“Who Says Manchester Says Cotton.”
Textile Terminology in the Oxford English Dictionary (1000-1960)
By Patrick Leech (Università di Bologna, Italy)
Abstract & Keywords
English: The study of terminology, which often aims at reinforcing specific denotation and precise relationships between words and things, entails a tendency towards synchrony, towards the collocation of terms within an overarching stable system. Historical analyses of terminology, on the other hand, can highlight the unsystematic nature of actual use, the chaos of conflicting and overlapping denotations from which subsequent synchronic systems are abstracted. This paper examines textile terminology in a specific field (textiles) using a historical dictionary (the Oxford English Dictionary) to highlight the problems raised by looking at terminology in a diachronic perspective. The study of the development of terminology in this field also points out some features of the processes of terminological innovation, in this case mostly through borrowings from other languages (in particular from French).
English: Nello studio della terminologia, che ha spesso lo scopo di rinforzare denotazioni specifiche e precise relazioni tra parole e cose, è implicita una tendenza alla sincronia, verso la collocazione di termini all’interno di un sistema fondamentalmente stabile. D’altra parte invece, analisi storiche della terminologia possono mettere in luce la natura non sistematica dell’uso reale, il caos di denotazioni in conflitto e sovrapposizione dai quali sono successivamente astratti i sistemi sincronici. Questo articolo prende in esame la terminologia tessile in un campo specifico (i tessili) usando un dizionario storico (l’Oxford English Dictionary) per mettere in luce i problemi dall’analisi terminologica in una prospettiva diacronica. Lo studio dello sviluppo della terminologia in questo campo mette in rilievo inoltre alcune caratteristiche dei processi di innovazione terminologica, in questo caso principalmente attraverso prestiti da altre lingue (in particolare dal francese).
Keywords: terminology, terminologia
©inTRAlinea & Patrick Leech (1999).
"“Who Says Manchester Says Cotton.”", inTRAlinea Vol. 2.
This article can be freely reproduced under Creative Commons License.
Stable URL: http://www.intralinea.org/archive/article/1610
The study of technical lexis, terminology, tends to focus on lexemes currently in use. This is in particular due to a bias in the discipline towards producing useful knowledge: descriptions of and prescriptions for the agreed uses of terms in specific and often technical fields. The aim of terminology studies is thus an applied one: to reinforce specific denotation and precise relationships between words and things. This aim entails a tendency towards synchrony, towards the analysis of the references of lexemes within a system conceived of as a stable chronological entity (see in particular Sager 1990: 39-40; Rey 1995: 23-47), and eschews, on the whole, the study of the historical development of terminological fields.
Historical linguistics shows a parallel blind spot. Studies in this discipline rarely focus on the historical development of language in a specific field. References and examples taken from specific fields abound (see, for example, Hughes 1988, ch. 2), but these are commonly placed within a wider framework than that of the historical development of lexis belonging to a particular field. Its scope, instead, is the larger one of the development of languages over time through formations from existing lexical items and grammatical and phonological shifts. This approach becomes problematic in particular in relation to the development of lexis in specific or technical fields, in which, given the need to create new terms for new objects or processes, neology often takes the form of borrowing from other languages rather than formation on the basis of existing lexemes (see Rey 1995: 63-84). The need to describe lexical innovation through contact with and contamination from other languages will become apparent when investigating a lexical field such as that of textiles, one which involves large numbers of objects crossing national boundaries as commodities.
The present study is located on the border between terminology and historical linguistics, and focuses on terminology in the specific field of textiles over a large time span. A full description of this field would involve a vast amount of detailed philological investigation; the research presented here, instead, provides an initial survey of the information on textile terminology available in the second edition of the Oxford English Dictionary (OED 1992 hereafter OED). The immediate aim of the research was to trace some aspects of the historical development of lexis in this field in order to provide a basis for further research in this field. A larger concern, however, was to see what happens to the study of terminology when pushed out beyond the haven of the stable synchronic system. The OED is a diachronic dictionary, and terms referring to textiles, it was found, vary enormously over time. A subsidiary focus was on the extent and modality of borrowings as a source of neology in technical fields such as this. The study was carried out on the basis of a corpus of terms taken both from historical dictionaries of textile terminology and from glossaries in standard historical works on English costume. The list of terms thus obtained was checked against the OED, and a data base was created to record the main information it gave on these terms (type of fabric, date of first and last illustrative quotations, etymology and examples of use in quotations).
2. Corpus and sources
The identification of a specific technical field in which to examine the development of terminology can be problematic. In particular, rather than starting from the definition of common formal characteristics as in the study of lexicography, the starting point for terminological analysis is the definition of common characteristics in the world outside the lexeme, in the referent of a term. The selection of a terminological field thus involves an act of inclusion (of terms belonging to set of referents established a priori) and a parallel one of exclusion (of terms which do not belong to this set). The criteria used to include and exclude terms from the corpus on which this survey is based, then, need to be outlined in some detail.
The present study has limited itself to terms denoting natural woven fabrics. This is, of course, only a part of the terminology relating to textiles. It excludes, for example, the immense range of terms referring to the transformation of raw materials (wool, linen, hemp, silk, cotton and so on) through spinning, reeling, drawing, weaving, dyeing and finishing. It is limited, in other words, to fabrics, not fabrication, to products not processes. It has also limited itself to natural fabrics, that is, it has excluded terms denoting fabrics made from the wide range of twentieth-century man-made fibres. Given the availability of such terms in contemporary trade dictionaries, the historical focus of the study would have been seriously compromised by the inclusion of such a large number of terms. It is also limited to finished woven fabrics, that is, it excludes terms which refer exclusively to yarn. A final exclusion is concerns terms denoting fabrics made of furs (although these are included in some historical dictionaries of textiles, see BECK, 1886). Such terms may be less interesting from a linguistic point of view as on the whole they are made up of homonyms of the animal from which the fur has been taken, used by extension to denote the fabric, as in the case of ermine or sable. Their exclusion is also the result, though, of the constraints of the definition adopted of woven fabrics, as furs are not, strictly speaking, fabrics which are constructed through a process of interlacing warp and weft on a loom. Terms relating to lace such as point net and orris have been excluded for the same reason.
While the constraints employed in choosing the terms for inclusion in the corpus were applied as strictly as possible, an eclectic approach was used as regards sources. The intention of this study was to provide an initial historical survey of textile terminology and it was thus decided to be as comprehensive as possible. The survey thus collected all terms referring to natural woven fabrics from a variety of sources: a series of textbooks on English costume (Cunnington and Cunnington 1952, 1954, 1955, 1957, 1959 and Mansfield and Cunnington 1973 respectively); a historical survey of mourning dress (Taylor 1983);  and two historical dictionaries of textile terms, an anonymous 1696 pamphlet entitled The Merchant’s Ware-House Laid Open, or the Plain Linnen Draper and S. William Beck’s The Draper’s Dictionary. A Manual of Textile Fabrics. Their History and Applications, published in 1886. The handbooks on English costume and the treatise on mourning dress are twentieth-century publications and need no introduction. The two historical dictionaries used, however, need a brief presentation.
