Form Unmodified/Uninflected “Many surnames formed from personal names without any prefix or suffix are derived from personal names which were either disused or rare by 1300. ..Most of these were Old English or Scandinavian personal names.” This type of unmodified surname was common “among the better-off sections of town populations from the 12th century onwards, and among small freeholders in the countryside,” and probably fairly common among bondmen too [Mackinley 1990, p ] Examples Common before 1300 in all areas: Alexander, Allen, Andrew, Austin, Bartholomew, Brian(t), Clement, Ellis, Gilbert, Goddard, Go(o)dwin, Hammond, Humphrey, Jarvis or Jervis or Gervase, Lambert, Laurence, Martin, Matthew, Mitchell, Osborne (though can be a placename), Payne or Pain, and Vincent. Old English examples: Allnutt, Allward, Brightiff, Edrich, Elphick, Godwin, Woolridge. Old Scandinavian examples: Allgood, Ingold, Kettle, Tooley, Tovey, Thurkill. Post-Conquest examples: Ansell, Baldry, Durrant, Goddard, Hammond. Gaelic: Duncan, Findlay. Modified forms Genitival (Possessive) See also: other genitival forms, e.g. occupational – Parsons; topographical – Mills; relationship – Cousins. “The genitive –s form was by no means exclusively patronymic. It could signify other relationships. A high proportion of those with such names in fourteenth century East Anglia were women, often widows, whose –s names refer to their relationships to their husbands, or former husbands, for example Alice Thomys (widow of Thom) and Katherine Wilkins. Others were servants, for example Adam les Prestes, Diot del Dawkins or Thomas at Adamys.” Wilson (1998) p 128 The possessive form appears to originally have been the name of “small free tenants, bond tenants, and the less affluent members of town populations”, who adopted surnames fairly late. (McKinley) This form was rare before 1270, but was numerous by 1350. Its heartland was the SW Midlands (Gloucestershire, Oxfordshire, Herefordshire). It remained relatively rare in the North of England. From the 16th century, this form became widely adopted in Wales where “the addition of ‘s’ seems to form a substitute for ap, so that it is not surprising that ‘s’ is not usually added to the class of names in which ap/ab have become absorbed (Bowen, Prosser, Pugh)” . However, exceptions are found in the English border counties with such names as Bevans and Beavans appearing in some quantity.” (Rowlands 1996, p ) -s, -es, as in Williams, Jones -s, as, -x, as in Rix, Dix Plural forms -kins as in Jenkins -sons: This form (personal name + sons) is scarce, and mainly found in the SW Midlands. It seems to have arisen through the migration of –son names into areas where –s forms predominated. e.g. Johnsons, Robertsons. Pet forms (hypocorisms) Viewpoint 1 Pet forms seem to have been used to express affection or status, especially in areas where the root personal name was frequent. For example, Adekyn Bron of Ossett was the son of Adam Bron in 1286, and seems to have been used in the sense of ‘junior.’ This practice Viewpoint 2 Pet-forms may have been used as a form of mockery by Anglo-Normans, and then taken up, as nicknames are, by those native ‘social inferiors’ who had adopted Norman ‘names.’ (suggested by P McClure in Nomina 26 and 28) The large number of pet forms (hypocoristics) of personal names reveals the playfulness and inventiveness of the medieval mind. Pet forms are manifested in three main ways: Abbreviation – simplification through the removal of a linguistic segment. This can be the deletion of a segment of any length, and can occur in any position e.g. Nicol>Col, Matilde>Til. Extension through the addition of a hypocoristic suffix to either the full personal name or a diminutive or pet form of it. See the suffixes below. Substitution. For example through the playful substitution of vowels and consonants, or the addition of a prosthetic rhyming consonant e.g Roger>Hodge, Richard>Dick, Agnes>An>Nan, Isabel>Ib>Tib (P McClure Nomina 21, 1998) “Names in -s or -son made particular use of hypocoristic forms. In a sample from Norfolk in 1330, 70% of names ending in –s stemmed from hypocoristics.” (Wilson, 1998, p ) The problem “We can be sure that baptismal names were used in a wide variety of hypocoristic or pet forms, especially by ordinary folk. The problem is to know which hypocorisms belong to which baptismal names.” P McClure Nomina 21 (1998), p 102 Diminutives English diminutives Examples Nicholas > diminutive ‘Col’ > surname form Collins Gilbert> diminutive ‘Gib’ > surname form Gibson Geoffrey> diminutive ‘Gep’ > surname form Gibson Isabel> diminutives ‘Tibot/Ibot’ > surname form Ibbotson (Redmonds 2004a, p ) Sub-types: with or without suffixes Diminutives without suffix are relatively rare: Hudd, Hobb. Hypocoristic suffixes were often appended to diminutives of personal names: Adkins, Dawson (which is derived paradoxically, in England from Ralph, but in Scotland from David. (Source Hough, 2003 citing MacClure) Gaelic diminutives Class ‘with-suffix’ under latter element? Welsh diminutives Beddoes<Bedo<Maredudd Gittins<Guto<Gruffydd Prefixes (of descent) Ab/Ap- [Welsh] “In the system of mutation of the Welsh language mab becomes fab, and the initial soft f sound (English v) was dropped in normal speech…The general rule which evolved was to use ab before vowels and ap before consonants, but finds many breaches in actual use in records.” (Rowlands 1996, p8 ) However, Ap or Ab did not agglutinate if the initial letter was unsuitable, as in the cases of Gruffydd, Llewellyn and Morgan. The distribution of ap or ab surnames “shows a heavy emphasis towards the border counties of Brecon and Radnor where collectively they are held by as many as 30% of the population.” (from a survey of marriages in Wales 1813-1837, Rowlands 2006, p108) Post-Conquest personal names: Probert, Pritchard, Prosser, Prandle Welsh personal names: Price (Ap Rhys), Bowen (Ap Owen) Fitz- (Continental-Germanic) Fitz is the Anglo-French equivalent of filius. However, it was used only in Normandy not in the rest of France. The second element is usually a post-conquest personal name. They were mostly originally the names of landowners of Norman origin.e.g. Fitz Alan, FitzWalter The use spread to other social classes, and it was in current use to about c1300. Some forms (FitzRoy, FitzJames, FitzClarence) were late (17th-19th century) bestowals on the natural children of Charles II, James II and Willam IV, respectively Gil + Saint’s name/Religious name Gaelic prefix gille = servant of Devotional names were often prefixed to the name of the saint on whose day the child was born. Gillies (Highland, Western Isles)< gille Iosa ‘servant of Jesus.’ Gilchrist < gille Christ “servant of Christ.’ Gil + occupational term Mac, Mc, M’ This group accounts for about 20% of Scotland’s surnames. (source: Dorward 1998: 116). Thirteen are in the top 100 names in Scotland Scottish or Irish origin? Mac + occupational term Many Mac- names reflect an association with a clan, and therefore may not be patronymic. Clan names in the top 100 (Scotland): MacDonald, MacGregor, MacIntosh, MacIntyre, MacKay, MacKenzie, MacLean, MacLeod, MacMillam. Mac + Gaelic personal name – MacEwen, McFadden Mac + Scandinavian personal name – MacIver, MacManus, MacLeod Mac + English personal name – MacJames, MacWilliam Mac + Gaelicisation – Thomas>Mac Támhais > MacTavish Mac + Anglicisation – MacAmbois > MacCambridge; MacShuibne > MacQueen Dropped Mac (IOM typifies) – MacOwen > Keown; MacCurtin > Curtin Scottish examples: Kennedy < mac Kenedi, Cowen < MacOwen < mac ghille Chomghain Keddie (Shetland) < MacKeddie < mac Adaigh Mac+ pet form – Macdickon, Macgibbon, Macjock Mac + Saint’s name – MacBride (devotee of Saint Bridget) native Arran surname, otherwise an Irish import. Mab-, Mag-, Map- (Welsh) In the lay subsidy lists of 1292-3 for Merioneth, 53% of taxpayers bore patronymics using Mab or Ap. Ni- (Irish); Nhic- Nic- (Gaelic) Anglicised forms of Irish Ní, daughter of, e.g. Maire Nic an Gaill O’ (Irish) Anglicised form of the Irish Ó/Ua, originally grandson/granddaughter of Qu + vowel (Manx) Verch (Welsh) Daughter of e.g. Nest verch Madog. “Verch becomes ferch in modern orthography; it was often shortened to vch or vz in documents and appears also as ach and ych.” (Rowlands 1996) This prefix, like ap-, could adhere to patronyms: Critchett < verch Richard Kedward < verch Edward But beware similar abbreviated Mac forms Suffixes “Shortly before 1250 the process was more or less completed by which insular (Old English and Old Norse) personal names were largely displaced by names introduced by Normans and their followers from the near continent, such as Bretons, Flemings and Picards. In origin the new name stock was partly Continental Germanic, of a West Frankish type, partly Romance (including many saints’ names) and partly from the Biblical languages, Hebrew, Aramaic and Greek (via Latin), not to mention the lesser contributions from Scandinavian and Celtic languages. All the names were expressed in Old French form and they came with ready-made hypocorisms or pet-forms, usually consisting of a short form