Skip to navigation – Site map

HomeAnglophonia28French Loanwords in Canadian Engl...

French Loanwords in Canadian English: A Usage-Based Approach

Julie Rouaud

Abstracts

This paper focuses on the semantic evolution of French loanwords adopted into Canadian English from the seventeenth century onwards. Our lexical database of Canadianisms of French origin, extracted from historical dictionaries and works on contemporary Quebec Gallicisms, was verified by other sources, such as a recent Canadian dictionary and a corpus of Canadian English. Our methodology allows us to observe French loanwords and their patterns of evolution from a semantic point of view. More traditional treatments of loanwords in terms of zero semantic extension, restriction or expansion of meaning (Filipovic 1968) soon find their limits and fail to offer a more comprehensive treatment of semantic change for the whole lexicon, since both the native lexicon and borrowings seem to follow similar paths in their semantic evolutions. The mechanisms of narrowing and broadening of meaning operating in lexical pragmatics for the native lexicon are the same as for loanwords once they have been integrated in the recipient language. While giving more importance to context and allowing a more unified lexical treatment, this approach raises several questions as far as the basic assumption of a linguistically-specified meaning is concerned. Consequently, we propose to adopt another perspective, a usage-based approach, which stresses the central role played by context and usage in the understanding of semantic change, particularly in a diachronic perspective. Our methodology is compatible with that type of approach. By positing the existence of exemplars which are stored in the mental lexicon with all their features, including meanings and inferences in context, we assume that meaning is not necessarily encoded linguistically and therefore fixed but that it can be stored and become conventionalized through use. This approach also allows us to explain cases of metonymy or metaphor we encountered and which can sometimes be challenging to tackle in other theoretical frameworks.

Top of page

Full text

Introduction

1Borrowing is defined as “the incorporation of foreign features into a group's native language by speakers of that language: the native language is maintained but is changed by the addition of the incorporated features” (Thomason and Kaufman 1988: 37). In other words, a recipient language (RL) borrows from a source language (SL) certain features, but most commonly lexical units, and integrates them into its system. Lexical borrowing is one way to increase the lexicon of a language. Once lexical borrowings---also called loanwords---enter the RL, they can undergo different degrees of phonological, morphological and semantic adaptations. This article focuses on the semantic evolution of French loanwords adopted into Canadian English from the seventeenth century onwards, in other words, Canadianisms of French origin. Since this study includes a diachronic perspective, there is a need to rely on historical dictionaries of Canadian English because historical corpora of Canadian English are very rare (Brinton 2015). French loanwords seem to follow semantic evolutions that can be grouped together. As far as semantics is concerned, borrowings do not seem to evolve randomly but rather follow similar patterns and can be paralleled with those found in the native lexicon. How to account for semantic change in loanwords in a diachronic perspective and to what extent are these loanwords subject to processes of semantic change similar to those that can be observed in the native lexicon of a language? In order to shed light on these questions, we first present our methodology to constitute and verify a lexical database of French borrowings in Canadian English. A more traditional way of dealing with semantic change in loanwords is then reviewed. Finally, we propose another approach based on usage by comparing a lexical-pragmatic to a usage-based framework in order to show that usage and context are key to account for processes of semantic change in loanwords as well as in the native lexicon.

Methodology

2A lexical database of French loanwords in Canadian English from the seventeenth century onwards was automatically extracted from different sources, among which historical dictionaries. The loanwords that were not extracted automatically were localized and integrated thanks to a manual verification. The Dictionary of Canadianisms on Historical Principles or DCHP 1 (Avis et al. 1967) is the first historical dictionary ever concerned with the specificities of Canadian English, published on the occasion of the centenary of the Confederation. DCHP 1 offers a semantic division with dated quotations going back to the sixteenth century and the dates of first written attestation for each given meaning. The Dictionary of Canadianisms on Historical Principles 2 or DCHP 2 (Dollinger and Fee 2017) is an updated online version of DCHP 1 displaying new Canadianisms that have appeared after 1966 with up-to-date quotations from 1966 to 2016. Other sources of French Canadianisms are studies on contemporary Gallicisms in Quebec English, carried out by different linguists studying contemporary corpora of English in Quebec, such as the Montreal Gazette (Boberg 2010; Fee 2008; Grant-Russell and Beaudet 1999; Josselin 2001; McArthur 1989; McArthur 1992; Poplack et al. 2006; Russell 1996; Yuen 1994). Table 1 shows the distribution of headwords in our lexical database (Rouaud 2019). Our lexical database is composed of 721 loanwords, among which 90% are nouns, which reflects a general tendency in the hierarchy of borrowability established as far back as Whitney (1881).

