Forthcoming. Mixed languages

1 Mixed Languages1 Felicity Meakins & Jesse Stewart 1. Introduction Mixed languages2 are a type of contact language, which results from two or more languages combining in a situation of bilingualism (or multilingualism). They arise during times of significant social change, serving as an expression of a new identity or the maintenance of an older identity. Their dual linguistic parentage or 'genetic ambiguity' (Thomason & Kaufman, 1988) means that mixed languages cannot be classified genetically according to the Stammbaum descent model. Beyond this general definition, mixed languages are difficult to characterise on socio-historical or typological grounds. They form in different types of socio-historical situations. Speakers may be the descendants of migrants or groups who underwent colonial incursions; they may be the children of mixed marriages or the descendants of a ethnic community undergoing language shift. Similarly, there is no single typological profile of a mixed language. They fall into three categories roughly: (i) L(exicon)G(rammar) mixed languages, where one language provides the grammar and another language contributes large amounts of vocabulary, for example Media Lengua and Angloromani; (ii) structural mixes where both languages contribute significant amounts 1 Glossing abbreviations: ABL=ablative, ABS=absolutive, ALL=allative, ACC=accusative, BEN=benefactive, COMP=complementiser, CONV=converb, DAT=dative, DET=determiner, DIM=diminutive, DIR=directional, F=feminine, FOC=focus, FUT=future, GEN=genitive, ILL=illative, IMP=imperative, IMD=immediate, IMPF=imperfect, IRR=irrealis, LOC=locative, M=masculine, OBJ=object, OBV=obviative, NOM=nominative, PERF=perfect, PL=plural, POSS=possessive, PROG=progressive, PRS=present tense, PRT=preterit, PST=past tense, SG=singular, SP=Spanish origin, SS =same subject, TOP=topic, TR=transitive, VAL=validator. 2 Mixed languages are also called 'bilingual mixed languages' by Thomason (1997c), 'split languages' by Myers-Scotton (2003), 'fused lects' by Auer (1999), and intertwined languages by Bakker and Mous (1994). For consistency we use the term 'mixed language', which is the most widely used term. 2 of grammatical (and lexical) material to the new language, for example Gurindji Kriol and Michif; and (iii) converted languages where a language maintains its lexicon but undergoes structural convergence with another language, for example Sri Lanka Malay (Bakker, 2015). Historically, mixed languages were not considered autonomous language systems and were often dismissed as cases of code-switching, creolisation, or adstrate influence. They were brought to the attention of linguistics again by Thomason & Kaufman (1988). As a result, a number of edited volumes drew together substantial amounts of mixed language data (Bakker & Mous, 1994; Matras & Bakker, 2003b; Thomason, 1997d), and Bakker's (1997) A Language of Our Own provided the first detailed account of a mixed language: Michif. More recently, Michaelis et al. (2013) have provided sketches of some languages, and a number of monographs and edited volumes now provide detailed accounts of Ma'á (Mous, 2003b), Gurindji Kriol (Meakins, 2011b) and Sri Lanka Malay (Nordhoff, 2009, 2012). In addition, four substantial review papers, a bibliography and textbook chapter explore the theoretical implications of mixed languages (Bakker, 2013, 2015; Matras & Bakker, 2003b; Meakins, 2013b, 2016a; Winford, 2003, Ch 6). The notion of LANGUAGE AUTONOMY is important to the identification of a mixed language. It refers to the ability of the language to function as a stand-alone linguistic system with only minimal continuing input from its source languages (Bakker, 2003). Following Saussure ([1916] 1983: 86), the parts of a language must be “synchronically interdependent.” Thus changes in the source languages do not feed into the mixed language and vice versa. This level of autonomy is difficult to demonstrate given that there is often a close synchronic and diachronic relationship between mixed languages 3 and other mixing practices, for example code-switching. Many mixed languages, referred to as 'symbiotic mixed languages' (N. Smith, 2000), also exist alongside one or both of their source languages. Nonetheless a number of independent developments of the source languages or mixed language have been demonstrated, including Light Warlpiri (O'Shannessy, 2013), Gurindji Kriol (Meakins, 2012), Bilingual Navajo (Schaengold, 2003) and Sri Lanka Malay (I. Smith, Paauw & Hussainmiya, 2004). Other measures of autonomy have also been proposed, including the stability of the language and children targeting the language in acquisition. These criteria are discussed in detail in Meakins (2013b). Mixed languages can be contrasted with pidgin and creole languages, and codeswitching through a number of criteria. They are created in situations where a common language already exists and communication is not at issue, whereas pidgin and creole languages are born out of the need for communication between people from a number of language groups (Golovko, 2003: 191; Muysken, 1997b: 375). As a result, pidgin and creole languages are formed from (usually) one dominant lexifier that came in contact with a number of different languages, which contribute to varying extents to the grammar and phonology. By contrast, mixed languages have two clear sources. Pidgin and creole languages are also (arguably) the result of successive generations of second language learners targeting the lexifier language, rather than a situation of bilingualism, which is the case for mixed languages. Code-switching, on the other hand, is also found in bilingual contexts, however mixed languages show more stability, i.e. predictability, in the sites of switches and have developed new structures which are not reflected in either source language. 4 This chapter begins with an overview of languages which have been classified as mixed (§2) and presents representative case studies of a number of languages within a typological classification of mixed languages (§3). It then discusses their contemporary functions (§4.1), their socio-historical origins (§4.2) and linguistic processes (§5) which led to their genesis. Much of this discussion focuses on the lexicon and morphosyntax of these languages. §6 provides the first detailed discussion of the phonology of mixed languages. As will be shown, mixed languages originate from a range of socio-historical settings and linguistic processes which do not obviously predict the resultant shape of the language. 2. An overview of the descriptive literature on mixed languages The following table lays out examples of contact varieties, which have been labelled as mixed languages. Table 1 Languages which have been identified as mixed languages LANGUAGE COUNTRY ETHNICITY OF SPEAKERS MIX SOURCES Angloromani England Romani Grammar: English Lexicon: English and Romani Barranquenho Border of Spain and Portugal Bolivia Barranquenho Grammar and lexicon: Portuguese with some Spanish influence Lexicon: Puquina Grammar: Quechua (Bakker, 1998; Boretzky & Igla, 1994; Hancock, 1970, 1976; Matras & Bakker, 2003a; Matras, Gardner, Jones & Schulmann, 2008; Smart & Crofton, 1875; Thomason, 2001) (Clements, 2009; Clements, Amaral & Luís, 2008) Chindo Indonesia Domari Iran, Egypt, Palestine Australia Peranakan Chinese Dom Callahuaya (Kallawaya) Gurindji Kriol Callahuaya Travelling Healers Gurindji Lexicon: Malay Grammar: Javanese Lexicon and grammar: Indic and Arabic VP: Kriol NP: Gurindji Lexicon: mixed (Hannß & Muysken, 2014; Juárez, 1998; Muysken, 1994a, 1997a) (Matras & Bakker, 2003a) (Dreyfuss & Oka, 1979) (Matras, 1999, 2007, 2012) (McConvell, 2008; McConvell & Meakins, 2005; Meakins, 2007, 2008a, 2008b, 2009, 5 Island Carib (Igneri) 2010, 2011a, 2011b, 2011c, 2012, 2013a, 2015, 2016b; Meakins & Algy, 2016; Meakins, Jones & Algy, 2016; Meakins & O'Shannessy, 2005, 2010, 2012; O'Shannessy & Meakins, 2012) (Hoff, 1994; Taylor & Hoff, 1980) Island Carib men Grammar: Arawak Lexicon: some Carib Javindo Lesser Antilles (Caribbean) Indonesian Javanese mothers, Dutch fathers Grammar: Dutch Lexicon Dutch and Javanese (M. de Gruiter, 1994; V. E. de Gruiter, 1990) Jenisch Germany Jenisch traders (Matras, 2000, 2003, 2009) Lekoudesch Germany Jewish cattle traders Light Warlpiri Australia Warlpiri Grammar: German 3 Lexicon: Rotwelsch, Hebrew, Romani, Romance Grammar: JudeoGerman Lexicon: some Hebrew VP: Ab Eng/Kriol NP: Warlpiri Lexicon: nouns mixed, verb Kriol Ma'á Tanzania Mbugu Grammar: Bantu Core Lexicon: Cushitic Media Lengua Ecuador Quichua Lexicon: Spanish Grammar: Quichua Mednyj Aleut Bering Strait (Russia) Aleut VP (finite): Russian NP: Aleut Lexicon: Russian and Aleut Michif Canada Metis (French fathers, Cree mothers) VP: Cree NP: French (Grammar and Lexicon) 3 Rotwelsch is camouflaged German not an independent language. (Matras, 2000, 2003, 2009) (Meakins & O'Shannessy, 2005, 2010, 2012; O'Shannessy, 2005, 2006, 2008, 2009, 2011a, 2011b, 2012, 2013, 2016; O'Shannessy & Meakins, 2012) (Brenzinger, 1987; Mous, 1994, 2000, 2003a, 2003b; Myers-Scotton, 2003; Thomason, 1997a, 1997d; Thomason & Kaufman, 1988) (Deibel, accepted; Gómez Rendón, 2008; Jarrín 2014; Lipski 2016 Muysken, 1981, 1994b, 1997b; Myers-Scotton, 2003; Stewart, 2011, 2013, 2014, 2015a, 2015b, 2018a, 2018b) (Golovko, 1994, 1996; Golovko & Vakhtin, 1990; Myers-Scotton, 2003; Sekerina, 1994; Thomason, 1997b) (Bakker, 1994, 1997; Bakker & Papen, 1997; Gillon and Rosen, 2016, 2018; MyersScotton, 2003; Papen, 1987a, 1987b, 2003, 2005; Prichard & 6 (Bilingual) Navajo United States Navajo New Tiwi Australia Tiwi Old Helsinki Slang Finland Finnish and Swedish gangs Papiamentu Curaçao (Caribbean) West Africans Petjoh Indonesia Malayspeaking mothers, Dutch fathers Philippine Hybrid Hokkien Philippines Reo Rapa Rapa Iti (French Polynesia) Ireland Shelta (Lee, 1987; McConvell, 2002) (Ceniccola, 2014; de Smit, 2010; Jarva, 2008; Paunonen, 2006) Upper Guinea Portuguese Creole with some Spanish lexicon Grammar: Malay Lexicon: Dutch (Jacobs, 2012) Chinese Filipinos Lexicon: Hokkien/ Tagalog/ Eng Grammar: Tagalog/ Hokkien Nominal: Hokkien/ Tagalog Verbal: Hokkien Maybe neither GL or VP (Gonzales, 2018) (Gonzales, 2017; Gonzales, 2018) Rapa Old Rapa, Tahitian (both Polynesian) (Walworth, 2015) Grammar: English Lexicon: Irish Forms: Contact variety of Malay Grammar: Tamil and Sinhala (Grant, 1994) Forms: Contact variety of Portuguese Grammar: Tamil and Sinhala Lexicon: Austronesian Grammar: Waskia (Trans New Guinea) Tibetanized variety of Northwest Mandarin Sri Lanka Malay Sri Lanka Irish Travellers Malay Sri Lanka Portuguese Sri Lanka Portuguese Takia Karkar Island (PNG) China Takia Wutun Grammar: Navajo Lexicon: Navajo and English VP: Tiwi NP: Aboriginal English/Kriol Grammar: Finnish Lexicon: 80% Swedish Shwayder, 2014; Rhodes, 1977, 1986, 1987, 2001, 2013; Rosen, 2000 2003, 2006, 2007, Strader, 2014) (Schaengold, 2003) Tibetan Buddists (Giesbers, 1995; van Rheeden, 1994) (Aboh & Ansaldo, 2007; Ansaldo, 2005, 2008, 2011a, 2011b; Bakker, 2003; Nordhoff, 2009, 2012; Slomanson, 2006, 2007; I. Smith, 2003; I. Smith & Paauw, 2006; I. Smith et al., 2004) (Bakker, 2003; I. Smith, 1977, 1979a, 1979b, 1984, 2001) (Ross, 2001, 2006) (Chen, 1986; Janhunen, Peltomaa, Sandman & 7 with some Bonan (Mongolic) influences Dongzhou, 2008; Lee-Smith & Wurm, 1996; Sandman, 2012; Sandman & Simon, 2016) The table is meant as an overview of the literature rather than a definitive statement on the categorisation of these languages, because the status of a number of these languages is questionable. For example, Barranquenho, spoken near the Portugal/Spain border, is most likely a dialect of Portuguese with some Spanish influence, such as clitic placement, rather than a mixed language. In any case, Barranquenho is derived from very closely related languages so the extent of horizontal versus vertical transmission is difficult to ascertain. The distinction between creoles and mixed languages is also not always clear. Sri Lanka Malay is generally considered a mixed language, but has also been called a creole language by Smith & Paauw (2006). Conversely, Papiamentu is generally classified as a creole but has been reanalysed as a mixed language by Jacobs (2012). Bakker (2015) has an extensive discussion of these cases. 