Introducing the Special Issue: Mixed Verbs and Linguistic Creativity in Bi/Multilingual Communities

In introducing this inaugural Special Issue for the open access journal Languages, it is important to understand the concept of linguistic creativity and how this relates to code-switching (henceforth CS)1, a common practice in bi/multilingual communities. It is essential to discuss this pertinent issue, as in the extant language contact literature, linguistic creativity is a phenomenon that has been argued to be especially associated with the emergence of mixed languages [1] and Creoles [2] but not necessarily CS.

In the formal study of intra-sentential CS, linguistic creativity has not played a prominent role given that for most scholars, CS primarily entails the combination of two autonomous grammatical systems that remain discrete in bilingual speech (for relevant discussion, see Gardner-Chloros and Edwards [3]). This suggests that CS is restricted to pre-existing structures and/or patterns in the component languages, and any degree of hybridity is envisioned as indexical of convergence. While this certainly reflects some bilingual contexts, it does not adequately account for multilingual contact situations where CS not only potentially involves more than two languages (see Bhatia and Ritchie, [4]), but where it is inclusive of hybrid structures and/or patterns that evince innovation and language change (see papers by Balam [5] and Yip and Matthews [6]).

In recent sociolinguistic work, several researchers advocate for a reframing of language and bilingualism that shifts away from monolingual frameworks and perspectives (see, for instance, the work of García [7]), and there is increasingly more interest in investigating and understanding the complex, hybrid language practices of bi/multilinguals. Thus, it is only apt we study how linguistic creativity relates to the dynamic CS interactional practices of bi/multilingual populations, especially in their use of mixed verbs, which have been less researched, compared to language mixing in the nominal domain. By mixed verb, we refer to a verbal complex where the lexical/content verb, drawn from one language, is integrated into the utterance through the mediation of a semantically light verb from another language. The former provides the lexical meaning while the latter hosts the grammatical information (e.g., tense and agreement inflections). In the literature, such mixed verbal complexes are commonly described as bilingual compound verbs, bilingual complex verbs, or light verb constructions, frequently involving the light verb ‘do’, as shown in the Hindi-English utterance in (1).

1.	merii	patnii	[vp saaRii	choose]	kar-egii
	my	wife	saree	choose	do-Fut.3.Sg.Fem
	‘My wife will choose a saree’.		Bhatia and Ritchie [4] (p. 6)

We were initially drawn to the study of these mixed verb constructions as they allow us to examine not only how bi/multilinguals seamlessly integrate their languages but how they creatively abstract away from pre-existing structures and/or patterns to devise novel ones. We know that a distinctive characteristic of mixed verbs is that they often evince structural features that cannot be easily attributed to one language or another; hence, revealing the flexibility, dynamism, and creativity of CS [3,8,9,10].

We take linguistic creativity to be the dual employment of language-specific and universal principles and mechanisms to devise innovative structures in bi/multilingual speech that may or may not have pre-existing equivalents in adult grammars. Linguistic creativity, therefore, brings to the fore the pivotal role that bi/multilinguals play in building switch sites in order to fully exploit the tools that are available in their linguistic repertoires. Particularly because the mixing of languages evidences an underlying “optimization” strategy at work, the bi/multilingual mind’s engagement in the creative use and manipulation of structure and meaning [4,11,12], may pave the way for contact-induced language change.

The present project on mixed verbs comprises seven articles, which collectively elucidate our understanding of the synchronic and/or diachronic dimensions of mixed verbs among different bi/multilingual populations across the world. In so doing, they explore (to varying degrees) key theoretical issues on the development and use of mixed verbs, and thereby contribute to our understanding of the creativity underlying language mixing in the bi/multilingual mind.