French Canadianisms (DCHP 1 & 2)

Quebec Gallicisms

Total

691

30

721

3Table 1: Distribution of headwords in the lexical database

4Once the extraction operated, those French loans were verified in another more recent general dictionary, the Canadian Oxford Dictionary or CanOD (Barber 1998) which is also concerned with the specificity of Canadian English, but also in different corpora, that are the press corpus Canadian Broadcasting Corporation (CBC News website), the spoken corpus PAC-LVTI [Montreal] (Rouaud 2016-2017) and the general Canadian corpus of reference Strathy Corpus of Canadian English (Strathy Language Unit 2011). In this article we are concerned with headwords from our database which are attested in the Strathy so as to study potential semantic evolutions in context from 1921 to nowadays. These loanwords represent approximately 33% of the database (234/721).

5Using dictionaries raises a certain number of issues that must be kept in mind, especially as far as semantic divisions and citations are concerned. Each lexicographical work is indeed often based on previous works. In the case of DCHP 1, the editorial team elaborated on previous works and materials gathered by Charles J. Lovell, who was part of the editorial team of A Dictionary of Americanisms published by Mitford M. Mathews in 1951. This dictionary was compiled following “the well-established tradition of The Oxford English Dictionary, A Dictionary of American English and A Dictionary of Americanisms, all three of which have been indispensable during the editing process. To follow this tradition, however, is to accept certain restrictions” (Avis et al. 1967: xii). Consequently, all those dictionaries make common choices that need to be considered.

6Firstly, there is generally no hierarchy in terms of usage in the semantic division. Each meaning is given independently of its actual frequency of use in Canadian English. A loanword such as lodge illustrates this fact.

1a. See also: beaver lodge
1b. See also: muskrat house
2a. an Indian dwelling, as a teepee, or wigwam. See also: Indian lodge teepee wigwam ((n.)) (def. 1a)
2b. See also: tent (def. 1)
3. See also: tent (def. 2)
4a. † See also: fishing lodge hunting lodge (def. 2)
4b. a hotel for accommodating guests in a resort area.

7Sense 1a corresponds to the earliest attestation (1744), while sense 4b corresponds to the most recent one (1958). Senses are ordered chronologically, which is coherent with a historical approach but does not reflect usage. Within the numbered division, they are grouped together by letters according to their semantics. 1a and 1b both concern animal dwellings, even though 2a (1765) was attested earlier than 1b (1896). Sense 2b is marked as Hist, indicating that this sense is not in use anymore except in specialized historical texts. It shows that usage is not a primary criterion in the semantic categorization of the dictionary.

8Secondly, quotations do not necessarily reflect actual use in Canadian English. Walter Avis explains in his users guide how quotations were selected and says that several quotations corresponded to “some other context not to be classed as evidence for the use of the term in English” (Avis et al. 1967: xviii). Several quotations for headwords are explanations of the French term in English and do not show any evidence of its actual use in Canadian English at all. The 1907 citation of the headword achigan is an example.

(1) achigan (1907) “Achigan ... A French-Canadian name of the small-mouthed black bass ... occasionally found in English writings.”

9Those particular contexts are not taken into account in our study. Another type of citations that is excluded from our study are personal etymological remarks, usually corresponding to folk etymologies, such as for aboiteau (2):

(2) aboiteau (1898) “I have always understood that the word "aboideau" came from the French words "aboi," "d'eau"; "aboi"--to keep at bay, "d'eau"--the water. It is a poetical expression taken from hunting--the moose keeping the dogs at bay. . ..”

10Another misleading fact as far as quotations and usage are concerned is the issue of overrepresentation of a headword. The number of quotations in a dictionary entry can give the dictionary user the impression that a word is relatively frequently used. Such is the case for pembina ‘a type of cranberry; its shrub’ which includes six quotations for one definition but the loanword does not appear in any other sources that were used in the verification process, suggesting that the term is not as commonly used as it seems.

11One of the most important questions in building a semantic division in a lexicographical work is how to decide when words with identical forms are to be treated as two separate headwords or not. On the one hand, we would have a case of homonymy between two independent lexemes, and on the other hand, we would be confronted to cases of polysemy of a same lexeme. Depraetere and Salkie (2017: 21) give the example of battery and state that “the boundary between accidental homonymy and motivated polysemy is sometimes hard to draw”. This difficulty to draw the line can have an incidence on semantic change. In our database, caribou is one of those cases with an extended semantic division but also a separation between two headwords. The choice operated in DCHP 1 was to separate common and proper nouns, creating two entries each with their own semantic divisions.