Both can be described as trade dictionaries, one of their principal functions being to provide a terminological guide to those buying, selling and manufacturing in the trade. The first, indeed, The Merchant’s Ware-House, has few ambitions beyond this. Published at a moment when the domestic market was being flooded by a host of new fabrics, many of them cottons from the Indian subcontinent (see in particular Thomas, 1926) its declared intent in the subtitle was to show “how to buy all sorts of Linnen and Indian goods”. The aims that it set out in the “Epistle” to the reader at the beginning of the work were explicitly those of what might now be termed applied terminology. It addressed in particular the difficulty “for most people to know good linnen from bad” and aimed “to prevent all people from buying one thing for another” (Merchant’s Ware-House, 1696: 1). 
The Draper’s Dictionary of 1882 compiled by S. William Beck is a rather different case. On the one hand, like The Merchant’s Ware-House, it can be assumed, its author had strong links to the linen trade: the dictionary had first been published in instalments in the Warehousemen and Draper’s Trade Journal. Its declared intent, though, went beyond the purely utilitarian function of a terminological glossary, and espoused an explicitly philological approach. Its aim was “to reduce the chaos of particulars relating to the history of textile fabrics to an exact study” (Beck, 1882: “Preface”). The nineteenth-century corporative pride of those involved in the drapery trade is evident in the production of a volume of 377 closely-written pages abounding in a host of philological surveys of individual terms relating to textiles (many of which obsolete) based on a wealth of historical source materials. Any serious philological study of textile terminology must take Beck’s volume as its starting point.
383 terms referring to natural woven fabrics appearing in the sources mentioned above can be found in the second edition of the OED. To enable an elaboration of the information contained in the OED regarding these items, a data base was created to record the following information: term, definition, type of fabric, dates of illustrative quotations, status (that is, whether the term is recorded as obsolete or not), etymology, origin (in the case of words taken from or influenced by other languages), illustrative quotations given in the dictionary and variant spellings.
The extraction of this information from the OED was not a straightforward task. The OED is a vast archive of material, but it is not a data base. The principal function of the OED is to provide full and accurate information on the history of the use and development of words considered to be part of the English language. This often involves the sort of description that does not fit precisely into categories which need to be established a priori in a data base. It is necessary, therefore, to discuss in more detail some of the methodological criteria adopted in the compilation of the data base used for the present study.
3.1. OED headwords and combinations
Two criteria were used to determine inclusion in the data base. The first concerns the status of the entry in the OED, the second the semantic range of the term. Since the advent of the CD-ROM version of the second edition of the OED, it has been possible not only to search for terms under headword entries and under combinations (combinations of headwords with other units giving rise to different senses, listed at the end of headword entries) but also to search the entire text of the dictionary for any string of letters, whether these appear as headwords, combinations or simply in the texts of the definitions, notes or exemplifying quotations. Thus damasquitte, mentioned in the Draper’s Dictionary, can be located in the illustrative quotations of the headword damassin by means of a full-text search. Information on terms which can be found in the OED only through a full-text search, however (terms which, in other words, do not have the status of headwords or combinations), is partial and fragmentary and these terms have thus been excluded from the data base.
As noted above, the semantic requirement for inclusion was that a term should refer to a natural woven fabric. More precisely, in relation to the use of the OED, this meant either that the definition of the term in one of the senses listed under one of the headword or combination entries in the OED should refer directly to this meaning, or that the illustrative quotations clarified any ambiguity in the definition and made specific reference to natural woven fabrics. The term botany,for example, is defined in the OED under the word’s second sense, as “short for Botany wool”. But the quotations used to support this definition provide no evidence that the term was used to denote fabrics produced by the use of botany wool rather than simply a yarn that could be used, possibly in combination with other yarns, to make woven fabrics. Indeed, the quotations following the definition support this meaning. A quotation from an 1884 work entitled Spinning tells us that “when the material is very short Botany, the advantage is not so great. It is confined to commission combers and Botany spinners.”  There can be no certainty, then, that the term was used to denote a woven fabric rather than simply a yarn. In cases such as these, where no definition or quotation in the OED testifies unambiguously to a term’s referent as a fabric, the terms were excluded from the sample. The definition of the terms superfines, to give a contrary example, gives merely “goods of superfine quality” but its illustrative quotation makes direct reference to the use of the term to denote a fabric, and has thus been included.
3.2. Textile Types
The data base aimed to assign each term to a higher order of generalisation in terms of the type of yarn making up the fabric. For all but five of the terms, in fact (see section 4.2 below), where information on textile type was given in the definitions, the yarn making up the fabric could be assigned primarily to one of the following superordinates or combinations of them: wool,  linen, hemp, silk and cotton. For a large number of terms, however (73), no indication of textile type was given.
Classification of textile terminology into types was by no means an unproblematic operation. In a large number of cases, it was found that terms could be used, either contemporaneously or in different historical epochs, to denote fabrics made up of entirely different yarns. The definition given for the term bombasine, for example, runs as follows: “A twilled or corded dress-material, composed of silk and worsted; sometimes also of cotton and worsted, or of worsted alone.” Worsted being a type of wool yarn (as well as a fabric made from this yarn), the term could thus refer to mixtures of silk and wool, cotton and wool, or wool alone. For such cases, it was necessary to use a second “textile type” category, leaving the first yarn mentioned as the primary textile type (in this case, silk and wool), if no other information went against this interpretation, and indicating in the second category the other yarns that could be used. It became clear from the start, in fact, that the link between term and fabric could not be based entirely on the type of yarn from which it was made.
The illustrative quotations presented in the OED are in chronological order and aim to document the first attested literary use of a term in the English language. By inclusion of the dates of these first quotations in the data base, it has been possible to present a general survey of the dates and periods in which textile terminology entered the language, the results of which are presented below. (Full lists of terms can be found in Appendix A, ordered chronologically by their date of entry into English, and Appendix B, ordered alphabetically.) Some of the early quotations for some of the entries, however, appear in the OED between square brackets. The OED itself has left it somewhat unclear what it means by this. Its main intention, as stated in the “General Explanations”, is to indicate the “fully developed use” of a term. But it also aims to document a term’s “origin, its gradual separation from allied words or senses, or even by negative evidence, its non-existence at the given date”, and for these “subsidiary purposes” square brackets are used, although without explanation (“the exigencies of space render this impossible”) (OED 1989, xxix). For this data base, a selection has been made. When the quotations in brackets do not illustrate the prior development of the word but rather appear to document the use of the word itself, the dates of the bracketed quotation have been taken to indicate the earliest possible documented use of a term in the English language. A 1613 quotation giving evidence of the use of the term dongerijns, for example, has been accepted as an early dating for the term dungaree, whereas the Latin reference to “Casula de panno Tarsico” in a work of 1295 has not been considered a date attributable to the English term tars. Dates of quotations including the attributive use of a noun have been included, if earlier than substantive uses.