of the name compounded with a diminutive suffix.” (P McClure Nomina 26, 2003, p 93) Some of these suffixes apply also to relationship surnames. Hypocorisms (Pet forms) The majority of suffixes are Old French, but there is also a small but significant Germanic component, mainly Flemish. No Old English suffix is yet evidenced as a hypocoristic suffix to an Anglo-Norman name. “For a twelfth-century Englishman the social advantages of bearing a ‘Norman’ Christian name would be embarrassingly compromised by the addition of an English hypocoristic suffix.” (P McClure, Nomina 28, p 33) -ate, atte -chon (Northern Old French) e.g. Huchon -cok and –cus “To Norman ears, the Englishness of –cok and –cus, together with their resemblance to Flemish –kin, can only have signified that the bearers of such names were not ‘one of us’.”… Were they “invented by the Normans, not for use toward their own kind, but to mock and belittle those English who, in increasing numbers during the twelfth century, bore ‘Norman’ names”? (P McClure, Nomina 28, p 33-4) -cok and –cus look like twin forms. -cok is widespread and popular as a pet-form for male names (but not exclusively). -cus is less widespread, rarer examples, possibly commoner with women’s names. (P McClure Nomina 28, 2005, p 30) -cok (-cock, -cox) Reaney and Wilson cite as head-forms: Adcock, Alcock, Bad(t)cock, Bawcock, Beacock, Dilcock, Elcock, Godcock, Hancock, Haycock, Heacock, (Hickox), Hitchcock, Hullcock, Jeffcock, Litcook, Lowcock, Maycock, Moorcock, Mycock, Palcock, Pilcock, Pitcock, Raincock, Silcock, Simcock, Tilcok, Wadcock, Walcock, Watcock, Whitecock, and Wilcock. (Cited in McClure, Nomina 28, 2005, p 7) “Middle English –cok first appears in the late-twelfth century, of uncertain parentage and ambiguous morphology. The earliest recorded examples of its use in forenames are of the late-twelfth and early thirteenth centuries: Salecoc the Jew (1193) and Alecoc or Alekok (1204).” (P McClure Nomina 28, 2005, p 6) “The earliest examples of hypocoristic –cok occur just before and just after 1200, and are probably from Old French personal names or Middle English versions of them. So too, I believe are all but a handful of those that are recorded thereafter.” (P McClure, Nomina 28, 2005, p 18) Ths suffix must have been coined in England as no evidence if use in Flanders or north eastern France. Reaney derives this suffix from OE cocc ‘cock, rooster, male bird’ and by usage for a ‘pert servant’, but McClure points out that there is no evidence to support the ‘servant’ hypothesis. Ewen sees it as an extension of the Old English suffix -oc. However no other OE suffix is known to have transferred to continental names, and it is such names that to be found as the earliest examples, and overwhelming found with continental names. [There is clearly an error in this sentence which makes its meaning difficult to fathom. -MS] “On the evidence that we have it looks as though the suffix belongs with continental rather than insular names, in which case we need to find a way of explaining how a suffix of apparently Germanic or Anglo-Saxon composition came to be invented for use with an essentially Norman name-stock.” (P McClure, Nomina 28, 2005, p 23) “that whereas Old English –oc was used exclusively with men’s names, Middle English –cok was used in the continental manner with names of women as well as men, although predominantly the latter.” (P McClure, Nomina 28, 2005, p 23) Sometimes “there is confusion with the toponymic –cote. From the 13th century onwards, scribes often wrote a –c- like a – t, and vice-versa, such that –cot and –kot are impossible to distinguish from –coc and –koc (in printed sources).” (P Macclure, Nomina 26, 2003, p116 ]) or nicknames from birds – Heathcock, Peacock, Woodcock, Grewcock. -cus Examples: Amecus, Edecusa, Ricus, Wilcus McClure gives evidence that this is a Middle English hypocoristic suffix. It must have been coined in England as there is no evidence of use in Flanders or north-eastern France. Alternative 1: an OE suffix for use with native English names. It is possibly a double diminutive , -k-us, and modelled on OE –us forms as pet forms of women’s names. Alternative 2: “to suppose that –cus, like –cok, was originally coined for use with Anglo-French rather than native English names” i.e. for hypocorising the new names that were introduced after the Norman Conquest, through re-modelling the existing OE –us forms. Further, many –cus forms also present as –kin forms, e.g. Alcus/Alkin. (McClure Nomina 28, 2005,pp 29, 32) -cot (Picard) Sarrecote, a pet-form of ME Sarre (Sarah?)