William Rothwell (1991)

‘Historical linguistics is everywhere in retreat’. This pessimistic assessment comes from Robert Burchfield’s latest book Unlocking the English Language.¹ As regards the history of English, such a retreat is greatly to be regretted, since the scope for research into the development of the medieval form of the language is still very great, at least in the areas of semantics and etymology. Not that Dr. Burchfield is unmindful of the work that still awaits future editors of the OED, as he shows when he writes that: ‘The shortcomings of the OED record for words of particular periods and from particular regions are also well known. For example, the vocabulary of the Middle English period, 1066 to about 1475, is being recorded in a much more ambitious way in the Middle English Dictionary‘ (p.169).² Yet these shortcomings are even greater than he recognizes, having a whole extra dimension not so far taken into account by researchers into the historical development of English.³ That extra dimension is Anglo-French, a vital, yet largely unknown factor in the English lexis.

That the Norman Conquest profoundly affected the vocabulary of English is no new discovery, but the precise nature of that transformation has so far been only imperfectly examined and its implications for the study of English etymology only partially understood. Up to the present time there has been no unequivocal acknowledgement that as a result of the events of 1066 there can be no rectilinear approach to the history of English as there is to the history of French. The French language can be taken back in a straight line without any breaks from the present day to The Strasbourg Oaths of 842. At no time during this whole period was the langue d’oil ousted in the northern half of the country from its position as the spoken and written language of the kings and nobles, the judiciary, the Church, the national and local administrations or the mercantile class. Dialectal variations were not ^p174 lacking, but all were dialects of the langue d’oil, what has become modern French. The only rival to French was the Latin used as a formal language of record, but never as a vernacular. In England the situation was vastly different. For some three centuries after the Conquest all the literate classes used French, both spoken and written, very often alongside their native English: a Romance language overlaid the original Germanic one. Written French was especially important in medieval England as being a principal language of record – alongside British Latin – so that the sheer volume of surviving documentary evidence in Anglo-French for this period is overwhelmingly greater than that left behind in English, especially up to the late fourteenth century. No less important than the quantity of Anglo-French is the the breadth of its use. Although scholarly attention has focused largely on its literary productions from the twelfth and early thirteenth centuries, the later non-literary works are arguably of geater importance for the development of English culture as a whole. The law, from Parliament down to the lower courts, the administration of government at national and local level, the commercial and financial framework of the country, all worked through French rather than English. What is more, French was used extensively for all the arts and sciences long before English. This penetration of French into the whole fabric of civilization in medieval England means that the study of English etymology cannot safely confine itself to tracing words back to a cut-off point in Middle English, or even to making a leap across the Channel in search of a ‘borrowing’ from medieval French in order to reach the origin of English terms. The Middle English Dictionary reveals on virtually every page the massive and very obvious debt owed by English to French: less obviously, however, it also reveals that this debt was not built up by ‘borrowing’ in the conventional sense and that in literally thousands of cases forms and meanings were adopted (not ‘borrowed’) into English from insular, as opposed to continental French. The relationship of Anglo-French with Middle English was one of merger, not of borrowing, as a direct result of the bilingualism of the literate classes in medieval England. Terms were adopted often unchanged, sometimes in translation – ‘hot-foot’ (chaut pas), ‘beforehand’ (avant main), ‘behindhand’ (ariere main), ‘send for’ (mander pur) etc.⁴ – but always as part and parcel of a living language in daily use in England, not as isolated, static units of a foreign language borrowed from across the Channel.

Furthermore, as J.D. Burnley has shown in a recent article,⁵ the very style of much of the official writing in English from about 1400 closely reflects that of the fourteenth-century Anglo-French whose mantle it assumed. This is particularly striking in the Rotuli Parliamentorum when English begins to be used instead of the traditional Anglo-French: articles, prepositions and so on are now English, but the bulk of the vocabulary and even the turns of phrase remain very obviously French with minimal anglicization. Nor is this continuity confined to official or administrative documents. It is evident in ^p175 the best-known writer of the Middle English period – Chaucer. Although the extent of the French element in his work has long been common knowledge, many of the meanings he attaches to the French terms he uses will be sought in vain for his period – and often for any other – in the Old French dictionaries of either Godefroy or T-L. Adequate treatment of this important aspect of Chaucer must await another occasion, but may perhaps be illustrated by just one word – ‘bachelor’. Chaucer’s use of this common Old French term in the modern English sense of ‘unmarried man’⁶ has no parallel on the continent, but is found in Anglo-French by the first quarter of the thirteenth century in The Song of Dermot and the Earl and again in the Liber Custumarum⁷ in the early fourteenth century. It is a pointer to the fact that the development of modern English, at least as far as the lexis and syntax are concerned, cannot be adequately researched without taking into account the whole corpus of Anglo-French.

‘Bachelor’ is, of course, just one example of the profusion of faux amis in English and French that are to be found scattered throughout the lexis of both languages and which cannot be adequately explained without recourse to Anglo-French. Since this phenomenon has been dealt with at some length elsewhere,⁸ one simple instance may serve here as a reminder of its widespread diffusion. The modern French appareil and the English ‘apparel’ are clearly related by their form, but show little semantic connection. To find that original connection it is necessary to go back to the fourteenth century and to England, where the semantic shift took place that has separated the words in the modern forms of the two languages. In Medieval French on both sides of the Channel apareil etc. meant ‘preparation(s)’, ‘thing(s) prepared’, hence ‘equipment’, ‘furniture’, ‘gear’ of various kinds (Godefroy, I, 317-8; AND 1, 32): only in Anglo-French, however, was the term applied specifically to clothing.⁹

