Lexical Borrowing and Code-Switching: The Case of archegay/hasegaye/harsegay in the Middle Ages and Later

This article presents a diachronic analysis of the word archegay, a weapons term that has its origins in early fourteenth-century French but first appears in written form in English around the end of the fifteenth century. I begin by considering the origins of the word, provide a new citation that antedates the earliest in the Oxford English Dictionary, and then examine evidence in French and Latin texts that show that the word’s referent was well known in England in the fourteenth century, even though no Middle English lexical item has yet been discovered. I discuss the implications that this has for our understanding of the currency of the term in the multilingual context of the later Middle Ages in England, then proceed to argue that later examples from the sixteenth century onwards (which appear to suggest that the word became accepted and even naturalised in English) are misleading, and that the word and its denotatum had by then become exotic and obscure. Finally, I demonstrate that the form harsegay, recorded in the nineteenth century, is a doublet borrowed from later French sources, and that it was frequently used by English writers with little comprehension of its meaning. 1. THE ORIGINS OF ARCHEGAY AS A LOANWORD IN ENGLISH Contact linguistics has demonstrated many of the difficulties to be found in the area of lexical borrowing. One issue that continues to be debated is the question of whether the nonce usage of a foreign word constitutes borrowing or merely code-switching. By extension, the question of when, or even whether, certain loanwords become naturalised in a language can be notoriously difficult to answer. At one extreme, a cluster of examples in a limited body of works must be considered inadequate proof that a loan has won general acceptance, even within a restricted or specialised group. At the other extreme, a larger number of attestations over several centuries need not necessarily indicate widespread adoption of a word, or even comprehension of its meaning in its adoptive language, as I shall argue in this paper. Furthermore, written evidence can be misleading, especially in a multilingual culture such as that of England in the Middle Ages, when (Middle) English, (Anglo-)French and (Anglo-)Latin comprised, in DOI 10.1515/ang-2012-0042 1 I am very grateful to Thom Richardson of the Royal Armouries, Leeds, for sharing information from his unpublished studies; also to John Kahn and Prof. Rajend Mesthrie for reading and commenting on early drafts of this paper. 2 See, for example, Sarah G. Thomason, Language Contact: An Introduction (Edinburgh: Edinburgh UP, 2003) 131–36. varying degrees, the common linguistic stock of most literate persons. In such a culture, it is conceivable that a word – or even a set of cognate words for the same concept – might have spoken currency in one or more languages without necessarily being written down in all of them. Likewise, a word originating in one of these languages might be used in either or both of the others without a clear sense that it belonged to the lexicon of any one of them specifically. In such circumstances, where code-switching might be a common occurrence in both spoken and written usage, many issues concerning borrowing and naturalisation become even more difficult to assess by the accepted criteria of most standard dictionaries. Many of the issues identified above are pertinent in various ways to the noun archegay, which is recorded as a rarity in English. According to the second edition of the Oxford English Dictionary (OED), this word, borrowed from French archegaie (< Spanish/Portuguese azagaya < Arabic alzaġāya) is first attested in English in 1523, in Lord Berners’ translation of Froissart’s Chroniques. Two citations from Berners are supplied, together with a single quotation from William Morris’s dramatic poem “Sir Peter Harpdon’s End” (1858), which itself is set in the world of Froissart’s Chroniques; that is, in Poitou in the late fourteenth century. In the OED the lemma is prefixed with a dagger, indicating that the word may be considered “Obs[olete] exc[ept] Hist[orical]”, and it is defined as ‘An ironpointed wooden dart; an assagai’. In the latest, updated online edition of the OED little has changed in this entry, except that a new citation has been added, from George Eliot Voyle’s Military Dictionary of 1876, though it should be noted that the word employed by Voyle has the divergent form harsegaye, for which the OED gives no explanation. I shall return to Voyle’s dictionary and harsegaye in the final section of this study. There are several points to make about English archegay and the OED’s treatment of it, but first a few observations are necessary concern3 See W. Rothwell, “The Trilingual England of Geoffrey Chaucer”, Studies in the Age of Chaucer 16 (1994): 45–67. 4 See Tony Hunt, “Code-Switching in Medical Texts”, Multilingualism in Later Medieval Britain, ed. David A. Trotter (Cambridge: Brewer, 2000) 131–47, at 131: “Linguists have frequently sought to identify borrowings in the languages of medieval Britain, but in the context of multilingual societies it can be unrealistic to attempt to distinguish code-switching from borrowing.” 5 The Oxford English Dictionary, 2nd ed., prepared by J.A. Simpson & E.S.C. Weiner (Oxford: Clarendon P, 1989), s.v. archegay. 6 This poem was published by William Morris in The Defence of Guenevere and other Poems (London: Bell & Daldy, 1858), a work not cited in the bibliography of the OED, 2nd ed. The poem itself mentions the death of Edward III (1377), and one of its main characters is Bertrand du Guesclin (d. 1380), so a putative date of 1378–79 may be posited for the period in which it is set. 7 See the online Oxford English Dictionary [accessed 28 December 2011], s.v. archegay n., available at: . LEXICAL BORROWING AND CODE-SWITCHING 265