? Alcot, a pet form of names like Alice, Alexander and Alan Simcot, pet-form of Simon. “It is of Picard origin and belongs to a group of hybrid double suffixes, including –quet and –coul, in which Germanic –ik has been extended by the addition of a French diminutive. The suffix was used in Picardy and Flanders to form pet-names like Hanecot (with Old French Jehan) and Wilecot (with Northern Old French William).” (P McClure, Nomina 26, 2003, p 115) “Often from the mid 13th c onwards, scribes write a ‘t’ like a ‘c’ and vice versa, so that it is often impossible to determine whether what looks like a –cot and –kot respectively, is in fact a –coc and –koc (variant spellings of –coq).” (P Macclure, Nomina 26, 2003, p116) -el (Old French) Simonel, Pagnell -en This (pet form + -en) originated in the Midlands. Reaney found examples in Warwickshire, Worcestershire, Gloucestershire and McKinley found them in west Oxfordshire. Ed(d)en, and Hawken are possible examples. -et(e), -ett(e) (Old French) Jonet, Luckett (<Luke), Garnett. Exception: ett can be a dialectic contraction of ‘head’ e.g, Dowsett < Douce Head (Sweet head) Exception: Naldrett is a topographical term – “at the alder grove.” -ie, -y “The construction of hypocoristic forms in –ie is characteristic of Scottish English, especially in the north east.” (Hough, 2003, p37) Examples: Beattie, Dickie, Finnie, Petrie, Rennie, Ritchie. -in (Old French) Robin, Rawlin -ings (Old English) -it(e), -tt(e) -ke (Flemish?) Hanke, Silke Viewpoint 1: These may be contractions of names in –kin and –cok. Viewpoint 2: They may be full Flemish pet-forms. The prime example of this form is Jakke (=Jack). “… discussed Lindeman’s evidence that Jack, Middle English Jakke, was not of English origin but derived from the hypocoristic usage of medieval Picardy and Flanders. I set out evidence to show that the suffix –ke can be found in other Middle English personal names and that it is one of a group of Flemish and Franco-Flemish hypocoristic suffixes, including –kin, -man and –cot, that were introduced into England after the Norman Conquest along with numerous Old French personal names and their pet-forms.” McClure, Nomina 28, 2005, p 5) Puzzle: the abundance of the name Jakke is out of proportion to the influence and number of Flemings and Picards. -ken -kin (Flemish) Wilkin “According to both Reaney and McKinley, its [-kin’s] first appearance is among the names of Flemings in the mid-12th century.” (Padel, 2003, p ) Widespread use of –kin pet-forms is not evident until after 1250, reaching a peak in the later 14th C. (though this may be illusory, as earlier it may have been recorded in standardised Latin forms). Puzzle: The abundance of the –kin forms is out of proportion to the influence and number of Flemings and Picards, and does not follow the known pattern of their geographical settlement. Question: Were the –kin pet-forms “imported from the continent or were they co-incidentally re-created in England?” (P McClure Nomina 26, 2003, p 98-99) P McClure throws up for consideration the suggestion that Normans may have used Flemish suffixes to mock and belittle those who were trying to ape and ingratiate through the adoption of ‘Norman’ names. There is an abundance of common nouns ending in -kin in the English language (bodkin, catkin, firkin, napkin, etc.) where it seems to be a simple diminutive, cognate with -chen in German, (Mädchen, Häuschen, Hündchen). The suffix would therefore be part of Old and Middle English, not just of Flemish. So the creation of -kin names in England would be natural and not co-incidental. -MS In Wales: “The English suffix –kin was taken up enthusiastically by the Welsh in the late 13th and early 14th centuries. They used it both in forms based on Welsh names, and in a good number of names drawn from the common European stock in use in England. But there appears to have been a change in its use during the 14th century, and it cannot entirely be accounted for by scribal practice. It was used in the opposite way to that which might have been expected – to refer not to sons, but to fathers, and it came to be felt as a semi-surname, especially in the patronymic context.” (Padel, 2003, p ) Beware confusion with –skin and -in e.g. *Buckskin. -lin -man (Old English or French) Examples: Bateman, Hikeman, Pateman Reaney felt that this was an occupational indicator for ‘servant of’, but the modern interpretation is to discount this (for the Middle English period, the 12th and 13th centuries) in favour of a pet-form explanation (though ‘servant of’ is