Caribou/Cariboo
1. an extensive region in south-central British Columbia, originally that region between the junction of the Quesnel and Fraser Rivers and the Cariboo Mountains, but later, especially in popular usage, a much larger region, extending from Lillooet almost to Prince George.
2. (usually plural) a name given to the Reformers in Upper Canada in the 1830’s.
2a. one of a branch of the Abenaki Indians of the Maritimes.
2b. a group of Eskimos living inland in the District of Keewatin, west of Hudson Bay. See also: Caribou Eskimo
3. a twin-engined aircraft designed and produced in Canada, used by the armed services and for low-cost passenger and freight service in areas not having first-rate airports with long runways, the Caribou being able to take off and land on strips less than 500 feet long.

caribou/cariboo
1. any of several species of North American reindeer, genus /Rangifer/, native to Canada, Alaska, and formerly to Maine and Mass. See also: carreboeuf moose-deer reindeer tuktu (def. 1) tundra deer
2. the flesh of the caribou; caribou meat.
3. the skin or hide of the caribou.
4. a potent mixture of whisky-blanc, and red wine. See also: loup-garou (def. 2) whisky blanc whisky-wine

12This separation raises a problem. Two different entries would suggest that Caribou and caribou are homonyms. Yet, the British Columbian region as well as the two tribes were actually named after the animal which was omnipresent in those areas. Is it then logical to treat them separately? On the other hand, a caribou also designates a drink (sense 4), especially in Quebec. Nothing in the definition nor in the quotations helps explaining the relation between the animal and the drink. Both definitions appear to be more semantically unrelated than the animal and the British Columbian region aforementioned. In that case, could the drink and the animal really be considered as cases of polysemy? Were they more related at some point but did their semantic evolution blur this connection?

13 The limits dictionaries present have to be taken into consideration in order to study how semantic change works for loanwords in a diachronic perspective.

  • 1 Poplack, Sankoff and Miller (1988) define nonce borrowings as single-item loanwords that occur o (...)

14Our second stage of verification involves using a corpus, the Strathy Corpus of Canadian English in order to alleviate the limits that we encountered with dictionaries, such as the fact that usage is not really taken into account, while it plays a key role in semantic change. The Strathy Corpus is a relatively small corpus with approximately 50 million words and composed of several genres, ranging from academic writings to transcriptions of spoken language. It also provides the researcher with the number of occurrences of a term (tokens). French loanwords can thus be verified in context with their token frequencies which allows us to determine if they are actually used in Canadian English and their frequency rate. Unsurprisingly, relatively low frequencies that is token frequency 20account for more than half of the headwords that are present in the Strathy Corpus. It can be said that many French Canadianisms are not part of the “core” vocabulary of Canadian English, since most of those loanwords correspond to a more specialized vocabulary pertaining to specific domains, such as local fauna and flora or the way of life of the First Nations. However, it has to be kept in mind that their presence in a relatively small sample corpus of Canadian English already shows that they are not simply cases of nonce borrowings1 so can potentially be more common than might be expected. Another advantage of using such a corpus is that lexical units appear in context and it can reveal new semantic developments, as is shown below.

Traditional approach to semantic change in loanwords

15 Traditionally, analyses of semantic changes of loanwords in any language would be carried out separately from analyses of the native lexicon. In his study of semantic change of English loanwords in Serbo-Croatian, Filipovic (1968) established a classification in which he identified three degrees of changes in terms of semantic extension: zero extension, restriction and expansion. As far as zero degree is concerned, the meaning remains unchanged once the loanword has been integrated and corresponds to the meaning of the source language or dialect, that is to say French or Canadian French in our case. For instance, barbotte [< Cdn F] is a large catfish in French and in English, as shown in examples (3), (4) and (5).

  • 2 The figure indicated between square brackets corresponds to the number of tokens in Strathy.

(3) La barbotte est une espèce souvent négligée et pourtant, sa chair douce est délicieuse. [05/05/2017 La Presse www.lapresse.ca]
(4)
He's always ketchin' barbotte, dat's w'at you call bull-pout. (Drummond Johnnie Courteau, 1901) [DCHP 1]
(5)
In summer they ate pickerel or bass from the river, sometimes even catfish, barbotte. (Canadian Fiction Magazine 87, 1994) [Strathy 12]

16“Zero” semantic extension can be said to correspond to the basic mechanism of borrowing. The RL borrows from a source language (SL) a term or expression with a particular meaning already attached to it. An “important observation which can be made is that loanwords are normally borrowed in one meaning only. This means that SL polysemies (and homonymies) normally disappear in processes of borrowing” (Winter-Froemel 2014: 73). Loanwords such as depanneur (“corner store” in Quebec) are not considered as cases of restrictions properly speaking, since the lexeme has been borrowed with one meaning from the start. It corresponds to the basic mechanism of borrowing (6).

