LINGUIST List 29.3516

Wed Sep 12 2018

Review: Applied Linguistics; Sociolinguistics: Poplack (2017)

Editor for this issue: Jeremy Coburn <>

***************** LINGUIST List Support *****************

Fund Drive 2018
28 years of LINGUIST List!
Please support the LL editors and operation with a donation at:

Date: 25-Apr-2018
From: Natalie Operstein <>
Subject: Borrowing
E-mail this message to a friend

Discuss this message

Book announced at

AUTHOR: Shana Poplack
TITLE: Borrowing
SUBTITLE: Loanwords in the Speech Community and in the Grammar
PUBLISHER: Oxford University Press
YEAR: 2017

REVIEWER: Natalie Operstein,

The overarching goal of ''Borrowing: Loanwords in the Speech Community and in the Grammar'' by Shana Poplack is to provide an empirical foundation for the much-theorized distinction between borrowing and code-switching. More narrowly, the volume centers on disentangling the differences between three categories of borrowed material, attested loanwords, nonce borrowings, and code-switches, with the emphasis on morphological and syntactic means for doing so. The volume is preceded by a foreword by Pieter Muysken and is divided into a preface and twelve chapters.


The first three chapters provide an extended introduction to the volume's goals and the methodology that underlies the research findings and theoretical discussions in subsequent chapters.

Chapter 1 ''Rationale'' introduces the purpose of the book, which is to provide a detailed characterization of the structural, sociolinguistic, synchronic and diachronic aspects of lexical borrowing; as well as its methodology, which centers on the process of integration of donor-language material into the morphological, syntactic and phonological structures of the recipient languages based on an analysis of spontaneous bilingual speech. The process of integration is to be examined by comparing the grammatical structure of borrowed words with that of the unmixed or monolingual stretches of the corresponding donor and recipient languages (''the benchmark varieties''). The chapter provides definitions of the operative terms, including ''donor language'' (LD), ''recipient language'' (LR), ''attested loanwords'' (loanwords already attested in dictionaries or word lists and not involving active borrowing by bilinguals), ''established loanwords'' (loanwords that are widely diffused in the community) and ''nonce words'' (''LD-origin items occurring only once''). The status of nonce words as borrowings or single-word code-switches, the central issue that unites most of the chapters, is to be determined by comparing their behavior with that of attested loanwords, unambiguous code-switches, and corresponding items in the benchmark varieties.

Chapter 2 ''A variationist perspective on borrowing'' elaborates on the methodology of variationist sociolinguistics that was used to obtain the raw data and guide its analysis. Large amounts of spontaneous bilingual discourse were recorded by trained in-group members in carefully stratified community samples. Following transcription of the recordings, the LD material was extracted and divided into two major categories, multiword LD sequences, identified as unambiguous code-switches, and lone LD items. The latter category was further divided into attested loanwords and ambiguous items. The latter were then systematically compared to attested loanwords and to their counterparts in the benchmark varieties (unmixed LD and LR of the same speakers as well as multiword code-switches) in order to clarify their status as nonce borrowings or code-switches. The comparison exploited the existence of ''conflict sites'', or areas of structural or distributional divergence between the LD and LR (such as nominal case marking in the Tamil/English language pair).

Chapter 3 ''Bilingual corpora'' describes the datasets. The Ottawa-Hull French Corpus, which underlies the research findings in Chapters 4 and 8 through 11, was collected in the early 1980s in Ottawa and the Hull sector of Gatineau. Samples of spontaneous speech totaling about 270 hours and 2.5 million words were recorded by in-group members from 120 representative individuals from five neighborhoods differing with respect to the status of French, as majority or minority language, and ratios of francophones to anglophones. Two additional corpora, to be used in Chapter 8 for the study of the diachronic evolution of loanwords, are audio recordings of the elderly Québécois born between 1846 and 1895 and collected in the 1940s and 1950s, and spontaneous speech recorded between 2005-2007 in a school located in one of the neighborhoods represented in the Ottawa-Hull corpus. Smaller bilingual corpora involving other language pairs provide the basis for the discussion in Chapters 5 through 7 and 9.