The dating of terms (rather than word-forms) is further complicated by the fact that the same headword may often have different if related meanings. The word-form crape, for example is listed as meaning both “a thin transparent gauze-like fabric, plain woven, without any twill, of highly twisted raw silk or other staple” (sense a), and “a sort of thin worsted stuff, of which the dress of the clergy is sometimes made” (sense b). In these cases, although the referents are two very different fabrics, the date of the earliest quotation given has been accepted as evidence of the first use of the word-form crape. It can be argued, of course, that this procedure runs contrary to the whole foundation of terminology studies, whose unit of analysis is not the formal unit (the word) but the term, understood as a formal unit with a conventionally stable reference in a given field (Rey 1995: 26). The justification for this apparently lax procedure is twofold. On the one hand, an accurate description and dating of the different referents of the same formal unit is a highly complicated philological task which goes beyond the scope of the present survey. But the task may in any case be impossible in an absolute sense. The case of crape, a term which refers both to silk and wool fabrics, is more typical than exceptional. The variety of different historical speech communities in which any given formal unit was used, and the myriad different ways in which these communities may have used these units militates against any strict application of the classificatory systems used in terminology studies (see, for example, Sager 1990), at least in historical descriptions of a given field.
Immediately after the information on the form of the headword, the OED provides, between square brackets, information regarding the etymology of the word. Included here is information about the derivation, the “subsequent form-history” of the term in English as well as “miscellaneous facts as to the history of the word, its age, obsolescence, revival, refashioning, change of pronunciation, confusion with other words” (OED 1989, xxvii). Some headword entries have no etymological annotation, others have just a few words, others run to several lines. Particularly complicated entries may be even longer: the entry for church, for example, runs to some 15 column inches (Berg 1991: 19). Given the interest of the present study in linguistic borrowings, particular attention has been paid to the derivations noted in this section. The data base classified the derivations of all entries according to the following categories: adoptions, adaptations, formations, corruptions, foreign, Old English, eponyms, toponyms, and “not given”. Some explanation of the criteria used for the application of these categories is necessary.
The OED indicates clearly when a word has been “adopted” or “adapted” by the immediate use of the abbreviation “a.” or “ad.” respectively, defining adoption as essentially a “popular process, at work whenever the speakers of one language come into contact with the speakers of another, from whom they acquire foreign things, or foreign ideas, with their foreign names”, and adaptation as “a learned or literary process” involving the conscious importing of a word from a foreign language and its adaptation to the structural constraints of the English language (Berg 1991: xxvii). These two categories, unlike the others employed, have only been used in the presence of these explicit indications in the etymological note. Thus terms such as caddis for which there is no explicit indication of adoption or adaptation, despite an etymological note strongly suggesting a French origin,  were classified as “not given” as regards their origin.
In most cases, formations are marked with the abbreviation “f.” These indicate the “combination of existing words or parts of words with each other” (Berg 1991: xxviii). Not all cases of formation are marked in this way, however. In cases in which formation is transparent, such as bath coating, these have been considered as such even when there is no explicit indication. In other cases, when the derivation of terms as formations is less clear (in the case, for example, of cogware, a woollen fabric), the etymology was recorded as “unknown”. The indication “short for”, on the other hand, was classified as a “formation” (for example, in the case of merv, short for merveilleux).
In many cases, headword entries of foreign words used in English are indicated with a double straight slash (||) before the headword entry itself. Terms prefaced with this indication have all been classified etymologically as “foreign”. There are a number of occasions, however, when this typographical indication is absent, although the etymological note begins its explanations by giving an indication that the word is a foreign one. The term charmeuse, for instance, is given the following brief explanation: “Fr., fem. of charmeur, agent-n. of charmer to charm.” In the absence of an explicit indication as to whether there has been adoption or adaptation, such terms have been categorised as “foreign”.
“Old English” has been used as a category when the term has no clear foreign derivation, but has developed autochtonously from an Old English or Middle English origin.
The notes sometimes indicate that the derivation of a term is a “corruption”. This category has also been used on other rare occasions when it has seemed appropriate, for example in the case of figuretto, which is indicated as “? error for It. figurato figured (stuff)”.
There are a few occasions on which the etymological explanation of a word shows a clear derivation from the name of a person. Although the OED has eschewed the category of eponym, this seemed to be a worthwhile one to employ in terms of charting the probably entry of some words denoting textiles into the English language.
There is no standard indication, either, of derivation from toponym. The derivation of a term from the name of a place with which it was originally associated, however, is of particular interest to the student of textile terminology. Some of the most common textile terms such as denim, and jean have their origin in place names (in Nîmes and Genoa respectively). It should be remembered, however, that this category cuts across the others, and in particular those of adoption and adaptation. The term denim, for example, is clearly etymologically a toponym which, in all probability entered the English language as an adoption of the native French term to indicate a product associated with this city (“de Nîmes”). Etymologically, rightly speaking the term is both an adoption and a toponym. The distinction to be made between the derivation of a term from a geographical locality and the entry of the term into English from another language is clearer and more important when the geographical locality and the language spoken in this locality differ. This is the case, for example, with sicilienne (a fine fabric made of mixture of silk and wool) a term whose geographical referent is Italian (Sicily) but whose route into the English language clearly passed through French. The data base thus noted both the language of origin of terms which derived from foreign sources and, in the case of toponyms, the country or area to which the toponym refers. It was not possible to do this in an entirely consistent manner. Sometimes the dictionary gives a clear indication of the geographical location of the place associated with the textile, but no mention of the language used to refer this place. On occasions, intuition has dictated that a toponym came into English through an adoption or adaptation of the name for the place involved in the language spoken in that place. This is the case with terms such as barège and valencia, terms denoting different types of fabric which surely derive from the French and Spanish names for these cities respectively. On others, a degree of circumspection has been maintained. Thus a term such as Donegal, which clearly refers to an area of Ireland but whose linguistic origin (English or Gaelic?) is not specified, and similar cases involving terms for Indian cloth such as surat or cashmere, have been classed as “unknown” in terms of their linguistic origin.  Finally, given the importance of geographical as well as linguistic origins for textile terminology (neology may be the result of the need to describe new objects, in this case new objects imported from abroad), the category of “toponym” has been stretched to its limits to include not only toponyms in the strict sense of words which also refer to places, but also to terms which, through their modification, give a clear indication of geographical origin. Thus the term Italian cloth, which refers to a specific type of fabric, has been included in this category.
The variety of ways in which etymology is described in the notes to the headwords, works against watertight consistency in the compiling of a data base. There is often some indication of uncertainty on the part of the OED, as when, for example, it gives the unnerving abbreviations “app.” (= apparently), “perh.” (= perhaps) or simply “?”. The categorisation of the etymology of terms described in these ways has necessarily been somewhat subjective on occasions. Whenever the notes have given more than one possible etymology, however, the categorization has erred on the side of caution and chosen the class “unknown”.