(6) He was arrested for those crimes when he returned to buy cigarettes at a depanneur he had just robbed. (Montreal Gazette 1998) [Strathy 10]

17As far as restriction is concerned, “it begins with specialization from a general to a technical sense” (Filipovic 1968: 110). According to Winter-Froemel, restriction is a frequent type of phenomenon as far as loanwords are concerned:

Un type de changement qui s’observe avec une fréquence relativement élevée dans les emprunts sont les restrictions de sens, dans lesquelles le mot de la langue cible acquiert un sens plus spécifique que le mot de la langue source. (Winter-Froemel 2012: 55)

18The word acquires a more specific sense than it has in the source language once it has been integrated.

(7) acclamation the election of a candidate without opposition. [DCHP 1]
(8)
Praised by many MPs after he cleared the way for an acclamation by bowing out of the race, Rae avoided scrums with reporters but said in a brief interview the acclamation was upbeat. (O’Neill The Windsor Star 2008) [DCHP 2]

19Acclamation was first borrowed in the general sense of “proclaim or announce with applause” (Oxford English Dictionary). It is still in use in that general sense of course in all varieties of English but as a Canadianism, this sense has been reduced to a political context when a candidate is elected without opposition (8). As an example, the members of a party would proclaim the victory of a candidate by applauding him/her.

20Finally, expansion is the opposite of restriction, going “from a technical to a general sense”. Consequently, “their meaning started to loosen its original preciseness” (Filipovic 1968: 110). This suggests that once integrated in the lexicon, a specific sense can become more general. For instance, Pequiste originally designates a person belonging to the Parti Québécois (9) but its meaning has been widened to apply to nationalist persons in general and not just those belonging to this political party (10).

(9) The Pequiste opposition to Senator-elect Stan Waters has nothing to do with the Reform Party's opposition to the Meech Lake accord. (The Edmonton Journal 1989) [DCHP 2]
(10)
“Even the most pequiste folks aren't so militant about it anymore” [2016 Reddit Forum www.reddit.com/r/Edmonton/comments/42iyej/how_are_french_canadian_andor_quebecers_seen_in/]

21Classifying loanwords in terms of technical and general senses does not prove sufficiently satisfactory to account for the complexity of semantic change in loanwords. Firstly, it gives the impression that loanwords and the native lexicon are treated separately because of the zero extension degree which cannot be applied to the native lexicon. Secondly, in the works cited, neither Filipovic nor Winter-Froemel insist on the importance of context and how it can affect the semantics of a term. Of course, examples can always be found showing a restriction or expansion of meaning since they are mechanisms that can find echo in other theoretical frameworks, but a more unified treatment would be required to study the processes of semantic change at hand in languages, and especially how contexts of use participate in the modulation of meaning.

Treatment in a usage-based framework

22Traditional categorizations of borrowings find an echo in frameworks based more on usage. Lexical pragmatics is a field that “investigates the mechanisms by which linguistically-specified […] word meanings are modified in use.” (Wilson 2004: 343) Wilson describes the processes of narrowing and broadening, which are the two main mechanisms by which a linguistically-specified denotation is modified in use. Both of them definitely parallel Filipovic’s restriction and expansion phenomena but apply to the native lexicon as well, while highlighting the fact that these phenomena depend on use. Joan Bybee, who adopts a usage-based framework, and more precisely exemplar theory, also talks about those two processes and she also insists on the importance of context in semantic change. “Semantic change always occurs in a particular discourse context, where the context helps to drive the interpretation in a certain direction.” (Bybee 2015: 197).

23Narrowing occurs when “a word is used to convey a more specific sense than the encoded one, resulting in a restriction of the linguistically-specified denotation” (Wilson 2004: 344). Two cases will be analyzed below. Bat(t)eau (1760) denoted four different types of boats which are recorded in DCHP 1. These distinctions seem to be blurred in Strathy and the word usually designates one kind of boat in particular, a sort of sloop (11).