This distinction between continental and insular French became increasingly marked from the later thirteenth century into the closing decades of the fourteenth. A quite new linguistic situation developed some two centuries after the Conquest as the residents of medieval England who used French in their trade or profession brought an increasingly creative approach to the handling of their adopted language. For them French – like Latin – was not some sacrosanct static structure borrowed from abroad that had to be imitated down to the last detail in order to conform exactly to the linguistic practices of its rightful owners across the Channel. The national dictionaries of Medieval Latin now being prepared up and down Europe are a reflection of the varying forms of that language which developed from one country to another, all very different from the Latin of Cicero or Virgil, but no less valid for that. Both British Latin and Anglo-French were languages of record, tools of their trade used by administrators of all kinds, as well as laywers, clerics, doctors etc., for any document that needed to be preserved. Contrary to received academic opinion, the medieval French of late thirteenth-century London was not just Parisian French marred by comic errors introduced by the ignorant natives of an offshore island.¹⁰ Dr. Burchfield’s ^p176 perceptive observation that ‘East and West German are already demonstrably dissimilar after a period of just over thirty years of political severance’¹¹ is no less applicable to the distinct varieties of continental and insular French that developed after the loss of Normandy in 1204 and were used not over a mere thirty-year period but for about two centuries as the societies of England and France grew further and further apart until the insular form was replaced by Middle English.

The failure of scholars to appreciate fully the importance of Anglo-French in the making of the English lexis extends from individual words right up into whole areas of the culture of medieval England. For example, for the authorities on English etymology ‘troglodyte’ is adapted from the Latin and first attested in the middle of the sixteenth century – an example, one might be tempted to conclude, of the well-known re-birth of scientific interest in many fields that characterized this period. However, the term is found in French, only thinly disguised, in a Cambridge manuscript that may date from the late twelfth century and that has been available in published form for nearly seventy years.¹² The adaptation of the word from the Latin took place in England some three and a half centuries before the date given by the authorities, but it was taken into the vernacular of the literate laity – French – not English. This example shows that a more measured approach may perhaps be called for when contrasting the ‘darkness’ of the Middle Ages with the ‘light’ of the Renaissance. Similarly, the noun ‘crescent’is recorded in a twelfth-century Anglo-French medical work centuries before being attested in English and wrongly ascribed by the authorities to continental French.¹³ Again, the ‘spaniel’ and the ‘terrier’ are both found in Anglo-French before appearing in Middle English, indicating that their names are not borrowed directly from continental French.¹⁴ Only the MED, however, amongst the English authorities refers to Anglo-French in connection with these words, and only for ‘spaniel’.¹⁵ Another simple example is the humble ‘dandelion’, said by the authorities to be first found in English in the sixteenth century and to be an adaptation of the French dent-de-lion. Yet Godefroy ‘s sole example of dandelyon (IX, Comp. 304b) comes from the Englishman Palsgrave, a dating which clearly refutes the claim that the sixteenth-century English word is ‘adapted’ from continental French. Even the authoritative Französisches Etymologisches Wörterbuch cannot be relied on in this case, with its rather desperate claim that the form comes wahrscheinlich from Lyon in the fourteenth century.¹⁶ Once again, the neglected Anglo-French has new evidence to offer. On p.99 of his Plant Names of Medieval England (Cambridge, 1989) Tony Hunt gives numerous examples of the word in various spellings in both Anglo-French and Middle English going back to the late thirteenth century, with others of roughly similar date in his Popular Medicine in Thirteenth-Century England (Cambridge, 1990), pp.317 and 324. Both these fundamental research books provide a wealth of new linguistic evidence ^p177 across the whole spectrum of botanical and medical terminology that must inevitably lead to a far-reaching revision of attitudes towards the degree of knowledge possessed by doctors/herbalists in the Middle Ages, towards the vocabulary of Anglo-French through which much of their teaching was transmitted and, consequently, towards the importance of Anglo-French in the historical development of both French and English.¹⁷ In this area historical linguistics still offers a great deal of scope for basic research.

The pair ‘carpenter’ and ‘joiner’ provide a good example of a more difficult case, where a knowledge of both insular and continental medieval French and of the pitfalls lurking in the dictionaries of medieval French are of prime importance for the etymologist. The OED puts ‘carpenter’ and ‘joiner’ on a par, giving the origin of both as Anglo-French, taken ultimately from continental Old French. They are not, however, of similar origin. Whilst ‘carpenter’ is found on both sides of the Channel in the Middle Ages, ‘joiner’ is an Anglo-French creation and not attested in continental French at all, although its method of formation from the French verb joindre is impeccably regular (cf. acheter/acheteur, porter/porteur, vendre/vendeur etc. etc.). The authorities on English etymology claim that the English ‘joiner’ comes from Anglo-French joignour, Old French joigneor. This faulty information can only come from Godefroy ‘s dictionary, but his Old French joigneor referred to is no more than a ghost word. His sole quotation under this entry comes from a text composed in England in 1396¹⁸ and contains only the oblique form joignour, the nominative head-word joigneor (IV, 648b) being Godefroy ‘s own creation, just one of the many unattested nominative forms he made up to satisfy the demands of Neo-Grammarian orthodox Old French grammar.¹⁹ Since the MED‘s first attestation of the English ‘joiner’ in the sense of a worker in wood is as early as 1322, the word can hardly be derived from either a form not attested at all or from one not attested until seventy years after the English. As far as present knowledge goes, there is no evidence whatever for the existence of a continental joigneor. There is, however, an Anglo-French nominative form joinere, attested for the later thirteenth century in the sense of ‘link, coupling’ on a cart,²⁰ for which T-L uncritically makes a separate entry immediately below that of the ghost-word joigneor, where it copies Godefroy ‘s quotation.²¹ The T-L thus muddies the waters by giving two nominative forms of the same word under separate headings, one genuine, the other a nineteenth-century invention, and so creates the misleading impression that there are two different nouns involved. They are, however, clearly different senses of the same word, an Anglo-French creation, found in both nominative and oblique forms in England, although this is not made clear in either of the dictionaries of Old French.

