2018 Friday Poster 6395

Friday, November 2, 2018 | Poster Session I, Metcalf Small | 3pm

Leaving obligations behind: Epistemic incrementation in preschool English
A. Cournane, A. Pérez-Leroux

Introduction. Modals like must express both meanings of root obligation and epistemic inference (Ex1a-b)[1,2]. Modals generally exhibit cyclic patterns of language change, where root modal lexemes first acquire epistemic interpretations[4,5], and eventually lose all root meanings[6] before disappearing from the language[5]. The process of increasing novel uses (e.g., more epistemic uses in more grammatical contexts) with successive generations is called incrementation[3,6]. While children supposedly play a central role in this process[3,6,cf.7], evidence is generally rare[8], and non- existent for semantic changes like root→epistemic.

We report 3 experiments testing the child incrementation hypothesis. Study 1 tests whether preschool children overextend epistemic interpretations of must. A picture preference task compared modals with bare verbs (Ex.1a), which can be interpreted either as root (Ex.1a-i) or as epistemic with habitual construal (Ex.1a-ii), to modal aspect-marked constructions (Ex.1c-d) which are epistemic[2]. Studies 2 and 3 use sentence-preference tasks to test whether children prefer sentences with must over non-modal counterparts in indirect evidence (epistemic) scenarios (Ex.2)[9], and obligation scenarios (Ex.3), respectively.

Methods. All participants were English monolingual preschoolers from Toronto, Canada (where the modal cycle for must is highly advanced[5]) and dialect-matched adults. Study 1 tested 54 children and 10 adults. Participants had to choose between root and epistemic scenes (Ex.4), presented on a laptop (MATLAB_R2014a, Psychtoolbox[10]). There were 8 aspect-marked and 8 bare modal items. In Studies 2 and 3 participants had to choose between must and no-must sentences (Ex.2-3, respectively), said by two puppets in relation to a storybook picture scenario. Study 2 tested 35 children and 9 adults. Each participant heard 16 test stories, split between indirect evidence vs. actual scenarios. Study 3 participants were 52 children and 10 adults. Materials contained 10 test stories, split between obligation and actual scenarios.

Results. Study 1 results indicate preschool children approach adult rates of epistemic interpretation for aspect constructions, but for the variable-meaning bare verb constructions they start from an adult-like root bias at age 3 and acquire an epistemic bias by 5 (Fig.1) (glmer, Epistemic~AgeGroups+(1|Participant), AGEGROUPFIVE-YEAR-OLDS [beta]=1.818, <0.001**). Studies 2 and 3 demonstrate that children generally avoid choosing must sentences. Children fail to discriminate in indirect evidence contexts, rarely choosing must at 5/6yos  (Fig.2) However, they reliably discriminate in  obligation contexts (Fig.3) (glmer, MustChoice~Condition+(1|Participant)+(1|Item), PICTUREACTUAL:[beta]=-0.622, <0.003**), which suggests that their semantic choices in Study 1 do not result from insensitivity to the presence of modals nor loss of obligation interpretation.

Discussion. Our studies provide data for an epistemic-bias stage, and an avoidance pattern indicative of the final phase of the modal cycle, according to the predictions of the child incrementation hypothesis. They constitute the first experimental child evidence for standard theories of language variation and change. Why do older preschool children overgenerate epistemic must? We propose that must sentences with grammatical aspect in English show must overtly preceding aspect (Ex.5), cueing a high epistemic interpretation[11,12]. Once children reliably map must to its higher epistemic position in their 4th year of life[13,14], they overextend this higher interpretation to ambiguous sentences, gradually incrementing root→epistemic→[null symbol] with each generation.

References

[1] Kratzer, A.1977. What must and can must and can mean. [2] Hacquard, V. 2006. Aspects of Modality. [3] Labov, W. 2007. Transmission and diffusion. [4] Traugott, E. 1989. On the rise of epistemic meanings in English.[5] Bybee, J., Perkins, R., Pagliuca, W. 1994. The evolution of grammar: Tense, Aspect, and Modality in the languages of the world. [6] D’Arcy, A. & Tagliamonte, S. 2007. The modals of obligation/necessity in Canadian perspective. [7] Dasher, R. & Traugott, E. 2005. Regularity in semantic change. [8] Smith, J., Durham, M., & Fortune, L. (2009). Universal and dialect-specific pathways of acquisition: Caregivers, children, and t/d deletion. [9] von Fintel, K. & Gillies, A. 2010. Must… stay… strong! [10] Brainard & Vision. 1997. The psychophysics toolbox. [11] Brennan, V.M. 1993. Root and epistemic modal auxiliary verbs [12] Cinque, G. 1999. Adverbs and functional heads. [13] Papafragou, A. 1998. The acquisition of modality. [14] Cournane, A. 2015. Modal Development.