and Osbert in post-Conquest records). An example of such an interchange is provided by the following forms given by SPNLY, p. 170: Ketelbern, Ketelburne de Keles 1212 Fees, Ketelbern de Keles 1218-1219 AssRoll, Ketelbert de Westkel, Keles 1218, 1219 FF, Ketlebern' militem 1220---1234 RotHug, Ketelbern de Keles, Kaleys 1226, 1234 FF, Ketelbertum de Keles militem 1237 RotRob, domino Ketelberto de Kal' (witn.) 1240---1257 RA (see PNDB, p. 303, n. 1). Note also the following: Ketelbern radman, Ketelbert radman TRE DB, fo. 174b (Wo); Chetelbernus TRE DB, fo. 24Ob, Chetelbertus ibid., fo. 239, Chetelberto, abl., ibid., fo. 241b (Wa) (PNDB, p. 304). 19

For the forms Chelbert(us), Chilbert(us) which almost certainly do not belong here see below, s.n. *Ky/fu-vQrÖr.

ON Ketilbi9rn, ODan Ketilbiorn, OSwed Kretilbiorn b) (i) Chetelbern fiber homo TRE DB, fo. 149b (Baconsthorpe in Attleborough). Ketelbern 1165-1166 P, p. 28, Ketelbern' 1166---1167 P, p. 28 (Lynn) - de catallis Ketelbern' 1176---1177 P, p. 137 (Nf or Sf). - Ketelbern 1222 ElyA, fo. 169a (Walpole).20 - terram Radulfifilii Ketelbern c. 1230---1240 Der, fo. 238a (Clenchwarton), de

19 The Wa entries refer to Ketelbern son of JEoelwine and brother of Turchil of Warwick (PNDB, p. 304, n. 4). 20 This part of ElyA is printed by SSMEA, Appendix Il, no. 1, p. 263.

254

KetilbiQrn, Ketilbiorn, Kretilbiorn hornagio Radulfi filii Ketelbern de Northclenchewartun', homagium Radulfi filii Ketelbern de Northclenchewardtun c. 1235 Der, fo. 226b(ii) (in campa de Sutlen, i.e. South Lynn; Northclenchewartun', Northclenchewardtun is North Clenchwarton), Rodulfo filio Ketebern de Clenchewardtun et heredibus suis, dat., Radulfo filio Ketebern de Clenchewardtun, dat., ibid., fo. 227b(i) (parish of St Peter of Lynn), Carta Radulfi filii Ketelbern de Clenchewardtun, Radulfus filius Ketelbern de Clenchewardtun ibid., fo. 233b(ii) (properties in South Clenchwarton, Saddlebow and Wiggenhall), Radulfo filio Ketelbern de Clenchewardtun, dat. (bis) ibid., fo. 235a(ii) (parish of St Peter of Lynn), de hornagio heredum Radulfi filii Ketelbern de Clenchewardtun, homagium heredum Radulfi filii Ketebern de Clenchewardtun c. 1240 Der, fo. 228b(ii) (Clenchwarton), Carta Rogeri filii Radulfi filii Ketelbern de Clenchewardtun, Rogerus filius Radulfi filii Ketelbern de Clenchewardtun c. 1250 Der. fo. 235a(i) (Pikenhamholmes in North Clenchwarton).

Personal names containing Ketil- as first element are rare in Norway and Iceland but are well attested in OSwed both in runic inscriptions and in medieval records (Wessen, Nordiska namnstudier, p. 107). ON KetilbiQrn appears sporadically in Iceland from the time of the settlement onwards and is represented in Norway by a few examples in records of the period 1352-1438 (Lind, cols. 683--684). ODan Ketilbiorn is also extremely infrequent (DR, no. 381; DgP, cols. 745-746). In OSwed Ktetilbiorn, Ketilbiorn, -birjJrn was frequent in Uppland but rare in western Sweden and not particularly common in Södermanland, Västmanland and Östergötland (SRSö, nos. 226, 229; SRU, nos. 585, 594, 637, 846, 961, 977, 1023, 1047; Lundgren-Brate, pp. 159-160; Modeer, Personnamn i Ktb, pp. 131-135; Modeer, Fornsvenska personnamn i en regestsamling, p. 13; Modeer, Östsmåländska bondenamn från 1500-talet, p. 59; Thors, p. 68; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 130). In England ON KetilbiQrn etc. forms the first element of Kettlebaston Sf and of certain minor names in WRY (DEPN, p. 274, s.n. Kettlebaston; PN WRY, ii, p. 107; PN WRY, iv, p. 78; SPNLY, p. 170). The name is also quite frequent in independent use in England both within and outside the Danelaw: NPN, pp. 79-80; ZEN, p. 54; PNDB, p. 304; SPNT, p. 143; ELPN, p. 78; DBS, p. 204, s.n. Kettleburn; SPNLY, pp. 170-171; Smart, Moneyers, p. 235. Cf. also the following English forms: Cytelbearn 963-984 (m. 12th) ASChR, no. 40 (surety in a transaction conceming landat Wittering Nth); Ketelbernus 1161-1177 (13th) Rams, iii, p. 260 (Elton Hu), Ketelburn de Stanch' ibid., p. 302, Ketelbernus ibid., p. 303 (Cranfield Bd; Stanch' is unidentified); De Tomafilio Ketelbern 1175-1176 P, p. 59 (Sa); Rogerus filius Ketelbern' 1176-1177 P, p. 190, 1178-1179 P, p. 36, Rogerusfilius Ketelberni 1177-1178 P, p. 90 (Sx); RobertumfiliumKetelbern 1200 CRR 20 (Cur, i, p. 139) (Lei); Ricardusfilius Ketelbern? e. 13th (m. 14th) Rams, i, p. 93, no. 282 (Upwood Hu); toftum Ketelberni 13th ? John or Hy3 Blyth, no. 300 (Great Houghton WHY); Margaretam filiam Ketelbern 1203 CRR 31 (Cur, ii, p. 280) (C); Ketelbern antecessor ejus (William son ofWilliam son of Godric), de illo Ketelbern 1211 CRR 54 (Cur, vi, p. 138) (Hopton WRY; claim by William son of William son of Godric against Alexander de Crevequer that William's antecessor Ketelbem bad held a carucate of landat Hopton

255

Ketill, Ketil, Kreti) WRY on the day on which Henry I had died, i.e. 1 December 1135); Johannem Ketelbern 1221 AssRoll 1021 (Eyre I, no. 949) (Worcester); Ketelbernus de Knichteton' pater Ricardi, Ketelbern 1221 AssRoll 1021 (Eyre I, no. 1023) (Knighton-on-Teme Wo; the assize had to decide, inter alia, whether this said Ketelbern had died after the first coronation of Henry Il, i.e. December 19 1154); Walterum Ketelbern 1221 AssRoll 948 (Eyre III, no. 360) (Wa); Alicia que fuit uxor Ketelberni de Edreston', Ketelbern vir suus, Ketelbern, hereditas Ketelberni viri sui, Agheuillam que fuit uxor ipsius Ketelberni, Siwardus pater Ketelberni, eidem Ketelberno, dat., 1221 AssRoll 948 (Eyre III, no. 715) (Edstone Wa); Simon Ketelbern 1244 (m. 14th) Rams, i, pp. 452,453 (Husborne Crawley Bd), Richerus21 Ketelbern c. 1250 (m. 14th) Rams, ii, p. 9 (Cranfield Bd).

The spelling Chetel-, Ketel- for Scand Ketil-, Kreti!- is due to the influence of the etymologically identical appellative OE (Angl) cetel m. 'kettle', while Cytel- in the OE form Cytelbearn is a direct reflection of WSax scribal practice, since it shows the modification of Scand Ketil-, Ka:til- through association with the specifically WSax cytel m. 'kettle' (see above, s.n. Äskell). Initial Ch- in the DB form Chetelbern represents the normal e. AN spelling for /kl before front vowels (PNDB, § 113; SPNLY, § 126; von Feilitzen, PNWinton, p. 225). For the treatment of the second element, ON -biQrn etc., in English sources see above, s.n. ÄsbiQrn.

ON Ketill, ODan Ketil, OSwed Kretil a) (i) Ketlestuna 1086 DB, fo. 168b, Ketleston' (p) 1199 Justiciar's Writ (Pleas, i, no. 3491), 1209 AssRoll 558 (Pleas, iv, no. 4046), Keteleston' (p) 1209 AssRoll 558 (Pleas, iv, no. 4410) (Kettlestone, Gallow H). - Ketelesmerewang 1146-1148 HarlCh 47 H 45 (printed, SSMEA, Appendix I, no. 4), Ketellesmerewang 1146-1148 Coxford Il, fo. 2a(ii), Ketelmerewang ibid., fo. 2b(i), c. 1180--1200 Coxford Il, fös. 3a, 4a, 22 c. 1200 Coxford Il, fo. 3a(i) (f. n. in East Rudham). - Ketelishil c. 1195 Der, fo. 113b(ii), Ketishil, Keteshil c. 1235 Der, fo. 115a(ii), Ketelhil 1239-1243 Der, fo. 12lb(i) (f. n. in East Walton). - Ketelesgraue c. 1210 W als, fo. 64a(ii) (f. n. in Hindringham). - Ketelesdiks c. l220-l250Hales MS. (f. n. in Kirby Cane) (OKS). -Ketellondm. 13th Coxford Il, fo. 46b, 46b(i), Keteleslond (bis) ibid., fo. 51a(iii) (f. n. in Tattersett). -Ketelesholm c. 1250 Der, fo. 76b (f. n. in Watlington). - Ketelesgate 1300 NRS I (f. n. in Morston) (OKS). - Keteleswong 14th Blackborough (f. n. in Middleton) (OKS). - Ketelishagh 1314, 13351pm, Ketelyssale 13161pm (f. n. in Hingham). -Ketelescroft 1321 Ct (f. n. in Hindringham) (OKS). b) (i) Ketel (nephew ofThurketel Heyng) c. 1020--1050 (l. 13th) ASWills, no. 25 (Ketel

Corrected in the MS. from Ricardus (Rams, ii, p. 9, n. 3). The Coxford Il, fo. 4a entry cannot be later than the death of Earl Hamelin de Warenne in 1202. 21

22

256

Ketill, Ketil, Kretil was bequeathed pat !ond pat Omund ahte at Ormesby). 23 - Kytel, Kytele, dat., 10421053, possibly 1046 (13th) ASWills, no. 32 (Ketel and his brother Wulkitel are here bequeathed Nf properties at W alsingham in East Carleton, East Carleton and East Harling by their mother Wolgip), 24 Ketel (bis), after Keteles day m. 11th (l. 13th) ASWills, no. 33 (in this document, the will of Ketel's uncle Eadwine, it is stipulated that Ketel was to succeed to an estate at Ashwellthorpe after the death of Eadwine and his brother Wlfric), Keteles guide 1052-1066 (l. 13th) ASWills, no. 34 (Ketel's will; he disposes of property in Nf at East Harling, Great Melton, Ashwellthorpe, Ketteringham, Hainford, Str(a)ttune and Moran as well as of various holdings in Ess and Sf), 25 Ketel teinnus. Stigandi TRE DB, fo. 254 (Walsingham in East Carleton), Ketel liber homo Stigandi ibid., fo. 264b (Stoke Holy Cross), Ketel. liber homo ibid., fo. 223 (East Harling), Ketel (liber homo) ibid., fo. 266 (Thurton, Loddon H, sub Stigando), Kitel ibid., fo. 243b (Hainford, sub Stigando), Ketel (bis) ibid., fo. 254b (Little Melton, Ketteringham), Godricus liber homo. Kitel ibid., fo. 254a (East Carleton). 26 - Ketel liber homo Eduini TRE DB, fo. 131 (Sisland). - Ketel. I. liber homo TRE DB, fös. 233, 233b, 234, 264b (Great and Little Walsingham, Whitwell, Witchingham, Holkham), Ketel (liber homo) ibid., fo. 271 (Wells next the Sea), Ketel ibid., fos. 122b, 233 (bis), 234 (Great Snoring, Great and Little Walsingham, Stiffkey, Scottow), Ketel ibid., fo. 184b (Ketel held one liber homo with 30 acres in Aldborough), Ketelli, gen., ibid., fos. 171 (bis), 172 (three liberi homines of Ketel held landat Trunch, North Repps and Little Barningham, respectively). I. liber homo Ketel TRE DB, fo. 112b (Morston). 27 - Ketel. I. liber homo TRE DB, fo. 206b (Ryston). - Ketel. I. liber homo TRE DB, fo. 201 (Billockby). - Ketel. liber homo TRE DB, fo. 264b (Hackford). - Chetel. liber homo TRE DB, fo. 230b (Barton Bendish). -Ketelfiedai 1086 DB, fo. 135b, /. liber homo Chetelfriedai2 8 ibid., fo. 273b (Raveningham). - Ketel presbitero de Welles, abl. (witn.) 1101-1107 Binham, fo. lb, Ketel de Warham et Otulfo fratre suo (abl.) (witn.) ibid., fo. 2a (grant by Peter de Valoignes to Binham Priory; Welles is Wells next the Sea and Warham is Warham). - Chetellus (witn.) 1108 Binham, fo. 20b (printed, SSMEA, Appendix I, no. 35) (Binham and Langham). -Ketello presbitero, abl. (witn.) c. 1115 CAcre, fo. 26b(i) (charter of Roger son of the Earl of W arenne' s seneschal Wimer by which he confirms the grants made to Castle Acre Priory by himself and by his father Wimer and his mother Gilla), Ketell' presbitero, abl. (witn.) c. 1135 CAcre, fo. 37a(i) (Gayton). - Ketello decano, abl. (witn.) c. 1115 (14th) CAcre, fo. lOla(i) (West Lexham). -Chetel c. 1131-1134 SCBI, xx, no. 1557 (moneyer, Norwich). - Godwinusfilius Ketel (witn.) 1134-1140 Holme, no. 139

23

For the family of Thurketel Heyng see below, s. n. Porkell.

24

In this document Wolgip also bequeathed Ketel and another brother, Elfkitel, the usufruct of

an estate at Stisted Ess for their lifetimes, after which it was to pass to Christ Church, Canterbury. 25 Str(a)ttune is probably Stratton Strawless while Moran is probably the DB Mora in Blofield H (ASWills, p. 204). 26 The Ketel of ASWills, nos. 32, 33, 34, and of the following DB entries given above appears with the byname Alder in the Latin headings of the Bury St Edmunds cartularies and in the list of benefactors to Bury contained in CUL MS. Mm. iv. 19, fo. 167 (ASWills, p. 201). The byname Alder may represent OE ealdor m. 'prince, lord, ruler', but it is much more probable that it stands for the comparative ofOE eald, ald (OE eldra, te/dra, ME elder, alder 'elder') (cf. ASWills, loc. cit.). A genealogical table of Ketel's family is given in an appendix to the present section, tig. 3. 27 Ketel of Morston, a liber homo of Earl Gyrth, was 'added' (additus est) to the royal manor of Holt after King Edward's death. 28 The bynamefir)iedai belongs to OE Frlgeda:g 'Friday' (Tengvik, pp. 218-219; DBS, p. 135, s. n. Friday(e); von Feilitzen, PNWinton, p. 157, n. 6).

257

Ketill, Ketil, Kretil (Hellesdon). - Ketello de Lechesham, abl. (witn.) c. 1150 CAcre, fo. 57a(ii) (grant to Castle Acre Priory by Hubert de Muntchenesy of two parts of his tithes in Cley next the Sea and in Holkham; Lechesham is Lexham). - totam decimam Ricardi filii Ketel c. 1150 CAcre, fo. 67b(i) 29 (confirmation by Bartholomew of Glanville of his father William's Grants to Bromholm Priory). -de terraKetelli Hals, 30 domus Katelli Hals c. 1155 Wals, fo. 12a(ii) (Walsingham). - Ricardus filius Ketell' c. 1160 CAcre, fo. 6la(iii) (grant by Wimerus of Helhoughton of his meadow of Oxewelle between Helhoughton and Raynham). - De Ketell' Brain 31 1165-1166 P, p. 33 (Nf or Sf). - Chetel preposito, abl. (witn.), Hermero filio Ketel, abl. (witn.) c. 1175 CAcre, fo. 86b(i) (Narborough). Ricardusfilius Ketel Ricl CRR 12 (Cur, i, p. 14), Ricardumfilium Ketell' 1198 FF (PRS NS, xxvii, no. 161) (Horstead). -pro fugaRogeri Ketel 1197 P, p. 243 (Nf or Sf). -Walterum filium Ketel 1198 FF (PRS NS, xxvii, no. 205) (Kyneholm in Buxton). - Ketel' filium Godwin' 1199 CRR 11 (CurR, i, p. 346) (Guist). -Johannemfilium Ketel 1199 CRR 17, 18 32 (CurR, ii, p. 144) (Nf). - terram Ketel c. 1200 Der, fo. 272b, Hugo filius Ketelli 1206 CRR 42 (Cur, iv, pp. 154, 196), Hugo Ketelli ibid. (Cur, iv, pp. 196-197), terram que fuit Hugonis Ketel c. 1215 Der, fo. 273a, Carta Henrici filii Hugonis Ketel de Ringelande, Henricusfilius Hugonis Ketel de Ringelonde 33 c. 1225 Der, fo. 273b(i) (Ringland). - scilicet xiiij acras cum pertinentiis quas Ketel de Skaft tenuit 1202 FF (PRS NS, xxvii, no. 385) (Reedham and Limpenhoe; Skaft is unidentified). 34 - Acelinam que fuit uxor Ketelli 1202 CRR 32 (Cur, iii, p. 77) (Nf). -(cum) tenemento quod tenuit Ketel pater predicte Maolt c. 1205 Wals, fo. 88b(iv) (grant by William son of William son of Rocelin to Walsingham Priory of his man John of Stratton Strawless and John's wife Maolt together with their issue and with all service and the tenement which Maolt's father Ketel had held). - Gaufridus filius Ketelli 1207 CRR 44 (Cur, v, p. 3) (Gaufridus son of Ketel was one of three jurors whose default caused a case of darrein presentment involving the church of Topcroft to be postponed). - terram Ketelli 1210 CRR 68 (Cur, vi, p. 108), terram lwainfilii Ketel' c. 1225 Co;iford 11, fo. 47a(vi) terram lwain filii Ketel ibid., fo. 49b(iii) (Tattersett). - Radulfus filius Willelmi Ketel de Lamesse c. 1215 Hickling, fo. 7b(iii) (Scottow; Lamesse is Lamas), Radulfumfilium Willelmi Ketel de Lamhesse c. 1215 Bromh, fos. 40b(iii), 48a(i), Radulfo filio Willelmi 35 Ketel de Lamhesse, dat., totam terram quam Willelmus 36 filius Ketel tenuit de me (charter of Bartholomew of Reedham) ibid., fos. 4 la(i), 48a(ii) (Lamas and Scottow). - Rogerus filius Ketel 1222 ElyA, fo. 193a (Pulham). - terram Bartholomei filii Ketel ? c. 1225-1240 Der, fo. 264b(i) (Holkham). -terram que fuit Ketel Scut37 in Estcroft c. 1225

The original charter of which this CAcre document is a copy is printed as Crawf, no. 16. The byname Hals belongs to OE heats (WSax, Kt), hals (Angl), ME hals(e) n. 'neck' (cf. Tengvik, pp. 316--317; DBS, p. 163, s. n. Halse). 31 The byname Brain belongs to ME brain, bra3en, brein, brane n. (< OE brtEgen) 'brain', perhaps used as an adjective (?short for brain-wöd) meaning 'mad, insane' (see MED, s. v. brain). 32 These two rolls are duplicates. 33 The MS. reads Henricus filius Hugonis concessi Ketel de Ringelonde but 'concessi' is underlined for deletion. 34 Skaft isa Scandinavianized form of OE sceaft m. 'shaft, pole' in which Scand /sk/ has replaced OE !Il. As a topographical term OE sceaft was probably used to denote a shaft or pole acting as a boundary (EPN Il, p. 99, s. v.). 35 MS., Bromh, fo. 41a(i) Willelmo. 36 MS. Willelmi. 37 The byname Scut may represent a ME *scut < OFr escoute 'a listener, scout', cf. e. ModE scout(e), skout, scowt, skowt (NED, s. v. scout) or it may be a nickname derived from the etymo29

30

258

Ketill, Ketil, Kretil Wals, fös. 89b(iv), 90a, partem Ketel Scut ibid., fo. 90a (Itteringham), Simonfilius Ketel Scut de lteringham ibid., fo. 90a(i) (three roods and twelve perches which were situated super holmas between Calthorpe and Itteringham), Adam Ketel (acc.) 1232 CRR 232 (Cur, xiv, p. 533) (ltteringham), Carta Walteri Ketel, Walterus Ketel c. 1240 Wals, fo. 90a(ii) (three perches que iacent apud holmes et habutant super aquam de bliclinge; bliclinge is Blickling which lies three miles south-east of Itteringham). - Gilberto Ketel de Hardele, abl., m. -1. 13thLangley, fo. 22b (court at Langley; Hardele is Hardley Street about two miles east of Langley).

ON Ketill, OEScand Ketil, Kreti! is an original byname belonging to the appellative ON ketill, ODan, OSwed kreti! m. 'kettle, cauldron' which is derived from Lat catillus m. (a diminutive of Lat catinus m. 'basin, bowl'), a Latin loan word which is common to both NGerm and WGerm, cf. OE ci(e)tel, cytel (WSax), cetel (Angl, Kt) m. 'kettle', OHG chezzil, chezzel m. 'kettle, vessel'. Wessen, Nordiska namnstudier, pp. 71-72, suggested that the personal name Ketill originally had the sense 'he whose head is shaped like a kettle'. Janzen, NK VII, p. 43, considered it to be probable that the personal name was connected with the appellative in the specific sense 'helmet, kettle shaped spiked helmet', possibly with a transferred meaning '(helmet adomed) chieftain', and he compared it generically with ON Hialmr, OSwed Hi::elm, a name which is identical with the appellative ON hialmr m. 'helmet'. Altematively, it has been suggested that the personal name Ketill is to be connected with the appellative in the sense 'sacrificial cauldron, cauldron used for ritual acts' (NK VII, p. 43; Möller, Studien zu den theriophoren Personennamen der Germanen, pp. 132-134). The use of the personal name Ketill appears to have begun in the Viking period, but the suggestion of Mosse, Revue celtique, 1 (1933), pp. 248-253, that its use is the result of Celtic influence from the British Isles, can be rejected without reservation (see Möller, loc. cit.). Mosse, loc. cit., cited as comparison Gallic peir 'kettle' which could have the transferred sense 'leader, ringleader' and presumed that it had an equivalent in Insular Celtic. Mosse's examination of the name is highly speculative and based on too little evidence to be tenable. 38 ON Ketill is frequent in both Norway and lceland throughout the Middle Ages from the Viking period onwards (NiyR, nos. 154, 228, 251, 272, 390, 451,493, 500; Lind, cols. 684-687). In ODan the name occurs as runic ketil, nom. (DR, nos. 14, 403, Mo. 51),

logically obscure word scut originally used to denote the tail of a hare and later the hare itself (see DBS, pp. 309, 310, s. nn. Scott, Scutt). 38 For further discussion of the semantic problems connected with the etymology of ON Ketill etc. see Modeer, SvP, pp. 21, 60, and Hald, Personnavne i Danmark: I. Oldtiden, pp. 63--65. See also the references given by NK VII, p. 149, n. 60. For ON Hialmr, OSwed Hia:lm, which as a personal name probably originally belonged to the appellative in the transferred sense 'protection', see NK VII, p. 42.

259

Ketill, Ketil, Kretil kitil, nom. (DR, no. 67), ke(t)il, nom. (DR, no. 59), ke(ti)l, acc. (DR, no. 391), kitils, gen. (DR, no. 107). It is also frequent in medieval Danish documents where it appears as Ketil, Ketillus, Katillus etc., and, from c. 1400, as Krel(l), Kel (DgP, cols. 739-745, 1684; for the OEScand syncope of Kreti!> Krel(l) see Noreen, AschwedGramm, § 156.lb).39 The name isa common first element of Danish place-names, both in its original form Kreti! and in its later syncopated form Krel(l), Kel (DgP, col. 745; DS Vl, p. 531; DS VII, p. 329; DS IX, p. 162; DS XI, pp. 92, 170, 216--217; DS XII, pp. 99, 164, 190; DS XIV, p. 88; DS XVI, p. 116; SkOA XVI, p. 113). In OSwed Kreti! is frequent in runic inscriptions and is well attested in medieval and 16th century records in Sweden and Finland (SRÖl, nos. 5, 12; SRÖg, nos. 90,124,230; SRSö, nos. 46,338; SRSmå, nos. 5, 109, 121, 148; SRVg, nos. 11, 13, 79; SRU, nos. 97, 247, 284,288,293, 351,371,421,423, 509, 798, 946, 987, 999, 1044, 1053, 1104, 1114; Lundgren-Brate, pp. 158159; Modeer, Personnamn i Ktb, pp. 27, 69, 131-135; Modeer, Östsmåländska bondenamn från 1500-talet, p. 65; Thors, pp. 67--68; Fredriksson, pp. 214, 216, 218, 224, 226, 231; Pellijeff, Ur Norrbottens äldsta mantalslängder, p. 170; Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183, 185; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 130, s.n. Käll, Kättil). The Latinized form Katillus has been taken to reflect an old non-mutated side form ofthe normal OEScand Kreti! (see DgP, col. 745, and the works discussed by Modeer, NoB, xl (1952), p. 70), but Modeer has shown that this is untenable and that Katillus and such compounds as Katilbernus, Katilmundus and Katilvastus have a in place of the normal re, e as a result of the influence of Lat catillus (Modeer, NoB, xl (1952), pp. 69-74; Modeer, Personnamn i Ktb, pp. 131-135; Modeer, SvP, p. 60). The syncope of Kreti/ > Krel(l) etc. seems to have been later in OSwed than in ODan. Swedish Käl(l), Kel is known in the late 16th century in Ångermanland, Dalsland, Medelpad, Norrbotten and Västergötland (Fredriksson, pp. 219,221; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 36, 41, 74, 77, 80, 130). The modem Kjell is recorded in a document of 1583 from Småland and becomes general in Sweden around 1600 (Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 20 and n. 16; Otterbjörk, Svenskaförnamn, p. 106). In England ON Ketill, OEScand Kreti/, Ketil is frequent both in independent use and as the first element of place-names and field names: NPN, p. 79; ZEN, p. 54; PNDB, pp. 304--305; Tengvik, p. 177; SPNT, pp. 143, 144, n. l; ELPN,pp. 78,129,195; DBS, p. 204, s.n. Kettle; SPNLY,pp. 166--170; Smart, Mon-

39 The 15th century Norwegian examples of Kels, Kiels, gen., given by Lind, col. 686, are doubtless the result of Danish influence.

260

Ketill, Ketil, Kretil eyers, pp. 229, 263; von Feilitzen, SCBI, xi.i, p. 12; Fellows Jensen, Lincolnshire Tenants, pp. 87, 94; Smart, Cnut's York Moneyers, p. 225; SCBI, xxi, pp. xlvi, xlvii, xlviii; von Feilitzen, PNWinton, p. 153; DEPN, pp. 269,270,273,274,275, s. nn. Kedleston, Kelthorpe, Kettleburgh, Kettleby, Kettleshulme, Kettlesing, Kettlestone, Kettlethorpe, Kexby YE, Kexmoor, Keyston; PN BdHu, pp. 243, 300; PN NRY, pp. 89, 332; PN Nth, p. 291; PN Ess, pp. 619,638; PN Wa, pp. xxii, xxiii, 327,345,348; PN ERY, pp. 225, 325; PN Hrt, p. xx; PN Nt, p. 292; PN Cu, pp. 26, 370, 505; PN 0, pp. 283,481; PN Db, pp. 580, 762; PN WRY, i, pp. 172,325; PN WRY, ii, pp. 82, 103, 148,187,215; PN WRY, iii, pp. 16, 23; PN WRY, iv, pp. 56, 112, 186, 195; PN WRY, v, pp. 68, 92, 132, 202,203,209; PN WRY, vi, pp. 97, 179,227,238,271; PN WRY, vii, pp. 295,301; PN Ch, i, p. 110; PN Ch, iii, p. 313; SSNY, pp. 61, 69; SSNEM, pp. 56, 138, 170, 192.

It should be remarked that some of the place-names, especially those which are non-genitival compounds, may rather contain the appellative OE ci(e)tel, cytel (WSax), cetel (Angl, Kt) m., ON ketill, OEScand kr:etil m. 'kettle', used topographically in the sense 'a deep valley surrounded by hills' (see SPNLY, p. 170; EPN I, p. 91, s. v. cetel; EPN II, p. 3, s. v. ketill). Ekwall, DEPN, p. 274, s.n., took Kettleburgh Sf (Cetelbirig, Chetelberia, Chettlebiriga, Ketelbiria, Kettleberga, Ketlebere, Ketdesbirig 1086 DB) to have the etymology 'Ketil's hill' but thought it just possible that the name is 'a Scandinavianized form' of an OE cetelbeorg 'hill by a narrow valley'. The forms, in fäet, correspond to the OE etymology and there is no need to suggest Scandinavianization (as was also done by A.H. Smith, EPN II, p. 3, s. v. ketill). The regular appearance of forms without genitival -es speaks against the personal name as first element. Ekwall remarked that the almost total absence of genitival s in the early forms of Ab Kettleby Lei, Eye Kettleby Lei and Kettleby L 'is somewhat remarkable' but pointed out that it has 'many analogies' (DEPN, p. 274, s.n. Kettleby). SSNEM, p. 56, points out that derivation of the first element of the L and Lei Kettlebys from OE (Angl) cetel or OScand ketill with the sense 'kettle, cauldron, cauldron-shaped valley or hill' is 'not appropriate' to the sites of these three vills. Note also the following English examples of ON Ketill, OEScand Kr:etil, Ketil: Cytel (brother ofl>urferö of Warmington) 963-984 (m. 12th) ASChR, no. 40 (surety in a transaction conceming property at Warmington Nth), Cytel (brother ofl>urferö Rolfes sune) (ter) 963-992 (m. 12th) ASChR, no. 40 (surety in transactions conceming properties at Bainton, Lutton and Warmington Nth); Cytel Clacces sune ret Wermingtune 963-984 (m. 12th) ASChR, no. 40 (Warmington Nth; Cytel son ofClac ofWarmington was a surety in a transaction conceming property at Warmington); Cytel c. 969-993 (m. 11th) ASChR, no. 47 (surety in a transaction conceming property at Stoke Canon D); Cvtel 1042-1044 SCBI, xviii, no. 766 (moneyer, Canterbury K); Ketel 1070--1087 (12th), 1070--1087 (13th) Urry, p. 445 (Canterbury K); Chetel 1087-1098 (c. 11751200) Bury, pp. 30, 38, 44 (Hessett, Honington, Brettenham Sf); Orm filius Ketell (witn.) 1094 (15th) Lancaster, p. 10 (grant by Roger de Poitou to St Martin of Seez of the church of St Mary of Lancaster and other properties in his fee of Lancaster); totam terram Edrici filii Chetelli, Chetellus pater ejus 1123 AC, no. 10, totam terram et

261

Ketill, Ketil, Kretil teneuram que fuit Edric' filii Chetelli, Chetellus pater suus4° 1136 AC, no. 20 (Gl); Chitel c. 1131-1134 SCBI, xvii, no. 666 (moneyer, Leicester); terram Turchilli filii Ketelli in Humeresfeld 1146--1174 CampbCh xii. 3 (printed, SSMEA, Appendix I, no. 19) (Homersfield St); Ketellus filius Gerardi 1157-1163 LaCh, p. 312 (Ketel son of Gerard was one of the jurors charged with ascertaining the bounds between Furness La and the barony ofKendal We); Chetel (abl.) (witn.) c. 1180--1190 RuffordCh, no. 115 (Abney Db); Ketellus de Kirkebi Lonesdalie, Stephano genera Ketelli, abl. (witn.) 1184-1190 Cockersand, p. 912, Adamfilius Ketelli 1190--1200 Cockersand, pp. 913, 914 (Kirkby Lonsdale We); Ailwi Ketel 1186--1188 (c. 1270) BuryS, p. 49 (Langham St), Edricus Keteles ibid., pp. 51, 52 (Hopton St); Robertusfilius Chetelli c. 1195 (13th) Rams, iii, p. 250 (Elsworth C); Petrus Ketel' et Ailwinus frater suus, misericordiam Petri Ketell', Willelmus filius Ketell' 1199 CRR 51 (Cur, i, p. 105) (Hangleton Sx); Ketell forestario, abl. (witn.) 1200--1204 (15th) Lancaster, p. 152 (Ramessorm in Prestunholme and Gremescheric, both prob. in Gressingham La), Adamfilius Ketelli de Gersingham, Benedicto filio Ketelli de Gersingham, abl. (witn.), 1200--1219 (orig.) Cockersand p. 923, Adam filius Ketelli de Gersingham, Benedicto filio Ketelli, abl. (witn.) 1200--1225 (17th) Cockersand, p. 922 (Bustocrig in Gressingham La); ostium Ketelli 1200--1225 Cockersand, p. 848, totam terram illam cum omnibus suis divisis quam Ketellus surdus quondam tenuit, terram Gilbertifilii Ketelli 1241-1251 Cockersand, p. 856 (Caton La); dimidiam bovatam quam Walterusfilius Ketel quondam tenuit e. 13th (15th) RuffordCh, no. 66,Johannemfilium Walterifilii Ketel de Bramton' 1245 (15th) RuffordCh, no. 74, unam culturam que fuit Matildis uxoris Ricardi filii Ketel 1223-1231 (15th) RuffordCh, no. 81, totum servicium Radulfifilii Ketil 1250--1279 (15th) RuffordCh, no. 92 (Brampton Db); Stephanumfilium Ketelli 1201 CRR 25 (Cur, ii, p. 48) (Stake St); Willelmusfilius Ketelli 1212 CRR 54 (Cur, vi, p. 194), Willelmum filium Ketelli 1212 CRR 56 (Cur, vi, p. 345) (Cu); Ketell' mercator 1212 CRR 57 (Pleas, iv, no. 4754) (C); Ketelle de Baddesle 1221 AssRoll 948 (Eyre III, no. 412), De Ketell' de Baddeleg' ibid. (Eyre III, p. 421) (Baddesley Clinton Wa); Ketellum de Warewic', Ketellus (bis) 1221 AssRoll 948 (Eyre III, no. 832), Ketell' de Warewic' ibid. (Eyre III, no. 833) (Warwick); Jordanus mercator auunculus Hugonis Ketel 1221 AssRoll 949 (Eyre III, no. 1498) (Coleshill Wa); Thomas Ketel 1222 ElyA, fo. 150b (Leverington C), Rogerus Ketel ibid., fo. 227b, Rogerus filius Ketel ibid., fo. 229a (Wetheringsett St); Ketell' filius fabri c. 1230 (c. 1270) BuryS, fo. 11 la (Munhale in Saffron Walden Ess); Ricardo filio Ketel et Henrico filio eius, dat., 1235-1256 (orig.) Burscough, no. 71 (Burscough La).

In Normandy the name is recorded in independent use in the 11th century in the form Chetellus and it has also been taken to be the first element of several Norman place-names (Adigard, pp. 119, 312, 404-405). The forms Cetel, Ketel, Chetel, Ketellus and Chetellus have been Anglicized through the influence ofthe appellative OE (Angl) cetel. Forms in Cytel, Cvtel and Kytel reflect WSax scribal usage since they show the influence of the OE appellative in the specifically WSax form cytel. A late survival of the WSax variant Cytel is contained in the O field name Kutelesmore 1235 FF, Cutelesmore 1244 FF (PN 0, pp. 283, 481). Cf. also the representation of OScand -kcetill as the second element of dithematic personal names by -cytel,

°For Eadric son of Chetel see AC, p. 18 n.

4

262

N.N. I

I

Eadwine (OEEadwine)

l Wlfric (OE Wu/frlc)

W olgyp & Elfwine (OE Wulfgyö) (OE!Elfwine) fem.

Elfkitel

Ketel

(AScand !Elfcytel, (ON Ketill etc.) OScand *Alfketill, cf. ODan Alfkil)

Wulkitel

Bote

Gode

Ealgip

(AScand Ulfcytel, cf. OSwed (run.) Ulfka:till)

(OEBote) fem.

(OEGode) fem.

(OE Ealdgyö) fem.

Fig. 3. The family connections of Ketel Alder. Ketel had a further brother Godric (OE Godric), whom ASWills, p. 198, suggests might have been a half-brother on account of his absence in ASWills, no. 32, the will of Ketel' s mother Wolgyp. Ketel also had a stepdaughter whose name is rendered as Algif (nom.), Algiue (acc.) (probably OE !Elfgifu, fem., though OE !Eöelgifu fem. and OE *Ealdgifu fem. should also be considered) (ASWills, no. 34). His wife may have been the Sifleda (OE Sigeflced fem.) who appears in his will (ASWills, no. 34 and p. 202). Sources: ASWills, nos. 32, 33, 34.

...~

F=

~

? N

°'

VJ

~ 5:

Klakkr, *Klak -cetel, -chetel, -ketel, -chetellus, -ketellus etc. in English records (see above, s.nn. Arnkell, Äskell). The 11th century Norman form Chetellus (Adigard, pp. 119, 312) doubtless reflects AScand influence from the Danelaw since its form shows the influence of OE (Angl) cetel. 41 The spelling Katelli, gen., in the form domus Katelli Hals c. 1155 Wals has a for e probably as a result of AN interchange of a ande (PNDB, § 9; SPNLY, § 15), though perhaps the influence of the Lat appellative catillus might also have to be considered (for the medieval Scand Katillus whose form is clearly the result of association with Lat catillus, see above). Ch- for K- in the forms in Chetel, Chetellus represents the normal e. AN spelling for /k/ before front vowels (PNDB, § 113; SPNLY, § 126; von Feilitzen, PNWinton, p. 225). The personal nomenclature used in the family of the 11 th century East Anglian landowner Ketel Alder reflects the mixture of English and Scand elements which took place in East Anglia since it includes both OE and Scand names, with the former predominating. A genealogical table is given as an appendix to the present discussion (fig. 3). Reaney, DBS, p. 202, s.n. Kell(s), took ME Chel(le), Kelle to be 'ON *Kel, OSwed Krel, shortened from Ketill.' As is pointed out above, however, the syncopated Krel(l), Kel< Kreti/, Ketil does not appear in Danish until c. 1400 and is later still in Norway and Sweden. On chronological grounds, therefore, the syncopated Scand Krel(l), Kel < Kreti/, Ketil can be ruled out as the base of ME Chel(le), Kelle (see also PN Ch, iii, pp. 27fr277). The most likely explanation of the name is that it is a ME short form of such forms as Asket < ON Äskell, Ulkel < OEScand Ulfkil etc. (see PN NRY, p. 18). Where Ketel is used as a byname and confirmatory evidence of a patronymic function is lacking it can also reflect ME ketel 'kettle', corresponding to OE (Angl) cetel etc.

ON Klakkr, ODan (p. ns.), OSwed (p. ns.) *Klak a) (i) Clakeslosa 1086 DB, fös. 212b, 236, 263, Clacheslosa ibid., fös. 190, 205b, 209, 230,238, 250b, Clachelosa ibid., fö. 149, Claclelosa ibid., fö. 159b, Clakeclose 10871098 (c. 1175-1200) Bury, p. 11, Clakeslose, Clakesloes 1209 AssRoll 558 (Pleas, iv, nos. 4049, 4051), Clakelose ibid. (Pleas, iv, nos. 4055, 4074, 4077, 4089, 4093, 4095, 4106,4118) (Clackclose H; förfurtherförms seeEHNI, pp. 74-75). -Clakestona 1086 DB, fös. 177, 177b, 203b, Clarestona ibid., fös. 177, 203b, Clakeston' 1200 CRR 16 (Pleas, i, no. 3177), Clakestun, Claxtone 1275 RH (OKS) (Claxton, Loddon H). Clakeswelle e. 13th Wymondham, fö. 89a(i) (f. n. in Wymondham). - Clackesdele m.

41

For AScand influence in Normandy see above, s.n. ÄsbiQrn, n. 67.

264

Klakkr, *Klak 13th Wals, fo. 18a(i), Clakesdele m. - 1. 13th Wals, fo. 44a(i) (f. n. in Houghton St Giles).

ON Klakkr, OEScand *Klak is an original byname which is to be connected with the appellative ON klakkr m. 'lump', a word which is used in place-names to denote the summit of a hill and which, it has been suggested, is identical with Jutland dialect klak 'low-lying area in a water meadow where no grass grows' (Hald, Personnavne i Danmark: I. O/dtiden, pp. 67-68, for further details see the works cited by Hald, ibid., p. 80, n. 68). ON Klakkr has been taken to occur in Bohuslän in the late 14th century form Reider Klagsson but it does not otherwise appear to have been found in the OWScand area (Lind, cols. 690-691). ODan *Klak is the first element of several Danish place-names (DgP, col. 749; Hald, Vore Stednavne, pp. 101, 147, 203; Hald, Personnavne i Danmark: I. Oldtiden, p. 67; DS XI, p. 184; DS XVIII.l, pp. 113-114, 142). The name may also be represented in ODan by the runic form klakR, acc. (DR, no. 325; for discussion of this form see L. Jacobsen and E. Moltke, NoB, xxiii (1935), p. 185; DgP, cols. 748-749; DR, col. 675; Hald, Personnavne i Danmark: I. O/dtiden, pp. 67, 80, n. 67). Note also the 9th century Danish king Harald Klak, whose name appears as Klakk-Haraldr in ON literary sources and as Harald Clac, Haraldus Klak, Klac Harald etc. in medieval Danish chronicles (DgP II, col. 569). An OSwed *Klak is known from place-names (Lundgren-Brate, p. 150).42 A weak form, ODan Klakki, OSwed Klakke, is also on record (DgP, col. 749; SRSmå, no. 11; SRVg, no. 169; Lundgren-Brate, p. 150). Note also the related ON byname Klakka, which is attested in Iceland from the time of the settlement until the latter part of the 12th century (LindBiN, cols. 201-202). Cf. also the Icelandic formations Klakk-Alfr and Klakka-Ormr and the Icelandic byname KlakkhQ[öi (LindBiN, col. 202). In England ON Klakkr, OEScand *Klak is quite well attested, both in independent use and as the first element of place-names and field names (NPN, p. 81; ZEN, p. 54; PNDB, p. 305; Tengvik, pp. 153,215; OBS, p. 74, s.n. Clack; SPNLY, pp. 172-173; DEPN, pp. 109, 110, 197, s. nn. Clawson, Claxby, Claxton, Claythorpe, Glassthorpehill; IPN, p. 185; PN BdHu, p. 300; PN NRY, p. 37; PN Nth, pp. 83, 149; PN K, p. 277; PN 0, pp. 87, 89,481; PN WRY, i, p. 36; SSNY, p. 56). Note the following examples of the name in independent use in the OE period: Clac 925-939 C.E. Blunt, The Coinage oj Athelstan, 924-939, pp. 129, 136 (moneyer, North-Eastem I group; Cytel Clacces sune ret Wermintune 963-984 (m. 12th) ASChR, no. 40 (Warmington

42 Note also the runic klakR, nom. (Lundgren-Brate, p. 150; SRSmå, no. 133). This can reflect an OSwed (run.) Klakkr but it can equally stand for GlreggR, GIQggr (SRSmå, pp. 279, 338).

265

Klakkr, *Klak Nth; Cytel son of Clac of Wannington was a surety in a transaction conceming property at Wannington), Clac ibid. (Nth, possibly Wittering), Clac on Castre (bis) ibid. (surety in two transactions in Nth, one of which possibly concems Wittering; Castre is Castor Nth), Clac ret Byrnewillan ibid. (surety, Nth; Byrnewillan, wceps 'wasp' 56 and this might also be considered here (cf. also PNDB, § 90).

DBS, p. 83, s. n. Copsey, assigns this form to Nf. Note, however, that Björkman, NPN, p. 87, presumed that the name originally had an abusive connection and that it is to be linked with ModSwed dialect kofsa 'prostitute, loose woman, slut'. 54 Note that the form Cofsi is recorded once in DB beside the usual Copsi (PNDB, p. 306; SPNLY,p. 175). 55 For this Scand development see Noreen, AislGramm, § 240.2, and Noreen, AschwedGramm, § 259.2b. 56 For this see Campbell, § 418. This change is also mentioned by PNDB, § 90, but, as is men52 53

273

KolbiQrn, *Kolbiorn

ON Kolbh;_,rn, OSwed (p. ns.) *Kolbiorn b) (i) Terra. Coleberni. presbiteri, Colebernus 1086 DB, fo. 263b (Humbleyard H). Colobernus (husband of Ansfrida) 1150 St William, p. 149 (Norwich). ON KolbiQrn is extremely common in Norway from the Viking period onwards but is only weakly represented in Iceland (Lind, cols. 702-704; LindS, cols. 578-584). An OSwed *Kolbiorn has been postulated from place-names (Lundgren-Brate, p. 153). In England the name is extremely rare (NPN, p. 83; ZEN, p. 563). Examples additional to the Nf forms given above are: Colbern Wml/Wm2 (c. 108&--1089) SCBI, xii, no. 186, SCBI, xvii, nos. 651, 652, Wm2 (c. 1092-1095) SCBI, xii, no. 203 (moneyer, Wallingford Brk); Colbern 1180-1181 P. p. 31, Colbern. Robertusfilius ejus 1181-1182 P, p. 29, Colbernus, Robertus filius ejus 1182-1183 P, p. 114, Robertusfilius Colberni 1183-1184 P, p. 76, Robertus filius Colbern 1184-1185 P, p. 158 (D); Celestia uxor Ricardifilii Colberni 1185-1186 P, p. 192 (K).

Interconsonantal -e- and -o- in the Nf forms Colebernus and Colobernus are AN svarabhakti vowels (PNDB, §§ 52, 53; SPNLY, 63). For the rendering of the second element ON -biQrn etc. by -bern(us) in English records see above, s.n. AsbiQrn. The above Nf forms have been Latinized by the addition of the gen. and nom. endings of the Latin second declension (see PNDB, § 153; SPNLY, § 148; von Feilitzen, PNWinton, p. 227).

ON (p. n.) *Kolfr a) (i) Couestuna 1086 DB, fo. 162b, Coluestune 12thLewes (OKS), Colueston 1234FF (OKS), Colveston 1248 Ch, 1302FA, 1341 FF, 1346FF, 1346FA, 13471pm, 1348 Pat, 1358 FF, 1553 Pat (OKS) (Colveston, Grimshoe H). LindBiN, col. 209, considered it to be possible that the last Norwegian place-name Kolfs rud in Eidskogen contained an ON byname *K6lfr which had the sense 'balt', cf. ON k6lfr m. 'the tongue in a bell, a balt'. Ekwall, DEPN, p. 118, s.n., interpreted Colveston Nf as follows: Possibly 'Kolf's TON'. There may be an ON byname Kolfr. Or the first el. may be an unrecorded OScand pers.n. *Kolfastr; cf. Kolbeinn etc. An OScand byname *K6lfr would fit the early forms of Colveston Nf and is

tioned above, von Feilitzen preferred here to take the English form Copsi to reflect the Scand development.

274

Koli, Kolli, Kolle, Kulle etymologically unproblematic. It is, therefore, probably to be preferred to an unrecorded OScand *Kolfastr. The DB form Couestuna shows AN loss of ante-consonantal -l- (see ANI, pp. 146--153; IPN, p. 113; PNDB, § 64; SPNLY, § 70)

ON, ODan, OSwed Koli or ON, ODan Kolli, OSwed Kolle, Kulle a) (i) Colebei 1086 DB, fo. 115b, Colebi (p) 1202 FF (PRS NS, xxvii, no. 375), Coleby 1199 FF, 1200 FF, 1202 FF (PRS NS, xxvii, nos. 221,253,375) (Colby, South Erpingham H). - Coletuna 1086 DB, fo. 166, Coletune 1182 Wymondham, fo. 10a, Coletun' 1202 FF (PRS NS, xxvii, nos. 309,327), Coleton' 1198 FF, 1203 FF, 1209 FF (PRS NS, xxvii, no. 154, PRS NS, xxxii, nos. 26, 185, 190, 198) (Colton, Forehoe H). - Colechirca 1086 DB, fo.191b, Colekirka ibid., fo. 197b, Kolekirch' 1209 AssRoll 558 (Pleas, iv, no. 4309) (Colkirk, Launditch H).

OScand Koli and Kolli are etymologically unconnected but are impossible to separate in post-Conquest English records on account of the frequent doubling and simplification of OE and OScand consonants practised by AN scribes in the ME period (see PNDB, §§ 146, 147; SPNLY, §§ 142, 143). OScand Koli is a weak side-form of Kolr, a name which can be either a short form of dithematic names in Kol- or an original byname, 'the dark, the swarthy complexioned one', belonging to ON kol m. 'coal, charcoal' (Liden, NoB, iv (1916), p. 117; Franzen, Vikbolandet, pp. 107-108; NK VII, p. 50; Modeer, SvP, p. 36; Sandred 1987:319). OScand Kolli is a weak side-form of Kollr, a name which is an original byname, 'the bald-headed' or 'he whose hair is soft or falls thickly about his head' (NK VII, p. 51). Koli is only attested sporadically in Scandinavia, though it is known in both OWScand and OEScand (Lind, col. 705; DgP, col. 776; Lundgren-Brate, p. 153; Modeer, Personnamn i Ktb, p. 127). Kolli is well attested in Iceland from the time of the settlement onwards but is infrequent in Norway and in the OEScand area (Lind, cols. 706--707; LindS, col. 584; DgP, col. 776; Lundgren-Brate, pp. 153, 156). Koli is taken by Ekwall, DEPN, p. 116, s. nn., to form the first element of Colby Nf, We and Coleby (2) L. 57 It may also form the first element of Cowthorpe WRY, though OE Cola has been suggested as a possible alternative (PN WRY, v, p. 22). 58 Kolli has been suggested as the first element of cer-

SSNEM, p. 41, also takes Ko/i to be the first element of the L Colebys. SSNY, p. 57, states that the first element of Cowthorpe WRY is the Scand Koli and does not give OE Cola as an alternative. 57

58

275

Koli, Kolli, Kolle, Kulle

tain field and minor names in Cu, Db and WRY (PN Cu, p. 370; PN Db, pp. 451--452; PN WRY, iii, p. 264; PN WRY, vi, pp. 52, 160). PN We, ii, p. 96, takes Colby We to contain either ON kollr m. 'a hill' or the personal name Kolli. Kolli may also form the first element of Coulby NRY but here Koli and the strong forms Kollr and Kolr have also been proposed as alternatives (PN NRY, pp. 170, 323; SSNY, p. 24). Note also the lost NRY place-name Cole(s)bi 1086 DB near Kirby Sigston which also appears to contain one of these four names (SPNLY, pp. 176-177; SSNY, p. 24). Kolli may also form the first element of some 13th and 14th century Db field names, though in some of these ME Colle, a shortened form of Nicholas, is a possible alternative (PN Db, p. 762). Indeed, ME Colle < Nicholas should perhaps also be taken into account as a possible alternative in the other ME field names which have been hitherto presumed to contain ON, ODan Kolli, OSwed Kalle, Kulle (cf. the field names which PN WRY, iii, p. 264, and PN WRY, vi, pp. 52, 160, suggest contain ON Kolli). In independent use ON Kolli etc. appears in DB TRE and in 1086 in Db (PNDB, p. 307). Ekwall, DEPN, pp. 117, 118, s. nn., takes Colkirk Nf and Colton Nf, WRY (2) to contain either the Scand Koli or OE Cola. The two WRY names are taken by PN WRY, iv, pp. 117, 223, to contain the OE name. It is also probably not to be excluded that the Nf and We Colbys and the L Colebys might also possibly contain OE Cola. It is doubtful, however, if ON, ODan, OSwed Koli is attested independently in English records. It may be contained in the London form Nicolaus fil. Algari Colessune 1138-1160 (m. 12th), though the strong variant Kolr is equally possible here (ELPN, pp. 78-79). Cola is quite frequent in independent use in 1. OE and in the early postConquest period. Von Feilitzen, PNDB, p. 217 and n. 4, listed OE examples from D, Du, Ha, Nf, W and Wo, and DB and early post-Conquest examples from Brk, Co, D, Db, Ess, Gl, Ha, L, Nf, Sf, So, Sr, Sx, W and Y. Von Feilitzen (PNDB, p. 217) considered that some of these instances may represent ON, ODan, OSwed Koli but that, on the whole, the local distribution of the material does not point to a Scand origin and that in many cases we are concerned with OE Cola, which is presumably an original byname from OE col n. 'coal' meaning 'the coal-black, swarthy one' or the like. OE Cola has been assumed to form the first element of numerous English place-names and field names both within and outside the Danelaw: DEPN, pp. 81, 116-118, s. nn. Calcot Brk, Calcutt W, Colaton Raleigh, Collaton St. Mary, Colleton Barton, Colbury, Colham, Collingham, Collington, Collingtree, Colney, Colthrop, Colton St; PN Bk, pp. 141,242; PN BdHu, p. 40; PN Wo, pp. 71, 91; PN Sx,pp.49,58,75,346,451,491;PND,pp.50--51,303,307,587;PNNth,p. 145;PN Sr, p. 401; PN W, pp. 42-43; PN Nt, pp. 160,203; PN Mx, pp. 41-42; PN C, p. 2; PN 0, pp. 80,477; PNWRY, iv, pp. 117, 174-175, 223; PN Gl, ii, p. 183; PNBrk, pp. 188, 302; lnsley, JEPN, x (1977-1978), p. 58.

276

Kolsten The distributional pattem of these place-names and field names confirms, by and large, the pattem taken by the examples of Cola found independently in English records. Confirmation of the English origin of the personal name Cola is provided by the earliest recorded form of Colham Green Mx, Colanhomm, which occurs in BCS, no. 400, a contemporary charter of 831 (PN Mx, p. 41). 59 Further confirmation is provided by the place-names Collingham Nt and WRY and Colthrop Brk, in which Cola is compounded with the archaic OE elements -ingahäm and -prop, respectively (PN Nt, p. 203; PN WRY, iv, pp. 174-175; PN Brk, p. 188). There is, thus, no need to regard examples of the personal name Cola, Cole found in English records of the 1. OE and e. ME periods as representing anything other than OE Cola. This applies also to the Danelaw despite the probability that the rare ON, ODan, OSwed Koli forms the first element of Colby Nf and of the L Colebys. Note also the Sf forms Cole 1087-1098 (c. 1175-1200) Bury, p. 41 (Ixworth Thorpe St), JE/stan Cole sune ibid., p. 43 (Litlecerce St). Tengvik, pp. 153-154, took English Cola, Cole to represent either OWScand, ODan, OSwed Koli or OE Cola and also suggested (p. 154) that an alternative base for the second of the above Sf examples would be OWScand Kolli with AN simplification of consonants. It might just be possible that the form Cole sune could reflect a similar formation to the London Colessune mentioned above and contain Scand Koli or Kolli, though in view of the rarity of these Scand names the OE Cola is probably to be preferred. ON, ODan, OSwed Koli has been suggested as the first element of several place-names in Normandy (Adigard, pp. 123, 406-408). Note, however, OHG Cola (Förstemann PN, col. 371), which may have had an unrecorded West Frankish counterpart. In view of the known AScand element in Normandy, OE Cola or OE col n. 'coal, charcoal' or OE col 'cool' may also have to be considered here. 60

OSwed Kolsten b) (i) Colstanus 1155-1158 Holme, no. 21 (Ranworth), Reginaldusf Colstan 1191 P, p. 98, 1192 P, p. 40 (Nf), Reginaldusfilius Colstan de Radewrde, Reginaldusfilius Colstan 1209 AssRoll 558 (Pleas, iv, no. 4157) (Ranworth), Regina/dus Colstan 1214 CRR 112 (Cur, vii, p. 259) (Nf), terram Willelmifilii Colstan 1214-1229 Holme, fo. 85a(v) (Ranworth).

An OSwed Ko/sten is represented by the Latinized forms Colstanus 1342

59

Fora full bibliography conceming this charter see Sawyer, no. 188.

°For AScand influence in Normandy see above, s.n. ÄsbiQrn, and n. 67.

6

277

Kolsten (Östergötland) and Culstanus 1298 (Ljunga, Småland) (Lundgren-Brate, p. 153).61 The name does not appear to have an equivalent in Norwegian, Icelandic and Danish records. 62 In England OSwed Ko/sten has a counterpart in Colstan(us), which occurs in independent use in Do as well as in Nf6 3 and which forms the first element of the L field name Colstandic 1202 SPNLY, p. 179. A further English equivalent of OSwed Ko/sten is contained in the first elements of the lost Y place-name Colestainthorpe 1249 and the NRY field name Golstaindale 1160 (Lindkvist, p. 68; PN NRY, p. 332; SPNLY, p. 179). It is noteworthy that both of these Y names preserve the original OScand diphthong ad, ei in the second element of the personal name. Lindkvist, p. 68, took the first element of Colestainthorpe to be an unrecorded OWScand counterpart of OSwed Ko/sten, *Kolsteinn. In view of the rarity of the name in OSwed and its apparent absence in OWScand and Danish sources, however, it may perhaps be more plausible to suggest that these English examples reflect an AScand formation which originated in the Danelaw rather than a name which was imported directly from Scandinavia itself. It is noteworthy that the field name Colstandic 1202 and the above Nf and Do examples of the name in independent use reflect a form of the name which contains OE -stän and not OScand -streinn, -steinn in the second element. In these cases the name is probably best interpreted as an AScand hybrid formed from OScand Kol- and OE -stän which might have arisen by analogy with other compounds in Kol-. The first element of the NRY field name Golstaindale 1160 and of the lost Y place-name Colestainthorpe 1249 is probably also an analogous formation of this type which arose in the Danelaw but with the Scand element -streinn, -steinn as second element. For Nf examples of Scand and AScand personal names in Kolsee above, s.n. KolbiQrn, and below, s.n. Kol(l)sveinn. Note also the following Sf forms: Colbein 1087-1098 (c. 1175-1200) Bury, p. 39 (Wattisfield St)(< ON Kolbeinn, ODan Kulben, OSwed (run.) Kulbreinn); Osbertus filius Colgrim (witn.) 1114-1119 (c. 1175-1200) Bury, no. 107 (charter of Abbot Albold of Bury St Edmunds conceming land which the deceased Gocelin of Loddon Nfhad held and landat Thurston St); Nicolausfilius Colegrim 11861188 (c. 1270) BuryS, p. 51, Nicole Colegrim, Nicolaus Colegrim ibid., p. 52 (Hopton St) (Colgrim, Colegrim is ON Kolgrfmr).

61 On the use of Latinized -stanus for OScand -sti:einn in personal names found in medieval Swedish records see Modeer, NoB, xl (1952), p. 65, and Modeer, Personnamn i Ktb, pp. 130--131, where it is shown that this feature is ultimately the result of English influence (cf. OE -stän). 62 Lind and LindS do not record an ON *Kolsteinn and an ODan *Ko/sten is not found in DgP. 63 Do examples are: Colestan' et Ediuam uxorem eius 1201 AssRoll 208 (Pleas, ii, no. 718), Editha Vidua que fuit uxor Colstan 1213, CRR 59 (Cur, vii, p. 4).

278

Kol(I)sveinn, Kulsvreinn, Colswein

ON Kol(l)sveinn, OSwed (run.) Kulsvreinn, AScand Colswein a) (i) Colesueinescroft c. 1215 Der, fo. 241a, c. 1220 Der, fo. 217b(i), c. 1235 Der, fo. 235a, Collewenescroft c. 1220 Der, fo. 218a(i), Colsweinescroft, Colesueinnescrof c. 1235 Der, fo. 227b(i), Collesueinescrof ibid., fo. 234a (f. n. in Saddlebow). Colswainyshil 1251 Ely (f. n. in Terrington St Clement) (OKS) (Sandred, OUÅ, 1972, p. 45). b) (i) Colsweyn (acc.) (sokeman) 1101-1107 (14th) Holme, 64 Appendix A (Saxlingham). - Coleswen de Gaywde (abl.) (witn.) c. 1175 Wymondham, fo. 53b(iii) (Gaywood). -Closwein de Bee (acc.), predictus Closwein, predicto Closwein, dat., Closwein (dat., gen.), predicto Colswein, dat., Colswein (bis) (gen.) 1209 FF (PRS NS, xxxii, no. 179) (Gaywood).

Several examples of a personal name Kollsveinn are recorded in lcelandic records (Lind, col. 709). ON Kollsveinn is taken by NK VII, p. 111, to be an original byname. It would belong, in this case, to the adjective ON kollsveinn m. 'a bald or shaven-headed youth'. It might, however, be suggested that the double -Il- in the name form Kollsveinn is merely orthographic and that the form stands for an original Kolsveinn which would be a compound of a similar type to ON KolbiQrn, Kolgr{mr etc., though the absence of spellings with a single -/- among the forms grouped by Lind sub Kollsveinn might tend to speak against this. Note, however, the Swedish runic form kulsu--ns (gen.) (SRU, no. 177). SRU, i, p. 274, gives the base ofthis form as an OSwed (run.) Kulsweinn but includes the possibility that the first component of this form might reflect Koll-, Kull- and campares the name with the Icelandic Kollsveinn. Björkman, NPN, pp. 85-86, was cautious in his discussion of English examples of Colswe3en, Colsuain etc. and suggested that, if these forms are to be regarded as standing for a 1. OE Colswegen (and not *Coll-swe3en), the possibility must be included that the name was formed in England through analogy with other name in Col- < OScand Kol- and -swe3en < OScand -sweinn, -sveinn. 65 In view of the rarity of the Scand Kol(l)sveinn, it would be highly plausible to regard the English Colswe3en, Colswein, Colswain, Colsuain etc. as an AScand formation which originated in the Danelaw. Colswe3en, Colswein, Colswain, Colsuain etc. is quite well attested in English records of the 1. OE ande. ME periods and also forms the first element of several medieval field and minor names (NPN, pp. 85-86; ZEN, p. 57; PNDB, p.

64 This document was drawn up in the period 1101-1107, but the entry relating to the sokeman Colsweyn concems events of 1089 or soon after (see Stenton, EHR, xxxvii (1922), pp. 225, 229 and n. 1). 65 Ekwall, ELPN, p. 79, took the base to be 'ON Kolsveinn, late OE Colswegen, Colsuen etc.' For further discussion of names containing OScand Kol- in English records see above, s. nn. Kolbi9rn, Ko/sten.

279

Korpr, Korp

307; ELPN, p. 79; SPNLY, pp. 179-180; von Feilitzen, PNWinton, p. 153; Lindkvist, p. 68; PN NRY, p. 41; PN D, p. 160; PN ERY, p. 324; PN Nt, pp. 278, 285; PN WRY, iv, p. 143; Insley, JEPN, x (1977-1978), pp. 52, 63). Note also the following examples: Colswegen Wml (c. 1072-1074) SCBI, xii, nos. 65, 66 (moneyer, London); Colswein (witn.) 1. 11th (before 1093) Exeter Book, fo. 6a (D); Colsuein 1087-1098 (c. 1175--1200) Bury, pp. 33, 43 (Woolpit, Litlecerce St); Colswein c. 1100 Exeter Book, fo. 7a (member of a gild at Woodbury D): Colsuain de Hagnebi (abl.) (witn.) c. 1170-1178 RevesbyCh, no. 2 (Stickney L; Hagnebi is Hagnaby in Bolingbroke wapentake L); Colsuanio coco, dat., 1201 CRR 25 (Cur, ii, p. 17) (The Bosses in Shenstone St); Colseino filio Godwini, dat., 1219 CRR 71 (Cur, viii, p. 142) (Lambeth Sr).

ON Korpr, OSwed Korp (byname) a) (i) Corpestih 1086 DB, fo. 158b, Corpestig ibid., fo. 225, Corpsty ibid., fo. 193, Corpesti 1202 FF (PRS NS, xxvii, no. 371) (Corpusty, South Erpingham H).

This byname belongs to the appellative ON korpr, OSwed korper m. 'a raven'. ON Korpr is recorded in Iceland at the end of the 12th century and in the 13th century, while in OSwed both a strong form, Korp, anda weak form, Korpe, are on record (LindBiN, col. 215; XenLid, p. 97; NK VII, pp. 251, 274; Franzen, Vikbolandet, p. 194; Modeer, SvP, pp. 96, 102). DBS, p. 84, s.n. Corp(e), gives 12th and 13th century examples of a byname Corp from Y, 0 and Sf, taking the Y and Sf examples to belong to ON korpr but taking the O example to probably belong to OFr corp 'raven'.

ON Krokr, ODan, OSwed Krök a) (i) Crokestona 1086 DB, fo. 169, Crokeston' 1219 CRR 71 (Cur, viii, p. 76) (Croxton, Gallow H). - Crokestuna 1086 DB, fo. 136b, Croxtun' 1202 FF (PRS NS, xxvii, no. 445) (Croxton, Grimshoe H).

This name is an original byname, 'crook-backed', cf. ON kr6kr m. 'a hook', or possibly 'cunning, artful, disposed to underhand ways' (LindBiN, col. 221; NK VII, p. 51; Modeer, SvP, p. 36; SPNLY, p. 181). ON Kr6kr appears infrequently as a personal name in 14th century Norway but is fairly well attested as a byname in Norway from the latter part of the 11 th century onwards and in lceland from the time of the settlement onwards (Lind, col. 721; LindBiN, cols. 220-221). 280

Krokr, Krök ODan Krok may be known from runic Danish and is found as a personal name in Danish records from the 15th, 16th, and early 17th centuries, in addition to forming the first element of several Danish place-names (DR, col. 676; DgP, cols. 804, 1685; DS Il, p. 142; DS IV, p. 546; DS IX, p. 169; DS XII, p. 82; DS xm, p. 245; DS XIV, p. 167; DS XVI, p. 126; SkOA m, p. 74; SkOA XVI, p. 131). Note also the ODan byname Krok which occurs in medieval Danish records from the second half of the 12th century onwards (DgP Il, cols. 616-618). In OSwed the personal name occurs as runic KrokR and also appears in medieval and 16th century Swedish records (SRÖg, nos. 98, 194; SRU, nos. 32, 191, 432, 604, 866, 1156; Lundgren-Brate, p. 155; Fredriksson, Västgötska dopnamn under 500 år, p. 182; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 21, 130). The byname Krok is also well attested in medieval Sweden (XenLid, p. 101; Modeer, SvP, p. 101). ON Kr6kr, ODan Krok, OSwed (run.) KrokR, OSwed Krok(er) has been taken to form the first element of several place-names in England (DEPN, pp. 132-134, s. nn. Crosby Li, Croxall, Croxby, Croxdale, Croxden, Croxley, Croxteth, Croxton; Lindkvist, p. 115; PN NbDu, pp. 58-59; PN La, p. 114; PN NRY, pp. 205-206, 268; PN Ess, p. 591; PN 0, p. 81; PN Db, pp. 631--632; PN Ch, ii, p. 236; SPNLY, p. 181; SSNY, pp. 24--25; lnsley, JEPN, x (19771978), p 43). 66 In independent use the name is infrequent in English records. In the OE period it is represented by Croc 1017-1023 SCBI, xiii, nos. 1373, 1374, 1023-1029 SCBI, xiii, nos. 1375, 1376, SCBI, xvii, no. 320, 1029-1035 SCBI, v, no. 261, 1038-1040 SCBI, v, no. 272 (moneyer, Chester). An early post-Conquest example is Croc p' (witn.) 1. 11th (1. 13th) ASChR, Appendix I, no. 5 (Brk). The name is also found in London in a document of 1084 and in L in a record of c. 1190--1200 (ELPN, p. 79; SPNLY, p. 181). It should be noted that it has been recently suggested that Crooksby NRY, Crosby NRY, Land Croxby L may all contain the Scand appellative kr6kr m. in the sense 'nook, remote corner', or, in some cases, a cognate OE *croc, and that Croxall Db, Croxden St and Croxton L, Lei (2), St, contain OE *croc, rather than the Scand personal name Kr6kr (Fellows Jensen, Onoma, xix (1976 for 1975), p. 456; SSNEM, pp. 42-43, 375). The general rarity of the personal name Kr6kr in independent use in English records makes it rather surprising that it should apparently form the first element of such a relatively large number of English place-names. This might be taken to support Fellows Jensen's interpretation, but the persistence and ubiquity of the genitival -estends, on the other hand, to support derivation from the personal name. Cf. also above, s. nn. Klakkr, Klyppr. 66 Note, however, that PN Hrt, pp. 81, 247, takes Croxley Hrt to contain an OE *Crocc as first element. Ekwall, however, took Croxley to contain the ScandKr6kr (DEPN, p. 133, s. n.).

281

*Kylfu-vc;_,rör Note also that Croc, Crock appears as a byname in the ME period. ON Kr6kr etc. may also occur in Normandy, though Adigard des Gautries has suggested that the French surname Croc(us) isa possible alternative here (Adigard, pp. 222-223, 366, 455-456). Cf. Rainaldus filius Croc, Rainaldus Croc 1086 DB, (tenant-in-chiefHa). Tengvik, pp. 179,216, gives the etymon of Croc in this case as ON Kr6kr etc. Rainaldus (filius) Croc was a Norman and there is no objection to taking the personal name Croc, Croch in 11th century Normandy to represent ON Kr6kr etc., though the byname Croc in Normandy is perhaps more plausibly explained as standing for the French Croc(us).

OScand *Kylfu-v9rör a) (i) Culuertestuna 1086 DB, fos. 127, 276b, Culuerstestuna ibid., fo. 153b, Kiluerdestun' 1202 FF (PRS NS, xxvii, no. 445), Kiluerstone 1257 Ass (OKS) (Kilverstone, Shropham H).

Ekwall, DEPN, p. 275, s.n. Killerby, interpreted the first element of Kilverstone Nf as the personal name Kilvert which he in tum regarded as an unrecorded ON byname *Kylfu-vr;rör, 'one who defends the prow of the ship'. The derivation of the first element of Kilverstone Nf from an unrecorded ON *Kylfu-vQrÖr presents no real difficulties. The -u- of the first syllable of the DB Culuer(s)testuna can be taken to reflect the AN spelling for OE, OScand /y/ (see PNDB, § 19; SPNLY, § 37). Subsequent spellings in -i- would then represent the normal ME unrounding of OE, OScand /y/. The spellings -uert-, -uerd- in the second component of the first element of this place-name must, if the first element is to be regarded as ON *Kylfu-vorör, be taken to reflect the replacement of ON -VQrÖr by AN -uert, -uerd < OE jerö < OE -freö < OE -friö, AScand jerö < ON jrt/)ör, -friör (see PNDB, § 14; SPNLY, §§ 21, 43). These considerations lead on to the problems raised by the English personal name forms Chelbert, Chilbert, Chiluert, Kilvert. 67 Von Feilitzen, PNDB, pp. 303-304, and following him, Fellows Jensen, SPNLY, § 16.vii, considered three different explanations of the initial Chel-, Chil-, Kil- in these forms to be equally possible. These are as follows: i) These spellings reflect the introduction of a short variant Kil-, Kel-, by the side of Ketel-, Kytel-, Kitel-, in AScand compounds with Ketil- as first element by analogy with

67 For examples of Chelbert, Chilbert see NPN, p. 80; PNDB, p. 302; SPNLY, p. 170. For examples of Chiluert, Kilvert see the following: NPN, p. 81; ZEN, p. 54; PNDB, pp. 215 and n. 4, 305 and n. 1; DBS, p. 205, s. n. Kilvert; SPNLY, p. 171; DEPN, pp. 136,275,276, s. nn. Culverthorpe, Killerby, Killerwick, Kilverstone, Kilwardby; PN NbDu, p. 127; PN La, p. 205; PN NRY,pp.84, 103-104,245;SSNY,p.31;SSNEM,p.57.