Chapter 4 ''Borrowing in the speech community'' presents the key research finding, to be further tested and refined in subsequent chapters, that the morphosyntactic integration of LD-origin words in LR occurs at their first mention or soon thereafter. The specific study reported in the chapter is based on nearly 20,000 spontaneously occurring lone word tokens of English origin extracted from the Hull-Ottawa French Corpus; these correspond to 0.83% of the tokens and 3.3% of the types in the 2.5 million word corpus. The study finds that English-origin nouns are assigned gender early and consistently, with the small amount of variation in this area mirroring similar variation in monolingual French. Parallel results obtain with respect to plural marking on English-origin nouns and inflection on English-origin verbs; all English-origin words are also found to be syntactically integrated into French. In contrast to its robust morphosyntactic integration, the phonetic integration of the borrowed material shows both inter- and intra-speaker variation. These findings lead the author to conclude that the loanword versus code-switch status of LD-origin words may be disambiguated based on their morphosyntactic, but not phonetic, integration.

Chapter 5 ''Dealing with variability in loanword integration'' puts under further scrutiny the hypothesis that variability in the morphosyntactic integration of loanwords mirrors internal variability in the LR, including with respect to its conditioning. The hypothesis is tested on spontaneous bilingual speech data involving the Tamil/English language pair. The study finds that English nouns borrowed into Tamil display variation in the accusative and dative case marking as well as with respect to preverbal placement when used as direct objects, paralleling similar variability in the unmixed Tamil of the same speakers. Tamil nouns borrowed into English were found to take on English inflection and determiners and to be placed postverbally when functioning as objects. The chapter also emphasizes the inherent ambiguity of morphologically bare LD-origin nouns with respect to the borrowing/code-switch dichotomy.

Chapter 6 ''The bare facts of borrowing'' takes up the issue of the borrowing/code-switch status of LD-origin nouns in LRs without overt nominal morphology. The method involves reliance on syntactic clues, with the focus on nominal modification patterns in the French/Wolof and French/Fongbe language pairs. The conflict sites exploited for this purpose are LD/LR differences in the placement of determiners and attributive adjectives as well as in their tolerance for bare (undetermined) nouns. The results indicate that, with regard to modification, French nouns in Wolof and Fongbe pattern like nouns in the recipient languages rather than as in unmixed French, which is taken as evidence that they are instances of borrowing rather than code-switching. This result is then confirmed by the study of nominal modification patterns in the English/Igbo language pair.

Chapter 7 ''Confirmation through replication'' uses similar morphological and syntactic means to determine the status of lone LD-origin items in other LD/LR language pairs. One of the reported studies exploits LD/LR differences in the noun-adjective, noun-possessor, and constituent orders to determine the status of English-origin nouns in Gulf Arabic. Another exploits LD/LR differences in the placement of attributive and predicative adjectives to determine the status of English-origin adjectives in Persian. The occurrence patterns of undetermined English-origin nouns in Spanish are found to adhere to LR rather than LD norms. English-origin verbs in Igbo are found to be furnished with Igbo inflectional morphology and to trigger vowel harmony in the affixes. In all these cases, adherence to the structural or distributional properties of the LRs argues for the borrowed status of the studied LD words. Two of the reported studies emphasize the need to compare bilingual speech with the unmixed speech of the same individuals rather than with the respective standard varieties. The studies examine case marking on English-origin nouns in languages with nominal case, Ukrainian and Japanese, and conclude that the occurrence of bare (non-case-marked) nouns is due not to the code-switch status of these nouns but rather to variable case marking in the LRs. The chapter closes with a discussion of French-origin nouns in Tunisian Arabic; these nouns' resistance to morphological integration is explained by reference to ''a higher-order community resistance to inflecting LD-origin nouns'' (p. 120).

Chapter 8 ''How nonce borrowings become loanwords'' answers the question posed in its title by empirically testing two assumptions, the assumption that lone LD-origin items enter LR as nonce forms and are subsequently diffused across the community, and the assumption that such items enter LR as code-switches and are then gradually integrated into LR in the process of their diffusion. Methodologically, the study relies on isolating nonce borrowings in three diachronic corpora of spoken Quebec French, recordings of the elderly Québécois born between 1846 and 1895 and representing nineteenth-century speech, the Ottawa-Hull corpus representing twentieth-century speech, and recordings made between 2005-2007. The major finding is that, based on such diagnostics as verbal morphology, plural marking on nouns, determiner realization, and assignment of nominal gender, the linguistic integration of nonce forms takes place abruptly at their first mention rather than gradually, as assumed by some theorists. A companion finding is that the linguistic integration of a loanword proceeds independently from its social integration, as measured by its intra-speaker recurrence and inter-speaker diffusion.