Again, the OED tells us that ‘gobbon’ is ‘presumably a[dapted from] O.F. *gobon, an unrecorded form’. Yet the Anglo-French – not Old French, i.e. continental – gobon, together with a translation into Middle English, has been recorded in print since the beginning of this century in the Transactions of the ^p178 Philological Society over the signature of no less a scholar than W.W. Skeat.²² In fact, another attestation from around the same period is found in Tilander’s Glanures lexicographiques.²³ The derived verb goboner occurs in an Anglo-French culinary recipe of the late thirteenth century.²⁴ The related ‘gobbet’ is taken back by the OED to gobon, but it too is found in Anglo-French in the fourteenth-century Livre de Seintz Medicines of Henry of Lancaster.²⁵ In fact, the MED shows just how close were the links between Anglo-French and Middle English when it gives ‘c.1330’ as the date for the introduction of ‘gobet’ into English, but the derivation is incorrectly given as ‘OF’, not ‘AF’.

A more complicated case is presented by the English verbs ‘to hack,’ ‘to hash’, and the nouns ‘hash’ and ‘haggis’, whose relationship the authorities find confusing. Indeed, the origin of ‘haggis’ is usually said to be unknown, and connections with the French hachiz ‘hash’ are denied. If, however, we bring some Anglo-French and Middle English evidence to bear on the question, a different picture emerges. Whilst the new OED still persists in deriving ‘to hack’ from Germanic sources, without any mention of French, the Middle English Dictionary is nearer the mark in attributing it to ‘O.F.’ (but not ‘A.F.’, as it should). Amongst the quotations in the OED is one from a book of cookery recipes dated c.1440 in the sense of ‘to cut up into small pieces’, i.e. ‘to hash’. The MED takes this sense back in time to c.1325, but still without any mention of Anglo-French. Yet in an Anglo-French medical text from the second half of the twelfth century a mixture of herbs is to be hachez sur un ais (‘chopped up on a board’), an expression repeated a little farther on.²⁶ About a century later a verse recipe for staunching blood from a wound recommends that: ‘Le ortie menuement hagee En eisil fort seit destempree’ (‘Finely chopped nettle should be soaked in strong vinegar’).²⁷ The form of the verb here – hagee – is worth noting in view of the OED‘s form ‘to hag’. As for the assertion that the origin of ‘haggis’ is unknown, the clear refutation of this has been in print for close on a century and a half in the first edition of one of the manuscripts of the Anglo-French Treatise of Walter of Biblesworth,²⁸ with a reminder being printed from another manuscript in 1929.²⁹

Under ‘trick’ as noun the OED has this curious note: Both sb. and vb. have in English had developments of signification unknown to F. triche and tricher . In the entry for ‘trick’ as verb the dictionary has another note that reads no less strangely. Commenting on the first English sense of ‘to deceive by a trick; to cheat’, the editors state that it did not appear until late in the sixteenth century and go on: ‘The date of appearance is too late to refer it to the Norman-Picard F. trikier, triquer ‘. In fact, there is no reason whatever to cross the Channel to find an explanation of English ‘trick’ as either noun or verb. The verb is well attested in both transitive and intransitive form in Anglo-French from the twelfth century.³⁰ That it now has meanings not found in the continental French triche and tricher is simply because it developed from Anglo-French on English soil in English society and is not just a foreign borrowing imported in isolation as a static item of vocabulary. As early as the thirteenth century Anglo-French was using tricheur, tricherie, ^p179 tricherus, tricherusement as well as the verb, a whole ‘semantic family’. The OED‘s reference to the noun triche illustrates once again the danger of using Godefroy without an adequate knowledge of medieval French: all his examples, though undated, are late, none being earlier than the end of the fourteenth century, only a few years before the noun ‘trick’ is attested for English, so triche had little chance of influencing the English ‘trick’. More important than this dating, however, is the fact that the OED contains in its own pages the solution to its false problem of the late date of the verb ‘to trick’. Under treche, trich, meaning ‘to deceive, cheat’, the dictionary gives quotations ranging between c.1230 and 1425. Add to these the forms ‘treacher’ (c.1290), ‘treacherous’ (c.1330), ‘treachery’ (c.1225) and ‘trichard’ (c.1327) quoted in the dictionary, and it will be clear that Middle English absorbed the Anglo-French network of terms for ‘trick’ etc., and handed them down in the normal way to modern English. It will be very surprising if the forthcoming ‘T’ fascicle of the MED does not provide ample evidence on this point. Middle English was no more a slave to spelling than was Anglo-French: semantics takes precedence over orthography.

A very curious case is that of the verb ‘to grouse’. The authorities maintain that it appears in the nineteenth century as a slang term and that its origin is unknown. They note, however, that it has an unexplained resemblance to a Norman dialect form of the Old French groucier. Indeed, it has. The verb grucer was in frequent use in Anglo-French from the second half of the twelfth century onwards, not as a slang term but in highly respectable literary works,³¹ and from it there developed nouns – gruz and grucement ‘grumbling’, gruçur and the feminine grouceresse ‘grumbler’ – as well as an adjective – gruçus ‘grumbling, ill-tempered’. By the thirteenth century a version of the Ancrene Riwle uses grusceure to mean ‘rumbling (of the stomach)’.³² In other words, a complete semantic unit was in use in French on English soil for centuries. Furthermore, the continental group of grouc(i)er and its cognates was in very wide use in France all through the Middle Ages and was not at all a mere dialectal peculiarity confined to Normandy. From Anglo-French the verb and its ‘family’ were taken into English where they are abundantly attested by the MED in serious literary works from the thirteenth century right into the middle of the fifteenth.