282

*Kylfu-vQrör the interchange of -ketel with -kel and -kil in compounds in which these form the second element. ii) These spellings reflect a phonological development, viz. 1. OE Ketel-, Kytel-, Kitel> Ket{-, Kyt!-, Kit!- > Kel- (Chel-), Kil- (Chil-) (with loss of the dental as a result of assimilation to the following -l- owing to AN influence). iii) Kil-, Kel- has nothing to do with ON Ketil- etc. but represents a completely different name element whose form and provenance cannot be settled by means of the material at present available. A.H. Smith, PN NRY, p. 103-104, suggested that OE Ceolfrip, Ceolferö might be a possible alternative toan ON (sic!) *Ketilferö for the etymon of the English Chiluert. On phonological grounds, however, the possibility of OE Ceolfrip, Ceolferö need not be considered here. 68 Björkman, NPN, p. 81, suggested that Chiluert might perhaps reflect an AScand hybrid formation *Chetelweard, *Ketelweard. Von Feilitzen, PNDB, p. 205, rejected this explanation and Ekwall's derivation of the name from an unrecorded ON *KylfuVQrÖr,69 since he considered that the invariable -uert, which represents a pre-DB jerö < jriö, in the second element of Chiluert, Kilvert, is incompatible with these explanations. Von Feilitzen, PNDB, p. 215, tentatively interpreted Chiluert, Kilvert as possibly from an ON *Ketilfnpör, ODan *Ketilfrith and (ibid., p. 302) grouped forms in Chelbertus and Chilbert(us) s.n. Ketilbert. Von Feilitzen's comments about the interpretation of the first element Chil-, Chel- have been noted above. Von Feilitzen, PNDB, p. 215, n. 4, also remarked that: Early spellings with u in the first syll. would seem to point to *Cylferö < anglicized *Cytelferö, though it is possible that AN scribal association with OFr culvert ( -e) (see PNDB, §§ 45, 150; SPNLY, § 147). For further examples and discussion of the Scand name in independent use in England see the following: NPN, p. 95; PNDB, p. 324; Tengvik, pp. 190,396; SPNT, p. 144; SPNLY, pp. 194-195; Coinage ofEdgar, p. 201; von Feilitzen, PNWinton, p. 165). An interesting example of an Anglicized form of ODan Manni is provided by the form Mannan eorl (acc.) 921 (c. 925) ASC (A). This Manna was one of the Danish jarls slain at the battle of

10

For further discussion of OE Manna in independent use see Redin, p. 52.

299

MolSi Tempsford in 917, and was the son of Toglos eorl, 11 who also fell in this battle. The evidence for ODan Manni, OSwed Manne in Normandy is ambiguous (see the discussion of Adigard, pp. 225-226, 367 and n. 65), and cannot safely be used in any discussion of this name.

ONMoöi a) (i) Mothetuna 1086 DB, fo. 113b, Modetuna ibid., fös. 128b, 129b (bis), Mothetun' 1202 FF (PRS NS, xxvii, no. 416), Mouton' 1209 AssRoll 558 (Pleas, iv, no. 4183) (Moulton St Mary, Walsham H).

Ekwall, DEPN, p. 333, s.n. Moulton Nf nr Yarmouth, gave an OE *Möda, a short form of OE names in Mod-, -mod, as the first element of Moulton St Mary, but added that it might alternatively contain ON M6öi, OSwed Modhi. ON M6öi is either an original byname, 'the angry one' (cf. ON m6ör m. 'anger, wrath'), or a short form of names in -m6ör (SPNLY, p. 197). Cf. OSwed (run.) Moöa fem., a short form of such OSwed fem. names as Arinmodh, Ingemodh and K~tilmodh (SRSö, no. 297). ON M6öi is recorded as a mythological name, and apparently forms the first element of several Norwegian place-names (Lind, col. 774; LindS, col. 626). Lundgren-Brate, p. 177, suggests an OSwed *Modhe as the first element of Mosta in Uppland.

AScand *Mrerle-Sveinn b) (i) Marlesuein 1100-1135 (13th) Rams, iii, p. 267, Guibertus Marlesuein 1161-1177 (13th) Rams, iii, p. 269 (Ringstead and Holme next the Sea).

This name is a compound in which the second element is the common Scand personal name ON Sveinn, ODan Swen, OSwed Sven.

11 Tog/os, which also forms the first element ofToulston WRY, is perhaps best explained as reflecting a Scand byname (or original byname) *Tauglauss or *Toglauss, 'one who has no rope for his ship' (ON taug f. 'a string, a rope', or ON tog n. 'a rope, a line, a tow-rope', + the negative suffix ON -lauss (see NPN, p. 142; PN BdHu, pp. 272-273; PN WRY, iv, p. 80; SPNLY, p. 286). I. Lehiste, PMLA, lxxiii (1958), p. 18, following an unpublished suggestion of Stefan Einarsson, proposed that Icel tog could also mean 'the long coverhair of sheep', and suggested that Tog/os could be a nickname. If this second explanation is followed, the nickname would have the sense 'without outer clothing' (SSNY, p. 130). The Tog/os eorl of ASC (A) has been identified with a Toli comes, who is recorded in theLiber Eliensis also as having been slain at the battle ofTempsford (PN BdHu, p. 272; Whitelock, Foreword to LibEI, p. xiv). If this is the case, then Tog/os in ASC (A) must reflect the byname of a man whose baptismal name was Toli (cf. PN BdHu, p. 273).

300

*Mrerle-Sveinn The first component is difficult. SPNLY, p. 196, suggests that it may be an OE personal name *M:irla, a short form of such names as OE M:irheard. Björkman, NPN, p. 94, suggested tentatively that Mcerle- might possibly be identified with the initial Mcerle- of the place-name form Mcerlebeorge 1110 (1121) ASC (E) (Marlborough W, the first element of which is taken by PN W, p. 298, to be the unrecorded personal name OE *M:irla), and goes on to pose the question whether 1. OE Mcerleswegen (< AScand *Mcerle-Sveinn) could be interpreted as 'Swegen from Mcerlebeorg'. Björkman cited as comparison the ON formations Koll-Barör and Molda-Gnupr, whose first components are shortened forms of the place-names Kolla}Jf)r and Moldatun, respectively (NPN, p. 94, n. 2). Derivation of the first component of *Mcerle-Sveinn from a place-name would appear to be more plausible than derivation from an unrecorded OE personal name *M:irla. The form would fit an OE (ge)m:ir-leah 'wood or clearing at a boundary'. OE (ge)m:ir-leah has given rise to the English place-names Marley D, Du, and Mearley La (DEPN, pp. 315,319, s. nn.; PN D, p. 310; PN NbDu, p. 139; PN La, p. 78). The first component of AScand *Mcerle-Sveinn, however, might be more plausibly interpreted as belonging to a last minor name rather than to one of the above-mentioned place-names. AScand *Mcerle-Sveinn would, thus, denote 'Sveinn who dwelt at the wood or clearing on the boundary'. The earliest bearer of the name to be recorded is the famous sheriff of L of Edward the Confessor's last years, who held extensive properties in Co, D, Gl, and So, as well as in Land Y (NPN, pp. 93-94; Björkman, Festermen, pp. 4, 15; Lindkvist, Anglia Beiblatt, xxxiii (1922), p. 142; PNDB, p. 326 and n. 3; SPNLY, p. 196). Subsequent bearers of the name were probably named after this first recorded bearer. In 1068 and again in 1070 this man fled to Scotland, and the name is evidenced in 12th century Scottish records (Barrow, The Kingdom of the Scots, p. 288 and n. 2).

301

N

ON, ODan Narfi, OSwed Narve (i) Naruestuna 1086 DB, fo. 255b (lost, in, Clavering H).

The personal name Narfi has been linked etymologically with an unrecorded ON *nQrr 'narrow, tight' (DgP, col. 981). ON Narfi is extremely frequent throughout the Middle Ages in both Norway and Iceland from the Viking period onwards (Lind, cols. 782-783; LindS, cols. 627--635). In OEScand the name appears in medieval records in both Denmark and Sweden, but it is very uncommon (DgP, col. 981; Lundgren-Brate, p. 180; Modeer, 'Fornsvenska personnamn i en regestsamling', p. 10).

302

0

AScand *Obbi a) (i) Obbiscroft 1. 13th Binham, fo. 99a (f. n. in Edgefield).

An AScand *Obbi would be an acceptable hypocoristic form of AScand Ös-

be(o)rn, Ösb(e)arn < Scand ÄsbiQrn. A parallel formation is OE *Obba, which is to be interpreted as a hypocoristic form of OE Ösbeorht, and which forms the first element of the C field name Obbeleie PN C, p. 337 (Insley, JEPN, x (1977-1978), p. 68). Cf. also OSwed Obbe, a hypocoristic form of Aupbiorn etc. Note also the English forms Leurico Hobbesune, dat., TRE DB, fo. 337 (a liber homo, Lefsune, at Somersham Sf was commended to this Leuricus), Leurico Hobbesuna, dat., ibid., fo. 404b (one liber homo, Walduinus, at Offton Sf, was commended to this Leuricus), Leurico Obbesune, dat., ibid., fo. 405b (one liber homo, Anschillus, at Bricett Sf was commended to this Leuricus). Tengvik, pp. 157-158, derived the first element of the patronymic (H)obbesune, -a, from an OE personal name *Hobba. It is possible, however, that the forms in Hobbe- contain an initial AN inorganic H-. If this is the case, then AScand *Obbi < AScand Ösbe(o)rn etc., OE *Obba < OE Ösbeorht, and OE *Obbe fem., a hypocoristic form of OE Ösburh fem., must all be considered as possible sources for the etymon of the first element of this patronymic formation.

ON Oddgeirr, ODan Oddger, OSwed (run.) OddgreiRR, OSwed Odger b) (i) Ogerus 1086 DB, fo. 167 (Didlington). -Turchetilfilius Ogeri 1165-1166 P, p. 22 (Lynn). 1 - Willelmo filio Ogg(er). Philippo fratre eius, abl. (witn.) c. 1185 CAcre, fo. 33a(i) (Tinedwde and Foxemere in Kempston), Philippusfilius Ogeri 1187-1188 P, p. 64 (Nt). -Decena Ogeri Siwini 1198 P, p. 88, 1199 P, p. 270, de decena Ogeri 1200 P, p. 135 (Nf or St). -Andreasf Ogeri 1199 P, p. 286 (Nf or Sf, prob. Nt). - de redditu Ogeri le palmer c. 1200 CAcre, fo. 60a(i) (South Raynham).

ON Oddgeirr occurs in Iceland throughout the Middle Ages from the time of the settlement onwards, albeit infrequently, but does not seem to have been in

1

This is the same person as Turchetilfilius Oggi 1166-1167 P, p. 21 (see below, s. n. *Oggi).

303

Oddgeirr, Oddger, OddgreiRR, Odger use in Viking and medieval Norway (Lind, cols. 800--801). There are several 14th, 15th, ande. 16th century Norwegian forms in Odgeir, Odgceir, Odger, Odgher, Odgerus, -garus, -gharus, but these are grouped by Lind s.n. Auögeirr (Lind, col. 97; LindS, col. 125). An ODan Oddger may be represented by forms of c. 1390, 1596 and 1613, but this is doubtful, and both OHG Autger, Ödger, and ODan 0thger have been suggested as possible altematives (DgP, cols. 1015, 1619). In OSwed the name occurs as runic otkaiR, acc. (SRU, no. 1123), and the forms Odgeer ij Ambatzbcerghe and Odgher Asmundzsson, from documents of 1359 and 1421 respectively, are given by Lundgren-Brate, p. 183, s.n. Odger. The form Odgher Asmundzsson, however, is from Jämtland, an area which was Norwegian during the Middle Ages, and is grouped by Lind, col. 97, s.n. Auögeirr (cf. above). For the most part, however, the above Nf forms in Ogerus, Ogg(er) are unlikely to reflect the rare ON Oddgeirr etc., but rather belong to a reflex of OHG Ötger etc. It is possible that the Scand name need only be considered in the case of Turchetil filius Ogeri 1165-1166 P, p. 22, whose name is rendered as Turchetilfilius Oggi in the Pipe Roll ofthe following year (1166-1167 P, p. 21). Oggi can be acceptably interpreted as an AScand hypocoristic form of Ogger < ON Oddgeirr etc. 2 The possibility that this Ogerus/Oggi was of native AScand descent might be strengthened by the fäet that his son has the name Turchetil < ON Porkell etc., though it should be noted that Turchetil corresponds to the form normally taken by ON Porkell etc. in Norman records in the period from the late 10th century to c. 1200 (see below, s.n. Porkell). OHG Ötger, West Frankish Autgarius, Otgarius, OSax Ödger etc. is well attested in Continental records of the Carolingian and post-Carolingian periods, though it is uncommon after the 12th century (Förstemann PN, cols. 192-193; Geuenich, pp. 60, 93,210; Longnon, i, pp. 286, 312-313; Morlet, p. 43; Störing, p. 193; Kremer, p. 73; Mansion, pp. 41, 64, 206, 210; TavemierVereecken, pp. 130, 132; Schlaug I, pp. 131-132; Schlaug Il, p. 136). In OFr West Frankish Autgarius, Otgarius became Oger, Ogier (Longnon, i, pp. 286, 312). The popularity of this name is doubtless due to the fäet that it was borne by a hero of romance, 0 gier le Danois, one of Charlemagne' s paladins in the Chansons de geste. 3 The form Ogerus already occurs in Flanders in a record of 987-994, which is preserved in an 11th century copy (Tavemier-

2 For AScand hypocoristic formations of this type see von Feilitzen, Notes, pp. 64----66, and above, s.n. Hab(b)e. It should be noted that in an AScand milieu Ogger < OHG Otger etc. could also have given rise to a hypocoristic variant Oggi by analogy with other hypocoristic formations of this type. For further discussion of Oggi, see below, s. n. 3 For a conspectus of forms relating to Ogier le Danois in the Chansons de geste, see Langlois, pp. 495-496.

304

Oddgeirr, Oddger, OddgreiRR, Odger Vereecken, p. 130). Note also the Norman form Oggerius (burgess of Rouen) (witn.) c. 1037-1040 (14th) Fauroux, no. 95. In England OHG Ötger etc. is represented in the OE period by Odgar (dat.) 942-951 (1. 13th) ASWills, no. 1 (the will of Bishop I>eodred of London). The post-Conquest forms in Ogerus etc. correspond to the French form of the name, and many of the bearers of the name in England must have been Frenchmen or of French descent. A clearly Continental bearer of the name was Oger the Breton, the lord of Bourne L in 1086 (Stenton, First Century, pp. 25-26). Cf. also Oger the seneschal (Ogerus Dapifer 1165-1166 P, p. 16 et seq), sheriff of Nf and Sf from 1163 to 1170,4 who would also have been of Continental origin. The use of the name in England must also, in many cases, reflect the popularity of the Chansons de geste, 5 cf. Oggerus, Oggirus, Oggerum le Daneis 1202 FF (SPNLY, p. 203) (Allerwash Nb), who was either named after Ogier le Danois or at least received his byname through association with this figure of romance. For further examples of Odger, Ogger, Ogerus etc. in English records see Forssner, p. 197; Tengvik, p. 192; DBS, p. 254, s.n. Odgear; SPNLY, p. 203. The name also forms the first element of the lost Hu place-name Ogerston (Ogerestan(e) 1185 (c. 1200), 1227, 1253, 13th, Oggereston 1189, Oggerston 1305, 1360, Oggeston 1335, Ogerston 1597) (PN BdHu, pp. 192-193). The French cycle concerning Charlemagne and his paladins was transmitted to Scandinavia in the course of the 13th century and gave rise in Norway to the Karlamagnus saga, in which Ogier le Danois appears as Oddgeir danski.6 In Denmark Ogier le Danois appears in the Latin annals of Lund in the period 1250-1265 as Ozsyarus Danus, de Dacia, in MDa translation Ozcer Dan, Orre/dan. The Norse Oddgeir danski of the Karlamagnus saga appears as Udger (Wdger, Ottger) Danske in a Danish version, the Karl Magnus Kn/mike of 1450-1480. From the end of the 15th century Ogier le Danois is known as Holger Danske in Danish popular tradition. It should be noted that the Ogier le Danois of the Chansons de geste does not appear to have been based on a Danish hero at all, but seems to reflect various Continental personages.7 The Ogier le Danois of the French romances, the Scand Oddgeir danski etc., and the English form Oggerus, Oggirus le Daneis 1202 FF, led Forssner, p. 197, to the conclusion that English Odger(us), Ogger, Oger(i)us 'may be of

J.H. Round, EHR, xxxv (1920), p. 489 and n. 6. For a convenient summary of the extant ME versions of the French romances conceming Charlemagne and his paladins see A.C. Baugh (ed.), A Literary History of England: I. The Middle Ages, 2nd ed. (London 1967), pp. 185-189. 6 For the Karlamagnus saga see KLNM, viii, cols. 286-290, s. v. 7 The account of the forms taken by the name of O gier le Danois in Scandinavia and the transmission of the stories connected with this hero of romance which is given here follows that of Niels Lukman, KLNM, vi, cols. 634---637, s. v. Holger Danske. 4

5

305

Oddi, Odde, Udde

ON as well as of OG origin'. Similarly, PN BdHu, pp. 192-193, derives the lost Hu place-name Ogerston from either OHG Ötger etc. or from ON Oddgeirr or ON Auögeirr. ON Auögeirr can be ruled out here due to the absence of any forms containing the diphthong au (or its AScand reflex ou) and the spirant [ö] in the first element. The regularity of the form Oger(i)us speaks for a French origin in the majority of cases, though the occasional confluence of the Norse Oddgeirr with the French Ogier, Oger must be reckoned with. The Norse Oddgeirr was used to render OFr Ogier in Norway in the 13th century in the Karlamagnus saga, and it might well be that some of the examples of Oger(i)us found in England may represent the assimilation of ON Oddgeirr to OFr Ogier, Oger on the part of Norman scribes, though there is no doubt that most of the examples of Oger(i)us found in England are better interpreted as standing for the OFr name.

ON, ODan Oddi, OSwed Odde, Udde b) (i) Rogerumfilium Ode 1202 FF (PRS NS, xxvii, no. 448) (Witchingham). - bruarium Robertifilii Ode c. 1215-1230 Bromh, fo. 79b(i) (Worstead).

ON Oddi etc. has been taken to be a short form of names in Odd- (DgP, col. 1016). Alternatively, it might be either a direct derivative of the appellative ON oddr, ODan od(d) m. 'point (of a weapon)' or a hypocoristic side-form of ON Oddr, ODan Odd, OSwed Udd, a personal name which also belongs to this appellative (Hald, Personnavne i Danmark: I. Oldtiden, p. 71). The interpretation of Oddi as a hypocoristic side-form of Oddr is supported by the frequent appearance of Oddi and Oddr as the name of the same person in Norwegian records (Hald, loc. cit.). ON Oddi is well attested in Iceland from the 10th century onwards, but is rare in Norway, where it first appears in independent use in a document of 1380 (Lind, cols. 801-802; LindS, col. 640). ODan Oddi is not particularly frequent, but is found in medieval Danish records, where it is apparently sometimes confused with Otte (DgP, cols. 1015-1016, 1689). OSwed Odde, Udde is recorded in several runic inscriptions and in medieval and 16th century Swedish records (SRSö, no. 164; SRSmå, nos. 48, 155; SRU, no. 229; Lundgren-Brate, pp. 183,287; Modeer, Östsmåländska bondenamn från 1500-talet, p. 66; Thors, p. 95; Fredriksson, pp. 213,215; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 19, 20, 23, 149). ON Oddi etc. is Anglicized in OE sources as Oda, cf. ODA bisceop (witn.) 934 (12th) ASChR, no. 25, Odan rercebissopres, gen., Oda rercebissop (witn.) 306

Oddi, Odde, Udde 958 (10th) ASChR, no. 32, Oda arce biscop 958 (m. 11th) ASC (D). These forms relate to Oda, a man of Danish extraction, who was bishop of Ramsbury from 927 or 928 to 941, and archbishop of Canterbury from 941 to 958. 8 However, Oda is also used in OE sources to represent the West Frankish name Odo, cf. Oda 887 (c. 900) ASC (A), a form relating to Odo, king of France from 888 to 898. Note also that OHG Otto is rendered in OE sources by Odda, which appears in the annal for 982 in ASC (C) 9 for the names of the German emperors Otto I and Otto II and that of Duke Otto of Swabia. There is also a genuine OE Od(d)a, a hypocoristic form of names in Ord-. Pre-Conquest examples of this name include the following: Oddan ducem prouinci~ Defenum 878 (1596, printed from an e. 11th MS.) JEthelweard, p. 43; Oddan, gen. (witn.) 899-924 (10th) Harmer, SEHD, no. 18; Odda minister (witn.) 901 (12th) BCS, no. 595, 904 (12th) BCS, nos. 611,613; Brihtnoö Oddan sunu 924939 (c. 1140), Brihtnodus Odonisfilius 924-939 (c. 1114) Liebermann, i, p. 181; 10 Odda minister928 (12th) BCS, no. 663,931 (12th) BCS, no. 674,931 (contemp.) BCS, no. 677,931 (13th) BCS, no. 682,932 (12th) BCS, no. 689,932 (13th) BCS, no. 692, 933 (m.-1. 10th) BCS, no. 694, 933 (12th) BCS, no. 695, 934 (contemp.) BCS, no. 702, 934 (14th) BCS, no. 703, 938 (12th) BCS, nos. 730,731,939 (contemp.) BCS, nos. 734, 741,939 (13th) BCS, no. 743,940 (contemp.) BCS, nos. 748,753,940 (13th) BCS, nos. 761,762,940 (12th) BCS, no. 764,941 (1. 12th) BCS, no. 767,941 (12th) BCS, no. 770, 942 (13th) BCS, no. 777, 942-946 (e. 12th) ASCh, no. 28, 943 (contemp.) BCS, no. 780,943 (13th) BCS, nos. 784,789,943 (12th) BCS, nos. 786,787,944? for942 (13th) BCS, no. 801, ODDA miN (witn.) 934 (12th) ASChR, no. 25, Odda öeyn (witn.) 932939 (14th) ASChR, no. 27, Oda minister (witn.) 930 (14th) BCS, no. 669, 938 (12th) BCS, no. 729, 940 (12th) BCS, nos. 758, 763, 941 (15th) BCS, no. 769; Oddan boc m. 11th (contemp.) ASChR, Appendix Il, no. 5 (list ofbooks, probably from Worcester).

A notable 11 th century bearer of the OE name Od(d)a was Odda of Deerhurst (ob. 1056), who was made Earl of D, So, Do and Co after Earl Godwine's exile in 1051, and who was possibly Earl of Wo (and perhaps Gl) for same time after Godwine's return in 1052.ll In general, Oda, Odda in OE records can be taken to represent OE Od(d)a, except in cases where we are clearly concerned with foreigners or persons of foreign descent. To the second category belong the use of Oda for the name of Odo, king of France from 888 to 898, and of Odda for the names of the Ger-

8 For Oda's family connections see above, s.n. Äskell, n. 143, and the works cited there. For the dating of Oda's episcopate and archiepiscopate see the discussion of M.A. O'Donovan, Anglo-Saxon England, i (1972), pp. 33-34, Anglo-Saxon England, ii (1973), pp. 98-99. 9 The MS. ism. 11th in date. 10 In the Latinized form Odonis, gen., West Frankish Odo has been substituted for OE Od(d)a. 11 For Odda of Deerhurst see ASChR, pp. 456---458. ASChR, p. 499, poses the question as to whether the above-mentioned form Oddan boc m. 11 th (eontemp.) ASChR, Appendix Il, no. 5, could denote a book which has been owned by or presented by Odda of Deerhurst. Okasha, no. 28, incorrectly took the name of Odda of Deerhurst to represent ON Oddi (see von Feilitzen, JEPN, iv (1970--1971), pp. 66--67).

307

*Oggi

man emperors Otto I and Otto Il in the Anglo-Saxon Chronicle. The same is also true ofthe name of Archbishop Oda of Canterbury. 12 According toa late 10th century Ramsey tradition, Archbishop Oda's father was a Dane, who had come to England in the 9th century with Ingvar and Ubba, and who had remained a pagan until his death. 13 In view of Oda's pagan parentage, any possibility that his name could represent OE Od(d)a is to be ruled out, and it can only be explained as an Anglicized form of the Scand Oddi. The post-Conquest form Ode is more complex. Cameron, Newark, p. 9, explained it as representing West Frankish Odo, adding, however, that it could represent OE Od(d)a. Reaney, DBS, p. 254, s.n. Oade, also took the Continental name to be the base. However, ON Oddi etc. cannot be ruled out. The simplification of -dd- to -d- was a frequent feature of AN orthography, and the replacement of the final -i of Oddi by -e can be attributed to Anglicization (PNDB, §§ 147, 150; SPNLY, §§ 143, 147). For further discussion of ON Oddi etc., West Frankish Odo and OE Od(d)a and their appearance in English sources see the following: NPN, pp. 99-100; Forssner, pp. 198-199; Ekwall, NoB, vi (1918), p. 175; PNDB, p. 333 and notes 1-5; Tengvik, pp. 161, 192, 398; SPNLY, p. 202; Smart, Moneyers, pp. 247, 253, 257, 270--271; von Feilitzen, SCBI, xi.i, p. 12 and n. 4; Okasha, no. 28; von Feilitzen, JEPN, iv (1970--1971), pp. 66--67; Coinage ofEdgar, p. 202.

AScand *Oggi b) (i) Turchetilfilius Oggi 14 l166-1167 P, p. 21 (Lynn).

This form appears in the preceding Pipe Roll as Turchetil filius Ogeri, and Oggi is, thus, here a hypocoristic form of Ogerus, which in this particular case is probably best interpreted as belonging to ON Oddgeirr etc., though on förmal grounds a reflex of OHG Ötger, West Frankish Autgarius, Otgarius etc. cannot be excluded (see above, s.n. Oddgeirr, and n. 2). 15 It is interesting to note that an OE Ogga, a hypocoristic form of OE Ordgär, is also on record (PNDB, p. 335; Coinage ofEdgar, p. 202, n. 5).

For a selection of forms see above. See D. Whitelock, Saga Book of the Viking Society, xii (1937-1945), pp. 169-170. 14 OggeCR. 15 The appearance of Turchetil filius Oggi as Turchetil filius Ogeri in the preceding Pipe Roll renders irnpossible the suggestion of von Feilitzen, Notes, p. 65, that Oggi is perhaps here an AScand hypocoristic form of ON Auögrimr. 12

13

308

61arr, ö1ar ON Olafr, ODan, OSwed Öläf a) (i) Unleues gare c. 1170-1180 CAcre, fo. 28b(i), Anleuesgore ibid., fo. 29a(i) (f. n. in Palgrave). b) (ii) terram Olofc. ll95-1200Der, fo. 102a(i), Oloffilii Godman (gen.),Robertifilii Olof, gen., Radulfifilii Olof, gen., c. 1205 Der, fo. 102b, Radulfusfilius Olof et Robertus frater eius c. 1210 Der, fo. 11 la(iii), cum hornagio Radulfifilii Olof et Robertifratris eius ibid., fo. 111 b (Hillington).

ON Olafr etc. belongs to a PrScand *Anu-laibaR (NK VII, p. 85; Otterbjörk, Svenskaförnamn, p. 115; SPNLY, p. 204). As Ölafr, Öleifr the name is extremely common in Norwegian and Icelandic sources throughout the Middle Ages (Lind, cols. 810-816; LindS, col. 650). A side-form Aleifr is represented by runic alaib (acc.) in a lost inscription of the first half of the 11 th century from Tanberg in Buskerud (LindS, col. 650; NiyR, no. 97). ODan Öläf, Ölef is also extremely frequent throughout the Middle Ages (DgP, cols. 1022-1043, 1689). The Sk}Qldunga saga has the form Aleifus for the name of one ofthe kings of Lejre (DgP, col. 1022). ODan also has two examples ofrunic ulaf(acc.), dating from the Viking period (DR Per. 2.2) (DR, nos. 289, 314). OSwed (run.) ÖläfR, OlreijR, ÄlreifR, and ÄlefR are frequent, and the name is well attested in medieval Swedish records (SRÖl, nos. 34, 37; SRÖg, no. 129; SRSö,nos.36,52-54,65,67, 163, 189,276,317,340;SRSmå,nos. 78,114; SRVg, nos. 9, 12, 51, 181,223; SRU, nos. 130, 145, 162, 195, 199,231,233, 235,285,317,460,481,565,685,687,695,705,713,718,867,878, 1075; Lundgren-Brate, pp. 185-186; Modeer, Personnamn i Ktb, pp. 12-13, 27-28, 31, 32, 63 and n. 2, 64, n. 1, 65 and n. 2, 69, 73, 80-81, 103-104; Per Henningsson, 'Till uppkomsten av namnformen Olof', Personnamn från medeltid och 1500-tal, ed. I. Modeer (Stockholm 1957), pp. 33-55; Modeer, SvP, pp. 28,43,53,58,97,98,99, 100,101,102,103,104,107) In the early Viking period ON Ölafr etc. seems to have been primarily a royal name, but from the 11 th century onwards it is frequent in all social groups, doubtless as a result of the popularity of the cult of St Olaf of Norway (ob. 1030). The Scandinavian sources themselves do not contain forms showing the original nasalization of the first element of the name, but such forms are well attested in the British Isles. This earlier nasalized form is contained in the runic forms qulaibr liutulbsunr (Aleifr Lj6t6lfssonr), which is contained in a Manx runic inscription of c. 950-1000, and alabr (Ölafr), which appears in an inscription of roughly the same period from Holy Island in the Hebrides (RIGB, pp. 189,229; NiyR, v, p. 230). In Ireland the name is known from the mid 9th century onwards in the forms Amlaip, Amlaim, Amlaib (Marstrander, 309

61arr, 01ar pp. 61-62, 88-89, 106; Smyth, Scandinavian Kings in the British Jsles 850880, p. 280). Note also the following forms from English sources: Anlaf se Swearta 911 (c. 1000) ASC (B), 911 (m. 11th) ASC (C) (Scand chieftain slain at the battle ofTettenhall in 910); Onalajball 913-915 (l. 12th) SD, i, p. 209 (Onalafball received lands in Du between Castle Eden and the mouth of the Wear from the HibemoNorse Viking leader Rregnald, who later established himself as king at York); Anlafe, dat., Anlafes, gen., Anlaf (nom.) 937 (c. 955) ASC (A), Anelaf (acc.) 937 (c. 1000) ASC (F), Onlafum 937 (c. 1165-1170), rexOnlafprima 939 (c. 1165-1170) SD, ii, p. 93, Onlaf(nom.) 940 (c. 1165-1170) SD, ii, p. 94,Anlaf ofYrland (acc.) 941 (for 939/940) (m. 11th) ASC (D), An/af (nom.) (bis), An/af cying 942 (for 941) (1121) ASC (E) (Olaf Guthfrithson (ob. 941 ), king of Dublin 934-941 and king of York 939/940---941 ); filius vero Sihtrici nomine Onlaf941 (c. 1165-1170) SD, ii, p. 94, Anlafe cyninge, dat., 942 (for 943) (c. 955) ASC (A), 943 (m. 11th) ASC (D), Anlafes, gen., 943 (c. 1000) ASC (B), Onlafes, gen., 943 (m. 11th) ASC (D), regem suum Onlaf943 (c. 1165-1170) SD, ii, p. 94,Anlaf Syhtrices sunu (acc.) 944 (c. 955) ASC (A),Anlaf(acc.) 944 (1121) ASC (E), Anlaf Cwiran 949 (1121) ASC (E), Anlaf cyning (acc.) 952 (1121) ASC (E) (Olaf Sihtricson, king of York 941-943, 944, 949-952, king of Dublin 952-980); Maccus filio Onlafi (abl.) 954 (1. 12th) SD, ii, p. 197 (Maccus was the murderer ofEin1a B16ö0x, the last Norse king of York, and Stenton, ASE, p. 362, thought it probable that he was the son of either Olaf Guthfrithson or Olaf Sihtricson); Anlaf 991 (c. 1125-1130), Analauo, abl., 991(c.1114) Liebermann, i, p. 220, Unlaf993 (for991) (1001) ASC (A), Anlaf(nom.),Anelafe cinge, dat.,Anelaf cingc (acc.),Anelaf(acc.) 994 (1121) ASC (E), Anlaf(nom.), Anlafe cynge, dat., Anlaf (acc.) (bis) 994 (m. 11th) ASC (C) (OlafTryggvason, king of Norway 995-1000); Onlaf 1009-1016 Smart, Moneyers, p. 262 (moneyer, Lewes). In DB the name appears TRE as Vnlof in Ha, L, and Nt (PNDB, p. 335). An/af also appears as the first element of Anglezark La, Anlaby ERY and the lost Nth place-nameAnlafestun 963-984 (m. 12th) ASChR, no. 39, retAnlafestune 963-992 (m. 12th) ASChR, no. 40 (DEPN, p. 10, s. nn. Anglezark, Anlaby; PN La, p. 48; PN ERY, pp. 216--217; IPN, p. 73; SSNEM, p. 187).

The form Onlof is recorded in the Reichenau list (Reichenau, pp. 7, 20). This may appear surprising in view of the relative ly late i) in e. ME see above, s.n. Äki, and the literature cited there. Note, however, that it is possible that the -o- of the second element of the Nf Olof could well be short in value due to shortening subsequent to the change ä > i), as a result of weak stress (see Jordan, § 137).

The scribe wrote Onlaf or An/af at first. See B. Dickens, 'The Cult of S. Olave in the British Isles', Saga Book of the Viking Society, xii (1937-1945), pp. 53-80. 18 The ending -ar appears to represent a survival of the Scand nom. ending -r here (cf. also above, s.n. Dagr, n. 2). 16

17

311

OIQf, Ölöf

ON Ol9f, ODan, OSwed Ölöf fem. b) (i) Ollova qwedamfemina TRE DB, fo. 232b (Svttuna near Mileham, Launditch H).

ON Olij etc. is a fem. counterpart of the masc. Ölafr, and corresponds to a PrScand *Anu-laibu (NK VII, p. 108). The name is well attested throughout the Middle Ages in Norway and lceland, but is much less common in ODan, where it appears sporadically in medieval records, and in OSwed, where it is known from runic inscriptions and medieval records (Lind, cols. 816-817; LindS, cols. 650-654; DgP, col. 1045; SRÖg, nos. 112, 198; SRSö, nos. 8, 32, 70,296; SRVg, nos. 13, 50; SRU, no. 864; Lundgren-Brate, p. 187). There is also an ON side-form Ä.ltjf (Lind, LindS, loc. cit.). The above Nf form has been Latinized by the addition of the nom. ending -a of the Latin first declension (see PNDB, § 157; SPNLY, § 152; von Feilitzen, PNWinton, p. 227). Doubling of -l- in the above Nf form reflects AN scribal usage, though such doubling of simple consonants is also found occasionally in 1. OE MSS. (see PNDB, § 146). Von Feilitzen explained -v- in Ollova as reflecting a value [v] taken by the -f of ON Olij (PNDB, § 89). This is, however, impossible since PrScand b (= [v]) bad already been unvoiced to [t] in final positions before 700 (Br!1Sndum-Nielsen, § 232a), and in the Nf Ollova the voiced [v] must be the result of analogy with the voicing of the final [t] of the nominative in the oblique cases in OE in such name elements as OE -läf, OE -wulf etc. (cf. PNDB, § 89).

ON, OSwed (run.) Ormarr, ODan, OSwed Ormar a) (i) Ormerescroft m. 14th Binham, fös. 97b, 98a (f. n. in Edgefield). b) (i) Lefwino Ormarifilio, abl. (witn.) c. 1170 CAcre, fo. 83b(iii) (Thorpland and Wiggenhall).

ON Ormarr occurs in lceland at the time of the settlement, and is known in Norway from medieval records and as the first element of several place-names (Lind, cols. 818-819; LindS, cols. 654-655). There are a few late medieval examples of ODan Ormar, which is also found as the first element of a place-name in -thorp, and perhaps also of a place-name in -lef (DgP, cols. 1050--1051; DS XII, pp. 53-54; S!1Sndergaard, pp. 95-96). In OSwed Ormarr has been noted in runic inscriptions, and the name also occurs in Swedish records of the 15th and 16th centuries, though it is extreme312

Ormarr,Ormar ly infrequent (SRÖg, no. 89; SRSö, no. 109; SRVg, no. 194; Fredriksson, p. 179 and n. 3; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 20, 136). In the 15th century there are occasional examples of Ormarus, Ormer being used to represent ODan Orm, and it seems that there was sometimes a certain confluence between Ormarr and Ormr in late medieval Scandinavia (see DgP, cols. 1049, 1050; Lind, col. 818; Fredriksson, pp. 178-179). In England any discussion of ON Ormarr etc. is complicated by the fäet that OE Ordmrer can be shown to occasionally appear as Ormar(us). Thus in DB the name of a C tenant, Ordmrer of Bellingham, appears as Ordmcer(us), but is rendered as Ormarus and Ormar in the ICC (c. 1180), as Ordmar in IE (A) (c. 1180), and as Hordmerus in IE (C) (1. 12th) (PNDB, p. 336). PNDB, p. 336 and n. 4, takes the St DB forms Ormar (bis), Ormarus and Ordmer all to refer to the same person, but it is not clear from the DB text whether this is, in fäet, the case. A clear case of the use of Ormarus for OE Ordmrer is provided by the Nf form Fulch(er(i)o) filio Ormari, abl. (witn.) c. 1135 CAcre, fo. 37a(i) (Gayton), which also occurs as Fulch(er(i)o)filio Ordm(er), abl. (witn.) c. 1140 CAcre, fo. 33b(ii) (grant by Roger, son of William the Seneschal, to Castle Acre Priory of a mill at Lexham and land at Weasenham), fo. 33b(iii) (confirmation by Drogo, son of William the Seneschal of Gressenhall, of various properties and renders to Castle Acre Priory). The loss of interconsonantal -d- here is almost certainly the result of AN influence, though it should be noted that Orlaf < OE Ordläf occurs as the name of a moneyer of Edward the Confessor (see PNDB, § 103). The occasional interchange of Ormarus etc. with Ordmarus etc. in postConquest records cannot, however, be taken to exclude the possibility that certain of the forms in Ormar(us), Ormer(us) might represent the Scand Ormarr. Von Feilitzen, PNDB, p. 336, grouped DB examples of Ormar, Ormer etc. to examples of Ordmcer sub OE Ordmrer, but added that, in some cases, derivation from ON Ormarr, ODan, OSwed Ormar is formally possible. In 1937, Ekwall took the Sf form Ormer 1087-1098 (c. 1175-1200) Bury, p. 43 (Huntefelde St) to be probably Scand, though he included the possibility that this form could also reflect OE Ordmrer (Ekwall, Selected Papers, p. 97). Ten years later, Ekwall grouped the London form Ormarus c. 1150 (1241-1242) sub OE Ordmrer with the comment: 'This is not OScand Ormarr, which would have given 12th century Ormer' (ELPN, p. 57). More recently, von Feilitzen has taken an 1148 example of Ormerus from Winchester to belong to OE Ordmrer (von Feilitzen, PNWinton, p. 167). It should be noted, however, that examples of OE Ordmrer in which the interconsonantal -d- has been preserved are quite frequent in English records of the post-Conquest period. Cf. the following forms: 313

Ormr,Orm Ordmer 1087-1098 (c. 1175-1200) Bury, pp. 28, 32, 35, 42 (Rougham, Woolpit, Rushbrooke and Hopton (near Thetford) St), Ordmer de silua ibid., p. 31 (Hesset St); Galfridus filius Ordm(er)i de Westrudham c. 1145 CAcre, fo. 40a(iv), Willelmusfilius Ordmari de Westrudham ibid., fo. 40a(v) (West Rudham Nf); de terra que fuit Ordmeri se Cat 1163-1167 Urry, p. 237 (Canterbury K); Ordmarus filius Brunstan (witn.) c. 1170 CAcre, fo. 20a(iv) (Massingham Nf); Ordm(er) de Neut' (abl.) (witn.) c. 1170-1180 CAcre, fo. 28b(i) (Palgrave Nf), Ordmaro de Neut', abl. (witn.) ibid., fo. 95b(iv) (Dichewde Nf; Neut' is Newton-by-Castle Acre Nf); de Willelmo filio Ordmari et Waltero fratre ejus 1176--1177 P, p. 183 (C or Hu); Ordmarusfilius Toche de Parua Massingamia c. 1180-1190 CAcre, fo. 20a(v) (Little Massingham Nf); Ordmerus 1186-1188 (c. 1270) BuryS, p. 51, Ordmarus, Ordmer Ove,fen ibid., p. 52 (Hopton (near Thetford) St); Thomamfilium Ordmer 1202 CRR 28 (Pleas, ii, no. 1133) (Nf); Baldewinumfilium Ordmer' et Robertumfratrem suum de Claydun' 1209 FF (PRS NS, xxxii, no. 524) (Hemingstone Sf; Claydun' is Claydon St); Ordmer 1222 ElyA, fo. 205b (East Dereham Nf). For further Nf and Sf examples of OE Ordnuer from post-Conquest records see Selten PN, p. 125.

In the light of the forms retaining initial Ord- in the ME period, care is necessary in dealing with post-Conquest forms in Ormar(us), Ormer(us). Such forms can only be safely assigned to OE Ordm:Er when they occur side by side with forms in Ordmar(us), Ordmer(us) as the narne of a single person. Where this is not the case, the possibility that they represent the Scand Ormarr cannot be excluded. ON Ormarr etc. may appear as the first element of South Ormsby L (Ormesbi 1086, Ormesbi, Ormeresbi c. 1115), though the possibility of the more common ON Ormr etc. cannot be entirely excluded (DEPN, p. 351, s.n.; Fellows Jensen, NoB, lvii (1969), p. 63). SSNEM, p. 62, prefers derivation from ON Ormarr etc. to derivation from the more common Ormr, regarding early forms in initial Ormes- as merely reflecting confusion with Ormr.

ON Ormr, ODan, OSwed Orm a) (i) Ormisby c. 1020-1050 (1. 13th) ASWills, no. 25, Ormesby 1086 DB, fo. 115b, Ormesbei (bis) ibid., fo. 113b, Ormesbey ibid., fo. 273, Osmesbei ibid., fo. 197, Omesbei ibid., fo. 174b, Orbeslei (sic) ibid., fo. 114, Ormesbi (p), Ormesbi 1202 FF (PRS NS, xxvii, no. 405), Ormesbi, Ormesby 1209 AssRoll 558 (Pleas, iv, nos. 4137, 4178, 4194, 4457) (Ormesby, East Flegg H). - Ormiswell 1272 Terrier (f. n. in Rushford) (OKS). b) (i) PhilippofilioHorm', abl. (witn.) c. 1175-1190 CAcre, fo. 28b (the documentconcems properties in South Acre, Rackness, Palgrave and Burstall). - Willelmum Orm 1229 CRR 101 (Cur, xiii, p. 378) (Costessey).

ON Ormr etc. is an original byname belonging to the appellative ON ormr, ODan orm, OSwed orm(b)er m. 'snake, adder' (NK VII, p. 46). ON Ormr occurs as the name of several of the original settlers in lceland, 314

Ormr,Orm and is well attested in both Norway and Iceland throughout the Middle Ages (Lind, cols. 819-820). ODan Orm is also common in medieval Denmark, and forms the first element of several Danish place-names (DgP, cols. 1048-1050, 1689; Hald, Vore Stednavne, p. 114; DS Il, p. 45; DS IV, pp. 233,239,551; DS VI, p. 347; DS VII, p. 238; DS VIII, p. 30; DS IX, pp. 227-228; DS XI, p. 185; DS XIII, p. 258). In OSwed the name is known from runic inscriptions, and is found in medieval and 16th century records and as the first element of place-names (SRSö, no. 101; SRU, nos. 345, 628; SRVs, no. 19; Lundgren-Brate, p. 189; Fredriksson, pp. 169-182; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 32, 36, 39, 41, 136; Sahlgren, NoB, xv (1927), p. 73). In Norwegian and Swedish medieval records the name is often Latinized as Ormerus (Lind, col. 820; Modeer, SvP, pp. 59, 60; Fredriksson, pp. 178-179; for the use of Ormarus, Ormer for ODan Orm in 15th century Danish records, see above, s.n. Ormarr). In England ON Ormr etc. is well attested, as also emphasised by Sandred 1987:316, both in independent use and as the first element of place-names and field names: NPN, pp. 105-106; ZEN, p. 65; PNDB, p. 337; DBS, p. 259, s.n. Orme; SPNLY, pp. 204--206; Okasha, no. 64; DEPN, pp. 351, 488, s. nn. Ormerod, 19 Ormesby, North Ormsby, South Ormsby, 20 Ormskirk, Urmston; PN NbDu, p. 152; PN La, pp. 37, 84, 121, 153; PN NRY, pp. 157, 333; PN Nth, p. 291; PN Wa, pp. xxiii, 326; PN Cu, pp. xxxii and n. 4, 79, 306--307, 348, 435, 505; PN Db, p. 762; PN WRY, ii, pp. 50, 318; PN WRY, iii, p. 193; PN WRY, iv, pp. 12, 199; PN WRY, v, p. 56.

It is interesting to note that ON Ormr etc. is uncommon in Land Nf, but is extremely frequent in Y and fairly well attested elsewhere in northem England. Cf. the following La examples: Ormesdyke 13th (c. 1390---1400) Burscough, p. 40 (f. n. in Latham); Ormesdich', Ormesdiche m. 13th (c. 1390---1400) Burscough, p. 107 (f. n. in Wrightington); Orm filius Ketell (witn.) 1094 (15th) Lancaster, p. 10 (grant by Roger de Poitou to St Martin of Seez of the church of St Mary of Lancaster and other properties in his fee of Lancaster); Warinusfilius Orme, Warinusfilii Ormi c. 1150 (1412) Furness, ii, p. 169 (Stapleton Teme), Guarino filio Orm, abl. (witn.) 1160---1180 (17th) LaCh, p. 407 (Heaton in Lonsdale); Ormo filio Magni, Ormo filio Sueni, abl. (wits.) c. 1160 Hoghton (facsimile of MS. in frontispiece of De H oghton Deeds and Papers, ed. J.H. Lumby, LCRS, lxxxviii ( 1936)) (document concems properties at Elswick, Clayton-le-Woods, Whittlele-Woods, Wheelton, Hoghton and Roddlesworth); Rogero filio Orm, dat., 1160---1180 (17th) LaCh, p. 407 (Heaton in Lonsdale), Rogerus filius Ormi 1220---1225 (1412) Fur-

19 It is not clear whether Ormerod La contains ON Ormr etc. or ON Ormarr etc. as first element (DEPN, p. 351, s. n.; PN La, p. 84). 20 For South Ormsby, however, see above, s. n. Ormarr.

315

6spakr, ÖspakR, Öspak ness, ii, p. 173, quod Rogerus filius Ormi tenuit 1230-1250 (1412) Furness, ii, p. 175 (Skerton); Ormo clerico, abl. (witn.) 1180-1190 (17th) Cockersand, p. 327 (Carleton), 1184-1190 (17th) Cockersand, p. 324 (Staynall); Ormo de Hotun, abl. (witn.) 11841189 (from original, penes Count de Caseja, Scarisbrick Hall) Cockersand, p. 497 (La; Hotun is Hutton), seruicium Suani filii Orm 1200-1236 Cockersand, p. 434, predicto Swanofilio Ormme de Hoton', dat., c. 1229-1245 (orig.) Burscough, p. 253 (Hutton); Orm de Witington (abl.) (witn.) 1189-1200 (17th) Cockersand, p. 335 (case conceming tithes at Bilsborrow; Witington is Whittington); Ormus de Haidoc 1190-1199 Cockersand, p. 673, de dono Ormi de Haidoc 1242-1268 Cockersand, p. 643 (Ince; Haidoc is Haydock); 21 cum tofto in quo mansit Orm Dragun 1190-1213 Cockersand, p. 573 (Ainsdale); unam londam in australi parte de Aspincnol qme fuit Roberti filii Ormi 1190-1213 Cockersand, p. 584 (Ainsdale); Emmafilia Ormi de Asteleye 1190-1221 Cockersand, p. 712 (Astley); Ormf Dolfin 1194 P, p. 124, de Ormf Dolfin 1195 P, p. 149 (La); OrmMuschetl 195 P, p. 150, 1196 CR, p. 100, 1197 P, p. 193, 1198 P, p. 196 (La); toftum Ormi 1196--1206 Cockersand, p. 813 (Overton in Lancaster); Orm de Eston' 1200 P, p. 238, 1201 P, p. 271, 1202 P, p. 160 (La), Rogerus de Writhintonfilius Ormi de Aston 1201-1210 Cockersand, p. 504 (Wrightington; Eston', Aston is prob. Ashton on Ribble); in orientali parte teme quam Orm tenuit 1200-1230 Cockersand, p. 455 (Little Hoole); Ormfilius Rogeri 13th, before 1268, Cockersand, p. 244 (Mugehalc in Ashton on Ribble); Orm de Wolueton (abl.) (witn.) e. 13th (c. 1342) Whalley, p. 570 (Garston; Wolueton is Much or Little Woolton); de catallis Orm de Worleg' fugitivi, Orm de Warl' 1203 P, p. 233 (La; Worleg', Warl' is unidentified); Ormus et Thomas filius ejus c. 1219 Cockersand, p. 768 (Birstangrintel in Ellel); Henricusfilius Petrifilii Ormi de Parua Wordhull c. 1230-1250 (c. 1342) Whalley, p. 156 (Little Wardle and Haworth); Ormus de le ffalenge, Ormus c. 1230-1250 (c. 1342) Whalley, p. 155, Andreas filius Orm de ffalenges m.-1. 13th (c. 1342) Whalley, p. 770, de hornagio Andree filij Orm, homagium Andree filij Orm de le jfalenges 1277 (c. 1342) Whalley, p. 788 (jfalenge(s) in Spotland); Symonefilio Orm, abl. (witn.) c. 1275-1303 (c. 1390-1400) Burscough, p. 116 (Melling, West Derby H).

Note also the form Orman ho 1001 (contemp.) PN Wa, pp. 134, 328. I have elsewhere derived the first element of this place-name from an OE *Orma, which I interpreted as a 1. OE weak derivative of ON Ormr. 22 OE *Orma is, however, perhaps better interpreted as a hypocoristic form of Ordmier.

ON Ospakr, OSwed (run.) ÖspakR, OSwed Öspak a) (ii) Unspakeshill c. 1300 (?) AD (A 5482), Unspakehel c. 1300 (?) AD (A 5495) (f. n. Brancaster). - Unspakeshowe 14th Terrier (f. n. in Bast Rudham) (OKS). b) (i) Anschetelfilius Uspaci 1086 DB, fo. 279b (Bamingham Norwood).

21 Orm of Haydock was probably seneschal of Makerfield. His son was Alfred of Ince, who in 1212 held three teamlands in Ince and Haydock in thegnage, and owed the service of finding two doomsmen or judges for the court of Newton in Makerfield (Cockersand, p. 673, n. 2). 22 Insley, JEPN, x (1977-1978), p. 49.

316

Ospakr, ÖspakR, Öspak b) (ii) Vnspac TRE DB, fo. 113b (Field Dalling). - Unspati, gen., TRE DB, fo. 185 (one fiber homo ofthis Unspac held in Antingham). -Ailwardusfilius Vnspac (witn.) 11531168 Holme, no. 176 (North Walsham). - Vnspac Spracheling 1165-1166 P, p. 30, Vnspac Sprachelin 1166-1167 P, p. 30 (Nf or Sf) - Thomas filius Willelmi Unspac de Dilham m.-1. 13th Bromh, fo. 89a(ii) (Dilham).

ON Öspakr, OSwed (run.) ÖspakR, OSwed Öspak is an original byname belonging to the adjective ON 6spakr, uspakr, OSwed i5spaker 'unwise; unruly, quarrelsome' (NPN, p. 171; SRSö, p. 178; NK VII, p. 34; SPNLY, p. 329). ON Öspakr is found in lceland from the settlement period to the late 12th century, and is also recorded in the Orkneys (11th century) and as the name of a man from the Hebrides who died in 1230, but is only known in Norway from the 14th century onwards (Lind, cols. 822-823; LindS, col. 655). In OSwed the name occurs in runic inscriptions, and forms the first element of a place-name in -thorp (SRSmå, no. 113; SRU, no. 881; Lundgren-Brate, pp. 191,291, s. nn. Ospak, *Uspak). A weak variant, Öspaki, occurs in a runic inscription in Södermanland (SRSö, no. 202). ON Öspakr etc. is, thus, relatively rare in both OWScand and OSwed. Its apparent absence in Danish records is noteworthy. In England Unspac occurs as the name of a York moneyer of Edgar, and as the name of one of LEthelred Il's Shrewsbury moneyers (Coinage of Edgar, p. 206; Smart, Moneyers, pp. 222, 230). In DB the name occurs TRE in Nt, as well as in Nf (PNDB, p. 340). In the post-Conquest period it appears as the name of a Lincoln moneyer of William I, and is found in 12th century Y, where it appears as Unspac, Vnspac, Vnspak, Uspac, Vspac (von Feilitzen, SCBI, xi.i, p. 13; SPNLY, pp. 328-329). In Normandy the name is recorded in independent use in the 11th century as Ospac, Uspac, Ospacus and Ospachus, and forms the first element of the place-name Le Mesnil-Opac (Maisnillum Ospac 1180--1189 (1310)) (Adigard, pp. 128-129, 313-314, 410--411). Initial Un-, Vn- shows the substitution of the cognate OE prefix Un- for the Scand negative prefix 6-, (; < *Un- (SPNLY, p. 329).

317

R

ON Ragnhildr, ODan Rregnhild, OSwed Ragnhild fem. b) (i) Ragenild' Springsus 1198 AssRoll 559 (Pleas, ii, no. 51) (Freebridge H). -Ranilda 1222ElyA, fo. 211b (Thorp in Shipdham).

ON Ragnhildr is very common in Norway from the 9th century onwards, but is comparatively infrequent in medieval Iceland (Lind, cols. 842-844; LindS, cols. 668-677). There are also several examples of ON Ragnhildr found in the Orkneys and Shetlands in the post-Viking period, and the name was borne by a daughter of Olaf Bitling, king of Man and the Isles from 1103 to 1153 (LindS, col. 677). A variant Ragnhilda is well attested in Norwegian records of the 14th and 15th centuries (Lind, col. 842). In ODan there are the runic forms raknhiltr, nom. (DR, nos. 209, 230), and r:.ehnldu, dat. (DR, no. 241), and there area few examples of Regnild(a) etc. in medieval Danish sources (DgP, cols. 1145-1146). It should be noted that the name is recorded in the Danish royal family by the Norse Knyt/inga saga in the form Ragnhildr (Ragnhild, daughter of Sven Estrithson, king of Denmark from 1047 to 1074, and Ragnhild, daughter of Erik Ejegod, king of Denmark from 1095 to 1103, and mother of Erik Lam, king of Denmark from 1137 to 1146) (DgP, cols. 1145, 1691). It is not clear whether the first element of the runic raknhiltr should be read as Ragn- or as the i-mutated R:.egn- (Brpndum-Nielsen, § 80, Anm. 2; DR, cols. 696-697). DgP, col. 1146, remarks that forms in R:.egn-, Regn- can reflect a native Scand development, but regards it as probable that such forms reflect a loan from German (cf. OHG Reginhilt, MHG Reinhilt). Both examples of runic raknhiltr are from the Viking period (DR Per. 2.2; the inscription type is 'Fpr-Jelling'), and they relate to the same person, Ragnhild sister of Ulf, who was first the wife of a chieftain in south-west Sjrelland, Gunnulf son of Nrerfi (DR, no. 230), and then of Alli, goöi of the sanctuaries of northern Fyn, a devout pagan, whose inscription contains an invocation of the god Thor (DR, no. 209). 1 These inscriptions belong to a clearly pagan milieu, and are a considerable distance from the areas of southern Schleswig, where the Danes came in direct contact with the Carolingian and Ottonian empires. They

1

See L. Musset, Introduction a la runologie (Paris 1965), pp. 273, 424--426, 427.

318

Ragnhildr, Rregnhild, Ragnhild also belong to a period long before that in which German influence became widespread in Denmark. Therefore, in this case the possibility of German influence can be ruled out, and there is no reason to suppose that runic raknhiltr, even if its first element is to be read as REgn- and not as Ragn-, is anything but Scand in origin. OHG Reginhilt became Reinhilt in the course of the 10th and 11 th centuries (see Braune-Eggers, § 149, Anm. 6; for the development in OSax, which is similar, see Schlaug I, p. 50; Schlaug II, p. 145). Low German penetration of Denmark became important only in the later Middle Ages, 2 and by this time OHG Regin-, MHG Rein- etc. would have been rendered by Rein-, Reyn-, and not by Regn-, in MDan, though the German form might possibly have had the effect of reinforcing a native ODan REgn-, i-mutated form of Ragn. It is significant that forms in Regn- occur in ODan already in the 12th century, that is, before large scale German penetration ofDenmark had begun, and that the unmutated Ragn- occurs as late as the 14th and 15th centuries (RagnildE 1392, Ragnyl 1475) (DgP, cols. 1145-1146). In OSwed Ragnhild(r) is only sparsely represented in the runic material, but is frequent in medieval records (SRVg, no. 142; SRU, nos. 112, 309, 310; Lundgren-Brate, p. 198; Modeer, Personnamn i Ktb, pp. 29, 121; Thors, p. 75). In the Viking period the use of ON Ragnhildr etc. seems to have been confmed to royal and aristocratic circles (cf. SRU, ii, p. 427), and to have only later spread to other social groups. In England the name is moderately well attested in Land Y, but is infrequent elsewhere (NPN, p. 111; ZEN, p. 68; PNDB, p. 347 and n. 1; ELPN, p. 80; SPNLY, pp. 213-215; Fellows Jensen, Lincolnshire Tenants, p. 87; DEPN, p. 382, s.n. Raventhorpe; PN Bk, p. 261; PN ERY, pp. 191-192; PN WRY, iv, p. 203). The forms Reinild, Reynild (L VD, King Horn), which are included by Björkman, NPN, p. 111, sub Rag(e)nhild, do not belong here, but reflect a reflex of OHG Reginhilt, MHG Reinhilt etc. It is theoretically possible that the above Nf form Ranilda 1222 ElyA could reflect ON Hrafnhildr fem., but the rarity of this name, which in Scandinavia itself is confined to Iceland (Lind, cols. 565-566; LindS, col. 449), renders this unlikely.

2 For the Low German penetration of medieval Denmark see Skautrup, Det Danske Sprogs Historie, i, pp. 187-188, 304, ii, pp. 2-5, 102-111, 113. For German personal names in Denmark see Hald, Personnavne i Danmark: Il. Middelalderen, pp. 21-32.

319

Ragn(v)aldr, Rc;_>gnvaldr, Rregnwald, Ragnvald

ON Ragn(v)aldr, Rc;_>gnvaldr, ODan Rregnwald, OSwed Ragnvald b) (i) toftam Ragenaldi c. 1230 Der, fo. 98b(i) (Upwell).

ON RQgnvaldr, Ragn(v)aldr is frequent from the earliest times in Norway, but is first recorded in Iceland at the end of the 12th century, appearing fairly sporadically from this time onwards in Icelandic records (Lind, cols. 861-864; LindS, cols. 688-693). The name was in use in the 11th and 12th centuries in the comital dynasty of Orkney, and is found in the 12th and 13th centuries in the Hebrides (LindS, cols. 691-692). In medieval Danish records the name occurs as Regnwaldus, Regna/dus, but it is fairly infrequent (DgP, cols. 1146--1147). Regna/dus is also recorded by Saxo Grammaticus as the name of a son of the famous 9th century Viking Ragnarr Loöbr6k (DgP, col. 1146). The medieval Danish forms Reynoldus, Reinoldus, Reynolt, Reinolt etc. do not belong here, but reflect a reflex of the cognate MHG Reinolt etc. (see DgP, cols. 1147-1148). In OSwed Ragnvaldr occurs in three runic inscriptions from Uppland, and the name is frequent in medieval Swedish records as Ragnald, Ragnwald, Rangwaldus, Raghwaldus, Rawald etc. (SRU, nos. 112, 309, 310; LundgrenBrate, pp. 196, 199-200; Modeer, Personnamn i Ktb, pp. 19, 97, 103, 108, 127-128; Thors, pp. 75-76). In the Viking period the use of ON RQgnvaldr etc. seems to have been confmed to royal and aristocratic families in Scandinavia, and only later does it seem to have passed into general use (see SRU, i, pp. 161-162). In English records the earliest bearers of the name were Rregnald I (ob. 921), a Viking king from Ireland who established himself as king at York in 919, and Rregnald II Guthfrithson, king of York 943-944. In DB ON RQgnvaldr etc. is recorded in Ch, L, Nt and Y, and it is also found in other English sources, though fairly infrequently (Björkman, Festermen, pp. 5, 15; Lindkvist, Anglia Beiblatt, xxxiii (1922), p. 142; PNDB, pp. 346--347; SPNT, p. 144; SPNLY, p. 213; von Feilitzen, SCBI, xi.i, p. 12; PN Db, p. 762). SPNLY, p. 213, groups the lost WRY place-name Renathorpe Hall (Raynaldesthorp, Raynaltorp 13th) sub Ragnaldr, but the early forms of this place-name show that it contains OFr Rainald, a name which corresponds to MHG Reinolt, and which is cognate with ON RQgnvaldr etc. (cf. PN WRY, i, p. 212; PN WRY, vii, p. 296).

ODan Ringulf, OSwed (p. n.) Ringolf a) (i) Ringoluesdole Edwl Rental (f. n. in Wiggenhall St Mary Magdalene) (OKS) (Sandred, OUÅ, 1972, p 45). -Ryngolsty 1487 Terrier (f. n. in Gaywood) (OKS).

320

Ringulf, Ringolf b) (i) Rincvlfv 1017-1023 SCBI, xiv, no. 3217,Rincvlfl029-1035 SCBI, xviii, no. 690, 1044-1046 SCBI, xviii, no. 1140, Rin(g)vlf 1040-1042 SCBI, ii, no. 978, Ricnvlf 10231029 SCBI, ii, no. 920, SCBI, xiv, no. 3216, Rinvlf 1035 SCBI, xiv, nos. 3218, 3219 (moneyer, Norwich). - Ringul. fiber homo TRE DB, fo. 267 (Rock:land All Saints). Ringulfus. I. fiber homo TRE DB, fo. 174b (Oby),Ringulfi, gen., ibid. (one liberafemina ofRingulfheld atBastwick), terram et aulam Ringolfz 1101-1107 (14th) Holme, Appendix A (Nf), Ringolf (noster homo) ibid. (Tibenham). 3 - terram Ringulfi c. 1090-1100 CAcre, fo. la(ii), terram que fuitRingulfz 1146--1147 CAcre, fo. 3b(i) (Castle Acre). Godricus filius Ringolfl. 1 lth- e. 12th (13th) Rams, iii, p. 299 (Godric son of Ringolf is recorded in this document, a survey of 1201, as having granted half of Curemer in Outwell to Ramsey Abbey), Folcardo filio Godrici filii Ringulfz, dat., 1114--1130 (m. 14th) Rams, i, p. 144, no. 70 (Outwell). -unum uillanumRingulfumnomine 1138-1147 CAcre, fo. 45a(iii), I. uillano Ringulfo nomine, abl. ibid., fo. 49b(iv), Ringulfo, abl., Ringulfum c. 1155 CAcre, fo. 50a(i), Ringulf (acc.) ibid., fo. 50a(ii), terram Ringolf ibid., fo. 50a(iii), (Creake). -Rigolf (witn.) 1146--48 HarlCh 47 H 45 (printed, Cheney, English Bishops' Chanceries, p. 154, no. 4 (Rudham, Marham, the mill of Torp). Ringolfus c. 1150 (1302) NorwCPCh, no. 259 (2), Ryngolphus c. 1150 (e. 14th) NorwCP, p. 64 (Norwich). - terram Ricardi filii Ringolfz 1164--c. 1185 CAcre, fo. 6b(ii) (Marham). - de Ringulf de Redehal' 1165-1166 P, p. 34 (Redenhall). - de catallis Ringulfz 1176--1177 P, p. 137 (Nf or Sf, prob. Nf). - de hornagio Godefridi et Walteri filiorum Ringolfz de Tilneia c. 1200 Der, fo. 18la(ii), homagia Godefridi et Walteri filiorum Ringolfz de Tiln' ibid., fo. 181b (Tilney). - domum Ringolfz c. 1200-1211 (13th) BuryS, no. 137 (soke ofCaister StEdmunds). -Cartalohannis Ringolf,Iohannes filius Ringolf e. 13th Der, fo. 195a(ii) (Wiggenhall St Mary Magdalene). - terram Ringulphi bercarii 1214--1229 Holme, fo. 73b(vi) (Westcroft, South Walsham), toftum Ringulfz bercarii in villa de Walsam c. 1230 Hickling, fo. 8b(ii) (South Walsham). Willelmifilii Ringolf, gen., de Willelmo Ringolf 1219 FF, no. 20 (South Walsham). Ricardus filius Ringolf c. 1225 Co:iford Il, fo. 30b(ii), Ric(ardus) Ringolf ibid., fo. 30b(ii, iii (bis)), terram Ricardi Ringolf ibid., fo. 3 la(v ), Radulfz Ringolf, gen., Radulfus filius Ricardi Ringolf c. 1230-1240 Co:iford Il, fo. 3 la(iv), de Radulfo Ringolf ibid., fo. 34b(vi), terram Galfridi Ringolf c. 1230 Co:iford Il, fo. 36b, Nich(olaus) Ringolf c. 1230-1240 Co:iford Il, fo. 30b(iv), Nicholaus Ringolf ibid., fos. 30b(iv), 3 la(i), Nicholaus Ringolf et Ailfled vxor mea ibid., fo. 3 la(ii), terram Nicholai filii Ringolf (bis) ibid., fo. 34b(iv), de Nicholao Ringolfibid., fo. 34b(v), Ailfleda quondam vxor Nicholai Ringolf c. 1245 Co:iford Il, fo. 3 la(iii) (West Rudham). - Talliam Willelmi Ringolffilii Radulfz c. 1230 JMN, p. 233 (Nf). -RyngolfGumbaud 1258 Ass (Selten PN, p. 99) (Nf). In Denmark Ringvlf, Ringuf appears as the name of a moneyer of the period 1080-1086 (DgP, col. 1170). In view of the strong AScand element among 11th century Danish moneyers' names, 4 it is uncertain, however, whether this isa genuine ODan name or whether it is an import from the Danelaw. In Sweden Ringoljforrns the first element of the place-name Ringolfsmala 1399 in Småland (Ödeen, p. 201). Stenton, EHR, xxxvii (1922), p. 229, took the Nf Ringolf, -ulfto be a Scand

3 Ringolf of Oby fled to Denrnark with Abbot JElfwald of St Benet of Holme (1064-1089) soon after the Norman Conquest (see Stenton, EHR, xxxvii (1922), pp. 227,233). 4 See Hald, Om Personnavnene.

321

Ringulf, Ringolf name without any further discussion. It is noteworthy, however, that the name is so rare in Scandinavia itself but is relatively frequent in Nf. The regional distribution of Ringolf, -ulf in England is also interesting. In DB the name is confined to Nf and Sf, and in the ME period it is rare outside Nf, only a few examples being known from L, Sf and Y (PNDB, p. 293, s.n. Hringwulf, SPNLY, p. 219, s.n. Ringulfr; Selten PN, pp. 99-100, s.n. Hringwul/). Von Feilitzen, PNDB, p. 293, took DB Ringulf(us) to reflect OE Hringwulf, and Selten PN, pp. 99-100, takes this OE name to be probably the main source of the examples of Ringolf, -ulf etc. found in East Anglia in the ME period, though he does note the existence of the Scand Ringulf, -alf and the Frankish (H)ringulfus, Ringolf. There are good reasons for preferring the OE Hringwulf to the OEScand Ringolf, -ulf for the etymon of the English Ringolf, -ulf etc. The use of OEScand Ring- as a personal name element is extremely infrequent. The dubious Ringulf is the sole example in ODan, while in OSwed Ringolf is only supplemented by Ringsten, Ringvast and Ringvidh (NK VII, p. 198; Modeer, SvP, pp. 22, 30). In England, however, the corresponding OE Hring- seems to have been more productive-as late as the early ME period the masc. names *Hringhere, *Hringstän and *Hringweard, and the fem. names Hringwaru and *Hringwynn are on record (von Feilitzen, NoB, xxxiii (1945), pp. 83-84; Reaney, Studier i modern språkvetenskap, xviii (1953), p. 95; Selten PN, p. 99).

322

s ON, ODan Saxi, OSwed Saxe a) (i) Saxthorp 1086 DB, fo. 132b, Sastorp ibid., fos. 146b, 148, Saxiorp ibid., fo. 146 (Saxthorpe, South Erpingham H). - Saxedele e. 13th Der, fo. 122b(ii) (f. n. in Gayton Thorpe). b) (i) Saxy 11 th (contained in a list of benefactors to the Abbey of St Benet of Holme drawn up in the period 1186--1210) Holme, no. 62 (Swanton Abbot). - totum tenementum quod fuit Saxi de Morleia I. 12th Wymondham, fo. 59b(i) (Morley). - Carta Saxi filii Thurstan de Walt', Saxi filius Thurstani de Waltone c. 1190 Der, fo. 119b(ii), 1 l 9b(iii), Carta Saxi filii Thurstan de Waltona, Saxi filius Turstani de Walton' ibid., fo. 119b(iv), terram Willelmi filii Saxi e. 13th Der, fo. 119a(i) (East Walton). - Gilebertus f Saxe 1198 P, p. 91, 1199 P, p. 272 (Nf or Sf, prob. Nf).

A characteristically OEScand name, Saxi is mast likely an original byname belonging either to ON sax n. 'a short single-edged sword, a scramasax' or to the etymologically identical tribal name ON Saxar m. pl. 'the Saxons', though the possibility that it might also be sometimes a short form of ON SQxulfr, ODan Saxulf should also be considered (Brjljndum-Nielsen, § 431; DgP, col. 1203; NK VII, pp. 199, 205; Modeer, Personnamn i Ktb, p. 25; Otterbjörk, Svenskaförnamn, p. 123; SPNLY, p. 228; Hald, Personnavne i Danmark: I. Oldtiden, p. 61). Kristian Hald, KLNM, xiii, col. 223, took Saxi to belong to the tribal name 'Saxon', and to have been originally used in Denmark to denote a Saxon immigrant. A further possibility is that the name was borrowed at an early Si-in ODan seems to appear first in the 12th century. Cf. the examples given by Brs,;,ndum-Nielsen, §§ 263, 392.1, and note that initial Sig(h)- is general in Runic Danish, for which see DR, cols. 710--711. The spellings -uerd' and -uert in the second element are normal AN spellings for OE -ferö < -friö and AScand -ferö < ON -jrf')ör etc. (PNDB, §§ 14, 107; SPNLY, §§ 21, 43, 116.iv, 117.iii; SPNLY, § 43, believes that forms in -uerd should, when they represent the Scand element, be almost certainly derived from -frför and not from -jrf')ör). For further discussion and possible examples of ON Sigfrf')ör etc. in English sources see the following: NPN, pp. 117-118; PNDB, p. 360; SPNLY, p. 231. Björkman, NPN, p. 117, believed the Scand name to be at least probable in the case of Siferö, Siefereö, the name of a moneyer of Eadred and Edgar, but Coinage of Edgar, p. 204, takes Siferö here to represent OE Sigefriö.

ON Sighvatr, OSwed Sighvat, AScand *Si(g)wate a) (ii) Siwatestofot 1209 FF (PRS NS, xxxii, no. 148) (f. n. in Palling). - Sywatesholm 14thE/y (f. n. in Walpole). b) (i) Carta Willelmi filii Siquati Druth, Willelmus filius Siquati Druth de Tiln' c. 1200 Der, fo. 157a(i), Carta Willelmi filii Siquati Druth de Tiln', Willelmus filius Siquati Druth de Tiln' ibid., fos. 172a(iii), l 72b(i), Carta Willelmi fil ii Siquati Druth de Tilneia ibid., fo. 172b(ii), Willelmus filius Siq(ua)ti Druth de Tiln' ibid., fos. l 72b(ii) - 173a, de hornagio Willelmifilii Siquati, homagium Willelmi Sigwat de Tiln' c. 1235 Der, fo. 145b(i) (Tilney). -Robertus Siwat pater ipsius Agnetis 1220 CRR 77 (Cur, ix, p. 254) (Gateley). - de Sygwat Katerbode, de Sygwat Radbode 1275 RH, i, p. 451 (Taverham H).

