1. Preliminaries

In linguistics today, stating that one does contact linguistics draws no more curiosity than saying that one studies, for instance, sign language. Anybody working on some aspect(s) of language contact should know that the statement is not informative enough, because, as we show in Section 2, there is a wide range of topics that are subsumed by this umbrella label. Literally, as well explained by Weinreich (Reference Weinreich1953) and Cohen (Reference Cohen1956), language contact is used in reference to the coexistence of languages in the mind of a speaker/signer and/or to their coexistence in a social or geographical space.

To be sure, no contact really takes place in the latter cases if segregation is so rigid that nobody in the coexistent populations of speakers learns the other ethnolinguistic group’s language. In reality, from a macroecological perspective, language contact at the population level obtains when there is at least one individual that learns the other group’s language and can spread its features in his/her group. Usually more than one “dispersing individual” are involved, driven by some reason to interact with members of the other community, though many other members of the relevant populations may not engage in such interactions. What is relevant is that language contact at the population level presupposes language contact at the level of at least some individuals. There are numerous population structures in which, owing to differences in economic and/or political powers between the coexistent ethnolinguistic groups, at least some members of one or both groups are expected to learn the other community’s language, such as in European colonies during the last few centuries (e.g., those who worked as intermediaries, interpreters, or colonial auxiliaries; see Samarin Reference Samarin1989, Lawrence, Osborn & Roberts Reference Lawrence, Osborn and Roberts2006, Mufwene Reference Mufwene2008, Reference Mufwene, Adamou and Matras2020a, Van den Avenne Reference Van den Avenne2017).

Contact linguistics is about various facets and consequences of the coexistence of languages in individual speakers’ minds and in particular populations. Instances of these include language shift at the level of either the individual or a population. A consequence of this may be language endangerment and loss (LEL), if most or all the members of the relevant population speaking a particular heritage language stop using it and thus converge in shifting to one that they either consider more advantageous to them or are forced by various ecological pressures to speak more often (e.g., Fishman Reference Fishman1971, Reference Fishman1991, Mufwene Reference Mufwene and Stanlaw2020b). The outcome of this convergence of behaviors can be compared to the emergence of new norms in a particular language in a contact ecology, when speakers converge in selecting variants (forms or structures) that come from another language they or some of them also use.

Such feature adoptions have generally been referred to as transfers or borrowings, depending on the scholar’s perspective. On the other hand, the convergences also produce norms even in monolingual populations, out of the contacts of different dialects or simply out of the contact of different idiolects, as speakers negotiate tacitly about features of their respective varieties. One can thus conclude that the fundamental contact is enabled by interacting individuals, at the level of idiolects, native and/or non-native; whatever else happens at the population level is produced by convergence (Mufwene Reference Mufwene2001). Labov (Reference Labov1972, Reference Labov2001) explains this well in showing how outliers sometimes bring into their social networks linguistic features they learned in another network. It takes a central or influential network member that the outlier interacts primarily with to copy the feature from him/her and spread it to other members (Eckert Reference Eckert1989). That is, the latter copy it from the central or influential member, or from other copiers, rather than from the outlier (now the “dispersing individual”) who brought it to their network. This is basically how change occurs, with some individuals innovating or introducing (different features of course) and others copying and spreading them (Croft Reference Croft2000, Mufwene Reference Mufwene2008, Fagyal et al. Reference Fagyal2010).Footnote ¹ Thus, we can argue later that contact is an important actuator of language change.

This approach can also help us explain why some language varieties are less, or more, “focused” than others (using LePage & Tabouret-Keller’s Reference LePage and Tabouret-Keller1985 terminology). For example, foreign-workers’ interlanguages differ from pidgins, at least the expanded ones that we know today, because they lack communal norms, although the speakers produce some of the same features in their interlanguages (Perdue & Klein Reference Perdue and Klein1992, Perdue Reference Perdue1995). The reason for this state of affairs is that (allowing some over-simplification) several migrant workers live in segregated groups in which they socialize in either their heritage language or some other language of their countries of origin while their children learn the local language from native speakers. Interlanguages remain individual-speaker phenomena, which the migrant workers (adults) produce only when they interact with people that do not speak the language(s) they are more competent in. By contrast, pidgins are, according to the received doctrine, communal varieties arising primarily from the regular interactions of non-native speakers with their trading partners and among themselves in the trade language to which they have had limited exposure.

Through their interactions, speakers influence each other – accommodating each other on different features – and converge toward some group or communal norms (despite some natural variation in the emergent system). The ecologies in which the migrant workers’ interlanguages emerge (Pfaff Reference Pfaff1981, see below) are not conducive to the emergence of communal norms, as they do not form a community with their changing interlocutors with whom they communicate in the host country’s language. Although the occasional accommodations made by fluent speakers to their deviations from the target language – a kind of foreigner talk at the workplace – may reinforce the departures at the individual level, they do not at some communal level. Communal norms emerge and can give rise to new varieties in contact ecologies in which the speakers interact or socialize frequently with one another, accommodating each other’s forms and structures.

As much as some literature on (naturalistic) L2 acquisition has focused on interlanguages, Mufwene (Reference Mufwene2010) concluded that this scholarship can only make a limited contribution to research on the emergence of creoles, because it is focused on individuals, whereas genetic creolistics deals with communal norms as the outcome of convergence. The same can be said of contact dialects. To be sure, one can learn about transfer and substrate influence (such as about possible trajectories of linguistic influence) but not about how some substrate elements attested in some interlanguages have converged into substrate influence in a creole, while some others do not.Footnote ²

Consistent with Braj Kachru’s (Reference Kachru, Quirk and Widdowson1985) distinction between the “Inner,” “Outer,” and “Expanded Circles” of World Englishes, the presence of communal norms helps distinguish the Englishes of the Outer and Expanded Circles. Because English generally functions as an official language in the Outer Circle – which consists of former British exploitation colonies – and the national elite have embraced it as an emblem of their social status and often socialize in it, national or regional norms have emerged in different parts of the world. This practice has generated what outsiders can identify as Indian, Singaporean, South African, and Nigerian norms, among other varieties misidentified as “indigenized Englishes.”Footnote ³ These norms are the outcomes of competition and selection in their respective communal (national or regional) feature pools, in which individual speakers accommodate each other and converge toward shared local, national, or regional features (Mufwene Reference Mufwene2001). The use of English in the Expanded Circle, on the other hand, does not lead to the emergence of national or regional norms. For instance, Japanese or German speakers of English in, respectively, Japan or Germany do not normally speak English among themselves.Footnote ⁴

The study of language contact at the population level entails an additional level of complexity that has usually been overlooked in the literature, which we all should be aware of, although this dimension is generally also missing from this Handbook. The omission appears to be a legacy of the beginnings of linguistics, with historical and genetic linguistics, in which languages were conceived of as organisms, therefore without internal, inter-idiolectal variation (without their social life). It is also a consequence of the coexistence of the research area with theoretical linguistics, so central to linguistics to date, where it is generally assumed (in fact since de Saussure Reference Saussure1916) that communal features are shared by all individual speakers. Numerous grammars have been published that are based on work with one or a couple of consultants! We have generally not compared idiolectal systems with their communal counterparts, which are typically constructs of convenience, in order to check, for example, whether they are isomorphic in their details. Such isomorphism would be contrary to what the study of interlanguages suggests. What is the status of features that remain idiosyncratic of particular idiolects and are not shared by all the members of the community of practice?

