i)Iscovi;hy 



Surnames and the 



Chronology of the English 



Vocabulary 



By Professor Ernest Weekley and Miss 

 Dorothy D. Pilkington, M.A. 



To students of the English language the date of first 

 appearance of such words as are not already recorded 

 in Anglo-Saxon is a matter of considerable interest. 

 The great authority in this matter of chronology is, of 

 course, the New Enghsh Dictionary.' But this monu- 

 mental work has made hardly any use of one of the 

 most valuable sources of information, viz. the sur- 

 names of our ancestors recorded in the Government 

 Rolls and in innumerable local documents such as 

 Abbey cartularies. Borough records, Manor Court 

 rolls, etc., of which great numbers have been pubUshed 

 by various antiquarian societies. Surnames are now 

 felt to be proper nouns which have no place in a dic- 

 tionar^' ; but this only applies really to patronymics 

 (Edwards, Johnson, Dicks, etc.), or to those taken from 

 the names of towns and villages (Bedford, Chichester, 

 Bingham, Woodley, etc.). All other surnames ori- 

 ginated as descriptions (Arrowsmith, Baker, Good, 

 Bullock, Partridge, Armstrong, Sheepshanks, Shake- 

 spear, Doolittle, Truelove, etc.) or addresses (Field, 

 Wood, Athill, Bytheway, etc.), and are formed from 

 the everyday colloquial language. If, then, we find 

 the name of Alan le arwesmyth in the Close Rolls for 

 1288-96, we may conclude that the first N.E.D. record 

 for c. 1400 must be at least one century, and probably 

 two or three centuries, too late, as the compound would 

 not be likely to be used as a description or means of 

 identification unless it were a well-understood word of 

 long standing. It is sufficiently clear that any word 

 in general use as a surname must have had a pretty 

 long previous existence in the language. 



In this article only one group of surnames is con- 

 sidered, viz. the interesting class of nicknames of 

 compound form. Such names may be (i) descriptive 

 of occupation, e.g. Bridgman, Cartwright ; (2) names 

 of wares which the bearer made or sold, e.g. Hors- 

 nail, Whitbread ; (3) verbal phrases descriptive of 

 trade, habit, habitual gesture, etc., e.g. Bought- 

 flower (=bolt, i.e. sift, flour), Wagstaff ; (4) physical 

 nicknames, e.g. Broadhead, Lightfoot ; (5) descriptive 

 of dress or equipment, e.g. Curthose, Longstaff ; 

 (6) animal nicknames, e.g. Duncalf , Woodcock ; (7) de- 

 scriptive of characters, e.g. Gentleman, Merry\vcather ; 



» Referred to in the article as N.E.D. A short account of the 

 New English Dictionary appeared in Discovery for July 

 1920, p. 196. 



(8) from abstract qualities, e.g. Fairser\'ice, Truelove. 

 There are also a great many miscellaneous compounds 

 which do not fit into this, necessarily rather rough, 

 classification, such as names derived from the vegetable 

 world, e.g. Peascod, Sweetapple, from coins and 

 money, e.g. Hal penny. Hallmark (for half-mark), and 

 from oaths and exclamations, e.g. Pardew (par Dieu), 

 Godbchere. How such names were acquired is not a 

 question which can be dealt with here. Those which 

 survive as existing surnames are only a very small 

 proportion, but there arc enough of them to give some 

 idea of the extraordinary' and fantastic character of 

 Middle English nomenclature. On the other hand, 

 a very large number of compounds which now exist 

 as common nouns only are found as mediaeval nick- 

 names, and very often, as we shall see, at dates long 

 anterior to the earUest records established by the 

 N.E.D. 



I will now quote a few examples from the various 

 classes of compounds, giving the date at which each is 

 found as a surname compared with the date of the 

 first record given by the N.E.D. of the same word as 

 a common noun. The material I have before me * 

 contains a list of nearly 250 examples which antedate 

 the N.E.D. by one century at least, and sometimes by 

 more than four centuries, besides some thousands of 

 compounds of which dictionaries show no record. It 

 should be premised that the great mass of Enghsh 

 surnames were the creation of the twelfth and thir- 

 teenth centuries (the Anglo-Saxon had usually a single 

 name), and began to become hereditary- and fixed from 

 the fourteenth century onward. Hence, as a general 

 rule, we may assume that any word now well estab- 

 lished as a surname must have been current in English 

 not very long after the Conquest. This may be illus- 

 trated by such a name as Crabtree, common in the 

 West Riding of Yorkshire. The N.E.D. records crab 

 (apple) c. 1420, and crabtree c. 1425 ; but the existence 

 of the latter as a mediaeval surname suggests that both 

 the compound and the simplex must have been familiar 

 words at least a century earlier. This name is also 

 an example of the impossibility of " explaining " sur- 

 names. Did the first Crabtree obtain his name from 

 the tree near which he lived (cf. Ash, Thorne, or Nash, 

 i.e. atten ash, Noakes, i.e. atten oaks), from his " \-inegar 

 aspect " (cf. Sweetapple), or from habituallj' carrying 

 a crabtree staff (cf. Hardstaff).? It is impossible to 

 say which, if any, of these explanations is the true one. 

 The same applies to many of the names dealt with in 



= This material consists of a thesis for which Miss Dorothy 

 Pilkington was awarded in 1920 the M.A. degree of the Univer- 

 sity of London. The character of the investigation was sug- 

 gested by me, and I also indicated to Miss Pilkington the chief 

 sources of which she could make use. The results she has 

 obtained seem to me of considerable value for the historical 

 study of our language. 