(ii) Siwate Hod 1100--1102/1107-l 112 (13th) Rams, iii, p. 289 (Walsoken). 8 - Siwate 1150 St William, p. 181 (Norwich). -Siwate 1158-1180 SCBI, xvi, nos.404, 407 (moneyer, Thetford). -Sywate Mungiw (abl.) (witn.) c. 1200 BodlNJCh, no. 357 (Little Plumstead). - quod Siquate Snowd tenuit c. 1220 Der, fos. 139b(i), 144b, quod Siquate Snouth tenuit c. 1235 Der, fo. 145b(ii) (Tilney).

ON Sighvatr is frequent throughout the Middle Ages in both Norway and Iceland from the Viking period onwards (Lind, cols. 879-881; LindS, cols. 698-

8 This document, a marginal note of c. 1174, appears in a hand which is later than the 13th century hand of the surveys printed by Rams, iii, pp. 218-328.

328

Sigmundr, Sighmund 703). A famous Icelandic bearer of the name, Sighvatr 1>6röarson (c. 9951045), was a skald of St Olaf of Norway. 9 In Sweden runic Sighvatr is frequent and as Sighuat, Sigwater etc. the name occurs in 14th and 15th century Swedish records (SRÖl, no. 41; SRSö, nos. 26,317; SRVg, no. 94; SRU, nos. 67, 128, 144, 152,180,237,245,372, 586,885, 1080; Lundgren-Brate, pp. 218-219; Thors, p. 83). The name does not appear to have been in use in Denmark. 10 In England ON Sighvatr, AScand *Si(g)wate etc. is well attested in L from the time of Edward the Confessor to the early 13th century, but it is only found sporadically elsewhere (NPN, p. 121; PNDB, p. 363; SPNT, p. 144; ELPN, p. 80; DBS, p. 338, s.n. Suett; SPNLY, pp. 232-233). In Sf the name is represented by the AScand forms Siwate 1186-1188 (c. 1270) BuryS, p. 27, Siuuate ibid., p. 32 (Westley St). A weak OScand variant *Sighvati is not on record, and, therefore, the English forms Siquate, Siwate, Siuuate and Sywate are to be regarded as representing an AScand variant which arose in the Danelaw. The loss of -h- in the element -hvatr, *-hvati in England may reflect an OE or an ODan or an AN development (see above, s.n. *Gunnhvati, and the references cited there). The forms Siquati (gen.) and Siquate (nom.) have AN q for /kl, which is here the result of the AN interchange of g and k (PNDB, §§ 118, 119, 128; SPNLY, §§ 128.iv, 135). For the development of OScand Sig(h)- > Si- in English sources see above, s.n. Sig(f)r(Jjor). For the use of -w-, -uu-, and -u- for OScand -v- (= /w/ in English sources) in English reflexes of OScand -hvatr, *-hvati see SPNLY, § 67.ii, and above, s.n. *Gunnhvati. The Nf RH form Sigwat Radbode (abl.) also appears as Syward Radbode (abl.) (below, s.n. Sigward, Sighwarth (ii)). Here Sigwat and the more familiar Siward, Syward < ODan Sighwarth, OSwed Sighvardh or OE Sigeweard have been confused.

ON Sigmundr, ODan, OSwed Sighmund b) (i) Symundo de Lundham, abl. (witn.) 1127-1134 HarlCh 44 E 19 (printed, Holme, Appendix E) (Heigham next Norwich), Simundus (witn.) 1134-1140 Holme, no. 132 (Grengeuile), 1153-1168 Holme, no. 176 (North Walsham), Symundo de Ludham, abl. (witn.) 1150---1160 Holme, no. 65 (tithes in Scottow), Simundo de Ludhamfilio Stigandi, dat., Simundus (bis) 1153-1166 (l. 13th) Holme, no. 128 (Burwood and Lud-

9

For Sighvatr l>6röarson see KLNM, xv, cols. 231-238, and the works cited there. An ODan *Sighwat does not appear in DgP.

10

329

Sigriör, Sigriöa, Sighrith, Sighridh ham), Simundus de Ludham (witn.) 1153-1168 Holme, no. 167 (Burwood), 1153-1168 Holme, no. 179 ((H)abelund), Thomasfilius Simund (witn.) 1163-1166 BodlNfCh, no. 605 (printed, Holme, no. 164) (Hasardes holm near Potter Heigham), Thome de Waltonafilio Symundi de Ludham, dat., 1175-1186 Holme, no. 209 (confirmation to Thomas de Waltona of land which bis father Simund has held in the soke of Ludham; Waltona is unidentified). -Quieta clamantia filii Simundi le Nobele c. 1210--1220 Der, fo. 238b(ii), Stephanus filius Simundi le Noble de Wigenhale ibid., fo. 239a (the document concems the homage of Paganus son of Goche of Lynn; Wigenhale is Wiggenhall).

ON Sigmundr is well attested from the Viking period onwards in both Norway and Iceland (Lind, cols. 881-883; LindS, cols. 703-708). ODan Sighmund is rare but is found in two runic inscriptions, and is also known as the first element of several Danish place-names, including two in -stath (DR, nos. 325, 398; DgP, cols. 1224-1225, 1693; Hald, Vore Stednavne, pp. 92, 145; Kousgård Sprensen, Bebygg. på -sted, pp. 110, 245; DS IX, p. 107; DS XIII, p. 221). In Swedish there are a few runic examples of the name and it is well represented in medieval records (SRÖg, no. 31; SRSö, no. 162; SRSmå, nos. 23, 59; SRU, no. 777; SRVs, no. 24; SRNä, no. 31; Lundgren-Brate, pp. 219220). For the development of OScand Sig(h)- > Si- in English records see above, s.n. Sig(j)rr/}Ör. It is, however, not clear whether the above Nf forms represent the Scand Sigmundr, since ME Simund(us) can equally represent OE Sigemund or an OFr reflex of Frankish Sigimundus or an AN form of the biblical name Simon (NPN, p. 120; PNDB, p. 363; SPNT, p. 146; DBS, p. 321, s.n. Simmonds; SPNLY, p. 233). Bast Anglian examples of ME Simond, Simund etc. are also examined by Selten PN, p. 145. Selten (loc. cit.) also points out that the above mentioned Nf Simund of Ludham also appears as Simon, Symonus de Ludham owing to confusion with the biblical name, but considers that his name may be Scand like that of his father Stigand.

ON SigriOr, SigriOa, ODan Sighrith, OSwed Sighridh fem. b) (i) pro salute anime Willelmi Gerold et Sigeride uxoris eius I. 12th--e. 13th Wals, fo. 83a(i) (prob. Gunthorpe).

ON Sigr{ör etc. is a compound of Sig(h)- and jr{ör (NK VII, pp. 103, 195; Brpndum-Nielsen, § 263; Modeer, SvP, pp. 26, 54). ON Sigr{ör is common throughout the Middle Ages in both Norway and Iceland from the Viking period onwards (Lind, cols. 885-887; LindS, cols. 709-722). There are a few examples of a weak form Sigr{öa in Norwegian records from the end of the 13th century onwards (Lind, col. 885). 330

Sigriör, Sigriöa, Sighrith, Sighridh

In Denmark the name is relatively infrequent, but is found throughout the Middle Ages from the earliest times (DgP, cols. 1226-1227). In Swedish there are several examples of runic Sigrlör, and the name is common in medieval and 16th century records (SRSö, nos. 73, 101, 106, 198; SRU,nos. 78,198,440,608,687,695,819,851,884,973, 1035, 1042;Lundgren-Brate, pp. 220---221; Modeer, Personnamn i Ktb, p. 30; Modeer, Fornsvenska personnamn i en regestsamling, p. 14; Thors, pp. 80---81; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 165, 169). ON Sigr{ör etc. is well attested in England, especially in Y (NPN, p. 118; ZEN, p. 73; PNDB, p. 364; SPNT, p. 144; DBS, p. 322, s.n. Sired; SPNLY, pp. 234-235; PN ERY, p. 328; PN WRY, iv, p. 140). Note also that the name is frequent in medieval La. Cf. the following forms: Sigerithwatholm 1190-1215 Cockersand, p. 826, 1221-1246 Cockersand, p. 828, Sigerith Watholm 1210-1241 Cockersand, pp. 850, 853, Sigeritwatholm (bis) 12201250 Cockersand, p. 860 (f. n. in Caton); Johannes filius Sigerith de Sutton, Sigerith mater prredicti Johannis 1190-1220 Cockersand, p. 596 (Sutton, south east of and partly in St Helens); terram Willelmi filii Sygrit 1200-1225 Cockersand, p. 129, domum Willelmifilii Sygrit 1240-1268 Cockersand, p. 126 (Staynall); Rogerusfilius Aswardi et Sigherid sponsa mea e. 13th Cockersand, p. 197 (Warton); Simonfilius Sigrith de Bothelton 1220-1250 Cockersand, p. 914 (Bolton-le-Sands); Syrith filia Osberti de Aynoluesdale 11 c. 1240 (c. 1342) Whalley, p. 573 (Altune on the Mersey in Garston; Aynoluesdale is Ainsdale); Willelmus de Tonnewright et Syerith uxor mea, Willelmus et Syerith, dicte Syerith (dat.) m. 13th (c. 1342) Whalley, p. 763 (Spotland); Benedictus filius Sigerith de Far/ton m. 13th Cockersand, p. 1058 (Farlton), Sigerithfiliam Walteri ibid., p. 1059 (Lonsdale or Kendale), Sygerit filiam Willelmi filii Eylsi ibid., p 1061

(prob. Thomton near Blackpool). Interconsonantal -e- in the above Nf form Sigeride (gen.) is a characteristically AN svarabhakti vowel (PNDB, § 52.3; SPNLY, § 63.iii; note, however, that such inorganic vowels do also occur in OE, for which see Campbell, § 367). The replacement of OScand -ö- in the Nf Sigeride by -d-, which is here an inverted spelling for [t], is an AN development (PNDB, § 107; SPNLY, § 116; above, s. nn. *Asjr(f)ör, Auöhildr). Final -e- in the Nf Sigeride represents the gen. sing. ending of the Latin first declension (PNDB, § 157; SPNLY, § 152; von Feilitzen, PNWinton, p. 227).

11 The form is incorrectly rendered by the editor of Whalley as Syrith filius Osberti de Aynolnesdale.

331

Sigward, Sighwarth, Sighvardh

ODan Sigward, Sighwarth, OSwed Sighvardh b) (i) Carta Iohannis Sigward, Iohannesfilius Sigwardi de Gunethorp I. 12th-e. 13th Wals, fo. 82b(vi) (Gunthorpe). (ii) Seiardus bar TRE DB, fo. 128 (East Beckham), fo. 223b (Sheringham), Seiar bar ibid., fo. 223b (Salthouse), terram Seiardi. bar ibid., fo. 128 (North Erpingham H). 12 Siuuardus 1086 DB, fo. 136 (Hunstanton), fo. 137 (Bittering). -Radulfusfilius Siwardi (witn.) c. 1170 CAcre, fo. 20a(iv), Radulfo filio Siward', abl. (witn.) c. 1185 CAcre, fo. 16a(v) (Massingham), Radulfusf Siwardi 1191 P, p. 44, Radulfusf Siward' 1192 P, p. 189, Radulfus f Siward 1194 P, p. 55 (Nf). - Siwardus de Len 1176---1177 P, p. 136, 1177-1178 P, p. 25, 1178-1179 P, p. 6, 13 1179-1180 P, p. 18 (Lynn). - Hodiernam filiam Siword' 1198 FF (PRS NS, xxvii, no. 55) (Syderstone ). - H enricus f Siward 1199 P, p. 275, Henricus f Siward' 1200 P, p. 139, Henricus f Siwardi 1201 P, p. 134, de Henrico f Siwardi 1202 P, p. 108 (Nf or Sf, prob. Nf). -Alanus f Siward' 1199 P, p. 286, 1200 P, p. 144 (Nf or Sf, prob. Nf). -Humfridumfilium Siward' 1200 FF (PRS NS, xxvii, no. 252) (Syderstone). - per Siward' de Meresie 1208 CRR 50 (Pleas, iv, no. 3212) (Nf; Meresie is probably Mersea Ess). - Johannes Siward 1209 AssRoll 558 (Pleas, iv, no. 4117) (West Winch). -portam 14 Seward' 15 de Lade 1209 FF (PRS NS, xxxii, no. 195) (Bottisham C; the fine is grouped to the Nf fines). - per Siward' de Ormesbi 1211 P, p. 5 (Ormesby). -Siwardum le Barker', Siwardus 1224 CRR 86 (Cur, xi, p. 309), Siwardum le Barker 1224 CRR 85 (Cur, xi, p. 413), Siwardum le Berker 1225 CRR 90 (Cur, xii, p. 125), Siwardum le Berker' et Ricardumfilium ejus, Siwardus 1225 CRR 88 (Pleas, xii, p. 271) (Brisley). -de hornagio Willelmifilii Syward, homagium Willelmi filii Siward Diruth c. 1230 Der, fo. 137b(i), homagium Willelmi filii Siward Druth ibid., fo. 138a(i) (Tilney). - Ricardus Syward, Willelmus f Syward 1250 Ass (Nf) (Selten PN, p. 147), Syward de Wrokkesham 1257 Ass (Nf; Wrokkesham is Wroxham) (Selten PN, p. 147). - de Syward Stert de Lincroft 1275 RH, i, p. 480 (Nf; Lincroft is unidentified). - de Syward Radbode 1275 RH, i, p. 528 (Taverham H). 16

This name reflects a PrScand *Si3i-warÖuR (NK VII, pp. 89-90, 113-114; Kousgård Sprensen, Bebygg. på -sted, p. 248). In OWScand the name occurs with a weakly stressed second element in the form Sigurör, which is extremely frequent in Norway and Iceland from the Viking period onwards (Lind, cols. 889-899). Examples of the strong Sig-

12 These TRE forms relate to Siward Barn, who was one of the leading landowners of the East Midlands in the time of the Confessor and who was probably of Scand descent, his narne appearing in the Anglo-Saxon Chronicle as Siward Barn 1071 (1121) ASC (E), Sigwaro Barn 1072 (I. 11th) ASC (D) (PNDB, p. 361 and n. 7; cf. also the comments of C. Fell, Anglo-Saxon England, iii (1974), pp. 184--186). From the form appearing in the D text of the Anglo-Saxon Chronicle, it would appear that all the forms representing the narne of Siward Barn have ODan Sighwarth, OSwed Sighvardh as their ultimate base. 13 1178-1179 P, p. 6, notes that Siward of Lynn had departed to Norway ("Sed abiit in Norweiarn"). He must have returned for he is recorded by 1179-1180 P, p. 18, as having settled his account with the exchequer. 14 The reading is uncertain and the word may be parcam. 15 This form shows confusion between Siward and ME Seward < OE Sreweard. 16 This person appears elsewhere in the RH as Sigwat Radbode (RH, i, p. 451; see above, s. n. Sighvatr).

332

Sigward, Sighwarth, Sighvardh varör, Sighwardh are also known from the OWScand area, but they first appear in the later Middle Ages, and the replacement of the usual OWScand -urör in these forms by -varör, -wardh etc. is the result of Danish or German influence (Lind, loc. cit.). ODan Sighwarth is common in medieval Danish records, where it usually appears as Siwardus, Sywardus etc. (DgP, cols. 1229-1236, 1693). In OSwed forms in both -urör (OSwed -urdh) and -varör (OSwed -vardh) occur, but the former is more frequent, and the name itself is rare in the Viking period, only becoming well attested later (SRSmå, no. 75; SRU, nos. 568, 854; Lundgren-Brate, pp. 222-223, 224, 225, 227-228, s. nn. Sighurdh, Sighvardh, Sighvordh, Siurdh; Modeer, Personnamn i Ktb, pp. 19, 64; Thors, pp. 81-82; Modeer, SvP, pp. 22, 31; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 140, s. nn. Sigurd, Si(g)vard). Ofthe above Nf forms only Sigward, Sigwardi (gen.) can be taken to clearly represent ODan Sighwarth etc. Forms in Siward, Siwardus, Siuuardus etc. are ambiguous, however, since OE Sigeweard also gives a ME Siward. Siward can, it is true, be shown to represent the Scand name on occasions, cf. Earl Siward of Northumbria (ob. 1055), who was a Dane, and whose name is rendered in the Anglo-Saxon Chronicle as follows: Sigwarö eorl 1043 (m. 11th) ASC (D), Siward eorl 1048 (1121) ASC (E), refter Siwarde eorle, on Siwardes eorldome 1052 (m. -1. 11 th) ASC (D), Siward eorl 1054, 1055 (m. 11th) ASC (C), 1054 (m. -1. 11th) ASC (D), 1055 (1121) ASC (E) (bis), Syhward eorl 1055 (m. - 1. 11 th) ASC (D). 17 Note also the forms taken by the name of the eastem English magnate Siward Barn (see above, sub b(ii) and n. 12, and, for a full conspectus of DB forms, PNDB, pp. 361-363). 18 In the Danelaw there was probably a certain convergence between OE Sigeweard and ODan Sighwarth, OSwed Sighvardh which resulted in ME Siward, though it is, of course, theoretically possible that ME Siward can reflect either of these names. Only in a few cases is a definite decision possible. East Anglian examples of ME Siward are collected and examined by Selten PN, p. 147, s.n. Sigeweard. Selten (loc. cit.) takes ME Siward to have three sources, viz. OE Sigeweard, ODan Sighwarth, and 'ContGerm' Sigiward, but considers that the frequency of the name in L and Y in SPNLY suggests that the main contribution may have been Scand. Normally the Continental option need not be considered, though Selten does provide an example from a 13th century Lynn Bede Roll, Siward de Lubek. This man is clearly by origin a native of Liibeck,

17 For the career of Earl Siward of Northumbria see ASWrits, p. 572. Earl Siward had the byname Digera < OWScand Digri < digr 'big' (Tengvik, p. 310). 18 Seiardus, Seiar, which is used in DB to render the Nf forms relating to Siward Barn, shows the replacement of ME Si- (here from ODan Sigh-) by OFr Sei- as a result of Norman influence (cf. PNDB, § 133).

333

Skalpr, *Skalpi

and as such is only of marginal consequence for the discussion of ME Siward. Selten considers Scand origin likely in the case of the Sf form Saxe f Siward 1228 Ass, since the name of the son is Scand, but this is a non sequitur. In cases where ME Siward represents the Scand name, -ward in the second element reflects the replacement of OScand -varör by the corresponding OE -weard (see PNDB, p. 100, n. 2). For further discussion and examples of ME Sigward, Siward etc. see the following: NPN, pp. 118-119; ZEN, p. 73; PNDB, pp. 361-363, 364; Tengvik, p. 198; SPNT, pp. 144, 146; ELPN, p. 62; Newark, pp. xiii, 2, 4, 13; DBS, p. 314, s.n. Seward; SPNLY, pp. 236-239, 355; Fellows Jensen, Lincolnshire Tenants, pp. 87, 94; von Feilitzen, PNWinton, p. 172; DEPN, p. 413, s. nn. Sewardstone, Sewerby; PN Bk, p. 260; PN BdHu, pp. 226,290; PN NRY, pp. 231-232; PN D, pp. 196,686; PN Nth, p. 291; PN Sr, p. 401; PN Ess, pp. 30, 540; PN ERY, p. 104; PN Cu, pp. 69-70, 505; PN 0, pp. 130, 480; PN Db, p. 762; PN WRY, v, p. 120; PN WRY, vii, p. 299; SSNY, pp. 37, 68; Insley, JEPN, x (1977-1978), pp. 61, 62). Cf. also the following Sf forms: Sywardescroft c. 1200--1211 (e. 14th) BuryS, no. 125 (f. n. in Fenstead); Siuuard 1087-1098 (c. 11751200) Bury, p. 30 (Hessett), ibid., p. 34 (Whelnetharn), Siuuardus, Siuuardus Mocesun ibid., p. 42 (Layham); Siward 1186-1188 (c. 1270) BuryS, p. 19 (Rougharn); Terra Siward 1186-1188 (c. 1270) BuryS, p. 32, Eufemme filie Syuardi de Westle, 19 dat., c. 1200--1211 (e. 14th) BuryS, no. 66 (Westley); Siward Palmere 1186-1188 (c. 1270) BuryS, p. 37, Siuardus Palmere ibid., p. 39 (Stanton), Siwardibid., p. 41 (West Stowe); terra Galfrid' f Siward, Ricardus f Siwardi 1228 FF, Saxe f Siward 1228 Ass (Sf) (Selten PN, p. 147).

For Nf and Sf examples from the ME period additional to those in the present work see Selten PN, p. 147, s.n. Sigeweard.

ON Skalpr, *Skal pi (byname) a) (ii) Skalpistille 1351 OKS (f. n. in Wymondharn). b) (i) Calpus. liber homo TRE DB, fo. 240 (Postwick), Calpus ibid., fo. 262 (Tunstall).

This byname belongs to ON skalpr m. 'the sheath of a sword' (LindBiN, col. 315). There area few examples of the strong variant Skalpr in independent use as a byname in both Norway and Iceland, but the weak form *Skalpi is only known from a Norwegian place-name (LindBiN, cols. 314-315). The name does not appear to have been in use in OEScand. In England the weak variant *Skalpi occurs as a personal name in DB TRE in Ess and Sf (PNDB, p. 365). The above Nf field name Skalpistille also belongs to the weak form. Von Feilitzen, PNDB, p. 365, grouped the Nfforms in

19

Var: Eufemmefilie Sywardi (e. 14th), Eufemiefil' Siwardi (13th).

334

Skeggi, Skregge Calpus under the strong Skalpr, but admitted the possibility that they might alternatively be derived from the weak *Skalpi. The loss of initial S- in the Nf DB forms in Calpus is the result of AN influence (ANI, pp. 67---68, 72-73; PNDB, § 112).

ON, ODan Skeggi, OSwed Skregge a) (i) Scengutuna 1086 DB, fo. 229b, Scedgetuna ibid., fo. 248, Scegeton' 1191 P, 1236 Fees (OKS) (Skeyton, South Erpingham H).

This is an original byname belonging to ON skegg, ODan skeg n. 'beard' (Lind, col. 911, s.n. Skeggi; DgP II, col. 971, s.n. Skegg). ON Skeggi is known throughout the Middle Ages as a personal name, though it should be noted that it is more frequent in Iceland than in Norway (Lind, cols. 910-911; LindS, cols. 733-737). There are also a few examples of a strong variant Skeggr in Norwegian records from the end of the 13th century onwards (Lind, cols. 911-912). A byname Skegg is on record in Norway from c. 950 onwards, and examples of Skeggi as a byname occur in both Norway and Iceland (LindBiN, cols. 319, 320, s. nn.). In Denmark Skeggi occurs in a 15th century record as a personal name, and both Skegg and Skeggi occur as bynames in medieval Danish sources (DgP, col. 1254; DgP II, cols. 970-971). Skregge is on record in medieval Sweden as a personal name and as a byname, but it is not frequent (Lundgren-Brate, p. 234; XenLid, pp. 101, 105; Modeer, SvP, p. 96). In England ON, ODan Skeggi, OSwed Skregge appears as the first element of the two Nt Skegbys and of Skegness L (DEPN, p. 424, s. nn. Skegby, Skegness; PN Nt, pp. 133, 190; SPNLY, p. 247). It has recently been suggested that the two Nt Skegbys and Skegness L may contain the appellative skegg in the sense of a projecting piece of land rather than the personal name Skeggi (SSNEM, pp. 68---69, 172). The persistent -e- between the first and second elements in the early forms of the Nt Skegbys, the L Skegness and the Nf Skeyton speaks against this, since it is too frequent to be merely a svarabhakti vowel and is best interpreted as going back ultimately to the gen. -a of the personal name Skeggi.

ME Sket < ON skiotr b) (i) Sketh 11 th (contained in a list of benefactors to the abbey of St Benet of Holme drawn up in the period 1186--1210) Holme, no. 62 (North Walsham). -ScheitTRE DB, fo. 234b (Sloley). - Schett. fiber homo TRE DB, fo. 245b (Dersingham). - Scet liber

335

Skrauti homo TRE DB, fo. 257b (Edgefield), Esket ibid. (Binham). - Eschet TRE DB, fo. 268 (Overstrand). -Scet (abl.) (witn.) c. 1135-1140 CAcre, fo. 16b, (Massingham), Schetus 1138-1147 CAcre, fo. 4a(i) (Nt), De terra Schet c. 1150---1170 CAcre, fo. 16a(i), unam acram que fuit Eudonis filii Scet c. 1175 CAcre, fo. 16a(iii) (Massingham). - Ricardus filius Schet 1165-1166 P, p. 29, Ricardus filius Schiet 1166-1167 P, p. 30 (Nfor Sf, prob. Nt). 20 - domum Galfrid' Schet c. 1180 Wymondham, fo. 33a(i) (Nt).

This name is an AScand formation belonging to the Scand loan word 1. OE seeot, ME sket < ON ski6tr 'fleet, swift', and, as was noted by Björkman, NPN, p. 123, it appears to have been mainly a Nf name (NPN, p. 123; ZEN, p. 76; PNDB, p. 366; DBS, p. 322, s.n. Skeat). The use of -eh- for /kl in (E)sehet(t) and Seheit is in accordance with early AN usage (PNDB, § 113; SPNLY, § 126; von Feilitzen, PNWinton, p. 225). The use of -e- for /kl in Seet may reflect survival of a traditional OE orthographic usage, or it may simply reflect scribal error (PNDB, § 115). SPNLY, § 127.iv, suggests that spellings in Se-, -se- for /sk/ before front vowels might possibly reflect Anglicization to /J/. This can be regarded as implausible in view of the relative frequency of such forms, and their appearance in the early forms of place-names which have retained the value /sk/ (cf. the early forms of Skeyton Nf given above, s.n. Skeggi). It should be remarked that although in AN orthography -eh- is usual for /kl before front vowels and -e- before back vowels this was by no means a hard and fast rule and exceptions do occur (cf. PNDB, § 115; SPNLY, §§ 126.ii, 127.iv). The use of -k- for /kl in Sketh and Esket is most plausibly interpreted as reflecting normal ME orthographic usage, though it should be noted that /kl is also very occasionally represented byk in OE and OFr sources (Jordan, § 178; PNDB, § 118; SPNLY, § 125). Initial E- in the forms Esket, Esehet is an AN prosthetic vowel (PNDB, § 51; SPNLY, § 65). The use ofprosthetic e before S + consonant occurs several times in DB (PNDB, loc. cit.). The DB form Seheit seems to have AN ei for ME ? < OE eo, though it might alternatively be interpreted as belonging to AScand Seegö < 1. OE seegö < ON skeiö f. 'a warship, a swift vessel' (PNDB, p. 366). This last name is represented by the form Seegö 1002-1004 (1. 11 th) ASWills, no. 17 (Palterton Db).

ON Skrauti (byname) a) (i) Scrouteby c. 1020---1050 (l. 13th) ASWills, no. 25, Scroutebei 1086 DB, fo. 197, Scrovtebey ibid., fo. 273, Scroteby ibid., fo. 220b (Scratby, East Flegg H; for further early forms see Lindkvist, p. 164).

20

DBS, p. 322, s. n. Skeat, assigns this Richard son of Schet to Nf.

336

Skuli, Skiili, Skiile

The ON byname Skrauti belongs to the same base as the substantive ON skraut n. 'ornament, show, display, splendour', and denotes a showy, magnificent person, a person fond of display (Lindkvist, p. 164). This byname is rare, but it is on record in Iceland in the settlement period and the period immediately following (LindBiN, col. 332). Cf. Sandred 1987:317. In England ON Skrauti is attested in independent use as a personal name in Wa in DB TRE (ZEN, pp. 76-77; PNDB, p. 366).

ON Skuli, ODan Skiili (p. ns.), OSwed Skiile a) (i) Scufatorpa 1086 DB, fo. 144b, Scufetorpa ibid., fo. 168 (Sculthorpe, Gallow H). - Scufetuna l 086 DB, fos. 164, 266b, 268 (Scoulton, Wayland H). - Scufedafe, Scufedaf 1201 (13th) Rams, iii, p. 295 (f. n. in Walsoken). -Scufeshawe 1252 Wafs, fo. 66b(ii), 1259 Binham, fo. 60b(i) (f. n. in Barney). -Scufesflet c. 1260-1280 Wafs, fo. 51b(iii), Scufeshirne (bis), c. 1290 Wafs, fo. 51b(ii) (f. n. in Holkham). b) (i) Scufe Leojwofdes sune (witn.) 1043-1044 (1. 13th) ASChR, no. 97 (Swanton and Hindolveston). -Scufa fiber homo TRE DB, fo. 221 b (lslington). - Sclufa TRE DB, fo. 232b (Pensthorpe). -Escufe homo Herofdi TRE DB, fo. 239b (Tunstall), Escufe. l. fiber homo ibid., fo. 240 (Postwick), Esculce, gen., ibid., fo. 279b (held the commendation of two fiberi homines with 60 acres at Postwick). - Scufa. fiber homo TRE DB, fo. 245b (Massingham), Scufa ibid. (Babingley). - Scufa TRE DB, fo. 280 (Fomcett St Peter). Scufa c. 1090-1100 CAcre, fo. 1b, decimam Scufe c. 1100-1120 CAcre, fo. 2a (Nf). Willefmo filio Scufe, abl. (witn.) c. 1135-1140 CAcre, fo. 16a(vi)-16b (Massingham), Willefmus ftfius Scufe 1138-1147 CAcre, fo. 4a(i) (Nf), De Willefmo ftfio Scufe, Willefmus ftfius Scufe (witn.) c. 1150-1170 CAcre, fo. 16a(i), in toftes que fuerunt Willelmi filii Scufe c. 1175 CAcre, fo. 16a(iii) (Massingham). - Scul' et Willefmus frater eius (witn.) 1146-1148 HarlCh 47 H 45 (printed, Cheney, Engfish Bishops' Chanceries, p. 153, no. 4) (Rudham, Marham, the mill of Torp). - De Radulfo filio Scufe, Radulfus filius Scufe c. 1160 CAcre, fo. 62b(i) (Little Ryburgh). - Schufe (abl.) (witn.) c. 1165 CAcre, fo. 57b(ii) (the document concems the mill of Holt and dues from Brichefeswrde and Cley next the Sea). -Rogerusfifius Scufe (witn.) c. 1170 CAcre, fo. 20a(iv),Rogero ftfio Scufe, abl. (witn.) c. 1200 CAcre, fo. 19b(i) (Massingham). - Escufo (reeve of Philipp son of Emisius), dat., 1170-1180 CAcre, fo. 16a(iv) (Nf). - de Godwin' Scufe 1174-1175 P, p. 123 (Nfor St). - Willefmusfifius Scufe 1179-1180 P, p. 21, de Willefmo ftfio Scufe 1180-1181 P, p. 86 (Nf). -Iordanusfifius Scufe c. 1190-1200 Binham, fo. 6a (printed, SSMEA, Appendix Il, no. 3, p. 270)21 (Binham). - ad capud crofte Roberti filii Scufe 1196 FF (PRS, xvii, no. 195) (Creake). - Willefmumfilium Scufe 1196 FF (PRS, xvii, no. 207), 1198 FF (PRS NS, xxvii, no. 33) (Witchingham), de Willefmo f Scufe 1199 P, p. 273 (Nf or Sf, prob. Nf). - crofta Scufe 1200 FF (PRS NS, xxvii, no. 250) (Southmere). -terram Scufe c. 1200 CAcre, fo. 31b, terras Scufe et Rogeri Peko c. 1215 CAcre, fo. 32a(i) (Rougham), terramRicardififii Scufe 1202 FF (PRS NS, xxvii, no. 440) (Nf), terram que fuit Ricardi fifii Scufe c. 1210 CAcre, fo. 30b(i) (Rougham), 21

Here the Scule of the MS. is incorrectly rendered as Seule.

337

Skuli, Skiili, Skiile

terram Thomefilii Scul' 1209 FF (PRS NS, xxxii, no. 197) (the land ofThomas son of Scul' was adjacent to Galterhill' in Rougham). - de dono Sculi de Sparham 1206 FF (PRS NS, xxxii, no. 111) (Sparham). ON Skuli etc. is an original byname which is to be connected etymologically with the verb ON skyla 'to conceal, to screen, to protect' (NK VII, pp. 53,245; SPNLY, p. 254). ON Skuli is well attested in Iceland from the 10th century onwards, but is later and less frequent in Norway, where it does, however, form the first element of several place-names (Lind, cols. 925-927; LindS, cols. 738-740). LindBiN, col. 335, has a single example of Skuli used as a byname from a Norwegian record of 1294. The name is not certainly attested in ODan, but it probably forms the first element of two Danish place-names in -thorp (DgP, col. 1261; DS XVI, p. 140). In OSwed runic Skuli is attested twice, and a few other examples of the name have been noted in other Swedish records (SRVg, no. 217; SRU, no. 614; Lundgren-Brate, p. 233; Fredriksson, Västgötska dopnamn under 500 år, p. 183). In England ON Skuli etc. was Anglicized as Scula, Scule, and several bearers of the name are recorded already in the OE period. The earliest example in English history was a Viking who, according to the Historia de Sancto Cuthberto, received from Rregnald, the later Norse king of York, the territories between Castle Eden and Billingham-on-Tees Du some time between 913 and 915. 22 A second bearer ofthe name is the comes Scule, who is recorded by the Liber Eliensis as having held properties at Brandon, Livermere and Southbourne Sf, and who is identified by Dorothy Whitelock with the dux Scule who attested several charters between 931 and 935 and between c. 946 and 949. 23 Cf. also Scvla 1023-1029 SCBI, xiii, no. 477, 1029-1035 SCBI, xiii, no. 478 (moneyer, Exeter). Scula also appears as the name of a York moneyer who was active in the reigns of Harald Harefoot, Harthacnut, and Edward the Confessor (SCBI, xxi, pp. xlvii, xlviii). In DB the name is recorded TRE in L, Nt, Sf and Y, as well as in Nf (PNDB, p. 366). SPNLY has a single postConquest example of the name in independent use from a L document of c. 1150-1155, and the name has also been noted in place-names in Y and in field names in L (SPNLY, p. 254; DEPN, p. 409, s.n. Sculcoates; PN ERY, p. 214; PNWRY, i, p. 26; PNWRY, iii, p. 31; PNWRY, vi, p. 96). ON Skuli etc. also forms the first element of the lost R place-name Scvletorp 1086 DB, and is contained in the first component of the Du place-name School Aycliffe

22 23

SD, i, p. 209. See also Stenton, The Danes in England, p. 137, n. 1. D. Whitelock, Foreword to LibEI, p. xiv.

338

Spakr,Spak (SSNEM, pp. 116-117; DEPN, p. 20, s.n. Ayclijfe; PN NbDu, pp. 8-9). 24 A post-Conquest Sf example of the name is contained in the form Le/stan Scule filius 1087-1098 (c. 1175-1200) Bury, p. 26 (Great Barton St). In Normandy ON Skuli etc. forms the first element of two place-names (Adigard, pp. 135,414). The use of -c- for /kl in the English forms Scula and Scule reflects normal English orthographic practice. In OE c is normally used to represent /kl in all positions, 25 and in the post-Conquest period it is the normal early AN and ME spelling for /kl before velar vowels (Brunner, AeGramm, § 207; Campbell, § 50.4; Jordan, § 178.1; Brunner, AbriB, §§ 5B, 38.1; PNDB, § 114; SPNLY, § 127). Final -a, -e in the English forms Scula and Scule is the result of Anglicization (PNDB, §§ 45, 150; SPNLY, § 147). Initial E- in the DB forms Escule (nom.) and Esculc:e (gen.) is an AN prosthetic vowel (Lindkvist, p. 173, n. 2; PNDB, § 51; SPNLY, § 65). The use of prosthetic e before S + consonant occurs several times in DB (PNDB, loc. cit.). In the DB form Sclula the first l is due to anticipation (PNDB, § 66).

ON Spakr, ODan, OSwed Spak (byname) b) (i) Petrus f. Aldith. Spac frater eius 1203 AssRoll 817 (NthRS, v, no. 802) (the document deals with property in Martham). -portam Walteri Spac 1209 FF (PRS NS, xxxii, no. 217) (Scratby ).

This byname belongs to the adjective ON spakr 'wise; peaceful', ODan spagh 'gentle, peaceful' etc. It is not common, though it has a fairly wide distribution, examples being known from Iceland, Norway, Denmark and Sweden (LindBiN, col. 350; DgP Il, cols. 1037-1038; XenLid, p. 103; Modeer, SvP, p. 104). A weak variant, ON Spaki, is on record in Norway and Iceland in the Viking period (LindBiN, cols. 349-350). A weak form, Spake, is also found in OSwed (XenLid, p. 103). In addition to the above Nf examples, ON Spakr occurs in England as the first element of the So place-name Spaxton (DEPN, p. 433, s.n.).

24 Both Ekwall (DEPN, loc. cit.) and Mawer (PN NbDu, loc. cit.) identified the Skuli whose narne forms the first component of School Aycliffe Du with the Scula mentioned above, who received the area between Castle Eden and Billingharn-on-Tees from Rregnald some time between 913 and 915, but there is no evidence to confirm this identification. 25 For the occasional use of k for /kl in OE manuscripts see Brunner, AeGrarnm, § 206.7d, and Carnpbell, p. 173, n. 1.

339

*Sprot(t)-Ulfr

AScand *Sprot(t)-Ulfr b) (i) Sprotulfus c. 1155-1158 Holme, no. 243, Cirographum Wlfrici filii Sprotulphi, Wlfrico filio Sprotulfi, dat., predictus Sprotulphus pater eius 1168-1175 Holme, no. 311, Wlfricusfilius Sprottolf (witn.) 1168-1175 BodlNfCh, no. 602 (printed, Holme, no. 312) (Yarmouth).

A personal name Sprat is recorded TRE and TRW in DB, and is recorded elsewhere in the post-Conquest period in England (PNDB, pp. 370--371; ELPN, p. 65; DBS, pp. 329-330, s.n. Sprat(t); SPNLY, p. 261). Von Feilitzen, PNDB, p. 370, took the distribution of the DB TRE forms (Db, fäs, Sf, Y) to suggest Scand origin, adding that Björkman (Minneskrift utg. af Fil. Samf. i Göteborg 1910, p. 78 f.) had compared this name with Norw dialect sprat(t) 'small pole, angling rod'. Von Feilitzen also included the possibility, however, that the name may, in some cases, belong to OE sprat n. 'sprout, shoot, twig, small branch' or OE spratt m. 'sprat', and he regarded the ME sumame Sprat, which occurs throughout England, to be doubtless largely native in origin (ibid., pp. 370--371 and p. 371, n. 1). Ekwall took the personal name and sumame Sprat(t) to be more probably of English rather than of Scand origin, and he linked it to OE sprat(a) 'shoot, twig' (ELPN, pp. 64, 65). Ekwall's interpretation is followed by DBS, pp. 329-330, s.n. Sprat(t). SPNLY, p. 261, follows von Feilitzen' s interpretation, but also poses the question as to whether the name could be an AScand formation. Selten, p. 76 and n. 95, takes Sprat to belong to OE sprat n. 'sprout, shoot'. In general, the evidence speaks in favour of a native OE origin for the personal name Sprat(t). OE sprot n. 'sprout, shoot, twig, small branch' and OE spratt m. 'sprat' could both have been used without difficulty to form an original byname, and the relative frequency of the sumame Sprat outside the Danelaw in the MF period rather speaks against a Scand origin. Von Feilitzen, PNDB, p. 371, took Sprat(t)ulf etc. to be a secondary compound of Sprat with OScand -ulfr or OE -wulf. Selten, pp. 23, 76, also takes Sprat(t)ulf to be a secondary formation and considers the base to be an OE *Spratwulf < Sprat + -wulf. Altematively, we may be concemed with an AScand formation in which OE sprat or OE spratt with the function of a byname is prefixed to the common personal name ON Ulfr, ODan, OSwed Ulf. 26 In addition to the above Nf forms, Sprat(t)ulfhas also been noted as a personal name in Sf and it also occurs in Nf and Sf in the ME period as a byname (PNDB, p. 371; Selten PN, pp. 150--151).

26

For formations of this type see above, s. n. *Fang-Ulfr.

340

Steingrimr

ON Steingrimr a) (i) Stangrimeshoec (bis) c. 1200-1210 Der, fo. 244a (f. n. in Lynn). b) (i) Stangrin presbitero de Inguluesthorp, abl. (witn.) 1101-1107 Binham, fo. 2a (grant by Peter de Valoignes to Binham Priory; lnguluesthorp is Ingoldisthorpe). Guillelmo filio Estangrin', dat., suus pater Estangrin' c. 1115 (1395-1396) SSMEA, Appendix I, no. 57, Willelmo filio Estangrin', dat., Estangrin' pater suus, idem Estangrin' c. 1125 (1395-1396) SSMEA, Appendix I, no. 58, Guillelmofilio Estangrin, dat., ibid., no. 59 (the soke of Necton). - magister Steingrimus (witn.) 1146----c. 1170 (15th) NorwCPCh, no. 130 (the document cancerns the homage of Richard son of Odo of South Elmham Sf and land which belonged to Osbern of Hindringham Nf), c. 11501159 (I. 13th) NorwCPCh, no. 134 (the document cancerns the mill at Pedeham which belonged to the manor of Langley), Magistro Stangr(im)o, abl. (witn.) c. 1150 CAcre, fo. 47a (the document cancerns the church of South Creake), Magistro Stangrin', abl. (witn.) ibid., fo. 47a(i) (Creake), (magistro) Stangrimo, abl. (witn.) c. 1160-1170 NorwCPCh, no. 131 (the document cancerns the church of Scratby), magistro Stangrimo, abl. (witn.) c. 1165 (15th) NorwCPCh, no. 124 (Lynn and Mintlyn), 27 Stangrimo (archidiacono), abl. (witn.) 1169-1174 (15th) NorwCPCh, no. 128 (Hoxne Sf), c. 1170-1174 (15th) NorwCPCh, no. 125 (the documentconcerns the church of Cressingham), Johannes de marisco filius Steingrim quondam archidiaconi c. 1200 CAcre, fo. 65a(iii) (Norwich). 28 - v. sol' in Rochelund de tenura Stangrim c. 1155 CAcre, fo. 18a(i), v. sol' in Rochelund de tenura Willelmi filii Steingrim c. 1175 CAcre, fo. 42a(i) (Rockland All Saints). -Steingrim l 165-1166 P, p. 25 Staingrim 1166--1167 P, 29 p. 25 (Lynn). - Stangrim nouus 1165-1166 P, p. 27, Stangrim Nouus 1166--1167 P, 30 p. 27 (Lynn). - Staingrimfilius Game! 1165-1166 P, p. 27, 1166--1167 P, 31 p. 27 (Lynn). Staingrim bonpain 1165-1166 P, p. 29, Staingrim Bonpain 1166--1167 P, p. 29, 11671168 P, p. 19 (Nfor Sf). -Alanusfilius Staingrim 1165-1166 P, p. 28, 1166--1167 P, 32 p. 29 (Lynn), Hugonefilio Stangrim. Alano fratre suo, abl. (witn.) c. 1190 CAcre, fo. 84a (Wiggenhall), mariscum que fuitAlanifilii Stangrim c. 1215 Der, fo. 240b(i), mariscum que fuit Alani filii Stangrimi de Northlen ibid., fo. 242a(ii) (North Lynn). - Stangrimo presbitero de Linna, abl. (witn.) c. 1165 (15th) NorwCPCh, no. 124 (Lynn and Mintlyn), 33 Stangrim presbitero, abl. (witn.) c. 1170 Lewes, fo. 285b(ii) (NfMarshland), Stangrimo clerico, abl. (witn.) c. 1180 CAcre, fo. 89b(iii) (Clenchwarton), Stangrino clerico de Linna, abl. (witn.) 1182 CAcre, fo. 8 la(ii) (Nf), Willelmus filius Stangrimi c. 1175-1180 Lewes, fo. 283b(ii) (Walpole), Willelmo filio Stangrimi sacerdotis de Lenna, abl. (witn.) c. 1190-1200 CAcre, fo. 82b(i), Willelmo filio Steingrin de Linna, dat. (witn.) ibid., fo. 83a(ii) (Wiggenhall), Willelmus filius Stangrim 1198 FF (PRS NS, xxvii, no. 12l)(Tilney), Willelmusfilius Stangrim' 1199 CRR 17, 18 34 (CurR, ii, p. 19) (Lynn), Carta Willelmi filii Stangrim, Willelmus filius Stangrim de Len c. 1200 Der, fo.

For the dating of this document see above, s.n. Arnkell, n. 34. For Steingrim, archdeacon of Norwich, see L. Landon, 'The Early Archdeacons of Norwich Diocese', Proceedings of the Suffolk Institute ofArchaeology, xx, Part 1 (1930), p. 24. 29 Steingrim CR. 30 Stangrim Noueman CR. 31 Steingrim clericusfilius Game/ CR. 32 Alanus filius Steingrim CR. 33 For the dating of this document see above, s.n. Arnkell, n. 34. 34 These rolls are duplicates. 27 28

341

Steingrimr 176b(iii) (draua ofTilney-draua of Terrington), Willelmumfilium Steingrim' 1202 FF (PRS NS, xxvii, no. 459) (Lynn and West Winch), Willelmum filium Stang' 1207 Lewes, fo. 284a(ii) (printed, SSMEA, Appendix I, no. 63) (Tilney and Islington; Linna, Lenna, Len is Lynn). - Stangrimus et Edricus frater eius c. 1175-1180 Lewes, fo. 283b(ii) (Walpole). - Petrusfilius Stangrini de Len' c. 1190 Lewes, fo. 289b (Saddlebow; Len' is Lynn). - excepta Wimercha uxore Steingrini de Wigehale c. 1190-1200 CAcre, fo. 82a(vi) (Wiggenhall). -Thurstanusfilius Stengrim de Dersingham c. 11901220 Co:iford Il, fo. 68b(iii) (Dersingham). - Iohanne filio Stangrim, abl. (witn.) c. 1190-1200 Lewes, fo. 269a(ii) (lslington and Spelhohe), terram que fuit Johannes (sic) filii Stangrim c. 1205 Der, fo. 134a(ii), terram que fuit lohannis filii Stangrim de Tiln' ibid., fo. 134b(i), c. 1220 Der, fo. 174a(i), terram que fuitlohannis filii Stangrim c. 1220 Der 163a(i), c. 1230 Der, fo. 147a(i), terram que fuit lohannis filius (sic) Stangrim c. 1220 Der, fo. 163a(i) (Tilney). -Steingrim leues 1197 P, p. 240 (Nf). -Hawisamfiliam Stangrim 1198 FF (PRS NS, xxvii, no. 32) (Watton). - Stangrim (gen.) (deceased husband of Emma of Appleton) 1198 AssRoll 559 (Pleas, ii, no. 33) (Freebridge H). Steingrim 1198 P, p. 87, (Steingrim) 1199 P, (p. 270),35 (Stein)grim 36 1200 P, p. 136, de Steingrim 1201 P, p. 133 (Nf or Sf, prob. Nf). -Hamonemfilium Steingrim 1199 CRR 11 (CurR, i, p. 346) (Guist). - Robertus filius Staingrim 1199 CRR 17 (Pleas, i, no. 2556) (Nt), Roberto filio Stangrim, dat., 1199 CRR 17, 18 37 (CurR, ii, p. 19) (Lynn). Stangrimpater Engelie 1200 CRR 16 (Cur, i, p. 177) (laHoge). -Ricardo filio Stangrimi, abl. (witn.) c. 1200 CAcre, fo. 80b(iii), 80b(iv), homagiumRicardifilii Stangrim de Len c. 1215 Der, fo. 241a (Lynn). - homagium Semanfilii Stangrim de Sadelbowe c. 1200 Der, fo. 21 la(i) (Saddlebow), Simonemfilium Stengrim 1206 CRR 43A (Pleas, iii, no. 2439) (Nf), Carta Willelmi filii Simonis Stangrim de Sadelbowe, Willelmus filius Symanis Stangrim de Sadelbowe c. 1220 Der, fo. 217b(i), Carta Willelmi filii Simonis Stangrim de Sadelbowe, Willelmusfilius Symonisfilii Stangrim de Sadelbowe ibid., fo. 218a(i) (the documents here concem the homages of various men of Wiggenhall and also deal with property in Saddlebow). -terram que fuit Stangrimi potarii, de tota terra de Pegesneweland' que est de feodo ipsius Godefridi. 38 que fuit Stangrimifilii Ede, terram que fuit Stangrimi filii Ede 1201 FF (PRS NS, xxxii, no. 1) (Tilney). - Stangrim (dat.) 1202 CRR 28 (Pleas, ii, no. 765) (Nf), Steingrim (bis) 1203 CRR 26 (Cur, ii, p. 199), Stangrimus, Stengrimum 1203 CRR 32 (Cur, iii, p. 58) (Corpusty). -Staingrimf Alberti 1203 P, p. 247 (Nf or Sf, prob. Nf). -Steingrimf Burewen 1203 P, p. 247, 1204 P, p. 240 (Nf). - Robertus filius Stangrim' 1205 CRR 38 (Cur, iii, p. 282) (Nf; the document concems Gooderstone). - Ranulfus clericus filius Stangrimi diaconi de Snetesham c. 1205 Wymondham, fo 45b(i) (Snettisham). - Carta Ricardi filii Stangrim de Tiln' de hornagio Petri filii Alani de Clenchewardtun, Ricardus filius Stangrim de Tiln' c. 1205 Der, fo. 233a(ii) (the document concems the homage of Peter son of Alan of Clenchwarton and the land which he held at Choisecroft in South Clenchwarton from Richard son of Stangrim of Tilney; Tiln' is Tilney), Ricardi (sic) filius Ricardi Stangrim de parochia Sancti Petri de Lenn' c. 1240 Der, fo. 233a(iii) (the document concems the homage of Willelmus Aleman de Lenn' and the land which he held from Richard son of Richard Stangrim of the parish of St Peter of Lynn in South Clenchwarton; Lenn' is Lynn). -Ailwardusf Staingrim 1208 P, p. 14 (Nf or Sf, prob. Nf). -Thomefilio Stan-

35

Supplied from 1198 P, p. 86.

36

(Stein) is supplied from 1201 P.

37

These rolls are duplicates. Godfrey de Lisewis.

38

342

Steingrimr grim de Tilneie, dat., c. 1215 Der, fo. 127a (Tilney). - in dal' ueteris Stangrim c. 1215 Der, fo. 175a(i), super dalam ueteris Stangrim ibid., fo. 241b, in prima dala heredum ueteris Stangrim, in secunda dala heredum ueteris Stangrim c. 1220 Der, fo. 128b, secundam magnam dalam heredum ueteris Stangrim ibid., fo. 174a(i), in secunda magna dala heredum ueteris Stangrim ibid., fo. 174b (Tilney). -terram que fuit Stangrimifilii Seueri c. 1220 Der, fo. 202b(i), terram que fuit Stangrimi filii Seuere ibid., fo. 203b (Wiggenhall St Mary Magdalene). - terram Rogeri Stangrim m. 13th Der, fo. 118b(i), Carta Thome Stangrim de Estwalt', Thomas Stangrim de Eastwaltune m. -1. 13th Der, 117a(ii), Cart' Thom' Stangrim, Thomas Stangym de Estwaltun' ibid., fo. 1 l 7a(iii) (East Walton). 39

ON Steingrfmr is frequent throughout the Middle Ages in Iceland from the time of the settlement onwards, and is also quite well attested in medieval Norway, though it seems to have been less popular there than in Iceland (Lind, cols. 952-953; LindS, cols. 754-755). ON Steingrfmr does not seem to have had an OEScand counterpart.40 In England the name is extremely uncommon outside Nf. In L ON Steingrfmr occurs TRE in DB and has also been noted in a record of 1231 (PNDB, p. 373; SPNLY, p. 264). Cf. also the following: Willelmum filium Steingrim Ricl CRR 12 (Cur, i, p. 1) (Gislingham St); Steingrim de Hunt' 1201 CRR 25 (Pleas, iii, no. 235) (Huntingdon). Forms in Staingrim, Steingrim contain the normal ME spellings for OScand cei, ei, though the co-existence of spellings in -ai- and -ei- might, to some extent, also reflect the interchange between ai and ei in AN orthography (see Jordan, § 130; PNDB, § 38; SPNLY, §§ 48, 49). Forms in Stan- are Anglicized, ON Stein- having been replaced by the corresponding OE Stän- (see PNDB, §§ 41, 150; SPNLY, § 52). Such Anglicized forms are doubtless the result of analogy with the 1. OE and e. ME group of personal names with OE Stän- as first element. These are well attested in East Anglia. For examples from Nf and Sf see PNDB, pp. 371-372, s. nn. *Stänflc:ed, Stänheard, Stänmc:er, *Stänwine, and Selten PN, pp. 151-153, s. nn. *Stänflc:ed, *Stängi(e)fu, -geofu, Stänheard, *Stänhere, *Stänhild, Stänmc:er, *Stänwine. Note also the following additional examples: OE Stänheard: Stannard et JElric frater eius 1087-1098 (c. 1175-1200) Bury, p. 25 (Great Barton St), Stanard (bis), Stanard Lefstani filius, Stanard Euengiue filius, Filii Stanardi ibid., p. 29 (Rougham St), Stanard cum filio ibid., p. 30 (Hessett St), Aldred Stanardifilius ibid., p. 34 (Whelnetham St), Goduine Stanardifilius ibid., p. 35 (Rushbrooke St), Stanard ibid., p. 36 (Timworth St), Stanard pelliciarius ibid., p. 36 (Fomham St Genevieve St), Stanard (bis) ibid., p. 41 (lxworth Thorpe, Coney Weston St);

Thomas Stangrim of East Walton was the son of Roger and Basilia. Lundgren-Brate, p. 243, deduced an OSwed *Stengr7m from the Bohuslän place-name Stengrimseröd. This evidence is, however, unacceptable since Bohuslän belongs to the OWScand area. 39

40

343

Steingrimr Aluuricumfilium Stanardifullonis c. 1170 (13th) Bury, no. 203 (St); Stannard 1186-1188 (c. 1270) BuryS, p. 9 (Pakenham St), terra Stanardi ibid., p. 17 (Bradfield St George St), Aluricus de Westmere, Stannardfrater suus ibid., p. 18, Stannardus, Stannard ibid., p. 19, Stannard Fulwar(e) ibid., p. 20 (Rougham St), De terra Stannardi et sociorum ibid., p. 25, Stannard et socii sui ibid., p. 26, Stannard ibid., p. 32 (Flempton St); Stanardus auunculus predicti Ricardi (Ricardus filius Ailbricht') 1199 Justiciar's Writ (Pleas, i, no. 3540), totam partem quam Stanhard habuit, de jure Stanardi auunculi ipsius Ricardi 1200 FF (PRS NS, xxvii, no. 237) (Upwell, Outwell Nf): OE *Stänhild fem: 41 Uluric Stanildefilius 1087-1098 (c. 1175-1200) Bury, p. 27 (Pakenham St): OE Stänrnlir: 42 Stanmer (bis), Stanmer Crispinifilius 1087-1098 (c. 1175-1200) Bury, p. 42 (Hopton St).

For further examination of personal names in OE Stän- see the following: Boehler, p. 116; PNDB, p. 371, s. v. Stän-; von Feilitzen, NoB, xxxiii (1945), p. 89; Reaney, Studier i modern språkvetenskap, xviii (1953), p. 102; DBS, p. 331, s. nn. Stammer, Stanberry, Stanhill, Stannard. Björkman's belief (NPN, p. 130, n. 2) that the name element Stän- is to be regarded as the result of the Anglicization of ON Steinn etc. is unfounded, though it might well be possible, as was suggested by von Feilitzen (PNDB, p. 371), that the use of OE Stän- as the first element of dithematic personal names in 1. OE is to some extent due to Scand influence, Scand personal names in Stein- being wellattested in English records. Selten PN, p. 152, considers it to be uncertain whether Staner' in the form Willelmus filius Staner' 1201 CRR 25 (Pleas, iii, no. 117) (Nt) reflects OE *Stänhere, since this Willelmus filius Staner' may be the same person as Willelmum filium Steingrim', acc., 1202 FF (PRS NS, xxvii, no. 459). An examination of the two documents in questions reveals that the William son of Staner' of 1201 CRR 25 is almost certainly the same person as the William son of Steingrim of 1202 FF. Staner' would thus be a mistake for Stancr', which would in tum be a mistake for Stang', an abbreviated form of Stangrim(i). Stengrim in the form Thurstanus filius Stengrim de Dersingham c. 11901220 Coxford Il shows AN monophthongization of OScand ei > e, and cannot, therefore, be used as evidence for the existence of the OEScand monophthongized form Sten- in English records (see Fellows Jensen, NoB, lvii (1969), pp. 69-70).

41 PNDB, p. 371, n. 4, suggested that English forms in Stanilda etc. may altematively belong to a corresponding Scand name as attested by the OSwed Stenhild. The absence of the Scand diphthong ad, ei speaks against this, and there is no reason to doubt that these forms represent an OE *Stänhild. A Scand *Steinhildr is, however, on record in England, forming the first element of the Db field name Steiniltre 1243 (PN Db, p. 762). 42 PNDB, p. 371, suggested that OE Stänrnrer might possibly, in some cases, go back toan ON *Steinmarr, but the absence of spellings showing the Scand diphthong in the first element speaks against this, and there is no need to doubt that OE Stänrnrer is a native formation.

344

Steinn, Sten

ON Steinn, ODan, OSwed Sten b) (i) terram Alfrici Stein 1155-1166 (1. 13th) Holme, no. 202 (the Flegg hundreds). Lefwinus Stein 1209 AssRoll 558 (Pleas, iv, no. 4094) (Barnbarn Broom).

The personal name ON Steinn, ODan, OSwed Sten is an original byname belonging to the appellative ON steinn, ODan, OSwed sten m. 'stone' (NK VII, p. 43; Otterbjörk, Svenska förnamn, p. 126). ON Steinn etc. is well attested throughout medieval Scandinavia (Lind, cols. 956-957; LindS, cols. 757-765; DgP, cols. 1281-1283; SRSö, nos. 88, 139, 254; SRSmå, nos. 61, 109; SRU, nos. 32, 119, 145, 361,376,451, 924; Lundgren-Brate, pp. 241-242; Modeer, Personnamn i Ktb, pp. 27, 28; Thors, pp. 84-85). In England ON Steinn, ODan, OSwed Sten is found both independently and as the first element of place-names and field names, though it is not particularly frequent (PNDB, p. 373; SPNT, p. 144; DBS, p. 333, s.n. Stein; SPNLY, p. 263; DEPN, pp. 436,441, s. nn. Stainsacre, Stainsby Db, Stenson; PN NRY, pp. 123, 171, 333; PN Db, pp. 270, 354, 616, 762; Insley, JEPN, x (19771978), p. 49). The appellative ON steinn probably occurs as a byname in the Norwegian form Porer stain (scribal error for stEin) in a document of c. 1400, though Lind posed the question whether this byname might derive from a place-name, Marstein, which belonged to the same parish as the document in question (LindBiN, col. 358). In Denmark Sten occurs in the later Middle Ages as the surname of a North Jutland noble family which had probably come from Sweden (DgP Il, cols. 1065-1066). OE stän, ME stijn m. 'stone' appears as a byname in the form Robertus Ston 1212 CRR 56 (Cur, vi, p. 385) (Bainton 0) (see OBS, p. 335, s.n. Stone). The byname Stein in the above Nf forms probably does not reflect the personal name ON Steinn etc., but is rather ON steinn, ODan, OSwed sten or a Scandinavianized form of an original OE stän used as a byname.

ON Stigandr b) (i) StigandTRE DB, fös. 158 (gen.), 163b (gen.), 171 (abl.), 205 (abl.), 251b, Stigand episcopi(gen.)ibid.,fö.184,Stigandusibid.,fös.116(2), 116b(3), 120b(2), 121, 135b, 136, 136b (2), 137b, 138b (2), 139b (2), 140b (2), 141 (2), 142 (3), 143, 143b, 162, 163b, 173b, 188, 188b, 190,191b, 195, 197b,201b,202b,207,221b,222,222b,226, 227b,228b,229b,230,232b,238,243b,244,250b,256,256b,271,274b(3),277,278b (2), Stigandus archiepiscopus ibid., fös. 118b, 137b, Stigandi (gen.) ibid., fos. 114, 116, 116b, 121 (2), 130b, 138 (2), 139 (6), 139b (3), 140, 141b, 143b (2), 144b, 150b, 151b (2), 152b (3), 153 (3), 159, 163b, 165, 174b, 175, 175b (3), 176b, 181b (2), 186 (5),

345

Stigandr 186b (3), 188b (2), 189, 189b, 190 (2), 195, 197b, 199,202 (2), 210b, 217, 227b (3), 230(3),240,243b,244,246,247,250,250b,252b,254,258b,259,259b,260,263b (2), 264b, 265 (2), 266b, 276, Stigandi archiepiscopi (gen.) ibid., fös. 123b, 124b, Stigandi episcopi (gen.) ibid., fös. 135b, 175b, Stigando (dat.) ibid., fo. 139b, Stigando (abl.)ibid.fos. 110, 117b, 125,127,140,143 (4), 145b, 159b, 165b, 166,173, 173b(2), 176, 176b(2), 177, 177b(2), 180b, 181,222b,224b,225b,227,227b(2),229b,234b, 243b, 245b, 248b, 252b, 259 (2), 259b, 266, Stigando archiepiscopo (abl.) ibid., fo. 170b, Stigando episcopo (abl.) ibid., fo. 175b, Stingandus ibid., fo. 140, Stingandi (gen.) ibid., fo. 210b (all these forms relate to Stigand (ob. 1072), bishop of Elmham 1043, 1044-1047, bishop of Winchester 1047-1070, archbishop of Canterbury in plurality with Winchester 1052-1070, who was deposed in 1070; fora full conspectus of DB forms see PNDB, pp. 374-375, whose arrangement is followed above). -Stigandus et Thuruert filius eius 1126-1127 Holme, no. 76, Stigandus presbiter et filius suus ibid., no, 83 (the documents cancern the grant by Stigand and his son of the church of St Michael, Norwich, to the Abbey of St Benet of Holme). - Simundo de Ludhamfilio Stigandi, dat., Stigandus de Stanbrigge (witn.) 1153-1166 (1. 13th) Holme, no. 128 (the document is a grant to Simund of Ludham son of Stigand of land in Burwood and Ludham in retum for a yearly rent of 42 pence; Stanbrigge is unidentified). - Transcriptum de terra que fuit Stigandi filii Godrici, 43 totam terram quam Stigandus filius Edrici tenuit, Stigandus, filius Stigandi Willelmus, bona uoluntate et peticione Stigandi 11681175 Holme, no. 307 (Felmingham). -Eluiuamfiliam Stighand' de Ormesbi 1202 FF (PRS NS, xxvii, no. 405) (Ormesby).

ON Stfgandr is an original byname, 'he who goes with long strides' or 'the swift footed one' (cf. ON stiga 'to go, to step, to step upwards') (NK VII, p. 51). ON Stfgandr has been noted in 14th and 15th century Norwegian sources, and Lind noted that its appearance in England predates its appearance in ON (Lind, cols. 963-964; LindS, col. 771). A weak form, Stfgandi, is represented in Iceland by Stfgandi Kotkelsson, the name of a man from the Hebrides in the Laxdrela saga, and occurs in a Norwegian record of 1521 in the form Stygende T orgysson, in addition to being used in a mythological context as the name of a giant in the Snorra Edda (Lind, col. 963). Stfgandi also occurs as a byname in the Eyrbyggja saga (Prandr stfgandi e. 11th) and in the Hcensa-P6ris saga (PorbiQrn st{gandi 10th) (LindBiN, col. 361). In OSwed a weak variant, *Saghande, has been noted as the first element of two place-names in Östergötland (Franzen, Vikbolandet, pp. 207-208). In England the weak form St{gandi forms the first element of Stainby L and of the lost Stighandebi 12th NRY (DEPN, p. 435, s.n. Stainby; PN NRY, p. 333; SPNLY, p. 266; SSNEM, p. 71). In independent use the name appears in the strong form Stigand or as Latinized Stigandus in English sources. The name is, however, infrequent in English records (NPN, p. 131; ZEN, p. 80;

43 The name of the father of this Stigand is unclear since the rubric of the charter takes it to be Godricus (OE Godr"ic) while the text of the charter takes it to be Edricus (OE Eadrfc).

346

Styreman

PNDB, pp. 374-375; Tengvik, p. 198; SPNT, p. 144; DBS, pp. 333, 338, s. nn. Stigand, Styan; SPNLY, p. 266; von Feilitzen, PNWinton, p. 173). It should be noted that the famous archbishop of Canterbury, Stigand, had strong connections with Bast Anglia, as is shown by the extent of his DB holdings in Nf and Sf. Stigand was succeeded as bishop of Elmham in 1047 by his brother .tEöelmrer who held the see until he also was deposed in 1070.44 Interestingly, the personal name Stigand is well attested in Normandy. In the period before 1066 six bearers of the name are known from Norman records, including a seneschal of the Conqueror (Adigard, pp. 138-139, 315317). In England Stigand, bishop of Selsey from 1070 to 1075 and bishop of Chichester from 1075 to 1087 was doubtless a Norman, and the same is probably also true of the 1086 L under-tenant Robertus filius Stigandi. 45

ODan Styreman b) (i) Matildis que fuit uxor Stermanni 1176-1177 P, p. 134, Matildis uxor Sturmanni 1177-1178 P, p. 24, 1178-1179 P, p. 5, 1179-1180 P, p. 18, 1180--1181 P, 46 p. 84, Matildis uxor Stermanni 1184-1185 P, p. 41 (Nf or Sf, prob. Nf). 47 -Steremanno le bone, abl. (witn.) c. 1190 BodlNjCh, no. 356 (Plumstead), de Steremanno le bee c. 1200 BodlNJCh, no. 357 (Little Plumstead).

These forms belong to an original byname formed from the appellative ON styrimaör, ODan, OSwed styreman m. 'steersman, skipper, sea-captain' (DBS, p. 337, s.n. Sturman). DgP, col. 1301, hasa single example of ODan Styrman used as a personal name in a document of 1412. There are also several examples of a byname Styreman in 14th and 15th century Danish records (DgP II, col. 1090). ODan styreman etc. also appears as a byname in England, in mast cases no doubt with the function of an occupational term (Tengvik, p. 271; DBS, p. 337, s.n. Sturman). It might well be that the use of this Scand appellative as a personal name in England was a development which arose in the Danelaw itself. The use of -e- for OScand -ji- in the forms Stermanni, gen., and Steremanno, abl., reflects the characteristic development of OE, OScand ji in the southeastern dialects of ME, a development which is also found in Nf, though here development to i is more typical, while the use of -u- for -ji- in the form Stur-

44

45

46 47

For Bishop !Eöelrmer of Elmham see PNDB, pp. 184-185, and ASWrits, p. 553. For this form see Tengvik, p. 198, and SPNLY, p. 266. Turmanni CR. DBS, p. 337, s. n. Sturman, assigns these Pipe Roll forms to Sf.

347

*StyrgarOr manni, gen., is intrusive to the area, since it reflects the south-westem and West Midland development of OE, OScand y in ME (Jordan, §§ 39--42; Selten, pp. 116--118).

AScand *Styrgarör b) (i) De Stirgard 1165-1166 P, p. 33 (Nf or Sf).

A Scand *Styrgarör is not on record, but Björkman, ZEN, p. 81, pointed out that it would be a regular formation and compared OEScand Gri6tgarör, Steingarör and Ulfgarör. Possibly *Styrgarör could have been an AScand formation which originated in the Danelaw. It is more likely, however, that we are concemed with an Anglicized form of the Scand Styrger with OE -gär having replaced the OScand -geirr (see below, s.n. Styrger), and with final -d being excrescent and the result of AN influence (see PNDB, § 101). The form Stirgard shows i for OE, OScand y which was usual in the ME dialect ofNf (see Selten, pp. 113-116).

ODan, OSwed Styrger a) (i) Stirgeresgore c. 1260 Coxford Il, fo. 35a(ii), Stiregeresgore c. 1290 Coxford Il, fo. 35b(iii) (f. n. in West Rudham). b) (i) Stergar huscarla. regis. E. TRE DB, fo. 266 (Saxlingham Nethergate). - Stirgaro coco, abl. (witn.) c. 1160-1170 Binham, fo. 173b (printed, SSMEA, Appendix I, no. 37) (grant by Adam filius Alueredi, grandson of Peter I de V aloignes and nephew of Roger de Valoignes, of properties in Ingoldisthorpe to Binham Priory). - Robertus filius Sterger c. 1190-1200 Binham, fo. 5a (printed, SSMEA, Appendix Il, no. 3, p. 267) (Binham). - Matillis filia Stirg' 1210 FF (PRS NS, xxxii, no. 210) (Bumham). - pratum Roberti Hordeme etfil(ii) Stirger 1237-1251 Holme, fo. 106b(iv) (Webfen in Clippesby).

ODan Styrger is rare. DgP, col. 1300, has two examples of the name, one from Falster in the 13th century Cadaster of King Valdemar and one from Skåne in a testament of 1351. In OSwed there are several examples of the name in medieval records, but it is fairly infrequent (Lundgren-Brate, p. 248; Modeer, Personnamn i Ktb, p. 8). The name has not been noted in England outside Nf. Interestingly, the name occurs in Normandy as the first element of the place-name Turcaville (Sturgavilla (p ), Sturgarvilla (p) c. 1048 (1. 17th), Surgavilla (p) c. 1048 (1713), Sturgar villa (p) c. 1048 (1764-1789)), which was wrongly taken by Adigard, pp. 139,416, to contain ON Styrkarr etc. 348

Styrr, Styr Forms in -gar, Latinized -garo, abl., are Anglicized forms in which OScand -geirr has been replaced by the corresponding OE -gär (NPN, p. 203; PNDB, § 41; SPNLY, § 52; above, s. nn. Alfgeirr, A.sgeirr). The replacement of OScand -geirr by -ger in the forms Sterger and Stirger is probably the result of analogy with Frankish names in -ger(= OHG -ger), and such forms cannot, therefore, be taken as evidence for the existence of the specifically OEScand -ger < -geirr in English sources (see Fellows Jensen, NoB, lvii (1969), pp. 70--71, and above, s.n. Alfgeirr, n. 22). Forms in Stir- show -i- for OE, OScand -y-, which is usual in the ME dialect of Nf (see Selten, pp. 113-116, and above, s.n. *Styrgarör). Forms in Ster- show -e- for OE, OScand -y-, which is characteristic of the south-eastem dialects of ME, but which also occurs occasionally in Nf (see Jordan, § 40; Selten, pp. 113-116; cf. also above, s.n. Styreman).

ON Styrr, ODan, OSwed Styr a) (i) Sterestuna 1086 DB, fös. 125b, 139, 186, 210b, 259, 263b (Starston, Earsham H). - Esterestuna 1086 DB, fös. 183, 235b, Steirtuna ibid., fo. 251b (Sturston, Grimshoe H).

ON Styrr, ODan, OSwed Styr is either a short form of names in Styr- or an original byname with the sense 'unruly youth' or 'pugnacious fellow' (NK VII, p. 91). 48 ON Styrr is sporadically attested in Iceland from the time of the settlement onwards, and is quite well attested in Norwegian records between the early 13th century and the early 16th century (Lind, cols. 973-974; LindS, col. 772). In Denmark a single example is recorded in independent use in a Lund obituary of 1241, and the name also forms the first element of the Falster place-name St0dstrup (DgP, col. 1300; DS XI, p. 186). In OSwed there are two runic examples of the name and Lundgren-Brate hasa single form from a medieval Uppland document (SRVg, no. 179; SRU, no. 1067; Lundgren-Brate, p. 247). The earliest bearer of the name in England is a York moneyer of the

48 Cf. ON styrr m. 'a stir, tumult, brawl, disturbance'. Note also the case of Vfga-Styrr (ob. 1008), who originally bore the name Arngrfmr Porgrimsson but who was called Styrr and then Vfga-Styrr because he was 'an excessively overbearing fellow' (LindBiN, col. 367). Here a nickname has superseded the proper baptismal name, and LindBiN, col. 367, took the view that the personal name Styrr probably originated as a result of this particular case. The appearance of the name in England as early as the period 975-978, however, speaks decisively against this argument.