A more serious problem relates to the fact that the etymological notes refer to the headword only and not its variant successive senses. Thus the etymological note refers to the way in which the word form entered the language. But later senses of words may have entered the language in a variety of ways, and these may bear little or no relation to the way in which the original form did. Thus word-form imperial may have entered the English language through French, as the OED states, but this has little to do with the extension of its meaning to denote a woven fabric. Similarly, eider-down, as a formal unit, is “ultimately” adopted from Icelandic. But sense (3), which gives “a heavily napped wool or cotton or man-made fabric of thick texture” is a metaphorical extension of a term used to describe something very different, made with the feathers of the eider duck. Etymology, in short, describes the origin of forms, but not their subsequent semantic wanderings. It tells us little, in other words, about the variety of ways in which semantic neology develops the sense of word forms which already exist in the language. 
The data base also included an indication of “origin”, understood as the language of origin of a term which was adopted, adapted, foreign, formed by the combination of foreign lexical units or referring clearly to a foreign person (in the case of eponyms). When no clear indication of a language of origin appears in the etymological note, the origin was noted as “unknown”. This is the case even with terms whose geographical origin is mentioned clearly in the definition (for example, moory, which is described as “a kind of Indian cloth”) but for which no linguistic origin is mentioned.
Many of the terms recorded are obsolete, or used now only in a historical sense. The data base included a specification as to status (obsolete or otherwise).
The compilation of a data base of the information available in the OED on 383 terms relating to natural woven fabrics made possible a series of calculations and generalisations regarding dating, obsolescence, textile type, etymology, and the languages of derivation. It should be emphasised that the conclusions that can be drawn from this set of data in terms of the history of textile terminology in English are provisional. The data refer only to information available in the OED, and no attempt has been made to compare these findings with the numerous other sources available on the history of textiles and on terminology used to describe individual fabrics. Moreover, a large number of the terms to be found in the historical dictionaries do not appear in the OED. Further research into the history of these terms, with appropriate illustrative quotations documenting and describing historical use, could fill what would appear to be a significant gap in the coverage of the dictionary.
4.1. Dating and obsolescence
Table 1 presents the datings of the terms in the corpus, according to the first quotation provided, in fifty-year groups. As may be expected, the take-off for the introduction of new terms referring to textiles, at least as illustrated in the OED, occurs only after 1350. The period up to 1350, in fact, saw the introduction of only 27 (7%) of the total terms in the corpus. The real increase in new terminology, however, only takes place in the 150-year period 1551-1700, when 115 (30%) of the terms in our corpus were introduced,  compared with just 50 (13%) for the next hundred years. The large number of terms that appear for the nineteenth century (100 - 26%), might reflect the bias of the original OED compilers towards terms in use in their own century. The paucity of terms for the twentieth century reflects both this and the historical nature of the corpus.
Table 1 also shows the number of words later to become obsolete entering the English language in a given period. In the period up to 1700, of the 222 terms which had appeared in the language, 101 (45.5%) would later become obsolete. These figures would suggest that right up until 1700, a textile term had only just over a 1 in 2 chance of survival. A total of 113 (29.5%) terms in the corpus were indicated as obsolete. Of these, as appears from Table 2, 96 (84.9%) disappeared in the period after 1600. These obsolete terms include some which indicated a particular type of fabric in the past but which now have only different meanings. Examples are blanket, rug, scarlet,  tinsel, vermilion, wad.
The remaining 270 terms, according to the OED, are still in use. But we may note at this point that only 153 terms (39.9%) out of the total 383 are included in the latest, authoritative dictionary of Textile Terms and Definitions (Textile Institute 1995). The corpus would appear to include, then, 117 terms (29.8%) which are still in use in the English language but which no longer function as descriptive terms within the current field of textiles. We may hypothesise that these units continue to exist as lexemes (rather than terms) which may still denote textiles for non-specialist users or which have acquired wider metaphorical uses or different uses altogether.
4.2. Textile types
The 383 terms were classified in the data base according to the primary or original textile type mentioned in the definitions in the dictionary, based on the five superordinate yarn categories of wool, linen, hemp, silk and cotton.  As has already been pointed out, many of the terms have a secondary (or later) textile type, and many defy accurate classification simply in terms of the yarn from which they are made. Table 3 shows the results of this classification.  As might be expected from a data base of English textile terms, those relating to wool predominate (108 - 28.2%). Given the popular association of English textile manufacture with the cottons of the Industrial Revolution, it may come as a surprise that the number of terms relating to cotton is only slightly higher than those relating to linen, 37 (9.6%) and 32 (8.3%) respectively, and considerable lower than those relating to silk (87 - 22.7%). Cotton production, of course, became important in the north west of England only in the course of the eighteenth century, and even the expansion of cotton imports from India towards the end of the seventeenth century was a relatively late development compared with the linen industry, which had developed to a considerable extent through contacts with Flanders from the late middle ages. Silk manufacture in England also has a longer history than cotton, flourishing as early as the first half of the seventeenth century (Warner 1921). The number of terms denoting fabrics made up of a mixture of wool and silk (16) may also surprise. On a technical level, this may perhaps be explained by the use of silk, the stronger yarn, as warp on which to weave the weaker but warmer woollen weft (see Rothstein 1990: 18).
A detailed discussion of the textile types referred to by these terms with regard to their linguistic origin (and toponymic referent) will be left to later. This may be the place, however, to present briefly the etymology of some of those terms in the data base which are still of in everyday use.
To take those referring to wool first, these include the following well-known terms: felt, stuff, worsted, broadcloth, serge, baize, denim, plaid, tartan, flannel and cashmere. The term felt, is one of the oldest in the data base, with an illustrative quotation in Latin that mentions the use of the word in English, and its derivation from the Latin “filtrum” meaning “to filter”. Stuff, on the other hand, as a fabric, would appear to be a relatively late development of the seventeenth century. The fifth of the meanings given notes it specifically as a woollen fabric, and a 1735 quotation describes it as “any Sort of Commodity made of Woollen Thread, &c. but in a particular Manner those thin light ones that Women make or line their Gowns of or with.” Worsted refers to a wollen fabric made from a wool that is combed rather than carded, and derives from the Norfolk village of Worstead where these processes were developed (although there is some discussion as to the precedence of the fabric or the place-name).  Broadcloth was the traditional west-country quality wool product so named because of the specification that it had to be two yards wide, although the OED definition notes that the term is “now used to imply quality rather than width.” All these terms can be considered to have their origins in Old or Middle English. Serge, instead, is a term that was adopted from Old French, and is used to denote a variety of different types of wool. We may note that the origin of the term is in the Latin serica (=silk) and thus that there has been a shift in the meaning of the term in its passage from Latin to French. Baize also comes from the French, and denotes a coarser type of wool. Denim, as we have seen, is a shortened form of serge de Nîmes, and thus originally referred to a woollen fabric, although it has now, through American usage, come to mean the “coloured twilled cotton material” famous throughout the world. The term plaid is unique in the data base as a word derived from Gaelic, meaning blanket, but by extension also the woollen cloth from which both the blankets and the traditional Scottish Highland costume were made. Tartan, used to describe the cloth of woven crossed stripes as well as the pattern, is of uncertain etymology, the OED mentioning a possible derivation from the French tiretaine (a kind of cloth, half wool, half linen or cotton) and another from the terms tartar or tartarin (also types of cloth) indicating an association with Tartary. A long etymological note on the term flannel discusses its possible origins in French, but also indicates that it may have a Welsh origin, given the fact that it was a well-known production in Wales as far back as the sixteenth century.