In his contribution [13], Pieter Muysken draws upon areal and universalist perspectives in his analysis of the wide-spread occurrence of bilingual compound verbs in the bilingual varieties of the languages spoken in the area between Sri Lanka and Greece. He proposes that Turkic and Indo-Iranian, two large and important language families in the geographically contiguous region, already had a productive [noun + verb] compound construction which grammaticalized (in bilingual varieties) to the [verb + verb] pattern under conditions of language contact. Muysken further posits that mixed verbs constitute a bilingual universal of adult second language acquisition and not child first language acquisition. He suggests, therefore, that since native [noun + verb] constructions very easily develop into [verb + verb] constructions in bilingual contexts, though not in monolingual ones, the locus of cross-generational evolution of mixed verbs is language use among adults rather than children.

Margaret Deuchar and Jonathan Stammers [14] explore the controversial distinction between CS and borrowing, which in previous discussions in the literature has largely centered on nouns (for relevant discussion, see Bhatia and Ritchie [15]). As verbs constitute the nuclei of bilingual clauses due to their encoding of relevant lexical, semantic, and grammatical information, it is vital to examine this distinction in relation to mixed verbs. Deuchar and Stammers contribute to the discussion on the status of mixed verbs by examining English-origin verbs in Welsh speech. Based on their findings, they argue that the distinction between CS and borrowing is not categorical. In their quantitative analysis, they show that linguistic integration as operationalized by the application of mutation on verbs is related to both frequency of usage and “listedness”2 (i.e., inclusion in a dictionary). Thus, in their data, the more frequent a Welsh or English-origin verb was, the more likely it was to be mutated. Furthermore, Welsh and English-origin verbs that were listed versus those that were unlisted were more likely to be mutated.

Two papers (one by Tej Bhatia and William Ritchie [4] and the other by Ji Young Shim [16]) examine the universal mechanisms underlying the syntax of mixed verbs. Bhatia and Ritchie emphasize the value of intuitional (i.e., grammaticality judgment) data in studying language mixing and propose that bi/multilingualism in the Indian subcontinent, where CS is a stable and time-tested phenomenon, is an ideal context to research the creative facet of the bi/multilingual mind. In their analysis, they consider a range of South Asian languages, focusing in particular on Hindi-English. Drawing upon the intuitions of competent bilingual speakers, they demonstrate that direct insertion of an English verb into a Hindi utterance is blocked unless a dummy or light verb kar ‘do’ is used for morphological integration with the verbal inflectional endings, and that formal syntactic principles govern the well-formedness of the structural derivation in such language mixing.

They propose that the light verb construction is an innovation in the bilingual grammar, as the [verb + verb] pattern is permitted only in code-switched Hindi-English verbal complexes where the lexical verb and the light verb are drawn from different languages. In their view, language contact is a prerequisite (though not a sufficient condition) for the emergence of this bilingual innovation. Bhatia and Ritchie also discuss experimentally elicited intuitional data gathered from second language learners with less than one year of exposure to Hindi and no prior experience with CS (i.e., a classic case of the poverty of the stimulus) to provide further evidence in support of their claims that knowledge of the syntactic mechanisms within Universal Grammar underlying the use of bilingual verbs is innate.

Drawing on experimentally elicited intuitional data from adult early bilinguals in New York City, Ji Young Shim [16] examines word order variation in Japanese-English and Korean-English CS with a focus on the verbal complex in light verb constructions. She argues that the realization of either Object-Verb (OV) or Verb-Object (VO) word order in the code-switched verb phrase in Korean-English and Japanese-English, contrary to previous accounts, does not depend upon the head directionality of the language that the light verb is drawn from. Instead, she proposes that the word order variation in mixed verb constructions in these typologically similar language pairs stems from the parametric variation of the functional category (v in Japanese and Korean versus ASP(ect) in English) that is lexicalized by the light verb.

The findings of Shim’s study highlight the importance of the distinction between heavy versus light verbs as well as the syntactic flexibility of the verb phrase in accounting for word order variation in CS. Shim proposes that bilingual speakers by virtue of their knowledge of different languages have access to a wider set of functional categories that may vary in their morphological realization. Thus, in relation to the light verb construction, bilinguals are privileged (in comparison to monolingual speakers) in that they can draw upon the additional resources available to them to generate a more diverse array of creative patterns.