(11) I have come down from the upper reaches with timber, you see it lying there on that raft; and presently I go back. How do you go back? By a canoe or batteau, m'sieu, there is no other way. (Sullivan Three Who Came to Ville Marie 1941) [Strathy 19]

24The loanword Metis originally designates a half-breed of Indian and French origin but is most often applied to a specific community living mainly in Manitoba and composed of descendants of Indian-French half-breeds, the Metis (12).

(12) Metis writer and storyteller Maria Campbell is full of evidence of an improvised Christianity, as expressed in the traditional old men's stories that she gathered. (Essays on Canadian Writing 2002) [Strathy 1545]

25On the other hand, a “word […] used to convey a more general sense, with consequent widening of the linguistically-specified denotation” (Wilson 2004: 344) is a case of broadening. The latter encompasses approximation and category extension, such as metaphors. Mush [< F. « marche.IMPER.2.SG »] corresponds to a deformation of the order given by French trappers to their sled dogs. The original imperative form was borrowed from French in English and used as an interjection addressed to sled dogs. It then became a full-fledged verb (1862) meaning “travel by or with a dog-sled” (13) and also a noun (1910) denoting a journey by dog-sled.

(13) Cities like Saskatoon and Edmonton won't be on the short list. And Cambridge Bay? Hook up the huskies and mush. (National Post 2001) [Strathy 4]

  • 3 In the Strathy corpus, the stick to play the game has become a lacrosse stick or a cross but not (...)

26Lacrosse also conveys a more general sense than its original denotation. The name of the popular Canadian sport comes from the stick used to play the game [< F. « la crosse »]. Example (14) illustrates this previous use of the loanword while (15) shows broadening of this original meaning3.

  • 4 The capital letter was in the original nineteenth century source and was reproduced in the dicti (...)

(14) he was received by the Committee, presented with an address, accompanied by a very handsome gold-mounted Lacrosse, and an elegantly bound copy of Mr. Beer's work on the game. (Canadian Illustrated News 1869) [DCHP 1]4
(15)
Hockey gear was too expensive, so it was soccer in the winter and lacrosse in the summer. (Times Colonist 2002) [Strathy 127]

27The cases mentioned above seem to be clear cases of narrowing and broadening of meaning. Others are potentially more ambiguous since they can be a combination of narrowing and broadening at the same time, as explained by Carston (2010: 158). “The pragmatically derived concept may be more specific or more general than the encoded concept; […] or it may be a combination, both extending the lexical denotation and excluding some part of it.” This implies that “lexical denotation” is a set of features that form a whole. That is why a part of it can be subject to a process while another can be subject to the opposite. For instance, snye [< Cdn F. « chenail »] meaning channel, was first narrowed down to designate a side-channel, bypassing falls or rapids and creating an island (16).

(16) At this place, there are numbers of islands formed by snies winding round the Falls. (MacTaggart Three Years 1827) [DCHP 1]

  • 5 The metaphorical use is the only token in Strathy.

28However, this specific meaning was then elaborated upon and used metaphorically5, as in (17):

(17) In my remarks on language standards, I am going to explore only one small snye of the murky and troubled waters, in hopes that it will give a hint of what is involved. (W.C. Lougheed In Search of the Standard in Canadian English 1986) [Strathy 1]

29This use highlights the winding and not straightforward aspects of the element described, while suggesting bypassing the “troubled waters” of language. A shanty [< F. « chantier »] was originally a crude dwelling or camp for loggers or miners (1822). It became a specific type of crude wooden dwelling, a shack (18), and later was applied to a building or a living area usually in rundown conditions (19).

(18) Out in the Canadian forest, not another white woman for fifty miles, beside a miserable little shanty. (John Price-Brown Laura the Undaunted 1930) [Strathy 109]
(19)
The twin shanties that through all the town's life have served as men's and women's bath houses are still there. (Wendy Reynolds and Vera Madden Drug Wise 1992) [Strathy 109]

30If we compare those examples, we note that the domain of application for shanty was first circumscribed to a specific type of dwelling, probably because the referent of the original meaning was made of wood and usually found in forests. After this shift in referent, another shift occurred. Shacks tend to be inhabited and are almost always in rundown conditions. We can deduce that parts of the referent in (18) started to apply more broadly to other types of accommodation.

31The processes of narrowing and broadening are compatible with a usage-based theoretical approach and are dealt with by usage-based theorists, such as Bybee (2015: 202). The difference with pragmatic frameworks lies more in what is posited as a starting point and the explanation of semantic change.