3. Typology of Mixed Languages Mixed languages are typologically diverse but can be broadly categorised as L(exicon)G(rammar) languages (§3.1), structural mixes (§3.2), or converted languages (§3.3). For more refined categorisations, see Bakker (2003, 2015). 3.1 L(exicon)-G(rammar) languages Most mixed languages exhibit a split between the lexicon and grammar with respect to the source language. Bakker (2003: 125) calls these L(exicon)-G(rammar) languages and lists 25 in a typological survey. L-G languages differ in whether the ancestral or introduced language provides the grammatical structure. Those which select their grammars from the introduced language include Angloromani, Javindo, Kallawaya and Ma'á, whereas those where the ancestral language provides the grammar include 8 Bilingual Navajo, Media Lengua, Old Helsinki Slang, Papiamentu and Wutun. Angloromani and Media Lengua are discussed in detail in §3.1.1 and §3.1.2. Just how much introduced lexical material is required to 'qualify' as a mixed language is unclear, because the proportion is never 100%. At the extreme end, 89% of Media Lengua's vocabulary derives from the introduced language, Spanish (Muysken, 1997: 378; Stewart, 2011: 37). The percentages are much lower for Angloromani but the use of two parallel lexicons distinguishes these languages from normal borrowing scenarios. The use of parallel lexicons also differs from code-switching because the speakers are not bilinguals, but only have control over a second limited set of words or stems. 3.1.1 Angloromani Angloromani is spoken by some Romani people in Britain and is considered endangered (Matras, 2010). Romani people continue to be a travelling population, many of whom live in caravans. Romani people who now live in permanent accommodation generally do not speak this mixed language (Matras 2010). Currently Angloromani is not the language of conversation, but rather is restricted to individual utterances. These utterances can be characterised as the use of a restricted set of Romani-derived lexicon, which Matras et al (2007) call a “lexical reservoir,” within an English grammatical frame. This lexical reservoir exists largely in parallel with English lexicon and is drawn on in situations where speakers want to mark a sense of solidarity or group cohesion. An example of the coupling of Romani-derived lexicon with English grammar is given in (1) and (2). These sentences contain Romani words inserted into an English frame, for example the nouns fowki 'people' and poshaera 'penny', as well as the pronouns lesti 'he' and mandi 'me' (note that these pronouns are etymologically locative forms; 9 however the case distinctions have been lost). The Romani-derived words are given in italics. (1) The poor fowki that haven't got a poshaera to their name! The poor people who don't have a penny to their name. (Matras et al., 2007: 115) (2) Lesti's laughing at mandi. He's laughing at me. (Matras, 2010: 114) Verbs and function words such as maw 'NEG' are also common, although Romani verb inflections are no longer used. Some Romani morphology remains, such as the genitive -engra suffix, which attaches to lexical roots to create a related word: masengra (from mas 'meat'). Matras et al. (2007) also observe that Angloromani speakers do not always use the definite article, aspect and existential auxiliaries, and co-referential pronouns in places where they would be expected in English. However, they argue that these features are not specifically Romani; they simply indicate that Angloromani has slightly different grammatical rules froms English. It is likely that Angloromani developed after the Romani had already shifted to English, as an attempt to reclaim their heritage language through the use of Romani words. However, Thomason & Kaufman (1988: 103-104) suggest that Angloromani is the result of the wholesale adoption of the English grammatical system coupled with the maintenance of lexical material from Romani. Romani has likewise fused with other European languages and evolved into other mixed languages (Carling, Lindell & Ambrazaitis, 2014). 10 3.1.2 Media Lengua An example of a mixed language which retains the grammar of the ancestral language is Media Lengua. It is spoken by an estimated 2000 people in several communities throughout the Ecuadorian highlands. Fundamentally, the language demonstrates a clear split between Quichua grammatical and Spanish lexical elements (Dikker, 2008; Gómez Rendón, 2005, 2008; Jarrín Paredes, 2014; Lipski, 2016; Muysken, 1980, 1981, 1997b, 2013a, 2013b; Stewart, 2011, 2014, 2015a, 2015b). In most cases, the Spanish lexical items take on the semantic roles of their Quichua counterparts while also conforming to Quichua phonology—a process known as relexification (Muysken, 1981). Word counts estimate the number of lexical replacements at roughly 89% based on a 200-word Swadesh list elicited by Muysken (1997b) and later by Stewart (2011). Percentages are higher when taking into account larger portions of the lexicon (93% as per Stewart (2015b)) based on spontaneous speech data). Deibel (accepted) also provides quantifiable experimental evidence in support of relexification as the foundation of Media Lengua’s formation. The morpho-syntactic frame of Media Lengua is essentially Quichua in origin and therefore agglutinating and predominantly following an SOV word order. Of the 63 grammatical elements found in Ecuadorian Quichua, Gómez-Rendón (2008: 68) identifies at least 49 (77%) while Stewart (2015b: 28) identifies 55 (87%). Nouns inflect for case (nominative, accusative, locative and other semantic cases) and number, while Spanish-derived gender agreement is rarely productive. Similarly, verbs are inflected for tense, aspect, person, and number agreement for subjects. Free pronoun forms are derived from Spanish, but typically conform to Quichua patterns, viz., there is a marginally 11 productive use of lexical gender and in/formal distinctions. Media Lengua also preserves phonological patterns of Quichua which have since changed in the region (e.g., the lack of post nasal stop voicing). The following examples demonstrate the pattern of Spanish stems with Quichuan suffixes (italicised). (3) Mio hermana-ka mio papa-su-ta terreno-man 1SG.POSS sister=TOP 1SG.POSS dad-DIM-ACC land-DIR compaña-shpa i-ju-n accompany-SS.CONV go-PROG-3 My sister accompanies my father to the plot of land. (Stewart, 2015b: 29) (4) Ahora=ka sobre vestimenta-mi dezi-gri-ju-ni uno now=TOP about clothing-VAL say-FUT.IMD-PROG-1S a poqu-ito-go-ta vestimenta=ka aqui bit-DIM.SP-DIM.Q-ACC clothing=TOP here nuestro pueblo Kayambi-pa=ka. Nosotros indigena-kuna-pa=ka our town Cayambe-GEN=TOP our anaco prens-ada-mi centro ahora=ka dezi-nchi. skirt press-PRT.SP-VAL centre now=TOP indigenous-PL-GEN=TOP say-1.PL Mas antes=ka, anaco dezi-na-mi ese more before=TOP skirt say-INF-VAL this our grandparent-PL ese anaco-kuna=ka de this skirt-PL=TOP dezi-n y mas say-3 and more of nuestro abuelo-kuna, lana de borrego-mi ka-shka wool of sheep-VAL be-PST pes-ado anaco-kuna… weight-PRT.SP skirt-PL 12 I'm going to talk a little bit about the traditional clothing of our people from Cayambi. Nowadays, our indigenous anaco4 skirts are called 'centre-pressed' skirts. In the past, it was said that our grandmothers' anaco skirts were made of sheep wool and weighed a lot more. (Stewart, 2013, p. 14) The two most studied dialects of Media Lengua are found in the province of Cotopaxi (Muysken, 1980, 1997b) and two provinces to the north in Imbabura (Gómez Rendón, 2005, 2008; Stewart, 2011, 2015b). It is thought that both developed through intense contact between Spanish and Quichua in the early 20th century (Gómez Rendón, 2005 for Imbabura; Muysken, 1997b: 374 for Cotopaxi), though the origins remain relatively elusive, especially in Imbabura. Stewart (2015b) notes that construction of the Ecuadorian railway between 1915 and 1929 may have brought Media Lengua speakers north to Imbabura. This can be seen in Cotopaxi surnames in the area and several linguistics similarities, which are difficult to dismiss as chance innovations. In Cotopaxi, Muysken (1997b) proposed that many young men started working in the construction industry in a nearby provincial town and learning Spanish. This was the group that created Media Lengua. Muysken claims that the genesis of this mixed language occurred because speakers did not fully identify with traditional Quichua culture or the urban Spanish culture. In Imbabura, on the other hand, Media Lengua speakers fully identify as Indigenous and not as Mestizos (J. Lipski, p.c., August 2015; Stewart, 2015b). Recent surveys of Cotopaxi (Shappeck, 2011; Stewart, 2011) suggest that Media Lengua is no longer spoken or at least is very rare in the region. In Pijal, Imbabura, where the Imbabura dialect originated, only people aged roughly 40 and above speak it, while 4 The name of a specific skirt worn by many the Kichwa women in the Andes. 13 younger residents of Pijal are either Spanish monolinguals or passive bilinguals (Stewart, 2011). Within a short 10-minute bus ride from Pijal, however, Media Lengua appears to be quite healthy in the communities of Angla and Casco Valenzuela where children are still acquiring the language. 3.2 Structural Mixes Other mixed languages demonstrate more structural fusion, as the following sketches of Michif and Gurindji Kriol demonstrate. In these languages, the source languages combine to create a composite morphosyntactic frame. On the surface, these languages bear a striking resemblance to codeswitching and indeed most likely originated in codeswitching. 3.2.1 Michif Michif is the result of a community in which mixed marriages were common between Plains Cree-speaking women and French Canadian fur traders. Its genesis probably occurred in the early 1800s, from the speech of bilingual children of nomadic families in the Red River Colony area, now Manitoba and North Dakota (Bakker, 1997). Michif shows a great degree of structural mixing. In this case, the locus of mixing occurs between the verb and noun systems. The verb system is from Cree (Algonquian, polysynthetic), which has four verb classes (in/transitive and in/animate) and affixes which mark a number of grammatical functions (e.g., clause type, tense/mood, voice/valency/direction/aspect, person and number agreement). On the other hand, the nominal phrase reflects that of French, including constituent order (Det-Adj-N or Det-NAdj) order and limited productivity in article and adjective gender agreement. 