349

SumarHör, -liöi periods 975-978 and 979-985 (Smart, Moneyers, p. 229; SCBI, xxi, p. xlvi). A further pre-Conquest bearer of the name was the Northumbrian magnate Styr Ulfsson, who attests charters between 989 and 1009.49 According to the chronicle attributed to Florence of Worcester, the name was also borne (in the form Stir) by the major domus of Harthaknut (NPN, p. 132). The name is recorded in DB TRE in Landa post-Conquest example is to be found in a L charter of 1144-1154 (PNDB, p. 377; SPNLY, p. 268). An example ofthe use of ON Styrr etc. as a byname is known from a Y charter of c. 1160---1165 (SPNL Y, loc. cit.). In addition to forming the first element of the Nf place-names Starston and Sturston, ON Styrr etc. also occurs as the first element of place-names in Db, La, NRY, WRY and We (DEPN, pp. 65, 440, 452, s. nn. Brigsteer, Stearsby, Sturston; PN Db, pp. 608--609; PN La, p. 164; PN NRY, p. 28; PN WRY, i, p. 22; PN We, i, p. 109).50

ON SumarHör, -liOi a) (i) Sumerdes croft c. 1215 Der, fo. 241a (f. n. in Setchey). - Sumerledescroft 14th Blackborough (f. n. in Wellingham) (OKS). - Sumerdeshil 14th Terrier (f. n. in Quarles). b) (i) Svmerlid 1009-1017 SCBI, vii, no. 1241, Svmrld 1017-1023 SCBI, xv, no. 3848, Svmrleö ibid., nos. 4221, 3849 (moneyer, Thetford). - Svmrd 1062-1065 SCBI, i, no. 952, Sumerd 1062-1065 SCBI, xx, no. 1297 (moneyer, Thetford). - seruitium Sumerledi sacerdotis 1146--1148 Har1Ch47 H 45 (printed, SSMEA, Appendix I, no. 4), totam terram Sumerledi sacerdotis c. 1180-1200 Coxford Il, fo. 3a (East Rudham). -Sumerda Cusin 1165-1166 P, p. 28, 1166--1167 P, p. 28 (Lynn). -Petrusfilius Sumerledi, Petrus filius Sumered I. 12th-e. 13th Coxford Il, fo. 33b(iii), terram Sumerledi ibid., fo. 34a(ii) (West Rudham). - Sumerd filio Christien, abl. (witn.) c. 1190 CAcre, fo. 44a (East Barsham), Carta Galfridi Sumerede de Barsham, Galfridus filius Sumeredi de Estbarsham c. 1210 Wals, fo. 19a(i), quam Gaufridusfilius Sumerede dedit c. 1225 Wals, fo. 13a(i) (Walsingham; Barsham, Estbarsham is East Barsham). - Nicholao filio Sumeredi, abl. (witn.) c. 1200 CAcre, fo. 40a(i) (Rudham). - Willelmus filius Sumerith 1201 (13th) Rams, iii, p. 292 (Walsoken). - Sumered (acc.), Sumered' (dat., nom.) 1202 FF (PRS NS, xxvii, no. 318) (lslington). - Carta /ohannis filii Sumerdi de Tiln', Johannes filius Sumerdi de Tilneia, terram que fuit Arketelli filii Sumerdi c. 1215 Der, fo. 127b(ii) (Tilney). - Vnfridus filius Sumerede, Sumerede filius Wace 1222 ElyA, fo. 163a, Sumerede Tallur ibid., fo. 163b, Sumerede ibid., fös. 165b (bis), 166b (Terrington). Sumerede 1222 ElyA, fo. 171a (West Walton).

For Styr Ulfsson see ASChR, p. 375, and Hart, ECNE, p. 360. Ekwall, DEPN, p. 65, s. n., took Brigsteer We, to contain ON Styrr etc., but A.H. Smith, PN We, i, p. 109, while admitting the possibility of this, remarked that the regular ME form -ster(e) and the later -stear(e), -steer would favour OE steor 'steer, young bullock', or the ME nickname Steer derived from it. 49

50

350

SumarHOr, -liöi ON Sumarliöi and its strong side-form Sumarlför are original bynames with the sense 'summer traveller', that is, one who went in summer on Viking expeditions (NPN, p. 133; SPNLY, p. 271). Cf. also ON Vetrliöi, -Uör, an original byname with the sense 'winter traveller'. It should be noted, however, that Assar Janzen, NK VII, p. 45, took ON Vetrliöi to be identical with the appellative ON vetrliöi m. 'a young bear going into its second year', and suggested that Sumarliöi was formed from Vetrliöi through 'variation in meaning' (betydelsevariation). Janzen's explanation is, however, less plausible than the usual explanation of Sumarliöi, -Uör and Vetrliöi, -Uör as original bynames with the sense 'summer traveller' and 'winter traveller', respectively. 51 The possibility that the use of the personal name Sumarliöi might originally reflect the influence of the appellative OE sumorlida m. 'a summer army' may also have to be considered. The use of the personal name ON Sumarliöi seems to have originated in the Norse settlements of the islands of the western Atlantic (Lind, col. 976). ON Sumarliöi is well attested in medieval Iceland from the time of the settlement onwards and is also known from the Faeroes, the Orkneys and the Hebrides (Lind, cols. 975-976). In Norway Sumarliöi is found in a document of 1361 and there are a few examples of the strong form Sumarlför in records of the 14th and 15th centuries (Lind, cols. 976-977). In Denmark Sumarlipi occurs as the name of an 11th century moneyer but, as Kristian Hald pointed out, this name is not otherwise known in OEScand, and it is probable that we are here concemed with an immigrant from the Danelaw (DR, Mp. 74a-c, 75a-b, 76a-c, col. 718; Hald, Om Personnavnene, p. 186). 52 In England, where it appears as Sumerlid, Sum(m)erled(e), -(a), Sumerd(e), -(a), Sumered(e) etc., ON Sumarliöi, -liör is well attested in independent use, and is also found as the first element of place-names and field names (NPN, pp. 133-134; PNDB, pp. 377-378; SPNT, p. 144; DBS, p. 338, s.n. Summerlad; SPNLY, pp. 270-271; Smart, Moneyers, pp. 229-230, 235, 240, 244; von Feilitzen, SCBI, xi.i, p. 12; Okasha, no. 98; Coinage of Edgar, p. 204; DEPN, pp. 430,431, s. nn. Somerby, Somerleyton, Somersby, Somerton Sf; PN ERY, p. 325; Insley, JEPN, x (1977-1978), p. 52). It has recently been suggested that Somerby L (3), Lei, and Somersby L contain the appellative ON sumarliöi 'summer traveller', perhaps used as a byname (SSNEM, p. 70), but, in view of the relative frequency of the personal

51

For further discussion see Miiller, Studien zu den theriophoren Personnenamen der Ger-

manen, pp. 137,227. 52 DR, col. 718, s. n. Sumarli]Ji, Anm. 1, points out that -le]Ja in sumrlejJa, a runic form of this moneyer's name reflects Anglicization.

351

Sunnulfr, Sunnolf, Sun(n)ulf

name ON Sumarliöi, -Uör in English records, this explanation is unnecessary. It has also been suggested that the WRY field name Sumer lethome 1541, which was taken by A.H. Smith, PN WRY, iv, p. 52, to be possibly a compound of ON Sumarliöi and the Scand place-name element -holmr, is a fossilized OE dat. pl. *sumor-hleoöum 'at the summer slopes' (SSNEM, loc. cit.). This field name form is, however, too late to allow more than tentative speculation and its etymology must be left open. Forms with final -e, -a (< i through Anglicization) go back to the weak Sumarliöi (PNDB, p. 377). The forms Sumerledi, gen., Sumeredi, gen., and Sumerdi, gen., have been Latinized according to the Latin second declension (PNDB, §§ 153, 154; SPNLY, §§ 148, 149). In Normandy ON Sumarliöi has been noted as the first element of a place-name (Adigard, pp. 141-142, 417).

ON Sunmilfr, ODan Sunnolf, OSwed Sun(n)ulf b) (i) Sunnulf 1165-1166 P, p. 25, 1166-1167 P, 53 p. 25 (Lynn), Sunnolfo, abl. (witn.) c. 1165 (15th) NorwCPCh, no. 124 (Lynn and Mintlyn), 54 Sunnulfus de Lenna 11741175 P, p. 119, de Sunnulfo de Lenna 1175-1176 P, p. 66, selde Sunnolfi c. 1180 Wals, fo. 124b(vii), fletam Sunnolui c. 1200 Der, fo. 245a(i), fletam Sunnolwi ibid., fo. 245b(i) (Lynn). -Sunolf Flathe 1203 AssRoll 817 (NthRS, v, no. 802) (Martham).

ON Sunnulfr etc. was taken by Lind, col. 980, to have perhaps been formed from the fem. ON Sunnifa, ODan Sunniva, a Scand loan of the OE *Sunngi-

Ju_ss

ON Sunnulfr is extremely uncommon in Iceland, but is quite well attested in Norway from the 13th century onwards (Lind, cols. 979-980; LindS, col. 774). The name also appears infrequently in Denmark and Sweden (DgP, col. 1311; Lundgren-Brate, p. 250). In England ON Sunnulfr etc. occurs as the name of a York moneyer of .tEthelred II and Cnut and appears sporadically elsewhere (ZEN, p. 81; PNDB, p. 378; SPNLY, pp. 272, 355; Smart, Moneyers, p. 230; von Feilitzen, SCBI, xi.i, p. 12; Smart, Cnut's York Moneyers, p. 227; SCBI, xxi, pp. xlvi, xlvii; PN NRY, pp. 66, 192; PN WRY, i, p. 334). It is noticeable that Sunnulfr etc. does not certainly appear in Scandinavia

Sunnulf CR. For the dating of this document see above, s.n. Arnkell, n. 34. 55 For ON Sunnifa, ODan Sunniva < OE *Sunngifu see NK VII, pp. 144, 228. For a conspectus of forms see Lind, cols. 978-979; LindS, col. 773; DgP, col. 1310. 53

54

352

Svanhildr, Swanhild before the 12th century. 56 It is also worthy of note that the personal name element Sunn- is not common in Scandinavia, whereas it is well represented in WGerm. It would, therefore, seem probable that this name element was a loan in Scandinavia from WGerm, i.e. from England (cf. ON Sunnifa, ODan Sunniva < OE *Sunngifu). 57 It might well be that ON Sunnulfr etc. is, in fäet, an AScand hybrid composed of OE Sunn- and OScand -ulfr, which originated in the Danelaw and which was passed from here to Norway. Altematively, we may go back to the suggestion of Lind, col. 980, and take ON Sunnulfr etc. to be a new formation which arose in Scandinavia by analogy with the fem. Sunnifa, itself an English loan.

ON Svanhildr, ODan Swanhild fem. b) (i) terram Swonild' c. 1195 CAcre, fo. lOa(i) (Castle Acre). - terram /ohannis Swonildm. 13thDer, fo. 118b(i) (Bast Walton). -Rogerus Swonild' e. 14thBinham, fo. 194b (Gunthorpe).

ON Svanhildr is probably a Continental loan, cf. OHG Swanahilda, Swanohilt (NK VII, p. 91). The name probably came to Scandinavia with the VQlsungNibelungen cycle, in the Scand versions of which Svanhildr occurs as the name of the daughter of Sigurör Fafnisbana. The Svanhildr of the Scand versions of the VQlsung-Nibelungen cycle was based on a Gothic personage, Sonilda, who appears in the 6th century Gothic history of Jordanes, and who was put to death by the famous 4th century Gothic king Ermanarich. The name of the legendary personage Svanhildr is recorded in both Norse and Danish, but Svanhildr as the name of actual historical personages is only found in OWScand (Lind, col. 984; DgP, col. 1313).58 In England the name is uncommon, but it is found both in the Danelaw and in regions which lay outside the area of extensive Scand settlement (ZEN, p. 82; Boehler, pp. 117-118; DBS, p. 340, s.n. Swannell; SPNLY, pp. 355-356; PN Nth, p. 291; PN WRY, vii, p. 301). Boehler, pp. 117-118, takes English Suanild etc. to be either Scand or Continental in origin, but also includes the

56 A Sunnolfr appears in one version of the Landnamab6k arnong the original settlers of Iceland, but his narne is rendered in other versions of this text as Gunnolfr (Lind, cols. 979-980; islendingab6k: Landnamab6k (lslenzk fornrit, i), ed. J. Benediktsson (Rekjavik 1968), p. 236). This single form might well be the result of scribal error and it is impossible to use it as evidence for the existence of ON Sunnulfr etc. in the Viking period. 57 Cf. NK VII, p. 112. For English names containing OE Sunn- see von Feilitzen, PNWinton, p. 172 and n. 6. 58 For a convenient summary of the development and the various versions of the stories centered on Sonilda/Svanhildr see H. Uecker, Germanische Heldensage (Stuttgart 1972), pp. 63---69.

353

*Svartbrandr

förmal possibility of OE origin. The Continental name does not appear in 10th and 11th century Normandy and Flanders, the areas of the Continent with which the English had direct contact, and, therefore, does not really come into question as a possible source for English Suanild, Swonild etc. The late appearance of these forms in England and the absence in OE literature ofreferences to the stories conceming Sonilda/Svanhildr speak against a native origin for these forms and favour a loan from Scandinavia, 59 where the name occurs in independent use and where the Sonilda/Svanhildr cycle is directly evidenced. English forms in Swonild, Swonild' show ME Q < OE ä, which is a result of the confusion of the first element of the Scand Svanhildr with the OE personal name Swän or the appellative OE swän m. 'shepherd, swineherd; peasant; young man, warrior'.

AScand *Svartbrandr a) (i) Swartebrondhil 1269 Ass (f. n. in Weston Longville) (OKS).

*Svartbrandr is not on record in Scandinavia itself but is a perfectly regular formation, cf. such Scand names as ON Svartkell, ON Äsbrandr, ON Porbrandr. The absence of *Svartbrandr in Scandinavia itself suggests that it is an AScand formation which originated in the Danelaw where it is not infrequent (NPN, pp. 135-136; ZEN, p. 82; PNDB, pp. 378-379; SPNT, p. 144; SPNLY, pp. 274-275).

OScand *Svartgeirr (ON Sverkir, S0rkvir, ODan Swerkir, OSwed Svrerker) b) (i) Wulnodfilius Suartgari 1165-1166 P, p. 23, Wulnod' filius Suartgari 1166--1167 P, 60 p. 23 (Lynn). -Basillamfiliam Swartgar' 1202 FF (PRS NS, xxvii, no. 325) (Massingham).

This name is taken by Janzen, NK VII, p. 104, to represent an original byname formation, 'Geirr the black'. *Svartgeirr is not on record in Scandinavia itself, but the later forms Sverkir, Sr/Jrkvir etc. 61 are well evidenced (Lind, cols. 1021-1023; LindS, col. 785;

59 It should be noted, however, that Beowulf, Widsith and Deor all contain references to Ermanarich. 60 Wlnod' filius Suardgar' CR. 61 Noreen, AschwedGramm, § 117 Anm., took OSwed Swarkir, Swa:rkir to have *Swark-weR

354

Svartr, Swart, Svart DgP, cols. 1330--1331; SRÖg, no. 67; SRSö, nos. 209, 300; Lundgren-Brate, pp. 254-255; Modeer, Personnamn i Ktb, p. 9; Modeer, Östsmåländska bondenamn från 1500-talet, p. 59; Thors, p. 86; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 143). In England the name occurs among the names of .tEthelred Il's moneyers at Stamford and has also been noted in Y (Smart, Moneyers, p. 258; PNDB, p. 379; SPNLY, p. 275; PN NRY, p. 333). Forms with -gar- for OScand -geirr in the second element show the replacement of this Scand element by the corresponding OE -giir (NPN, p. 203; PNDB, § 41; SPNLY, § 52; above, s. nn. Alfgeirr, Äsgeirr). In Ireland the name appears as Suartgair, Suairtghair6 2 in annals relating to the year 1014 (Marstrander, p. 54).

ON Svartr, ODan Swart, OSwed Svart b) (i) Suart. liber homo TRE DB, fo. 243b (held at Spixworth, sub Heroldo).

ON Svartr etc. is an original byname belonging to the adjective ON svartr, ODan swart, OSwed svart(er) 'black' (NK VII, p. 50; Modeer, SvP, p. 36). ON Svartr is frequent in lceland throughout the Middle Ages from the time of the settlement onwards, but it is uncommon in Norway (Lind, cols. 987989; LindS, col. 773). Only a few examples of this personal name are known from ODan and OSwed (DgP, col. 1314; SRU, no. 1118; Lundgren-Brate, p. 251). Note also that ON svartr etc. is used as a byname throughout Scandinavia (LindBiN, col. 373; DgP Il, cols. 1103-1104; Modeer, SvP, p. 101). A weak variant, Svarti, is well attested as a byname in Norway and Iceland, and has been noted as a personal name and as a byname in Sweden (LindBiN, cols. 371-373; Lundgren-Brate, p. 252; NPN, p. 135, n. 1; G. Setterkrans, 'Några medeltida adelsnamn', Personnamn från medeltid och 1500-tal, ed. I. Modeer (Stockholm 1957), p. 107). In England ON Svartr etc. occurs in the OE period in the forms Swart and Swert as a moneyer's name (York, Lincoln, Stamford), and, in addition to being found in Nf, has been noted in DB TRE in Sf and Y (NPN, p. 135; PNDB, p. 379; SPNLY, p. 274; Smart, Moneyers, p. 230; SCBI, xxi, pp. xlvi, xlvii).

as original base. Brate, SRÖg, p. 69, took OSwed Svcerkir, Olcel Srprkver to go back to *svarka-wfhaR and compared the first element of this with Olcel svarkr m. 'a proud, haughty woman'. Björkman, NPN, p. 136, n. 1, suggested the probability that OSwed Swcerkir, Swarkir belongs to *Suartger (i.e. Svartgeirr), and this is the interpretation favoured by more recent discussions of the name (Wessen, SRU, i, p. 30; NK VII, p. 104; Modeer, Personnamn i Ktb, p. 9; Otterbjörk, Svenska förnamn, pp. 127-128. 62 Suairtghair is scribal error for Suarrtghair (Marstrander, p. 54).

355

*Sveingrimr The OE form Swert shows the influence of the etymologically identical adjective OE sweart 'black'.

AScand *Sveingrimr b) (i) Sueingrim 1161-1177 (13th) Rams, iii, p. 286 (Wimbotsham and Downham Market).

*Sveingrfmr is not recorded in Scandinavia itself but is a perfectly regular formation, cf. such names as ON SveinbiQrn, ON Asgrfmr, ON Porgrfmr. The absence of the name in Scandinavia suggests that it is best regarded as an AScand formation which originated in the Danelaw (see von Feilitzen, Notes, pp. 60, 63, where this name is also examined). ON Sveinn, ODan Swen, OSwed Sven a) (i) Torp 1086 DB, fos. 188, 240b, 265 (bis), Sueinestorp ibid., fo. 205, Sueinestorp (p) 1198 AssRoll 559 (Pleas, ii, nos. 2, 8), Sweinestorp 1202 FF (PRS NS, xxvii, nos. 276, 299) (Swainsthorpe, Humbleyard H). - Sweyniscroft 1271 Lynn (f. n. in Terrington St Clement) (OKS). b) (i) Swegner (gen.) 63 c. 1020--1050 (l. 13th) ASWills, no. 25 (Swegn's sons were the nephews of Thurketel Heyng who bequeathed them property at Scratby in this document). 64 - Suein 1161-1167 (13th) Rams, iii, p. 267 (Ringstead and Holme next the Sea). - Suein de Gaiwuda 1165-1166 P, p. 25, Suein de Gaiwud' 1166--1167 P, p. 25 (Lynn), Swein de Gaiwuda (abl.) (witn.) c. 1165 (15th) NorwCPCh, no. 12465 (Lynn and Mintlyn; Gaiwuda, Gaiwud' is Gaywood). - Suein de San/ord 1165-1166 P, p. 25, Suein66 de Stanford 1166--1167 P, p. 24 (Lynn; S(t)anford is unidentified, but it might possibly be Stanford in Grimshoe H, though this is a considerable distance from Lynn). De Suein G6t 1165-1166 P, p. 35 (Nf or St). - Suein M ercator 1166--1170 RBE, ii, p. cclxviii, no. 2 (Castle Rising). - Swein de Hecham (dat.), predictus Swein 1175-1186 Holme, no. 214, domum Reineri filij Swein, quatuor acras terre quas predictus Swein tenuit 1186--1210 Holme, no. 277, Reinerifilij Swein, gen., iiij.or. acras terre quas Swein tenuit ibid., no. 278, Transcriptum Reineri Sweyn, Reinero filio Swein, dat., ibid., no. 282 (HeighamnextNorwich). -Suein de Risinga 1180--1181 P, p. 89,Suein de Risinges 1181-1182 P, p. 70, 1182-1183 P, p. 11 (Nf; Risinga, Risinges is Castle Rising or Woodrising). -Robertumfilium Sweini 1199 FF (PRS NS, xxvii, no. 229) (Stow Bardolph). - terram Swein e. 13th Der, fo. 20a(i) (West Dereham). - W alterus f Swein 1203 AssRoll 817 (NthRS, v, no. 871) (Hasleg'). - De Simone f Swain 1203 AssRoll 817

63 64 65

66

Swegner is certainly scribal error for Swegnes on the part of the 1. 13th century copyist. For the family of Thurketel Heyng see below, s. n. Porkell. For the dating of this document see above, s.n. Arnkell, n. 34. Swein CR.

356

Sveinn, Swen, Sven (NthRS, v, no. 780) (Nf), Simonf Swain 1203 AssRoll 817 (NthRS, v, no. 807) (Weston Longville). -Sueno (brotherofGodefridus f. Ailwini) 1203 P, p. 245 (Nfor St). -Margaretam filiam Swain 1204 CRR 6 (Pleas, iii, no. 1028) (Nf). - Sueinum 1207 CRR 45 (Cur, v,p. 98) (Nf).-SueindeClenchwarton' (dat.) 1207 CRR45 (Cur, v, p.117) (Nf; Clenchwarton' is Clenchwarton). - Swain 1209 AssRoll 558 (Pleas, iv, no. 4342) (Cune gton'). - Margareta que fuit uxor Gaufridi fil ii Swain' 1209 AssRoll 5 58 (Pleas, iv, no. 4347) (Nf). - Rogerus f Suein' 1209 P, p. 51 (Nf or St). - toftam Rogeri filii Swein c. 1210 Der, fo. 17a(ii), terram Rogerifilii Sweyn ibid., fo. 22b(ii) (West Dereham). - Thomas filius Suain 1222 ElyA, fo. 163a (Terrington). - Thede relicte Sweyn (gen.) c. 1230 Binham, fo. 169b(i) (Great Ryburgh). -Swaynum Spalle 1232 CRR 111 (Cur, xiv, p. 448) (Middleton).

The personal name ON Sveinn, ODan Swen, OSwed Sven is an original byname belonging to the appellative ON sveinn, ODan swen, OSwed (run.) sweinn, OSwed sven(n) m. 'young fellow, lad; free man in the service of another man; page' (NK VII, pp. 44, 267; Hald, Personnavne i Danmark: I. Oldtiden, p. 62; Otterbjörk, Svenska förnamn, p. 127). In Norway ON Sveinn is well attested from the 10th century onwards, but it did not become frequent in lceland until the later medieval period (Lind, cols. 993-996). In Danish and Swedish the name is frequent from the Viking period onwards (DgP, cols. 1315-1326, 1694; DR, nos. 1, 3, 90, 163, 184, 277, 328, 336, 384, 389,394,403, 404, M0. 83-90; SRÖl, nos. 9, 18, 26, 27, 39, 40; SRÖg, nos. 83, 92, 100, 140, 147, 161, 218, 220, 231; SRSö, nos. 4, 14, 25, 32,60, 127,136,142,151,165,198,212,214,215,229,240,262,276,286, 290,353; SRSmå, nos. 13, 32, 46, 64, 86, 122, 140, 142; SRVg, nos. 44, 49, 50,58,76,103,133, 165, 192,202;SRU,nos.24,32,34,79,80,86,103, 104, 106, 109, 111, 125, 135, 150, 163, 174, 181, 204, 237, 238, 240, 241, 247, 284,317,326,327,340,354,377,378,382,390,392,395,518,525,539, 566,567,580,624,684,687,727,775,879,899,905,913,915,964, 1065, 1122, 1149; Lundgren-Brate, pp. 252-253; Modeer, Personnamn i Ktb, pp. 28, 63, 65, 69; Modeer, Östsmåländska bondenamn från 1500-talet, pp. 65, 66, 67; Thors, pp. 86--87; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 142-143). ODan Swen is not found as the first element of place-names in -lej and -stath, but forms the first element of around 30 Danish place-names in -thorp (Hald, Personnavne i Danmark: I. Oldtiden, p. 62; Hjorth Pedersen, Bebygg. på -by, pp. 35-36; DS II, p. 29; DS VII, p. 353; DS IX, p. 73; DS XII, p. 22; DS XIII, p. 219; DS XIV, pp. 68,196,229; DS XVI, pp. 167-168; DS XVII.I, p. 225; DS XVIII.I, p. 115). The frequency of ODan Swen in Danish runic inscriptions and -thorp names is taken by DgP, col. 1326, to be a sign of the popularity of the name e, which, in any case, seems to belong toa period subsequent to that of the major Scand settlements in eastem England, but rather shows AN monophthongization of OScand cei, ei > e (see Fellows Jensen, NoB, lvii (1969), pp. 67-70). The largely WSax development of ceg, eg > ce, e before n, d and ö (for which see Luick, § 251; Brunner, AeGramm, § 214.3; Campbell, § 343), which would have given a 1. OE *Swen < Sweg(e)n, need not be considered as a source for Suen in East Anglia. The universal use in the above forms of -u- and -w- for OScand -v- merely reflects the fäet that OScand v was stilla bilabial /w/ at the time of its adoption into English (PNDB, p. 75 (§ 55) and n. 1; SPNLY, § 67). A form Swan(us), Suan(us) is also on record in English sources. It is usually taken to be an Anglicized form of ON Sveinn etc. (NPN, pp. 139-140; PNDB, pp. 66 (§ 41), 380-381; Tengvik, p. 199; DBS, p. 340, s.n. Swan; SPNLY, pp. 276-282, 356). In some cases this is undoubtedly true, e.g. Suano f Westmundi, abl. (witn.) 1150-1160 (1. 12th) YCh, Suanof Westmund, dat., c. 1160-1180 YCh (Wentworth WRY), Sueinus f Westmundi 1166-1167 P (Y) (SPNL Y, p. 279); Suanus carpentarius, Robertus filius Suani 1221 AssRoll 948 (Eyre III, no. 412) (Baddesley Clinton Wa), De Swein' carpentario, De Roberto filio Swein 1221 AssRoll 950 (Eyre III, p. 421) (Wa). A Nf example of this is Swanus bercarius de Clenchwarton' 1207 CRR 45 (Pleas, iv, no. 2769), who is identical with Suein de Clenchwarton' (dat.) 1207 CRR 45 (above, p. 357). In these cases ON Sveinn etc. has been influenced by the

68 The personal name is, of course, formally possible as a base for the byname, but, in view of the fäet that the appellative passed into the general vocabulary of English, it would be wrong to regard the personal name as the sole source, and in most cases we are doubtless concemed with the appellative. DBS, p. 339, s. n. Swain, takes the appellative as base only in forms of the type Walter le Swein, but this is too narrow a definition, and there is no reason why the appellative should not be the base of the byname in forms like the above-mentioned Turold Swein. The personal name could only be postulated with certainty here if the bearer of this name had been found elsewhere as Turold fllius Swein, but this is not the case.

359

Svertingr, Swrerting, SvartungR, SvrertingR corresponding English appellative OE swan m. 'shepherd, swineherd; peasant; young man, warrior'. In cases where Swan(us), Suan(us) occurs alone, however, there is no need to regard it exclusively as an Anglicized form of the Scand Sveinn, for in mast cases it is doubtless an original byname, OE *Swan, formed from the appellative OE swan. A Nf example is contained in the form Robertumfilium Swan' 1202 FF (PRS NS, xxvii, no. 304) (West Winch). The personal name OE *Swan is also the best explanation for the first element of Swannington Nf (Sueningatuna 1086 DB, Suenningeton', Sweinnigeton', Swenigtune, Swenigtun' 1198 FF, Sueington', Swaningeton' 1202 FF). Forms in Swein-, Suein-, Swen-, Suen- show the influence of the Scand personal name ON Sveinn etc. or of the appellative ON sveinn, ME swein, swain etc. To judge from the early forms, Swannington is a genuine -ingatun formation. Ekwall, DEPN, p. 455, s.n., however, wrongly interprets it as: 'Apparently the TON of Svein's people'. OE -inga- had long ceased to be productive in place-name formation at the time of the first Scand settlements in England, and it is thus impossible that the Scand personal name Sveinn could form the first element of a place-name containing this element. Cf. also Swannington Lei, an -ingtun name, which Ekwall, loc. cit., interpreted in the same way as he did Swannington Nf, but which B.H. Cox, PN LeiR, p. 402, rightly regarded as containing the OE personal name *Swan + connective -ing- (-ing 4 ) + tun. Lindkvist, p. 93, took Swannington Nf and Lei to contain an OScand personal name *Sveiningr, a name corresponding to ON Sveinungr, ODan Swening etc., but, in the light of the evidence set out above, this is not a viable alternative etymology. In the case of the Nf Swannington the persistent ME spellings in -inga-, -inge- make it clear that it is a genuine OE -inga- formation, while in the case of the Lei Swannington spellings in initial Swan-, Suan- predominate, and in both cases there is no reason to suppose any first element other than the personal name OE *Swan.

ON Svertingr, ODan Swrerting, OSwed (run.) SvartungR, SvrertingR b) (i) Swyrtinc 979-991, Sweartinc 991-997, Swerting, Swertinc 991-1003, Svertic 997-1003, Swertic, Swertg, S:ertig 1003-1009 Smart, Moneyers, p. 243, Swertinc 991997 SCBI, vii, no. 1033, 997-1003 SCBI, vii, no. 1034, 1017-1023 SCBI, xv, nos. 3222, 3223, Svertic 997-1003 SCBI, vii, no. 1032, Sw:ertg 1003-1009 SCBI, i, no. 736 (moneyer, Norwich). - Suartinc fiber homo TRE DB, fo. 206b (Stradsett). - totam terram Swarting' 1153-1168 Holme, no. 167 (Burwood in Ludham).

ON Svertingr etc. is a derivative of ON Svartr etc. formed with the patronymic suffix -ingr (NK VII, p. 55). ON Svertingr was frequent in Iceland at the time of the settlement but less 360

Srefari, Srefari common later, and is only found sporadically in Norway, where it is known from the early 13th century onwards (Lind, cols. 1000-1002; LindS, col. 774). A byname Svertungr has been noted in 15th century Jämtland (Olaff swertung 1460) (LindBiN, col. 375). In ODan the name is represented by Svertin and Suertins (gen.) in documents from Lund, and also forms the first element of a place-name in -thorp in Skåne (DgP, cols. 1331-1332). The form Suerting also occurs in Saxo Grammaticus as the name of a Saxon prince-this is apparently the same person as the Swerting of Beowulf, whose nefa was the Geatish king Hygelac, the change of nationality in Saxo being the result of confusion in the course of the transmission of the cycle of stories concerning the Geatish and SkjQldung kings (DgP, col. 1331; EN Beow, pp. 110-111). An unmutated ODan *Swarting is taken by DS VI, p. 479, and DgP, col. 1332, to be the first element of the Jutland minor name Svartingsh!iij (Schwartingshuy 1641). In Sweden runic Sw:ertingR has been noted in two runic inscriptions in Södermanland, and a side-form formed with the suffix -ungr, SvartungR, has been noted in a runic inscription in Uppland (SRSö, nos. 68, 217; SRU, no. 1006). In English sources ON Svertingr etc. is quite well represented, being found, inter alia, as a fairly common pre-Conquest moneyers' name, and occurring in DB TRE in Bk, Brk, Ha, L and Sf, as well as in Nf: NPN, p. 137; PNDB, p. 381; SPNT, p. 145; SPNLY, pp. 282-283; Smart, Moneyers, pp. 230,243; vonFeilitzen, SCBI, xi.i, p. 12; Smart, Cnut's York Moneyers, p. 227; PN BdHu, p. 300; PN Db, p. 762. Note also the following post-Conquest Sfforms: Suarting 1087-1098 (c.1175-1200) Bury,p. 30, terramquefuitSwartingi 1186--1188 (c.1270) BuryS, p. 14 (Hessett); Suartingi (gen.) 1186--1188 (c. 1270) BuryS, p. 5 (Beyton), Roggerus filius Swarting ibid., p. 11 (Thurston); Swarting (acc.) (natiuus) 1199 FF (PRS NS, xxxii, no. 286) (Staverton).

Forms in Sweart- and Swcert- are due to association with OE sweart 'black, dark'. Forms in Swart-, Suart- reflect association with OE sweart and OScand Svart(-) (PNDB, p. 381; SPNLY, p. 283). Initial Swyrt- for Swert- in the coin legend Swyrtinc 979-991 is probably the result of an error on the part of the die-cutter. The universal use of -w-, -u- and -v- (= -u-) in the above forms for OScand -v- merely reflects the fäet that this was still a bilabial /w/ at the time of its adoption into English (PNDB, p. 75 (§ 55) and n. l; SPNLY, § 67).

ON Srefari, OSwed (run.) Srefari a) (i) Sefereshowe c. 1215 CAcre, fo. 8b(ii), Sephareshowe ibid., fo. 1 la(i), Sefareshowe ibid., fo. 1 la(iv), Sephereshog' ibid., fo. 1 lb(iv) (f. n. in Castle Acre).

361

Srefari, Srefari b) (i) .v. solidatas terre. quas tenuit Sefara c. 1090-1100 CAcre, fo. lb(i), 1106-1119 CAcre, fo. 112a(i) (Castle Acre). -De Sefare de Suthacra, Sefare hominem meum (Herbert of South Acre) c. 1170 CAcre, fo. 28a(ii) (South Acre). - Carta Godefridi filii Seuerii de Len, Godefridus filius Seuerii de Len c. 1200-1210 Der, fo. 243b(iii) (the document concems the homage of Reginald son of Siric of Lynn69 and the tenement which he held from Godfrey son of Seueri of Lynn at Stangrimeshoec in Lynn). - Sejar' Brokesanke (acc.) 1202 CRR 28 (Pleas, ii, no. 881) (Castr'). - terram que fuit Stangrimi filii Seueri c. 1220 Der, fo. 202b(i), terram que fuit Stangrimi filii Seuere ibid., fo. 203b (Wiggenhall St Mary Magdalene).

This name is an original byname belonging to the appellative ON scefari m. 'seafarer' (SPNLY, p. 284). There are a few fictional examples of ON Scefari in OWScand literary sources, and this personal name, or the appellative to which it belongs, occurs as the byname of the father-in-law of King Harald Fairhair of Norway, Jarl Eysteinn srefari of Heiöm9rk (Lind, col. 1007; LindBiN, col. 378). A single example of OSwed (run.) Srefari is on record (SRU, no. 454). In addition to the above Nf examples, ON Srefari, OSwed (run.) Srefari is known in England as the first element of Searby L (DEPN, p. 410, s.n.; SPNLY, p. 284), though it has recently been suggested that this place-name might rather contain the gen. pl. of the appellative scefari (SSNEM, p. 68). Sf examples of ON Scefari, OSwed (run.) Srefari are: Sephare 1186--1188 (c. 1270) BuryS, p. 40 (Bardwell); Ricardus filius Sefare 1222 ElyA, fo. 198a (Glemsford). Reaney, Studier i modern språkvetenskap, xviii (1953), p. 100, wrongly took the second of these Sf examples and a So form Seuare 1185 Templars, which also has ON Scefari, OSwed (run.) Srefari as etymon, to belong to an unrecorded OE *Srefaru fem. The forms Savari L VD and Sauari de Lafford' 1202 AssRoll (L) are taken respectively by DEPN, p. 410, s.n. Searby, and SPNLY, p. 284, to also belong to ON Scefari, OSwed Safari. It is true that the name element OE Sre-, ON Sce- does sometimes appear as Sa- in English records (see PNDB, § 7; SPNLY, § 41), but the above-mentioned forms are perhaps more plausibly interpreted as belonging to the ContGerm (Visigothic and post-Visigothic Spain, southem Gaul) Sabaricus, Savaricus, OFr Savari (Forssner, p. 223). 70 The use of -e- for ON -re-, OE -re- in the first element of the above Nf forms reflects the 1. OE development of re > f (Luick, § 361 and Anm. 6; Jordan, § 48; PNDB, § 7; SPNLY, § 40). Final -a, -e for -i is the result of Anglicization (PNDB, §§ 45, 150; SPNLY, § 147).

69 de hornagio Regina/di filii Syricii de Lenne in Der, fo. 243b(iii), homag' Regina/di filii Siricii de Linne ibid., fo. 244a. 70 A famous bearer of this name in English history was King John's lieutenant, Savari de Mauleon, who was one of the leading barons of Poitou.

362

*Srefugl The form Seuerii (gen.) has been Latinized through the addition of the gen. sing. ending of the Latin second declension (see PNDB, § 155). The use of -ph- for -f- reflects MedLat orthographic practice (SPNLY, § 97).

AScand *Srefugl b) (i) Sefugel filius Alfyet 1182-1183 P, p. 12, 1184--1185 P, p. 34, Sefugelfilius Alfiet 1183-1184 P, p. 6 (Nf or Sf). -Sefugell' de Tunstall' (acc.), predictus Sefugell', eidem Sefugell', dat., 1198 FF (PRS NS, xxvii, no. 65) (Tunstall), Sefugellum 1203 CRR 32 (Cur, iii, p. 23) (recognitor in a case conceming property at Walesham; Walesham is probably South Walsham), Godwin' filium Sefugol et Alexandrum fratrem suum 1206 CRR 42 (Pleas, iii, no. 1977) (Nf), Godwin' filium Sifugell' 1206 CRR 42 (Pleas, iii, no. 2094) (Tunstall), Godwinumfilium Sefugel 1207 FF (PRS NS, xxxii, no. 116) (Wickhampton).

This name is an original byname, 'sea-bird' (SPNLY, p. 284). It is not attested in Scandinavia itself and is probably an AScand formation which originated in the Danelaw. Cf. also the following examples: Slefuel 1042-1050 SCBI, xxi, p. xlviii (moneyer, York); HalwferlJ. s/efu3alasuna c. 1050 (c. 1075) Björkman, Festermen (facsimile of MS. facing p. 4) (Y); Ediuuu Sleuugeles /af c. 1100--1130 Earle, p. 254 (D); homagium hominis mei Willelmi filii Sefugel (charter of Matthew de Praeres) c. 1165 (c. 1175-1200) Bury, no. 204. liberum homagium Willelmifilii Sefugel c. 1175-118071 (c. 1175-1200) Bury, no. 206 (Wainfleet L); Rannulfumfilium Sefugel 1200 CRR 16 (Cur, i, p. 223) (Sf). Note also: Slefugel 11th LVH, (domus) Safugel TRE LW, Sefuel 12th LVH (von Feilitzen, PNWinton, p. 171 and n. 1).

The above D and Ha forms were taken by von Feilitzen (PNDB, p. 90 (§ 87); PNWinton, p. 171) to reflect an OE Srefugel. An OE Srefugel, -ul does occur in genealogies of the royal line of Deira (see D.N. Dumville, Anglo-Saxon England, v (1976), pp. 30, 32, 35), but actual historical personages hearing the name Scefugel, Sefugel etc. do not appear until the 11th century, and there is no reason to suppose that these 1. OE and e. ME examples of Scefugel, Sefugel, Sefugol etc. represent anything other than an AScand formation which originated in the Danelaw (see NPN, p. 115; ZEN, p. 71; Björkman, Festermen, p. 16; Tengvik, p. 162; SPNLY, p. 284). Semantic parallels to AScand Scefugel, Sefugel etc. < Scand *Srefugl are provided by AScand Sume,fugel < Scand *Suma,fugl and AScand Winte,fugel < Scand *Vet,fugl (see NPN, pp. 133, 177, and von Feilitzen, SCBI, xi.i, p. 13 and n. 1).

71

Bury dates this record to c. 1180 but the MS. allows an upper limit of c. 1175.

363

SregriIDR, *Sregrimr

ODan (run.) SregrimR, AScand *Sregrimr b) (i) S;:egrim 1040---1042 SCBI,xv, no. 3846, S;:egriim 1042-1044 SCBI, xviii, no. 1232 (moneyer, Thetford). - Segrimo forestario, dat., eidem Segrimo, dat., l. 12th Wymondham, fo. 63a(i), Segrimo forestario, dat., I. 12th - e. 13th Wymondham, fo. 63a(ii, iii) (Wymondham). -Alan' filio Segrim, abl. (witn.) c. 1195-1200 CAcre, fo. 36a(ii, iii), Alanus filius Segrim (witn.) ibid., fo. 36b (Wesehamtorp', i.e. the jJorp attached to Weasenham). - Segrim de Corpesti 1209 P, p. 35 (Corpusty).

In Scandinavia itself this name is only represented by an ODan (run.) SregrlmR, which appears as the name of a Lund moneyer of Sven Estrithson (DR, M0. 67a-c). In the forms Sregrim, Segrim, however, this name is quite well attested in English records, and, in view of the strong AScand element among 11th century Danish moneyers, 72 it is most likely that the name of the Lund moneyer is a loan from the Danelaw and that the name itself is an AScand formation which originated in the Danelaw. 73 This probability is reinforced by the form taken by the name in Denmark, runic sekrim (nom.), which corresponds exactly to the English Segrim, a form which shows the 1. OE development of re > f. 74 In England Sregrim, Segrim has been noted in DB TRE in L, Nth (also 1086) and St, in addition to being found among the names of 11th century moneyers and to appearing sporadically in the post-Conquest period (NPN, p. 115; PNDB, p. 353; SPNT, p. 144; OBS, p. 310, s.n. Seagrim; SPNLY, p. 284; von Feilitzen, PNWinton, p. 171). Cf. also the following: Sreigrim 1035-1038 SCBI, xviii, no. 502, Sregrim 1038-1040 SCBI, xviii, nos. 498-501, 1040-1042 SCBI, xv, no. 3483 (moneyer, Nottingham); Segrim Wml (c. 10831086) SCBI, xvii, no. 592 (moneyer, Shrewsbury); Sagrim c. 1131-1134 SCBI, xii, nos. 247-249 (moneyer, Oxford).

AScand *Sreliöi b) (i) Se/ide le Copere 1180---1181 P, p. 91, Se/ide le Cupere 1181-1182 P, p.72,11821183 P, p. 11, 1183-1184P, p. 5, 1184-1185 P, p. 34, 1185-1186P, p. 60, 1186-1187 P, p. 53, 1187-1188 P, p. 56, 1190 P, p. 93, 1191 P, p. 36, 1192 P, p. 181, 1193 P, p. 16, 1194 P, p. 49, 1195 P, p. 65 (Nf or St). 75 Seledus Tingtor' 13th GuildL, p. 46 (Selten PN, p. 134). -Paganusfilius Selid' 1200 CRR 16 (Cur, i, p. 227) (Outwell), Paganusf Se/ed' 1200 Selden Society, iii, p. 2 (Selten PN, p. 134) (Nt), de Paganof Selide 1204

For the AScand element among 11 th century Danish moneyers see Hald, Om Personnavnene. von Feilitzen, PNWinton, p. 171, n. 9, took the Lund moneyer to be an Englishman and not a Dane. 74 For the I. OE development of x > ? see above, s. n. Sltfari, and the works cited there. 75 DBS, p. 304, s. n. Salitt, assigns Se/ide le Cupere to Nf. 72

73

364

*Sreliöi P, p. 246 (Clackclose H). -Rogerusf Sa/ide 1205 P, p. 253 (Nf or Sf). - de hornagio et tenernento Huberti filii Selede de Welles, hornagiurn Huberti filii Selede de Wellis c. 1210 Der, fo. 93a(ii), hornagiurn Huberti filii Selede, terrarn Huberti filii Selede ibid., fo. 94a(ii) (Outwell or Upwell). - de hornagio Lefquenfilie Se/ide de Lenn, hornagiurn Lefquen filie Se/ide de Len c. 1220 Der, fo. 195a(i) (Lynn). - Sed/ed de Sengham, Selled, Seledf Ricardi Champeneys, Henricus Se/ide de Ludeham l250Ass (Selten PN, p. 134) (Nf; Sengham is Shingharn and Ludeham is Ludharn). - dornus Se/ide la Carpentere de Depham 1295 GDR (Selten PN, p. 134) (Deopharn).

An OScand *Sceliöi would be an original byname, 'seafarer'. Von Feilitzen, PNDB, p. 353, regarded an ON *Sceliöi as the etymon ofthe form Salithi 12th L VD, fo. 42a, but took the etymon of the TRE DB form Salide (Ha) to be either the Scand *Sceliöi or an OE *Srelida, an original byname belonging to the appellative OE srelida m. 'seafarer, sailor, pirate'. Von Feilitzen (loc. cit.) considered post-Conquest forms in Selyde, Selide, Salida, Selida, Salidus and Selidus to be 'probably native'. Later von Feilitzen dropped the Scand option completely, for in PNWinton (p. 172) he gave the etymon of Ha forms in Selida and Selid' as 'OE Srelida'. Ekwall, ELPN, p. 59, also took London forms in Salidus and Saled' to belong to the OE name. Similarly, DBS, p. 304, s.n. Salitt, takes ME forms in Salide, Selide (including the East Anglian Selide le Cupere given above), Seled' and Saled to belong toan OE *Srelida. East Anglian examples of Se/ide etc. from the ME period are collected by Selten PN, p. 134, and are taken to belong toan OE *Srelida. Selten notes the existence of the Scand *Sreliöi in East Anglia in the form Selithe de Wenham 1275 RH (Wenham Sf) and considers that the existence of Scand names in -liöi, -lför may perhaps have facilitated the transfer of the appellative srelida to the category of personal names (Selten PN, loc. cit.). However, Dorothy Whitelock, SPNT, p. 146, prefers a Scand origin for the Liber Vitae of Thorney form Scelide, pointing out that the element -liöi enters inta several ON names. She campares especially ON Hafliöi, which is semantically identical with *Sreliöi. A further possibility is that a personal name *Sreliöi arose among the Scand settlers of the Danelaw and spread elsewhere, its form being modified owing to association with the appellative OE srelida to give ME forms in Selida, Se/ide etc. For Sf examples of Se/ide etc. see PNDB, p. 353, and Selten PN, p. 134. For initial Sa- for OE Sre-, ON Sre-, a feature reflecting London influence through the medium of the central administration, see ELPN, p. 185, and Selten, pp. 90-91.

365

T

ON Teitr, ODan, OSwed (p. ns.) *Tet a) (i) Teteslond c. 1190-1200 Wymondham, fo. 49b(i) (f. n. in South Wootton).

ON Teitr, OEScand *Tet is an original byname, 'the glad one, the cheerful one' (cf. the identical adj. ON teitr 'glad, cheerful') (NK VII, pp. 53, 245; Hald, Personnavne i Danmark: I. Oldtiden, p. 59). ON Teitr etc. is a personal name of considerable antiquity. The earliest recorded example is runic tAitR, nom., representing PrScand TaitR, which appears in the Norwegian runic inscription of Tveito in Telemark belonging to the first half ofthe 7th century (Krause-Jankuhn, pp. 202-203; for the date see KrauseJankuhn, p. 317). ON Teitr is well attested in lceland from the time of the settlement onwards and in Norway there area few late medieval examples, in addition to the early runic example from Tveito mentioned above (Lind, cols. 1026-1027; LindS, cols. 785-788). In OEScand *Tet forms the first element of the Danish place-names Testrup and Teestrup and probably also of the Swedish place-name Testa (DS IX, pp. 109, 204-205; Hald, Personnavne i Danmark: I. Oldtiden, p. 59; Sahlgren, NoB, xv (1927), p. 80). In England the name is recorded in independent use as Teit TRE DB, fo. 322b (Cotton Sf), anda byname Tait, Teyt, Teit, Tayte has been noted in postConquest records (NPN, p. 140; ZEN, p. 83; PNDB, p. 382; DBS, p. 342, s.n. Tait).

ODan *Tithi, OSwed *Tidhe (p. ns.) a) (i) Tytheby 1086 DB, fo. 270 (Tyby in Wood Dalling, Eynsford H).

ODan *Tithi, OSwed *Tidhe, which appears to be a short form of a name corresponding to OSwed (run.) Tiökumi or of some other name in Tiö-, has been noted as the first element of place-names in Skåne, Halland, Västergötland and Uppland (SkOA III, p. 112; SkOA XVI, pp. 132-133; SOH I, p. 171; SOSk, xiii, p. 13; Hellquist, Ortn. på -by, pp. 50, 57-58). In England ODan *Tithi, OSwed *Tidhe has been taken to form the first element ofTithby Nt as well as ofTyby in Wood Dalling Nf (DEPN, p. 484, s. 366

Tofa, Töva nn. Tyby, Tythby; PN Nt, p. 242; Sandred 1987:319). However, SSNEM, p. 75, considers that the DB form of Tithby Nt, Tiedebi (bis), indicates that the first element of this place-name may rather be an unrecorded OScand *Työi, perhaps a short form of names in Pi6ö-.

ON Tofa, ODan, OSwed Töva fem. b) (i)/. liberum hominemfeminam Touu TRE DB, fo. 202b (South Walsham).

This name is characteristically OEScand, whence it passed into OWScand, and is a short form of such fem. names as Porfriör, Porvarr, Porve and the like (NK VII, pp. 133, 209; Otterbjörk, Svenskaförnamn, p. 195; Hald, Personnavne i Danmark: I. Oldtiden, pp. 86--87). In Denmark Tova is old and frequent, being known from several runic inscriptions as well as being well attested in medieval records from the 12th century onwards (DR, nos. 55, 78, 98, 143, 317; DgP, cols. 1473-1475). In OSwed the name is known from several runic inscriptions and appears sporadically in medieval records from the mid 13th century onwards (SRÖl, no. 6; SRSmå, no. 76; SRU, nos. 199, 431; Lundgren-Brate, p. 282). ON T6fa is found infrequently in lceland from the end of the 10th century onwards and is extremely rare in Norway, where it is first noted in a document of 1320 (Lind, col. 1032). Lind, col. 1033, noted that the name is older and more common in Denmark and considered it to be probably a loan into OWScand from Danish. ON T6fa, ODan, OSwed Tova is rare in English records (NPN, pp. 140141; ZEN, p. 83). The sole DB example given by PNDB, p. 384, is the above example from Nf. For an examination of the possibility that ON T6fa etc. may also form the first element of place-names in Denmark and England see below, s.n. Toft. Von Feilitzen, PNDB, p. 384, posed the question as to whether the final -u of the Nf DB form Touu (acc.) could reflect the OScand acc. in -u.

OScand *Tofa-Hildr fem. b) (i) de Willelmo filio Touild' 1174--1175 P, p. 122 (Nf or Sf). 1 -Robertusfilius Touild' 1201 CRR 25 (Cur, ii, p. 57) (Nf; Robert son of Touild' was a recognitor in a plea conceming property in Rudham).

1 This form is wrongly rendered as Willelmusfilius Touildus by PNDB, p. 384, and as Willelmusfilius Touild' by DBS, p. 352, s. n. Tovell. Both PNDB and DBS assign this form to Sf, but it is not clear from the text whether it is to be assigned to Nf of Sf.

367

Tofi, Tövi, Töve

Von Feilitzen, PNDB, p. 384, took this compound to have the sense 'Hildr the daughter of T6fi', and compared the formation Töja-Anna (gen.) 'of Anni the son of Töfi', found in a medieval Danish runic inscription. 2 This explanation is probably correct as regards the ultimate etymology of *T6fa-Hildr, though the appearance of several examples in England suggests that it was here regarded as anormal font-name in its own right. The apparent absence of *Tofa-Hildr in Scandinavia itself might be taken to suggest that it is an AScand formation. In addition to the forms given above, the following examples have been noted: Touillda TRE DB, fo. 102b (Fordham Ess); Touilt 1086 DB, fo. 446b (Claydon H Sf). For further examination of *T6fa-Hildr see NPN, p. 14; PNDB, p. 384; DBS, p. 352, s.n. Tovell. Selten PN, p. 158, grouped the above Bast Anglian forms of 1174-1175 and 1201 sub OE *Tunhild fem., arguing, however, that these instances are uncertain since u and n 'cannot as a rule be safely distinguished', though conceding that if the reading is u then we are concemed with the Scand *T6faHildr and not with the OE *Tunhild. In English manuscripts of up to c. 1250, however, it is by no means as difficult as Selten supposes to separate u and n, and, in any case, if we were concemed with OE *Tunhild then we would expect forms in initial Tun- to also appear, but this is not the case. In view of these considerations, there is no reason to suppose that the above forms from Ess, Nf and Sf from the period TRE-1201 represent anything other than an unrecorded OScand (or perhaps AScand) *T6fa-Hildr.

ON Tofi, ODan Tövi, OSwed Töve b) (i) Toft Prude (named as executor in the will of .tElfric Modercope), 3 1042-1043 (1. 13th) ASWills, no. 28 (the document disposes of property at Loddon, Bergh Apton, Barton Turf, Purwineholm and Fuglholm). - Toui. I. fiber homo Gerti TRE DB, fo. 269 (Plumstead). - Toue fiber homo TRE DB, fo. 163 (held in Titchwell sub Heroldo). Toue fiber homo TRE DB, fo. 225b (Bircham Newton). -Touet fiber homo TRE DB, fo. 173b (Ringstead). - Terra Toui 1086 DB, fo. 264b, Tovius, Touius ibid., fo. 265 (properties and rights at Hackford, Holkham, Stoke Holy Cross, Swainsthorpe, Newton Flotman and Kenningham Hall and in Henstead H). - portam que porta Toui cognominatur 1128-1129 Holme, no. 126 (this gate was situated at the Abbey of St Benet of Holme itself which lay near the River Bure just west of Thume). - iuxta terram et domum Thouy c. 1153 Wals, fo. 8a(i) ((? Little) Snoring). -Toue de Pagraue (abl.) (witn.)

2 DR, no. 307 (Lund). This inscription is dated by DR to Periode 4 (Middelalder). DR Periode 4 stretches from the beginning of the 12th century to the mid 14th century, but DR no. 307 would seem to fall within the period 1100--1300 (see DR, cols. 1030--1034, esp. 1032-1034). 3 For the career of the Anglo-Danish magnate Tovi the Proud see ASWills, p. 186, and ASChR,

p.400.

368

Tofi, Tövi, Töve c. 1160 CAcre, fo. 22b(i) ((? Little) Dunham; Pagraue is Palgrave). - de catallis Toui Hering 1176-1177 P, p. 139 (Nf or Sf). - Francplegium Toui filii Algari de Renham 1178-1179 P, p. 8, 1179-1180 P, p. 19, Francplegius Toui filii Algari de Renham 1180-1181 P, p. 85, Francplegii Touifilii Algari de Renham 1181-1182 P, p. 7 (Raynham). Alanusfilius Toui 1184-1185 P, p. 38, de Alano filio Toui 1185-1186 P, p. 63 (Nf). Decena Huberti Toue 1197 P, p. 245, 1198 P, p. 87, Decena Ricardi Toui, pro fuga Hugonis Toui 1197 P, p. 243, Decena (Henrici4 ) (Ricardi 5 ) Toui, pro fugaHugonis Toui 1198 P, p. 88 (Nf or Sf). - Walterus filius Toui de Wychamtune, Walterus filius Toui 13th Langley, fo. 16a(iii) (de uicario de Hoghesmers et de Lamesholm, prob. in Wickhampton; Wychamtune is Wickhampton). - de hornagio Willelmusfilius (sic) Towi de Suttune, homagium Willelmi filii Towi de Sutthne e. 13th Der, fo. 23 la(iii) (charter of John son of Reginald of Clenchwarton; Suttune, Sutthne is probably Long Sutton L). Thoui Busting (acc.), de predicto Thoui, predicto Toui, dat., 1202 FF (PRS NS, xxvii, no. 410) (Grouele). - Godardum filium Toui 1202 FF (PRS NS, xxvii, no. 416) (Moulton St Mary). - Botildam filiam Touy 1202 FF (PRS NS, xxvii, no. 457) (Palgrave ). -Toui longus 1209 AssRoll 558 (Pleas, iv, no. 4145) (Toui longus was involved in a plea of novel disseisin falsely brought by John the parson of Reedham). - Reginaldus filius Toui de Holcham c. 1210--1220 Binham, fo. 146a(ii), Carta Reginaldi Thoue, Reginaldus filius Thoue de Holcham c. 1210--1220 Wals, fo. 5la(i), c. 1220-1240 Wals, fo. 5la(ii, iii, v), Carta Reginaldi Thoue, Reginaldus filius Thoue ibid., fo. 51a(iv), terram quam Reginaldus filius Thoue dedit infirmis de Len ibid., fös. 50a(v), 52a(iii), quas Reginaldusfilius Thoue nobis dedit (charter of Prior William of Walsingham) ibid., fo. 54a(i), de hornagio et seruicio Reginaldi filii Thoue ibid., fo. 53a(iii), homagium et totum seruicium Reginaldi filii Thoue de Holcham ibid., fo. 53b, de portione vnius crufte que fuit Reginaldi fil i i Thoue, Reginaldus filius Thome filii Reginaldi Thoue de Holcham c. 1285 Wals, fo. 5la(vi), in tota portione mea (i.e. Reginald son of Thomas son of Reginald Thoue of Holkham) illius crufte que quondam fuit dicti Reginaldi Thoue quondam aui mei ibid., fo. 51 b (Holkham). - Thoue filii W. Tutebaud (gen.) c. 1215-1220 Wals, fo. 7la(i) (Syderstone). -mesuagiumquodRogerusfilius Touy tenuit 1219 FF, no. 20 (South Walsham). -de tenemento Willelmifilii Tovi, tenementum quod Willelmus filius Toui de me tenuit in uilla de B(er)em(er)e (charter of Martin of Barmer) c. 1220 CAcre, fo. 53b(vi) (Barmer). - Bode Toue 1222 ElyA, fo. 206b (East Dereham). -pro anima Petri auunculi meifilii Lescelinefilie Thoue de Warham (quitclaim by Thomas son ofHugh ofWarham) c. 1250 Wals, fo. 58a(i) (Warham).

ON Toft, ODan Tövi, OSwed Töve is generally regarded as a short form of such names in Por- as Po,fastr, Po,finnr, Porvarör and Porviör, though Ekwall has suggested that it may also be a short form of P6lfr < P6r6lfr and Hald includes the possibility that it could also be a short form of Pörbiorn (SRSö, p. 249; DgP, col. 1476; NK VII, p. 209; Modeer, SvP, pp. 39, 40; Otterbjörk, Svenskaförnamn, p. 132; Ekwall, EtymNotes, pp. 91-92; Hald, Personnavne i Danmark: I. Oldtiden, pp. 94-95). In Denmark Tövi and its side-form Tuvi are old and frequent, being known from runic inscriptions, medieval records and place-names, though in some

4

5

Cancelled. Interlined.

369

Tofi, Tövi, Töve place-names the fem. Töva might also have to be reckoned with (DR, nos. 40, 48, 69, 82, 108, 125, 165, 228, 316, 362, 368, 391, 392; DgP, cols. 14751476, 1495-1504, 1696, s. nn. Tovi, Tuvi; DS II, pp. 79, 94--95, 103; DS IV, p. 192; DS V, p. 365; DS VI, pp. 233, 420, 528; DS VII, p. 262; DS IX, p. 129; DS XI, p. 111; DS XII, p. 79; DS XIV, pp. 175, 295-296). 6 In OSwed Tövi is known from runic inscriptions, though it does not appear to have been especially common in Runic Swedish, and Töve, Tuve is not infrequent in later Swedish records (SRÖg, nos. 103, 177; SRSö, nos. 27, 283, 356; SRSmå, no. 64; SRVg, no. 257; SRU, no. 398; Lundgren-Brate, pp. 282283, 285, s. nn. Tove, Tuve; NK VII, pp. 248, 254; Modeer, Personnamn i Ktb, pp. 24, 66, 87; Modeer, Östsmåländska bondenamn från 1500-talet, p. 66; Thors, p. 95; Fredriksson, Västgötska dopnamn under 500 år, p. 182; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 147, s. nn. Tove, Tue). The relative infrequency of Tövi in Swedish runic inscriptions suggests that in OSwed the name was a Danish loan which had only begun to achieve a certain measure of popularity in Sweden in the latter part of the Viking period. In OWScand, where it is also a loan from Danish, T6fi occurs as the name of a Norwegian in Brennu-Njals saga and is quite well attested in Norway from the 10th century onwards but is extremely uncommon in Iceland (Lind, cols. 1033-1035; LindS, col. 793; NiyR, no. 126). In England ON T6fi, ODan Tövi etc. is well attested in independent use and forms the first element of several place-names: NPN, pp. 141-142; ZEN, p. 83; PNDB, pp. 384-385; Tengvik, pp. 200--201; SPNT, p. 145; ELPN, pp. 81-82; DBS, p. 352, s.n. Tovee; 7 SPNLY, pp. 285-286; von Feilitzen, SCBI, xi.i, p. 13; DEPN, pp. 478,479, s. nn. Toton, Towthorpe, Towton; Ekwall, EtymNotes, pp. 91-92; PN NRY, p. 13; PN ERY, p. 134; PN Nt, p. 152; PN WRY, iv, pp. 72-73, 182; SSNY, pp. 70, 130; SSNEM, pp. 120, 196. Note also the following English examples: mihi fideli comiti Tofig, dat., 1048 (m. 13th) ASChBurton, no. 38 (grant by Edward the Confessor of two hides at Berghe to Tofig comes); 8 ret Touie, innan Touies bure c. 1100 Exeter Book, fo. 6a (D); Toui c. 1100 Exeter Book, fo. 7a (member of a gild at Broad Clyst D); Toui e. 12th LYD, fos. 42a, 43b; Toui surdus 1114-1118 (15th) Burton Survey, p. 219, Toui c. 1126 (15th) Burton Survey, p. 218, Inlanda ad opus quod

6 For Danish place-names in which it has been suggested that the fem. Töva is an alternative to the masc. Tövi see DS II, pp. 79, 94--95; DS IV, p. 420; DS IX, p. 129. Cf. also DS IX, p. 89, where it is suggested that Tovgaard in Viborg Amt (Towgardt 1524, Thoufgaard 1562, Thougaard 1578) could contain the masc. Tövi or the fem. Töva, but that it is also possible that we are concemed here with the number to 'two' or the appellative To 'wool'. 7 DBS, p. 352, s. n. Tovee, wrongly regards T6fi as a diminutive of ON Pi6ovaldr. The statement of DBS (loc. cit.) that the name was brought to England by Cnut's follower Tovi the Proud is also without foundation, there being no reason why the name should not have already been brought to England by the Danish settlers of the late 9th century. 8 Fora discussion ofthe identity ofTofig comes see ASChBurton, p. 75. Berghe is unidentified; the suggestion of Hart, ECNE, pp. 251-252, that it is probably to be identified with Bergh Apton Nf is not supported by any real evidence and is best treated with scepticism.

370

Toki, Töki, Töke prius tenuit Toui pro. xij. denariis ibid., p. 219 (Streeton near Burton upon Trent St); Toui cocus (witn.) 1121-1148 (c. 1175-1200) Bury, no. 116 (grant of Westriht between Saxbarn Sf andRisby Sf), terrarnquam Toui cocustenuitin Bertona 1148-1156 (l. 13th) Bury, no. 142, terram quam Toui cocus tenuit in Bertune 1156--1160 Bury, no. 144 (Great Barton Sf); illud quod Toui dedit in uita sua, per illud seruitium quod Toui mihi reddebat et antecessoribus meis, per concessum omnium heredum Toui, et concessu Agnetis uxoris eiusdem Toui. et filii eius quem habuit Toui de illa Agnete (charterof Adeliz wife of Gilbert fitz Richard of Clare) c. 1136--1138 NthCh, no. 18 (Lowick Nth and Raunds Nth); Robertus filius Toui de Cliua 1167-1168 P, p. 53 (King's Cliffe Nth), Toui ibid., pp. 147, 148 (two different tenants in Congresbury So), Toui doc 9 ibid., p. 196 (Sx); Toui 1174-1175 P, p. 129 (the tithing ofTeddington Wo); de misericordia Ricardifratris Toui 1176--1177 P, p. 185 (C or Hu); de Toui 1179-1180 P, p. 71 (Y); de Ricardo filio Robertifilii Toui 1181-1182 P, p. 127 (O); Ricardusfilius Toui 1181-1182 P, p. 53, 1182-1183 P, p. 64, 1183-1184P, p. 16, 1184-1185 P, p. 83 (L); Nigellusfilius Tovi 1190--1191 (c. 1270) BuryS, p. 62, DeNelfilio Tovi ibid., p. 63 (Babergh DoubleHundred Sf); terram que fuit Touifabri iuxta cimeterium de Wig' 1196--1203 (c. 1240) WorcesterCP, p. 453 (Worcester); Rogerus filius Touy 1208 CRR 49 (Pleas, iv, no. 3520) (L); per Touy de Stanton' 1208 CRR 49 (Pleas, iv, no. 3666) 10 (C; Stanton' is probably Long Stanton All Saints and St Michael C); Ricardusfilius Thove 1212 CRR 56 (Cur, vi, p. 372) (Beeston Bd); Tovi 1212 CRR 56 (Cur, vi, p. 395) (Laughem Wo).

Stenton, TRHS 4th Series, xxv (1943), p. 12, n. 1, remarked that there is nothing, for example, in Touetun, the DB spelling of Towton WRY, to show whether the first element is the masc. T6fi or the fem. T6fa. Ekwall, Smith, and Fellows Jensen, however, have all taken Towton WRY to contain the masc. T6fi and have not included the possibility that it might contain the fem. T6fa (DEPN, pp. 478, 479, s. nn. Toton, Towton; PN WRY, iv, pp. 72-73; SPNLY, p. 286; SSNY, p. 130). The rarity of the fem. T6fa in independent use in English sources and the relative frequency of the masc. T6fi, however, speak in favour of the latter as the etymon of the first element of Towton etc. (cf. also above, s.n. Asi). ON T6fi etc. has also been noted in independent use in 11th century Normandy (Adigard, pp. 148, 318).

ON Toki, ODan Töki, OSwed Töke a) (i) Toketorp 1086 DB, fos. 145b, 202b, Tokestorp, Tochestorp ibid., fo. 166 (lost, in Forehoe H). -Tocheswda 1146--1148 HarlCh 47 H 45 (printed, SSMEA, Appendix I, no. 4), 1146--1148 Coxford Il, fo. 2b(i), Tokeswda 1146--1148 Coxford Il, fo. 2b, Tokeswde c. 1180--1200 Coxfordl/, fos. 3b(i), 4a 11 (f. n. inEastRudharn). -Tochesgate

This word is underlined and omitted on the CR. This form occurs in a text which is marked for cancellation. 11 The Coxford Il fo. 4a entry carmot be later than the death of Earl Hamelin de Warenne in 1202. 9

10

371

Toki, Töki, Töke c. 1150 Crawf, no. 16, Tokesgate c. 1150 CAcre, fö. 67b(i) (road, prob. in Mundesley). -Tokescroft 1193Der, fö. 114a(ii), m. -1. 13thDer, fö. 117a(ii) (f. n. inEastWalton). -Thokeshull', Tokeshull' e. 13thDer, fö. 29a(ii) (f. n. in Wereham). -Thoketof c. 1215 Wafs, fö. 82b(iii) Gunthorpe). - salinam de Tokeshill 1222 EfyA, fö. 162b, Tokeshif 1251 Efy (OKS) (Sandred, OUÅ, 1972, p. 45) (f. n. in Terrington St Clement, now Tuxhill field (6")). - Tokescroft c. 1225 Wafs, fö. 6la(ii) (f. n. in Hindringham). - viam quem vocatur Tokesgate c. 1240 Bromh, fö. 49b(ii) (road in Edingthorpe ). - Thokesgap c. 1240 Hickfing, fö. 7a(i), Tokesgap ibid., fö. 3 la(ii) (f. n. in Waxham), Tokesgap ibid., fö. 33b(i) (f. n., ? between Bastwick and Burgh St Margaret). - Thokesthon' c. 1270 Bromh, fö. 57b(i) (f. n. in Witton). - Thokesheuedfond 1279 NID (f. n. in Ingoldisthorpe) (OKS). -Tokemill' e. -m. 14thBinham, fö. 98b (f. n. in Edgefield). b) (i) Toka. I. fiber homo Herofdi. commendatus TRE DB, fö. 186b (Broome). -/. fiber homo Stigandi Toka TRE DB, fö. 250 (Hales). -Tocho. I. fiber homo TRE DB, fö. 258 (Holkham). -Toka. fiber homo TRE DB, fö. 257b (Testerton). -Toche fiber homo TRE DB, fos. 160, 160b, 163 (East Walton, Castle Acre, Heacham), Toke fiber homo ibid., fös. 165, 165b (Gressenhall, Rougham and Fransham). - Todue TRE DB, fö. 161b (Harpley), Toka ibid., fös. 168 (bis), 168b (bis), 169 (bis), 169b, 170 (Sculthorpe, Barsham, Fulmodeston, Croxton (Gallow H), Rudham, Tattersett, Briston), Tocha ibid., fo. 169 (Burnham Thorpe). -Toka ingrenehoga TRE DB, fö. 122 (Stiffkey; grenehoga is North Greenhoe H). -Toke TRE DB, fö. 145b (held at Toketorp, sub Stigando). -Toca TRE DB, fo. 157b (Taverham). - Tocho de Wintretune TRE DB, fö. 179b (lngworth; Wintretune is Winterton). -Toka TRE DB, fö. 242b (held atEsnuterfea, now Blakeney, sub Herofdo). - Goduini. Tokesone, gen., TRE DB, fö. 246b (Godwine Tokesone 12 and another Godwine held eleven liberi homines at W oodton under King Edward and Earl Gyrth). -Toche de Withuna (witn.) 1108 Binham, fö. 20b (printed, SSMEA, Appendix I, no. 35) (Binham and Langham; Withuna is Wighton). - Tochius presbiter (witn.) c. 1130 CAcre, fö. 103b(i) (grant by Alanus de Swofham (Swaffham Nf) to the monks of Acre). - quedam Alditha Norwici manens Toke cerarii uxor 1150 St William, p. 147, quedam Afditha Thoche cerarii ofim uxor? 1153 St William, p. 217, quedam Botifda Norwici uxor Toche pistoris 1150 St William, p. 170 (Norwich). - terram Toche de Briges quam ipse Toche dedit c. 1150 Crawf, no. 16, CAcre, fo. 67b(i) (confirmation by Bartholomew of Glanville of his father William' s grants to Brornholm Priory; Briges is unidentified). -Adefwafd Jifio Estrifde et Toche fratre eius, abl. (witn.) c. 1160 CAcre, fö. 94b(i), Adefwafdo filio Estrifde et Toche fratre eius, abl. (witn.) c. 1160 (15th) CAcre, fö. lOlb(ii) (the wood of Dicheuude, Dycheuude, which pertained to Middleton). -Hugo Jifius Toke 1161-1177 (13th) Rams, iii, p. 267, Toke ibid., p. 269 (Ringstead and Holme next the Sea), Toke filius Botifd:E ibid., p. 290 (Walsoken). -Rogerus Jifius Toch' 1165-1166 P, p. 26, 1166-1167 P, 13 p. 26 (Lynn). -Taco capellanus 11681170 RBE, ii, p. cclxxi, no. 18 (Nf). -Thoche (acc., nom.) c. 1170 CAcre, fö. 20a(iv) (Massingham), Ordmarusfifius Toche de ParuaMassingamia c. 1180--1190 CAcre, fö.