Despite progress in the scholarship on L2 acquisition, studies of bilingualism and multilingualism are dominated by those focused on the population level.Footnote ⁵ Perhaps because urban youth “stylects” (see Section 3) have been treated as discourse phenomena (from the perspective of performance), the discussion of the relation between variation at the idiolectal level and variation at the communal level has not come up. We hope to make more sense of (some of) these questions in the next section, where we argue that research on language contact has evolved from being marginal in linguistics to acquiring a central position, with contact at the population or the individual level acknowledged as an actuator of language change. Thus, the challenge has been to articulate the extent to which contact dynamics (dis)favor the emergence of new communal norms. These apply to the emergence of contact varieties – including creoles, pidgins, other mixed varieties, new ethnolects, and interlanguages – and evidently also to convergence and divergence between languages. The social histories of populations in contact, including the roles that multilingual speakers play in their respective ecologies, draw attention to the complexity of language change and of the social aspects of linguistic behavior. In the next section, we provide a selective synopsis of how language contact as a research area has evolved since the nineteenth century.

2. A Historical Survey

The study of language contact has been part of linguistics since the late nineteenth century, at variance with what is now known as genetic linguistics. The latter was then preoccupied with the uniparental representation of language speciation in especially the Indo-European family, and indeed also with the classification of languages into families and subfamilies such as Romance and West-Germanic. The comparative method of the nineteenth-century Neogrammarians has remained the backbone of the research area to date, though it is now enriched with computational modeling techniques. The Stammbaum, now also identified as cladogram, continues to serve as the proud demonstration of the success of that particular scholarship.Footnote ⁶ Lexical borrowings were excluded from the comparisons. No question was asked as to why language contact was considered a spoiler of what otherwise should be perfect, recursive speciation processes corresponding to different time depths. Deviations from the regularities of change revealed by the comparative method, such as “sound laws,” were accepted as exceptions, as long as they were not numerous enough to question the accuracy of the relevant analyses.

The inconvenience of language contact became more obvious in the late nineteenth century, when European philologists and other precursors of (genetic) linguists noticed how English, French, and Portuguese – and of course also Dutch and Spanish – had metamorphosed into creoles and pidgins (interpreted then as broken languages) in, respectively, the European plantation settlement colonies around the Atlantic and in the Indian Oceans and in trade colonies of West Africa and the Pacific. Both kinds of new language varieties were outcomes of how differently the non-European speakers of non-Indo-European languages reproduced the Indo-European languages they had learned. Contact was obviously the reason; and the question of the role of language contact in language speciation needed to be addressed.

However, the social ideologies of the time, which assumed purity of species and languages, disavowed these new colonial varieties of Western European languages as impure “bastard tongues” – as Derek Bickerton provocatively also referred to them in the title of his Reference Bickerton2008 book – or aberrations that were exceptional in the way they had emerged. With some exceptions, such as Hugo Schuchardt (see below), those who described them attributed the structural changes to the mental and anatomical inferiority of their presumably less evolved speakers.Footnote ⁷ This is the explanation then advanced by Adam (Reference Adam1883), Baissac (Reference Baissac1880), Vinson (Reference Vinson and Bertillon1882, Reference Vinson1888), Bertrand-Bocandé (Reference Bertrand-Bocandé1849), and Gonzales (Reference Gonzales1922). Then, genetic linguists apparently treated language contact as irrelevant to language speciation, generally ignoring Schuchardt’s (Reference Schuchardt1882) position that creoles and pidgins suggest instead how language diversification actually occurs. More specifically, Schuchardt attributed the current structures of modern languages to what Roger Lass (Reference Lass1997) calls “imperfect replication” by successive generations of their speakers, under contact conditions. Notwithstanding some oversimplification of the actual restructuring process on our part, substrate influence during language shift, for example, accounts for how Vulgar Latin had evolved into the Romance languages, with the “reproductions” incrementally diverging from the lexifier, as in the case of creoles.

Ignoring Schuchardt, genetic linguists dismissed creoles and pidgins by fiat as “mixed languages” and therefore anomalies by contrast to European varieties of Indo-European languages, all the way into the first half of the twentieth century.Footnote ⁸ Likewise, Hjelmslev’s (Reference Hjelmslev1938) contention that all languages are mixed to some extent got negligible attention. To date, there is still a category of “mixed languages,” whose definition varies according to author. For instance, Thomason (Reference Thomason2001) offers a less restrictive interpretation, which includes creoles – unlike some creolists who do not fit them in this category – although she is still at variance with Hjelmslev. Language contact is the reason why it is still widely assumed that creoles do not belong in the same genetic families as their lexifiers (Thomason & Kaufman Reference Thomason and Kaufman1988).Footnote ⁹ One must wonder why no natural account has been presented for why, for instance, all the Indo-European languages do not have the same morphosyntactic template that they would have inherited from Proto-Indo-European and the grammars of the Romance languages do not represent a continuous rectilinear or unilinear evolution from that of Vulgar Latin. Yakov Malkiel (Reference Malkiel1978) is among the exceptions in invoking contact to explain differences between Western and Eastern Romance languages.

The first half of the twentieth century saw the development of structuralism in Europe, with Ferdinand de Saussure (Reference Saussure1916) showing a greater interest in synchrony than in diachrony and developing more focus on structure/system (langue in French) than on speech (parole in French), although he situated much of the actuation of language change in the latter. The same period also saw the development of descriptivism in Anglophone North America, also mainly focused on synchrony, which was prompted largely by the endangerment of Native American languages. Anticipating that they would soon vanish, North American linguists endeavored to describe these languages.

The linguists were also in a situation where, unlike their European counterparts, they could find so many non-Indo-European languages within their own borders and were sensitive to the negative impact that the dominant European languages brought by the colonizers exerted on their vitality. Notwithstanding this, Sapir (Reference Sapir1921) acknowledged “individual and communal variation in English” (p. 157), as well as “linguistic interinfluencing” (chapter 9) between the languages of neighboring populations, where influence tended to be in one direction, from the most dominant culture to the other. Bloomfield (Reference Bloomfield1933: 476) also noted that speakers adapt their speech habits to those of their interlocutors, where the conditioning factors are the “density of lines of communication” and “the relative prestige of [the] social groups” (p. 345) (Fagyal et al. Reference Fagyal2010: 263).

Both structuralist and descriptivist approaches to language yielded several descriptions of non-European languages, with the “system” as its object. No significant study of language contact emerged until the middle of the twentieth century, about the same time Noam Chomsky (Reference Chomsky1957) would revolutionize the field with generative linguistics. Although a turning point in modern linguistics, this new trend has also remained focused on structure/system (a static framework, Bailey Reference Bailey1973: 34–5) and reiterated commitment to study competence and the faculty of language (or Universal Grammar – corresponding to de Saussure’s faculté de langage) over performance (corresponding to de Saussure’s parole). The generative approach helped cement the view that internal factors were sufficient to achieve rigor in linguistic explanations.

It was at this critical moment, in the mid-twentieth century, that Charles Ferguson, Joshua Fishman, Einar Haugen, and Uriel Weinreich, among others, regenerated interest in language contact, albeit in different though related ways. While very much interested in the Chomskyan Revolution, Uriel Weinreich revived the study of bilingualism, identifying the individual, more specifically the speaker’s mind, as the locus of language contact, especially through his study of interference. He proposed distinguishing between three types of bilingualism: coordinate, compound, and subordinate. This terminology has since been used in studies of both second language acquisition and societal bilingualism. The research, which culminated in his book Languages in contact (Reference Weinreich1953), would also establish the connection between language contact and language change (a dynamic framework, Bailey Reference Bailey1973: 34–5), embedding both in the social context of speakers’ interactions and highlighting “the interrelations between language history and culture history” (Bleaman Reference Bleaman2017). This is reflected in the following famous statement of his: “In speech, interference is like sand carried by a stream; in language, it is the sedimented sand deposited on the bottom of the lake. The two phases of interference should be distinguished” (Weinreich Reference Weinreich1953: 11).