14 The language division of the noun and verb structures extends to the lexicon. Michif is composed of 83-94% French nouns and 88-99% of Cree verbs depending on the speaker. Interrogatives, postpositions, demonstratives and personal pronouns are mostly Cree; and prepositions, adjectives, possessive pronouns and numerals are almost exclusively French. The French (NP) vs. Cree (VP) split is clearly demonstrated in (5) and (6). Cree elements are italicised. (5) êkwa pâstin-am sa bouche ôhi le loup ê-wî-otin-át and open-he.it his.F mouth this.OBV the.M wolf COMP-want-take-he.him And when the wolf came to him, he opened his mouth. (Bakker, 1997: 5) (6) Le per ki:-li-bin-i-w M.SG priest PST-M.SG-bless-INF-TA.3SG li mu:d M.SG people The priest blessed the people. (Bakker, 1997: 116) As with all mixed languages, the origin of Michif is a matter of speculation. The normalisation of French-Cree code-switching has also been provided as a justification for its split by Drapeau (1991), although Bakker (2003: 129ff) argues against this for three primary reasons: (1) the difference in quantity of lexical material transferred during mixed language formation far exceeds languages with abundant code-switching; (2) the nature of the embedded lexicon within a code-switching matrix differs from that of a mixed language (i.e., borrowed lexicon in a CS matrix is typically not from core lexicon, while in an ML it is), and (3) the lack of documentation of a transitory phase between code-switching behaviour before the mixed language was established. 15 3.2.2 Gurindji Kriol Gurindji Kriol is another example of a mixed language, which shows a VP-NP structural split according to its source languages. It is spoken by Gurindji people in northern Australia and derives from Gurindji (Pama-Nyungan) and Kriol (English-lexified creole). Gurindji Kriol originates from a language shift situation where there was extensive contact between non-Indigenous colonists and Gurindji people on a cattle station where Gurindji people worked in slave-like conditions (Charola & Meakins, 2016). Codeswitching provided fertile ground for the formation of this mixed language, which is now the first language of all Gurindji people under the age of 40 years (McConvell & Meakins, 2005; Meakins, 2011b). Structurally, Kriol contributes much of the verbal grammar including tense and mood auxiliaries, and transitive, aspect and derivational morphemes. Gurindji supplies most of the nominal structure including case and derivational morphology (Meakins, 2011a, 2011b). In this respect the structure of Gurindji Kriol is quite similar to the VPNP split seen in Michif, however unlike Michif, in Gurindji Kriol nouns and verbs also come from both source languages. In terms of the lexicon, Gurindji Kriol derives its lexicon relatively evenly from both languages. Based on a 200 word Swadesh list, 36.6% of vocabulary is derived from Kriol, including nouns for colours and parents and their siblings, some animals and plants; and the most basic verbs; and 35% of vocabulary is derived from Gurindji, including nouns for artefacts, body parts, siblings, grandparents, and in-laws, as well as most animals and plants; and verbs denoting impact, motion, body functions. The 16 remaining 28.4% contain synonymous forms from both languages (Meakins, 2011b: 19). Some extracts are given below. Gurindji elements are in italics. (7) I=m teik-im rarraj det karu=ma nyanuny 3SG.S=PRS take-TR run the child=TOP 3SG.DAT ngarlaka-ngka an warlaku kanyjurra-ngka. head-LOC and dog bin down-LOC dat diya-ngku i the deer-NOM 3SG.S PST fall jak im na karu an 3SG.O FOC child and dog kanyjurra-k, klif-nginyi=ma. down-ALL, cliff-ABL=TOP warlaku (The deer) takes the child running on its head, with the dog below. The deer threw the child and the dog downwards off the cliff. (Meakins, 2011b: 18) Both languages also contribute small amounts of grammar to the systems they do not dominate in Gurindji Kriol. For example, the Gurindji continuative suffix is found in the VP, and Kriol determiners are common in the NP. Gurindji Kriol also has Kriol SVO word order, although Gurindji information structure also determines word order to some extent (Meakins, 2009; Meakins & O'Shannessy, 2010). Complex clauses are constructed using both Gurindji and Kriol strategies, for example coordinating and relative clauses use Kriol conjunctions and relative pronouns; and subordinate clauses are formed using Gurindji-derived case and factive marking. New structures have also developed, such as an asymmetrical serial verb construction (Meakins, 2010); and old structures have undergone change. For instance, the ergative marker is now an optional nominative 17 marker with discourse functions (Meakins, 2015), while the locative marker also marks the progressive in presentational clauses (Meakins, 2016b). 3.3 Converted languages Converted languages develop when the ancestral language maintains its lexicon but undergoes a complete restructuring of its morphosyntax which is patterned on that of an introduced language. They differ from the previous categories of mixed languages in that all of the surface forms including lexicon and morphology derive from one language. Converted languages are the result of a process called metatypy (Ross, 2006), which will be discussed in §5.3. Sri Lanka Malay (§3.3.1), Sri Lanka Portuguese, and Takia are examples of converted languages. 3.3.1 Sri Lanka Malay Sri Lanka Malay is spoken in a number of communities in Sri Lanka by the Malay minority who migrated from the Malay Archipelago centuries ago. Lexically Sri Lanka Malay consists almost entirely of words from a Malay-based trade language called Vehicular or Bazaar Malay (Austronesian) (I. Smith et al., 2004), however it has been heavily restructured under the influence of Tamil (Dravidian) and, more recently, Sinhala (Indo-Aryan), owing to sustained social contact with Tamil-speaking Moors and pervasive Malay-Tamil bilingualism among Malay descendants. The result is a language which is unintelligible to Malay speakers in the Malay Archipelago, despite its Austronesian lexicon (Ansaldo, 2008; Nordhoff, 2009). Structurally, Sri Lanka Malay has evolved from an isolating language to an agglutinating one under the influence of Tamil, from which it has also acquired an SOV 18 word order, postpositions, pre-nominal determiners, and adjectives (Ansaldo, 2008, 2011a, 2011b; Hussainmiya, 1987; Nordhoff, 2009, 2012). (8) Sir anak-pada-yang ruuma-nang e-luppa. teacher child-PL-ACC PST-send house-DAT The teacher sent the children to school. (Ansaldo, 2008: 27) (9) Maana-ka kuuli perajan where-LOC daily.wage work ara-kerja PRS-do Here do they do daily wage work? (I. Smith & Paauw, 2006: 164) The Sri Lanka Malays are descendants of immigrants who were brought to Sri Lanka at different times by Dutch (1656 onwards) and British colonists (1796 onwards). Although they are called Malays, they came from a number of places including Banda, Balu, Java, connected especially by their Malay trade language. Traditionally they have maintained close ties with the Tamil-speakers, who are also Muslims. There are different views on how Sri Lanka Malay developed. The first is that it was creolised by the children of mixed marriages between Malay men who spoke a Vehicular Malay and Tamil-speaking Moor women. Smith & Paauw (2006) suggest that the mothers tried to make Malay the language of the home and children then nativised the Malay pidgin in a process of creolisation. Smith (2012) later emphasises the role of untutored L2 acquisition in the development of Sri Lanka Malay as a creole language. In this scenario the children were learning an L2 variety of Malay from their Tamilspeaking Moor mothers. Smith & Paauw (2006) propose that this whole process occurred before Malay was reintroduced by the British in schools between 1802 to 1873 therefore 19 creating a diglossic situation. During this period Sri Lanka Malay had extended contact with the Malay language (Smith, Paauw & Hussainmiya 2004). Bakker (2003, 2012) and Ansaldo (2008, 2011a) interpret the structural outcome of Sri Lanka Malay as a converted language which developed through a process of metatypy (see §5.3). Ansaldo (2008) and Bakker (2012) challenge the claim that pervasive intermarriage occurred between Malay immigrant soldiers and Moor women. In particular, Bakker (2012) presents evidence from a study on the molecular genetics of Sri Lanka populations which suggests that there has been little intermarriage between the Malay and Tamil; thus, the Malay speakers have retained their pre-Sri Lanka genetic profile (Papiha, Mastana & Jayasekara, 1996). This observation suggests that Tamil and Sinhala would have been present but external to the Malay community, thus they were not acquired by the latter. This picture does not support an abrupt nativisation hypothesis (associated by some with creolisation) but rather change that took place over an extended period of contact. Under this hypothesis, the Malays were segregated from the more indigenous populations and must have been highly multilingual in order for such a pervasive restructuring to have taken place. 4. Social functions and origins of mixed languages The following sections consider commonalities in the social functions of mixed languages (§4.1) and the socio-historical contexts of their genesis (§4.2). 4.1 Functions of mixed languages Many mixed languages are spoken by new ethnic groups. Some of these groups find their origins in mixed marriages. For example, Michif speakers are the children of Cree mothers and French fathers. They call themselves Métis, which reflects the mixed identity 20 of their group. Some other mixed languages are spoken by people who do not constitute a separate ethnic identity and regard their language as emblematic of a continuing ancestral heritage. For example, speakers of Media Lengua do not distinguish themselves from Quichua people, although their children (non Media Lengua speakers) identify to some extent with the urban Spanish-speaking society. Speakers of Gurindji Kriol speakers also do not separate themselves from Gurindji people in general and refer to their mixed language as 'Gurindji' (Meakins, 2012: 109). This mixed represents an attempt to maintain Gurindji under the continuing colonial pressure of English from which Kriol derives. 4.2 Socio-Historical origins of mixed languages Although mixed languages derive from one of three socio-historical settings, viz., mixed marriages, migration, or a cultural incursion, none of these contexts is particular to the formation of mixed languages. For example creolisation and language shift occur in colonial settings, and bilingualism and code-switching are found in all of these contexts. Children of the mixed marriages are said to form their own distinct cultural identity, which is indexed by the mixed language. Michif is the classic example. The second and third categories involve cases where no mixing of ethnic groups has occurred but rather one has dominated the other. This can happen when a minority population has migrated to a new region where it is dominated by, or became powerless in relation to, another group, as in the case of the Malay speakers in Sri Lanka. People may have migrated to a new region to escape persecution or for economic reasons, as was the case for Angloromani. They also might have been brought by another group, for example Sri Lanka Malay. In other cases, mixed languages arise when groups are colonised and 21 become minorities in their own country. Gurindji Kriol developed as a result of the invasion of Australia by British colonists. 