12 Godwine Tokesone is identical with the Sf tenant Godwine son of Tuka, whose name is rendered as: Goduin filio Tuka, dat., TRE DB, fo. 333b (Godwine son of Tuka held the commendation of two liberi homines, Langabeinus and Goda, at Stiven St); Godwinus filius Tuka TRE, 1086 DB, fo. 335b (Shadingfield Sf; Godwine son of Tuka held here in the commendation of Earl Gyrth TRE). PNDB, p. 386, took the form Tuka to represent the variant ODan Tuki, OSwed Tuke, but it is perhaps better interpreted as merely reflecting the influence of AScand *Tukk(a) etc. 13 Tocce CR.

372

Toki, Töki, Töke 20a(v) (Little Massingham). - Radulfus filius Toke 14 1174-1175 P, p. 120, de Radulfo filio Toke 1175-1176 P, p. 67 (Nf ar Sf). -de Toke de Welles 1176--1177 P, p. 133 (Nf; Welles is Upwell or Outwell ar Wells next the Sea). -Toka de Langehala 1176--1177 P, p. 133, Toche 15 de Langehal' 1177-1178 P, p. 24, Toke de Langenhal' 1178-1179 p, p. 5, 16 Toke de Langehala 1179-1180 P, p. 18, de Toke de Langehala 1180---1181 P, p. 83 (Langhale). - de Normanno filio Toki 1179-1180 P, p. 21 (Nf). - Tokae Godala 1180---1181 P, p. 90, 1181-1182 P, p. 71 (Nf ar Sf). - Willelmo filio Thoki, abl. (witn.) c. 1185 CAcre, fo. 9lb(ii), Willelmofilio Toke, abl. (witn.) ibid., fo. 91b(iii) (Narfard). -Tokefilio Edelin, abl. (witn.) c. 1190Lewes, fo. 233b(i) (Heacham). -Toqus c. 1190--1200 Binham, fo. 5b, Johannes filius Toke ibid., fo. 6a (printed, SSMEA, Appendix Il, na. 3, pp. 269, 17 270) (Binham). - tatam terram quam Toco presbiter tenuit c. 1190--1200 Wymondham, fo. 49b(i) (Sauth Waattan), terram que fuit Thoke presbiteri c. 1210 Wymondham, fo. 54a(iii) (Narth Waattan). -Toke f IEdrici 1191 P, p. 45, 1192 P, p. 189, 1194 P, p. 55 (Nf ar Sf, prob. Nf). -terre que fuit Gilebertifilii Thoke in Derham 1193-1205 Der, fo. 12a(iii) (West Dereham). - terram Astinifilii Thoke c. 1195-1200 Der, fo. 13a(iii), terram Astini filii Toke c. 1200---1205 Der, fo. 23a(i), terram que fuit Astini filii Thoke c. 1205-1210 Der, fo. 23a(ii) (West Dereham). - Rogerum filium Toch' 1198 FF (PRS NS, xxvii, na. 40) (Sparham). -Galfridum Toke 1198 FF (PRS NS, xxvii, na. 53) (Hales). - Laurentium filium Toke 1198 FF (PRS NS, xxvii, na. 130) (Narth Lynn) ibid. (PRS NS, xxvii, na. 135) (Narth Lynn and Wadhoge). - Godiue filiam Thoke 1198 FF (PRS NS, xxvii, na. 197) (West Winch). -Radulfusf Toke 1199 P, p. 276, 1200 P, p. 139 (Nf ar Sf, prob. Nf). -Johannes/ Toke 1199 P, p. 277, 1200 P, p. 140 (Nfor Sf, prob. Nf). -Robertus Toke 1199 P, p. 285, 1200 P, p. 144 (Nfor Sf, prob. Nf). -Thoke cum barba (abl.) (witn.)? 1. 12th/e. 13thBromh, fo. 44a (Dilham). terramfilij Toke c. 1200 Wals, fo. 80a(iii) (Stibbard). - Regina/dus filius Te (sic) de Flicham, hamagium Regina/di filii Thoke de Flicham c. 1200 Der, fo. 109a(i) (Hillingtan; Flicham is Flitcham). - Eustachiumfilium Thoke c. 1200---1210 Der, fo. 1 lOa(i), terram Eustachii fil ii Thoke c. 1210 Der, fo. 106b(ii), Eustachius filius Thoke (a former tenant) c. 1215 Der, fo. 109b(iv), c. 1220 Der, fo. ll0a(ii)(Hillingtan). -damum Thoke ? e. 13th Wals, fo. ll 7b(i) (Reepham). - Ricardus filius Toch, Willelmus filius Toch 1201 (13th) Rams, iii, p. 292 (Walsaken). -Ricardumfilium Emelot et Tokefratrem eius, Toke (dat., nam.) 1202 FF (PRS NS, xxvii, na. 284) (Gateley). -Thoke etGodiuam uxorem suam, Thoke (nam., dat.) 1202 FF (PRS NS, xxvii, na. 289) (Harsey). - De Thoke de Torp 1203 AssRall 817 (NthRS, v, na. 780) (Nf), Thoke del Thorp, Thok 1203 AssRall 817 (NthRS, v, na. 831) (Molton', i.e. Maultan St Michael; Torp, Thorp is unidentified). -Ernisiusf Toke, Geruasiusf Toke 1203 AssRall 817 (NthRS, v, na. 802) (the document cancerns Martham). -Toke (acc.) 1203 CRR 32 (Cur, iii, p. 82) (the dacument cancerns Hunstantan). - Thoke filium 18 Gunselini de H., predictus Thoke filius Gunselin', mesuagium predicti Thoke c. 1205 Der, fo. 109a(ii), Thoke de Hillinget' 1206 CRR 43A (Cur, iv, p. 231) (Hillingtan). -Toke Wictlok' 1208 CRR 67 (Cur, v, p. 131) 19 (surety in a Grand Assize cancerning half a knight's fee with appurtenances in

14 d'Wotton' is interlined on the CR after Toke. Wotton' is unidentified, though it might possibly reflect North or South Wootton Nf. 15 Thoke, corrected from Thom' CR. 16 Langehal' CR. 17 Here the Toqus ofthe MS. is incorrectly rendered as Togus. 18 MS.filii. 19 The byname Wictlok' in this form is taken by Selten PN, p. 169, to have the OE personal name Wihtlac as base.

373

Toki, Töki, Töke Topcroft). - Radulfum filium Toke et Gunildam uxorem eius 1209 CRR 49 (Pleas, iv, no. 3744), Rogerum (sic for Radulfum) filium Toke ibid. (Pleas, iv, no. 3745) (Nf). Thoke filium Henrici 1209 CRR 49 (Pleas, iv, no. 3885) (Nf). - Nicholaum filium Thochi 1209 AssRoll 558 (Pleas, iv, no. 4142) (Moulton St Michael). -terrarn Toke de Wrotton' c. 1210 Der, fo. 38b(i), terram Thoke de Wretton ibid., fo. 38b(iv) (Wretton). -de crofta que fuit Toke, crofta que fuit Toke de Hindringham c. 1215 Wals, fo. 61b(iv) (Hindringharn). - Haraldus filius Toke de Titleshale c. 1215 Coxford Il, fo. 77a(ii) (Tittleshall). - De hornagio Rogeri filii Elferi nepotis Toke, hornagiurn Rogeri filii Elferi nepotis Thoke de Wendlinge c. 1215-1220 CAcre, fo. 102b(ii), Et in vna rod' prati quarn Elfer nepos Toke dedit 1245 (1. 13th) CAcre, fo. 39a(ii) (Wendling). - unam rodarn et dirnid' que iacent iuxta terrarn Thoke clerici et abuttat super uiam de Massingharn c. 1220 Der, fo. 11 la(ii) (Hillington). - praturn quod Toke filius Eluredi dedit dornui de Castelacra c. 1225-1235 CAcre, fo. 103a(i) (Wendling). - Sampsonfilius Toke 1222 ElyA, fo. 206b (West Dereharn). -terrarn Thome filii Toke c. 1225 Der, fo. 25a (West Dereharn). - terrarn que fuit Rogeri fabri fratris Thoke fabri 1226 JMN, p. 223 (Nf). Radulfo filio Toke, abl. (witn.) 1228 CAcre, fo. 106b(vii) (Tuddenham). -Radulfo filio Toche, abl. (witn.)? c. 1230---1240 Bromh, fo. 52b(i) (charter of Hersindis, daughter of Bartholornew of Palling). -terrarn Toke Fike c. 1235 Der, fo. 20b(iii) (West Dereharn). - terram Willelmi filii Toke c. 1240 Wals, fo. 58a(ii, iii) (Warharn). - terrarn Willelmi Tokern. l3thLangley, fo. 24a (Sisland). -terrarnEluiue Thoke rn. l3thLangley, fo. 41a, terrarn que fuitRogeri Thoke ibid., fo. 41b (Raveningharn).

The etymology of ON T6ki etc. has been the subject of much discussion. Attempts have been made to link it with the ON byname t6ki, which occurs once in the Lj6svetninga saga and which is doubtless identical with ModNorw toke 'fool' and probably also with the semantically obscure ODan substantive tokce. The name has also been linked with an ODan fätök(cer), a variant of ODanfäti)kcer 'poor'. Another explanation is that T6ki is either a hypocoristic form of Porkell, formed with the i-suffix, or a short form of some other name in Por-, formed with the -ki suffix. For a full discussion with bibliographical references see Hald, Personnavne i Danmark: I. Oldtiden, pp. 89-93, on which the above account is based. For further discussion see the following: Noreen, Vårt språk, iii, p. 90 n; Wimmer, DRM, iv.ii, p. lxv; Wessen, Nordiska namnstudier, p. 90, n. 2; Brondum-Nielsen,§ 83; DgP, col. 1462; NK VII, p. 209; Hjorth Pedersen, Bebygg. på -by, p. 38; Otterbjörk, Svenska förnamn, p. 132, s.n. Tyge, Tyke; Modeer, SvP, pp. 39, 40; SPNLY, p. 288. T6ki is a characteristically Danish name, and in Sweden and Norway it is a loan from ODan (Wessen, Nordiska namnstudier, p. 97). ODan Töki and its side-form Tuki are frequent both in independent use and as the first element of place-names (DR, nos. 24, 46, 53, 56, 58, 87, 91, 116, 121, 123, 129, 143, 145, 179, 212, 219, 234, 240, 241, 258, 266, 278, 295, 296, 297, 315, 324,343,412; DgP, cols. 1458-1463, 1480-1489; Hald, Vore Stednavne, p. 108; Hjorth Pedersen, Bebygg. på -by, pp. 37-38; DS Il, pp. 69, 85, 97; DS IV, p. 259; DS V, pp. 248,501; DS VII, pp. 179,386; DS VIII, p. 46; DS IX, pp. 121, 169, 181; DS XI, pp. 89, 111, 200; DS XIV, p. 83; DS XVI, pp. 28, 92, 117, 219-220; SOH Il, p. 301; SkOA XVI, pp. 99-100). 374

Toki, Töki, Töke Kousgård Sprensen has suggested that an ODan *Tykir, a side-form of Tyki, the late medieval palatal umlaut form of Tuki, forms the first element of Tpjstrup, Gudme Herred, Fyn, and also of same of the other Danish place-names of this type (Tygestrup, Tpjstrup, Tygstrup), which were taken by Gunnar Knudsen, DS IX, p. 169, and DgP, col. 1489, to contain Tyki with the secondary gen. in -s- (DS XIII, p. 151). The absence of ODan Töki, Tuki among the personal names forming the first elements of place-names in -lej and -stath suggests that the name originated in the course of the Viking period, when these elements had ceased to be used in place-name formation. In OSwed Töki is known from runic inscriptions and Töke, Tuke and Tyke are well represented in later records, the name being also found as the first element of several place-names (SRÖg, nos. 70, 104, 189, 209; SRSö, nos. 49, 145,252; SRSmå, nos. 2, 10; SRVg, nos. 104, 114, 197; SRU, nos. 14,201, 586, 755; Lundgren-Brate, pp. 281, 283-284, 285-286; Modeer, Personnamn i Ktb, p. 24; Modeer, Östsmåländska bondenamn från 1500-talet, p. 66; Thors, p. 94; Fredriksson, Västgötska dopnamn under 500 år, pp. 182 and n. 14, 183; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 144, 148; Hjorth Pedersen, Bebygg. på -by, p. 38). ON T6ki is doubtless a loan from Danish (see above). Though an isolated 11th century Norwegian example is on record, ON T6ki is only continually attested from the earlier part of the 14th century onwards, and is then not particularly common in Norway, and in Iceland it is only known as the first element of a place-name in records of 1479 and 1523 (Lind, cols. 1035-1037; LindS, col. 793). In England ON T6ki etc. is well attested, both in independent use and as the first element of place-names and field names: NPN, pp. 142-143; ZEN, pp. 83-84; PNDB, pp. 385-386; Tengvik, pp. 164, 200; SPNT, p. 145; ELPN, p. 82; DBS, pp. 351-352, s. nn. Took, Tookey; SPNLY, pp. 287288, 356; Smart, Moneyers, pp. 248, 254, 256; von Feilitzen, SCBI, xi.i, p. 13; Okasha, no. 111; Smart, Cnut's York Moneyers, p. 228; von Feilitzen, PNWinton, p. 174; DEPN, pp. 476,479,482, s. nn. Tockholes, 20 Toxteth, Tugby; PN La, p. 115; PN Bk, p. 261; PN BdHu, p. 300; PN NRY, p. 333; PN Nth, p. 292; PN Wa, p. xxii; PN Hrt, pp. xx, 257; PN Nt, pp. 277,284,291; PN C, pp. xxi, 221; PN WRY, i, p. 106; PN WRY, iv, pp. 53, 250; 21 PN WRY, v, p. 175; PN Gl, iv, p. 209; SSNY, p. 106; SSNEM, p. 75; lnsley, JEPN, x (1977-1978), pp. 49, 55, 62, 69.

20 Ekwall, DEPN, p. 476, s. n., considered that the first element ofTockholes La could be either the Scand T6ki or OE Toc(c)a. Earlier, however, he did not include the possibility of the Scand name and he took the etymon to be OE Toc(c)a (PN La, p. 75). 21 Tockwith WRY is taken by PN WRY, iv, p. 250, and SSNY, p. 106, to contain the Scand T6ki. In DEPN, Ekwall proposed OE Toc(c)a as the first element of this place-name, but he latter suggested that it may be either OE Tocca or ODan Töki (DEPN, p. 476, s. n. Tockwith; OE wic, p. 59).

375

Toki, Töki, Töke In Normandy ON T6ki etc. has been noted as the first element of several place-names (Adigard, pp. 149-150, 419-420). Initial Th- for /t/ in forms in Thoke, Thoche is the result of AN orthographical interchange between th and t for etymological [t] (ANI, pp. 47-49; IPN, p. 109, n. 1; PNDB, § 96; SPNLY, § 102). Forms in Tochi, Tocho, Tochre, T(h)oche, Toch' and Toch contain -ch- for Scand /k/, this being the normal AN spelling for /k/ before front vowels (PNDB, § 113; SPNLY, § 126; von Feilitzen, PNWinton, p. 225). For the use of -k- for /k/ as in Toke etc. in OE and ME orthography see Btilbring, § 471, Anm. 2; Jordan, § 178.2; PNDB, § 118; Campbell, § 427, n. 1; SPNLY, § 125. Forms in Toki(g) are known from OE records, cf. the following: Tokiges, gen., Tokig 1020--1038 (11th) ASChR, no. 80, Tokig 1038-1050 (11th) ASChR, no. 90 (the documents concem landat Halton Bk; Tokig's wife, who was already dead when the transaction recorded in ASChR, no. 80, took place, bore the OE name /Eöelf/,;ed); Care Tokies suna ( /kl, cf. the patronymic Toqisone found in the Exon Domesday (PNDB, § 118; SPNLY, § 128.iv). Final -e in Toke, T(h)oche is the result of Anglicization (PNDB, §§ 45, 150; SPNLY, § 147; cf. also von Feilitzen, PNWinton, p. 225). The DB form Toca is also an Anglicized form. It occurs also among the names of pre-Conquest moneyers (Smart, Moneyers, pp. 248, 254, 256; Smart, Cnut's York Moneyers, p. 228; von Feilitzen, SCBI, xi.i, p. 13). Forms in Toc(h)o have been Latinized through the replacement of final -i by the nom. ending -o of the Latin third declension (PNDB, §§ 154, 155; SPNLY, § 149).23 Cf. the medieval Danish Latinized form Toko, -onis (DgP, col. 1458). The form Tochius c. 1130 CAcre has been Latinized by the addition ofthe nom. ending of the Latin second declension (cf. PNDB, § 155). Though loss of final -i may explain forms in Toch, Toch' and Toqus (cf. PNDB, § 49; SPNLY, § 150), it may also be possible that they do not stand for ON T6ki etc., but reflect an unrecorded ODan *Tök (? byname), perhaps corresponding to ModSwed tok 'fool, simpleton', a name which has been proposed as the first element of several Danish place-names (see DS IX, p. 40; DS XII, pp. 22, 125; SPNLY, p. 288; Hald, Personnavne i Danmark: I. Oldtiden, pp. 90, 91).

22 This Tokig, who was killed saving the life of the Conqueror at the battle of Gerberoi in 1079, has been presumed to be the son of the famous DB magnate Wigot of Wallingford (Douglas, William the Conqueror, p. 239 and n. 1). 23 Further English examples of this are: Toco TRE DB, fo. 35 (East Molesey Sr); Tocho Loccher c. 1175 (1. 12th) Newark, p. 2, Thoco pistor ibid., p. 4 (Newark-on-Trent Nt).

376

Toli, Töli, TöliR, Tiile

ON Toli, ODan Töli, OSwed (run.) TöliR, OSwed Tfile a) (i) Tolisnap c. 1200 CAcre, fo. 60a(ii, iii, iv) (f. n. in Gateley). - Tolescroft c. 1220 Wals, fo. 66a(i) (f. n. in Thursford). -crucem Toli m. 13th Wymondham, fo. 94a, 94a(i), Tolyescros 1377 OKS, Tolyscros 1380 OKS (f. n. in Wymondham). -Tolyesplace 1464 NJD (f. n. Hingham) (OKS). b) (i) Toli uicecomes TRE DB, fo. 21 lb (Broome, where Toli the sheriff gave one carucate ofland to the Abbey ofBury St Edmunds TRE), Tohli uicecomitis, gen., ibid., fos. 140 (one fiber homo ofToli the sheriffheld in Mundham), 26424 (one fiber homo in the commendation of Toli the sheriff held one carucate of land at Langhale TRE). 25 - Toli. I. fiber homo TRE DB, fo. 230 (Barton Bendish), Toli. fiber homo (bis) ibid., fo. 230b (Wereham, Upwell), Toli ibid., fo. 232 (Cleythorpe). - Guida filius Toli 1161-1177 (13th) Rams, iii, p. 268 (Ringstead and Holme next the Sea), Galfridus filius Toli de Helm ibid., p. 291 (Walsoken; Helm is probably Elm C, about two miles south of Walsoken). - Willelmus filius Tholi I. 12th - e. 13th Wals, fo. 83a(i), Willelmus Toly ibid., fo. 83a(ii) (Gunthorpe). - carta Haconisfilii Toly e. 13th (m. 14th) Rams, i, p. 99, no. 399 (Redlingesmiddel in Outwell), Acunfilius Toli 1201 (13th) Rams, iii, p. 297 (Outwell), De Hacone filio Toly Preste 1206-1207 (m. 14th) Rams, ii, p. 319 (Upwell and Welney), mesuagium quod fuit Hacun Toly de Wellis c. 1225-1230 Der, fo. 80b(i) (Upwell). -Toli Sacerdos et Johannes frater ejus 1201 (13th) Rams, iii, p. 298 (Outwell), W alterus filius Toli ibid. (Chisebec in Outwell). - Ricardum Toli, Ricardus Toli, Ricardo Toli, dat., 1202 FF (PRS NS, xxvii, no. 301) (Wicklewood). -Tholy Wace (acc.), predictus Tholy, predicto Tholy, dat., 1202 FF (PRS NS, xxvii, no. 363) (Colescroft in Straton'; Straton' is either Stratton Strawless or belongs to the Long Stratton/Stratton St Mary and St Michael complex). - Radulfum filium Toly (capellanum )26 1202 CRR 28 (Pleas, ii, no. 1040) (? Gasthorpe ), plegii Radulfi Capellani filii Tholy 1203 CRR 29 (Cur, ii, p. 170) (Nf). - de Radulfo filio Tholy et de heredibus suis c. 1205 Bromh, fo. 47b(i) (rents in Saxlingeham; Saxlingeham is prob. Saxlingham, Holt H). - terram Baconis filii Toli c. 1205 Der. fo. 90a (Upwell). - Gaufridum filium Tholi 1206 CRR 43A (Pleas, iii, no. 2439) (Nf). -Petrum Toli (bis) 1206 CRR 43A (Pleas, iii, no. 2435) (Nf), Petra filio Tholi, dat., 1207 CRR 46 (Cur, v, p. 38), corpus Petri filii Tholi 1208 CRR 48 (Cur, v, p. 165) (Nf), Petrusf Toli 1209 P, p 50 (Nfor Sf, prob. Nf). -projusticiando Robertof Toly 1219 P, p. 45 (Nf). - Walterusfilius Toli 1222 ElyA, fo. 158b, Walterus Toliibid., fo. 159a (Outwell). -Robertusfilius Toly 1222 ElyA, fo. 171a (West Walton). - Ricardus filius Radulfi Tholy 21 de Dodelinton' 1224 CRR 85 (Cur, xi, p. 478) (Didlington). -terram Willelmi Toli c. 1225 Der, fo. 132b, c. 1240 Der, fo. 130a(ii) (Tilney). -terram Tolyfilii Maynard' 1229-1237 Holme, fo. 88b(i) (Potter Heigham). -a Thome filio Toly, Saluis heredibus predicti Thomefilii Toly 1229-1237 Holme, fo. 90b(i), Carta Thome filii Toly de Hecham, Thomasfilius Toly de Hecham ibid., fo. 90b(ii), de Thome filio Toly ibid., fo. 90b(iii) (Potter Heigham). -de dono Nicholaifilii Toly de Winterton' 1234 Bromh, fo. 3b (Winterton). - De Willelmo Toly, De Henrico Toly, De Toly Screm c. 1245 Wymondham. fo. 106a (these forms occur in a list of rents due to Wymondham

Here the -h- of Tohli is added above the line. For Toli, Edward the Confessor's sheriff ofNf and Sf, see Morris, pp. 23, n. 47, 43 and n. 21, and ASWrits, p. 575. For the DB entries relating to Toli the sheriff see PNDB, p. 386. 26 Interlined. 21 Toly CRR 87. 24

25

377

Toli, Töli, TöliR, Tiile Priory). -mesuagium Roberti Tholi m. 13th CAcre, fo. 12a(ii) (Castle Acre). - Willelmo Toly, abl. (witn.) c. 1260 BodlNJCh, no. 294 (Norwich).

T6li is a short form of such names as Porlakr, Porleikr, Porleifr, Porlaugr (Br0ndum-Nielsen, § 244, Anm. 3; NK VII, p. 209 Modeer, SvP, p. 39; SPNLY, p. 289). The name is characteristically EScand and first appears in Norway in the 13th century (Lind, col. 1037). ODan Töli and its side-form Tuli are well attested in Danish sources, and the name has also been taken to form the first element of several Danish place-names (DR, nos. 94, 106, 386, M0. 93; DgP, cols. 1463-1465, 14891492, 1696; DS VII, pp. 224,373; DS XI, p. 147; DS XII, p. 131; DS XIII, pp. 13,225; Kousgård S0rensen, APhS, xxiii (1957), pp. 110-111). A side-form *Tolli, *Tulli, showing shortening of the stem vowel and compensatory lengthening of -l-, is known from Danish place-names (DS II, pp. 94, 117; DS III, p. 31; DS IV, p. 555; Kousgård S0rensen, APhS, xxiii (1957), pp. 111112; Hald, Personnavne i Danmark: I. Oldtiden, pp. 93-94). In OSwed Tule is found in medieval records (Lundgren-Brate, p. 284). OSwed Tolle, corresponding to the Danish *Tolli, is known from late medieval and 16th century Swedish records (Lundgren-Brate, p. 281; Modeer, Östsmåländska bondenamn från 1500-talet, p. 66; Thors, pp. 94-95; Modeer, SvP, p. 65; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 144).28 ON T6li, a loan from OEScand, probably from Danish, is not certainly attested in Iceland and is extremely rare in Norway, where it is first attested in a patronymic formation of 1239 (Lind, col. 1037). Cf. also ODan *Tölir, *Tulir, which is known from place-names, and OSwed (run.) TöliR, which is usually held to be a short form of PorleifR, Porleifr (DS IX, pp. 111, 176; DS XII, pp. 67, 128; Kousgård S0rensen, APhS, xxiii (1957), pp. 109-110; Hald, Personnavne i Danmark: I. Oldtiden, p. 93; SRSö, no. 293; SRVg, nos. 35, 162, 169; SRU, nos. 11, 1005).29 In England ODan Töli etc. is quite well attested in independent use, as well as being known from place-names and field names (NPN, pp. 143-144; ZEN, p. 84; PNDB, p. 386; ELPN, p. 82; DBS, p. 352, s.n. Tooley; SPNLY, pp. 289-290; DEPN, pp. 476, 478, s. nn. Tolethorpe, Toseland; PN BdHu, pp. 272-273, 300; PN Wo, pp. 114-115; PN Ess, p. 637; PN ERY, p. 232; PN WRY, v, p. 83; PN Gl, iv, p. 209; SSNY, p. 70; SSNEM, p. 120; Insley, JEPN,

28 OSwed Tolle has been interpreted as a short form of OSwed Thorlever (e. ModSwed Tollev) (Thors, p. 95; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 144). However, Modeer, SvP, p. 41, takes Tolle to be not only a hypocoristic form of OSwed Thorlef (Thorlever) but also of Thorläk and Thorlogh. 29 DgP, col. 1492, groups examples of T6larr (from the Norse Knyt/inga saga) and Tullius s. n. Tuli, but it is extremely doubtful whether these forms betong here (Hald, Personnavne i Danmark: I. Oldtiden, p. 93).

378

Toli, Töli, TöliR, Tii.le x (1977-1978), p. 43). Cf. also the following examples: Toli 1114-1118 (15th) Burton Survey, p. 216 (Branston St); duas acras in quibus Toly filius Manstan manebat 1156-1175 (c. 1175-1200) Bury, no. 195 (Sf, grant by Walter of Thelnetham); 30 Walterus filius Toli de Burchildeberi 1175-1176 P, p. 216, Walterusfilius Toli 1176-1177 P, p. 195, 1177-1178 P, p. 133, 11781179 P, p. 123, 1179-1180 P, p. 39, 1180-1181 P, p. 138, 1181-1182 P, p. 3 (Bucklebury Brk; the entries up to and including 1178-1179 P are grouped under Sr, those subsequent to this under Brk); pro plegio Ailiue et Tole 11841185 P, p. 137, pro plegio Aileue et Tole 1185-1186 P, p. 25 (Bk); 31 Toli 1186-1188 (c. 1270) BuryS, pp. 51, 52 (Hopton Sf); ex libero tenemento Toli et Gunnild' vxoris sue 1196 FF (PRS, xx, no. 4) (Bisham Brk); De terra que fuit Toli c. 1200 Urry, p. 302, De terra que fuit Benedicti filii Toli c. 1206 Urry, p. 332 (Canterbury K); Haldanumfilium Toli, TolifiliumHugonis 1202 CRR 27 (Cur, ii, p. 126) (Rochester K); Nicolaus filius Toly 1205 CRR 39 (Cur, iv, p. 10) (Sf); Anketilfilius Toli 1205 CRR 40 (Pleas, iii, no. 1512),32 Simonemfilium Toli et Astin' fratrem suum 1206 CRR 42 (Pleas, iii, no. 1866) (Lei); Petrum filium Tholy 1206 CRR 43A (Cur, iv, p. 246) (the document concems Kingston Sr); Aliciamfiliam Thole 1210 CRR 52 (Cur, vi, p. 29) (Y); Alardum filium Toly 1211 CRR 54 (Cur, vi, pp. 165, 198), Alardum filium Tholy 1212 CRR 55 (Cur, vi, p. 249) (Hanslope Bk), Alardus filius Tholi ibid. (Cur, vi, p. 271) (Bk); Toly ad fordam 1222 ElyA, fo. 99a, Toly de fordam ibid., fo. 99b (Doddington C; if fordam is not a local name, it may stand for Fordham C, though this is some distance from Doddington). Henricus filius Toly ibid., fo. 109b (Linden End C), Johannes filius Toly, Toly molend' ibid., fo. 152b (Tydd St Giles C), Thomasfilius Toly ibid., fo. 156b (Elm C). ODan (p. ns.) *Tölir, OSwed (run.) TöliR has been noted in England as the first element ofTollesby NRY, which A.H. Smith, PN NRY, p. 163, incorrectly derived from the ON byname Tollr (SPNLY, p. 290; SSNY, p. 40). 33

Bury dates this record to 1156---1180 but the MS. suggests an upperlimit of c. 1175. The fem. Tala is also possible here. 32 This document is marked for cancellation. 33 The ON byname Tollr, which is first attested in a Norwegian record of 1338, belongs to the appellative ON tollr m. 'charge, due, toll' (LindBiN, col. 184). Note also the 12th century Icelander Tolla-P6rir '1>6rir ofthe tolls', whose prefixed byname also belongs here. ON tollr is ultimately of Latin origin, though it must have reached Scandinavia through English or Low German channels (cf. OE toll, OSax tol), and it is, therefore, unlikely that it was used to form bynames as early as the Viking period. 30 31

379

Töpi

ODan Töpi a) (i) Topecroft 1086 DB, fo. 211b, Topecropt ibid., fo. 246, Topecroft 1202 FF (PRS NS, xxvii, no. 384), 1208 FF (PRS NS, xxxii, no. 132) (Topcroft, Loddon H). b) (i) Ran' filius Topi de Becham (witn.) Hy2 D and C Norwich (charter in a case) (Nf; Becham is East Beckham) (FMS). - Topi et filius eius 1169-1170 P, p. 12 (Nf). Godardusfilius Topi (witn.) c. 1170 CAcre, fo. 9la(i) (Gateley, Pentney, Massingham and land held by Rainaldus de Waltun). - terram quam Topi tenet 1. 12th AddCh 22571 (Baconsthorpe) (FMS). -Robertus Topi 1199 P, p. 280, deRoberto Topi 1200 P, p. 138 (Nf or Sf).

In Denmark ODan Töpi is largely confined to Slesvig and is attested from the 13th century onwards (DgP, cols. 1467-1468). The name is a hypocoristic form of Pörbiorn (Hald, Personnavne i Danmark: I. Oldtiden, p. 94). Homby, NK VII, p. 209, also includes the possibility that Töpi isa loan from the south, i.e. from the Low German area. However, no name corresponding in form to ODan Töpi has been noted in OSax, 34 and, despite its southerly distribution in Danish, there is no reason to doubt that it is a regular Danish formation. The side-form Tiipi, a hypocoristic form of Purbiorn, is, significantly, known throughout Denmark (DgP, col. 1494). In England ODan Töpi occurs in DB TRE in D, Ess, Ht, Land Sf, but is otherwise infrequent (NPN, p. 144; PNDB, p. 386; Tengvik, pp. 164, 226; DBS, p. 352, s.n. Toop(e); Whitelock, Foreword to LibEl, p. xi; SPNLY, p. 291). The side-form Tiipi may form the first element of Tupton Db and Tupholme L, though DEPN and SSNEM also include the possibility that the etymologically obscure appellative ME tup 'ram' is an alternative etymon here (DEPN, p. 482, s. nn. Tupholme, Tupton; PN Db, pp. 317-318; SSNEM, pp. 173, 196).

ON Tosti, ODan Tösti, OSwed Töste b) (i) Tostius Wml DB, fo. 200b (in this entry it is recorded that there were three liberi homines at Somerton TRE but after Tostius left England Berardus had them). - Tosteus Sauer (witn.) c. 1115 (1395-1396) SSMEA, Appendix I, no. 57 (the soke ofNecton).

T6sti isa hypocoristic form of Porsteinn (NK VII, pp. 57, 95, 245; Modeer, SvP, pp. 39, 40). T6sti isa characteristically EScand name, being well attested in Denmark

34 The works of Schlaug on OSax personal names and those of Mansion and TavemierVereecken on MDu personal names contain nothing which could have given rise to ODan Töpi.

380

Tryggr, TryggR and Sweden from the Viking period onwards (DR, nos. 68, 78, 108,338,401; DgP, cols. 1469-1472; Hald, Vore Stednavne, pp. 135, 192; Hjorth Pedersen, Bebygg. på -by, pp. 38-39; DS IV, p. 607; DS VII, pp. 41, 284; DS VIII, p. 96; DS IX, pp. 112, 186; DS XI, pp. 106, 149, 210-211, 218; DS XIII, p. 139; DS XVI, pp. 246,260; DS XVIII.I, pp. 134, 148; Sydsl. Stedn. VII.l, p. 41; SkOA XVI, pp. 43-44, 68; SRÖg, nos. 70, 82,165,204,209,210; SRSö, nos. 145, 254; SRSmå, nos. 51, 62, 153; SRVg, nos. 87, 171; SRU, nos. 232,344,506; Lundgren-Brate, pp. 281-282; Modeer, Personnamn i Ktb, pp. 24, 90; Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 147). In OWScand the name is infrequent, though it appears in both Norway and Iceland, and is clearly a loan from OEScand (Lind, col. 1040; LindS, col. 794). In England T6sti is fairly infrequent (NPN, pp. 144-145; ZEN, p. 84; PNDB, p. 387; SPNLY, p. 291; PN Nth, p. 292; PN Cu, p. 266). The most famous English bearer of the name is Tostig (ob. 1066), earl of Northumbria from 1055 to 1065, and brother (and eventual enemy) of King Harold II Godwinesson.35 In Sfthe name is represented by Tostius liber homo TRE DB, fös. 84b, 435b (Bures St). Forms in Tostius, Tosteus have been Latinized by the addition of the nom. ending ofthe Latin second declension (cf. PNDB, § 155). The use of -e- for -i- in the Nf form Tosteus reflects Anglicization (see PNDB, §§ 45, 150; SPNLY, § 147; cf. also von Feilitzen, PNWinton, p. 225).

ON Tryggr, OSwed (run.) TryggR b) (i) Willelmus Trig c. 1235 (m. 14th) Rams, i, p. 417 (Brancaster).

ON Tryggr, OSwed (run.) TryggR is an original byname, 'the faithful one' or 'the safe one, the sure one' (NK VII, p. 53; cf. the adjective ON tryggr 'trusty, faithful, true'). ON Tryggr occurs in a 12th century runic inscription at Maeshowe in the Orkneys and has been noted in Norwegian records from the mid 14th century onwards (Musset, lntroduction a la Runologie, pp. 440-441; Lind, cols. 1041-1042; Linds, col. 794). In OSwed TryggR appears in a runic inscription in Uppland (SRU, no. 952). Note also the weak form ON Tryggvi, ODan Tryggi, OSwed *Trygge, though this can also be a short form of Sigtryggr or Ötryggr (NK VII, pp. 156,

35

For the career ofEarl Tostig see ASWrits, p. 575.

381

*Trygg-Ulfr n. 159,209; Lind, cols. 1042-1043; LindS, col. 794; LindBiN, col. 388; DgP, col. 1478; DS XIII, p. 259; Spndergaard, p. 111; SRÖl, no. 16; LundgrenBrate, p. 283, s. nn. Trighe, *Trygge). In England the strong variant Tryggr occurs as a personal name in the following forms: Trig 1185 Templar (Y); Trigo, abl. (witn.) 1. 12th (c. 1400) YCh (Y); Trig c. 1200-1211 RA (Oseby-in-Haydor L) (DBS, p. 354, s.n. Trigg; von Feilitzen, Notes, p. 61; SPNLY, p. 292). Both DBS, p. 354, s.n. Trigg, and von Feilitzen, Notes, p. 61, regard the personal name Trig as belonging directly to the adjective ON tryggr, ME trigg, but there is no reason why it should not represent a direct loan of the personal name ON Tryggr, OSwed (run.) TryggR. On the other hand, where Trig is used as a byname, as in the above Nf form, it is more likely to reflect ME trigg 'true, faithful' < ON tryggr, rather than the personal name ON Tryggr, OSwed (run.) TryggR, although on förmal grounds the personal name cannot entirely be excluded (cf. NPN, p. 145; ZEN, p. 84; DBS, p. 354, s.n. Trigg; SPNLY, p. 292). It is interesting to note that DgP II, col. 1148, took the late medieval Danish byname Trygge to belong to the adjective tryg rather than to the personal name Tryggi.

AScand *Trygg-Ulfr b) (i) Radulphus Thyrgolf c. 1235 (m. 14th) Rams, i, p. 421 (Brancaster), Leva, qucefuit uxor Radulphi Trigolf 1239 (m. 14th) Rams, i, p. 427, Willelmus Trigulf ibid., p. 426 (court at Brancaster).

A personal name *Tryggulfr is not on record in Scandinavia, and von Feilitzen, Notes, p. 61, took the above Nf forms of 1239 to probably reflect an AScand formation with ON tryggr > ME trigg 'true, faithful' as first element. In view of the absence of compound names in Trygg- in Scandinavia, it is probable that we are concemed here with an AScand *Trygg-Ulfr, in which the adjective ON tryggr is compounded as a prefixed byname with the common personal name ON Ulfr, ODan, OSwed Ulf. This compound would denote 'the trusty, the faithful Ulf' or the like, 36 and it would have been quite possible for it to have been taken into the regular name stock after its semantic significance had been lost and to have then been regarded as a normal dithematic personal name. This would have been facilitated by the fäet that the prefixed byname is a particularly Scand feature unknown to OE, and such a formation coined among Scand settlers in the Danelaw would doubtless have been understood by the English as a normal dithematic personal name without any particular semantic connotation. Thyrgolf in the form Radulphus Thyrgolfhas undergone metathesis. 36

For Scand formations ofthis type see above, s. n. *Fang-Ulfr.

382

Tubbi, Tobbi, Tubbi, Tobbe, Tubbe

ON Tubbi, ODan Tobbi, Tubbi, OSwed Tobbe, Tubbe a) (i) Tubbishirne 1480 OKS (f. n. in Gressenhall). -Tubbesgate 1485 Rental (f. n. in Gimingharn) (OKS). b) (i) Alanum Tubbi 1206 CRR 43A (Cur, iv, p. 244) (Nf).

Tobbi, Tubbi isa hypocoristic form of PorbiQrn (Noreen, AislGramm, § 241, Anm. I; Brpndum-Nielsen, § 244, Anm. 3; NK VII, pp. 95, 209; Hald, Personnavne i Danmark: I. Oldtiden, p. 94). 37 Tobbi, Tubbi is characteristically EScand, being found in Denmark and Sweden from the Viking period onwards (DR, no. 331; DgP, cols. 1479-1480, 1696; Hald, Vore Stednavne, p. 131; DS I, p. 69; DS II, pp. 97-98; DS IV, p. 325; DS VII, p. 225; DS XII, p. 19; DS XIII, p. 225; SRU, nos. 229, 232, 689, 1177; Lundgren-Brate, pp. 280, 283; Thors, p. 94; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 41, 147). In Norway Tubbi first appears in the 14th century and is relatively infrequent (Lind, cols. 1043-1044; LindS, col. 794). It is clearly a loan from the OEScand area. Lind and LindS have no examples of the name from medieval Iceland, but it is interesting to note that Tobbi has arisen independently in Modlcel as a hypocoristic form of Modlcel P6rbjörn (Hald, Personnavne i Danmark: I. Oldtiden, pp. 84-85). In England Tobbi, Tubbi is fairly infrequent (ZEN, p. 84; PNDB, p. 388; DBS, p. 356, s.n. Tubb; von Feilitzen, Notes, p. 60; SPNLY, p. 285).

AScand *Tukk(a) a) (i) Tukesorte 38 1204 FF (PRS NS, xxxii, no. 55) (f. n. in Wereham). -Tukesparroc c. 1220 Der, fo. 216b(i), Tuckesp(arr)oc c. 1225 Der, fo. 15lb(ii), Tuckesparroc ibid., fo. 161b (f. n. in Lynn). b) (i) Tukkefabrum, Wlfmerumfilium Tukke 1101-1107 (14th) Holme, Appendix A (Ashby, West Flegg H). 39 - Tocca fabro, abl. (witn.) c. 1135-1140 CAcre, fo. 16b (Massingharn), Tucche faber, Tuche apud pantern 1138-1147 CAcre, fo. 4a(i) (Nf, prob. Massingharn), de terra Tucche de Ponte, De Tucchefabro, Tucchefaber (witn.) c.

37 Janzen, NK VII, p. 58, considered that Tubbi is most likely a hypocoristic form of PorbiQrn but also included the possibility that it is a hypocoristic form of Porbergr and Porbrandr. 38 This reading is uncertain. 39 Though this document was drawn up in the period 1101-1107, most of the material which it contains relates to the latter part of the 11 th century. It is probable that Tukke the smith of Ashby is included among the six free men annexed to the estate of Oby which Ringulf had formerly possessed and Stanhard held of Roger Bigod in 1086 (Stenton, EHR, xxxvii (1922), p. 228, n. 1). Wulfmerfilius Tukke was probably the son ofTukke the smith of Ashby (Stenton, ibid., p. 228, n. 7). The entry relating to this Wulfmer is probably contemporary with the document itself.

383

*Tukk(a) 1150-1170 CAcre, fo. 16a(i) (Massingham). -Johannesfilius Tuch 1201 (13th) Rams, iii, p. 298 (Outwell). - ad gardun' Roberto40 Tucke, gardinum Roberti Tucke 1202 FF (PRS NS, xxvii, no. 280) (Nf). -pratum Tuck' filii Leuerun c. 1205 Der, fo. 88b(ii) (Upwell). - mesuagium Johannes Tuche de Upwell' c. 1205-1210 CAcre, fo. 78b(ii), mesuagium /ohannis filii Tuche ibid., fo. 78b(iii), terram Helewyse quondam uxoris /ohannis filii Tucke c. 1220 Der, fo. 83a(i, ii) (Upwell). - Radulfo filio Tucke, abl. (witn.) c. 1210-1220 Binham, fo. 146a(ii) (Holkham). - Walterus filius Tucke 1222 ElyA, fo. 159a (Outwell).

AScand *Tukk(a) is a hypocoristic form of Purkil, Purketel < ON Porkell, Porketill etc. (see DBS, p. 356, s.n. Tuck). The weak *Tukka is represented by the forms in Tukke, Tucke, Tucche, Tuche, and Tocca, while the strong variant *Tukk is represented by the forms in Tuck' and Tuch. 41 It is impossible to say whether the Nf field names given above contain the strong or the weak variant. Tocca in the form Toccafabro, abl. (witn.) c. 1135-1140 has -o- for -u-, probably as a result of the influence of Toke, Tochi etc. < ON T6ki etc. or of a side-form of *Tukka, AScand *Tokka, hypocoristic form of Porkil etc. Note also, however, that ODan Tokki, a variant of Töki whose stem vowel has been shortened, which has been noted in medieval Danish records and appears to form the first element of the Danish place-name Tokkerup, which occurs five times on Sjrelland and once on Lolland,42 may also have to be considered as a source of influence here. Cf. the Y forms Ricardus f Tocce, Tocche, Toke c. 1150--1180, c. 1154-1172 (15th), 1180--1195 YCh (SPNLY, p. 287), where it is not clear whether we are concemed with ON T6ki etc. or with ODan Tokki or AScand *Tokka. The influence of OE Toc(c)a can probably be ruled out in view of the rarity of monothematic names in OE sources after the end of the 9th century. It is interesting to note that whereas ODan has Tokki by the side of Töki, it has no record of *Tukki by the side of Tuki (Hald, Personnavne i Danmark: I. Oldtiden, p. 87). For further discussion and examples of AScand hypocoristic formations of the type *Tukka < Purkil, Purketel see von Feilitzen, Notes, pp. 64-66, and the present work, s. nn. Hab(b)e, *Obbi, *Oggi, *Turka. Note also the following examples: AScand *Svartki, hypocoristic form of ON Svartkell,

Scribal error for Roberti. Tengvik, p. 226, took Tuk in the form Besi Tuk 1051 (17th) KCD, no. 795 (L) to belong toan unrecorded ODan *Tiik, side form of Tiiki. KCD, no. 795 is a Crowland forgery and names which it contains should, therefore, be approached with caution. In any case, Tuk is better explained as belonging to the AScand *Tukk. 42 Hald, Personnavne i Danmark: I. Oldtiden, p. 90. It has, however, been suggested that the place-name Tokkerup is more likely to contain the substantive ODan thokki 'estimate', probably used ofunsurveyed land (DS XN, pp. 153---154; cf. also Hald, Vore Stednavne, p. 156). 40 41

384

Tumi, Tömi, Tummi, Tömi{fummi, *Törne, *Tumme Svartkollr; Suarche 1087-1098 (c. 1175-1200) Bury, p. 39 (Wattisfield St); AScand *Tuffa, hypocoristic form of Pwfast < ODan (run.) Po,fastr, Pu,ferö < ON Pmfref)ör etc; Tuffa homo comitis Wallef TRE DB, fo. 217b (Southill Bd),43 Robertus Tuffe c. 1175 (1. 12th) Newark, p. 3 (Newark-on-Trent Nt); AScand *Tugg, 44 hypocoristic form of Purgot < ON Porgautr, Purgar < ON Porgeirr, Purgrim < ON Porgrfmr etc; Dauid filio Tugge, abl. (witn.) Hy3 (1404) Burton Lazars, fo. 75 (Billesdon Lei) (FMS); AScand *Ulk(k)a, hypocoristic form of Ulkil, Ulkel < ODan, OSwed (run.) Ulfkil; Thoraldum filium Ulkke filii Radulfi Hy2 Alvingham, fo. 12b (Alvingham L) (FMS).

ON Tumi, ODan Tö.mi, Tummi, OSwed (run.) Tö.mi{fummi, OSwed (p. ns.) *Tfime, *Tumme a) (i) Tomestuna 1086 DB, fos. 164, 183, 264, 268, Tumesteda (sic) ibid., fo. 277, Tomestun' 1191 FF (PRS, xvii, no. 10), Tomeston' (p) 1201 CRR 25 (Pleas, iii, no. 32) (Thompson, Wayland H). A characteristically Danish name, Tumi, Tummi is a hypocoristic form of Pormar, Porm6ör, Pormundr (Brpndum-Nielsen, § 244, Anm. 3; DgP, col. 1493; NK VII, p. 209; Hald, Personnavne i Danmark: I. Oldtiden, p. 88). In ODan Tumi is known from several runic inscriptions, and Tummi, with shortened stem vowel and compensatory lengthening of -m-, is attested in medieval Danish records from the second third of the 12th century onwards (DR, nos. 98, 270, 276, 280, 282, 283, 294; DgP, cols. 1492-1493; Hald, Personnavne i Danmark: I. Oldtiden, p. 94). The name also occurs as the first element of several Danish place-names (Hald, Vore Stednavne, pp. 110, 145; Hjorth Pedersen, Bebygg. på -by, p. 39; DS II, pp. 12, 98; DS VI, p. 427; DS VII, pp. 235, 245; DS XII, p. 212; DS XIV, p. 77). In Sweden Tumi or Tummi is known from runic inscriptions and the name also forms the first element of several place-names (SRÖg, no. 123; SRSmå, no. 42; SRVg, nos. 3,212; SRU, no. 719; SOÄ, vii.i, p. 184; SOÄ, vii.ii, p. 48; .. .. .. 45 SOA, ix.ii, p. 86; SOA, x, p. 144; SOA, xiv, p. 146).

43 von Feilitzen, PNDB, p. 338, considered Tuffa in this DB form to be 'perhaps an AngloScand pet-form' of OScand Purfast, jin, -fero. Ekwall's suggestion (DEPN, p. 481, s. n. Tuffiey) that DB Tuffa may be a short form of OE TunfrijJ is unlikely in view of the rarity of personal names in Tun- in I. OE ande. ME. 44 Note, however, ODan Toggi, hypocoristic form of ODan Thörger and Thörgrim (Hald, Personnavne i Danmark: I. Oldtiden, pp. 90, 94). 45 SRÖg, no. 123, and SRSmå, no. 42, suggested that an OSwed (ron.) *Dömi may be a possible alternative base for runic tumi, but SRVg, nos. 3, 212, and SRU no. 719, do not include this as a possible alternative and merely give the base as Tumi, Tummi.

385

*Turka ON Tumi is known only from Icelandic, where it is not infrequent, the first recorded bearer of the name being one Tumi Kolbeinsson, who died in 1184 (Lind, cols. 1044-1045; LindS, col. 794). Lind, col. 1045, suggests that the name may have come to lceland from England. In England the name occurs in DB TRE as Tumi (nom.) (Wo) and possibly also as Tumme (gen.) (Y) (PNDB, p. 388). Björkman, NPN, p. 146, considered that it is hardly possible to distinguish the Y form Tumme (gen.) from OE Tumma, and Fellows Jensen, SPNLY, p. 293, also regarded OE Tum(m)a as a possible alternative here. Note also the York moneyer Tum(m)a 973-975, Tumme 979-985, whose name can be either ODan Tumi etc. or OE Tum(m)a (Smart, Moneyers, p. 230; Coinage of Edgar, p. 205). OE Tumma occurs in the e. 9th century part of the LVD (OET, pp. 160, 562). If the Y DB form Tumme (gen.) reflects ODan Tumi etc., the double -mm- of the form would reflect the doubling of simple consonants characteristic of AN orthography (PNDB, § 146; SPNLY, § 142), and final -e would be the result of Anglicization (PNDB, §§ 45, 150; SPNLY, § 147). SPNLY, p. 293, includes the possibility that Tumby L may contain the Scand Tumi, Tummi, but the forms favour the alternative etymon given here, OE tun (see EPN II, p. 193, s. v. tun; SSNEM, pp. 75-76). 46 In Normandy Tumi, Tummi has been noted as the first element of two place-names (Adigard, pp. 152-153, 442).

AScand *Turka b) (i) Adamfilius Turka c. 1190-1200 Binham, fo. 6b (printed, SSMEA, Appendix Il, no. 3, p. 271) (Binham). - croftum cum mesagio quod fuit Turke 1198 FF (PRS NS, xxvii, no. 120) (Burnham Thorpe).

*Turka is an AScand hypocoristic form of Purkil, Purketel < ON Porkell, Porketill etc. and several examples have been noted in English records (DBS, p. 357, s.n. Turk; von Feilitzen, Notes, p. 66). 47 An early example ofthe name is Turca (witn.) 1107-1137 Exeter Book, fo. Sa (D). A strong side-form is also on record (PNDB, p. 389; DBS, p. 357, s.n. Turk). 48

46 The proposal of Ekwall, DEPN, p. 482, s. n. Tumby, that this plaee-name eontains the Seand element tun as first element is not excluded by SSNEM, p. 75, but it ean, in fäet, be rejeeted without diffieulty in view of the fäet that tun had eeased to be a !iving plaee-name formative element in Denmark in the period of the Danish settlement in England (see Cameron, Scandinavian Settlement III, p. 149, n. 2). 47 von Feilitzen, Notes, p. 66, wrongly dates the Nf form Adamfilius Turka to e. 1100. 48 PNDB, p. 389, interprets the C forms Turch homo abbatis de Ramesy TRE DB and Turcus homo abbatis de Ramesio TRE (e. 1180) ICC as merely refleeting DB Turchel < ON Porkell with AN loss of -el, but these forms fit better to ASeand *Turk, the strong side-form of *Turka.

386

*Turka This personal name should be kept separate from the byname Ture, Tureh, Turk etc. which is known in English records from c. 1140 onwards and belongs to the ethnic name OFr ture 'Turk', perhaps used in a derogatory sense. 49 A Nf example is contained in the form Robertus Turk 1209 FF (PRS NS, xxxii, no. 210) (Bumham). Note also the Hrt field name Turkesfeld Ricl PN Hrt, p. 288, which can contain either the personal name *Turk(a) or the ethnic name Ture in its function as a byname (see Insley, JEPN, x (1977-1978), p. 55 and n. 21).

49 For the appearance of OFr turc 'Turk' as a byname in English sources, see ELPN, p. 170; DBS, p. 357, s. n. Turk.

387

I>

OSwed (run.) l>egn, l>iagn, OSwed Thiregn(e), Thregn(e) b) (i) Robertusfilius Thein 1165-1166 P, p. 23, Robertusfilius Theini 1166--1167 P, p. 22 (Lynn), Roberto filio Thein 1 et Roberto fratre eius, abl. (witn.) c. 1165 (15th) NorwCPCh, no. 124 (Lynn and Mintlyn), 2 Robertus Tein 1165-1166 P, p. 25, 1166-1167 P, p. 25, 1167-1168 P, p. 19 (Lynn).

OSwed (run) Pegn, Piagn, OSwed Thilegn(e), Thcegn(e) is an original byname corresponding to the appellative OSwed (run.) jJegn, jJiagn m. 'free man, warrior, hero' (NK VII, p. 267; cf. also the comments ofR. Kinander, SRSmå, pp. 109-110). The name is well known from Swedish runic inscriptions and occurs sporadically in later records in Sweden and Finland (SRSö, nos. 237, 349; SRSmå, no. 71; SRU, nos. 34, 131, 201, 353, 363,372,456, 935, 937, 990, 991,999; Lundgren-Brate, pp. 261,278; Modeer, Östsmåländska bondenamn från 1500-talet, p. 65; Thors, p. 88). It should be noted that the appellative jJegn occurs throughout Scandinavia, but its use as a personal name is confined to Swedish. 3 DBS, p. 345, s.n. Thain, takes the above Nf forms from the Pipe Roll of 1165-1166 to belong to the appellative OEjJeg(e)n m. 'a servant, a retainer, a nobleman with a wergild of 1200 shillings'. There is, however, no evidence that OE jJeg(e)n was ever used as a personal name, though its ME reflex is on record as a byname, 4 and jJegn- is only recorded as the first element of an OE dithematic personal name in Peg(e)nwine, the name of one of Cnut's Exeter moneyers. 5 It should be noted that OE Peg(e)nwine may have arisen by analogy with ContGerm names in Thegan-, two of which, *Theganfrid and Theganhard, are known from English records. 6 In view of these factors, and the

1 Var: Tein (e. 14th), Terry (e. 14th). Terry is the result of confusion with OFr T(h)ierri, Terri < Frankish Theuderic. 2 For the dating of this document see above, s.n. Arnkell, n. 34. 3 In OWScand Pegn occurs as the name of a mythical personage in the Rfgspula (Lind, col. 1122). 4 For examples see Thuresson, p. 49; DBS, p. 345, s. n. Thain. 5 Reaney, Studier i modern språkvetenskap, xviii (1953), p. 103; von Feilitzen, 'Some Continental Germanic personal names in England', p. 58; SCBI, xiii, nos. 479-482. Note, however, that -peg(e)n is the second element of such names as OE Leofpegn, Tilpegn etc. (fora list see OET, pp. 524-525). 6 For these names see von Feilitzen, 'Some Continental Germanic personal names in England',

388

Therir frequency of the personal name Pegn, Piagn in Swedish runic inscriptions of the late Viking period, there can be little doubt that the above Nf forms belong to this Swedish personal name and not to the OE appellative, though it should be noted that in these Nf forms the form of the Swedish personal name has fallen together with that of the OE appellative. A further English example of OSwed (run.) Pegn, Piagn, OSwed Thicegn(e), Thcegn(e) is Thein e. 12th LVD, fo. 45a.

ODan Therir a) (i) Therstuna, Sterestuna 1086 DB, fo. 181, Sterstuna ibid., fo. 260, Terston' 1209 FF (PRS NS, xxxii, no. 155) (Tharston, Depwade H). Ekwall, DEPN, p. 464, s.n., took the first element of Tharston Nf to be clearly a personal name, 'perhaps ODan Therir, for which there is some evidence'. DgP, col. 1353, groups several 14th and 15th century patronymic forms in Thcer-, T(h)er- sub Therir, though with the comment that these forms can equally reflect delabialized forms of *Th(Jrir or mutated and delabialized forms of Thorir (ibid., col. 1354). These late medieval forms, in fäet, reflect a Jutland variant of Thorir which has undergone palatal mutation and subsequent delabialization. 7 In any case, the early forms of Tharston Nf suggest rather that it contains the appellative OE jJyrs m. 'a giant, a demon', though the appearance of this element in a compound whose second element is the habitative term -tun is somewhat curious. A possible explanation could be that OE jJyrs is used here as a byname. In this connection it may be significant that the corresponding ON jJurs m. 'a giant' occurs as a byname in Olcel (LindBiN, col. 410). Cf. especially P6rir Purs, who is described by the Landnamab6k and by Egils saga Skalla-Gr{mssonar as having lived atPursstaöir. 8 Spellings in T(h)ers- and Ster(e)s- are fully compatible with derivation from OE jJyrs. The use of e for OE y is characteristic of the south-eastem dialects of ME, but it is also found occasionally in Nf (see Jordan, § 40; Selten, pp. 113-116; above, s.n. Styrger). Spellings in T(h)- reflect the AN replacement of /8/ by /t/ (ANI, pp. 41-43; IPN, pp. 108-109; PNDB, § 105; SPNLY, §§ 111, 112). The DB forms in Ster(e)s- contain inorganic initial S- as the result of AN influence (see ANI, pp. 68-70; PNDB, § 112; SPNLY, § 124.ii). pp. 57-58. *Theganfrid is represented by the form Teinfri/Je mine circwirhtan, dat., 1049-1066, prob. 1057-1066 (14th) ASWrits, no. 87 (Shepperton Mx), and Theganhard by Teinardo falconario, abl., 1129-1130 PR, p. 89. 7 Ex inf. J. Kousgärd Sprensen. Cf. the following: Brpndum-Nielsen, §§ 83, 139, 141; Spndergaard, p. 112. 8 fslendingab6k: Landnamab6k (fslenzk fornrit, i), ed. J. Benediktsson (Rekjavik 1968), p. 90; Egils saga Skalla-Grfmssonar (Islenzk fornrit, ii), ed. S. Nordal (Rekjavfk. 1933), p. 74.

389

l>ialfi, Thialvi, l>ialfi, l>elfi, Thirelve, Threlve

ON l>ialfl, ODan Thialvi, OSwed (run.) l>ialfl, l>elfi, OSwed

Thirelve, Threlve b) (i) Teluetuna 1086 DB, fo. 154b, Teluetunam ibid., fo. 215, Teluentuna ibid., fo. 269, Teluetaham (sic) ibid., fo. 154b, Thelueston' 1182-1183 P, 9 p. 15, 1183-1184 P, p. 7, 1184-1185 P, p. 35, Teluetune 1198 FF (PRS NS, xxvii, no. 86), Theluetone 1269 Ass (OKS), The/veton 1275 FF (OKS) (Thelveton, Diss H). - Thelfelond c. 1230 (c. 1270) BuryS, fo. 128a (f. n. in Tivetshall).

Pialfi (< PrScand *Peloan-) is probably a nomen agentis corresponding to Modlcel pjalfa (aö) 'to work' (SRG, p. 209). Janzen, NK VII, p. 244, interprets the name as an original byname, 'subjugator, conqueror'. Pialfi is a characteristically Swedish name. It is frequent in Swedish runic inscriptions and is also attested in medieval and 16th century records in Sweden and Finland and as the first element of Swedish place-names (SRÖg, no. 111; SRSö, nos. 194,271; SRSmå, nos. 80, 149; SRU, nos. 56,200,652,681, 778, 867, 875, 925, 948, 1052; SRG, p. 209; Lundgren-Brate, pp. 262, 278; Modeer, Personnamn i Ktb, pp. 23, 47, 120; Modeer, Östsmåländska bondenamnfrån 1500-talet, p. 66; Thors p. 88; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 69, 144, s.n. Tjälv; Sahlgren, NoB, xv (1927), p. 81). ODan Thialvi has been noted as a patronymic in the form Magnus Tj:.elvasson 1356 (Skåne) and probably also in the form Ysek Tr/Jlfison 1462 (Jutland) (DgP, col. 1356). In OWScand Pialfi occurs as the name of the servant of the god Porr in the Edda and forms the first element of a lost place-name in Norway and may also be the base of a byname in the Icelandic Landnamab6k (Lind, col. 1123; LindBiN, col. 407). In England the name has only been noted as the first element of the above Nf place-name and field name.

AScand l>ör, l>iir b) (i) Tor (liber homo) TRE DB, fo. 222b (held at Great Bircharn, sub Stigando). -Thor Steph SCBI, xx, no. 1623 (moneyer, Norwich). 10

This name is an AScand short form of names in Por-, Pur- and -porr (PNDB, p. 390; SPNLY, p. 296). It appears to have had no counterpart in Scandinavia itself. DgP, cols. 1413-1414, considers por in the Swedish runic form por, father of sihpor, to be presumably a short form of names in Thor-. This runic Telueston' CR. The moneyer's name on this coin is wom away, but it is from the same ora, Thora, Thura, Thora por is regarded by SPNLY, p. 296, as a personal name belonging to the name of the god 1>6rr. SRU, no. 440, has shown, however, that this runic por is to be interpreted as standing either for Porör or PoriR. DS IX, p. 24, considers that Tustrup in Jutland (Tustrup 1403 (1540-1549), 1468 (1540-1549), 1469 (1540-1549), 1540, 1664, 1844, Tostrup 1436 (1540-1549), Twstrup 1469 (1540-1549), 1501, Thustrup, Tvistrup 1538, Tuestrup 1610, 1683, 1688) may contain an ODan *Thur, formally identical with the name of the god, but probably a short form of names in Thur-. The early forms of this place-name are, however, too inconclusive to allow any definite decision as to its etymology. A plausible suggestion would be that it contains Danish Tue < ODan Tiivi. In England AScand Por, Pur is well attested in independent use, especially in Y (NPN, pp. 146-147; Björkman, Festermen, pp. 5, 17; PNDB, p. 390; Tengvik, p. 201; SPNLY, pp. 295-296; von Feilitzen, SCBI, xi.i, p. 13; SCBI, xxi, pp. xlviii, xlix). For the possibility that AScand Por, Pur may also occur in place-names see below, s.n. P6rir.

ON 1>6ra, ODan Thöra, Thiira, OSwed Thöra fem. b) (i) Radulfum filium Ricardi et Thure matrem suam, Thure (dat., nom. (bis)) 1198 FF (PRS NS, xxvii, no. 185) (Rudham).

ON P6ra etc. is a short form of fem. names in Por-, -p6ra, -d6ra (NK VII, p. 61). ON P6ra etc. is well attested throughout Scandinavia from the Viking period onwards (Lind, cols. 1134-1135; DR, no. 212; DgP, cols. 1374-1376, 1434-1435, 1695; SRÖg, nos. 83, 192,194,226; SRSö, no. 137; SRVg, nos. 30, 41, 66, 67, 106, 115, 171; SRU, nos. 32,168,465,682; Lundgren-Brate, p. 263). In English records the name is infrequent (NPN, p. 147; SPNT, p. 145; SPNLY, p. 296). The above Nf forms reflect the Danish variant Thura. Initial Th- in the above Nf forms probably reflects AN replacement of OScand /8/ by /t/, though the use of th for /8/ is not unknown in OE orthography (PNDB, § 105; SPNLY, § 111).

391

J>oraldr, Thorald

ON l>oraldr, OSwed Thorald a) (i) Thoraldes toft Hy2 (17th) BS, no. 330 (f. n. in Ketteringham). b) (i) Turoldus. I. fiber homo TRE DB, fo. 229b (Wolterton). -Toradre TRE DB, fo. 267 (Attleborough). 11 - Taraldus homo Willelmi de War(enna), Turaldus 1086 DB, fo. 133 (Hevingham), Turold' ibid., fo. 157b (Hackford), Turoldus ibid., fös. 158, 158b (ter), 159, 172 (bis) (held properties at Coltishall, Crakeforda in Banningham and Little Barningham, liberi homines with landat Irmingland, Paston, Filby and Plumstead, bordars with landat Wolterton and Banningham anda sokeman with landat Banningham). - Turold' 1086 DB, fo. 177b (Mundham), Turoldus ibid., fos. 177b, 178, 181 b, 182, 186b, 190 (Seething, Little Hockham, Hempnall, Heckingham, Baconsthorpe, South Erpingham H, Haddiscoe), Turaldus ibid., fo. 184 (held one fiber homo at Morston). Turoldus 1086 DB, fo. 228 (Tunstall). -Turoldus de Gimingeham c. 1100--1120 CAcre, fo. lb(i) (charter of William Il de W arenne, confirming, inter alia, Turold' s grant of the church of Witton, Tunstead H; Gimingeham is Gimingham). - Et Testurtuna. et Riburg. et Wodedallingga. et Saxselyngham que tenet Thuroldus 1101-1107 Binham, fo. la, Turoldo de Dallinga, abl. (witn.) ibid., fo. lb (grant by Peter de Valoignes to Binham Priory; Testurtuna is Testerton, Riburg is (Little) Ryburgh, Wodedallingga is Wood Dalling, and Saxselyngham is Saxlingham, Holt H), Turoldus de Dallinges (witn.) 1108 Binham, fo. 20b (printed, SSMEA, Appendix I, no. 35) (Binham and Langham), Turoldus de Dallinge, Radulfusfilius Turoldi (witn.) 1121-1135 BactonCh, no. 2 (charter of Bishop Everard ofNorwich), Thuraldus de Dallinges (witn.) c. 1130 Binham, fo. 2b (confirmation by Roger de Valoignes of grants made by his father Peter de V aloignes to Binham Priory), Turold' de Dalling (abl.) (witn.) ibid., fo. 24a(i) (London, charter of Walter de Valoignes ), Turaldus de Dalling (witn.) ibid., fo. 145a(i) (charter of Roger de V aloignes concerning churches and property at Dersingham and Great W alsingham and property at Gunthorpe and Wells next the Sea), per manum Radulfifilii Thoroldi 1146 - c. 1160 Binham, fo. 123b(ii) (charter of Bishop William of Norwich conceming the church of St Andrew of Wood Dalling), Radulfi filii Turoldi de Dalling, gen., 11611165 Binham, fo. 168a(ii) (charter ofBishop William ofNorwich concerning the church of Little Ryburgh), Carta Radulfi filii Thoraldi c. 1160 Binham, fo. 12la(iii), Radulfus filius Thoroldi de Wodedalling' ibid., fo. 121b (grant by Ralph son ofThorold ofWood Dalling to Binham Priory of the church of Little Ryburgh, rents at Little Ryburgh and Wood Dalling and landat Britrichescroft), ex dono Radulfi filii Thoroldi 1171-1174 Binham, fo. 44b(i) (charter of Bishop William of Norwich confirming, inter alia, the grant made by Ralph son ofThorold ofWood Dalling in Binham fo. 12la(iii)-12lb), Radulfus filius Toraldi (bis) ibid., fo. 19a(ii), Radulfus filius Thorald' ibid., fo. 19b (charter of Reginald de Warenne confirming Ralph son of Thorold's grant of rents at Little Ryburgh and Wood Dalling to Binham Priory), Carta Radulfi filii Thoraldi, Radulfus filius Toraldi de Wodedalling' c. 1160--1165 Binham, fo. 12la(i), Carta Radulfifilii Thoraldi, Radulfusfilius Turoldi de Woddalling ibid., fo. 164b(iii) (Wood Dalling, Little Ryburgh, Britrichescroft), Carta Radulfi filii Thoraldi, Radulfus filius Thoroldi de Dalling ibid., fo. 121a(ii) (printed, SSMEA, Appendix I, no. 36) 12 (Little

11 Toradre can be ON P6raldr with AN loss of -/- or it may altematively stand for ON P6roddr with -a- for -o- as a result of scribal error (PNDB, pp. 386--387). Final -re is probably an AN addition (PNDB, p. 387). ON P6roddr is well attested in Iceland but is rather infrequent in Norway (Lind, cols. 1202-1204; LindS, col. 877). 12 This document is wrongly dated by SSMEA, Appendix I, no. 36, to c. 1125.

392

l>oraldr, Thorald Ryburgh, Britriztoft), Carta Radulfi .filii Thoraldi, Radulfus .filius Thoraldi ibid., fo. 12lb(i) (the mill of Wood Dalling), Carta Radulfi .filii Thorald' de terra quam dedit Helewyse sorori sue, Radulfus .filius Thoroldi de Dalling ibid., fo. 12lb(ii) (the document isa grant of the land of Helewise of (Little) Ryburgh), Carta Radulfi de Dalling' .filii Thoraldi, Radulfus de Dalling' .filius Thorald' ibid., fo. 127b(i), Radulfus de Dalling .filius Thorald' ibid., fo. 146a(i) (the mill of Tyby in Wood Dalling), ex dono Radulfi .filii Thuroldi c. 1175 Binham, fo. 3b (Wood Dalling, Little Ryburgh Britrichescroft), ex dono Radulfi .filii Thoroldi et Rogeri .filii eius 1183 Binham, fo. 45a(i), ex dono Radulfi .filii Thoraldi et Rogeri .filii eius 1206 Binham, fo. 43a (Wood Dalling, Little Ryburgh, Britrikescroft), ex dono Radulfi.filii Thoroldi et Rogeri.filii eius 1230 (lnspeximus of the charter of 1183) Binham, fo. 43b(i) (Wood Dalling, Little Ryburgh, Britriskescroft). - Turoldus (presbiter) (witn.) 1135-1143 NorwCPCh, 13 no. 116, 1135-1143 (e. 14th) NorwCP, p. 58 (charter of Bishop Everard of Norwich to Norwich Cathedral Priory), Turoldus presbiter (witn.) 1136-1143 NorwCPCh, no. 117, Thuroldus presbiter 1136-1143 (e. 14th) NorwCP, p. 62 (charter ofBishop Everard of Norwich to Norwich CathedralPriory), Turoldo (clerico), abl. (witn.) 1139-1143 (15th) NorwCPCh, no. 119 (charter of Bishop Everard of Norwich to Norwich Cathedral Priory), 1146-1149 BodlNfCh, no. 224 (printed, Holme, no. 89) (charter of Bishop William of Norwich conceming the church of Lamas), Turaldo capellano, abl. (witn.) 1146-1147 CAcre, fo. 4a (Nf), Turoldus capellanus (witn.) 1146/c. 1149 (15th) NorwCPCh, no. 129 (Martham), 1146-1148 Har1Ch47 H45 (printed, Cheney,English Bishops' Chanceries, p. 154, no. 4) (Rudham, Marham, the mill ofTorp), c. 1150-1153 (e. 14th) NorwCP, p. 64 (Fibrigg, Norwich), Thorold' capellano, abl. (witn.) 1146-c. 1150 Hickling, fo. lOa(i) (the church of Waxham), Turold capellanus (witn.) 1146- c. 1157 (15th) NorwCPCh, 14 no. 127 (Sedgeford), Turoldus capellanus episcopi (witn.) 1146 - c. 1170 (15th) NorwCPCh, no. 130 (the document concems the homage of Richard son of Odo of South Elmham Sf and land which belonged to Osbem of Hindringham Nf), Toraldus capellanus episcopi c. 1150 (1302) NorwCPCh, no. 259 (2), Thoraldus capellanus episcopi c. 1150 (e. 14th) NorwCP, p. 64 (Norwich), Torald' capellano, abl. (witn.) c. 1150 CAcre, fo. 47a (the document concems the church of South Creake), Turoldus (witn.) c. 1150-1159 (1. 13th) NorwCPCh, no. 134 (the document concems the mill of Pedeham which was attached to the manor of Langley), Thoroldo capellano, abl. (witn.) 1153-1168 Holme, no. 95 (confirmation by Bishop William of Norwich of the grant to Thomas the clerk of Ludham of the church of Potter Heigham on the presentation of Abbot William Il of St Benet of Holme). - Toraldus, terram quam dedit eis Toraldus 15 c. 1135-1140 CAcre, fo. 16a(vi), terram quam dedit ipsis Toraldus de Massingeham 1138-1147 CAcre, fo. 4a(i), terram quam Toraldus dedit eis c. 1150 CAcre, fo. 16b(i) (landat Massingham granted to Castle Acre Priory). - donationem unius sochemanni nomine Turoldi de Scotehou qui reddit dimidiam marcam argenti 1138-1147 YCh, viii, no. 40, ad Scutehou terram unius sochemanni nomine Turoldi 1147 YCh, viii, p. 91 (note to no. 40) (Scottow). - Roberto .filio Turold', abl. (witn.) c. 1145 CAcre, fo. 80a(ii) (Lynn). -Rogerus.filius Turaldi (witn.) e. Hy2 CAcre,

13 The editor of NorwCPCh dates this charter to '(? 1141-1143)'. The dating given in the present work follows HRHEW, pp. 102, 104. 14 Dated by the editor of NorwCPCh to '1146-e. 1170, probably in the earlier part of this period'. 15 This fonn appears in the MS. as Toroldus but a mark under the second o and ana inserted over this o indicates that the fonn should be corrected to Toraldus.