Papers by Osmer Balam [5] and Tridha Chatterjee [17] examine the innovative morphosyntactic and lexico-semantic evolution of mixed verbs. Through the analysis of naturalistic speech data from Northern Belize Spanish-English CS, Balam [5] demonstrates that the most productive and complex use of mixed verbs, as evinced through syntactic verb type, is attested among adult bi/multilinguals whereas the most simplistic use of these structures is found among adolescents. His data suggest, therefore, that the emergence and propagation of novel mixed verb forms take place among post-adolescents and adults rather than adolescents, which supports Pieter Muysken’s suggestion [13] that mixed verbs constitute a bilingual universal of adult second language use.

In her analysis of film data from three different decades, Tridha Chatterjee [17] explores the evolution of mixed verbs in Bengali-English CS. Her analysis reveals a drastic increase in the use of Bengali-English mixed verbs in films produced in the 2010s versus those of the 1970s and 1990s. Importantly, Chatterjee also finds that in films from the 2010s, mixed verbs exhibit greater semantic productivity, as mixed verbs are used not only when speaking about academic- and work-related topics but when speaking about frequently occurring activities in informal contexts as well (e.g., the family setting). These notable changes in the use of mixed verbs reflect an increase in societal bilingualism and the more frequent practice of CS among younger Bengali-English bilinguals in India, especially in urban areas.

In their article, Virginia Yip and Stephen Mathews [6] analyze the code-mixed utterances of nine Cantonese-English simultaneous bilingual children in Hong-Kong. Focusing on the use of English verbs in Cantonese utterances, Yip and Mathews show that the children’s rate of mixing mirrors the rate of mixing in the parental input, and that different input conditions (one-parent/one-language versus one-parent/two- language contexts) also impact code-mixing. Similar to adults, the Cantonese-English bilingual children use Cantonese aspect markers with English verbs and insert English verbs in Cantonese utterances without the mediation of a light verb. Yip and Mathews attribute the use of this strategy, which Pieter Muysken [13] terms as “direct insertion”, to the typological properties of Cantonese and its congruency with English. At the same time, however, the bilingual children’s use of split verb particle constructions (as opposed to the non-split verb particle order) goes well beyond the input, in affirmation of their linguistic creativity.

In sum, the previous studies underscore the breadth and depth of concomitant issues on the use of mixed verbs that contributors address in their work. This Special Issue, to our knowledge, is the first attempt to bring together a collection of articles that draw on a variety of data sources to specifically investigate mixed verbs and linguistic creativity in diverse language contact situations. This project highlights the element of linguistic innovation and flexibility that may accompany the emergence and use of mixed verbs in bi/multilingual speech. In this regard, CS is a creative process in which speakers are not restricted to pre-existing structures in their languages. Across generations, however, while novel forms or uses of mixed verbs may develop, it may also be the case that typological factors may lead to the non-occurrence of bilingual light verb constructions, as in the case of Cantonese-English CS. In future work, it is imperative that we investigate other important questions that remain under-researched to better understand mixed verbs. We suggest a few areas where future research would be promising.

Most of the papers in this volume focus on bilingual varieties. Two exceptions are the papers by Balam [5] and Bhatia and Ritchie [4] who also discuss language mixing in trilingual contexts. Further research on mixed verbs in tri/multilingual contexts is crucial to help us fully understand how multiple languages work in the human mind. There is the possibility, for instance, that linguistic creativity involving mixed verb phrases may be more complex in cases of multilingual language mixing. As Bhatia and Ritchie [4] observe, a potential challenge to their account of the bilingual light verb construction is posed by the trilingual Tamil-English-Hindi intuitional data reported by Usha Lakshmanan [18]. In her personal communication with Bhatia and Ritchie, Lakshmanan points out that in the case of trilingual varieties of these languages, as shown in the contrasting examples in (2a-b), the Tamil lexical verb stem cannot be inserted directly into the Hindi verbal complex containing the light verb kar but must take the English causative suffix –ify for morphological integration with the Hindi light verb.