The terms relating to linen that are immediately familiar are cambric, diaper and buckram. The first of these reflects a clear origin in Flanders, being a development from the name of the town Cambray. Diaper, on the other hand, according to the OED, is an ancient term to be found in Latin, Greek and medieval French to indicate white. In this, the OED goes against the hypothesis, put forward tentatively in The Draper’s Dictionary, that the term derives instead from Ypres (or from the French “d’Ypres”) which would give it a toponymic status similar to that of cambric (Drapers’ Dictionary, 1882: “Diaper”). We may note in passing that the buckram in which the “four rogues” mentioned by Falstaff in Henry IV Part I were clothed, was a linen fabric also spelt bokeram.
Well-known terms referring to hemp include canvas, from the Middle English canevas, related to the Latin term for hemp (cannabis), hessian, from the area of Germany, Hesse, and sackcloth, clearly a formation based on the use of the coarse hemp cloth for making bags and sacks.
With the exception of velvet (adapted from the medieval Latin velvetum), all the best known terms denoting silk derive from French. Thus gauze is adopted from the French gaze, “of uncertain origin” (although the OED mentions an unsubstantiated hypothesis that it derives from the name of the city of Gaza in Palestine); grogram is adapted from the French gros grain (large or coarse grain); damask is probably derived from the Anglo-French Damasc, the Syrian city; chiffon is directly taken from the French chiffe (= rag); crape is the anglicized spelling of the modern French crêpe, found in English in a number of phrasal uses such as crepe de chine and crepe lisse; satin came to English via the French homonym (the OED denies any etymological link with the apparently synonymous Arabic zaitun, supposed to be derived in turn from Zaitun, the name of a city in China); and taffeta derives from the old French taffetas, or taphetas (“ultimately” derived from Persian taftah, indicating either silk cloth or linen clothing).
The most common terms for cotton cloth all betray the oriental origin of these fabrics. Two of the terms come to English via French, but both can be traced further east. Gingham, the common term for striped cottons, comes from the French guingan for the same fabric, but can be further traced to a Malay word meaning striped. Muslin derives from the French mousseline, the French name for the Iraqi town of Mosul where the fabric was originally made. Chintz and dungaree (formerly a cotton fabric before being extended to refer, in the plural, to the item of clothing) both come from Hindi words for Indian cotton cloth. The most extensive term for Indian cloth, of course, is calico, from the name of the city on the Malabar coast of India, Calicut, although it is “not clear” precisely how this term came into the English language. Nankeen, a common cotton cloth in the eighteenth century, derives directly from the name of the Chinese city.
The only term indicating a fabric made up of a mixture of yarns and of common use today, perhaps, is fustian (categorised in this data base as a mixed linen/cotton fabric, as it only later became made exclusively of cotton). This term derives from the Old French fustaigne, which is in turn “conjecturally” traced back to the name for a suburb of Cairo, Fostat, where it was supposed to have been originally made.
Textile terms, however, as has been pointed out, are far from stable in their denotation of textile types. Many terms remain in the English language unaltered in their forms but shifting considerably in terms of their denotation. 71 (18.8%) terms were classified as “unknown” with regard to their primary type, and in a number of cases this was due to the fact that the definition gives a number of different yarns from which these fabrics could be made. Drab, for example, which only later came to denote the colour (and later still to indicate dullness) is defined simply as “a kind of cloth”. 70 other terms, in fact, were noted as referring, either contemporaneously or in a different period, to fabrics made up of textile types different from the principle one recorded (see Table 4). Of these 70, 18 are even more problematic, having what can only be described as multiple referents in terms of textile type (Table 5).
Table 6 shows the distribution of terms according to textile type by the date of entry into the English language. A relatively large number of terms referring to wool, as can be imagined, entered the language early: 19 (17.4%) by 1400. We may note, however, that the nineteenth century saw the appearance of a further 33 terms (30.3%) referring to wool. The golden age of silk terminology, instead, would appear to be the seventeenth century, which saw the appearance of 23 (25.8%) of the total 89 terms referring to silk. Silk also enlarged its terminology in the later nineteenth century, with a further 16 terms (18%) in the period 1851-1900.
Table 7 illustrates the different derivations that the OED gives for the terms in this data base. It should be repeated here that the information given in etymological notes, when they are present, varies enormously and that the categories used here are not exclusive. The 100 toponyms, in particular, in the case of those which refer to geographical localities outside of Great Britain, are clearly adoptions, adaptations or foreign terms in their own right. Similarly a number of formations are formations from foreign forms. Eponyms, too, often refer to individuals who were not British. The status of proper nouns as “belonging” to one or other language is in itself, of course, somewhat debatable.
We may note, first of all, the high number of terms for which no etymology is given (67). The other significant categories are toponyms (100), adoptions (77) and formations (53). It is immediately evident, then, that adoptions, irregular borrowings based on non-formalised contact between speakers of different languages “from whom they acquire foreign things, or foreign ideas, with their foreign names” (Berg 1991: 19), is much more prominent than those acquired by the more formal process of adaptations (only 18 in this sample) or than those that continue in some way to be considered as foreign terms, although used within the English language (29 examples). The question of the origins of these imported terms will be dealt with later. Here we will take the opportunity to discuss briefly some of the other ways in which terms entered the language, namely through corruptions, eponyms and other non-classifiable means.
The category of corruption, as mentioned earlier, has been used to included variants on main forms, alterations and etymologies noted as possible errors.  10 terms in the corpus were classified as deriving from a process of corruption. Thus figuretto (an unidentifiable fabric called so because of its flowery pattern) is given as a possible error for the Italian figurato, meaning figured material. Lustring is an alteration of the French lustrine, a glossy silk fabric introduced into England in the late seventeenth century. Paduasoy, is not, it would appear, a toponym referring to silk originating in Padua, but a corruption of the French poult-de-soie (“a fine corded silk”). Sateen is simply a variant on satin, as shagreen is of chagrin (although the former alone refers to a silk fabric), whereas seersucker is more complicated: an “East Indian corruption” (although precisely what this means is not clear) of the Persian sh’r o shakkar, meaning milk and water, giving its name to a striped material. Drilling is given as a corruption of the German drillich, meaning threefold; grazet is a possible corruption of the French grisette, “a cheap woollen stuff of grey colour”; and mockado is a corruption of the Italian mocajardo, the nearest equivalent to the Arabic term mu’ayyar, meaning mohair, or cloth of goats’ hair. Tweed, often thought to be associated with the River Tweed, is instead an “accidental misreading” of tweel, the Scottish form of twill.