The need to understand the role of Anglo-French in the development of English is shown again in attempts to explain doublets. For instance, Dr. Burchfield states that: ‘Some Romance words, borrowed at different periods, produced doublets in English’.³³ His examples include ‘cadence/chance’ and ‘dignity/dainty’. This concept of ‘borrowing’ may be applicable to the adoption in our own day of terms such as haute couture or boutique into English, ‘marketing’ into French or ‘manager’ and ‘leasing’ into German. The linguistic situation in medieval England, however, produced a quite different kind of transfer of words from French into English, a transfer based on the fact that generations of educated Englishmen passed daily from English into French and back again in the course of their work. Very many ^p180 of the French terms they used had been developing semantically on English soil since 1066, were absorbed quite naturally with all their semantic values into the native English of those who used them and then continued to evolve in their new environment of Middle English. This is a very long way from the traditional idea of ‘linguistic borrowing’.

In the cases in point, ‘chance’ is not a ‘borrowing’ at all, being a normal part of insular French from the time of the Conquest onwards and used in various senses developed from its original etymological meaning of ‘fall’ – e.g. ‘throw (of dice)’, ‘(good/bad) fortune’, ‘accident’ etc.. In fact, by about 1340 we find it in what looks suspiciously like the etymological origin of the modern English ‘windfall’, tucked away in one of the Anglo-French Year Books: ‘ceo q’il avera luy avendra par volunté de homme ou par cheanse de vent’ (literally, ‘as a wind fall’).³⁴ This semantic development is not recorded for the continental form of medieval French at any period. Theoretically, the so-called learned borrowing ‘cadence’ ought to be introduced from the continent much later into a more highly educated society, but perversely appears in Chaucer’s English in its musical sense some two centuries before being attested in the Renaissance French of the continent. Its appearance may well be a reminder of the well-documented presence of Italian bankers and merchants in fourteenth-century England. To list the pair ‘cadence/chance’ as doublets, then, although this would be no more than a superficial categorization, explaining nothing, is at least not contradicted by the facts, whereas the attempt to explain them by chronology is demonstrably false.

Nor are ‘dainty’ and ‘dignity’ straightforward doublets with their form and sense determined by their provenance and date of entry into English. Deinté in the sense of ‘delicacy’, ‘choice morsel’ is attested in the early twelfth-century Anglo-Norman Voyage of St Brendan,³⁵ but the learned digneté is recorded only a few decades later in Adgar’s Marienlegenden.³⁶ The Middle English Dictionary records both ‘dainty’ and ‘dignity’ within a few years of each other in the first half of the thirteenth century, but, as with the ‘chance/cadence’ pair, they appear in the ‘wrong’ order, the learned ‘dignity’ being attested before the popular ‘dainty’. The semantic separation of the two was determined by the difference of register in which they were used in insular French, long before they passed into English. Both forms were developing specialized senses in the French used daily in medieval England over many decades, and it is from this French that they eventually pass into Middle English.³⁷ Once again, register and social use provide the key to understanding, not a false chronology. These examples may perhaps suffice to demonstrate that it is meaningless to refer historically to doublets in English without crossing the language barrier into the time when English was not yet the dominant language of the classes that determined linguistic development in England.

In numerous instances where the authorities on the etymology of English fail to pick up the Anglo-French link in the etymology of an English word the problem is caused by prefixation, as in the French attiser and modern English ‘to entice’. Although the OED puts the Old French enticier as the ‘probable’ origin ^p181 of ‘to entice’, and even mentions the now obsolete English forms ‘attice’ and ‘tice’, its original compilers did not have at their disposal the Anglo-French material which would have turned their probability into a certainty. Godefroy has no less than four separate entries for this verb – atisier (I, 478), enticier (III, 264), entiser (III, 266) and atiser (VIII, 229) – yet the quotations under each of them prove beyond any doubt that we are dealing with the same verb. The Anglo-French forms show a similar variety of prefix, and even a lack of prefix as mentioned for English in the OED – atiser etc., enticer etc., t(e)icer. Ignoring the prefixes, then, as irrelevant, both continental and insular forms use the verb figuratively from the twelfth century in the sense of ‘ to provoke, incite, stimulate’, both develop it syntactically to attiser quelqu’un que … ‘to provoke etc. someone to …’, and also semantically to the adjoining meaning of ‘to entice, allure’.³⁸ What has happened since the medieval period is that modern English has limited the sense of ‘to entice’ to that of ‘ to allure (by the offer of some inducement)’, whilst modern French has limited attiser in the figurative sense to the meaning of ‘to provoke, incite’.

Similarly, the Old French besiller is found in both the general and the legal register of Anglo-French in various spellings and also with the prefixes a-, em- and en- which do not alter the sense, as is quite normal in medieval French in general and Anglo-French in particular. The basic meanings of ‘to do away with’, ‘make away secretly with (goods or money belonging to another)’ are found in a wide variety of texts in French at the end of the thirteenth century and beginning of the fourteenth, a full century before the first attestation of ’embezzle’ in the OED, which is, in turn, nearly two centuries before the ‘1600’ given by Geoffrey Hughes³⁹ The contexts in which these various forms of ’embezzle’ are found refer not only to souls, charters, legal documents and a strong-box, but also to the disappearance of royal money given as a charitable bequest and to customs duty being fraudulently avoided and presumably pocketed by individuals,⁴⁰ these latter senses being very much in line with the modern English meaning of the term. As might be expected, the MED, under besilen and embesilen, shows that Middle English took up these verbs from Anglo-French and was using them abundantly in records after the first quarter of the fifteenth century. Common sense, of course, would suggest that the urge to ’embezzle’ did not suddenly descend on members of English society in 1600: the concept and also the modern word to describe it were common in England centuries earlier.⁴¹