an accepted explanation of –man names that arose in the 15th and 16th centuries, like Mathewman, Walterman). Hypothesis 1 “The formula ‘Personal name + man’ seems to be limited to post-conquest personal names. In fact, -man is the only Old English hypocoristic suffix to be used with post-conquest personal names.” (Redmonds, 2004a, p ) Hypothesis 2 Since no other OE suffix was transferred, it is more likely that the Middle English usage was adapted from Flanders and area, where the use of –man as a hypocoristic suffix has a long history in the West Germanic languages, e.g. Coleman for Nicolaus was commonplace. (McClure Nomina 26, 2003, p 110-111) There is often insufficient evidence to clearly determine whether –man is being used as a pet-form, occupational compound, or locative or nickname context. -mon -mon is a W Midlands variant of –man, a dialect version that extended from the W and NW Midlands into Cheshire and southern Lancashire. “In contrast, north of the Ribble, and on the eastern side of the Pennines, the voicing of the vowel was a”. In the peripheral areas, e.g. Derbyshire, by the late 14th C, -mon forms were being displaced by –man.” (Postles, 2007, The north through its names pp 107-8) -ment -oc (Old English) -on (Old French) -ota Female diminutive: Examples – Elisota, Emmota, Magota, Cristelota, Iselota, Emecota, Elcota, Mariota. -ot(e), -ott(e) (Old French) Annot, Philpot (Philip is often cited as a feminine name in the 14th century.) (Redmonds, 2004a, p13 ) Exception: Arlott =’lad, youth’ -uc (Old English) Whittock -un (Old French) -us (Old English) Suggestion: pet-form of Old English dithematic women’s names, e.g. Aldus, Edus and Godus, and Old Scandinavian women’s names. “the origin of this suffix is obscure, but in all probability it is a spontaneous hypocoristic formation current chiefly in the north-east Midlands” (Smith The place-names of the West Riding of Yorkshire Part V, p 45) -us is normally affixed to women’s names of Old English or Old Scandinavian origin, e.g. Aldus, Edus, Godus, Ricus. Otherwise it is the Latin nominative ending for male names in official records, e.g. Johannus = John. -ut(e), -utt(e) Pet forms – Double suffixes -el-in (Old French) Tommelin, Hamlin, Jacklin, Wakelin -el-et (Old French) Perelet -el-ot Giblett (< Gilbert), Bartlett (< Bartholomew) -en-el Parnell (<Per<Peter) -et-in Turketine -in-et (Old French) Picknett (Pik-en-et) -in-ot Adnett, Adinet Taxonomic considerations Should (pet forms+individual suffix) be under 1st or 2nd element? Should compound forms –kinson be classed separately or under –kin or –son, or all? Should all suffixes be grouped together, then divided by syllables? Are all the above suffixes evidenced as hypocorisms? Relationship suffixes “Vernacular male forms (-son) were deployed much more freely than female forms (-doghter, -wyf/-wif) which tended more frequently to be construed in Latin than the vernacular.” As a form of hypercorrection, just for the written record? And remember: “these name-forms were not assumed by women, but imposed upon them, perhaps rhetorically, by male clerks or male assessors; they were not fashioned by their bearers.” (Postles 2001, p40, 41n) -daughter, -dochter This suffixed form is found to be unstable, and no example has survived to today (though ‘Daughter’ and ‘Daughters’ were recorded in 1881). In the 14th century in Yorkshire and Lancashire it was fairly common as a byname. The simplex Dafter still exists -son, -ason, -eson, -ison, etc. This form is usually associated with the diminutive form, e.g. Wilson, Hanson Williamson, Dawson. Note that some –sons are in fact ellipses of toponymic –stons or –stone, e.g. Beeson from Beeston. Tyson is the nickname for a firebrand, Dallison fron D’Alencon, Mayson is a form of Mason, Pinson derives from the Old French ‘Finch.’ My son, my son: notes on the –son form The Scandinavian influence debate The –son form derives form the Anglo-Saxon ‘suna, sune or sunu’. The earliest recorded instance is in the Anglo-Saxon Chronicle, Hering Hussan sunu, in the early 7th century. Tengvik unearthed over a hundred Old English personal names linked to sune + variants in the period up to 1100 and twenty-four Scandinavian personal names + sune/sunu. In the period when second names were not the rule, sunu names were infrequent, but not a great rarity. They became increasingly more frequent in the 11th century. Between the early 12th century and the late 13th century, the sunu forms became eclipsed by the filius form, till that in turn was replaced by the –son form “At the level of everyday