Eponyms, the naming of words after individuals, have a particular interest for the social history of the development of textile terminology. This research, however, has so far only identified eight such cases. Challis might appear to be an adoption from the French, but the OED attests, although without evidence, that it is apparently “of English origin, and not improbably from the surname Challis.” Coburg, rather than being a toponym, is more likely to have been used to denote a type of worsted because of an association with Albert of Saxe-Coburg, the consort of Queen Victoria. The OED quotes The Draper’s Dictionary here in an interesting case of the use of an eponym for the new promotion of what was essentially an old product: Coburg was “... introduced shortly after her Majesty’s marriage with Prince Albert of Saxe-Coburg; most probably aiming at popularity through that event. It was merely a modification of what had previously been known as Paramatta cloth.” Doily was a textile named after a eighteenth-century linen draper of the same name who practised in the Strand and who, according to a 1712 quotation, “raised a Fortune by finding out Materials for such Stuffs as might at once be cheap and genteel.” Georgette was the name of a silk introduced in the early twentieth century which took its name after a French stylist, Mme Georgette de la Plante. Petersham is described as taking its name from Viscount Petersham in 1812 (although the same is supposed to hold true for the term’s other senses, that is, as an overcoat or a hat). Victoria crape, as has already been mentioned, takes its name from Queen Victoria. Zanella, finally, for which the OED only gives two quotations, dated 1876 and 1882, is indicated as possibly deriving from an Italian authority on wool, Antonio Zanelli, author of Le lane italiane (1878).
The category “other” has been used to indicate words whose etymological note gives no description of how these terms arrived into the English language. This is the case for plaid, derived in some way from Gaelic, and sameron, derived from Yorkshire dialect. Galatea, a striped blue and white cotton material took its name from the ship, H.M.S. Galatea, whose sailors were dressed in it, and satinisco is “pseudo-Spanish” for satin, on the model of Morisco.
4.4. Languages of origin
Table 8 and Table 9 show respectively the languages of origin of the terms in the corpus and the breakdown of each language by textile type. The preponderance of terms deriving from French (135 - 35.2%) is immediately evident, even in relation to “native” English terms (103 - 26.9%). The dominance of the French language in this field is all the more striking if we compare it to the next foreign language in terms of influence, Italian, with only 11 examples. Altogether, other foreign languages contribute only 66 terms (17.2%), less than half compared to those deriving from French alone. The number of terms for which the OED gives no origin, or else a contested one, is also high (79 - 20.6%).
Reading Table 9 horizontally, by language, we may note that loans from French were used in particular with regard to silk. These account, in fact, for 53 (39.3%) of the total 135 terms borrowed from French, with terms referring to wool products reaching only 30 (21.6%). Terms deriving from English were used for woollen fabrics in particular (40 - 38.8%), considerably more than for cotton (12 - 11.6%) and in particular silk (5 - 4.8%).
Reading the same table vertically by textile type, terms denoting wool were above all of English origin (40 out of 108, 37%), but a significant contribution was made by French (30 - 29.1%). Again, however, it is the case of silk that stands out. The 53 terms denoting silk (out of the total 87) deriving from French accounted for 60.9% of the total, the five examples of the nearest competitor, English, only being equivalent to 5.7% of the total. French dominated, too, in the category of wool-silk mixtures, with 8 out of the 17 terms.
Italian (along with Urdu) contributed the next highest number of terms (3) relating to silk (mantua, genoa and ferret), but also terms denoting cotton (dimity), a hemp-cotton mix (bergamot), two terms denoting wool (perpetuana, mockado) and a wool-cotton (zanella). Terms deriving from Latin were above all toponyms whose name in English was classified in this sense (silesia, a linen; saxony, a wool; and the unidentified venetian). Asbestos cloth is also derived from Latin, as is the descriptive sempiternum, a wool probably similar to the English version, everlasting. The term velvet, as already mentioned, is of Latin origin (velvetum). Spanish terms were used in particular for wool (vicuna, merino, alpaca), but also for silk (satinisco, frizado) and cotton (barragan). Linen fabrics accounted for over half of the terms imported from German (ticklenburg, osnaburg, garlits, drilling), the other term identifiable by type referring to hemp (hessian). Hindi terms denoted above all cottons (mulmull, dungaree, chintz, dorea), but also hemp (gunny) and silk (tussore). Three out of the four terms deriving from Dutch denoted linen (holland, gulix, duck), the other to a silk (lukes). Terms deriving from Urdu denoted silk (roma, lungi, culgee) and cotton (nainsook). Of the three words deriving from Chinese, two denoted silk (canton, shantung) and one cotton (nankeen). Flemish terms were used in English to denote a linen (cambric), a silk (dormick) and a well-known woollen fabric (duffle).
Other languages exporting textile terms into the English language are Persian, with the woollen fabric cheney (the Persian word for China), and the linen (later cotton) seersucker; Greek with antherine (a wool-silk mix); Arabic with mohair (a wool made from goat’s hair); Turkish with elatcha (silk); Icelandish with eider-down (wool); Gaelic with plaid (wool); Tamil with the unidentified pullicate; and Old Norse with wadmal (a coarse woollen cloth). Abaca is the “native name” for the hemp produced in Manila.
Table 10 gives a breakdown of the exactly 100 terms classified as toponyms by their geographical referent. (Appendix C, gives a complete presentation of these terms with reference to their specific geographical referent.) There is an immediately clear relation, as we might expect, between these and the principal areas from which England imported textile goods: France, the Middle East, Italy, India, Germany, China, Flanders, Holland and Spain. The largest number of toponyms, however, are English, referring to the wool centres of the west of England in particular. The preponderance of toponyms referring to silk textiles from French, Middle Eastern and Chinese locations may also not come as a surprise.
A full description of these toponyms would necessitate a detailed discussion also of the history of the British textile trade, far beyond the scope of this paper. Here, instead, attention may be brought to the gap between the geographical referents of some of the toponyms and the language through which, according to OED, these terms entered the English language. In other words, when a toponym was used to denote a textile fabric, these were often not the native names for the places but instead the version of another language, often English or French, of these names.
All 22 of the toponyms referring to English places, not surprisingly, are taken from English. The equivalent is true, also, of the 17 toponyms referring to French localities. The picture changes completely when we look at the toponyms which refer to places in the Middle East. Eight of these 12 terms entered English through French: turkin, tartarin, tars, tabby (a suburb of Bagdad), muslin (Mosul), levantine, damassin and damask (Damascus), illustrating, perhaps, the extent to which English contacts with the world of the Ottoman empire and the Middle East was heavily mediated culturally by French institutions (see Fukasawa 1987). Interestingly, French versions of place-names also dominate the 12 terms relating to Italian places, 6 of which come into English through French: sicilienne, pleasance (Piacenza), gros de Naples, fustian anapes (Naples), florence and florentine. Only three come directly from Italian: mantua, genoa and bergamot. Jean, according to the OED, derives from the Middle English name for Genoa, “Gene, Jene, Jeyne, Jayne, Jane.” Unfortunately, no indication of language origin is given for eight out of the 11 toponyms that refer to Indian places (calico, cashmere, cashmerette, cassimir, satara, surah, surat, bengals). Pullicate comes from the Tamil name for the town Pulicat. Bengaline is classified as a formation from the French. Kerseymere is an example of a toponym (referring to Kashmir) whose origin is a corruption due to a mistaken association with the English term kersey. Five out of the seven German toponyms come directly from German (brunswick, garlits, hessian, osnaburg, ticklenburg), and two from Latin geographical terms (silesia, saxony). Of the six terms which refer to places in Flanders, only three come from Flemish (cambric, dormick, duffle). Bruges in English is in all probability derived from the French name for the Dutch city Brugge, while the term lyre is derived from the Latin name Lyra for the city now known as Lière, in Brabant. Lukes is adopted from the Dutch name for the city now known as Liège. Crêpe de chine is clearly a French term used in English to refer to this type of Chinese silk, whereas pekin, was adopted from the name that French Jesuits gave to the city. Cheney, as we have seen, came into English via the Persian name for China. Nankeen, shantung and canton all derive from the Chinese names for these cities. Holland is the general term for the linen coming from this country, whereas gulix derives from the town now known as Juliers. Of the terms which refer to other places not covered in this categorisation, jap silk is clearly an English formation, marocain comes from French, and no indication is given for the linguistic origin of the Irish area of Donegal. Paramatta, a wool-silk or wool-cotton mix, is associated with the Austrialian town of this name, whose linguistic origin is not given.