393

l>oraldr, Thorald fo. 68b(i) (the document concems the cell of Mendham Sf with its appurtenances and landat Crochestune, probably Croxton, Grimshoe H, Nf). - Willelmo filio Torald', abl. (witn.) c. 1160 CAcre, fo. 16b(ii) (Massingham). - Turoldus Taneur 1165-1166 P, p. 24, Turold16 le Tanur 1166-1167 P, p. 23 (Lynn). -Turoldfilius /Elueue 1165-1166 P, p. 28, Turoldusfilius /Elueue, 17 /Edwardfrater Turoldi 1166-1167 P, p. 28 (Lynn). Turoldus Flobi 1165-1166 P, p. 28, Turold Flobi 1166-1167 P, p. 28 (Lynn). -Turoldus de Breteham 1169-1170 P, p. 12, Turoldus de Bretheham 1170--1171 P, p. 9, 1171-1172 P, p. 29 Toroldus de Breteham 1172-1173 P, 18 p. 125 (Brettenham). -Turoldus de Walesham 1169-1170 P, p. 12, 1170--1171 P, p. 9, 1171-1172 P, p. 29, Toroldus de Walisham 1172-1173 P, 19 p. 125, Turaldus de Walesham 1173-1174 P, 20 p. 44, 1175-1176 P, 21 p. 64, Turaldus de Walisham 1174-1175 P, 22 p. 115 (North or South W alsham Nfor Walsham le Willows St). - Robertus filius Torold' 1174-1175 P, p. 119, de Roberto filio Turoldi 1175-1176 P, p. 66 (Nf or St). - Turoldus clericus de Geistwit (witn.) c. 1175-1180 Binham, fo. 123a(i) (Wood Dalling, Little Ryburgh, Brittrichescrof), Thurold' de Geystweith (witn.) ibid., fo. 128a (Wood Dalling, Little Ryburgh, Brigtrichescrof (sic); Geistwit, Geystweith is Guestwick), Robertumfilium Turold' 1202 FF (PRS NS, no. 451), Robertusf Thoroldi 1203 AssRoll 817 (NthRS, v, no. 843),Robertusf Thorold' 1203 AssRoll 817 (NthRS, v, no. 844) (Guestwick). -de Turold' fabro et filiis ejus 1176-1177 P, p. 130 (Nf or St). - Rogerus frater Turoldi 1176-1177 P, 23 p. 131, 1177-1178 P, 24 p. 24,Rogerusfilius Torold' 1178-1179 P, p. 5, Rogerusfilius Turoldi 1179-1180 P, p. 17, 1180--1181 P, p. 83, 1181-1182 P, p. 67, 1182-1183 P, p. 9, 1183-1184 P, p. 4 (Nfor St). -Thoroldo j{ilio) Bartholomei, abl. (witn.) c. 1180--1185 Binham, fo. 49b(i) (Bamey). - Regina/dus Turald 1196 CRR 7 (Cur, i, p. 21) (Nf). - Reginaldumfilium Torroud, Regina/dus filius Torroud', Reginaldo filio Torroud', dat. (bis), de Regeginaldo (sic) filio Torroud 1196 FF (PRS, xx, no. 35), Reginaldumfilium Thorald' 1198 FF (PRS NS, xxvii, no. 60) (Mouton, i.e. probably Moulton St Michael), Reginaldo filio Torold, dat., 1212 CRR 55 (Cur, vi, p. 244) (recognitor in a case conceming property at Moulton St Michael). - Decena Turoldi de Totinhull' 1197 P, p. 246, 1198 P, p. 89, 1199 P, p. 271 (Tottenhill). -Henricumfilium Thuroldi 1198 FF (PRS NS, xxvii, no. 194) (Lynn). - iiij°'. acras quas Rogerus filius Thoroldi tenuit c. 1200 CAcre, fo. 30a(vi), iiii0 ' . acras terre quas Rogerus filius Toroudi tenuit c. 1210 CAcre, fo. 31b(i) (Rougham). -Matill' filiam Torald', Matill' filia Torall' 1202 FF (PRS NS, xxvii, no. 400) (Carleton'). -Estrildam que fuit uxor Ricardi filii Torold', de libero tenemento quod fuit Ricardi Turold' 1203 FF (PRS NS, xxxii, no. 39) (Lynn). - Willelmus .. f Thorold' 1203 AssRoll 817 (NthRS, v, no. 870) (Thorp). De Thoroldo f Rogeri 1203 AssRoll 817 (NthRS, v, no. 780), Thoroldus f Rogeri, Misericordia Thorold' 1203 AssRoll 817 (NthRS, v, no. 832) (Haveringland), Thoraldusfilius Rogeri 1209 AssRoll 558 (Pleas, iv, no. 4358) (Witchingham). -de libero tenemento quod fuit Thoroldi coci patris ipsius Agnetis (i.e. Agnes, wife of William of

Toroldus CR. Toroldusfilius Alfueue CR. 18 Turoldus de Bretheham CR. 19 Turoldus de Walesham CR. 20 Toraldus de Walisham CR. 21 Taroldus de Walesham CR. 22 Turaldus de Walesham CR. 23 Rogerusfilius Turoldi CR. 24 Rogerusfrater Torold' CR. 16 17

394

l>oraldr, Thorald Runhall) 1205 FF (PRS NS, xxxii, no. 75) (Mattishall and Burgh Mattishall), de libero tenemento quod fuit Thoroldi Coc patris ipsarum Cecilie 25 et Agnetis 1205 FF (PRS NS, xxxii, no. 76) (Hockering, North Tuddenham, Burgh Mattishall), Godamfiliam Thoraldi coci, de libero tenemento quod fuit Thoraldi coci patris predictarum Gode et Cecilie 1206 FF (PRS NS, xxxii, no. 89) (Burgh Mattishall, Mattishall, Hockering, North Tuddenham). - Willelmumfilium Toroldi 1206 CRR 42 (Cur, iv, p. 114) (Nf), Willelmum Torold' ibid. (Cur, iv, p. 130) (Sedgeford), Willelmus filius Toroldi ibid. (Cur, iv, p. 152), Willelmo filio Toroldi, dat., ibid. (Cur, iv, p. 156) (Nf). -Gerardusf Turoldi 1206 P, p. 34 (Nf or Sf). - Turold' sutor 1209 CRR 49 (Pleas, iv, no. 3778) (Nf). - Godricus pater Thoraldi, de ipso Toraldo, de ipso Thoraldo, Thoraldus 1209 AssRoll 558 (Pleas, iv, no. 4233) (Mort d'ancestor if Godric father of Thorald was seised in his desmesne as of fee of two acres of land with appurtenances in Booton on the day he died etc.). Thoraldum de Couton' 1209 AssRoll 558 (Pleas, iv, no. 4234) (Booton; Couton' is Colton). -Torald' de Tortun' 1209 CRR 49 (Pleas, iv, no. 3899) (Nf), Toroldi de Turtun', gen., 1210 CRR 52 (Cur, vi, p. 30) (the document concems the advowson ofthe church of Ingworth), Thoroldi de Turton', gen., 1212 CRR 55 (Cur, vi, p. 289) (Nf), Turoldum de Torton' 1225 CRR 88 (Cur, xii, p. 212) (recognitor; the document concems Cawston; Tortun', Turtun', Turton' and Torton' represent the name of a lost village near Witchingham).26-terramRobertifilii Thoraldi (bis) c. 1210 Der, fo. 277a(ii) (Themelthorpe). - Radulfus filius Thoroldi 1211 CRR 54 (Cur, vi, p. 152) (Wood Norton). - terram Thoraldi textoris c. 1215 Der, fo. 255b(ii), terram Thorold' textoris ibid., fo. 256a (Oxborough). - terram Hulme filii Thuraldi c. 1215 Der, fo. 257a(ii) (Oxborough). - terram Rogeri filii Thoraldi c. 1215 Wals, fo. 61a(i), Carta Rogeri filii Thoraldi, Rogerus filius Thoraldi de Hindringham ibid., fo. 63b(iv) Carta Rogeri filii Thoraldi c. 1215 (15th) Wals, fo. 142a (15th century index of the cartulary), terram Hugonis filii Thoraldi c. 1215 Wals, fo. 63a(iii) (Hindringham). -terram Toraldi c. 1220 CAcre, fo. 54a(v) (Barmer). - Purold Satt 1222 ElyA, fo. 205a (East Dereham). - Edmundus filius Thorald, Eliafilius Thorald 1222 CRR 81 (Cur, x, p. 277) (Sparham). -Thoraldo Honag, abl., 1230 CRR 106 (Cur, xiv, p. 44) (Weston Longville). - domum Thoroude 1237-1251 Holme, fo. 96b(i) (Great Hautbois). - Carta Thoraldi le Bretun de Wichingham, ego Thoraldus le Bretun de Wichingham et Auelina vxor mea ? 1245 Wals, fo. 22b(i) (grant by Thorald le Bretun and his wife Avelina to Walsingham Priory of the homage of Roger de Grestun and of rents in Great and Little W alsingham and of villeins in Holkham; Wichingham is Witchingham), Carta Thoraldi le Bretun et Aueline vxoris sue ibid., fös. 37b(ii), 48a(ii) (rubrics only of charters conceming rents in Great and Little W alsingham and homages in Holkham), Thoraldum le Bretun et Auelinam vxorem eius, predicti Thoraldus et Auelina (bis), de predictis Thoraldo et Auelina (ter), de dono predictorum Thoraldi et Aueline, predictos Thorald' et Auelinam 1245 Wals, fo. 22b(ii) (final concord between Abbot William of Walsingham and Thorald le Bretun and his wife A veline concerning land and rents in Holkham and rents in Great and Little Walsingham), Carta Thoraldi Britonis, Thoraldus Brita de Wichingham, Thoraldum m. 13th Wals, fo. 107b(iii)27 (Witchingham), et quarta parte molendini de Bleckemilne in Magna Wals' que Rogerus pater meus adquisiuit in feodo de Thoraldo le Bretun de Wichingham et Auelina vxore eius c. 1260 Wals, fo. 38a (Great Walsingham, charter of

25

26 27

Cecilia, daughter of Thorald the cook, was the wife of Richard of Tuddenham. See Darby and Versey, Domesday Gazetteer, p. 291 and Map 35 grid reference H. 4. In this document the name of the wife of Thorald the Breton of Witchingham is given as

Margar(eta).

395

l>oraldr, Thorald John son ofRoger de Grestone). -de Torald' Bening 1275 RH, i, p. 489 (Eynsford H). - in decena Thorald' de Quitting' (gen.), Thorald' de Qwitingg' (acc.) 1279-1299 W als, fo. 103b (the document concems a messuage with two cottages at the bridge of Sall; Quitting', Qwitingg' is unidentified). - terram que fuit Willelmi Thorald 1. 13th Holme, fo. 161 b(iii) (Paston).

ON P6raldr < Porvaldr is extremely frequent in medieval Norwegian records from the end ofthe 13th century onwards (Lind, cols. 1137-1139; LindS, cols. 831-839). An earlier example is runic pural(t), acc., found in an 11 th century runic inscription from Alstad, Opland Fylke (NiyR, no. 62). A weak form, P6raldi, is also on record in medieval Norway, though after c. 1300 it is not possible to separate it from the strong form (Lind, cols. 1135-1137). ON P6raldr and its weak side-form P6raldi are not attested in medieval Iceland, where the name is represented by the more conservative form Porvaldr, a variant which is also found sporadically in medieval Norway (Lind, cols. 12121214). In Sweden Thoraldus occurs as the name of a 12th century cellerarius from the Cistercian monastery of Nydala in Småland and Tarald- has been noted in 16th century patronymic formations from Österbotten in Finland (Thors, p. 89). 28 Lundgren-Brate, pp. 263-264, also hasa few 14th and 15th century examples of Porallder, Torall etc. from Bohuslän and Jämtland, but these are to be regarded as Norwegian rather than Swedish. DgP, col. 1353, s.n. Therels, suggested that the Danish place-name Torrendrup (Thoreldrup 1. 14th) may contain an ODan *Thorwald, corresponding to ON P6raldr, OSwed Thörald, but Kousgård Sprensen RJb, p. 240, considers that this place-name most likely contains the ODan fem. personal name Thcmelf. In England ON P6raldr etc. is one of the most frequent Scand personal names, being found both in independent use and as the first element of place-names and field names: NPN, pp. 160--161; ZEN, p. 87; PNDB, p. 390; Tengvik, p. 202; SPNT, p. 145; ELPN, p. 82; DBS, p. 347, s.n. Thorold; SPNLY, pp. 296-300; Smart, Moneyers, p. 220; von Feilitzen, PNWinton, p. 175; DEPN, pp. 460,467,478, s. nn. Tarleton La, Thorlby, Torrisholme; PN La, pp. 124, 138, 177; IPN, p. 92; PN NRY, pp. 175-176, 268; PN ERY, pp. 25, 125, 149; PN Hrt, pp. xx, 260,312; PN Cu, p. 358; PN Db, p. 238; PN WRY, v, p. 6; PN WRY, vi, p. 76; PN We, ii, p. 314; Ekwall, Selected Papers, p. 65; SSNY, pp. 39, 68; SSNEM, p. 127; Insley, JEPN, x (1977-1978), p. 55. Sf examples include the following: Turoldus Papilio (witn.) 1066-1087 (1. 13th) Bury, no. 168 (St); Turoldus de Halsted2 9 (witn.) c. 1115 (c. 1175-1200) Bury, no. 174 (Santon Downham), Turoldi de

28 The change P6raldr > Para/dr is a late medieval Norwegian development (Noreen, AislGramm, § 121). The 16th century examples of Tarald- in patronymic formations in Österbotten almost certainly reflect Norwegian influence. 29 Var: Thuroldus de Haustede (I. 13th).

396

l>oraldr, Thorald Halstede, 30 gen. (witn.) 1121-1135 (c. 1175-1200) Bury, no. 112 (the document isa charter of Abbot Anselm ofBury St Edmunds conceming various properties at Chebenhale in Fressingfield, Bury St Edmunds, Pakenham, Ingham, Stowe, Sidoluesmere, Semer, Chelsworth and Bradfield; Halstede is Hawstead); Turold Swein 1186-1188 (c. 1270) BuryS, p. 15 (Woolpit), carucata Turoldi ibid., p. 18, Turold ibid., pp. 18, 19 (Rougham), 30 (Nowton), Turoldus ibid., p. 37, Hendricusfilius Turoldi ibid., p. 38, Turold, Henricusfilius Turold ibid., p. 39 (Stanton); Johannemfilium Turold' et Geuam uxorem suam 1200 FF (PRS NS, xxxii, no. 290) (the document concems property at Ashbocking, Hemingstone, Ipswich and Rattlesden); Walterumfilium Toroldi, Gerardus filius Thoroldi 1203 CRR 29 (Cur, ii, p. 150) (the document concems Coddenham).

ON P6raldr etc. occurs as the name of no less than 28 persons in Norman records ofthe period 968-1066, appearing in the forms Torolde, Toroldus, Turoldus, Toraldus and Turaldus (Adigard, pp. 172-173, 342-347). The name continues to be frequent in Normandy in the late 11th and 12th centuries, when it appears as Toroldus, Turoldus etc., and it remains in use as a personal name in Normandy inta the second half of the 13th century, appearing latterly as Thoroudus, Theroldus, Theroudus etc. (Adigard, p. 172, n. 81). The name has also been noted in Norman place-names (Adigard, pp. 172, 432-433). Many of the English examples of ON P6raldr etc. reflect Norman forms of this name and many of the bearers of the name in post-Conquest English records were doubtless Normans or of Norman descent (cf. SPNT, p. 145, n. 5; SPNLY, p. 300). In this connection it is interesting to note that the name is fairly infrequent in England in the pre-Conquest period and that its popularity increases greatly after 1066. A clearly Continental bearer of the name from Nf is the 13th century landowner Thorald the Breton of Witchingham. Spellings in T- reflect the AN replacement of Scand /8/ by /t/, and the same is doubtless also true of spellings in initial Th-, though perhaps in same cases the fäet that th was sometimes used for /8/ in OE and e. ME may also have to be borne in mind (ANI, pp. 41-43; IPN, pp. 108-109; PNDB, § 105; SPNLY, §§ 111, 112; for th for /8/ in OE ande. ME see PNDB, § 105; Campbell, § 57.5; ASCh, p. xxix; Jordan, § 16, Anm. 2; cf. also SMED, p. 220). The ending -oldus in Norman Toroldus, Turoldus is due to the influence of Frankish names in -old < -(w)ald, though in English forms in -old(us) native sound developments must also be considered (cf. Luick, § 113; PNDB, § 2; Adigard, p. 245; SPNLY, § 4.ii; above, s.n. Ingialdr). Forms in -oud show AN vocalization of preconsonantal -l- > -u- (PNDB, § 61; SPNLY, § 69). The coexistence of an AScand variant beside the Norman Turoldus etc. in England in the ME period is confirmed by ElyA, where the AScand variant is represented by the Nf form Purold Satt and the AN variants by the following: 30

Var: Turoldi de Haustede (I. 13th).

397

*l>orbert, *l>urbert Turold' de stapel' 1222 ElyA, fo. 108a, Turold' ibid., fo. 109a, Torold' ibid., fo. 110a (Linden End in Haddenham C); Thuroldus filius Ricardi ibid. fo. 195a (Hartest Sf). Initial Pur- and T(h)ur- formally reflect OEScand (Danish) Pur-, though the appearance of spellings in -o- and -u- for the name of the same person in English records of the ME period suggests that the distinction between OScand Por- and the specifically OEScand Pur- had ceased to be meaningful in the ME period (see PN La, pp. 244-245; PNDB, p. 390; SPNLY, § 28).

Scand-West Frankish hybrid *l>orbert, *l>urbert b) (i) Torbertus fiber homo Stigandi TRE DB, fo. 176b (Shimpling). -Turbertus. I. fiber homo TRE DB, fo. 147b (Taverham). - Turbertus decanus (witn.) 1140--1145 BodlNfCh, no. 607 (printed, Holme, no. 161) (Grenesuilla), 1153-1168 Holme, nos. 168 (Rackheath), 177 (Norwich), 184 (Thwaite, Thurgarton, Paston), 188 (Thurgarton), Turbertus decanus et Thomas filius eius (witn) 1153-1166 Holme, no. 170 (Ranworth), 1155-1168, prob. 1163-1168, Holme, no. 180 (Carleton Forehoe and Barford), Turberto decano, abl. (witn.) 1153-1168, Holme, no. 169 (Rackheath). Roberto de Ludham fratri Thurberti, dat., Thurbertus decanus (witn.) 1146 - c. 115031 Holme, no. 153 (grant by Abbot Hugh of St Benet of Holme to Robert of Ludham, brother of Thurbert, on Robert's marriage to the daughter of Walter Tyke, of land in North Walsham and Antingham which Walter had held of the abbey, for three shillings annually), Thurbertus decanus (witn.) 1153-1166 (l. 13th) Holme, no. 128 (Burwood andLudham), 11531168 Holme, nos. 187 (Thwaite), 189 (Paston), 1153-1168, prob. c. 1163, Holme, no. 201 (Hardley), 1153-1166 Holme, no. 166 (the document concems landat Clippesby, Repps, Webfen, Flegg and Waxham), Thurberto decano, abl. (witn.) 1163, Holme, no. 193 (quitclaim to the abbey of St Benet of Holme of the villa of Heigham next Norwich), Thurbertus presbiter (witn.) 1146--1149 Holme, no. 151 (grant ofland at Beeston St Lawrence), Thurbertus de Ludham (witn.) 1146-c. 1150 Holme, 32 no. 146 (grant to Osbert the priest of Thurgarton, of land between Runton and Felbrigg for twelve pence of rent annually, which belonged to the four pounds due from Thwaite), Turbertus de Ludham (witn.) 1153-1168, prob. 1162-1168, Holme, no. 87 (confirmation to the abbey of St Benet of Holme of a third of the tithes at Ingham and W axham, two thirds of the tithes at Worstead and Brunstead, and a half of the tithes at Horsey; Ludham is Ludham). 33 - De Turb(erto) 34 craiariere 1165-1166 P, p. 34 (Nf or St). - Turbertus de Baenburch 1193 P, p. 26 (Bawburgh). -ThurbertumfiliumRogeri, predicto Thurberno,

Wrongly dated by the editor ofHolme to 1141-1149. Wrongly dated by the editor ofHolme to 1141-1149. 33 All these Holme examples of T(h)urbertus relate to the same person. His name is rendered in other Holme charters as T(h)urbernus and T(h)uruerdus, Turuerhdus. These forms are given below, s. nn. Porbiqrn, *Porfr~or. In these Holme forms *Porbert, *Purbert and AScand Purbe(o )rn < ON Porbiqrn etc. and AScand Puifero < ON *Porfr~or etc. interchange. Cf. the use of Purberd for Puifero in BCS no. 1344, a 14th century copy of a charter of 934 (see Crawf, p. 75; PNDB, p. 391, n. 5. Cf. also above, s. nn. Äsbiqrn, Ketilbert, Kylfu-vqror. 34 Turb(erno) is equally possible, in which case the etymon would be ON Porbiqrn etc. 31

32

398

l>orbiQrn, Thorbiorn dat., predictus Thurbernus 35 1198 FF (PRS NS, xxvii, no. 13) (Bunwell). - Regina/dus auunculus Turberti 1209 P, p. 34 (Nf), Turbertus de Holcham 1209 P, p. 34, 1210 P, p. 54 (Holkham). - Elias filius Thurberti, Robertusfilius Thurberti 1225 CRR 89 (Cur, xii, p. 59) (Sparham). 36

*Porbert, *Purbert is a hybrid formed from OScand Por-, Pur- and West Frankish -bert. Cf. also Ketilbert (above s.n.). *Porbert, *Purbert doubtless originated in Normandy, cf. the place-name Thouberville (Eure) (Tuberti villa c. 1060, Tubervilla 1175),37 and it is well attested in England in the postConquest period, though in same cases we may be concemed with ON PorbiQrn etc. owing to AN confusion of -bert and -bern (NPN, pp. 154-155; PNDB, pp. 390--391; ELPN, pp. 15, 97 and n. 1; DBS, pp. 356-357, s.n. Turbard; SPNLY, pp. 300-301; von Feilitzen, PNWinton, p. 174).

ON l>orbi\>rn, ODan, OSwed Thorbiorn b) (i) Turber. fiber homo TRE DB, fo. 146 (Briningham). -Turbern. I. fiber homo TRE DB, fo. 147b (Swannington). -Turbernus TRE DB, fo. 236 (Castle Acre/West Acre). -Testimonio Turberni presbiteri 1146 Holme, no. 142 (Mancroft in Norwich), Transcriptum Thomefilii Turberni, Thomefilio Turberni, dat., 1153-1168 Holme, no. 183 (grant to Thomas son of Turbem of the churches of Ludham, Potter Heigham and St Michael's in Norwich, for ten shillings annually), Transcriptum Thome filii Turberni presbiteri de firma de Hecham, Thome filio Turberni presbiteri, dat., ibid., no. 190 (Potter Heigham), Transcriptum Thome filii Turberni, Thome filio Thurberni, dat., 11751186 Holme, no. 249 (grant by Abbot Thomas of St Benet of Holme to Thomas son of Turbem of the churches of Ludham, Potter Heigham and St Michael's in Norwich, which his father had held, for ten shillings annually to the sacrist), Transcriptum Thome filii Turberni presbiteri, Thome filio Thurberni presbiteri, dat., ibid., no. 257 (Potter Heigham). 38 - De Turb(erno) 39 craiariere 1165-1166 P, p. 34 (Nf or St). - homagium Galfrid' filii Thurb(er)n c. 1190 (1. 13th) CUL MS. Ff ii. 33, fo. 134a(iii) (Depwade H). - Ricardum Turbern et Mariam uxorem suam 1198 FF (PRS NS, xxvii, no. 147) (Hudeston in Bunwell). - Willelmus filiaster Turberni c. 1200 CAcre, fo. 46a(i) (Reepham). -Turbernum 1202 CRR 28 (Pleas, ii, no. 1075) (Nf). - Umfridumfilium Turber'

35 ON PorbiQrn is equally possible here. Confusion of -bern and -bert is an AN feature which is probably due to the occasional loss 9fthe final consonant in these elements (PNDB, p. 391). For further examples, see above, s. nn. AsbiQrn, Ketilbert. 36 Robertus son of Thurbert also occurs as Robertus filius Thurberni 1222 CRR 81 (below, s.n. PorbiQrn). It is, therefore, not possible to decide whether the etymon is *Porbert, *Purbert or ON PorbiQrn etc. See also above and n. 35. 37 Adigard des Gautries, Annales de Normandie, v (1955), p. 26; von Feilitzen, PNWinton, p. 196, n. 5. 38 The T(h)urbernus of these Holme entries appears in other Holme charters as T(h)urbertus and as T(h)uruerdus, Turuerhdus (see above, s.n. *Porbert, n. 33). 39 Turb(erto) is equally possible, in which case the etymon would be the Scand-West Frankish hybrid *Porbert, *Purbert (q. v.).

399

l>orbiQrn, Thorbiorn 1209 FF (PRS NS, xxxii, no. 139), Humfridumfilium Turbern' 1209 FF (PRS NS, xxxii, no. 152), Turber pater Gaufridi 1209 AssRoll 558 (Pleas, iv, no. 4386)40 (Morton on the Hill). - Turbern' Swain' (acc.), Turbernus 1209 AssRoll 558 (Pleas, iv, no. 4136) (Ormesby). -Turbern Wiege (acc.), Turbern (bis) (acc., nom.) 1209 AssRoll 558 (Pleas, iv, no. 4290) (Weston Longville). - homagia Nicholai et Radulfi filii Thurbern de Wigehale c. 1220 Der, fös. 217b(i), 218a(i) (Saddlebow in Wiggenhall St Mary the Virgin; Wigehale is Wiggenhall). -Thurberno tanatori, dat., dictus Thurbernus, dicto Thurberno, dat., c. 1220 Bod!NfCh, no. 74 (Colegate). - Robertus filius Thurberni 1222 CRR 81 (Cur, x, p. 277) (Sparham). 41 - Turbernus Pytel c. 1240 (m. 14th) Rams, i, p. 407 (Ringstead). - Dom' Thurb(ern), domo Thurbern 1275 RH, i, p. 527 (Forehoe H).

ON PorbiQrn etc. is frequent throughout Scandinavia from the Viking period onwards (NiyR, ii, pp. 87-88, 95-96 (no. 107); NiyR, iii, pp. 204-205 (no. 239); NiyR, v, p. 221; Lind, cols. 1147-1150; DR, nos. 97, 99, 161; DgP, cols. 1376-1382, 1695; DS IX, p. 127; DS XIV, p. 261; SRÖl, no. 56; SRÖg, nos. 32, 120, 151; SRSö, nos. 48, 50, 61, 84, 159,190,202,229,232,360; SRSmå, nos. 99, 110, 133; SRVg, nos. 156,242; SRU, nos. 29, 37, 61, 70, 86, 151, 176,180,229,379,391,394,405,430,456,467,481,510,532,628,686, 838, 854, 938, 1012, 1031, 1034, 1094, 1111, 1149, 1159; SRVs, no. 27; Lundgren-Brate, pp. 264-265; Modeer, Personnamn i Ktb, pp. 8, 69, 77, 89, 102, 103; R. Dunås, 'Namn av typen Srestridh', Personnamn från medeltid och 1500-tal, ed. I. Modeer (Stockholm 1957), p. 29; Modeer, Östsmåländska bondenamn från 1500-talet, p. 59; Thors, p. 89; Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183, 185, 186, 187; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 145). ON PorbiQrn etc. is fairly frequent in English sources (NPN, pp. 147-148; PNDB, p. 392; Tengvik, p. 201; SPNT, p. 145; ELPN, pp. 82-83; OBS, p. 347, s.n. Thurban; SPNLY, p. 301; PN Db, p. 762; PN WRY, iv, p. 75). Note also the following Sf examples: Turebern 1186-1188 (c. 1270) BuryS, p. 19 (Rougham); Radulphusfilius Turben de Lafham 1190--1191 (c. 1270) BuryS, p. 57 (Cosford half-hundred; Lafham is Lavenham), Radulphus filius Turben ibid., p. 62 (Babergh double-hundred). ON PorbiQrn etc. was taken by Fabricius, p. 268, to be the first element of the Norman place-name Thouberville (Eure), but the evidence rather suggests that this place-name contains *Purbert (see above, s.n. Porbert and n. 37). For initial T(h)ur-, which formally reflects the OEScand (Danish) Pur-, see above, s.n. P6raldr.

40 Turber father of Geoffrey was already dead at the time of this document, which is a plea of mort d'ancestor. 41 Robert son of Thurbem also occurs as Robertus filius Thurberti 1225 CRR 89 (above, s. n. *Porbert, *Purbert). It is, therefore, not possible to decide whether the etymon is ON PorbiQrn etc. or the Scand-West Frankish hybrid *Porbert, *Purbert. See also above, s.n. *Porbert and n. 35.

400

l>orör, Thorth, Thordh For initial T(h)-for OScandP- see above, s.n. P6raldr. For the use of -bern for ON -biQrn etc. in English sources see above, s.n. AsbiQrn. Loss of final -n in Turber is a Romance feature (PNDB, § 78). Some of the post-Conquest examples of T(h)orbern(us), T(h)urbern(us) may ultimately go back to *Porbert, *Purbert, owing to AN confusion of -bert and -bern (see above, s.n. *Porbert, *Purbert and n. 35).

ON l>orör, ODan Thörth, OSwed Thördh a) (ii) Torddesmere c. 1215 Der, fo. 150b(i), Tordesmere ibid., fo. 241b, Thordesmere Edwl Rental (OKS) (Sandred, OUÅ, 1972, p. 45) (f. n. in Wiggenhall St Mary the Virgin). b) (i) Toreth. tennus TRE DB, fo. 250 (Wheatacre), Toret (bis) (nom., gen.) ibid., fo. 253 (Chedgrave), sub Toret ibid., fo. 253b (one liber homo held 30 acres at Carleton St Peter under Toret TRE). (ii) Tord (fiber homo) TRE DB, fo. 247b (Kerdiston), Tord. I. liber homo TRE DB, fo. 254 (Billingford near East Dereham). - Tort TRE DB, fo. 251b (Terrington). - Tort. liber homo TRE DB, fo. 256 (Babingley). - Thordo capellano, abl. (witn.) 1146 - c. 1150 Binham, fo. 123b(ii) (charter of Bishop William of Norwich concerning the church of St Andrew at Wood Dalling). - Thort Manet' 1162-1163 P, p. 29, Thord Manet' 1163-1164 P, p. 34, 1164--1165 P, p. 4, 1165-1166 P, p. 18, 1166--1167 P,42 p. 18, 1167-1168 P, p. 17, 1168-1169 P, p. 96, 1170--1171 P, p. 4, 1171-1172 P, p. 26, Thord' Manet' 1169-1170 P, p. 5, 1172-1173 P, p. 121, 1173-1174 P, p. 40, 1174-1175 P, p. 111 (Nfor Sf). -Walterusfilius Thord' 1174--1175 P, p. 121, de Walterofilio Thorö 1175-1176P, p. 67 (Nf or Sf). -Elfwjfilius Pord c. 1180 Wymondham, fo. 58b(i) (Wdecroft in Besthorpe).

ON P6rör etc. is a syncopated form of Porrrj}Ör < *PorfrrjJÖr (see NPN, pp. 149-150; Heusler, § 118; DR, cols. 728, 787 Anm. 3, 1009; Wessen, SRU, i, p. 181; NK VII, pp. 94, 95, 104, 190,195,258; Otterbjörk, Svenskaförnamn, p. 130, s.n. Tord; for further bibliographical references see NK VII, p. 173, n. 397).43 ON P6rör, ODan Thörth, OSwed Thördh is frequent throughout Scandinavia from the Viking period onwards (NiyR, nos. 26, 60, 126, 148,420; Lind, cols. 1152-1156; DR, nos. 84, 99, 156, 227, 264, 269, Mp. 112-113; DgP, cols. 1406-1414, 1695; Hald, Vore Stednavne, p. 135; DS XI, p. 156; DS XII,

42

TordMonet' CR.

The interpretation ofNoreen (ANF, vi (1890), p. 306 f.; AschwedGramm, § 250), that P6rlJr is to be derived from an original *P6rvgrlJr, *Pörw(jrlJR, must be rejected without reservation (see NPN, pp. 149-150; DR, col. 787 Anm. 3). 43

401

*l>orfr~Or, l>orfreOr pp. 85, 108; DS XVI, p. 106; SRSö, nos. 49, 175, 187; SRSmå, nos. 99, 126, 129; SRVg, nos. 23, 24, 25, 32, 131, 157,173,217; SRU, nos. 200,315,316, 678, 842, 864, 919, 922, 1070, 1107; Lundgren-Brate, pp. 265-266; Modeer, Personnamn i Ktb, pp. 20, 63, 64, 69, 120; Modeer, Östsmåländska bondenamn från 1500-talet, p. 64; Thors, pp. 89-90; Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183, 185; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 145, s.n. Tord). In Scandinavia itself the more conservative form Porr()Ör is known only from ON poetry (Lind, col. 1156). For the full form *Po,fr(Jör see below, s.n. In England forms representing both the syncopated P6rör and the more conservative Porr()Ör are on record (Crawf, pp. 148-149; NPN, pp. 148-149; PNDB, pp. 396--397; Tengvik, p. 201; SPNT, pp. 136, n. 2, 145; ELPN, p. 84; Newark, pp. 2, 14 (Tord); SPNLY p. 302; DEPN, p. 478, s.n. T01worth; PN D, p. 383; 44 PN Nt, p. 100; PN WRY, vii, p. 302). The DB forms in Toret(h) reflect the more conservative Porr()Ör (PNDB, p. 397). For initial T(h )- for OScand P- see above, s.n. P6raldr. The survival of OE, OScand P, here representing a pronunciation /8/, in the Nf form Elfwy filius Pord c. 1180 conforms to normale. ME orthographic usage (Jordan, §§ 16, 203). Final post-consonantal -t for OE [8], OScand [ö] in Thort is an AN spelling with the phonetic value [t], and final-din Pord (gen.), Thord, Thord' (nom.), and Thordo (abl.) is an inverted spelling for [t] (PNDB, § 107; SPNLY, § 116.iv; above, s.n. *Äsjr(Jör). Final-th in DB Toreth is an inverted spelling for -t (see PNDB, § 107; SPNLY, § 118.iv). Thordo in the Nf form Thordo capellano, abl. (witn.) 1146 - c. 1150 has been Latinized by the addition of the abl. ending of the Latin second declension (see PNDB, § 153; SPNLY, § 148; von Feilitzen, PNWinton, p. 227).

ON *l>orfr0ör, OSwed (run.) l>orfreör a) (i) Thuruertuna 1086 DB, fos. 151, 225b, 250b, Turuertuna ibid., fos. 141b (bis), 182b, 267b (Thurlton, Clavering H). b) (i) Thurveröus c. 980 (e. 13th) LibEl, pp. 114, 115 (Northwold and Pulham). DREFRD 1017-1023 SCBI, xv, no. 3224, DVREFERD 1017-1023 SCBI, xv, nos. 3225-3227, DVRFERD 1017-1023 SCBI, xv, no. 3228, 1023-1029 SCBI, xv, no. 3229, DVRVERD 1023-1029 SCBI, xv, no. 3154 (moneyer, Norwich), DVRFERD

44 PN D, pp. 383, 687, wrongly suggests that the place-name Drewstone in Bishop's Nympton D (Thordeston 1281 Ass (p)) contains AScandPorward, but the form of 1281 shows clearly that it belongs to ON P6rör etc.

402

*l>orfr0ör, l>orfreör 1023-1029 SCBI, xv, no. 3853 (moneyer, Thetford). - DVRFYRD 1048-1050 SCBI, xviii, no. 1143, DORORD 1050---1053 SCBI, xviii, no. 1141, DVRFIORD 1053-1056 SCBI, xviii, no. 1142 (moneyer, Norwich). - Toruert fiber homo TRE DB, fo. 256b (Ingoldisthorpe). -To,fortc. 1090---1100 CAcre, fo. lb (Nf). -Thuruerdus 1 lth-12th (contained in a list of benefactors to the Abbey of St Benet of Holme drawn up in the period 1186--1210) Holme, no. 62 (Nf). - Stigandus et Thuruertfilius eius 1126--1127 Holme, no. 76 (confirmation by Archbishop William of Canterbury ofthe grant to the Abbey of St Benet ofHolme by Stigand andhis son Thuruert, ofthe church of St Michael in Norwich). -Thuruerdus presbiter45 (witn.) 1146 Holme, 46 no. 143 (Saxlingham), Turuerhdus47 presbiter de Ludham (witn.) 1146 - c. 1150 BodlNfCh,48 no. 606 (printed, Holme, no. 141) (Fijbrege, Norwich), Turuerdo Decano, abl. (witn.) 1146-c. 1169 Bromh, fo. 12b(i) (the document concems the vicaria ofthe church of Witton and the churches of Keswick (Tunstead H), Paston and Dilham), Thome clerico de Ludhamfilio Thuruerdi, dat., 1153-1168 Holme, no. 95 (Confirmation by Bishop William of Norwich, of the grant to Thomas the clerk ofLudham, son ofThuruerd, ofthe church of Potter Heigham, on the presentation of Abbot William Il of St Benet of Holme).

The usual reflex of PrScand *PunrajriöuR in Scandinavia is the syncopated P6rör (q. v.). The full form is, however, represented by OSwed (run.) Porfreör (SRÖg, no. 235; SRU, no. 121). Note also the intermediate variant Porrt/)Ör, which is confined to ON poetic records in Scandinavia itself (see above, s.n. P6rör). In England the full form *Po,frt/)Ör etc. was borrowed into 1. OE to give AScand Pwferö, and is well attested in independent use and as the first element of place-names (Crawf, p. 75; NPN, pp. 155-156; Lindkvist, p. 171; PNDB, p. 392; Tengvik, p. 201; DBS, p. 351, s.n. Tollfree; SPNLY, pp. 303304; von Feilitzen, SCBI, xi.i, p. 13; Coinage of Edgar, p. 205; DEPN, pp. 468, 471, 472, s. nn, Thoroton, Thurlton, Thurvaston; PN D, p. 572; 49 PN Nth, p. 292; PN Nt, p. 229; PN Db, pp. 593-594; SSNEM, p. 195). A Sf example of the name is contained in the form Godric Turuerdifilius 1087-1098 (c. 11751200) Bury, p. 39 (Wattisfield). A late English example of ON *Po,frt/)ör etc. is contained in the form Johannes filius Puruerd 1222 ElyA, fo. 145b (Wisbech C). Forms in -ferö, -veröus, -uert, -uerd(us) reflect a Scand variant in -friör, while Toifort c. 1090-1100 stands for an original *Po,frt/Jör (PNDB, § 14; SPNLY, §§ 21, 43). For final-tand -d for OE [8], OScand [ö] in fort, -uert

45 This and the following Holme examples of T(h)uruerdus, Turuerhdus relate to the same person. His name is rendered in other Holme charters as T(h)urbertus and T(h)urbernus (see above, s.nn. *Porbert, PorbiQrn and notes 33, 38. 46 Wrongly dated by the editor of Holme to 1141-1146. 47 The Holme cartulary has Thuruerdus here (Holme, no. 141, n. 3). 48 Wrongly dated by the editor of Holme to 1141-1149. 49 PN D, p. 572, interprets Thorverton D as probably containing AScand Puifero, but Ekwall, DEPN, p. 470, s. n., took this place-name to be an OE Pornford-tun 'TON by a förd marked by a thorn-bush'.

403

l>orgautr, Thurgot, Thorgot

and-uerd(us) see above, s.n. P6rör. The use of -u-, -v- (= [v]) for OE, OScand -/- reflects the OE voicing of medial 1- between voiced sounds, and is in accordance with Latin, AN and ME orthographic usage (Jordan, §§ 17.2, 214; PNDB, §§ 84, 85, 87; Brunner, AbriB, § 36; SPNLY, § 96.ii). For initial Dur- and T(h)ur-, which formally reflect OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h)- for OScand P- see above, s.n. P6raldr. Initial D- for OScand p- may reflect 1. OE (southem) voicing of initial [8] or it may be merely an orthographic variant for P- (SPNL Y, § 110). In Normandy the name occurs in independent use as Twfredus, To,fredus and forms the first element of ten place-names (Adigard, pp. 156--157, 319, 320, 424-426).

ON l>orgautr, ODan Thurgot, OSwed Thorgot a) (i) Thurgotisshagh 13181pm (f. n. in Ketteringham). b) (i) Durgod 1066 SCBI, ii, no. 1211 (moneyer, (Thetford). -Lambertumfilium Turgod' 1205 CRR 40 (Pleas, iii, no. 1536) (Nt). - de feodo Thurgot c. 1250 Langley, fo. 45b (Heckingham).

ON Porgautr is frequent in Norway from c. 1300 onwards but is rare in Iceland, though it appears in Iceland already in the settlement period (Lind, cols. 1161-1163; LindS, col. 851). 50 Janzen, NK VII, p. 104, considered personal names in -gautr in OWScand to be largely borrowed from Sweden, though he added that in some cases they could also have been formed in the parts of south-eastem Norway bordering on Götaland. The name is frequent in ODan, where its first element often undergoes metathesis to give Thrugot, developing later to such forms as Trowcet (1424), Trowet (1450) etc. (DR, nos. 337, 354, Mp. 96--103; DgP, cols. 1426--1432, 1435-1436, 1695, s. nn. Thrugot, Thurgot; Kousgård Sprensen, Bebygg. på -sted, pp. 127-128, 242; Spndergaard, pp. 110-111; DS II, p. 59; DS IV, pp. 159, 339-340; DS V, p. 485; DS VI, pp. 195,480; DS IX, pp. 17, 140; DS XI, pp. 31, 131; DS XII, p. 85; DS XVIII.l, pp. 25, 140). In Sweden OSwed (run.) Porgautr is quite well attested and Thorgotus, Thorgut has been noted in medieval Swedish records, albeit extremely infrequently (SRÖg, nos. 22, 70, 181, 197; SRSö, nos. 111,268,336; SRSmå, nos.

50 Lind's earliest Norwegian example is dated to c. 1190, but the name only begins to be common in Norway a century or so later (Lind, Ioc. cit.). Note also the runic forrnjJ(o)r(g)autr fifil, nom., from Rauland, Buskerud Fylke, though this cannot be dated any more closely than to the period 1175-1350 (NlyR, no. 130).

404

l>orgeirr, Thorger, l>urgär 73, 89; SRVg, nos. 117, 128; SRU, nos. 13, 308, 646, 746, 958; LundgrenBrate, pp. 269-270). 51 ON Porgautr etc. is well attested in English records where it appears as Purgod, T(h)urgod, -got, T(h)orgot etc. (NPN, pp. 157-158; ZEN, p. 87; PNDB, p. 393; SPNT, p. 145 and n. 5; DBS, p. 348, s.n. Thurgood; SPNLY, pp. 304-305; Smart, Moneyers, p. 263; Fellows Jensen, Lincolnshire Tenants, p. 94; Insley, Nomina, iii (1979), p. 54; PN Nth, p. 291). Sf examples of the name are: Turgot liber homo TRE DB, fo. 378b (Tattingstone); Turgod 10871098 (c. 1175-1200) Bury, p. 39 (Wattisfield); ex parte Godwinifilii Turgod, Turgozus 1200 CRR 24 (Cur, i, p. 250)52 (Houghton in Cavendish). Cf. also Purgot mine cnihte, dat., 1043-1045 (1. 13th) ASWills, no. 31 (Ongar Ess; the document is the will ofl>urstan son of Wine). For initial Pur-, Dur- and T(h)ur-, which formally reflect OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h)- for OScand P- see above, s.n. P6raldr. For initial D- for OScandP- see above, s.n. *Porfrorgeirr, ODan, OSwed Thorger, AScand l>urgär b) (i) Durgartun 1044---1047 (1275-1302) KCD, no. 785,53 Turgartuna 1086 DB, fo. 185, Turgartunam ibid., fo. 216, Turgaitune ibid., fo. 179b, Turgarton' 1205 CRR 40 (Pleas, iii, no. 1596), 1206 FF (PRS NS, xxxii, nos. 104, 105) (Thurgarton, North Erpingham H; for further forms see Lindkvist, p. 94). b) (i) Turgar' de Boxton 1202 P, p. 113, 1203 P, p. 242 (? Buxton).

ON Porgeirr is well attested in Norway and Iceland from the Viking period onwards (Lind, cols. 1163-1166). In ODan the name appears as runic PurgiR (nom., acc.), and occurs in other

51 The account of the name given by Lundgren-Brate, pp. 269-270, is superseded by that given by R. Kinander, SRSmå, p. 190 (SRSmå, no. 73). 52 Turgod, Turgozus was the great-great-grandfather of one of the litigants in the present document and must, therefore, have lived in the middle or the latter part of the 11 th century. A genealogica! table illustrating the descent of the family is given by Selten, p. 42. 53 This charter is spurious (see Sawyer, no. 1055, for full bibliographical references). It is also printed as Holme, nos. 3, 4.

405

l>orgeirr, Thorger, l>urgär Danish records as Thorger(us), Thyrgerus etc. (DR, nos. 121, 328; DgP, cols. 1383-1384, 1441-1443, 1695, s. nn. Thorger, Thyrger). In Sweden OSwed (run.) PorgEiRR, PorgeR is well attested, and Thorgerus, Thyrgerus etc. occur in later Swedish records (SRÖg, nos. 131, 177, 233; SRSö, nos. 36, 342; SRU, nos. 116,126,509, 940; SRNä, no. 26; LundgrenBrate, pp. 267-268, s.n. Thorger; Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183 (Tyrger)). In England, where it is Anglicized as Purgar, with OScand -geirr having been replaced as second element by the corresponding OE -gar, the name is not especially frequent, though it occurs both independently and as the first element of place-names: NPN, p. 156; PNDB, p. 393; SPNT, p. 145; DBS, pp. 347-348, s.n. Thurgar; SPNLY, p. 305; DEPN, p. 471, s. nn. Thurgarton, Thurgoland; PN Nth, p. 291; PN Nt, p. 178; PN C, p. 349; PN WRY, i, pp. 154,314; PN WRY, ii, p. 231; PN We, ii, p. 314; SSNY, p. 106; SSNEM, pp. 195-196; Insley, JEPN, x (1977-1978), p. 69. Note also the following English examples: Purgar /Elfgares sunu (witn.) 1044---1048 (contemp.) ASChR, no. 103 (the document concems landat Oftham K), Purgar (witn.) 1048-1050 (contemp.) ASChR, no. 108 (Canterbury); assartum quod Turgarius tenuit Hy2 Stenton, First Century, p. 259 (Bk); Adelelm filius Turgari c. 1195 (13th) Rams, iii, p. 246 (Knapwell C), Willelmusfilius Turgari ibid., p. 251 (Elsworth C), Willelmusfilius Turgar' 1200 CRR 20 (Cur, i, p. 141) (assize of mort d'ancestor conceming property at Oakington C); Reginaldo filio Purgar, abl. (witn.) c. 1200 AOMB 31/94 (Fordingbridge Ha) (FMS); Osbertus filius Turgar' 1201 CRR 25 (Pleas, iii, no. 379) (Lincoln); 54 Simonfilius Turgari 1203 CRR 29 (Cur, ii, p. 167) (Street in Lympne K); de Torgero Pistore? 1206 Cur, iv, p. 101 (Lincoln); 55 Robertumfilium Turgar' 1206 CRR 42 (Pleas, iii, no. 1921) (Brk), Robertus filius Turgeri 1206 CRR 42 (Cur, iv, p. 186) (Bisham Brk).

For initial Pur- and T(h)ur-, which formally reflect OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h)- for OScand P- see above, s.n. P6raldr. For Anglicization through the replacement of OScand -geirr by the corresponding OE -gar as in Purgar, T(h)urgar(us) etc. see NPN, p. 203; PNDB, § 41; SPNLY § 52; above, s. nn. Alfgeirr, Äsgeirr, Styrger. For the replacement of OScand -geirr by -ger(us) see Fellows Jensen, NoB, lvii (1969), pp. 70--71, and above, s. nn. Alfgeirr, and n. 22, and Styrger. ON Porgeirr etc. appears in independent use in Normandy as Turgerius (PNDB, p. 393; Adigard, pp. 158, 320). In Ireland the name is recorded as Torgair, Torchair in an annal of 1171 (Marstrander, pp. 89, 117). For other forms relating to Osbert son ofTurgar see SPNLY, p. 305. This form occurs in a list of cash payments due from Lincoln which is stitched to CRR 37 and is in a hand which appears to belong to the same period as this roll (Cur, iv, p. 100, n. 1). 54 55

406

l>orgils, Thorgisl

ON l>orgils, ODan, OSwed Thorgisl b) (i) Turgis TRE DB, fo. 239b (held seven liberi homines in Brettenham). -Turgis. I. fiber homo 1086 DB, fo. 257b (Rudham). - Turgis 1086 DB, fo. 256, Apeltone quod tenet Turgisius 1101-1107 Binham fo. la (Appleton). -Turgisius de Eggemere (witn.) 1101-1119 (15th) NorwCPCh, no. 107 (charter ofBishop Herbert ofNorwich granting various properties and churches to Norwich Cathedral Priory), Turgidius 56 de Eggemere (witn.) ibid., no. 108 (charter ofBishop Herbert ofNorwich granting churches at Lynn, Yarmouth, Aldeby, North Elmham and Hemsby, with land and appurtenances, to Norwich Cathedral Priory; Eggemere is Egmere in Little Walsingham), Turgisius dispensator (witn.) c. 1110--1119 (15th) BactonCh, no. I (grant by William of Bacton Sf to Richard the archdeacon ((of Sf); Turgisius the dispenser is taken by B. Dodwell, BactonCh, p. 148, n. 1, to be probably identical with the Turgis of Egmere of NorwCPCh, nos. 107, 108), Turgisio, abl. (witn.) c. 1110--1116 (15th) NorwCPCh, no. 109 (Gnatingdon in Ringstead), Turgiso, abl. (witn.) 1121-1127 (15th) NorwCPCh, no. 12 (Eaton in Norwich). - Radulfus filius Turgisi (witn.) c. 1145 Binham, fo. 49b (printed, Stenton, First Century, p. 261) (Barney and Thursford). -De Turgis' 1165-1166 P, p. 33 (Nf or St). -Turgis c. II90--l200Binham, fo. 5b (printed, SSMEA, Appendix Il, no. 3, p. 269) (Binham). -Turgisius de Wiggehal' 1195 P, p. 73, 1197 P, p. 235, Turgisus de Wigehale 1196 CR, p. 133 (Wiggenhall), terram Turgis de Wigenale 1198 FF (PRS NS, xxvii, no. 155), terram que fuit Turgis de Wigenale c. 1200 Der, fo. 173a (Tilney; Wigenale is Wiggenhall), Willelmus filius Turgisii 1199 CRR 51 (Cur, i, p. 70) (Nf), Willelmum filium Turgisii de Wigehal' ibid. (Cur, i, p. 75), Willelmumfilium Turgisii ibid. (Cur, i, p. 81), Willelmumfilium Turgis 1199 FF (PRS NS, xxvii, no. 226) (Wiggenhall), Willelmumfilium Turgis' 1199 CRR 11 (CurR, i, p. 264 (Nf), Carta Willelmifilii Turgis de Wigehale et Ade filii eiusdem Turgis de Wigehal' de hornagio Iohannis filii Wlfrici, Willelmus filius Turgis de Wigehale, Adamfilius eiusdem Turgis de Wigehal' c. 1200 Der, fo. 189b(ii) (this document concems the homage of John son ofWulfric ofWiggenhall and the tenement which he held in Tilney; Wigehale, Wigehal' is Wiggenhall). -Turgis (acc.) 1198 CRR 8A (CurR, i, p. 139) (Nf). - Turgis de Pastan' 1209 CRR 49 (Pleas, iv, no. 3743) (Paston). - Henrici filij Thurgys de Caudecotes patris mei, gen. (charter of Henry of Caldecote) c. 1218-1219 Holme, no. 271, Henricus filius Thurgys de Caudecotes, Henricus filius Thurgys ibid., no. 272, Henricum filium Turgis de Caldecotes, Henricus filius Turgis', Henricus filius predicti Henrici filii Turgys 1219 FF, no. 40 (Hardley; Caudecotes, Caldecotes is Caldecote). - terram Turgis filii Aldithe 1247 Langley, fo. 19a (Erodeswelle near Loddon).

ON Porgils < Porgfsl is common in both Norway and Iceland from the Viking period onwards (Lind, cols. 1168-1172). In Danish the name is extremely frequent, being represented by runic Purkisl (nom., acc., gen.), Purils (nom., acc.), and by T(h)urgils, T(h)urgillus, Thrugillus, Troels etc. in other records, in addition to forming the first element of several place-names (DR, nos. 53, 58, 149, 229, 294, 314, 370; DgP, cols. 1384--1385, 1418-1425, s. nn. Thorgisl, Thrugils; Hald, Vore Stednavne, pp. 110, 135; Hjorth Pedersen, Bebygg. på -by, p. 37; DS I, p. 69; DS III, pp. 16, 126; DS IV, p. 89; DS V, pp. 32,541; DS VI, p. 527; DS IX, pp. 83-84, 113; 56

V ar: Turgisius (e. 14th).

407

l>orgils, Thorgisl DS XI, p. 219; DS XII, p. 145; DS XIV, pp. 51, 60, 293; DS XVIII.I, p. 78; Sydsl. Stedn. VII.l, p. 41; SOH I, p. 38). In Sweden OSwed (run.) Porglsl is well attested and the name is frequent in later records, where it appears as Thorgillus, Thyrghils etc. (SRÖg, no. 28; SRSö, nos. 64,287,336,382; SRSmå, no. 145; SRVg, nos. 79, 87, 97; SRU, nos.47,99, 124,144,481,663,774, 786,836,899;Lundgren-Brate,pp.268269; Modeer, Personnamn i Ktb, pp. 11, 89, 119, 127; Modeer, Fornsvenska personnamn i en regestsamling, pp. 15-16; Modeer, Östsmåländska bondenamn från 1500-talet, p. 60; Thors. pp. 90--91; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 147, s.n. Truts). In England the name is well attested, though in the post-Conquest period it normally appears in the Norman forms Turgis, Turgis(i)us etc. (NPN, pp. 156-157; ZEN, pp. 86-87; PNDB, p. 393; Tengvik, pp. 201-202; SPNT, p. 145;57 DBS, p. 337, s.n. Sturge; SPNLY, pp. 305-306; Fellows Jensen, Lincolnshire Tenants, p. 87; DEPN, p. 466, s.n. Thirkleby YE; PN ERY, pp. 125126; SSNY, p. 39). The original Scand Porgisl appears as Duregisel 10651098 (contemp.) ASChR, Appendix II, no. 7 (? Bury St Edmunds Sf). 58 Note also the following Sf examples of the Norman form Turgis: De terra Turgis 1186-1188 (c. 1270) BurgS, p. 5, Filii Turgis ibid., p. 16 (Felsham); Brichwi filius Turgis ibid., pp. 50, 51 (Hopton); Robert filius Turgis de Brethenham 1190--1191 (c. 1270) BuryS, p. 58, Robertfilius Turgis ibid., p. 59 (Brettenham); Robertusfilius Turgis c. 1230 (c. 1270) BuryS, fo. 106a (Felsham). Turgis bedellus, Turgis ibid., fo. 119b (the soke ofRedgrave, prob. Yaxley). In Normandy the name occurs in independent use as Turgis, Turgis(i)us, Torguis etc., and forms the first element of four place-names, including one in OEScand -/Jorp (Adigard, pp. 159-160, 321-322, 427). Turgis etc. is remarkable for the persistency of its use in Normandy, where its latest appearance as a personal name belongs to the period 1416-1423 (Adigard, pp. 159-160, n. 66). The replacement of OScand -g{sl, -gils in the second element by -gis, -gis(i)us in these Norman forms is the result of analogy with such West Frankish names as Ansegis(us), Winegis(us) etc. (see PNDB, p. 393, n. 4; SPNLY, § 70). The predominance of the Norman forms in post-Conquest English records is noteworthy and it is probable that many of the post-Conquest bear-

57 SPNT, p. 145, interprets the Liber Vitae of Thomey form, Tur3ysle. as reflecting a weak variant, ON *Porgisli. 58 The original Scand Porgisl is also clearly manifested in the early forms of the place-name Thirkleby ERY (Turgislebi 1086, -by 1109-1119, 1193-1205, 1228, Turgesleby c. 1225, 1228, Turgilebi 1086), and it has been noted in independent use in DB TRE in Db and L as Turgisle (cf., however, above, n. 57), and in post-Conquest Y in the forms Johanne f Turgisili, abl. (witn.) 1163-1169 (I. 12th) YCh, Turgisillo monacho de Sancta Trinitate, abl. (witn.) 1170--1183 (m. 13th) YCh, Thomaf Turgisel, abl. (witn.) 1185-1205 (14th-15th) YCh (PNDB, p. 393; SPNLY, pp. 305-306; PN ERY, pp. 125-126).

408

l>orgrimr, Thorgrim ers of the name in England were Normans or of Norman descent. A clear case of a Norman bearer of the name in England is provided by Turgis d'Avranches, who was appointed castellan of Saffron Walden castle Ess by Stephen in 1144.59

ON l>orgrimr, ODan, OSwed Thorgrim b) (i) Turgrim TRE DB, fo. 170 (Wiveton). - Dvregrim Wml (c. 1068-1070) SCBI, xxi, no. 1165, Wml (c. 1070-1072) SCBI, xii, no. 49 (moneyer, Norwich). -terramAylwardi Thurgrim 1214-1229 Holme, fo. 85b(iii) (? South Walsham).

ON Porgrfmr is frequent in Norway and Iceland from the Viking period onwards (Lind, cols. 1173-1174; LindS, cols. 855-863). In the OEScand area the name is rare. In the ODan area it is represented by a single form, Thordgrim Gandce, which occurs in Halland in the Liber Census Danice (DgP, col. 1385). In Sweden PorgrimR occurs in three runic inscriptions, and Thorgri(i)m is on record in the 14th century in Dalsland, a province which, it should be noted, borders directly on Norway (SRSö, no. 124; SRU, nos. 180, 192; Lundgren-Brate, p. 269). The distribution of Porgrfmr in Scandinavia shows it to have been a characteristically OWScand name. It might well be the case that the isolated examples of the name found in the OEScand area were the names of Norwegians or men of Norwegian descent, though on förmal grounds the existence of an OEScand counterpart of ON Porgrfmr cannot be excluded. In England Porgrfmr is not frequent, though it is attested both in independent use and as the first element of place-names (NPN, p. 158; ZEN, p. 87; PNDB, p. 393; SPNLY, pp. 306--307; Smart, Moneyers, p. 230; von Feilitzen, SCBI, xi.i, p. 13; Smart, Cnut's York Moneyers, p. 228; DEPN, pp. 466,468, s. nn. Thorganby, Thornthorpe; PN ERY, pp. 143, 263-264: PN WRY, ii, p. 72; SSNY, p. 68; SSNEM, p. 74). Interestingly, runic purkrim, nom., appears in an inscription probably from the first part of the 11th century, which was found in Ireland at Killaloe, County Clare (RIGB, p. 181; Musset, Introduction a la runologie, p. 442, no. 161). For initial Dvr- and T(h)ur-, which formally reflect OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial D- for OScand P- see above, s.n. *Porjr(Jör. For initial T(h)- for OScand P- see above, s.n. P6raldr.

59

R.H.C. Davis, King Stephen (London 1967), pp. 85-86 and n. 23.

409

~orgunnr,Thrugun,~orgunn

ON, ODan (run.) l>orgunnr, ODan Thrugun, OSwed (run.) l>orgunn fem. b) (i) Petrum molendinarium et Turgund' uxorem suam 1202 CRR 28 (Pleas, ii, no. 1032) (Nt), Petrum molendinarium. et Thurgund' uxorem suam, predicti Petrus et Thuregunde, predictis Petro et Thuregunde, dat., heredibus ipsius Thurgunde 1202 FF (PRS NS, xxxii, no. 5) (Mundford).

ON Porgunnr is late and rare and is confined to Norway (Lind, col. 1175). A weak variant, ON Porgunna, is rather more common and has been noted in both Norway and Iceland (Lind, cols. 1174-1175). In Danish ODan (run.) Porgunnr is represented by purkuni, dat., in a runic inscription of the Viking period (DR Per. 2.2), and a form with metathesis, Thrugun, is quite well attested in medieval Danish records (DR, no. 40; DgP, cols. 1385-1386, 1432-1433, s. nn. Thorgun, Thrugun). A weak variant, Thorgunna, is also known in Danish (DgP, col. 1386). In Sweden there area few examples of OSwed (run.) Porgunn(r), Porguör, and the name also forms the first element of a place-name in Dalsland (SRÖl, no. 17; SRÖg, no. 29; SRSö, nos. 73, 288; SRU, nos. 459, 789; LundgrenBrate, p. 269). In England the name is only found sporadically (SPNT, pp. 149-150; ELPN, p. 84). For initial T(h)ur-, which formally reflects OEScand (Danish) Pur-, see above, s.n. P6raldr). For initial T(h)- for OScand P- see above, s.n. P6raldr. In the above Nf forms OScand -gunnr has been replaced in the second element by -gund', -gunde as a result of association with the West Frankish fem. name element -gund, -gundis.

ON l>orhildr, ODan, OSwed (p. n.) Thörhild fem. b) (i) Henricum nepotem Turild 1198 FF (PRS NS, xxvii, no. 10) (Terrington). -terram Bartholomei filii Thurild' c. 1200 Wals, fo. 76b(ii), terram Bartholomei filii Thurild ibid., fo. 78b(iii) (North Barsham). - Thurildam, Thurilda 1224 CRR 86 (Cur, xi, p. 309), Thurildam 1224 CRR 85 (Cur, xi, p. 413), Ricardum filium Godelive et Thurildam60 sororem eius 1225 CRR 90 (Cur, xii, p. 125), Ricardum filium 61 Godlive et Thurildam 62 sororem ejus 1225 CRR 88 (Cur, xii, p. 271) (Brisley).

ON P6rhildr is quite well attested in Iceland from the time of the settlement until the middle of the 13th century, but it is late and rare in Norwegian Turillam CRR 91. filium is omitted in CRR 92 here. 62 Turildam CRR 92. 60 61

410

l>orir, Thörir, Thörer, Thöri, Thöre records, although its early appearance in lceland would seem to indicate that it was also in use in Norway before the settlement of Iceland (Lind, col. 1178; LindS, col. 864). ODan Thörhild is only known as the name of a figure of saga found in medieval chronicles and regnal lists (DgP, col. 1386). In Sweden the name occurs as the first element of a place-name (NK VII, p. 255). Runic jJorilter, nom., occurs in a recently discovered late 13th century inscription from Lödöse in Älvsborgs Län, Västergötland, but Elisabeth Svärdström regards it as a clear expression of Lödöse's close cultural connections with Norway (E. Svärdström, Fornvännen, lxvii (1972), pp. 261,262). In England ON P6rhildr etc. is relatively infrequent (NPN, p. 158; PNDB, p. 393; DBS, p. 358, s.n. Turrill; SPNLY, p. 307; PN WRY, ii, p. 247). For initial T(h)ur-, which formally reflects OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h)- for OScand P- see above, s.n. P6raldr. For the loss of -h- in OScand -hildr in the above Nf forms see above, s.n. Gunnhildr. The forms Thurilda, Thurildam have been Latinized by the addition of the endings of the Latin first declension (see PNDB, § 157; SPNLY, § 152; above, s.n. Gunnhildr).

ON l>orir, ODan Thörir, OSwed Thörer, ODan Thöri, OSwed Thöre a) (i) in terram que dicitur Thori in uilla de Kemest' c. 1220 CAcre, fo. 33a(ii) (f. n. (?), in Kempston). b) (i) Thori pistore, abl. (witn.) c. 1110 CAcre, fo. 59b(i) (grant of the church of Fulmodeston to Castle Acre Priory). - Willelmo filio Thori, abl. (witn.) 1164 - c. 1185 CAcre, fo. 6a(iv) (grant by Earl Hamelin de Warenne of tenementumRadulfifilii Athelwoldi de Burneham to Castle Acre Priory), c. 1175-1180 CAcre, fo. 46b(iii) (confirmation by Thomas de Bellofago of the church of South Creake to Castle Acre Priory), 1182 CAcre, fo. 81a(ii) (Nf), W.filius Thori (witn.) c. 1180 CAcre, fo. 9a(iv) (Puttockeshe and Longediche in (Castle) Acre). -Thori de Sacrist' (witn.) c. 1165-1178 CAcre, fo. 87a(i) (charter ofRobert of Watlington to Castle Acre Priory conceming terram quam

Ulfketel Stoche tenuit. Et illam quam Ricardus filius Gode tenet. Et terram quam Alfricus ruffus tenet, properties in Watlington and West Dereham, and properties and (servile) tenants in Wretton). - Henrico filio Thori, abl. (witn.) c. 1175 CAcre, fo. 28a(iii) (Hungerehille in South Acre ). - contra Thori Golde c. 1175-1190 CAcre, fo. 28a(iv, v) (? South Acre).

The name complex Pörir/Pöri has been the subject of much disci.Ission. Pörir is to be interpreted as belonging to an earlier *Pörver < PrScand *PunrawzhaR, although it has been suggested that it can also be an -ir derivative of the 411

l>orir, Thörir, Thörer, Thöri, Thöre

simplex P6rr, 63 while Pöri is either a secondary derivative of Pörir, which has undergone a change from the ia- stem to an an- stem, or is a short form of names in Pör- (Noreen, AislGramm, § 235.lf; Heusler, § 117.5; BrpndumNielsen, § 435.2.3; DgP, cols. 1387, 1388; NK VII, pp. 95, 191, 208, 241, 245; SPNLY, p. 309). 64 Interestingly, Pöri is confined to the OEScand area, whereas Pörir occurs throughout Scandinavia. ON P6rir is one of the most common personal names found in Norway and Iceland between the Viking period and the end of the Middle Ages (NiyR, nos. 42, 110, 225, 293, 351, 418, 449, 547, 590; Lind, cols. 1180--1182). In Danish Thörir and its side-form Thurir are rare, though it is interesting to note that the name forms the first element of no less than six place-names in -lef, as well as occurring independently and as the first element of place-names in -thorp (DR, nos. 109, 114, 117, 122, 127; DgP, cols. 1387-1388, 1440; Spndergaard, pp. 110, 111-112; Hald, Personnavne i Danmark: I. Oldtiden, pp. 44---47; DS Il, pp. 109, 134; DS XI, pp. 200, 210). In Sweden OSwed (run.) PöriR is frequent and Thorerus, Thorirus etc. are well attested in other Swedish records (SRÖl, no. 46; SRÖg, nos. 11, 27, 85,223; SRSö, nos. 31, 35, 161, 232,233,237,246,290,301,323; SRVg,nos. 11, 13, 15,50, 78,100,156, 169; SRU, nos. 104, 163, 199, 235, 283, 293, 429, 726, 908, 934, 996, 1143, 1144, 1176; Lundgren-Brate, pp. 266-267, 276-277, s. nn. Thorer, Thurer; Modeer, Personnamn i Ktb, pp. 10, 118, 120, 128; Modeer, Fornsvenska personnamn i en regestsamling, pp. 13, 14, 17; Modeer, SvP, p. 100 (Thorirus Piltir 1347)).65 ODan Thöri and its side-form Thuri are extremely frequent in medieval Danish records and are also known from Danish place-names (DgP, cols. 1386-1387, 1436-1439, 1695; Hjorth Pedersen, Bebygg. på -by, pp. 36-37; DS XI, pp. 140--141; DS XII, p. 115; DS XVI, p. 118). OSwed Thöre and its side-form Thure are also well attested (SRÖg, nos. 123, 184; SRVg, nos. 112, 114, 160; SRU, no. 275; Lundgren-Brate, pp. 266, 276; Modeer, Personnamn i Ktb, pp. 10, 20, 26; Modeer, Östsmåländska bondenamn från 1500-talet, p. 60 (Tore, Ture); Thors, p. 90; Modeer, SvP, p. 101 (Thore Neb 1431); Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183, 185, 186, 187 (Tore); Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 145-146, s.n. Tore). 63 NK VII, pp. 95, 118; SPNLY, p. 309. The absence of the simplex Porr as a personal name in Scandinavia (cf. above, s. n. Por, Pur) allows one to reject this without reservation. 64 DgP, col. 1388, and R. Homby, NK VII, p. 208, and NoB, xxxvi (1948), p. 42, also include the possibility that Porgeirr is an alternative to *Porver for the base of Porir, but Janzen, NK VII, p. 95, shows that this view is untenable. 65 Modeer, SvP, p. 60, considered the Latinized Thorerus in medieval Swedish records to be an archaism for Thore, but cf. also Modeer, Personnamn i Ktb, p. 120.

412

l>orir, Thörir, Thörer, Thöri, Thöre In England the OEScand Pori, Puri is well attested in independent use as Puri, T(h)ori, T(h)uri etc., while ON P6rir, OEScand Porir, Purir is known from several place-names, including Thoresby L (2), Nt, and NRY, and Thoresthorpe L, and occurs in independent use in the Liber Vitae of Thomey Abbey as Dorir, and has also been taken to form the first element of a considerable number of field and minor names: NPN, p. 158; ZEN, p. 87; PNDB, pp. 393-394; Tengvik, p. 202; SPNT, p. 145; ELPN, pp. 83, 129; ASWrits, pp. 423-424, 574; DBS, p. 347, s.n. Thory; SPNLY, pp. 307-309, 356; DEPN, pp. 466, 472, 483, s. nn. Thoresby, Thoresthorpe, Thoresway, Thursby, Thurston, Turton; PN La, p. 47; PN D, p. 325; PN Nt, pp. 92-93, 275,287; PN Cu, p. 154; PN WRY, i, p. 142; PN WRY, v, pp. 22, 126; PN WRY, vi, pp. 52-53, 88, 156; PN Gl, i, p. 66; PN We, ii, pp. 117-118, 314; PN Brk, p. 384; SSNY, p. 39; SSNEM, pp. 74,118,222; lnsley, JEPN, x (1977-1978), p. 62. 66

In Sf the OEScand Puri occurs in DB TRE as Thuri, Thure, Turius (PNDB, pp. 393-394). A post-Conquest Sf example is contained in the form Godric Turifilius 1087-1098 (c. 1175-1200) Bury, p. 27 (Pakenham). A late English example of OEScand Puri is contained in the form Walterus filius Puri 1222 ElyA, fo. 142a (Hatfield Hrt). SPNT, p. 145, n. 3, cites a document in which the form Turi is used to render the name of Archbishop Thurstan of York (1114--1140), and argues from this that the English form Thuri would appear to have been used sometimes as a short form of compounds in Thur-. It is, however, unsafe to argue from the case of a single individual, especially when, as in this case, the person in question was a Norman, 67 and there can be no doubt that English forms in T(h)uri normally stand for OEScand Puri. For initial T(h)- for OScand P- see above, s.n. P6raldr.