5. Structural processes involved in mixed language genesis As shown in §3, mixed languages are typologically diverse. This diversity is a reflection of both the particular kinds of languages that came in contact and of the linguistic diversity of processes that produced them. While some outcomes are extended practices of borrowing and code-switching, others evolved through relexification or metatypy. These processes can be distinguished by whether the form or structure of the source languages is replicated. Mixed languages, which are the result of code-switching or borrowing replicate both forms and structures from their source languages in the resultant mix. On the other hand, mixed languages which find their origins in relexification replicate only form; and those which stem from metatypy, only structure: Table 2. Processes which lead to the formation of mixed languages FORM STRUCTURE EXAMPLE BORROWING/CS Borrow Borrow Gurindji Kriol, Angloromani, Michif RELEXIFICATION Borrow Maintain Media Lengua METATYPY Maintain Borrow Sri Lanka Malay 5.1 Borrowing and codeswitching accounts The genesis of most L-G and structurally mixed languages lies in large-scale borrowing or codeswitching. In the case of Gurindji Kriol, empirical evidence exists for a codeswitching genesis scenario (McConvell & Meakins, 2005). Borrowing and codeswitching involve the replication of lexical and morphological material from one language into the another language. Borrowing theories of mixed language genesis, such 22 as by Thomason & Kaufman (1988), have many commonalities to codeswitching approaches, for example Myers-Scotton (2003). For a detailed discussion of different transfer scenarios and the history of code-switching approaches, see Meakins (2013b). Thomason & Kaufman (1988) base their theory of mixed language genesis on their borrowability scale of linguistic categories, which ranges from nouns, the most borrowable category, right through to relatively unborrowable structures such as inflectional morphology. The scale is implicational, viz., where derivational morphology has been borrowed, conjunctions have already been borrowed etc. Borrowing is a consolidation of insertional codeswitching practices where elements from one language are inserted into another language's morphosyntactic frame or matrix. Similar to Thomason & Kaufman, Myers-Scotton (2003) developed a scale of likely switches, with content words such as nouns easily switched and inflectional morphology impossible to switch. Thomason & Kaufman suggest that mixed languages represent different degrees of transfer, with some borrowing only vocabulary and others borrowing almost entire grammatical systems (along with vocabulary). Similarly Myers-Scotton (2003) theorises the move from insertional codeswitching to a mixed language within her Matrix Language Frame model, labelling the transition the “Matrix Language Turnover Hypothesis.” This hypothesis is concerned with the change in dominance of the participating languages. Myers-Scotton proposes that mixed languages arise when there is a turnover underway which does not go to completion. According to Myers-Scotton, mixed languages may stop at different places, which explains why they surface in different forms and with the split in different places. For example, in Gurindji Kriol, both 23 Gurindji and Kriol acted as matrix languages in the initial code-switching stage of the 1970s. Kriol because more dominant, providing the matrix language more often and eventually this pattern fossilised into the mixed language (McConvell & Meakins, 2005). Under both borrowing and codeswitching accounts, mixed language formation may halt at the least disruptive end of the scale and exhibit only lexical borrowings. The L-G languages are a good example because they are characterised by a clear division between the lexicon and the grammar. An example is Angloromani where a subset of vocabulary from the ancestral language (Romani) is maintained as a lexical reservoir and exists in parallel with the lexicon of the grammar language (English). A contemporary process of paralexification replaces vocabulary utterance by utterance. In this respect, paralexification occurs synchronically and is not a diachronic process; and although this process resembles codeswitching, paralexification does not require bilingualism. More controversially, Thomason & Kaufman (1988: 103–104) suggest that Angloromani is the result of the complete borrowing of the English grammatical system coupled with the maintenance of some vocabulary from Romani. More convincing cases of structural borrowing are Michif and Gurindji Kriol, which contain inflectional morphology from both source languages. Other borrowings such as lexical and more minor structural borrowings are also present as predicted by Thomason & Kaufman’s (1988) implicational hierarchy of borrowing. For example, inflectional morphology from both French and Cree is present in Michif. Verbal inflections are derived from Cree and, in the NP, Michif preserves both French plural marking and adjectival agreement. Similarly, Gurindji Kriol combines Kriol, the language of the verbal inflectional categories (tense and mood markers), with Gurindji 24 nominal inflections in the form of case marking, both syntactic (ergative, dative, possessive) and semantic (locative, allative, ablative). The situation described for Michif and Gurindji Kriol is exceptional given the empirical rarity of inflectional morphology transfer. Inflectional morphology is rarely borrowed and mostly derived from the more dominant language in insertional codeswitching. Indeed Matras (2003: 158) suggests that a particular feature of mixed languages is the seemingly unconstrained borrowing of grammatical elements, which in the past have been labelled as 'loan proof'. 5.2 Relexification Relexification is a process involving the relabelling of lexical entries from one language to another (Lefebvre, 2005, 2006; Lefebvre & Therrien, 2007; Muysken, 1981). It is a sub-type of borrowing where the form of a word or affix is borrowed, but the semantics or functional distribution of the native remains the same. Relexification takes place on a large scale and involves the relabelling of substantial portions of a source language's lexicon. This process, as opposed to adlexification in Shappeck's (2011) analysis of Quichua-Spanish contact, does not maintain synonymic or near-synonymic pairs from each language. According to Muysken (1981), the only essential information needed in the relexification process is the replacement of the phonological shell, while the transfer of other linguistic elements, known as translexification, e.g., semantic representation and syntactic, subcategorization and/or selectional features is nonessential. According to Lefebvre (1998, 2005), instead of the immediate relabelling of the ancestral language's phonological representation by that of the introduced language, both representations co-exist simultaneously. After an indeterminate amount of time, the 25 original phonological representation falls into disuse in favour of the introduced language. At this stage, the lexical entry of the source language is made up of mixed elements from each language. Lefebvre also makes it clear that the phonological representation of the target language is adapted to the phonological grammar of the source language. For example, the phonological shell of the introduced Spanish word quer-er 'to want, to love' adopts the semantic features of the ancestral Quichua word muna-na 'to want, to like, to love, to enjoy' while conforming to Quichua phonotactics as kiri-na [kiˈɾina] (Stewart, 2011:57). See the verb 'like and like/want' in (10): (10) Spanish: Te gusta pescado frito? 2.OBJ like.3SG fish fried Quichua: Chaluwa frei-shka-ta muna-ngi-chu? fish fry-PRT-ACC like/want-2-Q Media Lengua: Pescado cozna-shka-ta kiri-ngi-chu? fish cook-PRT-ACC like/want-2-Q 'Do you like fried fish?' (Stewart, personal database) 5.3 Metatypy Converted languages are the result of the diachronic process of metatypy, which is the typological restructuring of one language on the model of another while maintaining its forms (Ross, 2006: 95). The language which undergoes restructuring is emblematic of the speech community's identity, viz., their ancestral language, and the language whose structures are borrowed is the one used to communicate with the other speech community (Ross, 2001: 146). 26 Sri Lanka Malay is most likely the outcome of metatypy. It maintains Malay vocabulary and morphology, while its morphosyntax is patterned on Tamil. According to Smith, Paauw & Hussainmiya (2004: 2004), Malay prepositions have become postpositions on the model of Tamil’s suffixing and dependent-marking patterns. 6. Phonology Traditional phonological analyses (Bakker, 1997; Muysken, 1997b) and theoretical accounts (van Gijn, 2009) of mixed languages suggest that their phonology can be predicted based on their morphosyntactic arrangement. That is, mixed languages have two prototypical arrangements. In the first, words systematically adopt the phonological structure of the language that provides the grammar; while, in the second, they preserve the phonological shape from the language they were selected from. In the case of L-G mixed languages (§3.1), phonology is considered part of the grammatical system. Lexical items from language A are regularized to the phonology of language B, the source of the grammar. Therefore, the Spanish lexicon in Media Lengua should sound like that of Quichua. For the structural mixes (§3.2), Michif is often analysed as having two coexisting phonologies, with the French phonology applying to French-derived elements and Cree phonology to Cree-derived ones (Bakker, 1997; Rhodes, 1986). In the case of converted languages (§3.3), little information exists on the phonology of Sri Lanka Malay. However, detailed work on a similar language, Sri Lanka Portuguese, which derives its forms from Portuguese but its structure from Tamil, shows the vowel system to be of Portuguese origin, at least regarding number and place of articulation. On the other hand, it has eliminated the nasal contrast found in Portuguese in favour of the length contrast in Tamil (I. Smith, 1978). 