2.	Hindi-Anglicized Tamil; (Matrix language-Hindi)
a.	main	tanni	kudicc-ify	karuungii
	I	water	drink.VBP-Eng suffix	do-FUT.1FEM.SG
	‘I will drink water’.
b.	* main	tanni	kudic	karuungii
	I	water	drink	do-FUT.1FEM.SG
	‘I will drink water’.
c.	* main	tanni	kudicc-uungi
	I	water	drink.VBP-FUT.1.FEM.SG
	‘I will drink water’.		Lakshmanan [18]

The fact that the Hindi tense and agreement endings cannot be directly attached to the Tamil lexical verb stem, as (2c) illustrates, lends support to Bhatia and Ritchie’s proposals that formal syntactic principles govern the well-formedness of the structural derivation in such language mixing. At the same time, inserting the Hindi light verb kar alone without the mediation of the English suffix –ify would not result in grammatical language mixing, thus highlighting the potential for variability in mixing with light verbs depending on the properties of the languages involved. For example, it would be relevant to compare bilingual speakers of say Tamil and Hindi with trilingual speakers of Tamil, English, and Hindi to determine to what extent the structure of the mixed verbal complex is simply the result of trilingualism per se or a result of cross-linguistic differences, language dominance, and/or other sociocultural factors.

The contribution by Yip and Mathews is the only study featured in this collection that investigates mixed verbs in simultaneous bilingual children. Research on bilingual verbs has focused largely on adult bilinguals. As Yip and Mathews report, the Cantonese-English bilingual children in their study, similar to adults, favored direct insertion of an English verb into a Cantonese utterance without the use of a light verb, which they attribute to typological and congruency factors. Further longitudinal studies on mixed verbs in the developing grammars of simultaneous bi/multilingual children exposed to typologically different languages is needed to help disentangle the role of innate universal mechanisms from other factors such as language experience and language proficiency. Crucially as well, while bi/multilingual children’s code-switched speech may exhibit novel verb structures or patterns not present in adult grammars, longitudinal research must be carried out to investigate whether such innovations endure the test of time and become part of children’s developing grammars, or whether they are abandoned after a certain age. Such investigation could further contribute to our understanding of cross-generational CS innovation among child, adolescent, and adult bi/multilingual populations.

Other issues deserving of attention in future work include research comparing the processing of mixed verbs (compared to mixed nominals and/or monolingual verbs) by children and adults, studies of the use of mixed verbs in CS by bi/multilinguals with language disorders such as Aphasia and dementia, and the challenges posed by language mixing in the verbal domain for automated bilingual parsing and translation. Last but not least, it is crucial to comparatively examine the nature of linguistic creativity across generations in contact situations where mixed verbs are overwhelmingly used to incorporate other-language verbs versus those where other strategies (i.e., absence of light verbs) are the norm. This work would contribute to our understanding of linguistic innovation and language change in relation to differential switching patterns in the verbal domain.

In closing, we hope that the papers in this collection will spur renewed interest in the topic of mixed verbs and linguistic creativity in bi/multilingual speech. In her well-known metaphor, Valdés-Fallis [19] (p.126) notes that “It is helpful to imagine that when bilinguals code-switch, they are in fact using a twelve-string guitar, rather than limiting themselves to two six-string instruments”. It is important to highlight, however, that because of the potential for linguistic creativity, CS entails more than simply using a unified combination of strings. Using this musical metaphor, we can view CS as a dynamic process that may entail the skillful adaptation of strings to produce novel types of sounds, which may be more innovative in terms of resonance in cases of multilingualism, where speakers have an even richer linguistic repertoire. In bringing out this Special Issue, we make a call for further research that focuses on complex language contact situations, in particular contexts where multilingualism has thrived for generations. It is from these CS contexts, where this sociocultural practice is not stigmatized but tacitly accepted and embraced by its speakers, that we still have much to learn in relation to how mixed verbs evolve both across time and space.