‘Embezzle’ calls to mind ‘bribe’ in the same semantic area. Although this is found in Chaucer and his contemporaries, its meaning at that time is not unequivocal. There is no clear semantic path from the continental bribe ‘small piece’, ‘scrap’ to the modern English ‘bribe’. The earliest date given in the MED for the modern sense is near the middle of the fifteenth century. In Anglo-French, however, bribe was in use in its modern English sense without any ambiguity in the mid-fourteenth century.⁴² Once again Anglo-French is the missing link in the etymological ^p182 chain.⁴³ Yet, as with ’embezzle’, the introduction of the word bribe certainly cannot be interpreted as marking a new attitude towards the corruption of officials brought about by the nefarious activities of William of Windsor in fourteenth-century Ireland. In 1258, although no generic term for ‘bribe’ is mentioned, the Annales de Burton⁴⁴ stipulate that no sheriff shall accept presenz such as money, corn, wool, lambs etc. and that, in order to facilitate this, his term of office shall not exceed one year (p.454).

In all the cases referred to above a true appreciation of the role played by Anglo-French in the formation of the English lexis would have obviated a great deal of error. Yet English scholars can hardly be blamed for not fully understanding its importance. The widespread ignorance about this lost language of medieval England is not surprising given the neglect that has marked the attitude of medievalists towards it for half a century and more. Specialists in Old French have tended to confine their interest in what is usually called Anglo-Norman to the first century and a half following the Conquest, considering that the later period – the second half of the thirteenth up to the early years of the fifteenth – is one of linguistic and literary degeneracy that may safely be ignored. The result of this attitude has been that Anglicists have had no adequate guidance in assessing the role played by French in the formation of English vocabulary during the medieval period as a whole. Moreover, were they to attempt to tackle the question by their own efforts, they would come up against a number of difficulties. In the first place, the current dictionaries of medieval French draw preponderantly on continental sources, their infrequent insular quotations being taken almost exclusively from the earlier period, roughly before the middle of the thirteenth century. Secondly, when they do quote an insular text, the dictionaries do not identify it as such or provide a date, so that the reader needs to know his way about the extensive literature in French on both sides of the Channel during the medieval period before he can distinguish between the two types of French. Thirdly, as was mentioned earlier, both the old Godefroy and the as yet incomplete T-L follow the Neo-Grammarian practice of often using as head-words forms which are theoretically in conformity with the ‘sound laws’ claimed to determine the historical development of French, but which are not attested – a potential source of serious error. This means that there can be no uncritical reliance on the standard works on English etymology written a generation ago, since their compilers did not have access to a great deal of very necessary material in both Middle English and Anglo-French.

The danger of failing to appreciate the true nature of the linguistic situation in medieval England, however, goes beyond individual words, doublets and faux amis. Without an understanding of insular French, English scholars are liable to go badly astray in assessing the history of whole areas of ^p183 their native language. French cooking has long enjoyed a high reputation, but authorities are not agreed as to when it first came into prominence in England. Dr. Burchfield writes that: ‘The culinary revolution, and the importation of French vocabulary into English society, scarcely preceded the eighteenth century’. Professor Hughes, in his Words in Time, would move the date of this ‘culinary revolution’ back to the fifteenth century (p.43), but both these dates accord ill with two recently-published thirteenth-century culinary collections in Anglo-French,⁴⁵ which contain sufficient new terms and new techniques specific to England, as the editors emphasize, to show that an important advance in this area of domestic science had taken place centuries before. This is hardly surprising in the light of the close connections of all kinds between medieval France and England. False chronology leads to another serious error of cultural interpretation when Professor Hughes refers on p.60 to our modern meaning of ‘courtesy’ being recorded c.1513 and deriving ‘from the pragmatic Renaissance ethos of self-improvement, evidenced in the publication of numerous courtesy-books.’⁴⁶ Without waiting for the Renaissance, this type of book, written in Anglo-French, had been currently in use by the educated classes in England from the first half of the thirteenth century.⁴⁷

The dangers attendant upon this lack of knowledge about Anglo-French and its role in the formation of English may be clearly seen whenever attempts are made to link social and lexical development. The dating of social changes and innovations by reference to the semantics of a national language is valid only if the linguistic system used as the point of reference has been uninterruptedly in place during the whole of the period studied. As has been explained earlier, in France this condition is fulfilled, but not in England, where the social status of French meant that it was used extensively in preference to English for written records of all kinds from the twelfth to the fifteenth century. As a result, given that English was the native language of the majority of those who wrote this form of French, many hundreds of words would have been in daily use in spoken English for generations without necessarily being committed to parchment or paper, the people who used them being bilingual in varying degrees, but using only one of their two vernaculars – French – to set down in writing their decisions, judgements, transactions etc. for posterity.

This applies in particular to the language of English law. From the time of the Leis Willelme (1110-1120) and Magna Carta⁴⁸ onwards the legal register of Anglo-French developed in parallel with the society it served. When Middle English took over the role of Anglo-French as a language of record, it simply adopted the existing legal terms. Yet, although the all-pervading presence of French in English legal terminology has been accepted as a fact for generations, the specifically Anglo-French nature of English law has not been generally recognized outside the legal profession. From ‘abet’, ‘assault and battery’ through ‘impeachment’, ‘suit’, ‘try’ and ‘trial’ to ‘vicinage’, ‘vicarage’, ‘void’ and ‘waste’, the legal register of English contains literally hundreds of terms whose meaning and/or form is different from that found ^p184 in continental Old French. This is no mere linguistic quibble: ignorance of this fact leads to serious errors in the interpretation of social development by writers on the history of English.