66

PN NRY, p. 266, wrongly gives the etymon of the first element of Thoresby NRY as 'ON

Porr' (cf. SSNY, p. 39), and PN Nt, p. 93, allows for AScand Pör, Pur as an alternative to OEScand Purir for the first element ofThoresby Nt. Similarly, PN Cu, p. 154, takes Thursby Cu, which is taken by DEPN, p. 472, s. n., to have ON Porir etc. as first element, to contain OEScand Pöri, Puri. On the other hand, PN La, p. 47, wrongly interprets Thurton Nf as containing AScand Pör, Pur or OEScand Pöri, Puri, but, as is shown by DEPN, p. 472, s. n., the first element of this place-name is OE ]Jyrne f. 'a thom-bush'. Similarly, PN Ess, p. 353, wrongly considers OEScand

Puri to be a probable first element for Thonington Ess, which is correctly interpreted by DEPN, p. 470, s. n., as containing either OE ]Jorn m. 'a thom, a thom-bush' or OE ]Jyrne. It is normally not possible to separate ON Porir etc. from OEScand Pöri, Puri and AScand Pör, Pur in field and minor names in ME, cf. PN D, p. 325, and the field and minor names given by PN NRY, p. 333, as containing 'ON Porr' (i. e. AScand Pör, Pur), but which are grouped by SPNLY, p. 308, sub Porir. The inclusion by SPNLY, p. 309, of the name of the god Porr as a further possible first element in this place-name/field name group can be rejected without reservation. Some of the ME field and minor names in Turse-, Thurs- may altematively contain ON ]Jurs m. 'a giant' (see EPN Il, p. 217, s. n. ]Jurs; SPNLY, p. 309). 67 Archbishop Thurstan of York was a native of Bayeux (SPNLY, p. 315).

413

l>orke(ti)ll, Thorkil, Thurkil, Thyrkil, Th0rkil

ON l>orke(ti)ll, ODan Thorkil, Thurkil, Thyrkil, Th0rkil, OSwed (run.) l>orkretill, l>orkell, l>orkil, OSwed Thorkil, Th0rkil a) (i) Thurketeliart 1086 DB, fo. 230, frutecta de Turketelgiyard 68 ll01-ll33 69 (15th) NorwCPCh, no. 20 (a lost vill in Aldeby, Clavering H). -Thurketeliscroft c. 1210 Der, fo. 82a(ii) (f. n. in Upwell). -Turkildesslade 1318 Bromh, fo. 90b (f. n. in Witton or Bacton or Paston). b) (i) Durkilles here 1009 (m. 11th) ASC (C), 70 Purcyl, Purkyl 1013 (m. 11th) ASC (C), Purkil, Purcyl 1013 (1121) ASC (E), Purkylle, dat., 1017 (m. 11th) ASC (C), ASC (D), Purcylle, dat., 1017 (1121) ASC (E), on Purkilles eorles (gewitnysse) 1017-1020 (l. 11th) ASWrits, no. 26 (writ of Cnut declaring that he has confirmed the liberties of Christ Church, Canterbury), Purkil eorl 1020 (m. 11th) ASC (C), Purkyl eorl 1020 (m. 11th) ASC (D), Porkyll eorl (acc.) 1021 (m. 11th) ASC (D), Purkil eorl (acc.) 1021 (1121) ASC (E), Purcil, Purcille, dat., Purciles sunu 1023 (m. 11th) ASC (C) (these entries all refer to the famous Viking Thorkel the Tall, who was Cnut's earl of East Anglia from 1017 to 1021).71 - Turkillo (father of Aelfwen), abl., 1014 (c. 1100) UANG, p. 243 (Nf), Thurketel, mid Thurkitel seluen c. 1020--1050 (1. 13th) ASWills, no. 25 (in this document, Thurketel, who bore the byname Heyng, 72 bequeathes property at Caister, Thorpe, 73 Ormesby and Scratby, and makes bequests of cash and personal freedom). - Purketel (bis), Purfketel (sic) (nephew of the preceding) (dat.) 1020--1040 (1. 13th) ASWills, no. 2474 (in this document, J:>urketel bequeathes properties at Roydon (near Diss) and Shimpling Nf, and at Palgrave, Whittingham, Wingfield and Thrandeston Sf; his nephews Vlfketel and J:>urfketel received an estate at Wingfield Sf). - (P)urkil m. 11th (1. 13th) ASWills, no. 36 (bequest by J:>urkiF 5 and

68

Var: Turketelisiard (e. 14th), Turktelisiard (e. 14th), Thurketelgiard (I. 13th), Turketelgiard

(I. 13th).

Probably in the later part of this period (NorwCPCh, loc. cit.). The ASC (C) entries from 1009 to 1017 are taken from Ashdown. 71 For Thorkel the Tall see Ashdown, pp. 293-294, and ASWrits, p. 574. Note that the entries in the present work do not relate specifically to Nfbut to Thorkel's activities as Viking leader and as Cnut' s earl and follower. 72 Thurketel's byname Heyng, which appears in the Latin rubric of this document, belongs to ON hceingr m. 'a male salmon' (ASWills, p. 180; cf. also LindBiN, col. 169; Tengvik, p. 363). Thurketel 's family shows an interesting mixture of OE and OScand personal nomenclature. His daughter bore the name Alfwen (OE /Elfwynn), and his nephew the name Ketel (ON Ketill etc.). His great-nephews were the sons of Sweg(e)n (ON Sveinn etc.) and Alemund (OE (Angl) Alhmund). 73 ? Thorpe Abbots or Momingthorpe (see ASWills, p. 180). 74 Sawyer, no. 1527, dates the present document to the 11th century, 'probably before 1038'. However, since l>urketel bequeathes property to St Edmund, it must be dated to after the foundation ofthe Abbey ofBury St Edmunds in 1020. The date of 1013 cited by Hart, ECEE, p. 66, no. 90, from a Bury list ofbenefactors is, thus, too early. The upper limit of 1040 is determined by the fäet that l>urketel bequeathed half of Roydon Nf to his wife Lefquen, who is probably identical with the Lefwenn, who is recorded in the will of Bishop JElfric, a contemporary document of 1035-1040 (ASWills, no. 26; Sawyer, no. 1489), as having sold landat Roydon to this bishop. The confusion between Lefquen (OE Leofcwen) and Lefwenn (OE Leofwynn) is probably the result of an error on the part of the late 13th century copyist of ASWills, no. 24. 75 For the possibility that this Purkil may be identical with the DB tenant Turchillus who held 69

70

414

l>orke(ti)ll, Thorkil, Thurkil, Thyrkil, Th~rkil Apelgit7 6 to the Abbey of Bury St Edmunds at Wereham). - Turchillus. I. liber homo TRE DB, fo. 230b (Barton Bendish), Turchillus ibid., fos. 230b, 231 (bis) (Crimplesham, Boughton, Shouldham).77 - Turchillus TRE DB, fos. 149, 222 (bis), 223b (East Walton, Middleton, (North) Runcton, Gayton, (North) Creake), Turchillus liber homo ibid., fos. 222, 274b (lslington, Gayton Thorpe). - Turkil haco TRE DB, fo. 223b (Beeston Regis). - Turchillus TRE DB, fo. 245 (Oxborough). 78 - Turchillus TRE DB, fo. 267 (Attleborough). - xx. acras tenuit Turchetel ad socham sancte Adeld' TRE DB, fo. 205b (Marham), Turchetel fiber homo ibid., fos. 205b, 206b, 207 (ter), 207b (Barton Bendish, Terrington, (North) Runcton, Great Ellingham, Litcham, Whinburgh), Turchetel ibid., fos. 206 (ter), 207, 208 (Wormegay, West Briggs, Thorpland (near Downham Market), Stow Bardolph, Islington, (East) Tuddenham). - Turchetel TRE DB, fo. 233 (Wiveton). - Turketel TRE DB, fo. 258 (Bamey). - Turchetel. liber homo TRE DB, fo. fo. 266b (lngoldisthorpe). -Turkellus Rufus (witn.) 1108 Binham, fo. 20b (printed, SSMEA, Appendix I, no. 35) (Binham and Langham). - Turkillus avunculus ejusdem Rogeri (i.e. Rogerus filius Mowini) 1114-1130 (m. 14th) Rams, ii, p. 260, no. 372, terram Thurkilli de Hirst, quam habuit in Fordham de feudo Abbatis 1133-1151 (m. 14th) Rams, ii, p. 267, no. 384 (Fordham; Hirst is unidentified). - xxli acras terre quas tenebat Turchillus rufus c. 1115 CAcre, fo. 26b(iii) (Sporle). -Bondo; Torkilfilius ejus (witn.) c. 1120 [before 1122] (12th) France, p. 414, no. 1149 (Sporle andMileham). - Turchill' capellanus episcopi (witn.) 1121-1135 BactonCh, no. 2 (charter of Bishop Everard ofNorwich), Turchillus capellanus (witn.) 1136--1143 NorwCPCh, no. 117 (charter of Bishop Everard of Norwich to Norwich Cathedral Priory). -Turchill' presbiter de Swofham, Achi prepositus et Turchillusf(rater) eius (witn.) c. 1130 CAcre, fo. 103b(i) (grant by Alanus de Swofham to the monks of Acre; Swofham is Swaffham Nf). - Warino presbitero et Turchill' filio eius, abl. (witn.) c. 1135-1140 CAcre, fo. 16a(vi) (Massingham). -Turchilli, gen., c. 1140 CAcre, fo. 26a, terram Turchilli c. 1140 CAcre, fo. 33b(iii), 1146--1147 CAcre, fo. 3b(i) (Kempston). -Turchillus de Binham (witn.) c. 1150 Binham, fo. 168a (Great Ryburgh; Binham is Binham). - Turkillus de Ingham (witn.) 1153-1168, prob. 1162-1168, Holme, no. 8779 (confirmation to the Abbey of St Benet of Holme of a third of the tithes at lngham and W axham, two thirds of the tithes at Worstead and Brunstead, anda half of the tithes at Horsey; Ingham is Ingham). -(terram) Walterifilii Turchil (bis) Hy2 HarlCh 50 C 14 (printed, SSMEA, Appendix I, no. 8) (? Brockdish and Shimpling). - domum Turkilli Hy2 (17th) BS, no. 330 (Ketteringham). - Ricardo filio Turkil, abl. (witn.) c. 1160--1170 Binham, fo. 173b (printed, SSMEA, Appendix I, no. 37) (grant by Adam filius Alueredi, grandson of Peter I de V aloignes and nephew of Roger de Valoignes, of properties in Ingoldisthorpe to Bin-

TRE at Barton Bendish, Crimplesham, Boughton and Shouldham (below and notes 77, 78) see ASWills, pp. 205-206. 76

OE/Eöelgyö.

For the possibility that this Turchillus is identical with the Purkil of ASWills, no. 36, see above, n. 75, and the reference given there. 78 Turchillus of Oxborough may be identical with the Turchillus of Barton, Bendish, Crimplesham, Boughton and Shouldham (and, therefore, perhaps also with the Purkil of ASWills, no. 36, cf. above, n. 77). Oxborough lies close to Barton Bendish etc., but in 1086 the property of Turchillus of Oxborough belonged to a different fee to that of Turchillus of Barton Bendish etc., and, therefore, it is perhaps best to leave the question of identity open. 79 The witness Turkil of Ingham does not occur in the copy of this charter contained in the Holme cartulary itself, but has been supplied by the editor of Holme from the copies contained in the British Library Cotton Rolls and the Bodleian Rolls. 77

415

I>orke(ti)ll, Thorkil, Thurkil, Thyrkil, Th0rkil ham Priory), Ricardo filio Turchil. et Milone filio eius, abl. (witn.) c. 1175 Binham, fo. 4a (charter of confirmation granted by Robert de V aloignes to Binham Priory), Ricardus filius Turchilli. et Milo filius eius (witn.) c. 1175-1180 Binham, fo. 123a(i) (Wood Dalling, Little Ryburgh, Brittrichescrof). - Turchetillo filio Eilwine, Turchil filio Fabri, Iohanne filio Turchilli, abl. (witn.) c. 1160 CAcre, fo. 16b(ii), terram que fuit Iohannis filii Turchill' in meduecroft c. 1200 CAcre, fos. 18b, 19b(i) (Massingham). -Thurkillus 1161-1177 (13th) Rams, iii, p. 288 (Hilgay and Snore Hall). - Turchetel filius Goch' 1165-1166 P, p. 23, 1166--1167 P, 80 p. 23 (Lynn), Radulfo filio Goche, 81 Turchetello 82 et Sirico et Hugone fratribus suis, abl. (witn.) c. 1165 (15th) NorwCPCh, no. 124 (Lynn and Mintlyn). 83 - Turchetil filius Ogeri 1165-1166 P, p. 22, Turchetil filius Oggi84 1166--1167 P, p. 21 (Lynn). - Wlnodfilius Turchil 1165-1166 P, p. 26 Wulnod' filius Turchil 1166--1167 P, p. 26 (Lynn). -Gaufrido j{ilio) Turchil, abl. (witn.) c. 1165-1180 CAcre, fo. 85b(ii) (the document concems Ulfchetel son of Aildricus of Marham). Transcript' Turchil' de Gernemuta, Thurkil' fratri Sewine de Gernem', dat., 1168-1175 Holme, fo. 73b(ii),85 (lease to Thurkil of land given to the Abbey of St Benet of Holme by Hardmannus; Gernemuta, Gernem' is Yarmouth). - Turchillo carpentario, abl. (witn) c. 1170 Binham, fo. 23a(i) (Lynn). -Thurchillo moppi, abl. (witn.) c. 1170-1175 Binham, fo. 121b(i) (the mill of Wood Dalling). - Willelmus filius Turchil (witn.) c. 1170-1180 CAcre, fo. 16a(iv) (Massingham). -Lamberto pistore. Turchillo. et Reinaldo fil(ibus) eius, abl. (witn.) c. 1175 Binham, fo. 4a (charter granted by Robert de Valoignes to Binham Priory), Turchilfilius Lamberti et Reginaldusfrater eius (witn.) c. 1175-1180 Binham, fo. 123a(i) (Wood Dalling, Little Ryburgh, Brittrichescrof). terram que fuit Thurchilli presbiteri 1176--1189 Wymondham, fo. 19a (Colton), terram que fuit Torkilli presbiteri 1182 Wymondham, fo. 10a (Coston), terram que fuit Thurkilli presbiteri 1230 Wymondham, fo. 22a (Carleton Forehoe). - Roberti filii Turchill' de Bradeham, gen., 1180-1181 P, p. 89, 1181-1182 P, p. 70, Roberti filii Turkill' de Bradeham, gen., 1182-1183 P, p. 11,86 Robertifilii Turkilli de Bradeham, gen., 11831184 P, p. 5 (Bradenham). -Turchillo de la Gren, Turchillo malebron de Binham, abl. (witn.) c. 1180-1185 Binham, fo. 49b(i) (Bamey; Binham is Binham). -Aliciamfiliam Thurkil c. 1185 Wymondham, fos. 37b(ii), 39a(i), Aliciamfiliam Turkil ibid., fo. 4la(ii) (Nf, prob. Wicklewood). -Turchillo le bedel, dat.,? 1186--1210 Holme, Appendix G iii (Estfen, Potter Heigham). - Thurchillus filius Edmod c. 1190-1200 Binham, fo. 5a, Thurchil carpentarius ibid., fo. 5b (printed, SSMEA, Appendix Il, no. 3, pp. 267,269) (Binham). -Edwinusf Turkill' 1193 P, p. 26,!Edwinusf Turkill' 1194 P, p. 60 (Nf or Sf). -Alano filio Turchill', abl. (witn.) c. 1195 CAcre, fo. 61a(i) (Kipton). - Willelmum filium Thurkil 1196 FF (PRS, xvii, no. 208) (Stanford and Hingham). - Turkil Mirewaud' (gen.) 1197 P, p. 246 (Nf or St). -pro fuga Turkil de Wikes 1197 P, p. 244, pro fuga Turkilli 1198 P, p. 89, pro fuga Turkill' 1199 P, p. 271 (Nf or Sf; Wykes is unidentified). - Gaufridumfilium Turketel 1198 FF (PRS NS, xxvii, no. 20) (West Winch), Gaufridofilio Turketelli, dat., 1200 CRR 22 (Cur, i, p. 261) (Lynn). -super Ricard' Turchetel? 1198 FF (endorsement to PRS NS, xxvii, no. 50) ((Nf; PRS NS, xxvii, no. 50,

Turchetelfilius Goche CR. Var: Cache (e. 14th). 82 V ar: Turgetello (e. 14th). 83 For the dating of this document see above, s.n. Arnkell, n. 34. 84 Ogge CR. 85 This charter is printed M Holme, no. 310, where the present forms are incorrectly rendered as follows: Transcriptum Turkili de Gernemuta, Thurkilo fratri Lewine de Gernemut', dat. 86 Turchill CR. 80 81

416

l>orke(ti)II, Thorkil, Thurkil, Thyrkil, Th0rkil is concemed with property in Runham). - Benedictum filium Thurkild 1198 FF (PRS NS, xxvii, no. 67) (Marlingford). -Thurketell' filius Edwini, predictus Thurketell' 1198 FF (PRS NS, xxvii, no. 104) (Martham). -Johannemfilium Turketel 1198 FF (PRS NS, xxvii, no. 134) (Lynn), Johannemfilium Turekilli 1213 CRR 59 (Cur, vii, p. 15),Johannifilio Turketilli, dat., 1214 CRR 59 (Cur, vii, p. 64) (Nf). - Walterumfilium Turketill' 1199 CRR 11 (CurR, i, p. 297) (Nf). -Thurketil' filius Aky 1199 CRR 17, 1887 (CurR, ii, p. 22) (Wiggenhall). -Galfridusf Turketill' 1199 P, p. 285, 1200 P, p. 143, Galfridus f Turkill' 1201 P, p. 135, 1202 P,88 p. 109, 1203 P, p. 239 (Nf or St). - terram lohannis filii Turkil (bis) c. 1200 Wals, fo. 80a(ii) (Stibbard). - homagium Wlfrici filii Turkil cum toto tenemento suo in Redesnewelond in Wigeh' c. 1200-1205 Der, fo. 135a(i) (Wiggenhall), hornagio Wlfrici filii Turkil de Wigehale ibid., fo. 190a, homag' Wlfrici filii Turkil de Wigehale ibid., fo. 190b (Wiggenhall St Mary Magdalene), de hornagio Wlfrici filii Thurkil, homagium Wulfrici filii Turkil ibid., fo. 191b(ii) (Wiggenhall St Peter), De Woluricf Turkil 1203 AssRoll 817 (NthRS, v, no. 780) (Nf), Woluricusf Turkill' 1203 AssRoll 817 (NthRS, v, no. 818) (Wiggenhall), Wlfuricusf Turkill' 1203 P, p. 249, Wluricus f (Turkill') 89 1204 P, p. 241, de Wulurico f Turkill' 1205 P, p. 246 (Nf). - Gaufridus filius Purkilli c. 1200 CAcre, fo. 31b(i), Aileue filia Turchill' de Rucham ibid., fo. 3 lb(ii), hereditas Galfridi filii Turchill' et Aleuie sororis sue c. 1210 CAcre, fo. 30b(iii), Rogerusfilius Turchil' de Rucham c. 1215 CAcre, fo. 30b(iv), Rogerus de Ruchamfilius Turchil ibid., fo. 30b(v), Ascelin' u.xorem Purkilli dusing c. 1200 CAcre, 3 lb, terram Purkil dusing ibid., fo. 3 lb(ii) (Rougham). 90 - Radulfusfilius Turkil de Castr' e. 13th Hickling, fo. 32b(iii) (Nf; Castr' is Caister next Yarmouth). - Thurchillus Faber, Thurkillusfilius Lefsi 1201 (13th) Rams, iii, p. 292 (Walsoken), Thurkillus ibid., p. 298 (Outwell). -Bernardusfilius Turkill' 1202 CRR 28 (Pleas, ii, no. 881) (Bemard son ofTurkil was a surety for Sefar Brokesanke in a plea conceming property at Castr'). - Johannes f Turkil 1202 P, p. 114 (Nf or St). - masagium quod Thurkill' tenuit 1202 FF (PRS NS, xxvii, no. 314) (Ovington). - de Hugone filio Turkill' et heredibus suis 1202 FF (PRS NS, xxvii, no. 319) (Palling). -Turkill' filium Ricardi, jus predicti Turkill', predictus Thurkil 1202 FF (PRS NS, xxvii, no. 446), terram Thurkilli c. 1215 Der, fo. 256a (Oxborough). - Reinerus filius Turkilli 1206 CRR 42 (Cur, iv, pp. 154, 196) (Ringland). - (de) Thurkillo filio Brittild' 1206-1207 (m. 14th) Rams, ii, p. 320 (Chisebeche, in Outwell). - Turchil' Brind' (acc.), Turkill' 1209 AssRoll 558 (Pleas, iv, no. 4188) (Turchil Brind was a villanus, holding at Ingham). - terram Radulfi filii Thurkel c. 1210 Wals, fo. 64b (Hindringham). - Beatrix Thurkild c. 1210-1230 Hickling, fo. 2a(iv) (Horsey), de Beatrice vxore Turcildi ibid., fo. 21a(ii) (Hickling and Horsey), de Beatrice vxore Turkildi et heredibus suis, De Willelmo filio Beatric' Turkild ibid., fo. 42b(iv) (Nf), de Beatrice vxore Thurkildi et heredibus suis ibid., fo. 49b(ii) (Hickling and Horsey), Goscel(in)us filius Thurkild de Hikelinghe c. 1250 Bod!NjCh, no. 592 (Yarmouth; Hikelinghe is Hickling). -Turkillusfilius Johannis 1214 CRR 58 (Cur, vii, p. 153) (the document concems a tenement in Southburgh). - in crofta que fuit Turkil c. 1215-1220 Wals, fo. 93a(iii) (Runton-Beeston Regis). -Purketel 1222ElyA, fo. 210a, Walterus Turkil ibid., fo. 212a, Ricardusfilius Purkil ibid. fo. 213a (Thorp in

These two rolls are duplicates. Galfridusf Turkil CR. 89 Interlined. 90 Turchil oj Rougham and Purkil dusing appear to have been two different persons. CAcre, fo. 3lb(ii) is printed as SSMEA, Appendix I, no. 29, where the forms from this charter which are given above are wrongly rendered as follows: Aileue filius Turchilli de Rucham, terram Thurkil Dusing. 87 88

417

l>orke(ti)II, Thorkil, Thurkil, Thyrkil, Th0rkil Shipdham). - Ricardo filio Turkilli, abl., (Editha) soror predicti Turkilli, Turkillus filius suus (i.e. of William son of Wulmerus Bese), pater predicti Ricardi, Thurkillus pater suus (i.e. ofRichard) (bis) 1225 CRR 89 (Cur, xii, p. 13)91 (Panworth in Ashill). -terram Rogeri filii Turchild c. 1225 Hickling, fo. 15a(ii) (Palling). - terram que fuit quondam Thurkilli le heyward c. 1225-1240 Der, fo. 25b(i) (West Dereham). - Radulfi filii Thurkill', gen., c. 1230 Binham, fo. 169b(i) (Great Ryburgh). -Thurkillum de Saxstorp 1232 CRR 232 (Cur, xiv, p. 532) (ltteringham; Saxstorp is Saxthorpe). -terram que fuit Turkilli Bedelli c. 1235 Der, fo. 20b(ii) (West Dereham). This name is frequent throughout Scandinavia in the Viking and medieval periods, usually appearing with syncopation of the second element as ON Porkell, ODan (run.) Porkil (runic purkil, nom.), ODan Thorkil, Thurkil, Thyrkil, Th(Jrkil, OSwed (run.) Porkell, Porkil, OSwed Thorkil, Th(Jrkil etc. (Lind, cols. 1184--1187, 1304--1305; NiyR, v, p. 232; DR, nos. 132, 157, 161, 269, 376, M~. 104--107; DgP, cols. 1389-1395; Hald, Vore Stednavne, pp. 114, 135-136, 178; DS II, p. 81; DS III, p. 83; DS IV, pp. 38,325,468; DS V, pp. 93,648; DS VI, p. 373; DS XI, pp. 169-170; DS XVIII.l, p. 41; SRÖg, nos. 32, 81, 82, 84, 165; SRSö, nos. 34, 54, 112, 305; SRVg, nos. 9, 35, 40, 143; SRU, nos. 145, 435, 453, 454, 455, 566, 653, 914, 933, 1066, 1072; Lundgren-Brate, pp. 270-271; Modeer, Personnamn i Ktb, pp. 11, 50, 89-90, 127; Modeer, Fornsvenska personnamn i en regestsamling, p. 15; Modeer, Östsmåländska bondenamn från 1500-talet, p. 61; Thors, p. 91; Fredriksson, Västgötska dopnamn under 500 år, pp.182, 183, 185; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 146, s.n. Torkel). The more conservative full form, ON Porketill, OSwed (run.) Porka:till is infrequent in Scandinavia, where it occurs in Norse skaldic verse and in runic inscriptions in Södermanland and Uppland (see above, s.n. Arnkell, and the references given there). 92 In England forms representing both the full form Porketill and the later Porkell etc. are on record, the latter being the more frequent: NPN, pp. 151-153; ZEN, p. 85; Björkman, Festermen, pp. 5, 17; PNDB, pp. 394-395; SPNT, pp. 134, 140--141, 145; ELPN, p. 83; Newark, pp. 4, 14 (Turchel, Turcheti); DBS, p. 348, s. nn. Thurkell, Thurkettle; SPNLY, pp. 309-311, 356; Smart, Moneyers, pp. 230, 235-236, 240; von Feilitzen, SCBI, xi.i, p. 13; Coinage of Edgar, p. 207 and n. 1; SCBI, xxi, pp. xlvi, l; von Feilitzen, PNWinton, p. 174; von Feilitzen PNOEO, pp. 35-36; Insley, Nomina, iii (1979), pp. 53, 56-57; DEPN, pp. 466, 471, s. nn. Thirkleby YN, Thirtleby, Thruxton, Thurcaston; PN BdHu, p. 300; PN Wo, p. 398; PN NRY, pp. 49,189,333; PNNth, p. 292; PN Sr, p. 402; PNEss, pp. 592,635; PNERY, pp. 29, 5152, 325; PN Cu, pp. 192,506; PN 0, pp. 270,481; PN Db, p. 762; PN WRY, i, p. 267; PN WRY, iv, pp. 239,252; PN WRY, vi, p. 272; PN We, ii, p. 314; SSNY, p. 39; SSNEM, p. 195; Insley, JEPN, x (1977-1978), pp. 44, 59. Sf examples include the fol-

91 The part of this document containing the form Thurkillus pater suus (bis) has been supplied by the editor of Cur xii from Bracton (see Cur, xii, p. 13, n. 8). 92 For an examination of the -ka:till/-kell/-kil complex and its development in Scandinavia see above, s. n. Arnkell.

418

l>orlåkr, l>orleikr, l>orläkR, Thorläk lowing: Turchetlestuna 1086 DB, fo. 420, Turketeleston' 1202 FF (PRS NS, xxxii, no. 366) (lost village in Shotley, Samford H); Turchil 1087-1098 (c. 1175-1200) Bury, p. 34 (Stanningfield), Turchildus presbiter ibid., p. 38 (Honington), Thurchildus ibid., p. 40 (Wattisfield); Turkillus presbiter (witn.) 1121-114893 (1429-1446) Bury, no. 129, Turkellus Presbiterde Pakenham (witn.) 1148-1156 (1. 13th) (Great Barton; Pakenham is Pakenham), Turkil Presbiter (witn.) 1156-1160 Bury, no. 144 (the document concems estates at Coney Weston, Hopton near Thetford, Great Barton and Bury St Edmunds); terram Turchilli filii Ketelli in Humeresfeld 1146-1174 CampbCh xii. 3 (printed, SSMEA, Appendix I, no. 19) (Homersfield); Turkil de Pacheham (witn.) 1161 SibtonCh, no. 2 (land near Freckenham; Pacheham is Pakenham); Rogerus filius Thurketel (hus band of Gunnilda) 1. Hy2 SibtonCh, no. 8 (Sf; grant to Sibton Abbey ); 94 v acras de ware in uilla de Rucham quas Turkillus aliquando tenuit per seruitium quod terre prescripte debent annuatim ad hundredum de Thetward(estre) c. 1182-1200 (13th) BuryS, no. 38, Turkil 1186-1188 (c. 1270) BuryS, pp. 18, 19 (Rougham); 95 Johannes filius Turkilli 1186-1188 (c. 1270) BuryS, pp. 10, 11 (Pakenham, Thurston), Turkild filius Norman, Turkil filius Norman ibid., p. 11 (Thurston), Turkil ibid., p. 13 (Hessett - Beyton), Turkild, Turkil ibid., p. 19 (Rougham), Aluricus filius Turkilli ibid., p. 41 (West Stow), Turkil ibid., p. 46 (Honington); Thurkill' fil' Edwardi (acc.), Thurkill' (ter) 1209 FF (PRS NS, xxxii, no. 519) (Semer).

In Normandy the full form Porketill occurs independently as Turketil, Torchetil, Torchetillus, Torchitillus, Turchetillus, Turchitillus etc., remaining in use until c. 1200, and forms the first element of three place-names, while the later Porkell, Purkil etc. is only known as the first element of a single place-name (Adigard, pp. 164-165, 322-326, 429). As Turcall, Torcall the name has been noted in Irish annals of 1124, 1160 etc. (Marstrander, pp. 65, 80). For initial Pur- and T(h)ur-, which formally reflect the OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h)- for OScand P- see above, s.n. P6raldr. For the survival of OE, OScand initial P- in ME see above, s.n. P6rör. For the second element, OScand -ketill/-kelll-kil etc., and its appearance in English records, see above, s. nn. Arnkell, Askell.

ON l>orlakr, l>orleikr, ODan (run.) l>orläkR, ODan Thorläk, OSwed (run.) l>orlreikR, l>orläkR, AScand l>urläc a) (i) Turlocyweit (bis) c. 1210 Binham, fo. 50a (f. n. in Barney). -Thurlakesdele 1302 AddCh 59947 (f. n. in Burnham Norton).

Towards the end of this period. This charter hasa late 12th century endorsement Rogerifilii Thur ketel, and it hasa seal hearing the legend- SIGILL' ROGERI FILII THVRKETEL. 95 For this Turkil(lus) and his tenement see BuryS, p. 18, n. 3. 93

94

419

l>orlakr, l>orleikr, l>orläkR, Thorläk b) (i) pret fen pe Purlac me sealde in i'Elmham pa preostas to foddan 1035-1040 (contemp.) ASWills, no. 26 (Nf; !Elmham is North Elmham; the document is the will of Bishop i'Elfric ). - apud Plomstedam terram Turlaki que reddit per annum x sol' 11181138 CAcre, fo. 3a(ii), apud Plumstede terram Thurlac. que reddebat x sol' 1146-1147 CAcre, fo. 3b(i) (Plumstead). - iiii sol' in Fincham de filiis Turlach c. 1165-1180 CAcre, fo. 85b(i) (Fincham). - terram Turlac de la Lunde c. 1195 CAcre, fo. 88b(iii) (Fincham). -Radulfumfilium Turlac 1199 CRR 17 (Pleas, i, no. 2368) (Nt), 1199 Justiciar's Writ (Pleas, i, no. 3516) (this writ concems a suit brought by Robert de Sudale against his cousin Ralf son of Turlac conceming 21 acres at Plumstead, Wolterton and Bamingham), 1201 FF (PRS NS, xxvii, no. 264) (quitclaim conceming the 21 acres at Plumstead, Wolterton, and Bamingham), 1206 CRR 41 (Cur, iv, p 76) (Plumstead), Radulfus filius Turlac 1200 CRR 19 (Pleas, i, no. 2687), Radulfum Turlac ibid. (Pleas, i, no. 2713), Simonem de Turlac ibid. (Pleas, no. 2643), Simon Turlac' ibid. (Pleas, i, no. 2683) (Nt), Simonem Turlac, 96 Bonde avus predicti Roberti97 et Simonis Turlac, Turlac, Turlac junior frater 98 et pater predicti Simonis 1200 CRR 24 (Cur, i, p. 297) (suit brought by Simon son ofRobert de Sudale, acting for his father, against Simon son of Turlac concerning the 21 acres in Plumstead, Wolterton and Bamingham). 99 -1ohannes Turlak 100 1229 CRR 101 (Cur, xiii, p. 374), terram quam commune uille de Vtwelle dedit Johanni Thurloc et sociis suis c. 1240 Wals, fo. 132b(i) (Outwell). - terram Thurlac c. 1230 Der, fo. 53a(iii), terram Ricardi Thurlac c. 1250-1260 Der, fo. 52b(i), terram Baldewini Purlac ibid., fo. 55b(iii) (Barton Bendish).

This name is predominantly OWScand in its distribution in Scandinavia itself. ON Porlakr, -leikr is frequent in Iceland from the time of the settlement onwards, and is common in Norwegian records after c. 1300 (Lind, cols. 11871190; LindS, cols. 864-867). In OEScand the name is rare, occurring in Danish and Swedish runic inscriptions and appearing sporadically in other Danish records (DR, nos. 160, 354,384,400; DgP, col. 1397; SRÖg, nos. 23, 156; SRSmå, no. 128; SRU, no. 753). In England ON Porlakr etc. is rare (NPN, p. 159; ZEN, p. 87; DEPN, p. 471, s.n. Thurloxton; PN Nth, p. 291; PN Wa, pp. xxii, 329; PN Db, p. 438; Insley, JEPN, x (1977-1978), p. 49; SSNEM, pp. 164-165). Note especially the following examples from the OE period: Thurlac 925-928/938-939 C.E. Blunt. The Coinage of Athelstan, 924-939, pp. 121, 139 (moneyer, Athelstan's two-line type); Purhlac feröing 963-984 (m. 12th) ASChR, no. 40 (surety in a Peterborough land transaction, probably conceming landat Wittering Nth), Purlac ibid. (surety in a transaction conceming landat Wittering Nth), Purlac (bis) 963-992 (m. 12th) ASChR, no. 40 (surety in transactions conceming land at Peterborough Nth and Maxey Nth), Leofsie Purlaces sune

Simonemfilium Turlac CRR 22 and 23. i. e. Robert de Sudale. 98 i. e. of Elfere son of Bonde. 99 Fora genealogical table of the family of Ralf and Simon, sons of Turlac, see above, s. n. B6ndi, fig. 2, p. 101. HlO For Johannes Turlak CRR 102 has Turlac (Cur, xiii, p. 374, n. 4. 96

97

420

l>ormoör, Thormöth, Thormödh (bis) ibid. (surety in transactions conceming landat Bainton Nth and Anlafestune Nth). In Normandy ON Porlakr etc. forms the first element of two place-names, Tourlaville (Manche) and Tout-la-Ville (Calvados) (Adigard, pp. 166, 429430). For initial Pur- and T(h)ur-, which formally reflect OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h)- for OScand P- see above, s.n. P6raldr. For the survival of OE, OScand initial P- in ME see above, s.n. P6rör.

ON l>ormoör, ODan Thormöth, OSwed Thormödh a) (i) Thurmodetun 1202 FF (PRS NS, xxvii, no. 454) (lost village, grouped in the present document with Saxlingham Nethergate ). 101 b) (i) Tarmoht. I. fiber. homo. regis. E. TRE DB, fo. 246 (Denton). - Turmod etfilius ejus 1165-1166 P, p. 29, 1166-1167 P, p. 24 (Nf or Sf). - terram Hacunfilii Thurmod c. 1215Der, fo. 256a, superforeramHaconisfiliiThurmodibid., fo. 257b (Oxborough).

ON Porm6ör is extremely frequent in Norway and Iceland throughout the Viking and medieval periods (NiyR, nos. 114, 154, 245, 259, 273; Lind, cols. 1198-1200; LindS, cols. 868-877). In Denmark the name occurs as runic purmupr, nom., and it appears occasionally in medieval Danish records (DR, no. 150; DgP, cols. 1398-1399). In Sweden the name is found independently in medieval and 16th century records and forms the first element of several place-names, occurring most frequently in Västergötland (SRVg, no. 246; Lundgren-Brate, p. 271; Modeer, Personnamn i Ktb, pp. 15, 69, 89; Fredriksson, pp. 199-206; Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 146, s.n. Tormod). In England ON Porm6ör etc. is quite well attested, occurring independently and as the first element of place-names: NPN, pp. 159-160; ZEN, p. 87; PNDB, p. 395; SPNLY, p. 311; von Feilitzen, SCBI, xi.i, p. 13; Coinage of Edgar, p. 207; DEPN, pp. 467, 471, 472, s. nn. Thormanby, Thornaby on Tees, Thrumpton, Thurmaston; PN NRY, pp. 26-27, 172; PN Nt, pp. 90, 250; PN We, ii, p. 210; Cameron, Scandinavian Settlement 111, p. 150; SSNY, p. 40; SSNEM, pp. 195, 198; Insley, Nomina, iii (1979), p. 57. ON Porm6ör etc. occurs in DB in SfTRE (PNDB, p. 395), and has also been noted in Sfin the post-Conquest period. Cf. the following post-Conquest Sf examples: Turmod 1087-1098 (c. 1175-1200) Bury, p. 39 (Wattisfield); Carta Turmodi de Hegesete, Thurmodus de Hegesete? I. 12th (13th, after 1250) Stake, fo. 133a (Hessett); Walterus filius Turmod 1186-1188 (c.

101 The editor of the present document suggests that Thumwdetun may possibly have been located in Raveningham (PRS NS, xxvii, p. 225, sub Thurmodeston).

421

Thorn, *l>orn 1270) BuryS, p. 44 (Troston); Turmodum 1199 CRR 17 (Pleas, i, no. 2247) (St); per Turmot' 1199 CRR 17 (Pleas, i, no. 2503) 102 (St); terram Thurmod' 1201 FF (PRS NS, xxxii, no. 306) (Brettenham); Purmod' lerond 1222 ElyA, fo. 178b, Idem Purmod ibid., fo. 179a (Rattlesden).

For initial Pur- and T(h)ur-, which formally reflect OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h)- for OScand P- see above, s.n. P6raldr. For the survival of OE, OScand initial P- in ME see above, s.n. P6rör. The spelling -mod(us) for OScand -m6ör reflects the replacement of this Scand name element by the corresponding OE -mod (NPN, p. 159, n. 1; PNDB, p. 100, n. 2). The -a- in the Nf DB form Tarmoht is scribal error for -o- or -u- (see PNDB, p. 395 and n. 3).

ODan Thorn, AScand *l>orn (?) b) (i) Torn fiber homo TRE DB, fo. 174 (Hunstanton). -Torn TRE DB, fös. 248b, 249 (ter) (Hempnall, Boyland, Hudeston in Bunwell, Raveningham).

ODan Thorn is rare and late, the earliest example in DgP having been noted in a record of 1434, and it is best interpreted as an original byname belonging to the appellative ODan thorn m. 'thom', though DgP also includes the possibility that it could be a short form of a name in -thorn (DgP, col. 1399; NK VII, p. 207). The lateness of the first appearance of ODan Thorn practically rules out any direct connection between this name and the above Nf DB forms in Torn. These Nf DB forms reflect an original byname which belongs either to OE porn m. 'thom, thom-bush' or to ON, OEScandjJorn m. 'thom, spike', and, though ultimately etymologically identical with the ODan Thorn, they represent an independent English development. Torn also occurs in Y in DB TRE and a form Crinan f Thorne has been found in a NRY record of 1100 - c. 1115 (PNDB, p. 395; SPNLY, p. 312). For initial T- for OE, OScand P- see above, s.n. P6raldr.

ON 1>6r6lfr, ODan (run.), OSwed (run.) l>örulfä, OSwed Thörulf a) (i) Thuroluesacre c. 1180 Wymondham, fo. 9lb(ii) (f. n. in Great Melton). -Thurolfes milt' m. 13th Wals, fo. 38b(i) (? mill, Great Walsingham).

102

422

Turmod' CRR 18.

1>6r6lfr, l>örulfR, Thörulf b) (i) Torolf fiber homo Stigandi TRE DB, fo. 202 (Tasburgh). - Ricardo filio Turolf1° 3 et Aschetillo 104 fratre eius, abl. (witn.) c. 1165 (15th) NorwCPCh, no. 124 (Lynn and Mintlyn). 105 -Paul' filio Thurolf, abl. (witn.) c. 1200 CAcre, fo. 80b(iii), Paul' filio Turrolf, abl. (witn.) ibid., fo. 80b(iv) (Lynn). - Ricardus filius Turolf 1201 CRR 25 (Cur, ii, p. 64) (Nt).

ON P6r6lfr is extremely frequent in Iceland from the time of the settlement onwards and is also fairly well attested in Norwegian records (NiyR, nos. 110, 121, 160; Lind, cols. 1204-1205; LindS, cols. 877-881). The name also occurs in a Manx runic inscription as jJurulfs hins raujJa, gen. (RIGB, p. 217). In OEScand Pörulf(R) is rare. Runic Danish has the forms jJurlf, nom. (DR, no. 1) and jJurulfs, nom. (DR, no. 46). In Swedish there are the runic examples jJuruljR, nom. (SRSö, no. 328), jJurulf, nom. (SRU, no. 201), and jJurul---, nom. (SRU, no. 890), and the forms Thorulphus, Thuruluus occur in records of 1312 (Lundgren-Brate, p. 273). The name is, however, well attested in OEScand in its syncopated form, ODan Thölf, OSwed Thölf, Thulf, which was borrowed into Norway as ON P6lfr, frequent after c. 1300 (DR, nos. 77, 201, 329; DgP, cols. 1364-1366, 1695; Kousgård Sl?)rensen, APhS, xxiii (1957), pp. 106--109; Sl?)ndergaard, p. 110; DS XVI, p. 216; SRÖg, no. 230; SRSö, no. 109; SRVg, no. 70; SRU, nos. 756, 838; Lundgren-Brate, pp. 263, 276; Modeer, Östsmåländska bondenamn från 1500-talet, pp. 64-65 (Tolffuer); Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183, 185 (Tolv); Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, p. 144, s.n. Tolv; Lind, cols. 1133-1134; LindS, cols. 825-831). In England ON P6r6lfr, OEScand Pörulj(R) is well attested both independently and as the first element of place-names: NPN,pp. 162-163; ZEN, p. 88; PNDB, p. 396; Tengvik,p. 201; SPNLY,p. 317; Smart, Moneyers, pp. 230, 239; SCBI, xxi, p. xlvii; DEPN, pp. 466,471, s. nn. Tholthorpe, Thulston, Thurland, Thurlby, Thurlston; PN La, p. 183; PN Bk, pp. 49-50; 106 PN Wo, p. 399; PN NRY, pp. 21-22; PN Nth, p. 291; PN Db, pp. 462, 762; PN WRY, i, p. 339; SSNY, pp. 68, 130; SSNEM, pp. 75, 118, 195; Insley, JEPN, xi (1978-1979), p. 63.

In Normandy the name occurs in independent use as Turulfus, Turolfus, Torulfus, and forms the first element of four place-names (Adigard, pp. 170171, 340-341, 431-432). In England the OEScand syncopated variant P6lfr is recorded in DB TRE in Db, Do, Ha, Sf, St and Y, and appears in the Liber Vitae of Thomey Abbey

103 Var: Thurold (e. 14th); Thurolf (e. 14th). The form Thurold shows confusion with ON P6raldr etc. 104 Var: Astentell (e. 14th); Archetill (e. 14th). The form Archetill shows confusion with ON Arnkell, ODan Arnketil etc. (see above, s. n. Ä.skell, n. HlO). 105 For the dating of this document see above, s.n. Arnkell, n. 34. 106 Cf., however, DEPN, p. 483, s. n. Turweston.

423

*l>orp as Dolf (PNDB, pp. 389-390; SPNT, p. 145; SPNLY, p. 295). It also forms the first element of Thoulstone W (PN W, p. 156). For initial T(h)ur-, which formally reflects OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T(h )- for OScand P- see above, s.n. P6raldr. For final -olf, which may possibly reflect the Scand variant -6lfr, see PNDB, § 17; SPNLY, § 32.

OScand *l>orp b) (i) Torp TRE DB, fo. 251b (Sturston).

PNDB, p. 396, interprets the above Nf form as belonging to the appellative and place-name element ON, OEScand porp 'a secondary settlement, a dependent outlying farmstead or hamlet', and as having had the function of a nickname for a man living at a Porp or at a place called Porp (> Thorp, Thorpe). This interpretation is rejected by Kökeritz, NoB, xxvi (1938), p. 38, who takes Torp to be identical with the Tort (< ON P6rör etc., above, s.n.), who appears on the same folio of DB, final -P having been misread as -p. For initial T- for OScand P- see above, s.n. P6raldr.

AScand *l>orric b) (i) lohannesfilius Thorici de Pagraue c. 1230---1240 CAcre, fo. 29a(iv) (grant by John son of Thoric of Palgrave to Castle Acre Priory of rents from Philip Emald of Great Palgrave ), Cristianafil' Thorici de Magna Paggrave 1240 Selten PN, p. 161 (Great Palgrave).

This name is probably best interpreted as an AScand hybrid formed from OScand Por- and OE -rzc, though it could, theoretically, be a true Scand formation, since T oricker is recorded in a document of 1492 from the Swedish province of Dalecarlia (Selten PN, p. 161). T(h)urrich occurs as a surname in Sf in records of the period 1308-1328 (Selten PN, loc. cit.). For further AScand hybrids of this type see below, s. nn. *Purw(e)ard, *Purwzg, *Purwine. For initial Th- for OScand P- see above, s.n. P6raldr.

424

l>orsteinn, l>orsten, Thorsten, l>orstreinn

ON l>orsteinn, ODan (run.) l>orsten, ODan Thorsten, OSwed (run.) l>orstreinn, OSwed Thorsten, AScand l>urstän, AN Turstin (us) a) (i) Turstanestuna I 086 DB, fös. 121 b, 166b, Turestuna ibid., fo. 167, Thurstuna ibid., fo. 183b, Turstuna ibid., fo. 214b, Thustuna ibid., fo. 121b, Toruestuna ibid., fo. 174b (Thuxton, Mitford H). -Thurstangate c. 1200 Wals, fo. 79a(i) (f. n. in East Barsham). - Thurstaneshowe 1214-1229 Holme, fo. 78a(iii) (f. n. in Belaugh). -Thurstaneshoge c. 1225 Coxford Il, fo. 33b(iv) (f. n. in West Rudham). - Thurstanesmers 12701pm (f. n. in Halvergate). b) (i) Dorstan 1009-1016 Smart, Moneyers, p. 243 (moneyer, Norwich). - Purstan Wine sune, Purstan, Purstanes, gen., 1043-1045 (l. 13th) ASWills, no. 31 (in this document, his will, l>urstan son of Wine disposed of properties in C, Ess, Nf (Shouldham is named) and St). - Dvrstan 1050---1053 SCBI, xviii, no. 1144, 1062-1065 SCBI, xvii, no. 470 (moneyer, Norwich). -Turstinus liber homo TRE DB, fo. 178 (Great Snarehill). -Turstinus liber homo TRE DB, fo. 274 (Wallington). -Turstinusfilius Guidonis (bis) 1086 DB, fo. 179b ((North) Creake, Quarles), Turstinus filius Widonis ibid., fo. 183b (held four liberi homines with half a carucate at (North) Creake), Torstinus.filius. Wd' ibid., fo. 185 (held three liberi homines with one anda half carucates at Antingham), Torstinus ibid., fo. 185 (held one liber homo with 40 acres at Shipden), Turstinus ibid., fo. 223b ((North) Creake). -Turstinus burgensis 1086 DB, fo. 173 (Thetford), ad feudum Turstini. de Tedfort, Turstinus ibid., fo. 277b (Great Snarehill). - terram unius sochemanni nomine Torstini 107 1086 (c. 1180) IE (A) (Hamilton, p. 140) (Didlington). Tostano (de Achra), abl. (witn.) c. 1090---1100 CAcre, fos. la(ii)(Castle Acre), lb (charter of William Il de W arenne in which the grant of Hugh de W anci of the churches of Deepden and Barsham and the tithes of Hugh's francigeni were confirmed to Castle Acre Priory), Tustano, abl. (witn.) c. 1100---1120 CAcre, fo. 2a (Nf), Tustanus (witn.) ibid., fo. 2a(i) (the charter cancerns various properties in Nf granted to Castle Acre Priory by Lambert de Rosei), Turstano de Acra, abl. (witn.) c. 1115 CAcre, fo. 26a(iii) (Kempston, Sporle, Newton near Castle Acre; Achra, Acra is Castle Acre, West Acre or South Acre ). - terram Turstini qui dicitur diaconus 1101-1119 (e. 14th) NorwCPCh, no. 110, terram que fuit Turstini diaconi apud Tornham 1136--1141 NorwCPCh, no. 117, Terram quoque que fuit Thurstani 108 diaconi in Thornham 1139-1143 (15th) NorwCPCh, no. 119, terra que fuit Turstini dia(coni) apud Tomam 1146--1174 NorwCPCh, no. 123, terram que fuit Thurstani dyaconi 1157 (e. 14th) NorwCPCh, no. 35, laicum feodum quod fuit Thurstini diaconi 1205 NorwCPCh, no. 170, de dominico Turstini quondam diaconi in Tornham 1227-1236 NorwCPCh, no. 193, de dominico Thurstini quondam diaconi in Thornham 1245 NorwCPCh, no. 212 (Thomham). -Turstano coco, abl. (witn.) c. 1115 CAcre, fo. 26a(i) (charter of Roger son of the Earl of W arenne' s seneschal Wimer by which he confirms the grants made to Castle Acre Priory by himself and by his father Wimer and his mother Gilla). -Thurstani filii Boyselyn, gen. (witn.) 1126--1127 Holme, no. 124 (this document isa grant of landat Grenseuille made by Abbot Conrad of St Benet of Holme to William de Curcun in retum for various renders, including two parts of the tithes of Blakeuurde and Fritton). - Turstino

107 108

Var: Thurstanus 1086 (I. 12th) IE (C) (Hamilton, p. 140, n. 2). Var: Turstini (c. 1290), (e. 14th).

425

l>orsteinn, l>orsten, Thorsten, l>orstreinn Reuello, abl. (witn.) 1127-1134 HarlCh 44 E 19 (printed, Holme, Appendix E) (Heigham next Norwich). - terram illam de Fauertonefeld' quam Turstinus filius Algari dedit predicte ecclesie in elemosinam 1138-1147 (1315) HarlCh 57 E 32 (printed, BS, no. 416) (charter of John de Warenne, earl of Surrey, issued April 4 1315, inspecting and confirming the charters ofhis predecessors William, Hamelin, William and William for the priory ofthe Holy Sepulchre at Thetford; Fauertonefeld' is unidentified). -Torstanus de Helengeia c. 1145 CAcre, fo. 92b(ii) (Polwere; Helengeia is Hilgay). -Turstani dispensatoris, gen., c. 1150 Crawf, no. 16, CAcre, fo. 67b(i) (the tithes ofTurstan the dispenser are here confirmed to Bromholm Priory). -Turstanus presbiter de Bucheham (witn.) 1151-1152 Har!Ch 83 D 9 (Nf; Bucheham is Buckenham, Blofield H). Turstain c. 1158-1161 SCBI, xvi, no. 405, c. 1161-1165 SCBI, xvi, no. 407, c. 11651168 SCBI, xvi, no. 408, Turstein c. 1161-1165 SCBI, xxi, no. 1264, Turstanus (monetarius de Tedford') 1167-1168 P, p. 26 (moneyer, Thetford), Turstano monetario de Tejf et Radulfo filio eius, abl. (witn.) c. 1175 CAcre, fo. 43a (ii) (charter in which Ralph de W anci confirms grants of his father and grandfather to Castle Acre Priory and grants to the Priory land at Depden St). - totam terram Turstani filii Sifled quam tenuit in Estunia. et Hunigham. et in Torp' 1159-1167 YCh, iv, no. 57 (Easton, Honingham and Honingham Thorpe). - de molendino de Holt quod Turstanus molendinarius tenet c. 1160 CAcre, fo. 57b(i), de molendino de Holt quod Turstanus tenuit c. 1165 CAcre, fo. 57b(ii), molendinum de Holt quod Turstanus tenuit ibid., fo. 57b(iii) (Holt). - Thurstanus 1161-1177 (13th) Rams, iii, p. 262, Thustanus ibid., p. 264 (Brancaster, Bumham, Burnham Deepdale), Thurstanus, Thurstanus filius Ailmari ibid., p. 267, Thurstanus Parvus ibid., p. 268 (Ringstead and Holme next the Sea), Thurstanus filius Haraldi 109 ibid., p. 288 (Hilgay and Snore Hall). - lohanne filio Turstein', abl. (witn.) 1164 - c. 1185 CAcre, fo. 6b(ii) (charter of Earl Hamelin de Warenne concerning a messuage in Thetford and a meadow between the land of the canons of W est Acre and the land of Richard son of Ringolf in Marham). -Turstinus Baril 1165-1166 P, p. 22, 1166-1167 P, p. 22 (Lynn). -Turst' filius Rolfl 165-1166P, p. 26, Turstinusfilius Rolf1166-1167 P, p. 26 (Lynn). - Turstein de Wiggehal' 1165-1166 P, p. 23, Turstinus de Wiggehal' 1166-1167 P, p. 22 (Lynn), De Turstano et Game! de Wigehale, Turstanum et Game! fratrem suum de Wigehale homines meos c. 1175 CAcre, fo. 82a(iii) (Gaufredus de Snetesham issues this charter), Turstan de Wigehale (witn.) c. 1185 CAcre, fo. 82a(iv) (the charter isa grant by Peter of Bexwell of Matilda, daughter of Snelling of Wiggenhall, with issue etc., to Castle Acre Priory; Wiggehal', Wigehale is Wiggenhall). - De molendino de Wording' et Turstano molendinar', molendinum de Wordinga cum Turstano molendinario c. 1165-1175 CAcre, fo. 63b(i), molendinum de Wordinga cum Turstano molendinario ibid., fo. 63b(ii), de molendino quod Turstanus tenuit c. 1195 CAcre, fo. 63b(v), et ex alia parte. vi. acras terre cum parvo prato quas Turstanus tenet c. 1165-1175 CAcre, fo. 63b(ii) 110 (Worthing). -Turst' de Winrthingh' (abl.) (witn.) c. 1175 CAcre, fo. 63b(iv) (grant by Hubert de Ria to Castle Acre Priory of the land etc. of Philip Belet; Winrthingh' is clearly corrupt and the result of scribal error). - Thurstano presbitero de Rucham, abl. (witn.) c. 1175 CAcre, fo. 86b(ii) (West Winch;

109 Written Halardi in the MS. Afterwards /ar was subpuncted and ral interlined (Rams, iii, p. 288, n. 5). 110 This part of CAcre fo. 63b(ii) appears at the end of the charter after the witness list and it is uncertain whether the Turstanus named here is the same person as Turstanus the miller of Worthing, who appears in the main body of the charter.

426

I>orsteinn, l>orsten, Thorsten, I>orstreinn Rucham is Rougham). - de Turstino palmario 1176--1177 P, p. 132 (Nf or Sf). - de catallis Turstin' 1176--1177 P, p. 137 (Nf or Sf). -Johannes filius Turst' de Tedford 1184-1185 P, p. 39, 1185-1186 P, p. 68,Johannesfilius Turst' de Thedford 1185-1186 P, p. 64, Johannesfilius Turst' 1186--1187 P, p. 55 (Thetford). - pro plegio Turst' celarii 1184-1185 P, p. 40, 1185-1186 P, p. 69 (Nf). -Rogerifilii Turst(ini), gen., 11871188 P, p. 63 (Nf). -CartaSaxifilii Thurstan de Watt', Saxifilius Thurstani de Waltone c. 1190 Der, fo. 119b(ii), 119b(iii), Carta Saxi fil ii Thurstan de Waltona, Saxi filius Turstani de Walton' ibid., fo. 119b(iv) (East Walton). -Thurstan c. l l90-1200Binham, fo. 5a (printed, SSMEA, Appendix Il, no. 3, p. 267) (Binham). - salinam quam Thurstanus presbiter tenuit c. 1190-1200 Wymondham, fo. 49b(i), Rogerum filium Turstini hominem meum, ipsam salinam quam tenuit predictus Turstinus c. 1200 Wymondham, fo. 54b(i) (South Wootton; Wymondham fo. 54b(i) isa charter of Hodiema of South Wootton). - Thurstanus filius Stengrim de Dersingham c. 1190-1220 Co:iford Il, fo. 68b(iii) (Dersingham). -Adekinf Turst' 1191 P, p. 44 (Nf). -Robertusf Turst' 1192 P, p. 191 (Nf or Sf). -terramRicardifilii Thurstani c. 1195-1205 Der, fo. 13a(iii) (West Dereham). -Turstanus Charite 1195 P, p. 74, 1198 P, p. 83, 1199 P, p. 267, Turstinus Charite 1196 CR, p. 134, 1197 P, p. 235 (Nfor Sf). -pro fugaRicardif Turston' 1197 P, p. 246 (Nf). - Decena Turstan' le seruier 1197 P, p. 245, de decena Turstani le seruier 1198 P, p. 87 (Nf or Sf). -DecenaLeuricif Turstani 1197 P, p. 244, 1198 P, p. 89, 1199 P, p. 271 (Nf or Sf). -Turstan Pigace 1198 AssRoll 559 (Pleas, ii, no. 30) (Beeston, Launditch H). - Thurstanum filium Wlurici, jus et hereditatem predicti Thurstani, predictus Thurstanus 1198 FF (PRS NS, xxvii, no. 8) (Wiggenhall). - Elfled sororem Turstein' 1198 FF (PRS NS, xxvii, no. 48) (Limpenhoe). -Rogerumfilium Thurstani 1198 FF (PRS NS, xxvii, no. 102) (Burlingham). - Leuare filiam Turstani 1198 FF (PRS NS, xxvii, no. 111) (Long Stratton). -Thurstanum de Grimeston', predictus Thurstanus 1198 FF (PRS NS, xxvii, no. 118) (Grimston). -Johannemfilium Turstani 1198 FF (PRS NS, xxvii, no. 195) (Wiggenhall), terram que fuitAlanifilii Thurstan de Wigehale c. 1210 Der, fo. 173b (Tilney; Wigehale is Wiggenhall). 111 -Turstan' Clober (acc.) 1199 CRR 11 (CurR, i, p. 414), Turstan' Cloker (acc.) 1199 CRR 17, 18 112 (CurR, ii, p. 13) (Fransham). - Turstanus de Fransham 1199 CRR 11 (CurR, i, p. 446) (Nf), Turstanus de Frousham 1199 P, p. 287 (Nf), Turst' de Fransh' (abl.) (witn.) c. 1200 CAcre, fo. 26b(ii) (the charter concems lngritha, wife of Aylmer Kele ofKempston), Turstanus de Fransh' ibid., fo. 26b(iii) (Kempston), Turstano de Frousham, abl. (witn.) ibid., fo. 98b(ii) (confirmation by Gilbert ofFransham to Isabel, wife ofReginald, son of Adelwald of Fransham, of eight acres in tofta sua and four acres in Tullescrofte and Caulinge; Fransham, Frousham, and Fransh' represent Fransham), Turstan' (abl.) (witn.) ibid., fo. 99a(i) (Fransham). -Turstanus Blanchome 1199 P, p. 287, 1200 P, p. 145 (Nf or Sf). - Radulfus filius Turstani 1199 CRR 17, 1200 CRR 19 (Pleas, i, nos. 2592, 2741), 113 Radulfusfilius Turstani apud Fraceham 114 1200 CRR 19 (Pleas, i, no. 2952), Radulfusfilius Turstani de Freteh' 1200 CRR 20 (Cur, i, p. 171) (Nf; Fraceham and

111 The John son of Turstan of 1198 FF and the Alan son of Thurstan of Wiggenhall of Der fo. 173b are shown to have been brothers by the 1198 docurnent. Cf. also the Thurstan son of Alan, whose landat Wiggenhall St Germans is mentioned in Der fo. 195a c. 1235. This Thurstan may well be the son of the above-mentioned Alan and be named after bis grandfather. 112 These rolls are duplicates. 113 Pleas, i, no. 2741, is cancelled.

1 14

Freceham CRR 20.

427

l>orsteinn, l>orsten, Thorsten, l>orstreinn Freteh' represent Frettenham). -Agneti filie Turstani Norwicensis, dat.,? 1. 12th/e. 13th Bromh, fo. 43b(iv), Benedictus et Walterus fil(ii) Thurstani presbiter(i) de Dilham, Angnes filia alterius Thurstani qua fuit quasi uxor Bartholomei presbiter(i) ibid., fo. 44a (Dilham; Norwicensis = 'ofNorwich'). -Rogerusfilius Turstani 1200 CRR 24 (Cur, i, p. 310) (Witton). - Willelmo filio Turstan', abl. (witn.) c. 1200 CAcre, fo. 3la(v) (Rougham). -Lambertusfilius Sturstani de Weseh(am)torp' c. 1200 CAcre, fo. 36a(iv) (Wesehamtorp', i.e. the porp attached to Weasenham). - Walterus filius Thurstani nepos meus c. 1200 Wymondham, fo. 62b (Wadeker in Wymondham; the charter was issued by Richard, son of Ralph the clerk of Bunwell). - Thurstanus de Marisco 1201 (13th) Rams, iii, p. 295 (Walsoken; Thurstanus de Marisco isa contemporary of the grandfather of a tenant in the present document of 1201, and so he must have lived towards the middle ofthe 12th century). -Turstanus Taissun 1201 CRR 25 (Pleas, iii. no. 427) (Nf). -Godwinumfilium Turstan' 1202 FF (PRS NS, xxvii, nos. 279,351) (Lynn; Fodderstone, Clackclose H (lost)). -Haie filium Turstan' 1202 FF (PRS NS, xxvii, no. 407) (Holt). -Rogerumfilium Turstan' 1202 FF (PRS NS, xxvii, no. 430) (Setchey). perplegiumRicardif Turstani 1203 AssRoll 817 (NthRS, v, no. 780) (Nf), Ricardusf Turstan' 1203 AssRoll 817 (NthRS, v, no. 831) (Motton', i.e. Moulton St Michael). Turstanus Paumer, Petrusf Turstein 1203 AssRoll 817 (NthRS, v, no. 802) (the document concems Martham). -Decena Turstani 1203 P, p. 247, 1204 P, p. 240, de decena Turstani 1205 P, p. 246 (Nf or Sf). - super mesuagium Thurstani Duue c. 1205 CAcre, fo. 102a(ii) (Kirtlinge near Bittering). -Turstanus molendinarius 1206 P, p. 34, 1208 P, p. 9, 1209 P, p. 52, Turstanus Mol' 1207 P, p. 175 (Nf or Sf). - Gaufridumfilium Turstan' et Matillidem uxorem suam 1207 FF (PRS NS, xxxii, no. 122), Willelmumfilium Turstan'. et Godam uxorem suam, Gaufridumfilium Turstan' 1206 FF (PRS NS, xxxii, no. 123) (Docking). -Turstanfilius Botild' 1209 AssRoll 558 (Pleas, iv, no. 4151) (Nf). -Thurstan Brun 1209 AssRoll 558 (Pleas, iv, no. 4344) (Carlefled; Carlefled is unidentified but was situated on the coast). - Willelmusf Turstan' 1209 P, p. 53, de Willelmo f Turstin' 1210 P, p. 51 (Nf or Sf). - per Turstanum de Munfort 1210 P, p. 30 (Mundford). - ad portam Willelmi Thurstani c. 1210 Wals, fo. 63b(iii) (prob. Hindringham), terram Willelmi filii Thurstani c. 1215 Wals, fo. 63a(iii) (Hindringham). - Gileberto rufo et Turstano fratre eius, abl. (witn.) c. 1215 CAcre, fo. 94a(iv) (Haringeshage). God' filio Turstan', abl. (witn.) c. 1215-1220 CAcre, fo. 75b(iii) (Halech, prob. in Emneth; in this document Ricardus filius Estmundi de Wel!' confirms to Peter son of Gilbert of Emneth land which the said Peter held in Halech ). - Thurston' 1222 ElyA, fo. 125b (Northwold). -Achard filius Purstan' 1222 ElyA, fo. 157b (Emneth). - Radulfus filius Purstan' 1222 ElyA, fo. 182a, Ricardusfilius Purstan' ibid., fo. 183a (Bridgham). - Purstan'. stiward' 1222 ElyA, fo. 186b, Purston Elwin' ibid., fo. 189a (Pulham). Turstan'. dealbator 1222 ElyA, fo. 201 b (East Dereham). - Willelmo Turstan, abl., 1226 CRR 94 (Cur, xii, p. 361) (Norwich). - Ricardus filius Thurstan de Bunewelle 1227 JMN, p. 210 (Bunwell). -Reinerumfilium Thurstani 1229 CRR 105 (Cur, xiii, p. 283), Reinero filio Thurstani, dat., ibid. (Cur, xiii, p. 324) (Nf), Reinerus filius Thurstani, Thurstanus pater Reineri, Thurstanus pater suus (bis) ibid. (Cur, xiii, p. 346). (the document is a plea conceming the right of presentation at the church of East Beckham). Reginaldumfilium Thurstani 1230 CRR 104 (Cur, xiii, p. 570) (the document concems Beechamwell). -de Thurstano palmera, de predicto Thurstano, eidem Turstan', dat., c. 1230--1240 Wals, fo. 15b(iii) (Little Walsingham), terram ... scilicet que fuit quondam Thurstanni le paumer c. 1250 Wals, fo. 13b(i), de terra que fuit quondam Thurstanni palmen· ibid., fo. 14a(i) (Nf), Quiet' clam' Iohannisfilii Thurstan' palmerij, Johannes filius Thurstani palmerij de Parua Walsingh' ibid., fo. 20b(ii) (Walsingham; Parua Walsingh' is Little Walsingham). - terram quod (sic) fuit Thurstanfilii Alani c. 1235

428

l>orsteinn, l>orsten, Thorsten, l>orstreinn Der, fo. 195a (Wiggenhall St Germans). 115 - terram Thurstani fabri 1237-1251 Holme, fo. lOOa(ii) (Nf; the document isa grant by Richer of Stratton 116 to the Abbey of St Benet of Holme). - terram Thurstani pistoris c. 1240-1250 Der, fo. 59a(i) (Barton Bendish). - Carta Thurstanni de Bekeswelle, Thurstanus filius Hugon' de Bekeswell', predictus Thurstanus c. 1240-1250 Der, fo. 58b(iii), Carta Thurstanni Cachare, Thurstanus filius Hugonis de Bekeswell ibid., fo. 58b(iv), predictus Thurstanus de Bekeswell' ibid., fo. 59a (Barton Bendish), terram Thurstani fratris mei (ter), terram Thursta(ni)fratris mei ibid., fo. 59a(i) (Barton Bendish and Boughton; the charter was issued by Michael Cachehare de Bekeswell), terram Thurst' de Bekeswel', terram predicti Thurstan' ibid., fo. 59a(ii), Confirmacio Thurstanni de Bekeswelle, Thurstanus filius Hugonis de Bekeswell' ibid., fo. 59b(ii), de fedo quondam Michael et Thurstani Cachehare 1285 Der, fo. 67a(i) (Barton Bendish; Bekeswelle, Bekeswell' etc. represent Bexwell).

ON Porsteinn etc. is extremely frequent throughout Scandinavia from the

Viking period onwards (NiyR, nos. 42, 148, 151, 175, 178, 454, 510, 536, 592; Lind, cols. 1207-1211; DR, nos. 116, 130, 169, 117 260, 372, 384, mp. 108-111; DgP, cols. 1400-1406, 1695; DS IX, p. 112; DS XII, p. 45; SRÖl, no. 46; SRÖg, nos. 42, 96, 105, 188, 225, 237; SRSö, nos. 29, 54, 55, 61, 84, 110,208,248,254,307,338,343,360,363; SRSmå, nos. 93, 96; SRVg, nos. 3,4,6, 10, 17,116,175; SRU,nos. 7,53,93, 104,131,144,146,209,275, 279,338,360,372,418,426,448,481,482,510,628,653,732,840,843, 844,848,907,917,947,955,987,998, 1034, 1075, 1152, 1157, 1165;Lundgren-Brate, pp. 272-273; Modeer, Personnamn i Ktb, pp. 17, 83, 89-90, 139140; Modeer, Östsmåländska bondenamn från 1500-talet, p. 63; Otterbjörk, Vår äldsta nämnd, pp. 89-90; Thors, p. 92; Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183, 185, 186, 187, 196 (Torsten); Fredriksson, Svenskt dopnamnsskick vid 1500-talets slut, pp. 146-147, s.n. Torsten). ON Porsteinn etc. is one of the most frequent Scand personal names in English records, where it most commonly appears in the Anglicized forms Purstan, T(h)urstan etc., in which OScand -stceinn, -steinn has been replaced by the corresponding OE -stan, and, in the post-Conquest period, in the Norman form Turstin(us), and it is also well attested as a first element of English place-names and field names: NPN, pp. 161-162; ZEN, pp. 87-88; Ashdown, p. 295; PNDB, p. 396; Tengvik, p. 202; SPNT, p. 145 and n. 6; ELPN, pp. 74, 83-84; Newark, pp. 1, 14 (Turstein buc); DBS, p. 348, s.n. Thurstan; SPNLY, pp. 313-317, 356; Smart, Moneyers, pp. 230,236,238, 243; Fellows Jensen, Lincolnshire Tenants, pp. 93, 94; SCBI, xxi, pp. xlvi, xlvii, xlix; von Feilitzen, PNWinton, p. 175; Insley, Nomina, iii (1979), pp. 54, 56, 57; Lindkvist, pp. 95-96; DEPN, pp. 470, 471, 472, s. nn. Thrislington, Throxenby, Thrussington, Thurstaston, Thurston Sf in Hawkedon, Thurstonfield, Thurstonland, Thuxton; PN

115 116

117

For Thurstan son of Alan see also above, n. 111. Probably Stratton Strawless. For DR no. 169 see also NiyR, v, pp. 234-235.

429

l>orve, Thorwe NbDu, p. 196; PN La, p. 192; PN Wo, p. 398; PN NRY, p. 110; PN Ess, pp. 634---635; PN ERY, pp. 257, 322; PN Nt, pp. 283,293; PN C, p. 340; PN Cu, p. 128; PN WRY, ii,pp. 108-109, 251-252, 317; PNWRY, iv,pp. 109-110; PNWRY, vi, p. 192; PN Gl, iv, p. 209; PN Ch, iv, pp. 279-280; SSNY, pp. 7, 106; SSNEM, p. 195; Insley, JEPN, x (1977-1978), pp. 44, 52, 62, 69. In Sfthere are numerous examples ofthe name in the Bury cartularies and elsewhere.

In Normandy Adigard des Gautries lists no less than 60 bearers of the narne from the period between the latter part of the 10th century and the Conquest, and the narne is frequent in later Norman sources, though, rather surprisingly, it forms the first element of only three Norman place-names (Adigard, pp. 168-169, 326-340, 431). From c. 1030 the usual Norman form is Turstinus, which gives way in the course of the 12th century toa secondary derivative, Tustinus (Adigard, pp. 168-169, and p. 169, n. 48). The use of the narne in Normandy was of prolonged duration, and it occurs regularly throughout the 13th century as Toustain, Tostain, Tustinus, Tustanus etc. (Adigard, p. 169, n. 48). Many of the post-Conquest English bearers of the narne, especially in the form Turstin(us), in the knightly classes and in the ecclesiastical hierarchy were Normans or of Norman descent. An early exarnple of a Nf bearer of the narne of Norman origin is the 1086 DB tenant Turstin fitz Guy. For initial Pur- and T(h)ur-, which formally reflect OEScand (Danish) Pursee above, s.n. P6raldr. For initial T(h)- for OScand P- in T(h)urstan, Turstein etc. see above, s.n. P6raldr. For the survival of OE, OScand initial P- in ME as in Achard filius Purstan' etc. 1222 ElyA, see above, s.n. P6rör. For the replacement of OScand -st::einn, -steinn by the corresponding OE -stän see NPN, p. 203; PNDB, § 41; SPNLY, § 52. Von Feilitzen, PNDB, § 40, incorrectly took -inus in the Norman Turstinus to reflect the Danish change of weakly stressed ::ei > e > i, though he added that the influence of the Latin suffix -inus should perhaps also be considered. The form Turstinus only begins to become dominant in Normandy around 1030 (see above), and final -inus here clearly represents a native Norman development, possibly due to association with the Latin suffix -fnus. 118

ON l>orve, ODan Thorwe fem. b) (i) Ulricus filius Turuue 1138-1147 CAcre, fo. 4a(i) (Nf).