Alain Rey’s work has stressed that the study of terminology is necessarily interdisciplinary, given that it entails the description and analysis of a system of terms relating to specific fields (Rey 1995: 23-47).. Terms are not names, functioning in isolation, but units which cut up the material world of the field in question according to more or less strict criteria. The study of terminology aims to draw demarcations between referents of different terms, where the reference of one term ends (or “terminates”) and another begins. In other words, it is constantly occupied with trying to solder the signifier to its signified. As such, terminology studies necessarily has to move outside the discipline of linguistics and mix with others such as cognitive psychology, the sociology of knowledge, and in this case, history.
The diachronic approach to the study of terminology in a specific field, however, raises, in a particularly striking way, a problem that is at the heart of the discipline. Specifically, it problematises the notion of system that is at its basis. In a diachronic study such as this, change is the dominant feature. This diachronic survey of textile terms in the OED shows that there can be a variety of different referents for individual terms. The notion of a closed system, which is of necessity synchronic, fits the data uneasily if at all. It may be possible to map out systems of textile terminology in given places and at given moments. It may be that something approximating to a terminological system could be charted for, say, woollen products in England in the sixteenth century. The wide chronological scope of the present survey, however, only enables us to glimpse this possibility. Such systems, moreover, are likely to differ themselves in structure. The terminological system conventionally agreed amongst wool merchants in pre-industrial Britain is likely to be considerably more heterogeneous (and “unsystematic”) than one relating to cotton products during the industrial revolution. The problem of the nature of terminological systems, in other words, raises the parallel historical problem of standards and standardisation within the textile trade (Styles 1983: 533-534) and the larger problem of changes in classification systems (Foucault 1973).
Historical approaches to terminology such as the present one, then, have little relation to synchronic terminological descriptions. Rather than a term’s demarcation within a relatively stable system, such approaches instead raise two related philological questions. One concerns the temporal scope of terms. If the term serge, for example, had a more or less agreed referent in England and France over a wide temporal span (from the fourteenth century to the present), what term was used to describe its nearest equivalent before this date? If gingham is a common term for striped cottons, when and how did it replace or accompany other terms with similar referents such as bengals? Another relates to the parallel issue of geographical scope. If we can assume that serge and gingham were common and conventional terms in England and France, to what extent were they current use also in the German-speaking, Slavic and Middle Eastern textile trades?
This leads us to another feature of this study of textile terminology which emerges clearly: the international nature of communication within this trade. From the links between south-eastern England and Dutch linen manufacture in the Middle Ages, to the influence of French Huguenots on silk manufacture in London in the seventeenth and eighteenth centuries, and the import of raw materials first from the Middle East and then from India and China, the English textile trade was an international affair, and this finds unambiguous reflection in its terminology.  Even in the wool trade, England’s best established textile manufacture, raw materials were imported from Spain and wool products were exported throughout Europe and the Middle East (Mann 1971). In the eighteenth century, a Norwich manufacturer of worsted damasks received visitors from Cadiz, Venice, Leipzig, Weimar, Zurich, Frankfurt, Cologne, Bremen, Lubeck, Copenhagen, Oslo and Stockholm all in one year (Fawcett 1985: 154). To trace the international validity of any of the terms involves a trail leading both through a daunting number of languages and into the dense, although not impenetrable, forest of textile history.
The “Norwich manufacturer of worsted damasks” points us to another problem concerning our material: the relation between toponymic terms and the actual geographical origin of fabrics. The Norwich manufacturer in question was well-known for producing a type of material whose name indicates two places, one near to Norwich (Worstead) the other far away (Damascus). Toponyms used in this sense, as a moment’s reflection confirms, only rarely indicate that a given product comes directly from the geographical place from which it takes its name. A Weiner Schnitzel is a veal cutlet which takes its name from a recipe probably originating in Vienna; the cutlet itself is nearly always of entirely different provenance. The toponymical referent, in other words, indicates only an indirect link, which in textile terminology can signify style, raw materials, pattern, weaving methods and so on. Thus German serges were manufactured in Wiltshire and cassimeres (later kerseymeres) in Bradford (Mann 1971: 6, 50-51) The analysis of toponymic referents of terms, rather than indicating the movement of objects, tells us about the movement of styles, skills and symbols.
It is clear from these examples that the problem here is one of neology, the process by which new words are introduced into a language,  in this case, through the specific medium of a specialised field. In some cases, neology in textile terminology takes place through regular grammatical extension, through affixation on previous lexemes. This would appear to be the case with terms such as cashmerette (from cashmere) or velveteen (from velvet). But such forms are not numerous. More frequent are cases of formation through combination of existing lexemes. Examples of these abound, from victoria crape and plain-backs to bath-coating and sackcloth. But a far greater number, it emerges, are borrowed terms from other languages, often in the form of topoynms. We may ask ourselves, at this point, why this should be so, and we may conclude by putting forward two related hypotheses.
The first is linked to the fact that neology is indissolubly tied to the emergence of new realities, in this case new textile products. The case of coburg, a new term for the old product known as paramatta is probably an exception, although an illuminating one (see above, section 4.3). New products, however, rarely arrive on the international market entirely unbaptised. We may assume that most of the new imports into Britain came already labelled in some way amongst the textile manufacturers and traders of the country of origin. This is surely the case with a large number of the terms that derive from languages other than English. Thus the French product termed gros grain was imported into England along with its name, which became corrupted into grogram. In short, new products rarely become public property without already having acquired a name of some sort, and the process by which a new product earns itself a standard or conventionally recognised name is not an abstract process of naming but a choice made within a highly restricted field of already existing possibilities.
Second, given this restricted field, it is likely that an already existing foreign term has a greater chance of being adopted than a new native one. Goods are pregnant with all sorts of social meaning (Douglas and Isherwood 1980) and textiles in particular are important signifiers of fashion, itself an indicator of a series of related signifiers linked to urban or metropolitan life-styles, access to information, exoticism and so on. In this context, in particular, the terminology of English textiles faithfully reflects both the cultural hegemony of France and the exoticism of empire.