Just one example may suffice to make this point. We read in Words in Time that: ‘As the explicit terms for sexual activity became unacceptable and then taboo, numerous general latinized words were drawn into the “semantic vacuum”. Among them were rape (1482) …’. However, leaving aside the awkward fact that ‘rape’ is given in the MED for 1425, as early as c.1289 rape is defined in the French of the English lawyers as the forcible abduction of a woman;⁴⁹ in c.1292 the law defines it, again in French, as male violence against a woman’s body.⁵⁰ Therefore, for well over a century before the first attestation of ‘rape’ in Middle English, the law of England, expressed in French but executed by English justices, had been using these definitions throughout English society. Rape is an Anglo-French term, not found on the continent. Admittedly, the Latin rapum is found even earlier than the French, but it was the widespread use of French in the actual pleading and detailed written accounts – as distinct from the brief formal Latin record – of cases in the English courts of law from the second half of the thirteenth century onwards⁵¹ that has resulted in so very many English legal terms like ‘rape’ having a French look about them. The date 1482 – even if it were correct as representing the earliest attestation of ‘rape’ in English, which it manifestly is not – would merely indicate a linguistic appearance: it has no validity as a semantic marker for the introduction of a new legal or social concept, nor does it fill a semantic gap. As far as ‘rape’ is concerned, there never was, in fact, a ‘semantic vacuum’ as claimed by Hughes. In the early twelfth century both the offence of rape and its penalty were laid down in the Leis Willelme,⁵² but the term used was purgesir. This verb continued in widespread use for another two hundred years and more. In the second half of the thirteenth century the verb ravir, originally ‘to plunder, steal’, takes on the sense of ‘to rape’.⁵³ What happened was that in the later decades of the thirteenth century a growing and increasingly influential body of English lawyers codified the law of the land as it stood and proceeded to modify and refine it over several generations in many hundreds of cases argued out in the courts and recorded in the Year Books, Cases in the Court of King’s Bench etc. etc.⁵⁴ Similar remarks could be made about many other legal terms – e.g. ‘larceny’.⁵⁵ Not English, not continental French, but later Anglo-French is the key to an understanding of the legal register of English, even today.⁵⁶

On occasion, the very fact of an English word having a legal connotation at all is not recognized. ‘Place’ is handled by all the etymological authorities as a general term of French origin, and the OED refers to the parliamentary expression ‘another place’ without mentioning any suggestion of a legal sense, even though it gives a quotation from Palsgrave (1530): ‘Place where justyce is mynystred, parlement ‘. The reference behind ‘another place’ is to Parliament as a court of law, as Anglo-French evidence from the early fourteenth century onwards makes clear, with numerous examples of place ^p185 meaning ‘court’, together with chef place ‘High Court’, commune place ‘Court of Common Pleas’, place le roi ‘King’s Bench’, place de la Corone ‘Crown Pleas Division’.⁵⁷

Not only the technical language of English law, however, requires to be approached through Anglo-French. The etymology of everyday English also needs a comprehensive, multi-lingual study of semantic fields in the medieval period. As an example we may take the case of ‘garble’ and its ‘family’. Its history shows how words moved almost imperceptibly from one language to another at the crucial period when Anglo-French was giving way to Middle English. ‘To garble’ originally meant not ‘to confuse’, but ‘to remove the rubbish (from imported consignments of spices)’, and it is first found back in the time when many commercial records were kept in Anglo-French. As far as present knowledge goes, it is from 1393 that any investigation of ‘garble’ must start. In the Records of the Grocers’ Company are to be found a whole set of terms based on ‘garble’.⁵⁸ The verb garbeler, the adjective garbellable and the noun garbelour (the man who cleans the spices) are all found in 1393 in Anglo-French, with the product garbelure (the rubbish left behind after the cleaning) coming in 1436.⁵⁹ The Grocers’ Records have the English verbal form ‘garbelyd’ by 1420, the noun ‘garbell’ by 1434, the compounds ‘garbelure’ in 1425 and ‘garbelage’ in 1431. That this is an Anglo-French development, not a borrowing from continental French, is made clear not only by the obvious passage from Anglo-French to Middle English forms in the same set of records, but by the fact that the dictionaries of medieval French provide no continental evidence to antedate the insular references given above. Godefroy has only two examples of gerbelle from the mid-fifteenth century, both of them misunderstood and consequently mistranslated as ‘sorte d’épice’ (IV, 265a), together with two examples of the verb grabeler used figuratively in the sense of ‘to search’ dating from the middle of the sixteenth century;⁶⁰ the new, predominantly literary T-L offers no evidence at all for the whole ‘garble’ group. Moreover, the modern English ‘garbage’ is connected to the ‘garble’ group – as well as to the modern French grabuge (also found as garbuge and gaburge). This means that the history of the group in Anglo-French can be taken back to 1318, at least. In the Household Ordinance of York of that year⁶¹ reference is made to the sergeant garbagere whose task is to clean the royal poultry for the cooks. His title must imply the existence of a basic Anglo-French noun garbage ‘offal’ which, although not so far attested, may reasonably be postulated to antedate the function of the garbagere who deals with it.⁶²

Turning to the related world of the economy, the role of Anglo-French in the development of the English vocabulary of commerce and finance has not been generally appreciated up to the present. Abundant documentary evidence from Anglo-French sources, both literary and administrative, suggests that the vocabulary necessary for a money economy was in place from the thirteenth, if not the late twelfth century.⁶³ In the mid-twelfth century the insular Proverbes de Salemon has the verb porchacier in the sense ‘to ^p186 acquire, purchase’,⁶⁴ and by the late thirteenth century Britton was making a legal distinction between land obtained by inheritance and by purchaz, that is, by transference of rents, etc..⁶⁵ The fourteenth-century Livre de Seintz Medicines of Henry of Lancaster contains an interesting confirmation of the early progress of the verb purchaser towards a monetary sense. Henry says that it is folly to refuse Paradise and ‘avoir et purchacer dolour pur joie’.⁶⁶ The variant manuscript at this point has achater instead of purchacer. This Anglo-French evidence provides the necessary background to that found in the MED, where ‘purchase’ is shown to carry the modern sense of ‘buy’ from 1340 onwards.

