Suzanne Aalberse Sociolinguistics and semilexicality in heritage languages

Suzanne Aalberse Sociolinguistics and semi-lexicality in heritage languages

Overview 1) Scenario-approach/different bilingual optimization strategies in different scenarios (social setting affects structure, different types and different extents of openness to innovation/vulnerability) 2) Semi lexicality: progressive/present tense marker + demonstrative/article 3) Moro 2016: proximity matrix of linguistic features: co-activation as a source of Dutchness (and not proficiency/fluency, WPM) 4) Mandarin data: no cluster of Dutchness, Different sources of innovation for superficially same innovation, more in depth structure plus more detailed scenario necessary, not one heritage scenario

• Languages do not interact in a single way, but rather in many different ways, depending on the social setting of the contact (Thomason and Kaufman (1988) , Van Coetsem (1988), Muysken (2010, 2013). The scenario approach & bilingual optimization strategies (Muysken 2013: 710)

• UP • L 1 • L 2 • L 1/L 2 Bilingual optimzation strategies Typ hier de footer 4

• • • Factors in determining bilingual optimization strategy Typ hier de footer • • Prestige L 1 & L 2 Shared knowledge L 1 Shared knowledge L 2 Language distance Number of situations where both L 1 and L 2 can be used Age of acquisition L 2 (Ross 2013) Language use patterns Linguistic beliefs (purism) Literacy Metalinguistic awareness 5

Bilingual language play • People use language to portray an image of themselves • Inserting contact effects (words from a different language, literal translations, accent) can be an act of identity. Typ hier de footer 6

SEMI-LEXICALITY: Language contact studies show contact-induced grammaticalization, of interest now: demonstratives as articles and progressive markers as present tense markers. Typ hier de footer 7

Formal factors: CP domain, partial overlap, interface Heritage scenario and contact induced grammaticaliz ation Social: immigrant situation stimulates convergence (Matras 2009): • maintain language out of language loyalty, rather than change from above, ‘stability from above’ • reduce load of selection and inhibition associated with speaking two languages by allowing patterns in the languages becoming more similar. Eg to converge. • Wish of language loyalty different for different speakers • Need for selection and inhibition load different in different speakers

Semi-lexicality and incipient contactinduced grammaticaliza tion (as a form of convergence) Numeral one/indefinite article /demonstrative/definite article (cf. Backus, Dogruoz & Heine 2011, Aalberse & Moro 2014, Moro 2016, Moro 2018, Aalberse Zou & Andringa 2017, Aalberse, Andringa, Faber & Lippe 2020) Progressive marker as present tense marker (Moro 2016, 2017)

Moro (2016, 2018): heritage Ambon Malay: proximity matrix, Dutch features cluster together related to sociolinguistic profile (sequential bilingual outside wards: coactivation)

How about progressives and demonstrives in Mandarin in NL? NL quantity: significantly more demonstrativces NL qualitative differences demonstratives: encoding all arguments: NL no overall extension progressives, but extension in some lexical verbs

Results: no cluster like in Malay • High rates of demonstratives do not predict higher rates in progressives in Mandarin in NL • Sociolinguistic profile cannot predict high/low use in either domain • Sociolinguistic message: linguistic structure looked at too crude • Social situation more diverse Chinese in NL than Moluccans in NL Typ hier de footer 12

Future plans (with Shi Meng, Josje Verhagen & Francesca Moro) • fluency cover-ups: contexts that trigger disfluency (low frequency verbs) • Convergence/contact induced grammaticalization: finite contexts/ infinite contexts. Sybesma (2007) suggest that control verbs like zhunbei ‘prepare’ shefa ‘try’, dasuan ‘plan’, quan ‘persuade’, bi ‘force’ and qing ‘invite’ select non finite clauses. One would expect zai not to occur in clauses selected by these verbs if zai would function as a present tense marker or a finiteness marker. • If finiteness is motivated by parallel input as argued in Moro (2016), one would expect that the increased use of zai could be primed in a cross-linguistic priming experiment (Kootstra & Muysken 2018).

Overview 1) Scenario-approach/different bilingual optimization strategies in different scenarios (social setting affects structure, different types and different extents of openness to innovation/vulnerability) 2) Semi lexicality: progressive/present tense marker + demonstrative/article 3) Moro 2016: proximity matrix of linguistic features: co-activation as a source of Dutchness (and not proficiency/fluence, WPM) 4) Mandarin data: no cluster of Dutchness, Different sources of innovation for superficially same innovation, more in depth structure plus more detailed scenario necessary, not one heritage scenario

Scenario approach lessons • Include scenario of your participants, think what factors motivate innovations (lack of use, language play, frequent co-activation, lack of literacy) • Realize your outcome can probably not be replicated in other social context • What you think is completely impossible, becomes more impossible given the right scenario (‘you must self the slingers uphanging’/)

Thank you! s. p. aalberse@uva. nl