The distinction Weinreich made between individual and societal bilingualism led him to pay more attention to various social factors that govern the functional distribution of the coexistent languages in their speakers’ lives, viz., which communicative domains are associated exclusively or primarily with which particular language (see also Ferguson Reference Ferguson1959). Indirectly, he also advocated for what we are calling “communal norms” when he wrote: the “impact of interference phenomena on the norms of a language may be greater if the contact occurs through groups of bilinguals” (Weinreich Reference Weinreich1953: 3).

In collaboration with Marvin Herzog and William Labov, he drew attention to the actuation of language change, situating this not only in language-internal variation but also in language contact, a facet of the social context of language practice. He characterized this approach to language change as speaker-oriented. Anticipating one interpretation of uniformitarianism in linguistics today, their seminal essay, titled “Empirical foundations for a theory of language change” (Reference Weinreich, Labov, Herzog, Lehmann and Malkiel1968), argues that “the same social and linguistic factors that condition linguistic variation today also account for historical change,” as paraphrased by Bleaman (Reference Bleaman2017). Perhaps this explains partly why variationist sociolinguists like talking about “language variation and change” and even have a journal named this way.

However, the phrase is problematic from an evolutionary perspective, as variation provides materials on which evolution works – in the case of language, through interactions among idiolects, i.e., inter-idiolectal contacts. However, variation is not necessarily indicative of change in progress. Nonetheless, Weinreich, Labov & Herzog (Reference Weinreich, Labov, Herzog, Lehmann and Malkiel1968) were correct in arguing that “finding order and structure” in “linguistic heterogeneity” (which they also characterized as “orderly heterogeneity,” p. 100) can say something about the “course of [language] development” (p. 99). The order was to be found in the “social and stylistic determinants” of heterogeneity that regulate what speakers do.

Einar Haugen developed an interest in bilingualism among Norwegian immigrants to the USA and its impacts on their heritage language. Embedding (American) Norwegian and its speakers in the population structure in which English is the dominant and economically more powerful language, he went on to study its attrition. This was at a time when, as noted by Baran (Reference Baran2018), European immigrants who arrived after the American Revolution were not immediately recognized as American and all people of European descent were not yet united under the label of “White Americans.” With his The Norwegian language in America: A study in bilingual behavior (Reference Haugen1953), Haugen pioneered the study of language endangerment and loss (LEL) in North America. In his subsequent book Bilingualism in the Americas (Reference Haugen1956), he distinguished languages depending on their social history: native, colonial, immigrant, or creolized.

Haugen also distinguished what he considered true language loss from language mixing. The former is produced by language shift, while the latter is a consequence of attrition in the heritage language and represents language change. Rejecting the nineteenth-century ideology of language purity, he would have thus disagreed with those creolists according to whom creoles coexisting with their lexifiers are dying by “decreolization” (e.g., Hazaël-Massieux Reference Hazaël-Massieux1999), i.e., the process by which speakers substitute acrolectal features for basilectal ones. (We return to this topic below.) Note that both language shift and language mixing can be identified in, for instance, Nancy Dorian’s (Reference Dorian1981, Reference Dorian2010) work on the obsolescence of Sutherland Gaelic and certainly also in the endangerment and loss of the languages of immigrants from other nations in European settlement colonies since the fifteenth century, where the language of a particular nation has become the dominant one (e.g., Clyne Reference Clyne2003 for Australia; Brown, in press, for the United States).

Haugen’s investigation of language practice among Norwegian immigrants led him to develop, among other things, seminal scholarship on borrowing, which he characterizes as process rather than outcome. According to him, the latter can be any number of loan forms or structures and does not rule out congruence of particular forms or structures between the languages in contact. His article “The analysis of linguistic borrowing” (Reference Haugen1950) is a classic that any student of borrowing who is eager to figure out whether they are innovating or working in the footsteps of this pioneer must read. He situates borrowing in the behavior of bilingual speakers and is critical of the metaphor of “language mixture,” which can be misunderstood as a “concoction” coming out of a “cocktail shaker,” i.e., out of an unconstrainted combination of various ingredients (p. 211).

Partly anticipating theories such as Carol Myers-Scotton’s (Reference Myers-Scotton1993) frame analysis in terms of “matrix” and “embedded” languages, Haugen (Reference Haugen1950: 211) argues that speakers always speak one language while resorting to the other language for assistance. This position is apparently corroborated by today’s students of urban youth language varieties in sub-Saharan Africa. For Haugen, the phenomena misidentified as “mixed” or “hybrid language” are particular consequences or outcomes of language contact and are principled, obeying specific structural constraints (the kinds of things that genetic creolists have been struggling to capture in various accounts of the emergence of creole vernaculars).

Haugen privileged the historical fold of borrowing over its synchronic one according to which one could identify loan forms and structures with certainty, without invoking the history of the language. It is this historical take that led him to develop his “ecology of language” (Reference Haugen1971), which evolved into a collection of papers put together under the title The ecology of language (Reference Haugen1972) by Anwar S. Dil. He expanded a biology-inspired approach to language change, including LEL, that had been pioneered a few years earlier by C.F. Voegelin, F.M. Voegelin, & Noel W. Schutz, Jr. in Reference Voegelin, Voegelin, Schutz, Hymes and Bittle1967. In an article titled “The language situation in Arizona as part of the Southwest culture area,” the latter discuss, among other things, the role of boarding schools in the endangerment and loss of Native American languages. The schools constituted ecologies of intense language contact where the students were forced to speak only the colonizer’s language. All these pioneers direct attention to local interactional dynamics within specific population structures in which speakers evolve in order to account for how the structures and/or vitality of their languages change.

This is the right juncture to also mention Pour une sociologie du langage (Reference Cohen1956) by Marcel Cohen. To be sure, the author anticipated sociohistorical linguistics – aka historical sociolinguistics – more than he did any of the present-day developments in the study of language contact. However, his emphasis on embedding language change in its sociohistorical context, in which the population structure plays a central actuator role, is a significant contribution to the scholarship on language contact as the field developed from the works of Uriel Weinreich and Einar Haugen in particular. The same sets of ethnolinguistic groups coming in contact under different ecological conditions will not produce the same thing.

Also, changes in socioeconomic structures can trigger changes in terms of how one group speaks the other’s language, and whether in the first place its members remain “invested” in the language. Among several examples, Marcel Cohen cites the situation of Jews in colonial French North Africa. The weakening of segregation between the Jews and the Arabs led the former to speak Arabic in ways more similar to its heritage speakers’. However, the rise of French as the new language of formal economy motivated them to become more invested in French than in Arabic. Marcel Cohen also highlighted similarities between language contact and dialect contact – now further developed by Peter Trudgill (e.g., Reference Trudgill1986, Reference Trudgill2004) – which should lead us to realize, as we noted in Section 1, that the fundamental level of contact in language is between idiolects, regardless of whether they are native or non-native (Mufwene Reference Mufwene2001). One may also note similarities between Cohen’s typology of various ecologies of language contact with the sociology of language developed by Joshua Fishman (discussed below).Footnote ¹⁰

In the mid-1960s, Joshua Fishman (Reference Fishman1968a) enriched the scholarship on language contact with insights on the complexity of multilinguals’ behaviors in communities where members share similar repertoires of the same languages, although their competences in each and/or preferences for one or the other vary to some extent. In an effort to better understand language maintenance and language shift, he highlights the fact that factors such as situation, domain, topic, and interlocutor influence their language choices at particular speech events, consistent with research on the ethnography of speaking in monolingual communities by linguists such as Hymes (Reference Hymes, Gladwin and Sturtevant1962) and Ervin-Tripp (Reference Ervin-Tripp1964). Like Ferguson in dealing with “diglossia” (see below), Fishman (Reference Fishman1972) directed attention to the fact that one language is usually preferred for informal interactions while another is required in formal settings, which may include one’s professional life. Independent of diglossia, he highlighted the fact that a speaker may also feel more at ease discussing a particular topic in one language than another, owing to, for instance, the fact they were trained on or have experienced the topic in the language they currently prefer to use, that in which they have developed the best command of the relevant terminology or idiomatic phrases.