27 Drawing on descriptions of Media Lengua, Callahuaya, Mednyj Aleut, and Michif, van Gijn (2009) claims that the phonological arrangement of mixed languages can be reasonably predicted based on the unmixed phonological domains which directly correspond to their place on the prosodic hierarchy (Nespor & Vogel, 1986). According to Van Gijn, agglutinating mixed languages like Media Lengua appear to conform to the phonology of the grammar’s source language, because nearly all words contain morphosyntactic elements from both languages, e.g., Spanish stems and Quichua suffixes. Michif, a highly synthetic language on the Cree side and fusional on the French side, contains a greater degree of 'unmixed' words, because the verb phrases, Creederived, remain syntactically separated from noun phrases, largely French-derived. Because of this division of labour, van Gijn claims that French phonological rules can apply to the French-derived elements while Cree phonological rules can apply to the independent Cree-derived elements. He then maps these observations onto the prosodic hierarchy. Here, because both Media Lengua and Michif contain elements from both languages at higher prosodic levels (e.g., the intonational phrase and above) there should be suprasegmental material from both languages. At the mid-levels (e.g., the phonological phrase and prosodic word), he claims Media Lengua should still conform to Quichua since the language still shares elements with it at these levels, while Michif is still considered 'divided'. Finally, at the lower levels (e.g., syllable and foot5), van Gijn claims that both Media Lengua and Michif should be phonologically stratified. While van Gijn's (2009) analysis reflects various impressionistic aspects of the surface-level phonologies of mixed languages, it falls short at predicting the actual 5 Van Gijn does not include the foot level in his analysis, though Muysken (2013b) does. 28 phonetic production and perceptual realities of these languages. From a phonetic stand point, mixed language phonology is a complex arrangement of the phonologies of the source languages. Analyses of Media Lengua (Stewart, 2014, 2015a, 2015b, 2018a, 2018b), Gurindji Kriol (Buchan, 2012; Jones & Meakins, 2013; Jones, Meakins & Buchan, 2011; Jones, Meakins & Muawiyath, 2012; Stewart, Meakins, Algy & Joshua 2018, Stewart, Meakins, Algy, Ennever & Joshua, forthcoming), and Michif (Rosen, 2006, 2007; Rosen, Stewart & Cox, forthcoming) suggest there exists a propensity for phonological material to assimilate to the phonology of the ancestral language (e.g., Quichua for Media Lengua, Cree for Michif, and Gurindji for Gurindji Kriol). In other words, the language, which was acquired originally as an L2 essentially conforms to the L1 phonological system in much the same way a mid to late bilingual might acquire the phonology of their L2 (i.e., acquired after puberty). At the same time, the introduced language appears to feed in phonological aspects that appear beneficial for maintaining contrasts. The arrangements of the source phonologies, however, do not always conform to traditional notions of adaptive dispersion models, which predict that when a new category is established, crowding of the phonetic space occurs causing dispersion to maintain contrasts (Flege, 2007; Johnson, 2000; Liljencrants & Lindblom, 1972; Lindblom, 1986, 1990; Livijn, 2000). Instead, there are near-mergers, overlapping categories, categorical assimilation, categorical maintenance, and overshoot of target categories at the segmental level, in addition to prosodic assimilation, possible preservations of archaic patterns, and innovation at the suprasegmental level. The following sections describe these processes for obstruents (§6.1), vowels (§0) and suprasegmentals (§0). 29 6.1 Obstruents For Gurindji Kriol, with coexistent Gurindji voiceless stops /p,t,c,k/ and Kriol contrastive pairs /p-b, t-d, k-g/, Jones & Meakins (2013) examine whether voice onset time (VOT) values of Gurindji Kriol reflect those in English cognates. Their results show that, regardless of the source-voicing category, stops in Gurindji Kriol are produced with short lag VOT. In contrast, Stewart et al. (2018) show that perception of voicing in Gurindji Kriol is currently developing, owing to increasing contact with mainstream English. Their results show listeners are able to perceive voicing contrasts between labial stops, [p-b], whereas contrast between the alveolar and velar stops, [t-d] and [k-g], is perceived only by an estimated 39% of the participants tested. The rest of the participants consistently perceived the voiceless alveolar and velar but not their voiced counterparts. Kriol listeners tested in the same experiment showed greater degrees of perception of the contrast but not to the extent that would make it significant in the phonology of the language. A similar story can be told for fricatives in Gurindji Kriol. Butcher (2006) shows the majority of Australian languages including Gurindji lack phonemic fricatives. However, Sandefur (1979) shows the production of fricatives in Kriol is highly variable in comparison with their stop counterparts. Buchan (2012) specifically investigates the production contrasts between voiceless fricatives and stops [f-p] & [s-t]) with an analysis of maternal speech in Gurindji Kriol. While trends suggest variability across place/ manner of articulation, mother’s speech of word-initial fricatives became more prototypical, when communicating with older children. According to a perceptual study of this same conflict site by Stewart et al (forthcoming), perceptions of [f-p] and [s-t] 30 were also quite variable, with little over half the participants showing a strong contrast between the pairs while the other half either had consistent responses to the fricative stimuli but random responses to the stops. Yet others only showed consistent responses to the fricatives. Kriol listeners showed similar results. Finally, for Media Lengua, with coexistent source stop systems similar to Gurindji Kriol, consisting of voiceless /p,t,k/ phonemes from Quichua and contrastive /pb, t-d, k-g/ phonemes from Spanish, Stewart (2014, 2018b) also explores the production and perception of stop voicing. Unlike Gurindji Kriol, Media Lengua has clearly adopted the voicing contrast in both production and perception, while older Quichua speakers have a tendency to weaken the voiced series during production and older Quichua listeners do not consistently identify the voiced series perceptually. 6.2 Vowels Several recent studies provide intra and/or inter-language comparisons between the vowel systems of mixed languages and those of their source languages. Gurindji Kriol has coexistent vowel systems consisting of /ɪ,ɐ,ʊ,ɛ,ɔ/ monophthongs from Gurindji and /ɪ,e,æ,ɔ,ʊ,ɐ/ monophthongs, including a length contrast in five of the positions (/iː,ɜː,oː,ʉː,ɐː/), from Kriol. Jones, Meakins & Muawiyath (2012) demonstrated there exists greater formant (both F1 & F2) overlap in the mainstream Australian Englishsource front vowels /æ/ & /e/ and back vowels /ʉː/ & /oː/ in Gurindji Kriol compared to their English cognates in Kriol – a result which may suggest that Gurindji Kriol is expanding its vowel inventory. Jones, Meakins & Muawiyath (2012) also show that the duration differences between the Gurindji Kriol (e.g., /ɪ/ and /i:/) are also shorter compared to those in mainstream English. 31 For Media Lengua, with coexistent vowel systems consisting of /i,a,u/ phonemes from Quichua and /i,e,a,o,u/ phonemes from Spanish, Stewart (2014, 2018b) describes an partially overlapping yet stratified system. In this analysis both Quichua-source and Spanish-source high and low vowels (/i,a,u/) co-exist as near-mergers while Spanishsource mid-vowels (/e,o/) co-exist with Quichua-source high vowels (/i,u/) with considerable overlap. Stewart (2014) also shows that the arrangement of Spanish-source vowels in borrowed words into Quichua are dissimilar. Here, Quichua-source and Spanish-source high and low vowels (/i,a,u/) underwent complete merger while Quichuasource high-vowels maintained a negligible contrast with Spanish-source mid-vowels (/i,e/). Unlike Gurindji Kriol and Michif, Media Lengua is often described as a mixed language with few stratified elements at the phonological level (Gómez Rendón, 2005; Muysken, 1997b). These results however, call into question such analyses since Media Lengua appears to be operating two vowel systems based solely on the language of origin, albeit with very low functionality. Michif has source vowel systems consisting of /i,e,a,o/ phonemes with a length contrast (/iː,eː,aː,oː/) from Plains Cree and /i,y,e,ø,ə,ɛ,œ,a,ɑ,ɔ,o,ʊ,u/ phonemes with four nasal contrasts /ɛ,̃ œ̃ ,ɑ̃,ɔ/̃ and one length contrast /ɛː/ from Métis French. Rosen (2007) provides a synchronic description of the Michif phonological system, showing that it is unnecessary to focus on the source languages to accurately describe its underlying phonology. Rosen, Stewart & Cox (forthcoming) also investigate phonological stratification with an acoustic analysis of F1 and F2 formants. Based on the oral series, their results reveal that only two French vowels appear significantly different from their Cree counterparts (/ɛ,ɔ/). 32 6.3 Suprasegmentals Regarding suprasegmental phonology, Stewart (2015a) describes a variety of intonation patterns based on fundamental frequency (F0) contours in Media Lengua. He suggests that the overwhelming majority conform to Quichua-like prosody and those that did not were either innovations or preserved patterns no longer used in present day Quichua of the region where Media Lengua is spoken. Furthermore, there were no patterns that appeared to match Spanish-like prosody, that were not already shared with Quichua. For Michif, Rosen (2006) concludes that its stress patterns are a convergence of both Cree and French stress systems. According to her, Michif stress assignment is strikingly similar to that of Cree, except that Michif is quantity sensitive at the word level, while Cree is quantity insensitive at every level (Rosen, 2006: 186-7). 6.5 Discussion The results from the acoustic studies presented heretofore suggest that stratification at both the segmental and suprasegmental level is more complicated than a simple clear-cut division between source languages. Many of the different phonological arrangements found throughout these acoustic analyses are non-conventional in the sense that in order to maintain phonological contrasts, we would expect categorical dispersion. On the other hand, if contrasts are not important for phonological optimisation, we would expect the complete loss of a sound— not overlapping systems or covert contrasts, which are perceptually contrastive only in one position but not in others, weaker degrees of categorial discrimination and identification than would be expected for fully contrastive phonemes, and variability in production and perception. 33 These findings suggest that when stratification is observed, it is most likely a byproduct of different underlying processes such as the age of acquisition of the L2 during the genesis of the mixed language, the level of proficiency of the introduced language, exposure and extra-linguistic influences (e.g., prestige), and the level of functional load required to maintain an optimum level of phonemic and prosodic contrast in the mixed language. As such, many of the phonological arrangements reflect those in speakers who learned an L2 later in life in that phonological conflict sites are either not fully acquired, assimilated, or acquired but not to the same degree as by native speakers. This can also be seen in how mixed languages overwhelmingly conform to the phonological system of the L1 source language spoken by the community before the L2 language was present. The fact that some contrasts are adopted while others are not might also suggest that cognitive factors are at play which benefit cognitive processes such as distributing functional load, levelling out phoneme frequency, and allowing for a greater number of contrasts leading to greater phonological optimization. The unruly phonological systems of mixed languages should not come as a surprise since the phonological shells of entire linguistic systems and/or categories undergo intentional transfer to a new system in an incredibly short period of time before becoming nativised, oftentimes within less than a single generation. 