Abstract Zaefferer & Sodian

This project aims to examine a new approach to understanding language universals, tracing them back partly to universals of prelinguistic cognitive development. Therefore it is designed as an interdisciplinary cooperation between linguistics and developmental psychology. It conceives of the relation between thought and language neither as strictly modular (Fodor) nor as holistic (Langacker), instead it regards the mental representation of language as an integrated module, a composite of cognitive systems rooted both phylo- and ontogenetically in other capacities (sensorimotor skills and prelinguistic concepts), with which it co-develops and continues to be closely connected via an intense exchange of information taking place not only inside individuals (entrenchment), but also across them (conventionalization) in the form of Mental Content Sharing.

Accordingly, it is necessary to ascertain (a) which other components of human cognition its linguistic components are based on, (b) how the concepts to be investigated are coded cross-linguistically, and (c) whether and, if so, how those codings retroact on non-linguistic cognition (Whorf effects). This way an Archimedean point outside of language can be reached from which the circularity of purely linguistic reasoning can be overcome and new insights can be gained.

The reason for selecting the example of triadic interactions (Agent, Co-agent, Referent) as object of investigation is the assumption that this domain includes categories which are essential and specific for humans. The interest in modal concepts, i.e. concepts with propositions in the referent role, is motivated by their central role in the conceptual system of our species with its very elaborated Theory of Mind. Finally the focus is on participant roles because these are always codable with grammatical means and for the latter an especially high degree of entrenchment and conventionalization can be assumed.

Since cross-linguistic research on the coding of modal categories is mainly focused on modal auxiliaries and adverbs (Frawley 2006; for an exception cf. Zaefferer 2001, 2005), the expression of such concepts by verbs and nouns is rather understudied. With regard to the non-linguistic conceptual development, it is conceivable that action-related, epistemic and volitional concepts are ontogenetically prior to deontic concepts. Recent research on the socio-cognitive development in early childhood suggests early representation of action intentions (Woodward 2005; Gergely & Csibra 2003; Tomasello et al. 2005). Already in their first year of life infants represent action goals of individuals as well as goals of different participants in social interaction. Numerous findings indicate early cooperation and representation of epistemic states of participants in cooperative communication (Liszkowski et al. 2008). Gaze duration patterns found in habituation experiments are consistent with the interpretation that 13 to 18 month-old babies can represent false belief and hence possess a Theory of Mind (Surian, Caldi & Sperber 2007). However, since alternative interpretations based on behavioural rules and association patterns have been proposed as well, cf. Perner & Ruffman (2005), Perner (in press), further investigations are required.

The literature on ditransitives, constructions that code triadic events with participants such as Agent, Theme and Recipient, (e.g. Malchukov et al. 2007) shows that there is a strong typological preference for indirective alignment (the marker for Recipients differs from those of all other major roles), an asymmetry which Haspelmath (2005) proposes to explain in terms of the model function of local transfer. We submit that considering the model role of propositional content sharing (a triadic modal category) alongside with physical transfer and its prelinguistic categorization provides a more realistic explanation of this universal preference. Finally, since Boroditsky (fortch.) has shown that even highly arbitrary linguistic features such as grammatical gender influence nonlinguistic behaviour, we want to examine whether speakers of indirective alignment languages tend to separate the concepts of Recipient and Patient more sharply than speakers of other languages.

The hypothesis to be examined is based on the presumption that concepts which are universally established in the prelinguistic period are both very deeply entrenched and universally distributed. Linguistic codings should be preferred to the degree they structurally reflect such concepts. (Dis-)Proving this hypothesis will be conducted through a broad variety of methods, such as cross-analyses of reference grammars, corpus analyses, elicitation and experiments, including habituation and ERP studies. Thus, converging evidence is to be seen as the guiding methodological principle.

 

References

Boroditsky, L. (forthc.), “Linguistic Relativity”, in: Encyclopedia of Cognitive Science. MacMil-lan Press.

Frawley, W., ed. (2005), The expression of modality, Berlin: Mouton de Gruyter

Gergely, G., & G. Csibra (2003), “Teleological reasoning about actions: The one-year-old’s naïve theory of rational action”, Trends in Cognitive Sciences 7, 287-292.

Haspelmath, M. (2005), “Argument Marking in Ditransitive Alignment Types”, Linguistic Discovery 3, 1-21.

Liszkowski, U., M. Carpenter & M. Tomasello (2008), “Twelve-month-olds communicate helpfully and appropriately for knowledgeable and ignorant partners”, Cognition 108, 732-739.

Malchukov, A. et al (2007), Ditransitive constructions: a typological overview. Max-Plank-Institut für Evolutionäre Anthropologie, Leipzig.

Perner, J. et al. (2005), "What does 'that' have to do with point of view? The case of conflicting desires and 'want' in German", in: J.W. Astington & J. Baird, eds., Why language matters for theory of mind, New York/NY: Oxford University Press, 220-244.

Perner, J. (in press), “Who took the Cog out of cognitive science? - Mentalism in an era of anti- cognitivism”, in: P.A. Frensch et al., eds., ICP (International Congress of Psychology) 2008 Proceedings, Psychology Press.

Perner, J., & T. Ruffman (2005), “Infants’ insight into the mind: How deep?”, Science 308, 214-216.

Surian, L., S. Caldi & D. Sperber, (2007), “Attribution of beliefs by 13-month-old infants”, Psychological Science 18, 580–586.

Tomasello, M. et al. (2005), “Understanding and sharing intentions: The origins of cultural cognition”, Behavioral and Brain Sciences 28, 675 - 691.

Woodward, A. L. 2005. The infant origins of intentional understanding. In R. V. Kail (ed.) Ad-vances in Child Development and Behavior, Volume 33, Oxford: Elsevier, 229-262.

Zaefferer, D. (2001), "Modale Kategorien", in: M. Haspelmath, E. König, W. Oesterreicher, W. Raible, eds., Language Typology and Language Universals. An International Handbook, Berlin: de Gruyter, 784-816.

Zaefferer, D. (2005), "A general typology of modal categories." Tidsskrift for Sprogforskning 3/2, tema Modality. Linguistic, Philosophical and Logical Aspects of a Universal Category, 19-49.