Nonetheless, one may also “violate” this division of labor when interacting with particular multilingual individuals in the same community, to whom they feel socially close, even to the point of mixing codes, as explained in Fishman’s (Reference Fishman1965a) article “Who speaks what language to whom and when?” Fishman was already anticipating the fascination of linguistic anthropologists such as Michael Silverstein (Reference Clyne2003) and his students and colleagues with the indexical role of language and, mutatis mutandis, of particular registers, sociolects, topolects, and other varieties.

The focus on the complexity of the multilingual behavior also led Fishman to shed light on the gradual and non-uniform way in which a language dies. If the opposition in status can be cast in terms of socially weak vs. strong or dominant languages (viz., demographically, economically, or politically), particular patterns of social interaction can lead speakers to bring the dominant language into domains associated with the weak one, or from the public sphere into the private domain, such as the home. For instance, because members of the family use the same repertoire, some speakers may find it unnecessary to switch back to the heritage language at home. Fishman also showed that language shift at both the individual and population levels does not occur abruptly and wholesale but rather gradually and piecemeal, from one domain to another, affecting some individuals before others, until it affects the whole community.

In another article, “Bilingualism, intelligence and language learning” (Reference Fishman1965b), Fishman dispelled the myth that multilingual speakers are more intelligent than monolinguals. He did likewise with another myth that claimed the existence of a state of perfect multilingualism in which no language mixing occurs. This was in reaction to the distinction between “coordinate” and “compound bilingualism” mentioned above, which was then gaining ground. If one thinks about it carefully, phenomena such as interference/transfer and borrowing would be impossible in speakers whose bilingualism was coordinate, as there would be no osmosis between coexistent languages.

Fishman evolved to become a strong defender of linguistic diversity and even wrote about the possibility of reversing language shift (Reference Fishman1991), thereby underscoring the significance of ecological factors in maintaining or giving up a language. Nonetheless, we cannot ignore the fact that in “Sociolinguistics and the language problems of developing countries” (Reference Fishman, Fishman, Ferguson and Gupata1968b), he presented multilingualism as a problem for nation-building and political unity in the newly independent nations of Africa. In this line, Stewart (Reference Stewart and Fishman1968a) also presented his widely cited “A sociolinguistic typology for describing national multilingualism.”

Following Fishman, among others, language economists such as Ginsburgh & Weber (Reference Ginsburgh and Weber2011) have interpreted linguistic diversity in the developing world as linguistic “fragmentation,” aka “fractionalization” (Alesina et al. Reference Alesina, Devleeschauwer, Easterly, Kurlat and Wacziarg2003). For them, this stands in the way of economic development, a position at odds with linguists’ advocacy for sustaining linguistic diversity and affording every child an opportunity to be schooled in their heritage or ancestral language, especially if this is also their mother tongue.

To be sure, Fishman did not argue against linguistic diversity. He described the difficulties that societally multilingual developing nations were facing in having to choose between, on the one hand, maintaining the colonial language as the official language, although it is not accessible to large segments of the national populations, and, on the other, promoting one of the many indigenous languages as the official language and thereby (potentially) disadvantaging speakers of the other languages. One must wonder why only the Western model of nation building (with a nation united by one national/official language) had to be considered – though Fishman must be praised for paying attention to practical problems associated with language maintenance and economic development.

In papers such as “Nationality-nationalism and nation-nationism” (Reference Fishman, Fishman, Ferguson and Gupata1968c) and “National languages and languages of wider communication in the developing nations” (Reference Fishman1969), Fishman highlights the diglossic relation that has been obtained between the colonial languages that continue to function as official languages and the indigenous languages. In his view, (rigid) diglossia has contributed to preventing the displacement of indigenous languages by the colonial or dominant language, despite some current language policies that have favored the use of official languages in schools. He thus anticipated some “heretics” such as Mufwene (Reference Mufwene2017a) who argue that the population structures of former European exploitation colonies, in the Global South – which fosters socioeconomic segregation between the Natives and the colonizers and now the masses of the population and the ruling elite – is not conducive to the endangerment of indigenous languages by so-called killer languages (see below), for instance English and French.

Contributions to the scholarship on language contact really flourished in the mid-twentieth century. Following Nikolai Trubetzkoy (Reference Trubetzkoy1923, Reference Trubetzkoy1939), Murray Emeneau (Reference Emeneau1956) argued for language convergence, aka “Sprachbund” and “language/ linguistic area,” this time in South Asia. He focused on shared features of Indic and Dravidian languages that are the outcomes of language contact, in a way similar to the sharing of structural features among the Balkan languages (Sandfeld Reference Sandfeld1930). These studies and many more since then highlight the role of population movements and contacts, as well as the ensuing population structure, in the rise of areal features. They are actuators of convergent language change – with two or more coexistent but genetically or typologically different languages evolving shared structures – as is evident from some contributions to this volume, especially those by Alexandra Aikhenvald and Thiago Costa Chacon on the Americas and by Victor Friedman on the Balkans. As is also evident from Gumperz & Wilson (Reference Gumperz, Wilson and Hymes1971), another pioneer classic essay, areal features show how languages can wind up sharing forms and structures not inherited from a common ancestor, simply through mutual influences enabled by language contact. Recent book-length publications on the subject matter in Europe and Africa include Heine & Kuteva (Reference Heine and Kuteva2005), Heine & Nurse (Reference Heine and Nurse2008), Drinka (Reference Drinka2017), and, in South America, Cerrón Palomino (Reference Cerrón Palomino1994). The relevant contributions to this Handbook and its companion to follow just show how current the subject matter has remained in dealing with regional formal and structural affinities among languages.

Still in the mid-twentieth century, John Gumperz (Reference Gumperz and Lunt1964) also pioneered the study of code-switching. Among other things, he explored a distinction between “code-mixing” and “code-switching.” This has of course generally been ignored in the literature since the 1980s, in which “codeswitching” has become the default umbrella term for both; the distinction has apparently been harder to sustain at the level of discourse. Likewise, his attempt to show that creoles are particular outcomes of language-mixing has generally not been referred to, at least not in creolistics, although, based on the late-nineteenth-century ideology of language purity, creoles have been assumed to be mixed languages (as noted above in this section). Assuming an essentially synchronic perspective, Gumperz argued, more successfully, that code-switching and, more generally, language mixing among plurilinguals are normal and principled or rule-governed behavior.

A speaker can alternate deliberately between languages for all sorts of reasons, including identifying interlocutors in their narrative, quoting particular characters, or highlighting particular attitudes (e.g., sarcasm, comical effect, attitude toward the interlocutor). Gumperz did not, of course, exclude the possibility that a speaker may have to use as the “matrix language” (Myers-Scotton Reference Myers-Scotton1993) one in which he has less competence than another or has less ease in discussing a particular topic, such as when he/she is conversing with interlocutors who are not familiar with the language that he/she is more competent in. In sub-Saharan Africa, where the ideology of language purity is hardly embraced, this accounts for the mixed-code nature of utterances of speakers who are more competent in the European official language of their country than in an indigenous lingua franca spoken in some region of the same polity (Kamwangamalu Reference Kamwangamalu2000). It is also common when such a speaker has acquired a particular knowledge through formal education in the European colonial language but must convey relevant information in an indigenous vernacular not associated with the domain.