By the later twelfth century the names of various coins, of continental as well as domestic origin, are mentioned in Anglo-French documents without explanation of their meaning, indicating that they were a normal, accepted part of society.⁶⁷ Their presence confirms that medieval England was very much part of a money-based trading community extending over Western Europe. In the mid-thirteenth century penalties were announced for those found ‘clipping’ coins,⁶⁸ and one of the Provisions of Oxford involved the reform of the Royal Mint in London.⁶⁹ In 1271 the Jews in England were accused of being behind the counterfeiting that was rife at the time, and Christians as well as Jews were accused of involvement in dealings in ‘clipped’ coins.⁷⁰ By 1279 the amount of pure silver to be used in the new livre is laid down in the Red Book of the Exchequer, whilst at the close of the century Edward I found it necessary to reform his currency in an attempt to check the activities of the counterfeiters and ‘clippers’ who were damaging the country’s economy.⁷¹ His Statute of 1299 also proclaims that any person could change valid foreign currency of any kind at his Exchange. By 1300 provision is being made in the Red Book of the Exchequer for royal mints to be set up in various cities in England, each with its chaungur. The mestre de la moneye le Roi is to pay the wages of the man who engraves the coins and – illustrating the international dimension of this economy – the Frescobaldi of Florence are not only given permission to run the mints at Hull, Newcastle-upon-Tyne and Exeter, but are to receive a fee and have the right to buy up all manner of counterfeit coins and take them to be changed into good money at the king’s mints.⁷² Even before this, in 1279, Archbishop John Peckham had written in French to Edward I in the first year of his consecration to the see of Canterbury, reminding him that he had granted the Archbishop the ancient right to run the mint at Canterbury.⁷³ This mention of an ancient right shows clearly that the money economy and the involvement of the Church in its operation goes back to a time much earlier than the later thirteenth century. Peckham’s letter refers to trois coynz ‘three dies’ for stamping coins, and in 1320 mention is again made of punches necessary to stamp the coins at the mints.⁷⁴ Not only was the money economy the norm before the fourteenth century, but there was such a shortage of coin in 1339 that the acceptance of foreign coins as legal tender alongside the esterling was recommended in Parliament.⁷⁵ Moreover, towards the end of the thirteenth century Nicole Bozon refers to percentage interest being paid.^{76 p187} Around the same time there is clear evidence of the movement of money on an international scale by means of paper transactions. At the beginning of the fourteenth century the bishop of Norwich has to tell Edward I that a shortage of cash makes it impossible for him to send the 400 marks requested, because various people have paid their tithes not in cash but by credit.⁷⁷ In 1331 the prior of Christ Church, Canterbury, writes to an English cleric in France, asking him to pay to the merchants of the Bardi Company in Paris the money earned by the priory’s wines in France and get the Bardi to instruct their members in London to pay the equivalent to Christ Church.⁷⁸ By the middle of the century there is a letter in French from Edward III protesting against the circulation of fraudulent bills of exchange involving Lombard and French merchants as well as their English counterparts.⁷⁹ In this connection, the Maniere de Langage ⁸⁰ of 1396 gives an imagined conversation in a French market between the English traveller and a market trader. The bargaining involves not only payment in cash but also promissory notes with financial penalties for failure to pay by the agreed date and the threat of imprisonment (pp. 75-6).

The hold exerted by the Lombard brokers or middlemen on the economy of England by 1376 is evident from the xenophobic attempt to have them expelled in that year on the ground that they were behind a web of clever financial trickery centred on usury.⁸¹ Significantly, the same text of 1376 goes on to accuse these Lombard brokers of being Jews, Saracens and spies, an implausible combination perhaps, but a clear indication that the time-table for the development of the semantic field of xenophobia in England needs to start back in Anglo-French no less than does the study of the legal register.⁸² Paen etc. ‘pagan’ and Jeu etc. ‘Jew’ are regularly used in a derogatory sense from the time of the Anglo-French Comput of the early twelfth century onwards,⁸³ whilst bugre occurs with the derogatory meaning ‘sodomite’ in the thirteenth century.⁸⁴ In 1283 Archbishop John Peckham warns the Queen that it would be wrong to keep any financial gain deriving from Jewish usury, saying that anyone who would have her understand otherwise is bugre.⁸⁵ The heady mixture of religious bigotry and financial jealousy widespread in the medieval period is responsible for much of this xenophobia, so it is not surprising that jeuerie ‘Jewry’ should find its way into opprobrious expressions by the thirteenth century.⁸⁶ The xenophobic forces at large on the darker side of English society were not left without a voice before the end of the Middle Ages simply because English had not yet established itself as the sole vernacular in use in England.

A perfect example of the interlocking of historical linguistics and history is provided by the reckoning of time, where English is out of step with the rest of western Europe. This again can be explained only in terms of the languages of medieval England, including Anglo-French. One of the most fundamental changes brought about by a late medieval advance in technology – the invention of the striking clock – can be precisely illustrated by linguistic evidence. In brief, until the late fourteenth century the passage ^p188 of time was usually estimated roughly by the progress of the sun round the sky, with the ringing of the canonical hours in church or monastery for the religious offices providing the sole mechanical indication of the time of day, at least for those who could interpret the ringing.⁸⁷ Only when the striking clock became a feature of important public buildings from about 1370.⁸⁸ do we find the emergence of our modern English expression ‘o’clock’ and the generalization of the day of twenty-four equal hours. The earliest evidence for this change is given by the Middle English Dictionary in a quotation from 1370 – ‘hegh none smytyn by ye clocke’ – with the first reference to a clock hour coming in 1389 – ‘vij of the clokke’. All the evidence from this period shows that the passage of time was recognized aurally, not visually: it was the striking of the clock, not the sight of a finger on a dial, that is reflected linguistically.⁸⁹ The earliest Anglo-French quotations mentioning clock hours confirm this.⁹⁰ A few decades later, in 1428, we find the earliest record of clocks being imported: ‘iij orlages, valor iiij li’.⁹¹ Clocks costing just over a pound cannot have been the large municipal ones referred to above, so this importation may indicate the spread of domestic clocks. Bound up with this change was the name of the clock itself, an extension of meaning from the Anglo-French cloche/clocke ‘bell’ to cover the instrument responsible for the ringing of the bell.