32In order to be able to deal with the semantic evolution of loanwords in terms of narrowing or broadening of meaning, a starting point is obviously needed. In lexical pragmatics, a linguistically-encoded meaning is posited and is then modulated in context. First, how do we define it? It turns out not to be an easy task because “as well as not being lexicalized, there isn’t a linguistic phrase that fully encodes them either, and the paraphrases are intended as just a rough indication to aid readers in understanding what we have in mind in particular cases.” (Carston 2010: 165) Would linguistically-encoded meanings then be ad hoc concepts? In our case of a historical lexicographical database of loanwords, what would we consider as “linguistically-encoded meanings”? Because of the lexicographical materials we use for our study, already established semantic divisions have to be taken into account. To a certain extent, we adopt what can be called a “dictionary” view of semantics, as Lemmens (2017: 106) states it: “In a ‘dictionary’ view of semantics, an expression’s linguistically relevant meaning is limited to the minimal literal sense, as recorded in dictionary definitions. This linguistic meaning then gets modulated in context”. Nevertheless, using a corpus does not only provide words with contexts which modulate a dictionary sense taken as granted, but also allows us to verify dictionary definitions and question the relevance of linguistically-encoded meanings. Consequently, even though a “dictionary” view seems unavoidable, it is not blindly carried out without confronting to actual use in corpus. That is why our own methodology is not incompatible with a usage-based theoretical approach.

33Usage-based theories place usage at the center of all linguistic phenomena. Consequently, a linguistically-encoded denotation is to be questioned. It has “never been very clear what exactly the basis is for postulating a given linguistic meaning (apart from intuition which, however, is quite unreliable) but also a full specification of linguistic meaning is generally simply absent.” (Lemmens 2017: 104).

34However, this does not mean that there is no basis for linguistic meaning that can be taken into account. Bybee (2010: 52) states that “frequently made inferences from the context can become part of the meaning of an expression or construction. This suggests no clear divide between aspects of the meaning that are derivable from context and those that are inherent to the lexical item or construction.” The author talks about inherent and contextual aspects of meaning but implies the difficulty to draw a clear line between the two, especially since these “inherent” aspects come from context as well. They are developed through form-meaning pairings which are stored in the mental lexicon as units through repeated uses. “In an exemplar model the combinations of meaning in real utterances are registered in memory and if repeated, become conventionalized as possible interpretations” (Bybee 2010: 55). Consequently, “cognitive linguistics does not deny the existence of pragmatic processes or that linguistic expressions have a stored meaning, they clearly do” (Lemmens 2017: 107), through a rich memory process. Rich memory is a domain-general cognitive process in which “the meanings, inferences and aspects of the context relevant to meaning are also stored with exemplars” (Bybee 2010: 29). As a consequence, a “word is represented as a set of phonetic exemplars with a small range of variation associated directly with a set of meanings” (Bybee 2010: 19).

  • 6 Langacker (2000: 3) uses the term entrenchment to designate a psychological process facilitating (...)

35Once exemplars with their set of meanings are categorized or matched according to their similarity with other exemplars, they are stored in memory. The more exemplars are to be found in a category, the more entrenched they become in the mental lexicon, rendering further categorization of novel occurrences faster and more efficient. In that respect, a linguistically-specified denotation proves rather illusory to the extent that repetition is key to linguistic meaning and therefore cannot be fixed and can be subject to change through usage. The most frequent meanings in corpora which are more likely to be the most entrenched6 exemplars in the mental lexicon form a semantic core that subsumes more peripheral contextual variants.

36In a diachronic perspective, the oldest attested meanings in our database can serve as a basis for semantic change on condition that they are still in use nowadays, either in recent dictionaries or in the Strathy corpus. The most frequent meanings found in Strathy are considered as the most entrenched or central exemplars due to the effect of frequency (Bybee 2010) and provide an explanation for potential semantic evolutions. Semantic changes that are described earlier in examples (11)-(19) can also be tackled in terms of domain-general processes which are at hand in semantic change for the native lexicon and loanwords, rendering a linguistically-encoded meaning that would be modulated in context unnecessary.

37Each of the aforementioned examples can be considered as clusters of exemplars presenting a set of characteristics attached to it. Through time and repeated uses, some characteristics became more entrenched than others and allowed novel occurrences to be categorized with this cluster by analogy. Sometimes, semantic change occurs at the expense of the oldest senses. For instance, lacrosse designating the stick used to play lacrosse is not found in Strathy, suggesting a shift to the name of the sport. Instead of regarding snye and shanty as cases of narrowing and broadening at the same time, their semantic evolutions find an explanation in exemplar theory. The sets of features attached to the category stored in rich memory are more or less entrenched. As for shanty, one of its most frequent characteristics inferred from context is that shanties are in rundown conditions, hence a wider application to any type of accommodation. The metaphorical use of snye in (17) can also be due to the semantic features attached to the exemplars of the category SNYE. that Snies are usually not straightforward since they are winding. The more this metaphorical use will be used, the more it will become conventionalized. Even though many loanwords in our database are low-frequency lexical items, they are stored in the mental lexicon the same way as more frequent terms, but are generally less affected by the reducing effect of frequency from a phonological point of view. In this case, their phonological forms tend to remain unadapted.