Beck, S. William (1882). The Draper’s Dictionary, London,Warehousemen and Draper’s Corporation.
Berg, Donna Lee (1991). A User’s guide to the Oxford English DictIonary, Oxford and New York, Oxford University Press.
Chaudhuri, K.N. (1978). The Trading World of Asia and the East India Company 1660-1760, Cambridge, Cambridge University Press.
Cunnington, C. Willett and Cunnington, Phillis (1952). Handbook of English Mediaeval Costume, London, Faber and Faber.
Cunnington, C. Willett and Cunnington, Phillis (1954). Handbook of English Costume in the Sixteenth Century, London, Faber and Faber.
Cunnington, C. Willett and Cunnington, Phillis (1955). Handbook of English Costume in the Seventeenth Century, London, Faber and Faber.
Cunnington, C. Willett and Cunnington, Phillis (1957). Handbook of English Costume in the Eighteenth Century, London, Faber and Faber.
Cunnington, C. Willett and Cunnington, Phillis (1959). Handbook of English Costume in the Nineteenth Century, London, Faber and Faber.
Douglas, Mary and Isherwood, Baron (1980). The World of Goods. Towards an Anthropology of Consumption, Harmondsworth, Penguin.
Fawcett, Tevor (1985). “Argonauts and commercial Travellers: the Foreign Marketing of Norwich Stuffs in the later Eighteenth century,” Textile History, 16 (2), pp. 151-182.
Foucault, Michel (1973). The Order of Things. An Archaeology of the Human Sciences, New York, Vintage Books.
Fukasawa, Katsumi (1987). Toilerie et commerce du Levant au XVIIe siècle. D’Alep à Marseilles, Paris, CNRS.
Hughes, Geoffrey (1988). Words in Time. A Social History of the English Vocabulary, Oxford, Basil Blackwell.
Mann, J. De Lacy, The Cloth Industry in the West of England from 1640 to 1880, Oxford, Oxford University Press, 1971.
Mansfield, J. And Cunnington, Phillis (1973). Handbook of English Costume in the Twenteith Century, London, Faber and Faber.
Munro, John H. (1983). “The Medieval Scarlet and the Economics of Sartorial Splendour”, in J.H. Munro (ed.), Textiles, Towns and Trades. Essays in the economic history of late medieval England and the Low Countries, Aldershot and Vermont, VARIORUM, V, pp. 13-70.
Merchant’s Warehouse (1696). The Merchant’s Ware-House Laid Open, or The Plain Linen Draper, London.
OED (1989). The Oxford English Dictionary. Second edition, vol. 1, Oxford, Oxford University Press.
OED (1992). The Oxford English Dictionary. Second edition on Compact Disc, Oxford, Oxford University Press. (OED online: http://www.oed.com)
Rey, Alain (1995). Essays on Terminology, Amsterdam/Philadelphia, John Benjamins.
Rothstein, Natalie (1990). Silk Designs of the Eighteenth Century, London, Thames and Hudson/Victoria and Albert Museum.
Sager, Juan (1990). A Practical Course in Terminology Processing, Amsterdam/Philadelphia, John Benjamins.
Spufford, Margaret (1984). The Great Reclothing of Rural England. Petty Chapmen and their Wares in the Seventeenth Century, London, The Hambledon Press.
Styles, John (1983). “Manufacturing, consumption and design in eighteenth-century England”, in John Brewer and Roy Porter (eds.), Consumption and the World of Goods, London and New York, Routledge, pp. 527-554.
Taylor, Lou (1983). Mourning Dress. A Costume and Social History, London, Allen and Unwin.
Textile Institute (1995). Textile Terms and Definitions, 10th edition, ed. J. McIntyre and P.N. Daniels, Manchester, The Textile Institute.
Thomas, P.J. (1926). Mercantilism and the East Indian Trade, London, P.S.King.
Willinsky, John (1994). Empire of Words. The Reign of the OED, Princeton, New Jersey, Princeton University Press.
Warner, Frank (1921). The Silk Industry of the United Kingdom. Its Origins and Development. London.
 Given the predominance of silk as a mourning fabric, this may have given a slight bias towards terms denoting silk fabrics in this corpus.
 For a commentary on The Merchant’s Ware-House Laid Open, or the Plain Linnen Draper, see Spufford (1984: 107-113).
 All definitions or explanations of terms in this article are from entries of the OED (secondo edition) unless otherwise specified.
 “Wool” was taken to include also materials made from goat’s hair such as cashmere, angora etc., but not from horse’s hair (crinoline).
 “Here two words are apparently mixed up: 1 (sense 1), properly cadas, cadace, OF. cadaz, cadas, cf. Cotgr. cadarce ‘the tow or coarsest part of silke, whereof sleaue is made’; cf. Irish cadas = cadan cotton; 2 F. cadis (15th c. in Littré) ‘sorte de serge de laine, de bas pris’. Of both, the ulterior history is unknown” (OED).
 This is intended to indicate, of course, that the derivation is “unknown” in terms of the information that the OED gives as to the origin of terms. In many cases it is probable that further research into the linguistic origins of such toponyms may come up with a clear answer.
 See here the criticisms of the OED in Willinsky (1997: 87).
 These figures may be slightly skewed by the OED’s reliance on the Merchant’s Ware House (1696): the first of occurrence for 13 terms is taken from this source.
 The original meaning of scarlet as a fabric rather than a colour is disputed in a fascinating article by John Munro (1983). He argues for the dominance of the colourings in the luxury wool products, the dyes representing 40% of the cost of the finished product. For this reason, he says, it is likely that the colour gave its name to the luxury cloth reserved for it, rather than vice-versa.
 The order of the categories is loosely based on the chronology of the importance of these textile types in British textile production and consumption.
 Four terms have been classified as “other” in terms of textile type. These are crinoline, made from horse hair, pine wool, made from “the spun fibres of pine-leaves”; scotch cloth, “said to have been made of nettle fibre”; and asbestos.
 The Draper’s Dictionary notes the possibility that the term derives instead from the Dutch ostade, “a manufacture introduced by Flemish weavers” (Drapers’ Dictionary, 1882: “Worsted”).
 The OED makes a distinction, in principle at least, between corruption, which indicates an element of error, and the more conscious process of adaptation.
 See for Dutch influences, Mann (1971: 11-13); for the importance of French Huguenots on English silk manufacture, Rothstein, 1990; for English and French trade in textiles with the Middle East, Fukasawa (1987); and for British trade with India and China, Chaudhuri (1978).
 I use this definition of neology, amongst the various ones available, as the most suitable to this context. See Rey (1995: 63-93).
Appendices & Tables
©inTRAlinea & Patrick Leech (1999).
"“Who Says Manchester Says Cotton.”", inTRAlinea Vol. 2.
This article can be freely reproduced under Creative Commons License.
Stable URL: http://www.intralinea.org/archive/article/1610