This name is extremely uncommon (Lind, cols. 1216-1217; DgP, col. 1415),

118 The earliest Norman examples of ON Porsteinn etc. occur as Torstingus, Torstenc, Thorstenus, Turstenus etc. (see Adigard, pp. 168-169, 326 ff.).

430

*I>urw(e)ard, *l>orward, I>orvarör, Thorwarth though the etymologically identical ODan Thyrwi is well attested (DgP, cols. 1444--1447). The above Nf example is the only example of ON Porve, ODan Thorwe to have been noted in England. For initial Tur-, which formally reflects OEScand (Danish) Pur-, see above, s.n. P6raldr. For initial T- for OScand P-, which is an AN feature, see above, s.n. P6raldr.

OScand *l>rykki < ON l>ryörikr a) (i) Trukebei 1086 DB, fo. 135, Trikebei ibid., fos. 180, 197, 225b (Thrigby, East Flegg H).

Ekwall, DEPN, p. 470, s.n., takes Thrigby Nf to contain as first element a Scand personal name *Prykki, a short form of ON Prjör[kr. ON Prjör[kr is attested in medieval Norway and also occurs as pririks, gen., in an 11 th century runic inscription (Lind, cols. 1224-1225; NiyR, no. 68), and there seems no reason to doubt Ekwall' s explanation of the first element of the Nf Thrigby as a short form of this name, which is also accepted by Sandred 1987:317f.

AScand *l>urw(e)ard, *l>orward, ON l>orvarör, ODan Thor-

warth b) (i) ... and at Metheltone into pere kirke pe Purwerd ahte m. 11th (l. 13th) ASWills, no. 33 (Great Melton). (ii) Carta Thorward filii Herlewini de Crimplisham, Thurword filius Herlewini de Crimplisham e. 13th Der, fo. 16b(i) (Fulewelle in West Dereham; Crimplisham is Crimplesham).

The OE form Purwerd is an AScand hybrid formed from OEScand (Danish) Pur- and OE -weard. The 13th century Thorward, Thurword may reflect a similar hybrid compound of OScand Pär-, OEScand (Danish) Pur- and ME -ward < OE -weard, or it may stand for ON Porvarör, ODan Thorwarth. ON Porvarör is frequent in Iceland from the time of the settlement onwards and is well attested in Norwegian records after c. 1300 (Lind, cols. 1214-1216). ODan Thorwarth is rare, occurring sporadically in Danish records ofthe late 13th and 14th centuries (DgP, col. 1415). For further discussion and examples of AScand *Purw(e)ard, *Porward, ON Porvarör, ODan Thorwarth in England see the following: NPN, p. 163; SPNLY, p. 317; PN D, pp. 424-425, 687; PN Nth, p. 292. 431

*I>urwig

AScand *l>urwig b) (i) terrarn Ricardi filii Turewi c. 1170 AddCh 15508 (printed, SSMEA, Appendix I, no. 15). 119

*Purwig is an AScand compound of OEScand (Danish) Pur- and OE -wig (von Feilitzen, Notes, p. 64). For initial T- for OScand P-, which is an AN feature, see above, s.n. P6raldr.

AScand *l>urwine a) (i) Purwineholm prob. 1042/1043 (l. 13th) ASWills, no. 27 (lost,? in Loddon).

*Purwine is an AScand hybrid compound of OEScand Danish) Pur- and OE -wine.

OScand *l>ymli a) (i) Timeltorp (bis) 1203 FF (PRS NS, xxxii, no. 36), PimeljJorpe c. 1220-1240 CAcre, fo. 44a(iv) (Themelthorpe, Eynsford H).

Ekwall, DEPN, p. 465, s.n., interprets the first element of Themelthorpe as being either OE *Pymel or ON *Pymli, bynames formed from OE pymel 'thimble' and ON pumall 'thumb', respectively. The absence of spellings with medial -e-, reflecting an original genitive, between the first and second elements in the early forms of Themelthorpe suggests rather that the first element of this place-name is, however, the appellative OE pymel 'thimble' used in a topographical sense. Cf. also Thimbleby L, NRY, which do have spellings with medial -e- between the first and second elements and which are usually interpreted as containing the personal name (or byname) ON *Pymli, though SSNEM has also raised the possibility that these place-names could also contain an appellative identical in form with the personal name/byname OE *Pymel or ON *Pymli (DEPN, p. 465, s.n. Thimbleby; PN NRY, p. 214; SPNLY, p. 318; SSNY, p. 38; SSNEM, p. 74). Note also that ON Pumli < ON Pumall 'thumb' is on record as a byname (LindBiN, col. 410).

119 The present document is dated by SSMEA, Appendix I, no. 15, and von Feilitzen, Notes, p. 64, to the reign of Henry Il, but it can be more closely dated to c. 1170.

432

u ON (p. ns.), ODan, OSwed (run.) Ubbi, OSwed Ubbe b) (i) Matillam que fuit uxor Vbbe 1188 CAcre, fo. 114a(ii) (Kempston).

Ubbi is an EScand name. It is perhaps best explained as a hypocoristic form of

Ulfr, or possibly as a short form of names in Ulf-, -ulfr, though attempts have also been made to derive it from a root *ub 'unfriendly' or 'great' and to connect it with ON ufr 'hostile' and Norw ubben, ybben 'rough, severe' (Lind, col. 1047; SRSö, p. 223; DgP, col. 1511; NK VII, pp. 59, 159, n. 220; SPNLY, p. 319). ODan Ubbi is frequent in independent use from the Viking period onwards and is well attested as the first element of Danish place-names in -thorp (DgP, cols. 1507-1511, 1696). In Swedish the name appears in a runic inscription in Södermanland and is well attested in medieval records (SRSö, no. 255; Lundgren-Brate, pp. 182, 286, s. nn. Obbe, Ubbe). In OWScand Ubbi is known as the first element of Norwegian place-names (Lind, col. 1047). The most famous bearer of the name to appear in English sources is the notorious (H)ubba, son of Ragnar Loöbr6k, and one of the leaders of the Danish Great Army, which was active in England in the 9th century. 1 The name is, however, rare in England. In addition to the above Nf example, it has been noted as the name of a TRE tenant in Db in DB, and it forms the first element of Ubbeston Sf and possibly also of Obthorpe L, though in the case of the latter OE Ob(b)a and OE Ub(b)a have been suggested as altematives (PNDB, p. 397; DEPN, p. 465, s.n. Ubbeston; SPNLY, pp. 319-320; SSNEM, p. 114).

ON Ulfkell, ODan Ulfkil, OSwed (run.) Ulfkretill, Ulfkell, AScand Ulfcytel a) (i) Vlketelescroft m. 13th Der, fo. 264b (f. n. in Holkham). b) (i) Ulfcytel (quater) 1004 (m. 11th) ASC (C), Vlfcytel, Vfcytel 1004 (c. 1100) ASC (F), Ulfkytel, Ulfcytel (bis) 1004 (1121) ASC (E), Ulfcytel 1010 (m. 11th) ASC (C),

1 For this Ubbi/(H)ubba see Smyth, Scaruiinavian Kings in the British Istes 850-880, pp. 2, 2223, 28, 59,195, 197,200,202-204,228-230,269.

433

Ulfkell, Ulfkil, Ulfkretill, Ulfkell, Ulfcytel 1010 (1121) ASC (E), Ulfcytef onEastengfum 1016 (m. 11th) ASC (C), Ulfkytef ofEast Angfan 1016 (m. 11th) ASC (D), Ulfcytef of East Angfan 1016 (1121) ASC (E), 1016 (c. 1100) ASC (F) (these entries relate to Ulfcytel (ob. 1016), who commanded the forces ofEast Anglia against the Danes from c. 1004 to 1016). 2 - Ulfcetef 1009-1016 Smart, Moneyers, p. 243 (moneyer, Norwich). - Wufkitefe, dat., 1042-1053, possibly 1046 (13th) ASWills, no. 32 (Wulkitel and his brother Ketel are here bequeathed Nf properties at Walsingham in East Carleton, East Carleton and East Harling by their mother Wolgip).3 - Eadwine Vlfketeles sune ( witn.) 1043-1044 (1. 13th) ASChR, no. 97 (Swanton and Hindolveston). - Vfchetef fiber homo TRE DB, fo. 149b (Harling). - Ufketef. I. fiber homo. Edrici commendatus TRE DB, fo. 154 (Shotesham), Ufketel. I. fiber homo Edrici commendatus, VIII. liberi homines. Ufketelli. commendati ibid., fo. 155 (Woodton), Ufketef fiber homo Edrici commendatus ibid., fo. 174b (Burgh St Margaret), Ofketef. fiber homo Edrici de Laxefefda. antecessoris Roberti Malet ibid., fo. 260 (Fritton). - Ofchetef fiber homo TRE DB, fo. 165 (Lexham). - I. liber homo Ufketef commendatione tantum TRE DB, fo. 167 (Stanhoe). - Ufchetef. I. fiber homo Afgar comitis commendatus TRE DB, fo. 176 (Framingham). - Ufketef (homo Herofdi) TRE DB, fo. 228b (Ulketel and another man ofHarold, Withri, held sokemen and bordars in Woodbastwick), Ufketef ibid., fo. 228b (Woodbastwick). - Vfketef TRE DB, fo. 234 (Haveringland). - Terre. Vfchetelli 1086 DB, fo. 270, Vfchetell TRE, 1086 DB, fo. 270 (Larling). - Vfchetef TRE DB, fo. 275 (Stoke Ferry). - Ufketef TRE DB, fo. 124 (held the commendation of one fiber homo and half the commendation of another in Bramerton or Rockland St Mary), Ufketel (gen.) (bis), Ulketelli, gen., ibid., fo. 175 (held 30 liberi homines in Surlingham, and the commendation of 14 integri fiberi homines in Rockland St Mary and 9 liberi homines in Bramerton), Ufketef (gen.) ibid., fo. 175b (held 3 integri fiberi homines in Kirby Bedon), Ufketef (gen.) TRE, Vfchetef 1086 DB, fo. 176 (held 4 fiberi homines at Kirby Bedon; Ulketel also held fiberi homines at Holverston (TRE), Yelverton (TRE, 1086), Shotesham, Stoke Holy Cross and Surlingham, and integri liberi homines at Poringland (all TRE)), I. fiber homo. Ufketelli TRE DB, fo. 176 (Rockland St Mary), I. liber homo Ufketelli. commendatus TRE DB, fo. 176 (Bixley), VIII. liberi homines Ufketelli TRE, Vfketef 1086 DB, fo. 177 (Mundham), Idem Vfketef TRE, ? 1086 DB, fo. 177 (Seething), XIII. fiberi homines Ufketelli TRE DB, fo. 177 (Afgamundestuna; Ulketel also held fiberi homines at Claxton and Woodton (TRE)), Vfketef 1086 DB, fo. 177 (held one fiberafemina at Norton Subcourse), I. liber homo Vfketelli TRE, ? 1086 DB, fo. 177b (Seething), Vfchetef fiber homo TRE, de feudo Vfchetelli 1086 DB, fo. 182b (Norton Subcourse), Vfchetef TRE, feudum Vfchetelli 1086 DB, fo. 182b (one fiber homo, Bondo, with 30 acres in Heckingham), Ufketef (acc.) TRW DB, fös. 176b, 177, iste Ufketef, Vfketef 1086 DB, fo. 177 (Mundham), Vfchetef 1086 DB, fo. 176 (Framingham). 4 - Vfchetef homo Hermeri TRW DB, fo. 213 (claimed landat Fodderstone, Clackclose H (lost) against the Abbey of Ely). - Vlfcitef Wml (c. 1083-1086) SCBI, xviii, no. 1334 (moneyer, Norwich). -Vlfkitefo homini nostro de Dilham, abl., 1101-1107 (14th) Holme, Appendix A (Dilham), dimidium possessionis Vlfketefi ibid. (Thume). 5 - Hulfketello, abl. (witn.) c. 1115 CAcre, fo. 26a(i)

For Ulfcytel of East Anglia see ASE, pp. 380-381, 383, 392. Fora genealogical table for this family see above, s. n. Ketill, fig. 3, p. 263. 4 The full DB entry here reads; 'In Framingaham ten(uit) Ulchetel. I. liber homo Algar comitis commendatus. I. carucatam teme. modo tenet Vlchetel.' The absence of the word 'idem' before the second Vlchetel indicates that he is probably not to be identified with the first who is, therefore, entered separately above. 5 Though this document was drawn up in the period 1101-1107, the entries conceming Ulfketel 2

3

434

Ulfkell, Ulfkil, Ulfkretill, Ulfkell, Ulfcytel (charter of Roger son of the Earl of W arenne' s seneschal Wimer by which he confirms the grants made to Castle Acre Priory by himself and by his father Wimer and his mother Gilla). - Uljkytello presbitero, abl. (witn.) 1127-1134 HarlCh 44 E 19 (printed, Holme, Appendix E) (Heigham next Norwich), Ulfketel presbiter (witn.) 1127-1134 (14th) Holme, no. 130 (Rollesby). - Vlchetellus presbiter 1129-1130 PR, p. 92, de pecunia Vlchetelli monetarii ibid., p. 94 (Nf or Sf). - (De) Vlchetello de Heringeshae, Vlchetellum c. 1130 CAcre, fo. 39b(i), de Ulchetello de Heringishae ibid., fo. 94a(vii), In Haringeshage Vlchetellum cum liberis suis et omni tenura sua 1146-1147 CAcre, fo. 4a (Heringeshae, Haringeshage). -Hulf Ketel Ruffo de Wertona (abl.) (witn.) c. 11501200 (17th) BS, no. 215 (Stoke Ferry; Wertona is unidentified, unless it is Wretton, which lies adjacent to Stoke Ferry). - Robertus clericus filius Vlfketel (witn.) 11531168, prob. 1162-1168 Holme, no. 876 (confirmation to the Abbey of St Benet of Holme of a third of the tithes at Ingham and Waxham, two thirds of the tithes at Worstead and Brunstead, anda halfof the tithes at Horsey). - Ulfketelus (bis) 1161-1177 (13th) Rams, iii, p. 263, Wlfketelus ibid., p. 265 (Brancaster, Burnham, Burnham Deepdale ). - Vlfketellus mercator c. 1165-1175 CAcre, fo. 35b(iii) (Wesenhamtorpe, i.e. the jJorp attached to Weasenham). - De terra quam Ulfketel Stoche tenuit, terram quam Ulfketel Stoche tenuit c. 1165-1178 CAcre, fo. 87a(i) (prob. Watlington). - De Ulfketel filio Aildrici, Ulfketel filium Aildrici de Marham c. 1165-1180 CAcre, fo. 85b(ii) (Marham). - Vlfketilfilius Gunne 1174-1175 P, p. 121, Ulfketelfilius Gunne 11751176 P, p. 67 (Nf or Sf). - totam terram illam quam Vlfkitellus dapifer dedit deo et ecclesie beati Benedicti et monachis ibidem 1175-1186 Holme, no. 228 (Yarmouth). Husketel de Snarehala 1180-1181 P, p. 89, Willelmusfilius Vlketel de Snareshala ibid., p. 91 (Ul)ketel de Snareshala 1181-1182 P, p. 70, Willelmus filius Vlketell' de Snareshala ibid., p. 72, Ulfketel de Snareshala, Willelmusfilius Ulfketell de Snareshala 1182-1183 P, p. 11, Ulfketel de Snareshala, Willelmus filius Ulfketel de Snareshal' 1183-1184 P, p. 5 (? Snarehill).7 - Ulfketelfilius Eu(er)ardi de Upwelle c. 1180 CAcre, fo. 78b(iv) (Upwell). - hominem meum Alanum videlicet filium Wuluiue del Haghe cum tota terra sua que fuit Ulfketelli parui de Norwude c. 1185 Wymondham, fo. 37a(ii), Alanum filium Wuluiue de Hage cum tota terra sua que fuitllketelli parui de Northwude ibid., fo. 37b(i) (charters of Agnes de Risley), hominem nostrum Alanum nomine filium Wuluiue del Hage cum tota terra sua que fuit Ulfketelli parui de Northwude ibid., fo. 37b(iii) (charter ofGilbert de Malet and his wife Agnes, i.e. Agnes de Risley), cum tota terra que fuit Ulfketel parui de Northwde ibid., fo. 39a(i) (charter of Robert de Risley) (Nf, prob. Wicklewood; Nor(th)wude, Northwde is unidentified). - Vlketel de Sutton 1199 P, p. 277, de Vlketel 1200 P, p. 138 (Nf or Sf; Sutton can be Sutton near Mileham Nf, Sutton near Stalham Nf or Sutton St). - et quatuor acras terre quas Ulfketel filius Godrici tenuit 1200 FF (PRS NS, xxvii, no. 243) (Pickenham). - terram Ricardi filii Ulfketel c. 1200 CAcre, fo. 36a(i) (Wesehamtorp, i.e. the jJorp attached to Weasenham). - terram que fuit Reinerifilii Wlfketele e. 13th Der, fo. 75a(i) (Watlington). - Ulfketel 1201 (13th) Rams, iii, p. 299 (Outwell). - Ulfketell' de Derebi et Hawisam uxorem

of Dilham and Ulfketel of Thume relate to a period somewhat earlier, in the latter part of the 11 th century (see Stenton, EHR, xxxvii (1922), pp. 226 and notes 1 and 2,228 and n. 5). 6 The witness Robert the clerk son of Ulfketel does not occur in the copy of this charter contained in the Holme cartulary itself, but has been supplied by the editor of Holme from the copies contained in the British Library Cotton Rolls and the Bodleian Rolls. 7 von Feilitzen, Notes, p. 57, incorrectly takes the form Husketel de Snareshala 1180-1181 P to imply an ON *Husketill. The other forms relating to this person show that the etymon in Ulfke(ti)ll and that the -s- of the 1180-1181 form isa mistake.

435

Ulfkell, Ulfkil, Ulfkretill, Ulfkell, Ulfcytel suam (acc.), Uljketell' (dat.) 1202 FF (PRS NS, xxvii, no. 341) (Bilney; Derebi is unidentified).8 - Vlketel Burnel 1203 P, p. 114 (Nf or St). - Ulketellum Tibel, Ulketellum de Stakes 1203 CRR 32 (Cur, iii, p. 8) (Wereham; Stakes is Stoke Ferry). - Uljketellum Bud' 1203 CRR 32 (Cur, iii, p. 56), Wlketellus de Well' ibid. (Cur, iii, p. 73), Hulketel Bude (acc.), predictus Hulketel (bis), predicto Hulketel, dat., 1204 FF (PRS NS, xxxii, no. 44) (Upwell). - ad drouam illius terre Uljketll' Elismoder9 c 1205-1210 CAcre, fo. 78b(ii), ad drouam illius terre que fuit Uljketel Elesmoder ibid., fo. 78b(iii), pratum quod fuit Hulketel Helesmoder 10 c. 1210 Der, fo. 87a(i) (Upwell). - De terra Uljketel Crovai, in terra de Heringeshae que quondam fuit Vljketel Crouaj aui mei c. 1210 CAcre, fo. 94a(v) (Heringeshae; the document isa charter of John son of Richard of Letton). - Willelmus Rufus de Wretton' filius Vljketelli c. 1210-1220 Der, fo. 38a(i) (the document concems the homage and tenement of Vincent of Wereham; Wretton' is Wretton). - terram Uljketel filii Estmundi c. 1210-1220 Der, fo. 92a(i) (Upwell). Wimer Vljketel 1222 ElyA, fo. 123b (Feltwell). -Radulfusfilius Vljketel 1222 ElyA, fo. 171b (West Walton). - Vljketel Must 1222 ElyA, fo. 203b, Ricardus Vljketil ibid., fo. 204a (West Dereham). -Lefquen Vljketil, Vljketel Tuleboi 1222 ElyA, fo. 209b, Adam Vljketel ibid., fo. 210a, Alanus Vljketel ibid., fo. 21 la, Vljketel fab' ibid., fo. 212a, Elfstonfilius Vljketel ibid., fo. 213a (Thorp in Shipdham). - Bonde Vljketel, Bonde filius Huljketel de Rondeshil c. 1225 Co:xford Il, fo. 71 b(iv) (Lynn). - terram que fuit Wljketel Koc c. 1225 Der, fo. 201a (Wiggenhall). -terram que fuitHulketel auunculi mei c. 1235 Der, fo. 21 la(i) (Wiggenhall St Mary Magdalene; the document is a charter of Landricus Grey son of Eudo Grey). - Walingerum filium Ulkeliti (sic) 1230 CRR 107 (Cur, xiv, p. 176) (Lynn). - Gilbertus Ulketel c. 1235 (m. 14th) Rams, i, pp. 416,417 (bis) (Brancaster). - de hornagio Hulketel filii Radulfi, homagium Huljketel filii Radulfi c. 1235 Der, fo. 221 b(ii) (Wiggenhall). - de hornagio Edmundi filii Vlketel Brung', homagium Edmundi filii Vlketel Brunger de Wig' c. 1235-1245 Der, fo. 208a(i), Carta Edmundi filii Vlketul Brunger, Edmundus filius Vlketel Brunger ibid., fo. 208a(ii), Edmundo filio Hvlketel Brunger de Wig', dat., ibid., fo. 208b (Wiggenhall). - Ulfeketel Mayl, Ulfeketel Mayle (acc.) 1236 CRR 116A (Cur, xv, p. 429) (Nf, Clackclose H). -Hugo filius Uljketel c. 1245 (13th) Bec, p. 116, Stephanus filius Uljketel ibid., p. 117 (East Wretham). (ii) Siger Vljkel 1222 ElyA, fo. 209b (Thorp in Shipdham).

ON Ulfkell occurs as the name of a thrall in the Landnamab6k, but the nationality of this person is unknown, and so it is not clear whether the name was native to OWScand (Lind, col. 1053). In OEScand the name is infrequent. ODan Ulfkil occurs as an 11 th century moneyer's name at Lund, as the name of a 12th century priest at Lund, and as the name of a bishop of Århus who died in 1134 (DR, M0. 118-123; DgP, col. 1528). In OSwed the name is known from runic inscriptions in Uppland, where both the full form Ulfkcetill and the contracted Ulfkell, Ulfkil are on record (Ulfkcetill: SRU, nos. 160, 633: Ulfkell, Ulfkil: SRU, nos. 100, 161, 227,479). 8 It is just conceivable that this IBtketel may have been a migrant to Nf from Derby Db or from Darby L. West Derby La can probably be ruled out as a source on grounds of distance. 9 Sic! 10 MS. hulketelhelesmoder.

436

Ulfr, Ulf In contrast to its relative rarity in Scandinavia itself, Uljke(ti)ll, Uljkil etc. is frequent in England, forms with syncopation of the second element to -kel, -kil etc. predominating in the Northern Danelaw and full forms in -ketel, -chetel etc. being practically universal in Bast Anglia: NPN, pp. 168-169; ZEN, p. 91; Björkman, Festermen, pp. 5, 17; PNDB, pp. 399--400; SPNT, p. 145; ELPN, p. 84; Newark, pp. 3, 14 (W)ulfchel; DBS, p. 359, s.n. Uncle (ii); SPNLY, pp. 325-327; Smart, Moneyers, pp. 230,236,243; von Feilitzen, SCBI, xi.i, p. 13; Fellows Jensen, Lincolnshire Tenants, p. 94; Smart, Cnut's York Moneyers, p. 228; von Feilitzen, PNWinton, p. 175; Insley, JEPN, xi (1978-1979) p. 68; DEPN, pp. 347, 354, s. nn. Oakerthorpe, Ouston Du, Nb nr Stamfordham; PN NbDu, p. 153; PN Nth, p. 292; PN Cu, p. 343; PN Db, pp. 335-336; PN WRY, ii, p. 247; PN WRY, iii, p. 158; PN WRY, v, pp. 22, 38; PN WRY, vi, p. 127; SSNY, p. 71; SSNEM, p. 130. Cf. also the following Sf examples: Ulfchetel 1087-1098 (c. 1175-1200) Bury, pp. 25, 41 (Great Barton, Coney Weston); Ulketel 1186-1188 (c. 1270) BuryS, p. 8 (Fomham St Genevieve). Thomas filius Ulketel ibid., p. 44 (Troston); de tota terra cum pertinentibus que fuerat Vlfketelli presbiteri in Mildenh(all) c. 1189-1198 (l. 13th) BuryS, no. 109 (Mildenhall).

Initial Wul- and Wl(f)- reflect the replacement of OScand Ulf- by the corresponding OE Wulf-. Loss of interconsonantal 1- in English forms in (H)ulketel etc. may reflect an English or a Scand development (PNDB, § 92; SPNLY, § 100.i; Bnzmdum-Nielsen, § 377). Initial H- in Hul(f)ketel etc. is inorganic and the result of AN influence (PNDB, § 138; SPNLY, § 140.i). For the development of the second element, OScand -ketill/-kell/-kil in England and Scandinavia see above, s.nn. Arnkell, Askell. The form Ilketelli, gen., c. 1185 Wymondham is the result of confusion of Uljketill etc. with *Yljketill, an unrecorded side-form with secondary i-mutation, which also forms the first element ofllketshall Sf (DEPN, p. 262, s.n.).

ON lJlfr, ODan, OSwed Ulf a) (i) Uluestuft 1202 FF (PRS NS, xxvii, nos. 273, 303) (f. n. in Tivetshall). -Oluesbrom c. 1210 Wals, fo. 92b(ii) (f. n. in Erpingham). -a cruce Vlfc. 1210 Wals, fo. 113a(iv) (in Wood Norton). b) (i) Vlf ::et Welle (witn.) 1043-1044 (1. 13th) ASChR, no. 97 (Swanton and Hindolveston; Welle is unidentified). -Vlf(witn.) 1043-1045 (l. 13th) ASWills, no. 31 (Nf). Vlfus TRE DB, fo. 121b (Cranworth). - Vlfus TRE DB, fo. 128 (Burnham Ulph). - Vlf. I. liber homo TRE DB, fo. 131b (Thuming). - Olfus homo antecessoris Roberti Malet TRE DB, fo. 135b (Raveningham). -Olfus teinnus TRE DB, fo. 150 (Hethersett), Olfus. teinnus ibid., fo. 180 (Hethel), idem Olfus ibid., fo. 180b (Ketteringham), Olfus ibid., fo. 180b (East Carleton and landat Flordon and Newton Flotman belonging to the desmesne ofHethel), ibid., fo. 181 (Fomcett (St Mary), Ulfi, gen., (8x) ibid., fo. 185b (held

437

Ulfr, Ulf liberi homines at Shotesham, Rockland St Mary, Surlingham and Saxlingham Nethergate and the commendation of liberi homines at Fritton, Framingham, Bramerton and Shotesham), sub Olfo (bis) ibid., fo. 178b (liberi homines in East Carleton and Flordon were in commendation and owed fold soke sub Olfo), Leustan. I. liber homo Ulfi commendatus ibid., fo. 259 (Starston). - Ofl liber homo TRE DB, fo. 178b ((North) Lopham). - Vlf. I. liber homo TRE DB, fo. 226 (Hargham). - Olfus TRE DB, fo. 229 (Taverham). - VlfTRE DB, fo. 247 (Hindringham). - Olfus. liber homo TRE DB, fo. 252b253 (Wicklewood). - Ulf 1101-1107 (14th) 11 Holme, Appendix A (Fleg, i.e. the hundreds of East and West Flegg). - Wlfo presbitero de Barsam, abl. (witn.) 1101-1107 Binham, fo. lb (grant by Peter de Valoignes to Binham Priory; Barsam is Barsham). terram Wlf de la Wella 1118-1138 CAcre, fo. 3a(ii), terram Ulf de la Walle 1146-1147 CAcre, fo. 3b(i) (Wickmere). -terram Ulf c. 1140 Holme, no. 66 (Nf). - Ulf (decanus) 1135-1143 NorwCPCh, no. 116 (charterof Bishop Everard of Norwich). - Radulfo filio Ulft, abl. (witn.) c. 1160 CAcre, fo. 57b(i) (the document concems the mill ofHolt held by Turstan the miller), de Radulfo filio Vift, de Radulfo filio Ulft c. 1165 CAcre, fo. 57b(ii) (Cockley Cley). - Uluus Bunting 1161-1177 (13th) Rams, iii, p. 262, Wlvus ibid., p. 265 (Brancaster, Bumham, Bumham Deepdale). - .tEdwinus et Vlf frater eius 1165-1166 P, p. 30, 1166-1167 P, p. 30 (Nfor Sf). -Iohannesfilius Vlf 1176-1177 P, p. 131, 1177-1178 P, p. 24, 1178-1179 P, p. 5, 1179-1180 P, p. 17, 1180---1181 P, p. 83 (Nf or Sf). -Rogero filio Hulft, abl. (witn.) c. 1180 HMC VarColl, vii, p. 230, no. 325 (Sustead). -deRobertofilio Vlfl 179-1180 P, p. 21 (Nf). -terram que fuit Vlfpresbiteri 1182 Wymondham, fo. 10a (Colton), ex donoHugonisfilii Vlf ibid., fo. 10b (Nf). -(pro plegio) Vift 1184-1185 P, p. 40, 1185-l 186P, p. 69 (Nf). -Ricardofilio Ulf, abl. (witn.) c. 1190 CAcre, fo. 44a (East Barsham). - Vlfus mercator c. 1190---1200 Binham, fo. 5b (printed, SSMEA, Appendix Il, no. 3, p. 269) 12 (Binham). - terram que fuit Ulft Fretehois c. 1190---1205 HarlCh 45 F 15 (printed, SSMEA, Appendix I, where it is dated to the late 12th century) (Hunworth). - Rannulfus f Vlf 1195 P, p. 75, Rannulfus f Vlf 1196 CR, p. 134 (Nf or Sf), Rannnulfumfilium Ulf 1209 AssRoll 558 (Pleas, iv, no. 4297) (Nf). - Vlfum de Sueinestorp et Willelmum filium eius, Vlfo et Willelmo filio eius, dat., Vlfus 1196 FF (PRS, xvii, no. 214), in terra et hereditate Vlfi 1204 P, p. 246 (Swainsthorpe). - Decena Simonis f Vlfi 1197 P, p. 243, 1198 P, p. 88 (Nf or Sf). Hardewinumfilium Ulf 1198 FF (PRS NS, xxvii, no. 30 (Witchingham), Hardwin 'filius Ulf 1206 CRR 42 (Pleas, iii, no. 1919) (Nf), Hardwinumfilium Ulft 1209 AssRoll 558 (Pleas, iv, no. 4354) (the document concems Witchingham). - UlffiliumAsketel,jus et hereditatem predicti Ulf, predictus Ulf, predicto Ulf, dat., 1198 FF (PRS NS, xxvii, no. 89) (Berham). -Amel' filium Ulft de Linna 1198 FF (PRS NS, xxvii, no. 101) (Lynn). - Vlfus 1199 P, p. 289, 1200 P, p. 145, 1201 P, p. 136, 13 1202 P, p. 108, 1203 P, p. 239 (Nf or Sf). - Walterusfilius Ulf, W.filius Ulf c. 1200 Wymondham, fo. 62a(iii) (Wreningham). -homagium Vift et uxoris sue et liberorum suorum c. 1200 Wals, fo. 126b(ii) (charter of William son of Constantine son of Godric ofLynn). - UlfumLuuel et Leciam uxorem suam, jus et hereditatem ipsorum Ulft et Lecie, predicti Ulfus et Lecia, de predictis Ulfo et Lecia, prefatis Ulfo etLecie, dat., 1202 FF (PRS NS, xxvii, no. 315) (Snettisham). - Stannardum filium Hulf 1202 FF (PRS NS, xxvii, no. 376) (Hardley). - ter-

11 Though this document was drawn up in the period 1101-1107, most of its contents relate to the period between the Conquest and the end of the 11 th century (see Stenton, EHR, xxxvii (1922), pp. 225-235). 12 SSMEA, Appendix Il, no. 3, p. 269, wrongly renders the present form as Ulfus mercador. 13 1201 P, p. 136, supplies this form from the corresponding CR.

438

(Jifr, Ulf ram Hugonis filii Hulf 1203 FF (PRS NS, xxxii, no. 23) (Tittleshall). - Regina/dum et Stephanumfilios Ulfi 1203 AssRoll 817 (NthRS, v, no. 874) (Harling). - medietatem servicii Ulfi de Hehedich 1203 CRR 31 (Cur, ii, p. 289) (West Winch), Vlfus de Hegeding 1205 P, p. 253 (Nf; Hehedich, Hegeding is unidentified). - Robertus f Vlfi 1205 P, p. 251 (Nf or Sf). - Roberto filio Ulf, dat., 1209 AssRoll 558 (Pleas, iv, no. 4398) (Kelling). - Ediuam filiam Ulf 1209 FF (PRS NS, xxxii, no. 158) (Stratton', i.e. Stratton St Mary and St Michael or Stratton Strawless). - domum Vlf Gegge c. 1210 Wals, fo. 52b(i) (Holkham). -terram VlfGriffin c. 1210 Wals, fo. 92b(iii) (Erpingham). -Leuiuamfiliam Ulf 1212 FF (PRS NS, xxxii, no. 272) (Field Dalling). - Gilebertus filius Vlf de Nortuna c. 1215 Wals, fo. 113b(iii) (Winnocescroift in Wood Norton), totum tenementum quod Gilbertusfilius V/ftenuit c. 1240 Bromh, fo. 6la(i) (Wood Norton). -Aliciafilia Ulfi de Crec mater Robertifilii A. de Creic c. 1220 CAcre, fo. 48a(iv), Al' filia Ulfi de Creik mater R.filii A. de Creik ibid., fo. 48a(v) (Creake). -Radulfus Vlf 1222 ElyA, fo. 193a (Pulham). - Thomas Vlf 1222 ElyA, fo. 205a, Thomas filius Vlf ibid., fo. 206b (West Dereham), - V/f 1222 ElyA, fo. 209b, Galfridusfilius V/f ibid., fo. 212b (Thorp in Shipdham). -terramRobertifilio Vlfi (bis) c. 1225 Wals, fo. 89b(iv), de Roberto filio Vlfi (bis), terram Roberti filii Vlfi ibid., fo. 90a (ltteringham). - Ulfum de Dallinge, Willelmumfilium Ulfi 1233 CRR 113 (Cur, xv,p. 113)(WoodDalling). -Ulfo filio Elualdi, abl. (witn.) c. 1235 CAcre, fo. 48b(iii) (South Creake), terram Bonde filii Vlf, terram Vlffilii Alwald c. 1235 Wals, fo. 69b(ii) (South Creake). -octo acras terre in villa et in camp' de Sudkrec quas Emma filia Vlfi tenuit de me c. 1235 Wals, fo. 69b(iii), Emme libere homini mee de Sudkrec filie Vlf filii lwani, dat., ibid., fo. 69b(iv) (South Creake; the documents are charters of William son of Philip of Bodham). - Rogerus filius Vlf, de Rogerofilio V/fe. 1235 Wals, fo. 70a(i) (South Creake). -Alanumfilium Ulf c. 1235 (m. 14th) Rams, i, p. 412 (inquisitor at Brancaster, Bumham Deepdale and Angrethorpe), Alanus filius Ulf ibid. p. 423 (Brancaster). - Ex dono Rogeri fil i i Warini filii Vlf m. 13th Langley, fo. 46a (Thurlton). - de Symone filio Ulf Henricus filius Simonis Ulf 1239 (m. 14th) Rams, i, pp. 426, 427, de terra Simonis Ulf, de Simone filio U/fibid., p. 427 (court at Brancaster). -Rogerusfilius Vlfi de Holcham c. 1250 Wals, fo. 52b(iii) (Holkham). ON Ulfr etc. is an original byname belonging to the appellative ON ulfr m. 'wolf' (NK VII, p. 44). The name is common throughout Scandinavia in the Viking and medieval periods (Lind, cols. 1054-1055; LindS, cols. 795-804; DR, nos. 70,230,271,288,329, Mp. 116, 117; DgP, cols. 1520-1526; SRÖg, nos. 166,214,232; SRSö,nos. 9, 72, 75,104,108,211,239,260,265,281, 291,294, 347, 350, 367; SRVg, nos. 55, 78, 125; SRU, nos. 22, 61, 153, 155, 160, 161, 203, 207, 231, 236, 237, 238, 252, 256, 277, 336, 343, 344, 363, 371,388,390,413,444,471,489,501,502,510,560,637,651,657,662, 803, 843, 871, 876, 957, 1043; SRNä, no. 32; Lundgren-Brate, pp. 288-289; Modeer, Personnamn i Ktb, pp. 136--137; Thors, pp. 95-96; Fredriksson, Västgötska dopnamn under 500 år, pp. 182, 183). ON Ulfr etc. is also extremely frequent in England, where it occurs both independently and as the first element of place-names and field names: NPN, pp. 165-166; ZEN, pp. 89-90; Björkman, Festermen, pp. 5, 17; PNDB, pp. 400402; Tengvik, p. 205; SPNT, p. 145; DBS, p. 359, s.n. Ulph; SPNLY, pp. 321-324; Smart, Moneyers, pp. 230, 236; Okasha, no. 1; SCBI, xxi, pp. xlvi, xlvii, xlix, l; von 439

Unbein < AScand *Ubeinn Feilitzen, PNWinton, p. 175; DEPN, pp. 353,354,485,486,495, s. nn. Oulston, Ousby, Ouston Nb nr Whitfield, Ulceby, Ulleskelf, Ullesthorpe, Ullswater, Ulverscroft, Ulnes Walton; PN NbDu, p. 153; PN La, pp. 136-137, 253 (Ulvesthweit); PN Bk, p. 172; PN NRY, pp. 192,249,333; PN D, pp. 316,353; PN Nth, p. 292; PN ERY, pp. 181,328; PN Nt, pp. 172,293; PN Cu, pp. 36, 134,228,327,423,428,457,506; PN Db, p. 762; PN WRY, i, p. 263; PN WRY, ii, p. 300; PN WRY, iii, pp. 13, 237; PN WRY, iv. pp. 36, 67, 186, 193; PN WRY, v, p. 162; PN WRY, vi, pp. 53, 154, 181, 192; PN We, i, pp. 17-18, 107; PN We, ii, p. 314; SSNY, pp. 107, 129; SSNEM, pp. 62,120,229.

ON Ulfr etc. may also occur in Normandy, since Ulf, Edward the Confessor's chaplain and bishop of Dorchester, was of Norman origin, hut in this context the existence of a Frankish (W)ulf must also be noted (Adigard, pp. 233-234, 368, 460-462). The form Wlfo, abl., 1101-1107 Binham has been modified through the influence of the appellative OE wulfm. 'wolf' or of the OE name element Wulf-. The influence of the rare OE monothematic personal name Wulf can be ruled out.

1. OE, ME Unbein < AScand *Ubeinn b) (i) Herewardus f Umbein. Robertus frater eius, Unbein Piteman 1203 AssRoll 817 (NthRS, v, no. 802) (the document concems Martham). - Godwin' Vnbein 1222 ElyA, fo. 185a, Godwinus Vnbeyn ibid., fo. 186b (Pulham).

This name belongs to ON ubeinn 'crooked, perverse' or 'inhospitable', and is also recorded in DB TRE in Db and occurs as a pre-Conquest moneyer's name (NPN, pp. 169-170; PNDB, p. 397; Smart, Moneyers, pp. 231,236; Coinage of Edgar, p. 206). The name has also been noted in a L field name and a weak side-form is contained in the Y form Porcetel Unbainasu(na) c. 1050 (c. 1075) Björkman, Festermen, facsimile of MS. facing p. 4 (SPNLY, p. 320; Lindkvist, AnglBeib, xxxiii (1922), p. 135). Unbein < *Ubeinn is best interpreted as an AScand formation which arose in the Danelaw. In Denmark the name occurs as runic ubein, the name of a Lund moneyer of the period c. 1065-1075 (DR, Mp. 114), hut DR, col. 731, s.n. Uben, Anm. 1, takes this to be a loan from English. Un-, Um- probably represent substitution of the OE negative prefix Un- for Scand U- (SPNLY, p. 320).

440

*Ungolfr

AScand *Ungolfr b) (i) Galfridus filius Vngolfi c. 1190---1200 Binham, fo. 5b printed, SSMEA, Appendix Il, no. 3, p. 270) 14 (Binham).

This is an AScand compound which was formed in the Danelaw from ON ungr, ODan, OSwed ung 'young' and the Scand name element -6lfr, possibly by analogy with the OE name *Geongwine. 15

14

SSMEA, Appendix II, no. 3, p. 270 wrongly renders the present form as Galfridus filius

Ungolf. 15 For OE *Geongwine and its appearance in sources of the ME period see von Feilitzen, NoB, xxxiii (1945), p. 81; DBS, p. 396, s. n. Yonwin; Selten PN, pp. 83-84.

441

V/W

ONVestarr b) (i) Wester liber homo Guert TRE DB, fo. 267, Wester. liber homo. Guert ibid., fo. 270 (Kelling), W ester. I. liber homo ibid., fo. 270 (Sall).

ON Vestarr occurs in Iceland in the Landnamab6k and is found in a Norwegian patronymic of 1334 (Lind, col. 1086). In DB ON Vestarr also occurs TRE in Y, where it appears as Westre (PNDB, p. 403; SPNLY, p. 332). SPNLY, p. 332, suggests that the Y form may altematively represent a Danish byname Wester, comparable with ODan vester 'the west', vestre 'western' and fairly well attested in Danish, hut this explanation is unnecessary and unlikely in view of the fäet that the Danish byname Wester is first recorded in the 15th century.

ON Vi0arr, ODan Withar b) (i) Wither. I. liber homo TRE DB, fo. 157 (Stintonin Sall), ibid., fo. 157b (Hackford). - Transcriptum Wyther Turnel, Wyther cognomento Turnel, dat., Wyther 1134--1140 Holme, no. 137 (lease to Wyther Tumel ofland in South Erpingham H, and land in Tunstead H pertaining to St Benet ofHolme's desmesne in Scottow), Transcriptum Wyther Turnel, Wither Turnel. et Richero filio eius cum Eluiua uxore sua (dat.) 1153-1168 Holme, no. 168, Richero filio Wither, dat., ibid., no. 169 (Rackheath). - Willelmo filio Wyther de Wichingham, dat., 1146 - c. 1150 Holme, 1 no. 150, Willelmo filio Wither de Wichingham, dat., 1168-1175 Holme, no. 308 Galfridus filius Wither de Wikingeham 1179-1180 P, p. 22, de Galfrido filio Wither de Wikingeham 1180-1181 P, p. 86 (Witchingham), in terra que fuit Gaufridi filii Wiöer 1197 FF (PRS, xx, no. 109), domum Hugonis filii Wither 1202 FF (PRS NS, xxvii, no. 359) (Swannington). -Godman. filius Wiöer 1165-1166 P, p. 26, Godman Wiöer 1166-1167 P,2p. 27 (Lynn).

ON Vföarr is well attested in Norway hut is rare in Iceland, though it does appear in the Landnamab6k (Lind, cols. 1090-1091; LindS, col. 807). In Denmark the name is fairly uncommon, hut its considerable antiquity is indicated by its appearance as the first element of a place-name in -lej (DgP, col. 1591; Spndergaard, p. 117).

1 2

Wrongly dated by the editor ofHolme to 1141-1149. Godeman filius Wicher CR.

442

Vigulfr, VigulfR, Vigholf In DB ON Viöarr, ODan Wzthar occurs TRE in L as well as in Nf (PNDB, p. 406). A pre-Conquest example in England is Wiöer 1015 (contemp.) ASWills, no. 20 (in this document, the will of the lEöeling lEöelstan, an inlaid sword which had belonged to Wiöer is included among the bequests). 3

ODan Wigot, OSwed (run.) Vigautr, Vigötr b) (i) Wigoth de Horreo (abl.) (witn.) c. 1175-1190 CAcre, fo. 28b (the document concems properties in South Acre, Rackness, Palgrave and Burstall). - Wigot mercatorem, predicto Wigot, dat., Idem Wigot 1196 FF (PRS, xvii, no. 193) (Burston). - terram Roberti filii Wygot in Ouby 1196 Holme, no. 300 (Oby). - cum hornagio Godwini filii Wigot Pog de Tiln' c. 1215-1220 Der, fo. 128b(i), Carta Godwinifilii Wigot de Tiln', Godwinusfilius Wigot de Tilneya c. 1225-1230 Der, fo. 18lb(i) (Tilney). - Wigot 1222 ElyA, fo. 214a (Thorp in Shipdham).

ODan Wigot is fairly uncommon but has been noted in both runic inscriptions and medieval records (DR, nos. 96, 97; DgP, col. 1571). OSwed (run.) Vlgautr, Vlgibtr is also relatively infrequent (SRÖg, no. 197; SRSö, nos. 25, 285; SRSmå, no. 111; SRU, no. 706). In England the name occurs as Wigot(us), Wigod(us), and is quite frequent, especially in post-Conquest L (NPN, p. 176; ZEN, p. 94; PNDB, p. 404; Tengvik, p. 165; ELPN, p. 84; OBS, p. 383, s.n. Wiggett; SPNLY, pp. 335336; von Feilitzen, PNWinton, p. 177 and n. 2; PN BdHu, p. 300; PN Sr, p. 402). Von Feilitzen, PNDB, p. 404, suggested that we may, in some cases, be possibly concerned with the rare OG Wihgoz, but this is not supported by the evidence, and there is no reason to doubt that ODan Wigot, OSwed (run.) Vlgautr, Vlgibtr is the base ofEnglish forms in Wigot(us), Wigod(us). Forms in Wigotus, Vigotus appear in Normandy and Adigard des Gautries, while admitting the possibility that these forms belong to the rare Frankish Widgaud, prefers to derive them from ODan Wzgot etc. (Adigard, pp. 238239, 369-370, 463).

ON Vigulfr, OSwed (run.) Vigulfä, OSwed Vigholf b) (i) Viulfus. I. fiber homo Edrici TRE DB, fo. 171b (Aylmerton), Wiulfus ibid., fo. 198b (Wiulf held one fiber homo with 15 acres at North Bamingham). - terram Wiolf, dimidiam acram quam Wiolftenet e. 13th Wals, fo. 46a(iv) (Egmere). -Alwiuamfiliam Wiof 1209 AssRoll 558 (Pleas, iv, no. 4297) (Nf). - Wiolf de Thorp 1210 CRR 63 (Cur, vi, p. 5) (Nf; Thorp is unidentified). - de terra Sewale Wyolf 1214-1229 Holme, fo.

3

A man called Wioer also appears as a witness in a charter of 1005 (ASWills, p. 172).

443

Wregge 75b(i) (Yarmouth). - Gaufridus filius Wiolf4 1220 CRR 74 (Cur, ix, p. 35) (recognitor in a case conceming land in Rollesby and Caister).

ON V{gulfr has been noted in independent use in Norwegian records of 1346 and 1424 and forms the first element of a Norwegian and an Icelandic place-name (Lind, col. 1104). In Sweden OSwed (run.) VigulfR is on record and there area few examples of OSwed Vigholver in medieval records (SRU, no. 863; Lundgren-Brate, p. 298, s.n. Wigholf, Thors, p. 97). In England the name also occurs in Sf in DB TRE (PNDB, pp. 404-405). An earlier English example is contained in the form Wiulf Sunta sune 963-992 (m. 12th) ASChR, no. 40 (surety in a transaction concerning landat Maxey Nth) (see also PNDB, p. 404, n. 8). ON Vfgulfr etc. also forms the first element of the NRY field names Wyolfesdic c. 1200 and Wyolmire 1206, which are wrongly interpreted by PN NRY, p. 332, as belonging toan OE *Wigwulf. For the reduction of OE Wlg-, OScand Vig- > Wl- in English see PNDB, § 133; SPNLY, § 136.iii.

ODan Wregge (byname) b) (i) Alexander f Wege 1203 AssRoll 817 (NthRS, v, no. 832) (Haveringland).

Forms in Wegga, Weg(g)e have been noted in Land Y and they are interpreted by SPNLY, p. 332, as reflecting an original byname comparable with ODan w;egge 'wedge', which is the base of a Danish byname. 5 Von Feilitzen, PNDB, p. 411, suggested tentatively that Wegga etc. may possibly be associated with OE wecg 'wedge, mass of metal, lump, piece of money' or with wecgan 'to move, agitate, drive hither and thither' if it isa native name, having noted earlier that the distribution of these forms suggests Scand origin.

4

5

Wiholf CRR 75. For the ODan byname Wegge < ODan wa?gge see DgP Il, col. 1176, s. n.

444

y

ON Ylfingr (fictional), OSwed Ylving b) (i) Iluingus 1086 DB, fo. 185 (held one liber homo with 12 acres at Thurgarton). De Iluing mereatore 1203 AssRoll 817 (NthRS, v, no. 780), Yluing mereator 1203 AssRoll 817 (NthRS, v, no. 865) (Nt). -Rogerus Yluing et Aliciam uxorem suam 1203 AssRoll 817 (NthRS, v, no. 839) (Norwich).

Ylfingr is formally a variant in -ingr of Ulfr with secondary i-mutation. Note also the plural form Ylfingar, a tribal name which is the Norse counterpart of the Wulfingas of Widsith and the Wylfingas of Beowulf.' ON Ylfingr is known as a personal name and byname in Norse legendary history (Lind, col. 1116; LindBiN, col. 405). In Sweden Lundgren-Brate, p. 305, has examples of OSwed Ylving from records of 1336 and 1338. It is formally possible that English forms in Iluing(us), Yluing etc. belong to ON Ylfingr, OSwed Ylving, but the rarity of the Scand names may suggest that we are here concemed with an AScand name which was formed from ON Ulfr etc. in the Danelaw. Cf. the appearance of *Ylfketill, a side-form of Ulfketill with secondary i-mutation, as the first element of Ilketshall Sf (above, s.n. Ulfkell).

AScand *Yric b) (i) Irieus (deeanus) (witn.) 1135-1143 NorwCPCh, no. 116 (charter of Bishop Everard ofNorwich). -frie (abl.) (witn.) c. 1165-1180 CAere, fo. 85b(ii) (the document concems Ulfchetel son of Aildricus of Marham). -Herueius Yrieh 1199 P, p. 288, de Herueio Yrieh 1200 P, p. 138 (Nfor St). -Reginaldusf Yrieh 1199 P, p. 288, (deReginaldo f Yrieh) 2 1200 P, p. 138, Regina/dus f Yrieh ibid., p. 145 (Nf or St). -Radulfus filius Yrieh 1209 FF (PRS NS, xxxii, no. 210) (Burnham). -homagium Wlnod' filii Yrie de Tiln' c. 1215 Der, fo. 137a(i), crofta que fuit Wlnod filii frie, a terra quam Wlnod filius frie dedit c. 1225 Der, fo. 147b(ii) Galfridus Wlnodfilius Yrie de Tiln' c. 1230 Der, fo. 137b(i) (Tilney).

Von Feilitzen, PNDB, p. 299, shows that the base of English Yric, Iric etc. is an AScand *Yrlc < OScand *(Oyrikr, *Eyrikr, which he incorrectly considered 1 For the Ylfingar/Wulfingas/Wylfingas group see EN Beow, pp. 120--122, and Malone, Widsith, pp. 213-214. 2 Cancelled.

445

*Yric to correspond to OSwed (Z)rik. 3 The replacement of Scand ~y. ey by OE y is paralleled by OE yre < ON eyrir and OE.fysian < ON *feysa (PNDB, pp. 299300). In addition to the above Nf examples AScand *Yrzc has been noted in DB TRE in D and Sf (PNDB, p. 299). For the use of Yric etc. to render the name ofEarl Eirikr of Hlaöir in English sources see above, s.n. J6rekr, and the references given there.

3

See above, s. n. I6rekr.

446

Appendix: Name Forms from Sources Excerpted Subsequent to 1980

ODanAggi b)(i) croftam Agge Falbe c. 1225 (e. 14th) Creake, fo. 7b (North Creake).

ONÅki b) (i)Hachi c. 1170 (1. 12th) Caen, p. 51,Achiibid., pp. 52, 54 (Horstead). -Adamfilius Ache 1251 Ely Reg, fo. 184b (Terrington). (ii) Petrumfilium Oky (inquisitor, Walton) 1251 Ely Reg, fo. 199a, Petrusfilius Oky iibid., fo. 202a (Walton), Petrus Oky ibid., fo. 204b (Wiggenhall). -Agnes et Basille heredes Ellickfilii Oky, Adamfilius Eluredy Oky 1251 Ely Reg, fo. 204a (Emneth).

AScandAnki b) (i) Alanusfilius Anche 1251 Ely Reg, fo. 261b (Northwold).

This is a short form of names in Arn- and is also found in L and Y (SNPLY, p. 10).

ODan, OSwed Anund, ON Qnundr b) (i) Galfridusfilius Hanundi c. 1170 (1. 12th) Caen, p. 51, Gaufridusfilius Anunt ibid., p. 52 (Horstead). (ii) Walterum filium Anandi (inquisitor, Pulham) 1251 Ely Reg, fo. 209a, Walterus Anant ibid., fo. 218b, Willelmus filius Anant ibid., fo. 219a (Gidelardeslete in Pulham).

ON Arnkell b) (i) Godefrido filio Arketel', abl. (witn.) 1. 12th (e. 14th) NorwCPCh Il, no. 121 (Gateley).

447

ON ÅsbiQrn a) (ii) ... pro quadam terra iacente in Brisingham que vocatur Osbernis e. 13th (1. 14th) SfCh, Il, no. 172 (f. n. in Bressingham). - Osberno coco, abl. (witn.) 1120---1150 NorwCPCh Il, no. 498 (grant of 8 acres next Hohes between Hellesdon and Catton to the Hospital of St Paul).

ON Åsgautr b) (ii) Herveumfilium Osgot' (serf: Sedgeford) c. 1200---1250 NorwCPCh Il, no. 7. Osegodfilius Laurencii 1251 Ely Reg, fo. 197b (Walpole).

ON Åsgeirr b) (i) Asgarus c. 1170 (1. 12th) Caen, p. 53, Asgerus ibid., p. 54 (Horstead). -Asgero clerico de Reppes, abl. (witn.) e. 13th NorwCPCh Il, no. 363 (Martham; Reppes is [North/South]repps).

ON Åskell b) (i) Ascetillus mercator (witn.) c. 1150---1168 (1. 13th) Acta, no. 55 (land pertaining to the manor of Cressingham in Ashill and Panworth [Hall]). (ii) Ascili (gen.) 1. 13th [before 1296] KLynn, p. 109 (no. 85) (entry in the Kalendar and obit roll of the hospital of St Mary Magdalen, Lynn [Gaywood Hospital]). (iv) Rogero filio Anchetil, abl. (witn.) 1161-1173 Acta, no. 109 (terram Turchilli filii Ketelli in Humeresfeld' [Homersfield Sf]). (v) Iohannesfilius Astini e./m. 13th KLynn, p. 296 (no. 379) (entry in the bede roll of the gild of Holy Trinity, Lynn). - terram Astini filii Gilberti c. 1222 KLynn, p. 77 (no. 11) (Lynn).

ONBondi b) (i) Bondo, nom. (witn.) 1158-1160 NorwCPCh Il, no. 492 (Pedham in Langley). decimam Ricardi fil ii Bondi prepositi, decimam Radulfi filii Bondi 1175-1189 (13th [? c. 1250---1260]) Acta, no. 193 (South Creake).

ON Brynhildr fem. a) (i) Brunylddesheg ? e. Hy3 (1. 13th) NorwCPCh Il, no. 488 (f. n. in Rushall).

448

OScand *Dottr a) (i) Dotescroft 1285 NorwCPCh Il, no. 135 (f. n. in Gateley).

ON Eirikr b )(i) Airicus c. 1170 (1. 12th) Caen, p. 54 (Horstead).

ONGauti a) (ii) Waltero filio Gothe de Hasseleya, dat., pater suus Gothe c. 1150-1168 (1. 13th) Acta, no. 55 (land pertaining to the manor of Cressingham in Ashill and Panworth [Hall]; Hasseleya is Ashill).

AScand *Gufa, ME *Guve fem. b) (i) Willelmi Iustise et Guwe uxoris eius (gen.) 1. 13th [before 1296] KLynn, p. 113 (no. 85) (entry in the Kalendar and obit roll ofthe hospital of St Mary Magdalen, Lynn [Gaywood Hospital]).

This name is also attested in L (SPNLY, p. 113).

ON Gunna fem./ON Gunni masc. a) (i) Gunnegore 1150-1158 (e. 14th) NorwCPCh Il, no. 84 (f. n. in Fring).

ON Gunnhildr fem. a) (i) Gunnildeslake (fishery at Welles [UpwelVOutwell]) 1251 Ely Reg, fo. 94b. b) (i) unum mesuagium quod fuit Gunnild' de Pundfold 1251 Ely Reg, fo. 186b, duas acras terre que fueruntAlanifilii Gunnilde ibid., fo. 189a (Terrington). -Gunnildafilia Briani 1251 Ely Reg, fo. 252b (Bridgham). -Godmanni Woderowe et Gunnilde uxoris eius (gen.) 1. 13th [before 1296] KLynn, p. 116 (no. 85) (entry in the Kalendar and obit roll of the hospital of St Mary Magdalen, Lynn [Gaywood Hospital]).

ON Hakun, -kon b) (i) Willelmus filius Haconis de Trousse m. 13th NorwCPCh Il, no. 442, Willelmus filius Hakun de Trouse m. 13th (e. 14th) NorwCPCh Il, no. 451, Willelmus Hakun ibid.,

449

no. 441 (Trowse). -Hakonem pany (inquisitor, Welles [UpwelVOutwell]) 1251 Ely Reg, fo. 94b, Haco pany ibid., fo. 95b (Welles [Upwell/Outwell]).

ONHalfdan b) (i) Ricardusfilius Halde[n] c. 1170 (I. 12th) Caen, p. 52 (Horstead).

ONHamundr b) (i) Hamundus de Burringeham et Gaufridus frater eius (witn.) c. 1114-1160 [towards the end of this period] NorwCPCh Il, no. 365 (Eaton near Norwich; Burringeham is Burringham L).

ONHaraldr b) (i) Willelmusfilius Harald et Mariota vxor eius 1251 Ely Reg, fo. 250b (Rudham). de Harald Swart 1309 KLynn, p. 419 (no. 467) (inquest at Lynn).

ON lngolfr [or Frankish Ingulf] b) (i) Ingulfus prior 1119 (later addition to Reg. 1 [e. 14th]) NorwCPCh Il, no. 496) (grant ofhalfthe tithe ofTaverham),Hugonefilio Inguifi, abl. (witn.) 1121-1145 (later addition to Reg. 1 [e. 14th]) NorwCPCh Il, no. 83 (Fring).

ONivarr b) (i) Ywar domino de Rollesby, abl. (witn.) e. 13th NorwCPCh Il, no. 363 (Martham; Rollesby is Rollesby).

ON Ketill a) (i) Keteleshawe (p) (bis), Keteleshaue 1251 Ely Reg, fo. 210a, Keteleshal' (p) ibid., fo. 210b (f. n. in Pulham). b) (i) Willelmusfilius Chetel c. 1170 (I. 12th) Caen, p. 52, Ricardusfilius Cetel ibid., pp. 53, 54, Adamfilius Cetel ibid., p. 54 (Horstead). - terram Ketel Hunewyne c. 1235 (e. 14th) Creake, fo. lOb(ii) (North Creake).

450

ON Kolsveinn a) (i) Colswainyshil 1251 Ely Reg, fo. 183b (f. n. in Terrington).

ON Ospakr a) (ii) sex acras terre quas Johannes filius Unspac tenuit 1188-1200 Acta, no. 222 (Thomage).

ON Ragn(v)aldr, R9gnvaldr b) (i) Stephanusfilius Rawenoldi 1251 Ely Reg, fo. 207b (Emneth).

The form Rawenoldi (gen.) shows the development of ME [a] + [y] > [au] (see Jordan, § 113).

ODan Ringulf [or OE Hringwulf] a) (i) Ringolfestopht c. 1200 NorwCPCh Il, no. 262, Ringolfestoft ibid., no. 263 (f. n. in Cringleford).

ON Sighvatr a) (i) Sywatesholm 1251 Ely Reg, fo. 193a (f. n. in Walpole). b) (i) pro anima ... Siwati presbiteri de Plumstede, Willelmus heres Siwati presbiteri 1146--1174 (e. 14th) NorwCPCh Il, no. 202, Milone filio Seeuuati, abl. (witn.) 12041205 (e. 14th) NorwCPCh Il, no. 203 (Plumstead).

ON Steingrimr b) (i) Stangrimus mag' (witn.) c. 1114-1160 [towards the end of this period] Norw CPCh Il, no. 365 (Eaton near Norwich), magistro Stangrino, abl. (witn.) c. 1154-1168, prob. c. 1154-1159, (l. 13th) Acta, no. 92 (coAf"rrmation to Eye Priory St), 1154-1170 (13th [? 1250--1260]) Acta, no. 74 (institution of the persona of the church of South Creake), c. 1159 (13th [? 1250--1260]) Acta, no. 75 (church ofWestbriggs), 1161-1173 (13th [? 1250--1260]) Acta, no. 76 (Massingham), 1161-1173 Acta, no. 109 (terram Turchilli filii Ketelli in Humeresfeld' [Homersfield Sf]), magister Steingrimus (witn.) 1158-1160 NorwCPCh Il, no. 492 (Pedham in Langley), magistro Steingrim', abl. (witn.) 1161-1173 Acta, no. 108 (confrrmation to Bartholomew, son ofRobert the cook

451

of Homersfield Sf, of his father's tenures and the lease of unam acram terre que fuit Leve inter domum Ricardi filii Wljketel et domum Udardi fabri), magister Stangrimus (witn.) 1161-1173 Acta, no. 111 (confirmation to Lewes Priory of its possessions in the diocese ofNorwich), [magistro] Staingrimo, abl. (witn.) 1161-1173 Acta, no. 112 (confirmation to Lewes Priory of the church of All Saints, Foulden), magistro Stangrimo, abl. (witn.) c. 1158-1169 (1. 13th) Acta, no. 87 (institution ofthe persona ofthe church of Bridgham), 1161-1173 Acta, no. 112A (church of Colveston), c. 1164--1172 Acta, no. 140 (Dedeholm, prob. in East or West Somerton), magistro Staingrino, abl. (witn.) 1166-1170 (13th [? 1250-1260]) Acta, no. 77 (tithes ofMethwoldHythe). -Aylwardus filius Stangrim de Sechford e./m. 13th (e. 14th) NorwCPCh Il, no. 9 (quitclaim; Wolvesdele in Sedgeford). -Regina/dus Stangrimfilius, Henricus (?)filius Stangrim e./m. 13th KLynn, p. 296, Laurencius filius Stangrim ibid., p. 298 (no. 379) (entries in the bede roll ofthe gild ofHoly Trinity, Lynn). - ... scilicet in Briggegate. quam Stangrimus pelliparius auus meus dedit [charter of Alexander, son ofWilliam Aswy], Reginaldo filio Staingrim, abl. (witn.) 1217-1224 KLynn, p. 75 (no. 7) (Lynn).

ON SumarliOr a) (i) Sumerdeshil 1251 Ely Reg, fo. 184a (f. n. in Terrington). b) (i) Godefridusfilius Sumeredi 1251 Ely Reg, fo. 203b (Tilney).

ON Sunnulfr b) (i)Robertofilio Sunnolfi, abl. (witn.) 1. 12th/e. 13th (c. 1290-1300) KLynn, p. 85 (no. 49) (Lynn). - ... concessionem et donationem patris mei Sunnolf [charter of Willelmus Flathe of Martham] e. 13th NorwCPCh Il, no. 363 (Martham).

OScand *Svartgeirr b) (i) ... a vico molendiniSwartger [in Lynn] c. 1220 (c. 1290-1300) KLynn, p. 86 (no. 51).

ONToki a) (i) Tokeshil 1251 Ely Reg, fo. 184a (f. n. in Terrington). b) (i) inter terram Robertifilii Toke et terram Radulphi fratris sui 1. 12th (c. 1290-1300) KLynn, p. 84 (no. 45), Symon filius Roberti filii Toke 1. 12th/e. 13th (c. 1290-1300) KLynn, p. 84 (no. 46) (Lynn). -Andreasfilius Toke c. 1230 (e. 14th) Creake, fo. 1 lb(i), Andreasfilius Toke de Northcr' ibid., fo. llb(ii), Ricardusfilius Toke ibid., fo. llb(iii) (North Creake).

452

ONToli a) (i) Toliesnap' 1180-1182 (13th [? 1250-1260)) Acta, no. 184 (f. n. in Gateley). b) (i) Magister Tholy 1. 13th KLynn, p. 305 (no. 379) (entry in the bede roll ofthe gild ofHoly Trinity, Lynn).

ODan Töpi b) (i) Caterinafilia Topy 1251 Ely Reg, fo. 203b (Tilney).

ON l>oraldr, Norman Turold b) (i) Turoldus capellanus (witn.) 1146--1174 (e. 14th) NorwCPCh Il, no. 202 (Plumstead), Thoraldo capellano, abl. (witn.) 1154-1158 (13th [? 1250-1260)) Acta, no. 71 (South Creake), Toroldo capellano, abl. (witn.) 1146--1163 (l. 13th) Acta, no. 86 (church of Melton), 1154-1170 (13th [? 1250-1260)) Acta, no. 74 (institution of the persona ofthe church of South Creake), c. 1158-1169 (1. 13th) Acta, no. 87 (institution of the persona of the church of Bridgham), c. 1159 (13th [? 1250-1260)) Acta, no. 75 (church ofWestbriggs), Toroldus capellanus (witn.) c. 1150-1168 (1. 13th) Acta, no. 55 (land pertaining to the manor of Cressingham in Ashill and Panworth [Hall]). - terram Torald Penting' c. 1230 (e. 14th) Creake, fo. 9b(i, iii) ([North] Creake).

ON l>orbi'i_)rn b) (i) Que omnia facta fuerunt vivente Thurberno tunc temporis rectore ecclesie supradicte [Martham] after 23/9/1273 NorwCPCh Il, no. 357 (Thurbem certainly lived at the end ofthe 12th century and possibly also at the beginning ofthe 13th century. After his death, Bishop John de Grey ofNorwich [1200-1214] bestowed the said church ofMartham on Geoffrey, deacon ofNorwich).

ON l>orgils, Norman Turgis b) (i) Turgiso [milite], abl. (witn.) 1. 12th/e. 13th (e. 14th) NorwCPCh Il, no. 222 (Postwick).

ON l>orir b) (i) Thure c. 1170 (l. 12th) Caen, pp. 53, 54 (Horstead).

The form here represents the OEScand variant Puri.

453

ON l>orketill/l>orkell b) (i) Turkillo capellano, abl. (witn.) 1135-1141 (e. 14th) Acta, no. 48 (confirmation to St Mary's Priory, Thetford). -Turchildus Brunus c. 1170 (l. 12th) Caen, p. 51, Turchillus brunus ibid., p. 52, Thurchild' mercator ibid., p. 53, Thurchild', Thurchildus ibid., p. 54 (Horstead). - Willelmus filius Thurketel, Willelmus Thurketel 1251 Ely Reg, fo. 239a (Shipdham).

ON l>orsteinn b) (i) Thurstanus Parmentarius e./m. 13th KLynn, p. 297 (no. 379) (entry in the bede roll of the gild of Holy Trinity, Lynn). - terram Thurstani filii Galfridi c. 1230 (e. 14th) Creake, fo. 9b(i), terram Thurstan' filii Galfridi ibid., fo. 9b(iii) ([North] Creak:e). - de terra Thurstani le stiward 1251 Ely Reg, fo. 212a, Thurstanus stiward ibid., fo. 213b, Gaufridus filius Turstan' ibid., fo. 217b (Pulham). - unam acram que fuit Thurstani 1251 Ely Reg, fo. 224a (East Dereham). -Radulfusfilius Thurstani 1251 Ely Reg, fo. 250b (Bridgham). -Nigello filio Thurstani de Frenge, abl. (witn.) 1271-1289 (e. 14th) NorwCPCh Il, no. 31, c. 1280-1286 (e. 14th) NorwCPCh Il, no. 33 (Sedgeford; Frenge is Fring), Nigello filio Thurstani de eadem [Frenges, i. e. Fring], abl. (witn.) 1272-c. 1280 NorwCPCh Il, no. 37, Nigello filio Thurstani de eadem [Frenge, i. e. Fring], abl. (witn.) 1275 (e. 14th) NorwCPCh Il, no. 38 (Sedgeford), c. 1280-Feb. 1283 NorwCPCh Il, nos. 73, 74 (Thornham), Nigello filio Thursta[ni], abl. (witn.) c. 1280-1283 NorwCPCh Il, no. 30 (Sedgeford). - terram que fuit Thurstani Tanatoris, Dictus Thurstannus I. 13th [before 1296] (15th) KLynn, p. 166 (no. 174) (Rondeshil in Lynn). - de Thurstanno Er!, Thurstannus I. 13th [before 1296] (15th) KLynn, p. 164 (no. 174), Thurstannus Eerl ibid., p. 165 (Dampgate in Lynn), Thurstanus Er! I. 13th KLynn, p. 303 (no. 379), Johannes filius Thurstani Er! I. 13th/e. 14th KLynn, p. 307 (no. 379), /ohannesfilius Thurstani ibid., p. 309 (entries in the bede roll ofthe gild ofHoly Trinity, Lynn).

ON Ulfkell b) (i) iiii acras prati et xx acras terre quas tenuit Uljketel 1161-1173 Acta, no. 111 (? Merton). - pratum Huljketel de Monasterio e. 13th SfCh, VIII, no. 284, pratum Huljketel del Muster ibid., no. 285 (Shelfanger).

ODan Wal(re)grest b) (i) PIELGJST 1009-1017 SCBI, vii, no. 1244, xxv, no. 563, xxvi, no. 1220, 10171023 SCBI, xv, no. 3858, PIELGISD 1009-1017 SCBI, vii, no. 1245, xxv, no. 564. PALGIST 1009-1017 SCBI, vii, no. 1246, xxxvi, no. 569, xxxvii, no. 151, PALGIT 1009-1017 SCBI, xxxvii, no. 152, PLG/ST 1017-1023 SCBI, xv, nos. 3875-3876, xxxvi, no. 659) (moneyer, Thetford).

454

DgP, col 1547, has two medieval examples of ODan Walegrest, but the name of the above moneyer from Thetford is perhaps better regarded as being of Continental origin and as going back ultimately to a Frankish *Walhagfs (cf. Remiremont, p. 276 for the forms Uualchis, Uualhis).

Bibliography and Abbreviations Acta

=

Caen

=

Creake

=

ElyReg KLynn

= =

NorwCPChll

=

SfCh

=

English Episcopal Acta, VI: Norwich 1070-1214, ed. C. HarperBill (London 1990). Charters and Costumals ofthe Abbey of Holy Trinity Caen, ed. M. Chibnall. Records of Social and Economic History New Series 5 (London 1982). The cartulary of North Creake Abbey (British Library, Add. 61900) (e. 14th). See the main bibliography. The Making of King's Lynn: A documentary Survey, ed. D.M. Owen. Records of Social and Economic History New Series 9 (London 1984). The Charters of Norwich Cathedral Priory Pt. 2, ed. B. Dodwell. PRS NS 46 (London 1985). Sujfolk Charters (Suffolk Record Society 1979 ff. I = Leiston Abbey Cartulary and Butley Priory Charters, ed. R. Mortimer [referred to in the main body of the present work as 'Leiston']; 11-III = Blythburgh Priory Cartulary, ed. C. Harper-Bill; IV-VI= Stake by Clare Cartulary, ed. C. Harper-Bill and R. Mortimer; VII-X= Sibton Abbey Cartularies and Charters, ed. P. Brown; XI = The Cartulary of the Augustinian Friars of Clare, ed. C. Harper-Bill; XII= Eye Priory Cartulary and Charters Pt. 1, ed. V. Brown).

455