38Metonymy is a special case that usually represents a real challenge but can find an explanation in usage-based frameworks. “Metonymic uses of language raise new issues since they are plainly not cases of lexical/phrasal narrowing or broadening” (Carston 2010: 176). It “occurs when a term for one concept is used for an associated concept” (Bybee 2015: 199). In our database, metonymic uses are quantitatively less important and mainly concern the flesh and hides of animals. The caribou is a large North American reindeer. In Strathy, the majority of the 1055 tokens of caribou refer to the animal, but we can find uses of the word designating its skin or meat, as in example (20).

(20) Standing there in my caribou parka, I surveyed what would be my home for the next two weeks: a tent, a small igloo and a solidly frozen seal (the last food item for the coming fortnight). (Up Here 1996) [Strathy 1055]

  • 7 Chandler (2015) presents differences between exemplar and prototype theories within a usage-base (...)

39Exemplars that are members of the category CARIBOU do not necessarily share all features. The most central exemplars present the most recurrent characteristics attached to the animal. Others are more or less peripheral7. Here the meat and skin of the animal are not the most central exemplars of the category but are still closely linked to it, being parts of the whole animal.

40A usage-based approach offers another perspective in dealing with semantic change in loanword adaptation. The most central exemplars, that is to say the most frequent ones, constitute the semantic core of a category but this core can evolve through time. The constant addition of new exemplars that are more or less peripheral at first can become entrenched and trigger semantic change.

Conclusion

41The processes at hand in loanword semantic change can be paralleled with the processes operating within the native language, once they have been integrated. They follow similar paths. That is why they should be dealt with in the same way as in the native lexicon and not separately, as can be found in more traditional treatments of semantic extension in loanwords. Other theoretical frameworks, such as lexical pragmatics, offer a more unified treatment of the semantic adaptation of loanwords through context-dependent processes that can be applied to both lexical borrowings and the native lexicon. A usage-based approach to semantic change gives usage a key role, especially in a diachronic perspective. Exemplars are stored in the mental lexicon with their meanings drawn from contexts of use, thus questioning the assumption of a fixed linguistically-encoded meaning which would be modulated in context. The usage-based approach can also provide an explanation for cases or phenomena that would otherwise be more difficult to tackle by adopting another perspective in terms of exemplars categorized according to a bundle of features that are more or less peripheral in relation to the most central exemplars. This is the case for metonymic or metaphorical uses. This approach could also help rethink semantic divisions in our database by taking into account usage in the organization of dictionary entries.

Top of page

Bibliography

Avis, Walter S., et al., editors. A Dictionary of Canadianisms on Historical Principles. Gage, 1967.

Barber, Katherine, editor. Canadian Oxford Dictionary. First, Oxford University Press, 1998.

Boberg, Charles. The English Language in Canada: Status, History and Comparative Analysis. Cambridge University Press, 2010.

Brinton, Laurel J. ‘Studying Real-Time Change in the Adverbial Subjunctive: The Value of the Bank of English’. Transatlantic Perspectives on Late Modern English, edited by Marina Dossena, John Benjamins, 2015.

Bybee, Joan. Language Change. Cambridge University Press, 2015.

Bybee, Joan. Language, Usage and Cognition. Cambridge University Press, 2010.

Carston, Robyn. ‘Lexical Pragmatics, Ad Hoc Concepts and Metaphor: A Relevance Theory Perspective’. Italian Journal of Linguistics, vol. 22, no. 1, 2010, pp. 153–80.

Chandler, Steve. ‘The Analogical Modeling of Linguistic Categories’. Language and Cognition, vol. 9, 2015, pp. 52-87.

Dollinger, Stefan, and Margery Fee, editors. DCHP-2: The Dictionary of Canadianisms on Historical Principles, Second Edition. 2017, www.dchp.ca/dchp2.

Fee, Margery. ‘French Borrowing in Quebec English’. Anglistik: International Journal of English Studies, vol. 19, no. 2, 2008, pp. 173-88.