Furthermore, the move of the canonical hour nonne (‘nones’, the ninth hour) from mid-afternoon to midday, as evidenced in the quotation from the MED given above, is again found in Anglo-French, as is the meal eaten at that time – nonsion, now ‘luncheon’.⁹² Until around the middle of the thirteenth century insular texts, like continental ones, make a clear distinction between midi and nonne.⁹³ In about 1260, however, an Anglo-French religious text shows the equivalence of nonne and midi on a Saturday, nonne being moved forward so that workers who finished work on week-days at nonne (mid-afternoon) could stop at midday on Saturday to honour the Virgin.⁹⁴ This move in time is confirmed by a medical receipt of about the same date, in which the medicine is to be taken three times a day – in the morning, at noon and at night: ‘Chaut le beive plain mazelin Au matin, a noune et au seir’.⁹⁵ Some thirty or forty years later later the same three time markers are found in a court case in which the bailiffs tried to take a distress against a recalcitrant farmer in Tideswell – ‘de prime joekes au noune … de noune jekes a seir’.⁹⁶ In 1334 a petition presented in Parliament complains that at Berwick-on-Tweed: ‘les marchantz … ne mariners ne pount entrer ne issir a hour de novene en matyne, ne au soir, fors a grante daunger’.⁹⁷ The addition of en matyne here precludes the interpretation ‘mid-afternoon’. However, the move forward in time of nonne must be credited to the religious, who had a stronger incentive than other people, an incentive more basic than the noble desire to honour the Virgin: in the monastic life this particular canonical hour was inseparably linked to the time of the only substantial meal of the day. One of the Anglo-French versions of the Ancrene Riwle sets out detailed instructions as to when the religious are to say their offices: the times vary between summer and winter and also between ^p189 feast days and fast days, but eating the main meal of the day and the office of nonne are always referred to together. The key phrase concerning the timing of this office is this: ‘En yver, avant mangier quant vous junez’.⁹⁸ This means that on the many fast days in the cold of winter the religious would have to remain with rumbling stomachs until the office had been said in late afternoon or else they would have to ring nonne earlier. That this is no figment of the imagination is shown by the following quotation from a thirteenth-century treatise on the Seven Deadly Sins: ‘trop ardant feim fet humme manger devant houre e tot (i.e. ‘prevents’) june e abstinence’.⁹⁹ A better illustration of the bond linking semantics and society would be hard to find. Equally, it would be difficult to find a clearer example of the need for a trilingual approach to sociolinguistic questions in medieval England. Anglo-French was not alone in moving forward the time of nones: the links with Middle English and the basic fact that the majority of Anglo-French users were English means that similar trends may be found in both languages, as, indeed, also in British Latin, given the importance of the religious community in the change. We are dealing here with a social phenomenon that transcended linguistic boundaries.

A comparison between the linguistic situation in England during the second half of the thirteenth century and that obtaining in the south of France during the same period provides an enlightening contrast and a telling illustration of the interaction between politics, society and language. In England an imported second language was adopted as the universally intelligible medium of the literate, hence dominant, classes, to serve as a unifying factor in a country linguistically divided by deep dialectal variations: this adopted language would profoundly transform not only the appearance but the very character of English for centuries to come. In France, on the other hand, a language deeply rooted in the soil of the south, every bit as old as the northern langue d’oil and forming the basis of a culture in no way inferior to that of the north, was to be systematically attacked and downgraded for centuries in the name of national unity. It is perhaps ironic that in our own day, some seven hundred years later, we are seeing the possible re-emergence of occitan – with the closely-related and similarly persecuted Catalan – as the working language of a new, powerful economic (and political?) axis – Toulouse, Marseille, Barcelona – within a European community in which English, with its strong Anglo-French historical component, will be the lingua franca.

This difference in historical circumstances betweeen France and England means that any failure by the FEW to take full account of the Provençal element in French would be far less damaging than the failure of English etymologists to take advantage of all the wealth of information now available not just in the MED but also, increasingly, in the AND. As more and more Anglo-French documents are brought to light from official collections such as the Public Record Office, the more important will it be to have an adequate dictionary for reference and the more necessary it ^p190 will become to include Anglo-French amongst the sources examined by English etymologists.

For the past hundred years or so we have had the unsatisfactory situation whereby historians of various specialities seeking first-hand knowledge of medieval England have been left to contend largely on their own with a wealth of very diverse records in later Anglo-French, and researchers in the history of English law have been obliged to wrestle unaided with the often opaque Law French of the medieval legal works that begin to appear in considerable numbers from the later years of the thirteenth century. Similarly, Middle English scholars often find themselves working in a trilingual field for which they are less than adequately prepared. All these groups have received little in the way of practical help from most of those whose speciality was the study of medieval French language and literature. The learned disquisitions on phonology and morphology found in historical French grammars will do little to assist the historian to cope with, say, the Rotuli Parliamentorum,¹⁰⁰ to smooth the path of the legal scholar engaged in publishing new volumes of the Year Books or of the Anglicist dealing with texts that have an Anglo-French dimension. If historical linguistics really is in retreat everywhere, as Dr. Burchfield states, it cannot be for lack of material on which to work, at least as far as the lexis of Anglo-French is concerned.

W. ROTHWELL

---