7. Concluding remarks The category of MIXED LANGUAGE is a mixed bag, to say the least. They have different lexical profiles, ranging from languages such as Media Lengua which have absorbed extraordinary amounts of vocabulary from a second language historically to languages such as Angloromani which selectively replace lexical items as a synchronic process in 34 some language shift scenarios. Mixed languages also have different structural profiles. In some mixed languages such as Media Lengua and Angloromani, the structure is clearly derived from one language, whereas in others such as Michif and Gurindji Kriol, the two languages contribute to the structure of the mix in a way that contrasts dramatically with other language contact scenarios, for instance, when inflectional morphology from both languages is often present. There are yet other mixed languages such as Sri Lanka Malay which are the outcomes of the restructuring of the grammar of one language on the model of another. One ramification of mixed languages is the inapplicability of historical classification methods, viz., the comparative method or newer phylogenetic tools. Lexico-statistical methods do not reveal a straightforward genetic signal for mixed languages due to the often mixed nature of basic vocabulary. Similarly, morphological methods cannot be used because they are based on categories which are generally resistant to transfer, e.g. inflectional paradigms. In this respect, one of the definitions of a mixed language relies on a negative criterion, viz., their inability to be classified according to a single language family. Mixed languages also broadly share one genesis story, in that they have all emerged in situations of bilingualism where a common language was already present. In this respect, they did not develop to serve a communicative function, but rather as a marker of an in-group identity. This identity was either a new identity created through mixed marriages or groups (Michif, Sri Lanka Malay) or the maintenance of an old identity, which is under threat (Angloromani, Gurindji Kriol, Media Lengua). To add to the lack of coherence in this class of contact language, it seems that the socio-historical 35 cradle of the language does not help predict the resultant typological profile of languages. For example, Michif and Gurindji Kriol are both structural mixes, which split along the NP and VP but Michif has resulted from mixed marriages and Gurindji Kriol from language shift in a single cultural group undergoing colonisation. Finally, mixed languages provide a unique opportunity to study the extremes of language contact and change, e.g. implicational hierarchies in studies of borrowing and predictive theories of code-switching. Although it was originally thought that mixed languages were the result of special processes (e.g. Bakker's theory of intertwining), it seems more likely that they are simply the "the extraordinary result of ordinary processes" (Thomason, 2003). The structure of most mixed languages can be attributed to earlier codeswitching practices, and their phonology is the result of well documented L2 learning processes. 8. References Aboh, Enoch & Umberto Ansaldo. (2007). The role of typology in language creation. In Umberto Ansaldo, Stephen Matthews & Lisa Lim (Eds.), Deconstructing Creole (pp. 39-66). Amsterdam: John Benjamins. Ansaldo, Umberto. (2005). Typological admixture in Sri Lanka Malay: The case of Kirinda Java. University of Amsterdam. Ansaldo, Umberto. (2008). Sri Lanka Malay revisited: Genesis and classification. In K. David Harrison, David Rood & Arienne Dwyer (Eds.), Lessons from documented endangered languages (pp. 13-42). Amsterdam: Benjamins. Ansaldo, Umberto. (2011a). Metatypy in Sri Lanka Malay. In Rajendra Singh (Eds.), Annual Review of South Asian Languages and Linguistics (pp. 3-16). Berlin: Walter de Gruyter. Ansaldo, Umberto. (2011b). Sri Lanka Malay and its Lankan adstrates. In Claire Lefebvre (Ed.), Creoles, their Substrates, and Language Typology (pp. 367-382). Amsterdam: Benjamins. Auer, Peter. (1999). From codeswitching via language mixing to fused lects: Toward a dynamic typology of bilingual speech. Journal of Bilingualism, 3(4), 309-332. Bakker, Peter. (1994). Michif, the Cree-French mixed language of the Metis buffalo hunters in Canada. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 36 case studies in language intertwining (pp. 13-33). Amsterdam: Uitgave IFOTT. Bakker, Peter. (1997). A language of our own: The genesis of Michif, the mixed CreeFrench language of the Canadian Métis New York: Oxford University Press. Bakker, Peter. (1998). Para-Romani language versus secret languages: Differences in origin, structure and use. In Yaron Matras (Ed.), The Romani element in nonstandard speech (pp. 69-96). Wiesbaden: Harrassowitz. Bakker, Peter. (2003). Mixed languages as autonomous systems. In Yaron Matras & Peter Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 107-150). Berlin: Mouton de Gruyter. Bakker, Peter. (2012). Sri Lanka Malay: New findings on contacts. In Sebastian Nordhoff (Ed.), The Genesis of Sri Lanka Malay (pp. 53-84). Leiden: Brill. Bakker, Peter. (2013). Mixed languages. Oxford Bibliographies online. Retrieved from http://www.oxfordbibliographies.com website. Bakker, Peter. (2015). Typology of mixed languages. In A. Y. Aikhenvald & R. M. W. Dixon (Eds.), The Cambridge Handbook of Linguistic Typology. Cambridge: Cambridge University Press. Bakker, Peter & Maarten Mous. (1994). Mixed languages: 15 case studies in language intertwining. Amsterdam: Institute for Functional Research into Language and Language Use (IFOTT). Bakker, Peter & Robert Papen. (1997). Michif: A mixed language based on Cree and French. In Sarah Thomason (Ed.), Contact languages: A wider perspective (Vol. 17, pp. 295-363). Amsterdam: John Benjamins. Boretzky, Norbet & Birgit Igla. (1994). Romani mixed dialects. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 35-68). Amsterdam: Uitgave IFOTT. Brenzinger, Matthew. (1987). Die Sprachliche und Kulturelle Stellung der Mbugu (Ma'a). (MA), Cologne. Broch, Ingvild & Ernst Hakon. (1983). Russenorsk: The Russo-Norwegian pidgin. In Shirô Hattori & Kazuko Inoue (Eds.), Proceedings of the XIIIth International congress of linguists (pp. 792-796). Tokyo. Buchan, Heather. (2012). Phonetic variation in Gurindji Kriol and Northern Australian English: A longitudinal study of fricatives in maternal speech. (PhD), University of Wollongong. Bundgaard, Rikke, Carmel O'Shannessy & Brett Baker. (2015). A marriage of necessity: Light Warlpiri obstruent perception. Paper presented at the 46th Australian Linguistics Society Conference, Sydney. Butcher, Andy. (2006). Australian Aboriginal Languages: Consonant-salient phonologies and the “place of articulation imperative.” In Jonathan Harrington & Marija Tabain (Eds.), Speech Production: Models, Phonetic Processes and Techniques (pp. 187–210). New York: Psychology Press. Carling, Gerd, Lenny Lindell & Gilbert Ambrazaitis. (2014). Scandoromani: Remnants of a Mixed Language. Leiden: Brill. Ceniccola, Serena. (2014). Contaminazioni anglo-americane nell’uso della lingua finlandese contemporanea. (Laurea Magistrale), Università di Bologna. Charola, Erika & Felicity Meakins. (Eds.). (2016). Yijarni: True Stories from Gurindji Country. Canberra: Aboriginal Studies Press. 37 Chen, Naixiong. (1986). Guanyu Wutun Hua (An outline of the Wutun linguistic structure) Journal of Asian and African Studies, 31, 33-52. Clements, Joseph. (2009). Linguistic legacy of Spanish and Portuguese. Cambridge: Cambridge University Press. Clements, Joseph, Patricia Amaral & Ana Luís. (2008). Cultural identity and the structure of a mixed language: The case of Barranquenho. Proceedings of the Annual Meeting of the Berkeley Linguistics Society, 34S. de Gruiter, Miel. (1994). Javindo, a contact language in pre-war Semarang. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 151-159). Amsterdam: Uitgave IFOTT. de Gruiter, Victor. (1990). Het Javindo: De verboden taal. The Hague: Moesson. de Smit, Merlijn. (2010). Modelling mixed languages: Some remarks on the case of Old Helsinki Slang. Journal of Language Contact, 3, 1-19. Deibel, Isabel. (accepted). Adpositions in Media Lengua: Quichua or Spanish? – Evidence of a lexical-functional split. Journal of Language Contact. Dikker, Suzanne. (2008). Spanish prepositions in Media Lengua: Redefining relexification. In Thomas Stolz, Dik Bakker & Rosa Salas Palamo (Eds.), Hispanisation: The Impact of Spanish on the Lexicon and Grammar of the Indigenous Languages of Austronesia and the Americas (pp. 121-148). Berlin: Mouton de Gruyter. Dreyfuss, Gail & Djoehana Oka. (1979). Chinese Indonesian: A new kind of language hybrid? Papers in pidgin and creole linguistics (Pacific Linguistics), A-57, 247274. Flege, James. (2007). Language contact in bilingualism: phonetic system interactions. Laboratory Phonology, 9, 353-380. Giesbers, Herman. (1995). Dutch Indonesian language mixing in Jakarta. In M. d. Dikken & K. Hengeveld (Eds.), Linguistics in the Netherlands 1995 (pp. 89-100). Amsterdam: John Benjamins. Gillon, Carrie & Nicole Rosen. (2016). Critical mass in Michif. Journal of Pidgin and Creole Languages, 31(1), 113–140. Gillon, Carrie & Nicole Rosen. (2018). Nominal Contact in Michif. Oxford: Oxford University Press. Golovko, Evgeniy. (1994). Mednyj Aleut or Copper Island Aleut: An Aleut-Russian mixed language. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 113-121). Amsterdam: Uitgave IFOTT. Golovko, Evgeniy. (1996). A case of nongenetic development in the Arctic area: The contribution of Aleut and Russian to the formation of Copper Island Aleut. In Ernst Jahr & Ingrid Broch (Eds.), Language Contact in the Arctic. Northern Pidgins and Contact Languages (pp. 63-77). Berlin: Mouton de Gruyter. Golovko, Evgeniy. (2003). Language contact and group identity: The role of 'folk' linguistic engineering. In Yaron Matras & Peter Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 177-208). Berlin: Mouton de Gruyter. Golovko, Evgeniy & Nikolai Vakhtin. (1990). Aleut in contact: The CIA enigma. Acta Linguistica Hafniensia, 72, 97-125. Gómez Rendón, Jorge. (2005). La Media Lengua de Imbabura. In Hella Olbertz & Pieter 38 Muysken (Eds.), Encuentros y Conflictos: Bilingüismo y Contacto de Lenguas en el Mundo Andino (pp. 39-58). Madrid: Iberoamericana. Gómez Rendón, Jorge. (2008). Typological and social constraints on language contact. Amerindian languages in contact with Spanish. Utrecht: LOT Dissertation Series. Gonzales, Wilkinson Daniel. (2017). Language contact in the Philippines: The history and ecology from a Chinese Filipino perspective. Language Ecology, 1(2), 185– 212. Gonzales, Wilkinson Daniel. (2018). Philippine Hybrid Hokkien as a postcolonial mixed language: Evidence from nominal derivational affixation mixing (MA Thesis). National University of Singapore, Singapore. Grant, Anthony. (1994). Shelta: The secret language of Irish travellers viewed as a mixed language. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 123-150). Amsterdam: Uitgave IFOTT. Hancock, Ian. (1970). Is Anglo-Romanes a creole? Journal of the Gypsy Lore Society, 49, 41-44. Hancock, Ian. (1976). The pidginization of Angloromani. In George Cave (Ed.), New Directions in Creole Studies (pp. 1-23). Georgetown: University of Guyana. Hannß, Katja & Pieter Muysken. (2014). Reduplication in Andean languages. In Gale Goodwin Gómez & Hein van der Voort (Eds.), Reduplication in Indigenous Languages of South America (pp. 39-76). Leiden: Brill. Hoff, Berend. (1994). Island Carib, an Arawakan language which incorporated a lexical register of Cariban origin, used to address men. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 161-168). Amsterdam: Uitgave IFOTT. Hussainmiya, Bachmiya. (1987). Melayu Bahasa: Some preliminary observations on the Malay creole of Sri Lanka. In Bachmiya Hussainmiya (Ed.), Lost Cousins: The Malays of Sri Lanka (pp. 153-172). Malaysia: Universiti Kebangsaan. Jacobs, Bart. (2012). Embedding Papiamentu in the mixed language debate. Journal of Historical Linguistics, 2(1), 52-82. Janhunen, Juha, Marja Peltomaa, Erika Sandman & Xiawu Dongzhou. (2008). Wutun. München: Lincom Europa. Jarrín Paredes, Elena. (2014). Estereotipos Lingüísticos en Relación al Kichwa y a la Media Lengua en las Comunidades de Angla, Casco Valenzuela, El Topo y Ucsha de la Parroquia San Pablo del Lag. Pontificia Universidad Católica Del Ecuador, Quito. Jarva, Vesa. (2008). Old Helsinki Slang and language mixing. Journal of Language Contact, 1, 52-80. Johnson, Keith. (2000). Adaptive disperion in vowel perception. Phonetica, 57, 181-188. Jones, Caroline & Felicity Meakins. (2013). Variation in voice onset time in stops in Gurindji Kriol: picture naming and conversational speech. Australian Journal of Linguistics, 33(2), 194-217. Jones, Caroline, Felicity Meakins & Heather Buchan. (2011). Comparing vowels in Gurindji Kriol and Katherine English: Citation Speech Data. Australian Journal of Linguistics, 31(3), 305-327. Jones, Caroline, Felicity Meakins & Shujau Muawiyath. (2012). Learning vowel categories from maternal speech in Gurindji Kriol. Language Learning, 62(4), 39 997–1260. Juárez, Geraldo. (1998). Los Kallawayas: Medicina Indígena en los Andea Bolivianos. Spain: Ediciones de la Universidad de Castilla-la Mancha. Lee, Jennifer. (1987). Tiwi today: A study of language change in a contact situation. Canberra: Pacific Linguistics. Lee-Smith, Mei & Stephen Wurm. (1996). The Wutun language. In S. Wurm, P. Mühlhäusler & D. Tryon (Eds.), Atlas of Languages of Intercultural Communication in the Pacific, Asia and the Americas Vol. 2 (pp. 883-897). Berlin: Mouton. Liljencrants, Johan & Björn Lindblom. (1972). Numerical simulation of vowel quality systems: The role of perceptual contrast. Language, 48, 839-862. Lindblom, Björn. (1986). Phonetic universals in vowel systems. In John Ohala & Jeri Jaeger (Eds.), Experimental phonology (pp. 13-44). Orlando (Florida): Academic Press. Lindblom, Björn. (1990). Explaining phonetic variation: A sketch of the H&H theory. In William Hardcastle & Alain Marchal (Eds.), Speech Production and Speech Modeling (pp. 403-439). Dordrecht, Holland: Springer Science & Business Media. Lipski, John. (2016). Language switching constraints: More than syntax? Data from Media Lengua. Bilingualism: Language and Cognition, 25. Livijn, Peder. (2000). Acoustic distribution of vowels in differently sized inventories – hot spots or adaptive dispersion? PERILUS, 23. Matras, Yaron. (1999). The state of present day Domari in Jerusalem. Mediterranean Language Review, 11, 1-59. Matras, Yaron. (2000). Mixed languages: A functional-communicative approach. Bilingualism: Language and Cognition, 3(2), 79-99. Matras, Yaron. (2003). Mixed languages: Re-examining the structural prototype. In Yaron Matras & Peter Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 151-176). Berlin: Mouton de Gruyter. Matras, Yaron. (2007). Grammatical borrowing in Domari. In Yaron Matras & Jeanette Sakel (Eds.), Grammatical borrowing in cross-linguistic perspective (pp. 151164). Berlin: Mouton. Matras, Yaron. (2009). Language Contact. Cambridge: Cambridge University Press. Matras, Yaron. (2010). Romani in Britain. The afterlife of a language. Edinburgh: Edinburgh University Press. Matras, Yaron. (2012). A grammar of Domari. Berlin: Mouton de Gruyter. Matras, Yaron & Peter Bakker. (2003a). The study of mixed languages. In Yaron Matras & Peter Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 1-20). Berlin: Mouton de Gruyter. Matras, Yaron & Peter Bakker. (Eds.). (2003b). The mixed language debate: Theoretical and empirical advances. Berlin: Mouton de Gruyter. Matras, Yaron, Heather Gardner, Charlotte Jones & Veronica Schulmann. (2007). Angloromani: A different kind of language? Anthropological Linguistics, 49(2), 142-184. McConvell, Patrick. (2002). Mix-im-up speech and emergent mixed languages in Indigenous Australia. Texas Linguistic Forum (Proceedings from the 9th Annual 40 Symposium about Language and Society), 44(2), 328-349. McConvell, Patrick. (2008). Mixed Languages as outcomes of code-switching: Recent examples from Australia and their implications. Journal of Language Contact, 2, 187-212. McConvell, Patrick & Felicity Meakins. (2005). Gurindji Kriol: A mixed language emerges from code-switching. Australian Journal of Linguistics, 25(1), 9-30. Meakins, Felicity. (2007). Case marking in contact: The development and function of case morphology in Gurindji Kriol, an Australian mixed language. (PhD Dissertation), University of Melbourne, Melbourne. Meakins, Felicity. (2008a). Land, language and identity: The socio-political origins of Gurindji Kriol. In Miriam Meyerhoff & Naomi Nagy (Eds.), Social Lives in Language (pp. 69-94). Amsterdam: John Benjamins. Meakins, Felicity. (2008b). Unravelling languages: Multilingualism and language contact in Kalkaringi. In Jane Simpson & Gillian Wigglesworth (Eds.), Children’s language and multilingualism: Indigenous language use at home and school (pp. 247-264). New York: Continuum. Meakins, Felicity. (2009). The case of the shifty ergative marker: A pragmatic shift in the ergative marker in one Australian mixed language. In Jóhanna Barddal & Shobhana Chelliah (Eds.), The Role of Semantics and Pragmatics in the Development of Case. (pp. 59-91). Amsterdam: John Benjamins. Meakins, Felicity. (2010). The development of asymmetrical serial verb constructions in an Australian mixed language. Linguistic Typology, 14(1), 1-38. Meakins, Felicity. (2011a). Borrowing contextual inflection: Evidence from northern Australia. Morphology, 21(1), 57-87. Meakins, Felicity. (2011b). Case marking in contact: The development and function of case morphology in Gurindji Kriol. Amsterdam: John Benjamins. Meakins, Felicity. (2011c). Spaced out: Inter-generational changes in the expression of spatial relations by Gurindji people. Australian Journal of Linguistics, 31(1), 4377. Meakins, Felicity. (2012). Which Mix? - Code-switching or a mixed language - Gurindji Kriol. Journal of Pidgin and Creole languages, 27(1), 105-140. Meakins, Felicity. (2013a). Gurindji Kriol. In Susanne Michaelis, Philippe Maurer, Martin Haspelmath & Magnus Huber (Eds.), The survey of pidgin and creole languages, Vol III (pp. 131-139). Oxford: Oxford University Press. Meakins, Felicity. (2013b). Mixed languages. In Yaron Matras & Peter Bakker (Eds.), Contact Languages: A Comprehensive Guide (pp. 159-228). Berlin: Mouton. Meakins, Felicity. (2015). From absolutely optional to only nominally ergative: The life cycle of the Gurindji Kriol ergative suffix. In Francesco Gardani, Peter Arkadiev & Nino Amiridze (Eds.), Borrowed Morphology (pp. 189-218). Berlin: Mouton de Gruyter. Meakins, Felicity. (2016a). Mixed languages. In Mark Aronoff (Ed.), Oxford Research Encyclopedia of Linguistics. Oxford: Oxford University Press. Meakins, Felicity. (2016b). No fixed address: The grammaticalisation of the Gurindji locative as a progressive suffix. In Felicity Meakins & Carmel O’Shannessy (Eds.), Loss and Renewal: Australian Languages Since Colonisation (pp. 367396). Berlin: Mouton de Gruyter. 41 Meakins, Felicity & Cassandra Algy. (2016). Deadly reckoning: Changes in Gurindji children's knowledge of cardinals. Australian Journal of Linguistics, 36(4). Meakins, Felicity, Caroline Jones & Cassandra Algy. (2016). Bilingualism, language shift and the corresponding expansion of spatial cognitive systems. Language Sciences, 54, 1-13. Meakins, Felicity & Carmel O’Shannessy. (2005). Possessing variation: Age and inalienability related variables in the possessive constructions of two Australian mixed languages. Monash University Linguistics Papers, 4(2), 43-63. Meakins, Felicity & Carmel O’Shannessy. (2010). Ordering arguments about: Word order and discourse motivations in the development and use of the ergative marker in two Australian mixed languages. Lingua, 120(7), 1693–1713. Meakins, Felicity & Carmel O’Shannessy. (2012). Typological constraints on verb integration in two Australian mixed languages. Journal of Language Contact, 5(2), 216-246. Michaelis, Susanne, Philippe Maurer, Martin Haspelmath & Magnus Huber. (Eds.). (2013). The Survey of Pidgin and Creole Languages Vol III. Oxford: Oxford University Press. Mous, Maarten. (1994). Ma'á or Mbugu. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 175-200). Amsterdam: Uitgave IFOTT. Mous, Maarten. (2000). Selective replacement is extreme lexical reorientation. Bilingualism: Language and Cognition, 3(2), 115-116. Mous, Maarten. (2003a). The linguistic properties of lexical manipulation and its relevance or Ma'á. In Yaron Matras & Peter Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 209-236). Berlin: Mouton de Gruyter. Mous, Maarten. (2003b). The making of a mixed language: The case of Ma'á/Mbugu (Vol. 26). Amsterdam: John Benjamins. Muysken, Pieter. (1980). Sources for the study of Amerindian contact vernaculars in Ecuador. Amsterdam Creole Studies, 3(66-82). Muysken, Pieter. (1981). Halfway between Quechua and Spanish: The case for relexification. In Arnold Highfield & Albert Valdman (Eds.), Historicity and Variation in Creole Studies (pp. 52-78). Ann Arbor: Karoma. Muysken, Pieter. (1994a). Callahuaya. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 207-211). Amsterdam: Uitgave IFOTT. Muysken, Pieter. (1994b). Media Lengua. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 201-205). Amsterdam: Uitgave IFOTT. Muysken, Pieter. (1997a). Callahuaya. In Sarah Thomason (Ed.), Contact languages: A wider perspective (Vol. 17, pp. 427-447). Amsterdam: John Benjamins. Muysken, Pieter. (1997b). Media Lengua. In Sarah Thomason (Ed.), Contact languages: A wider perspective (pp. 365-426). Amsterdam: John Benjamins. Muysken, Pieter. (2013a). Media Lengua. In Susanne Michaelis, Philippe Maurer, Martin Haspelmath & Magnus Huber (Eds.), The survey of pidgin and creole languages, Vol III (pp. 143-148). Oxford: Oxford University Press. 42 Muysken, Pieter. (2013b). Two linguistic systems in contact: Grammar, phonology, and lexicon. In Tej Bhatia & William Ritchie (Eds.), The Handbook of Bilingualism and Multiculturalism. Malden, USA: Blackwell. Muysken, Pieter. (Ed.) (2008). From Linguistic Areas to Areal Linguistics. Amsterdam: John Benjamins. Myers-Scotton, Carol. (2003). What lies beneath: Split (mixed) languages as contact phenomena. In Yaron Matras & Peter Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 73-106). Berlin: Mouton de Gruyter. Nespor, Marina & Irene Vogel. (1986). Prosodic phonology. Dordrecht, Holland: Foris Publications. Nordhoff, Sebastian. (2009). A Grammar of Upcountry Sri Lanka Malay Utrecht: LOT Dissertation Series 226. Nordhoff, Sebastian. (Ed.) (2012). The Genesis of Sri Lanka Malay: A Case of Extreme Language Contact. Leiden: Brill. O’Shannessy, Carmel. (2005). Light Warlpiri: A new language. Australian Journal of Linguistics, 25(1), 31-57. O’Shannessy, Carmel. (2006). Language contact and children's bilingual language acquisition: Learning a mixed language and Warlpiri in northern Australia. (PhD), University of Sydney, Sydney. O’Shannessy, Carmel. (2008). Children's production of their heritage language and a new mixed language. In Jane Simpson & Gillian Wigglesworth (Eds.), Children's language and multilingualism: Indigenous language use at home and school (pp. 261-282). New York: Continuum. O’Shannessy, Carmel. (2009). Language variation and change in a north Australian Indigenous community. In Dennis Preston & James Stanford (Eds.), Variationist Approaches to Indigenous Minority Languages (pp. 419-439). Amsterdam: John Benjamins. O’Shannessy, Carmel. (2011a). Competition between word order and case-marking in