Filipovic, Rudolf. ‘Semantic Extension Changes in Adaptation of English Loan-Words in Serbo-Croatian’. Studia Romanica et Anglica Zagrabiensia, no. 25–26, 1968, pp. 109-19.

Grant-Russell, Pamela, and Céline Beaudet. ‘Lexical Borrowings from French in Written Quebec English: Perspectives on Motivation’. U. Penn Working Papers in Linguistics, vol. 6, no. 2, 1999, pp. 17-33.

Josselin, Amélie. L’emprunt Lexical En France et Au Canada : Le Cas Particulier Des Anglicismes et Des Gallicismes et Leur Traitement Lexicographique. Master’s Thesis. Université de Lyon II, Centre de Recherche en Terminologie et Traduction, June 2001.

Lemmens, Maarten. ‘A Cognitive, Usage-Based View on Lexical Pragmatics’. Semantics and Pragmatics: Drawing a Line, edited by Ilse Depraetere and Raphaël Salkie, Springer International Publishing, 2017, pp. 101-14.

McArthur, Tom. ‘Quebec English’. The Oxford Companion to the English Language, edited by Tom McArthur, Oxford University Press, 1992.

McArthur, Tom. The English Language as Used in Quebec: A Survey. Strathy Language Unit, Queen’s University, 1989.

Poplack, Shana, et al. ‘An English “like No Other”?: Language Contact and Change in Quebec’. Canadian Journal of Linguistics, vol. 51, no. 2/3, 2006, pp. 183-213.

Rouaud, Julie. Lexical and Phonological Integration of French Loanwords into Varieties of Canadian English Since the Seventeenth Century. PhD Thesis. Université Toulouse Jean Jaurès, 2019.

Russell, Pamela. ‘An Investigation of Lexical Borrowings from French in Quebec English’. LACUS Forum, vol. 23, 1996, pp. 429-40.

Strathy Language Unit. Strathy Corpus of Canadian English. 2011, http://corpus2.byu.edu/can/.

Thomason, Sarah G., and Terrence Kaufman. Language Contact, Creolization and Genetic Linguistics. University of California Press, 1988.

Whitney, William D. ‘On Mixture in Language’. Transactions of the American Philological Association, 1881, pp. 5-26.

Wilson, Deirdre. ‘Relevance and Lexical Pragmatics’. UCL Working Papers in Linguistics, vol. 16, 2004, pp. 343-60.

Winter-Froemel, Esme. ‘Formal Variance and Semantic Changes in Borrowing: Integrating Semasiology and Onomasiology’. New Perspectives on Lexical Borrowing: Onomasiological, Methodological and Phraseological Innovations, edited by Eline Zenner and Gitte Kristiansen, Mouton de Gruyter, 2014, pp. 65-100.

Winter-Froemel, Esme. ‘Néologie sémantique et ambiguïté dans aa communication et dans l’évolution des langues: défis méthodologiques et théoriques’. Cahiers de Lexicologie, vol. 100, no. 1, 2012, pp. 55-80.

Yuen, Adrienne L. Gallicisms: An Analysis Leading towards a Prototype Gallicisms Checker. University of Ottawa, 1994.

Top of page

Notes

1 Poplack, Sankoff and Miller (1988) define nonce borrowings as single-item loanwords that occur only once in their corpus. This means that they are supposed to be “on the spot” borrowings.

2 The figure indicated between square brackets corresponds to the number of tokens in Strathy.

3 In the Strathy corpus, the stick to play the game has become a lacrosse stick or a cross but not a lacrosse anymore.

4 The capital letter was in the original nineteenth century source and was reproduced in the dictionary quotation presented here.

5 The metaphorical use is the only token in Strathy.

6 Langacker (2000: 3) uses the term entrenchment to designate a psychological process facilitating re-occurrence by leaving a kind of trace in the speaker’s mind.

7 Chandler (2015) presents differences between exemplar and prototype theories within a usage-based framework, among which the fact that the central elements are prototypes abstracted away from occurrences in context in prototype theory. This is why we do not use this terminology in our study.

Top of page

References

Electronic reference

Julie Rouaud, French Loanwords in Canadian English: A Usage-Based ApproachAnglophonia [Online], 28 | 2019, Online since 20 December 2019, connection on 08 December 2021. URL: http://journals.openedition.org/anglophonia/2508; DOI: https://doi.org/10.4000/anglophonia.2508

Top of page

About the author

Julie Rouaud

Université Jean Jaurès – CLLE-ERSS
julie.rouaud@univ-tlse2.fr

By this author

Top of page

Copyright

Licence Creative Commons
Anglophonia – French Journal of English Linguistics est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search