Gumperz also initiated the search for constraints in code-switching, which Carol Myers-Scotton, in collaboration with especially Janice L. Jake (Myers-Scotton & Jake Reference Myers-Scotton and Jake2016), just like Shana Poplack (Reference Poplack1980, Reference Poplack2017), would elaborate in several publications to date. Auer (Reference Auer1999) presents a more nuanced conception of what code-switching involves, especially regarding whether the grammars used in code-switched utterances are the same as the juxtaposition of those of monolingual speakers of the relevant languages. Anthologies such as Stell & Yakpo (Reference Stell and Yakpo2015) enrich the scholarship with diverse perspectives, including the psychological and social.

Gumperz certainly deserves more credit for underscoring the social and contextual functions of code-switching (see also Eerdmans, Prevignano & Thibault Reference Eerdmans, Prevignano and Thibault2003). This is apparently one of the aspects of code-switching on which the now growing scholarship on translanguaging (e.g., Williams Reference Williams2002, Garcia & Li Reference Garcia and Li2014, Reference Garcia, Li and Chapelle2018, Lewis, Jones & Baker Reference Lewis, Jones and Baker2012) has capitalized, especially in the classroom context.

The scholarship on language contact at the discourse level has also evolved into the study of urban youth language varieties, marked by switching and mixing, on which this volume includes a short discussion, in Alamin Mazrui’s chapter, regarding Sheng and Engsh in Kenya. Based on the grammar of a particular indigenous urban vernacular, such as Nairobi Swahili, Kinshasa Lingala, or Cameroon Pidgin English, the youth varieties have been characterized as audience-driven and context-dependent “stylects,” to display the speaker’s/performer’s “authenticity” rather than command of a particular communal norm. They are best produced at the street corner where the relevant youth socialize, rather than in private spaces, especially those including non-group members.

Students of youth language varieties (also identified as “youthspeak”), which are not recognized as separate languages or dialects, argue that the mixed varieties have an indexical role. Their speakers – who are otherwise disenfranchised economically from the elite and affluent white-collar speakers of European colonial languages – claim their speech indexes modern urban culture, by contrast with traditional ways of speaking the related indigenous languages (Mesthrie et al. Reference Mesthrie2021). Commenting on Sheng, Githiora (Reference Githiora2018: 1) states that “The speech code exists on a continuum of ways of speaking Swahili within a complex and stratified multilingual society in search of a modern identity.” Authenticity in the singularity of one’s utterances seems to carry more weight than the use of some stable norm, as their producers keep innovating with competing lexical synonyms. These performances appear to be consistent with Peter Auer’s (Reference Auer1999) interpretation of language mixing as associated sometimes with group identity.

While such varieties have been studied especially in sub-Saharan Africa and Europe, one is prompted to also consider varieties such as Spanglish, Heblish, and Media Lengua, although the latter has disputably been associated with relexification since Muysken (Reference Muysken, Highfield and Valdman1981).Footnote ¹¹ It is otherwise evident that language contact, through code-switching, code-mixing, “fused lects” (according to Auer’s Reference Auer1999 typology), or otherwise, plays a role in language speciation. In the case of at least sub-Saharan African urban “stylects,” contact apparently just expands the continuum of the variation between the languages in contact, producing new, ever-changing synonyms (Mesthrie et al. Reference Mesthrie2021). The study of language contact has thus expanded to include a greater focus on creative performance, at the discourse level, far beyond what Uriel Weinreich’s and Einar Haugen’s interest in interference considered as deviations from a specific (native) norm.

We did not intend to provide a comprehensive survey of studies of language contact in this section. Our goal was simply to highlight some ways in which the mid-twentieth century was pivotal to the expansion of research on language contact, diverging from the focus of structuralist and descriptive linguistics on language structures, chiefly phonology and morphology, until the Chomskyan Revolution made syntax the central concern of “formal linguistics.” The same impetus sparked renewed interest in the emergence of creoles and pidgins, starting with publications such as Lorenzo Turner’s (Reference Turner1949) Africanisms in the Gullah dialect and Robert Hall, Jr.’s (Reference Hall1958) “Creole languages and genetic relationships.”Footnote ¹² In a way these were also rejoinders to work produced by Haitian scholars Suzanne Sylvain (Reference Sylvain1936) and Jules Faine (Reference Faine1937), who had concluded, respectively, that Haitian Creole was Ewe relexified with French vocabularyFootnote ¹³ or that it was a new Romance language. The momentum, marked by several other publications by Robert Hall, Jr. (Reference Hall1958, Reference Hall1962, Reference Hall1966) on Haitian Creole and Melanesian Pidgin, and some studies tracing modern creoles back to some hitherto unattested Portuguese pidgin that had putatively evolved from the Mediterranean Lingua Franca (Thompson Reference Thompson and LePage1961, Whinnom Reference Whinnom1965), led to the publication of the seminal volume edited by Dell Hymes titled Pidginization and creolization of languages (Reference Hymes1971).

It was simply assumed, not without controversy, that new languages had emerged under contact conditions created by European plantation and trade colonies – in which non-Europeans were the overwhelming majority – and that they were not genealogically related to their lexifiers, because they had very different grammars.Footnote ¹⁴ In the legacy of the nineteenth century, the emphasis was then on the role of substrate influence or of the language bioprogram – acting in the minds of children that putatively transformed their parents’ broken pidgins into creoles – in accounting both for their structural divergence from their lexifiers and for structural similarities among them.

In addition to the question of how creoles and pidgins emerged, Pidginization and creolization of languages determined many of the issues that have preoccupied creolists to date, including whether creoles are defined by a particular set of structural features that are unique to them, as Creole Exceptionalists continue to claim. In addition, the book made it clear that the terms creole and pidgin had been extended to also include “contact languages” that were not lexified by European colonial languages. The book includes chapters on Mbugu, Lubumbashi Swahili, Chinook Jargon, and convergence between Indo-Aryan and Dravidian languages in India. This inclusion followed the publication of Status and use of African lingua francas by Bernd Heine in Reference Heine1970, which includes a long list of languages, some of which have since been counted as creoles or pidgins, such as Lingala and Sango. Along with other publications, these works would make it necessary to make some distinctions that are current today, such as not confusing “intertwined languages” and cases of “Sprachbund/convergence” with creoles and pidgins.

Also, the significance of inter- and intra-idiolectal variation in identifying what was misnamed “post-creole continuum” started with Pidginization and creolization of languages, which includes a seminal chapter on the subject matter by David DeCamp (Reference DeCamp and Hymes1971). Note, however, that the decreolization hypothesis (picked up by Bickerton Reference Bickerton1973 and Rickford Reference Rickford1987, among others), with which it was too hastily associated, can be traced back to Hugo Schuchardt (Reference Schuchardt1914). It had also inspired Robert Hall, Jr.’s (Reference Hall1962) hypothesis of creoles’ “life-cycle.” The basic idea was that if a creole continued to coexist with the acrolect of its lexifier, it would substitute many of the latter’s features (grammatical and lexical) for its basilectal counterparts – thus, it would “debasilectalize.”