interpreting grammatical relations: A case study in multilingual acquisition. Journal of Child Language, 38(4), 763-792. O’Shannessy, Carmel. (2011b). Young children's social meaning making in a new mixed language. In Ute Eikelkamp (Ed.), Growing up in central Australia: New anthropological studies of Aboriginal childhood and adolescence (pp. 131-154). New York: Berghahn Books. O’Shannessy, Carmel. (2012). The role of code-switched input to children in the origin of a new mixed language. Linguistics, 50(2), 305-340. O’Shannessy, Carmel. (2013). The role of multiple sources in the formation of an innovative auxiliary category in Light Warlpiri, a new Australian mixed language. Language, 89(2), 328-353. O’Shannessy, Carmel. (2016). Entrenchment of Light Warlpiri morphology. In Felicity Meakins & Carmel O’Shannessy (Eds.), Loss and Renewal: Australian Languages since Colonisation (pp. 217-251). Berlin: Mouton de Gruyter. O’Shannessy, Carmel & Felicity Meakins. (2012). Comprehension of competing argument marking systems in two Australian mixed languages. Bilingualism: Language and Cognition, 15(2), 378-396. Papen, Robert. (1987a). Can two distinct grammars coexist in a single language? The 43 case of Metif. Paper presented at the Proceedings of the 10th Annual Conference of the APLA. Papen, Robert. (1987b). Linguistic variation in the French component of Metif grammar. Paper presented at the Papers of the 18th Algonquian Conference. Papen, Robert. (2003). Michif: One phonology or two. University of British Columbia Working Papers in Linguistics, 12, 47-58. Papiha, S., Mastana, S. & Jayasekara, R. (1996). Genetic variation in Sri Lanka. Human Biology, 68, 707-737. Paunonen, Heikki. (2006). Vähemmistökielestä varioivaksi valtakieleksi. In Kaisu Juusela & Katariina Nisula (Eds.), Helsinki Kieliyhteisönä (pp. 142-161). Helsinki: Helsingin yliopiston suomen kielen ja kirjallisuuden laitos. Prichard, Hilary & Kobey Shwayder. (2014). Against a Split Phonology of Michif. In University of Pennsylvania Working Papers in Linguistics (Vol. 20, p. 1). Pennsylvania: University of Pennsylvania. Retrieved from http://repository.upenn.edu/pwpl/vol20/iss1/29 Rhodes, Richard. (1977). French Cree—a case of borrowing. In William Cowan (Ed.), Actes du Huitième Congrès des Algonquinistes (pp. 6–25). Ottawa: Carleton University. Rhodes, Richard. (1986). Métchif: A Second Look. Paper presented at the Proceedings of the 16th Congress of Algonquianists. Rhodes, Richard. (1987). Les Contes Metif—Metif Myths. In William Cowan (Ed.), Papers of the Eighteenth Algonquian Conference (pp. 297–301). Ottawa: Carleton University. Rhodes, Richard. (2001). Text Strategies in Métchif. In H. C. Wolfart (Ed.), Papers of the Thirty-second Algonquian Conference (pp. 455–469). Winnipeg: University of Manitoba. Rhodes, Richard (2013). The Michif Dictionary and Language Change in Métchif. In Karol Hele & J. Randolph Valentine (Eds.), Papers of the Forty-First Algonquian Conference. New York: SUNY Press. Rosen, Nicole. (2000). Non-Stratification in Michif. University of Toronto, Toronto. Rosen, Nicole. (2003). Demonstrative position in Michif. The Canadian Journal of Linguistics/ La Revue Canadienne de Linguistique, 48(1), 39–69. Rosen, Nicole. (2006). Language contact and Michif stress assignment. Sprachtypol. Univ. Forsch (STUF), 59, 170-190. Rosen, Nicole. (2007). Domains in Michif Phonology. (PhD), University of Toronto, Toronto. Rosen, Nicole, Jesse Stewart & Olivia Cox. (forthcoming). A comparative analysis of Michif, Métis French, and Cree vowel spaces. Ross, Malcolm. (2001). Contact-induced change in Oceanic languages in north-west Melanesia. In A. Y. Aikhenvald & R. M. W. Dixon (Eds.), Areal diffusion and genetic inheritance: Problems in comparative linguistics (pp. 134-166). Oxford: Oxford University Press. Ross, Malcolm. (2006). Metatypy. In Keith Brown (Ed.), Encyclopedia of language and linguistics (pp. 95-99). Oxford: Elsevier. Sandefur, John. (1979). An Australian creole in the Northern Territory: A description of Ngukurr-Bamyili dialects (Part 1). Darwin: SIL. 44 Sandman, Erika. (2012). Bonan Grammatical Features in Wutun Mandarin. SuomalaisUgrilaisen Seuran Toimituksia = Mémoires de la Société Finno-Ougrienne, 264, 375-387. Sandman, Erika & Camille Simon. (2016). Tibetan as a “model language” in the Amdo Sprachbund: Evidence from Salar and Wutun. Journal of South Asian Languages and Linguistics, 3(1), 85-122. Saussure, Ferdinand de. ([1916] 1983). Course in General Linguistics (R. Harris, Trans.). London: Duckworth. Schaengold, Charlotte. (2003). The Emergence of Bilingual Navajo: English and Navajo Languages in Contact Regardless of Everyone's Best Intentions. In Brian Joseph, Johanna DeStefano, Neil Jacobs & Ilse Lehiste (Eds.), When Languages Collide (pp. 235-254). Columbus, Ohio: Ohio State University Press. Sekerina, Irina. (1994). Copper Island (Mednyj) Aleut (CIA): A mixed language. Languages of the World, 8, 14-31. Shappeck, Marco. (2011). Quichua-Spanish language contact in Salcedo, Ecudor: Revisiting Media Lengua syncretic language practices. University of Illinois, Illinois. Slomanson, Peter. (2006). Sri Lankan Malay morphosyntax: Lankan or Malay. In Ana Deumert & Stephanie Durrleman (Eds.), Structure and variation in language contact (pp. 135-158). Amsterdam: John Benjamins. Slomanson, Peter. (2007). On the areal character of Sri Lanka Negation. Paper presented at the Society for Pidgin and Creole Linguistics, Amsterdam. Smart, Bath Charles. & Henry Thomas Crofton. (1875). The Dialect of English Gypsies. London: Asher. Smith, Ian. (1977). Sri Lanka Creole Portuguese Phonology. International Journal of Dravidian Linguistics, 7, 248-406. Smith, Ian. (1978). Sri Lanka Creole Portuguese Phonology. Trivandrum: Kerala University Co-operative Stores Press Ltd. Smith, Ian. (1979a). Convergence in South Asia: A creole example. Lingua, 48, 193-222. Smith, Ian. (1979b). Substrata versus universals in the formation of Sri Lanka Portuguese. In Peter Mühlhäusler (Ed.), Papers in Pidgin and Creole Linguistics 2. Canberra: Pacific Linguistics. Smith, Ian. (1984). The development of morphosyntax in Sri Lanka Creole Portuguese. In Mark Sebba & Loreto Todd (Eds.), Papers from The Creole Conference, York University (pp. 291-301). York University: York University. Smith, Ian. (2001). Creolization and convergence in morphosyntax: Sri Lanka Portuguese and Sourashtra nominal marking typology. In Peri Bhaskararao & Karumuri Venkata Subbarao (Eds.), The Yearbook of South Asian Languages and Linguistics, 2001: Tokyo Symposium on South Asian Languages : Contact, Convergence and Typology (pp. 391-409). New Delhi: Sage. Smith, Ian. (2003). The provenance and timing of substrate influence in Sri Lanka Malay: Animacy and number in accusative case marking. Paper presented at the SALA 23, Austin, Texas. Smith, Ian. (2012). Hijacked constructions in second language acquisition: Implications for Sri Lanka Malay. In Sebastian Nordhoff (Ed.), The Genesis of Sri Lanka Malay (pp. 195–232). Leiden: Brill. 45 Smith, Ian & Scott Paauw. (2006). Sri Lanka Malay: Creole or convert. In Ana Deumert & Stephanie Durrleman (Eds.), Structure and variation in language contact (pp. 159-182). Amsterdam: John Benjamins. Smith, Ian, Scott Paauw & Bachamiya Hussainmiya. (2004). Sri Lankan Malay: The state of the art. In R. Singh (Ed.), Yearbook of South Asian Languages 2004 (pp. 197215). Berlin: Mouton de Gruyter. Smith, Norval. (2000). Symbiotic mixed languages: A question of terminology. Bilingualism: Language and Cognition, 3(2), 122-123. Stewart, Jesse. (2011). A Brief Descriptive Grammar of Pijal Media Lengua and an Acoustic Vowel Space Analysis of Pijal Media Lengua and Imbabura Quichua. (Master of Arts), University of Manitoba, Winnipeg. Stewart, Jesse. (2013). Stories and traditions from Pijal: Told in Media Lengua Charleston: CreateSpace. Stewart, Jesse. (2014). A comparative analysis of Media Lengua and Quichua vowel production. Phonetica, 7(3), 159-182. Stewart, Jesse. (2015a). Intonation patterns in Pijal Media Lengua. Journal of Language Contact, 8(2), 223-262. Stewart, Jesse. (2015b). Production and perception of stop consonants in Spanish, Quichua, and Media Lengua. University of Manitoba, Winnipeg (Canada). Stewart, Jesse. (2018a). Voice onset time production in Spanish, Quichua, and Media Lengua. Journal of the International Phonetic Association, 48(2), 173–197. Stewart, Jesse. (2018b). Vowel perception by native Media Lengua, Quichua, and Spanish speakers. Journal of Phonetics, 71, 177–193. Stewart, Jesse. (2013). Stories and traditions from Pijal: Told in Media Lengua (1st ed.). Charleston: CreateSpace. Stewart, Jesse. (2018a). Voice onset time production in Spanish, Quichua, and Media Lengua. Journal of the International Phonetic Association, 48(2), 173–197. Stewart, Jesse. (2018b). Vowel perception by native Media Lengua, Quichua, and Spanish speakers. Journal of Phonetics, 71, 177–193. Stewart, Jesse, Felicity Meakins, Cassandra Algy & Angelina Joshua. (2018). The development of phonological stratification: Evidence from stop voicing perception in Gurindji Kriol and Roper Kriol. Journal of Language Contact, 11(1), 71–112 Stewart, Jesse, Felicity Meakins, Cassandra Algy, Thomas Ennever & Angelina Joshua. (forthcoming). Fickle Fricatives: Fricative and Stop Perception in Gurindji, Gurindji Kriol, Roper Kriol, and Standard Australian English. Languages in Contrast Strader, Kathleen. (2014). Michif Determiner Phrases (Masters). University of Manitoba, Winnipeg. Retrieved from http://mspace.lib.umanitoba.ca/handle/1993/23822 Taylor, Douglas & Berend Hoff. (1980). The linguistic repertory of the Island-Carib in the Seventeenth Century: The men's language: A Carib pidgin? International Journal of American Linguistics, 46(4), 301-312. Thomason, Sarah. (1997a). Ma'a (Mbugu). In Sarah Thomason (Ed.), Contact languages: A wider perspective (Vol. 17, pp. 469-487). Amsterdam: John Benjamins. Thomason, Sarah. (1997b). Mednyj Aleut. In Sarah Thomason (Ed.), Contact languages: A wider perspective (Vol. 17, pp. 449-468). Amsterdam: John Benjamins. 46 Thomason, Sarah. (1997c). A typology of contact languages. In Arthur Spears & Don Winford (Eds.), Pidgins and creoles: Structure and status (pp. 71-88). Amsterdam: John Benjamins. Thomason, Sarah. (2001). Language contact: An introduction. Edinburgh and Washington, DC: Edinburgh University Press and Georgetown University Press. Thomason, Sarah. (2003). Social factors and linguistic processes in the emergence of stable mixed languages. In Yaron Matras & Peter Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 21-40). Berlin: Mouton de Gruyter. Thomason, Sarah. (Ed.) (1997d). Contact languages: A wider perspective. Amsterdam: John Benjamins. Thomason, Sarah & Terence Kaufman. (1988). Language contact, creolization, and genetic linguistics. Berkeley: University of California Press. van Coetsam, Frans. (1988). Loan Phonology and the Two Transfer Types in Language Contact Berlin: Walter de Gruyter. van Gijn, Rik. (2009). The phonology of mixed languages. Journal of Pidgin and Creole languages, 24(1), 93-119. van Rheeden, Hadewych. (1994). Petjo: The mixed language of the Indos in Batavia. In Peter Bakker & Maarten Mous (Eds.), Mixed languages: 15 case studies in language intertwining (pp. 223-237). Amsterdam: Uitgave IFOTT. Walworth, Mary. (2015). The Language of Rapa Iti: Description of a Language in Change. (PhD), University of Hawai'i at Manoa. Winford, Don. (2003). An introduction to contact linguistics (Vol. 33). Malden, USA: Blackwell.