usage, it seems that there must have been a large degree of continuity with ‘filius’ acting as a translation of ‘sunu’ and then ‘-son’.” (Wilson, 1998, p127) Reaney showed that although names ending in son are characteristic of the ‘north’ [of England], that this phenomenon was not due to any Scandinavian influence. He cites: the continued instability of by-names in the north; Old English –sunu bynames had been more prevalent in the south than the north and that these were compounded with OE personal names, rarely Scandinavian ones. that -son names declined in the 12th century, but resurged in the 13th century, and were compounded with Continental-Germanic names. He concluded that names ending in son were inherently neither ‘Northern’ nor Scandinavian but became associated with the ‘North’ because of instability and the lateness of the adoption of hereditary surnames. (Postles, Defining the North: some linguistic evidence Northern History 38, 2001, p xx) Sörenson’s subtlety Sörenson suggested that the predominance of –son forms in northern England “resulted from the influence of Scandinavian settlement which obliquely reinstitutionalized the Old English suffix –sunu, since there were approximations between Scandinavian and English patronymic systems. In the south compounded patronyms and metronyms in the vernacular form with –son, failed to persist or develop until the later middle ages, perhaps when they were reintroduced by migration, since there was no Scandinavian promotion in that region.” (Postles (1994) “At Sørensen’s request: the formation and development of patronyms and metronyms in late medieval Leicestershire and Rutland”, Nomina 17, p55) A Middle Way? “Two influences have been advanced to explain this onomastic phenomenon: the more extensive deployment of Middle English combined with the greater flexibility in and of bynames and surnames in the north; and, on the other hand, a persistent Scandinavian influence.” “Hitherto these two influences have been proposed by some as antithetical. In particular, the influence of Middle English has been propounded to the exclusion of any Scandinavian context. It would seem, nevertheless, that these two concepts should no longer be regarded as independent and exclusive, but as complementary. In particular, the continuity of the –son element in the 12th century was quite often extended through compounding –son with a Scandinavian name form. Reaney – the proponent of an exclusive Middle English explanation – himself cited Grimmesune and Fornessun in 12th century Yorkshire” (Postles, 2007, The north through its names, p62) This –son form was rare before the late 13th century, but prevalent by 1350. Names were still being formed as late as 1650 (in isolated parts of S. Lancashire). [-Son] “formations are earlier [in the north] than further south, evident from the first decade of the fourteenth century – and in sporadic cases even earlier – accumulating in the 1320s and the 1330s and widespread by the 1340s… The appearance further south is first really apparent in the 1330s, but more particularly in the 1340s. Moreover the densest concentration in lay subsidies occurred in the Cumberland lay subsidy.” (Postles, 2007 North through its names p52) In Lowland Scotland, and areas of Scandinavian settlement (e.g. Shetland and Orkney), the personal name plus the –son form is preferred to the Mac- prefix; “Many –son names are common to southern Scotland and northern England, but Donaldson, Ferguson, Henderson (from Henry) and Patterson (from Patrick) appear to have originated exclusively in Scotland.” (Hough, 2003, p36) Other examples are Manson (from Magnus), Swanson (from Sveinn). “the combination of personal names with –son was not confined to the most common forenames. Variety is evident. For compounds with the most frequent personal names to have become so extensive by the late middle ages suggests that these earlier formations must have been unstable and disappeared. In all these earlier formations, the personal name elements in the byname or surname were not yet concentrated. An inherent variety thus characterised the formation of northern vernacular patronyms and metronyms in the late fourteenth century. Whilst there was some frequency of compounds on the most popular personal names, there was not the concentration that ensued later.” (Postles, 2007, North through its names p53) -son (elliptical): Dixon -son (compound): Jenkinson -son (surname form): Rare examples of –son being added to a surname or byname already in use, e.g. Ballardson, Grayson, Lambson, Spinkson.