Based on his study of Gullah, Mufwene (Reference Mufwene and Morgan1994) disputes the historical validity of this “decreolization” hypothesis. This was not a rejection of the continuum nor of the useful terms basilect, mesolect, and acrolect, which Stewart (Reference Stewart and Shuy1965) coined to capture it. What he rejected instead was the claim that the mesolect was the outcome of contacts between the basilect and the acrolect. He argued instead that the basilect and the acrolect are idealized poles of the continuum captured by the mesolect, and that the continuum had existed since the early stages of the speciation of the new vernaculars from their lexifiers. There just was no “post-creole” continuum; creoles had always existed as continua, as noted by Mervyn Alleyne (Reference Alleyne1980). As a matter of fact, the continuum obtains everywhere a distinction is made between a standard variety and nonstandard varieties connected by a mesolect, often characterized as colloquial speech. Much of the variation on which language evolution operates is to be found in this continuum.Footnote ¹⁵

Studies of African American Vernacular English (AAVE), especially by William Stewart (Reference Stewart and Shuy1965, Reference Stewart1968b) and J.L. Dillard (Reference Dillard1972, Reference Dillard1977), also fueled debates on the role of substrate influence to explain the divergence of its grammar from that of American standard English. (Ironically, standard English was definitely not its lexifier, just like in the case of English creoles!) It was assumed that AAVE had originated in an erstwhile Gullah-like creole spoken earlier by enslaved Africans on the rice fields and cotton and tobacco plantations and that this ancestor could ultimately be traced back to a West African Pidgin English. (This position is disputed in Mufwene Reference Mufwene and Lanehart2015.) Marked by several polemics, the scholarship on the emergence of creoles and AAVE has thrived, generating many more publications than on any other nonstandard dialect of European languages in the ex-colonies.

The exception to the above generalization can be cited from the scholarship on the indigenization of English in the former British exploitation colonies of Africa, Asia, and the Pacific, which prompted us to invite contributions on the spread of Western European languages in both the settlement and exploitation colonies for this Handbook. These include the chapters by John Lipski on the Romance languages (a comparison between Europe and the Americas), J. Clancy Clements on Portuguese, Robert Papen on the contact of French and English in North America, Cécile B. Vigouroux on French in Africa, and Edgar Schneider & Sarah Buschfeld on English.

In the case of English, the catalysts in the trend regarding what is now known as “World Englishes” (a term that applies to all modern English varieties) are Braj B. Kachru, Yamuna Kachru, and Larry Smith’s joint effort to expose what has been essentially a race bias (like in the case of creoles!) in disenfranchising the Englishes of the Global South as “new,” “nonnative,” and “indigenized.” The terms suggest inaccurately that these particular varieties are deviations or anomalies from the normal evolutionary process followed by the varieties spoken by the European-majority populations in the colonies. “Indigenized Englishes,” spoken in the Global South, would putatively be the only ones to have changed and diverged significantly from the British norm by appropriation by non-heritage speakers. It is for a good reason that Kachru (Reference Kachru2017) chose “World Englishes” as the umbrella term that could show all modern English varieties as evolving and new. At the same time he highlights the political and economic power hierarchy among the English national and regional varieties with the opposition (from Kachru Reference Kachru, Quirk and Widdowson1985) between the “Inner Circle” (which includes all countries of the Global North, where English is spoken as a vernacular), the “Outer Circle” (which includes all former British exploitation colonies, where English is the High variety and functions as an official language), and the “Expanding Circle” (which applies to the remaining countries that have adopted English as a lingua franca for communication with the outside world).Footnote ¹⁶

The 1980s were pivotal to the emergence of World Englishes as a research area, with two journals – World Englishes and English World-Wide – as well as a book series, English Around the World. They all bridged well with the publication in, and thanks to, creolistics of the journals Etudes Créoles, the Journal of Pidgin and Creole Languages, and the Journal of Language Contact, as well as the book series Creole Language Library, Cambridge Approaches to Language Contact, and Brill Studies in Language Contact and Dynamics of Language. A four-volume Atlas of pidgin and creole language structures has also been published and is now also accessible online (Michaelis et al. Reference Michaelis, Maurer, Haspelmath and Huber2013), in the spirit of The world atlas of language structures (Haspelmath et al. Reference Haspelmath, Dryer, Gil and Comrie2005, Dryer & Martin Reference Dryer and Haspelmath2013), The world loanword database (Haspelmath & Tadmor Reference Haspelmath and Tadmor2009), and South American indigenous language structures (Muysken et al. Reference Muysken, Hammarström, Krasnoukhova, Müller, Birchall, van de Kerke, O’Connor, Danielsen, van Gijn and Saad2016), reflecting the interest of the Max Plank Institute at Leipzig in both language typology and language contact.

The 1980s are also marked by Sarah Thomason and Terrence Kaufman’s book entitled Language contact, creolization, and genetic linguistics (Reference Thomason and Kaufman1988), which promoted a comparative approach to the study of language change. The book advocated for “the importance of extralinguistic alongside of linguistic factors in the study of language change (advocated by Schuchardt in the 19th century and in Weinreich Reference Weinreich1953)” (Escobar Reference Escobar2008: 200–1). By directly addressing contact phenomena, Thomason and Kaufman brought to the forefront the intersection of contact linguistics, sociolinguistics, dialectology, and historical linguistics. We return to this below.

By contrast, the 1990s and early 2000s will be remembered for the renewed interest in the LEL of “Indigenous languages,” to be interpreted narrowly and accurately as those Native to former European settlement colonies but more broadly as “non-European,” with a dangerous Eurocentric bias. If one takes into account the fact that the late wave of the Indo-European expansion over the past half-millennium had nothing to do with the same negative processes in China and other parts of the world, this evolution must be associated generally with population movements and language contact in ecologies in which one economically (and/or politically) powerful population subjugates the other(s).

An advocacy movement to stop or reverse language shift, the immediate cause of LEL, has followed a workshop organized by Kenneth Hale and Michael Krauz at the annual meeting of the Linguistic Society of America in 1991, the proceedings of which were published in a 1992 special issue of Language on the subject matter. Interestingly, unlike in the early twentieth century, when the interest was in describing the dying Native American languages, this time the focus was on sensitizing linguists to what was presented as a disaster for humanity in the form of loss of linguistic and cultural diversity, like in the case of endangered species.Footnote ¹⁷ The advocates initially laid the blame on “killer languages,” apparently ignoring everything that Einar Haugen and Uriel Weinreich had said about the causes of LEL. Publications such as Nettle & Romaine (Reference Nettle and Romaine2000) and Crystal (Reference Crystal2000) just focused especially on the powerlessness of “Indigenous” people in the face of the bulldozing effect of European colonial languages, especially English. In the early 2000s, UNESCO brought together specialists to implement a methodology for assessing language vitality and endangerment in the world. This led to the online UNESCO interactive atlas of the world’s languages in danger (Moseley Reference Moseley2010).

As often pointed out by Mufwene (e.g., Reference Mufwene, Filipovic and Pütz2016, Reference Mufwene2017a), the contribution to understanding how language contact can negatively affect the vitality of some coexistent languages – and under what conditions – has been marginal. (However, see Bradley & Bradley Reference Bradley and Bradley2019 for a healthy change.) The omission of any comparison of the loss of “Indigenous languages” with that of competing European languages in the settlement colonies, precisely what Einar Haugen has focused on in The Norwegian language in America (Reference Haugen1953), may be part of the explanation for this shortcoming.

However, we cannot ignore that LEL has become a productive research area that fits squarely in language contact and has the natural potential to contribute to a broader, more inclusive understanding of language evolution. The latter research area deals not only with the emergence and loss of structures but also with maintenance and loss of vitality, aside from the traditional concern with the emergence of languages and their speciation in evolutionary linguistics (Mufwene Reference Mufwene, Fill and Benz2018). A research area called “ecolinguistics” (Mühlhäusler Reference Mühlhäusler2003) is now thriving, with the shortcoming that it is more moralizing about the importance of maintaining the relevant ecology without explaining how or why. To be sure, the recently published Routledge handbook of ecolinguistics, edited by Alwin Fill & Hermine Benz (Reference Fill and Benz2018), includes also topics in the tradition of Einar Haugen and beyond.