Chapter 9 ''Distinguishing borrowing and code-switching'' confronts nonce forms with their counterparts in multiword code-switches. (Multiword code-switches were chosen for the comparison instead of the more logical single-word code-switches because of the researchers' inability to identify more than two potential examples of the latter (p. 156).) The study finds that the two categories are both psychologically and structurally distinct, including with respect to their part-of-speech composition, inflection on verbs, plural marking on nouns, assignment of nominal gender, placement of attributive adjectives, and determiner realization.

Chapter 10 ''The role of phonetics in borrowing and integration'' examines the potential of the phonetic integration of LD material to serve as a diagnostic for distinguishing between different types of bilingual behavior. Methodologically, the study focuses on the verbal production of speakers from the Ottawa-Hull French Corpus who engaged in both nonce-borrowing and code-switching behaviors. The conflict sites exploited for the study are the segments represented orthographically as <th>, <h>, <r> and VOT in voiceless stops. The major finding is that all three categories of language-mixing showed variability in phonetic integration, with nonce forms being phonetically integrated into French only 26% of the time, attested loanwords only 57% of the time (both results are contrary to the expected 100% for <th>, <h>, <r> and 88% for VOT), and that code-switches were also phonetically integrated 39% of the time (where 0% was expected). The overall conclusion is that phonetic integration proceeds independently from morphosyntactic integration and cannot be used for distinguishing LD-origin material.

Chapter 11 ''The social dynamics of borrowing'' uses the Ottawa-Hull French Corpus to look at the influence of extra-linguistic factors on the introduction and diffusion of borrowed material. The factors examined are the speakers' age, gender, education, occupation, neighborhood of residence, and proficiency in English. A major finding is that, where the borrowing rate and type are concerned, individual bilingual ability is outweighed by the norms of the speech community. The chapter also emphasizes the need to recognize the explanatory potential of community norms for the observed cross-community differences in the borrowing strategies, particularly with respect to some speech communities' failure to integrate LD-forms morphologically and/or their preference for some but not other integration strategies.

Chapter 12 ''Epilogue'' summarizes the key findings and methodological contributions of the volume and outlines future research directions.


''Borrowing'' brings a wealth of empirical evidence and proposes a concrete and falsifiable conceptual framework to bear on the much-discussed theoretical distinction between borrowing and code-switching. It describes in detail the employed datasets and methodological tools and innovations, making the reported studies replicable and applicable to other donor/recipient language pairs. It also does much more than that: by putting the data into the theoretical context of prevailing assumptions, and by systematically situating the volume's findings against the background of prevailing expectations, it contributes concrete empirical materials for testing proposals in a number of areas, including cross-linguistic borrowability of different part-of-speech categories, the gradualness theory of loanword integration, the impact of community norms on bilingual behavior and integration strategies, and the wealth of proposals about the phonetic and phonological integration of loanwords. But perhaps the most outstanding aspect of the volume is that it raises new and important issues for future research by, among other things, inviting comparison with thematically related studies.

One interesting issue concerns the heart of the methodology, the use of morphological integration as a criterion for differentiating between borrowings and code-switches, in light of Heath's (1989: 24) caution against automatically taking morphological integration as evidence of borrowing (since ''in cases of prolonged language contact, speakers of Lx [LR] may develop productive routines for spontaneously inserting Ly [LD] stems into Lx [LR] frames''). The types of morphological integration discussed in the volume include integration into majority inflectional patterns in the receiving languages (e.g. assimilation of English verbs into the first conjugation in French), integration into non-canonical patterns (e.g. assimilation of English verbs into an analytical construction reserved for non-verbs in Persian), and lack of integration (e.g. non-inflection of French nouns for number in Tunisian Arabic). Participation in all three types of integration is viewed as evidence of the borrowing rather than code-switch status of the affected items because ''[a]ll of these strategies derive from LR and are absent from LD'' (p. 121). It would be interesting to see how this approach applies to borrowing situations in which different degrees of integration are present in the same datasets, and whether it may be refined to incorporate the factors that govern such variation. Some published studies suggest that different degrees of integration of borrowed material may be correlated with both linguistic and extra-linguistic factors; for example, a combination of factors is invoked by Hafez (1996: 16) with respect to different pluralization patterns of borrowed nouns in Egyptian Arabic: ''The choice could be dictated by the degree of conformity of the loanword to E[gyptian] A[rabic] patterns. Such degrees of integration could also reflect language attitudes. For instance, the use of broken plurals (where there is a sound-plural form available) could mean that the user is less educated while use of the sound-plural form could be regarded positively as educated or negatively as affected and foreign''.