Other developments have also marked the expansion of the scholarship on language contact since the second half of the 2000s, including the study of “super-diversity.” In Reference Vertovec2007, Steven Vertovec published an issue of Ethnic and Racial Studies titled “Super-diversity and its implications.” He discussed the situation brought about in British (and other Western European) cities by “an increased number of new, small and scattered, multiple-origin, transnationally connected, socio-economically differentiated and legally stratified immigrants who have arrived over the last decade” (1024). To be sure, Lesley Milroy & Peter Muysken (Reference Milroy and Muysken1995) had anticipated Vertovec in referring to “the influence of migration from the Third World on language practices in the industrial West,” which they connect to the interlanguages of guest/migrant workers in Western Europe studied in the 1970s by, for instance, Wolfgang Klein & Norbert Dittmar (Reference Klein and Dittmar1979).

Regarding interlanguages, research conducted by Carol Pfaff (e.g., Reference Pfaff1981) shows that there are many factors that account for why foreign workers in Europe have not been able to learn the local language fluently. They include residential isolation from – and hence no socialization with – the host population; concentration of speakers of the same language in the same dormitories; little communication with native speakers at work, especially when this takes place through interpreters; and sometimes negative attitudes of members of the host population to the foreign workers. Population structure, especially regarding permeability between the host and foreign populations, affects how much of the local language they and their children can learn. Regarding the linguistic structures the foreign workers produce, we also learn that substrate influence and foreigner talk (used by autochthonous speakers) is only part of the story (Meisel Reference Meisel and Felix1980). There’s always more to learn about various aspects of language contact, which, as an aspect of human social and mental behavior, appears to involve more complexity than some may have imagined.

Returning to super-diversity, several papers, led especially by linguistic ethnographers Blommaert & Rampton (Reference Blommaert and Rampton2011), have been published that underscore the unprecedented multilingualism that has arisen in Western European urban centers as a consequence of foreign immigrations. As shown by Mufwene (Reference Mufwene2017c), the interest in this aspect of language contact generally lacks a historical perspective. The relevant scholars have generally overlooked ways in which European colonization had already also changed the linguascapes of the conquered territories, not only by introducing European languages but also with the addition of languages spoken by contract laborers brought to the colonies, such as in Fiji, Tanzania, and South Africa, without overlooking creole-speaking colonies (transformed earlier ethnolinguistically) such as Mauritius, Guyana, and Trinidad. It also changed the colonies with the production of new urban centers where, in exploitation colonies, even indigenous people from diverse ethnolinguistic backgrounds came to coexist with one another, which fostered the emergence of indigenous urban vernaculars and regional lingua francas.

Generally, we can now also learn a great deal about the role of economic power and costs of long-distance communication in determining, among other things, which languages survive and for how long. These are certainly factors that bore on the loss of European languages other than the dominant ones in, for instance, Anglophone North America, Brazil, and Australia. May they not also shed light on why Biblical Hebrew survived only in the written form in the Jewish Diaspora? At the same time, the Diaspora produced so many language varieties, called “Jewish languages” by Bernard Spolsky (Reference Spolsky2013), which have been lexified by languages other than Hebrew. This Handbook includes a chapter on the subject matter by Marie-Christine Bornes-Varol & Anne Szulmajster-Celnikier. Dirk Hoerder & Henry Yu complement the latter study with two chapters on how diasporas emerge from population movements and introduce or sustain linguistic diversity under specific population structures. In some cases, they foster linguistic areas.

Along with the study of multilingualism associated with super-diversity in Western European cities came interest in the way immigrants’ children speak the host populations’ languages. Reflecting the neighborhoods in which they live, they include autochthonous nonstandard features and elements from their parents’ heritage languages. French scholars have identified the French varieties of these allochthonous speakers as “les parlers/accents des banlieues” (‘suburban language varieties’, see, e.g., Fagyal Reference Fagyal2010), after the economically destitute neighborhoods, where large proportions of non-affluent immigrants live. On a larger scale, Jenny Cheshire & Penelope Gardner-Chloros’s (Reference Cheshire and Gardner-Chloros2018) “Multicultural London English and Multicultural Paris French” project studies youth varieties in immigrant populations in Paris and London to compare ways in which they are singular in relation to autochthonous norms. These immigrant varieties appear to reveal a great deal about the significance of population structure and economic affluence as determinants of the varieties targeted by the immigrants. Overall, this research area shows that, despite their stigmatization, the immigrants’ children learn the autochthonous sociolects quite faithfully, although, as noted by Lass (Reference Lass1997), there is really no perfect replication even in the acquisition of a mother tongue. That is, the immigrants’ children acquire much, much more of the host population’s language than they are credited for. They have simply been innovative in other ways, perhaps to index their social identities.

It is in a similar context that, starting with Ben Rampton (Reference Rampton1995), the study of the phenomenon identified as “crossing” has become fashionable. The allochthonous youth assimilate linguistically, but they also maintain certain ethnic peculiarities that are salient to their autochthonous peers. As they socialize also across ethnic lines, within their socioeconomic classes, the autochthons borrow some of these xenolectal ethnic peculiarities too. This observation shows how the study of language contact can help understand social behavior, with linguistic features functioning as ethnicity markers also revealing the extent to which autochthonous and allochthonous peers are open to mutual influences. Some of the xenolectal features may index age and patterns of socialization. Nonetheless, we must emphasize that, linguistically, the allochthonous children have assimilated more significantly than their parents and speak the local vernaculars of their communities fluently, if not natively. The features on which sociolinguists have focused are a small subset associated with their ethnic backgrounds.

The last fold of how language contact as a research area has expanded reconnects us with the late nineteenth century. Then, genetic linguists resisted Hugo Schuchardt’s position that the emergence of creoles, out of the contact of European languages with non-European ones – outside Europe! – were prompting them to revisit the subject matter of how current European languages had evolved out of population movements and language contact. This thesis would be resurrected in the 1970s by Charles-James N. Bailey & Karl Maroldt (Reference Bailey, Maroldt and Meisel1977) and Brigitte Schlieben-Lange (Reference Schlieben-Lange and Meisel1977), with the former arguing that Middle English was a creole and the latter that the Romance languages had evolved by the same restructuring processes, under contact conditions, like creoles.Footnote ¹⁸ Although these linguists of the late 1970s were generally also ignored by their colleagues, who stuck to what DeGraff (Reference DeGraff2003, Reference DeGraff2005) calls “Creole Exceptionalism,” a uniformitarian position on the emergence of creoles related to what Schuchardt’s view that has been championed by especially Chaudenson (Reference Chaudenson1992, Reference Chaudenson2001), Mufwene (Reference Mufwene2001, Reference Mufwene2008), Aboh (Reference Aboh2015), Aboh & DeGraff (Reference Aboh, DeGraff and Roberts2017). According to them, creoles have emerged and evolved by the same restructuring processes as other languages and do sometimes prompt us to ask contact-related questions about the evolution of some non-creole languages. We just need not identify as creole every other language (variety) whose emergence is associated with contact (and language shift), such as French, Spanish, Irish, and Amish English. Contrary views to the uniformitarian approach are summarized in publications by self-proclaimed exceptionalists such as Bakker et al. (Reference Bakker, Borchsenius, Levisen and Sippola2017) and McWhorter (Reference McWhorter2018).

Outside creolistics, we must recognize Bernd Heine & Tania Kuteva’s (Reference Heine and Kuteva2005) Language contact and grammatical change, according to which much of Western Europe is actually a linguistic area, owing to population movements and language contact in the region. They have been joined in this position by, among many others, Bridget Drinka, with her Language contact in Europe (Reference Drinka2017), in which the same position is supported with the study of the “periphrastic perfect.” Precedents to these developments can be found in Areal diffusion and genetic inheritance: Case studies in language change, edited in 2001 by Alexandra Aikhenwald & Robert Dixon. Noteworthy in this publication are especially the contributions by Calvert Watkins, Randy Lapolla, James Matisoff, and Bernd Heine & Tania Kuteva, who all underscore the significance of language contact as an actuator of change. Mufwene (Reference Mufwene, Fill and Benz2018) has gone as far as to conjecture that, from the perspective of the evolution of language (as the abstraction from the diversity of languages) since the dispersal of Homo sapiens out of Africa, population movements and language contact are critical to explaining how Proto-Indo-European and Proto-Bantu speciated (successively) into present-day “daughter languages.” The Stammbaum presents the outcome of speciation, not the process itself (Joseph & Mufwene Reference Joseph and Mufwene2008), contrary to what is suggested by the uniparental assumption that came along with it since its conception by August Schleicher in the mid-nineteenth century. The chapters in Part One of this Handbook reflect this bridge between the study of language contact and that of language evolution as approached by genetic linguists.

The study of language contact has definitely evolved in non-unilinear and non-rectilinear manners since the late nineteenth century. The range of research topics has increased in breadth and diversity; and the field has evolved from being peripheral to the study of the mechanical aspects of languages, diachronically and/or synchronically, to being one of the many legitimate and equally important ways of studying languages. A language contact approach grounds change in the emergent population structures, relating it to how speakers interact with each other across ethnolinguistic lines or other boundaries. We could not cover every aspect of the evolution of contact linguistics in this synopsis; we apologize for overlooking some other developments that some readers may have expected us to discuss too.

3. The Chapters

We would be lying if we claimed that we had perfectly planned which chapters to include in this Handbook when we invited contributions.Footnote ¹⁹ The preceding section, which exposes some of our omissions, was written only at the conclusion of the project, which had already grown so big that we had to split it into two thick books. Perhaps we will have another opportunity to fill the gaps with a third book. However, three things were clear on our minds at the conception stage: 1) we did not want to replicate literally the contents of the chapters in handbooks on language contact that had already been published; 2) we wanted a book that would reflect much of the recent trends in the scholarship on language contact, such as LEL, translanguaging, and super-diversity and societal multilingualism; 3) and we wished to underscore the role of language contact as an actuator of change, including language speciation. Some of these topics are covered in the companion handbook.

The diachronic orientation appeared so critical to us that we also decided to start this Handbook with the part on “Language Contact and Genetic Linguistics.” And we found it also fitting to lead this part of the Handbook with the chapter by Brian Joseph, “Language Contact and Historical Linguistics,” in which he explains how historical linguistics itself has evolved to recognize (to a larger extent) the relevance of language contact to its subject matter. This perspective is further substantiated with the chapters by Randy LaPolla, Bonny Sands, and Bridget Drinka, which also contribute geographical diversity and some complementarity toward the big picture. For those reading them in the same order as in the table of contents, we thought it would help to start with information from outside the Indo-European territory, contrary to the dominant tradition in linguistics. Doing things differently may help some readers ask questions about the speciation of Indo-European from the perspective of the emergence of language families elsewhere as they have been driven by population movements and language contact.

The breadth of historical linguistics itself as we interpret it explains the large size of this Handbook, considering the wide range of phenomena that one cannot make much sense of without a diachronic perspective. One may also conclude that, as research areas, historical linguistics and language contact are interconnected, as we hope this book makes obvious, although we still could not cover everything, as we explain above. We hope, nonetheless, that the reader will be happy with the wealth of information this volume provides.

We owe part of the wealth and diversity of the contents to the overlap, by design, between several chapters, whose foci are not necessarily the same. They provide richer perspectives on the areas of overlap, such as how and where linguistic areas have emerged (see, in particular, the chapters by Victor Friedman, Alexandra Aikhenvald, Thiago Chacon, and Hans Henrich Hock), whether lingua francas have emerged the same way everywhere (see the chapters by Hildo do Couto and by Nicholas Ostler in the companion handbook), and what impacts diasporas exert on language evolution, beyond spreading languages outside their homelands. In the latter respect, the reader is encouraged to go beyond the chapters by Dirk Hoerder & Henry Yu, Joseph Sung-Yul Park, and Sherman Lee, and to read the chapters by James Collins, Koen Bostoen & Hilde Gunnink, Alamin Mazrui, and Jonathan Owens.

For comparison’s sake, the reader is also exhorted to check the chapters on the emergence and spread of some European languages, by John Lipski, J. Clancy Clements, Robert Papen, Cécile Vigouroux, and Edgar W. Schneider & Sarah Buschfeld. This comparison may shed light on the difference between, on the one hand, language spread associated with migrations of heritage speakers who stick to their cultural traditions, regardless of whether or not they maintain their languages as their primary means of communication among themselves (such as in the Chinese, Indian, and Korean diasporas), and, on the other hand, language spread associated with the adoption (with or without shift) of a particular language (such as a European colonial language) by speakers of other ones. From the point of view of language evolution, the comparison should inform us about whether the changes undergone by the spreading languages are of different kinds, aside from the fact that heritage speakers may wind up giving up their language in the diaspora.

Bonny Sands’ chapter, “Tracing Language Contact in Africa’s Past,” definitely makes obvious the limitation of the comparative method in genetic linguistics, as kinship in forms and in structures can be attributed sometimes to borrowing, hence language contact, rather than only to inheritance from a proto-language – a position articulated earlier by, for instance, Meillet (Reference Meillet1900) and Tremblay (Reference Tremblay, Fussman, Kellens, Francfort and Tremblay2005). Africa can be looked at as a vast geographical setting that supported internal population movements, where successive kingdoms and empires grouped and brought in contact peoples speaking different languages, generating not just coexistence but sometimes spread, loss, or birth of some languages at different time depths – in addition to form and structural changes that have traditionally drawn most of historical linguists’ attention. Sands exposes so much that remains unknown and constitutes potential research areas in the future, at the intersection of contact and genetic/historical linguistics.

Likewise, Randy LaPolla explains, in “The Chinese Expansion and Language Coexistence in Modern China,” how successive migrations from especially the Yellow River to various directions and layers of cultural assimilation (involving language contact) wound up creating the political construct of Chinese people speaking a cluster of languages collectively identified as Chinese. A concomitant of this evolution was the emergence of minority populations speaking non-Sinitic languages, some of which are now considered as endangered. We found this chapter a natural candidate for the position after Bonny Sands’ chapter, as in both of them the history of population movements and layers of language contacts helps us understand the present state of coexistent languages from genealogical and typological perspectives, as well as from that of language vitality.

In the same vein Bridget Drinka and Koen Bostoen & Hilde Gunnink invoke successive migrations, from different origins, and contacts in order to account, with the former, for the gradual emergence of the Indo-European language family and differences between its subfamilies and, with the latter, for how contacts with the Khoi and San populations account for the speciation of the Bantu family into so many languages and subfamilies in both Central and Southern Africa. Drinka adduces genetics and archaeology to bear on her arguments, as do Bostoen & Gunnink, though to a lesser extent. The chapters underscore the significance of practicing interdisciplinarity to prove past contact. The reader should note that Drinka uses stratification with the meaning of “periodization.” Bostoen & Gunnink also discuss factors that brought some Khoi and San languages to extinction, including cultural assimilation to the dominant population – not only the Bantu newcomers – and genocides by European settlers.

Overall, the chapters in this Handbook show how far the study of language contact has evolved since the late nineteenth century and how much it has diversified, while enriching our understanding of human linguistic behavior, especially from a diachronic perspective. They highlight the social and psychological factors that influence or drive the relevant behaviors and show largely how we may not fully understand language change without looking into social change produced by population movements and contacts both within and across particular territories.