A related issue is how the integration-based diagnostic of the borrowing/code-switch status may be applied to borrowing situations in which the morphology of the loans (or lack thereof) ends up shaping that of the recipient language rather than the other way around. Smeaton (1973) reports that English loans in Hasawi Arabic, which reflect only about a decade of contact, ''exhibit a whole spectrum of integrational changes'' (103 fn. 185). The least integrated forms in Smeaton's (1973) study ''must be regarded as stable members of a special atypical (or marginal) morphological category'' (89). Similarly, the pluralization pattern of Spanish ''xenonyms'' like 'lócker(s)' ''locker(s)'' and 'menú(s)' ''menu(s)'' is atypical when compared with that of ''domestic'' nouns like 'líder(es)' ''leader(s)'' and 'tisú(es)' ''tissue(s)'' (Harris 1992: 69-70). Complete failure to integrate the loans morphologically may also lead to the emergence of new inflection classes in the LR; e.g., in Russian most borrowed vowel-final nouns do not enter the system of nominal declension, constituting a stable and growing class of indeclinables (Ungebaun 1947). It also seems that the integration-based diagnostic of the borrowing/code-switch status may need to be refined to accommodate cases in which the words are borrowed together with their LD inflectional formatives, thereby creating new inflection classes in the LRs. A well-known example of this kind in English are nominal pluralization patterns derived from foreign sources, such as alumnus / alumni, phenomenon / phenomena, and kibbutz / kibbutzim. Outside nominal number, paradigm borrowing has been described for nominal case as well as for adjectival, pronominal and verbal morphology (Kossmann 2010; Seifart 2013). Borrowing of inflectional formatives appears to be correlated with the degree of structural congruence between the donor and recipient languages and is influenced by a range of ''cognitive, communicative, and sociocultural constraints'' (Gardani, Arkadiev and Amiridze 2015: 10).

These are some of the questions that arise from reading this richly-documented and thought-provoking work. A synthesis of several decades of original research based on large sets of data, the volume offers empirical, theoretical, and methodological insights and raises interesting new questions relevant not only to the distinction between borrowing and code-switching but also to a range of additional issues in language contact, bilingualism, and diachronic linguistics research.


Gardani, Francesco, Peter Arkadiev, and Nino Amiridze. 2015. Borrowed morphology: an overview. In Borrowed Morphology, Francesco Gardani, Peter Arkadiev, and Nino Amiridze (eds), 1-23. Berlin/Boston: De Gruyter Mouton.

Hafez, Ola. 1996. Phonological and morphological integration of loanwords into Egyptian Arabic. Égypte/Monde Arabe 27-28: 383-410.

Harris, James W. 1992. The form classes of Spanish substantives. Yearbook of Morphology 2: 65-88.

Heath, Jeffrey. 1989. From Code-Switching to Borrowing: Foreign and Diglossic Mixing in Moroccan Arabic. London/New York: Kegan Paul International.

Kossmann, Maarten. 2010. Parallel system borrowing: Parallel morphological systems due to the borrowing of paradigms. Diachronica 27: 459-487.

Seifart, Frank. 2013. AfBo: A world-wide survey of affix borrowing. Leipzig: Max Planck Institute for Evolutionary Anthropology. (Available online at, Accessed on 2018-04-24.)

Smeaton, B. Hunter. 1973. Lexical Expansion Due To Technical Change: As Illustrated by the Arabic of Al Hasa, Saudi Arabia. Bloomington, IN: Research Center for the Language Sciences, Indiana University.

Unbegaun, Boris O. 1947. Les substantifs indéclinables en russe. Revue des Études Slaves 23: 130-145.


Natalie Operstein is the author of ''Consonant Structure and Prevocalization'' (2010) and ''Zaniza Zapotec'' (2015) and co-editor of ''Valence Changes in Zapotec: Synchrony, Diachrony, Typology'' (2015) and ''Language Contact and Change in Mesoamerica and Beyond'' (2017). Her research interests center on language change, phonology, and language contact.

Page Updated: 12-Sep-2018