Understanding linguistic categories: Temporality at the interfaces
Description of the proposed research

What are categories and why is it important to have them? Since Aristotle, we have known that speakers use categories (semantic categories and formal-grammatical categories) to describe the world and linguists use (corresponding) categories to describe languages. However, it is far from obvious what categories are and how many of them exist. Is there a universally available pool of “language categories” from which every language may choose a subset? Are apparently equivalent categories across languages really the same? For example, can we really equate the Polish perfective with what is usually referred to as “perfective” in Hindi? If grammatical categories such as ‘perfective’ or ‘modal verb’ or ‘lexical verb’ turn out to be language-specific (as is claimed, e.g., by Haspelmath 2007, 2008), we need a solid comparative base in order to define categories and consequently be able to do comparative research. Comparative research on categories should therefore be based on universally available semantic concepts (but see Newmeyer 2007 for the position that crosslinguistic formal categories exist; what is more, formal categories are – according to Newmeyer – necessary for an adequate linguistic typology). This is not an easy task given that even “categories” which are usually taken for granted, as, for example, word class categories (noun, verb, adjective ….), might turn out to be difficult to define. Many questions arise in this context. For example, are word class categories primitive or should they be defined in terms of features (e.g. N = [+N, -V], V = [-N, +V], Adj = [+N, +V] etc.)? Do lexical categories arise in a constructionist way as proposed, for example, in Marantz (1997), Borer (2005), among others? Are such categories discrete or should they be understood as “fuzzy” or “scalar” concepts? (see, among others, Haspelmath 2001, Croft 2003 for an overview). Speaking, for example, about nouns, are some nouns more typical (and thus more “nominal”) than others (cf. the prototype approach to word classes by Croft 2001, 2003)? Are word class categories universally available in all languages? That is, do all languages have nouns and verbs (and possibly adjectives) (cf. Haspelmath 2001; see also Dixon 1977, Schachter 1985, Langacker 1987, Hengeveld 1992, Sasse 1993a,b, Baker 2003, Croft 2003, among others)? All these questions are a matter of debate in the current literature (see the recent Linguistic Review volume (The Linguistic Review (2008), vol. 25), which is dedicated to the issue of linguistic / language universals).

Main research questions and aims of the project

Given that even such apparently “basic” categories are linguistically problematic, more complex concepts such as temporality pose an even bigger challenge for linguists. The task of understanding temporality is all the more demanding from a linguistic point of view since there are good reasons to think that temporality is a basic semantic concept.1

Firstly, humans seem to be the only ones that are able to talk about things that are not happening now (temporal / modal displacement)2 (cf. Hockett 1960).

Secondly, reference to past or future events is made even in the most “primitive” stages of such language systems as deaf children’s homesigns (Goldin-Meadow 2005) or the most basic pidgins or creoles (Bickerton 2008, Aitchison 1991, 1996). Thirdly, even so-called “tenseless” languages such as Kalaallisuut (Bittner 2005) or the frequently cited Hopi language (Whorf (see Carroll 1956:58)), which are claimed not to have any tense specification in the grammar, are not really “tenseless” in the sense of not being able to express temporality / temporal reference (see, e.g., Pinker 1995:63 for counterevidence against Whorf’s claim)). 3

If temporality is conceptually basic, how is it expressed in languages of the world? Are temporal relations expressed universally in the same way in different languages or do we find cross-linguistic variation in this respect? Are there any restrictions imposed on this (possible) variation? What universal patterns can be identified? By answering these and similar questions we will be able to better understand how humans linguistically categorize temporal relations. What is required to reach this goal is an integrated, multidisciplinary perspective across various linguistic interfaces. Just by studying only one language or by concentrating only on a single linguistic module (say only morphology), we will not be able to arrive at a complete picture of the phenomenon under investigation. Therefore, in our linguistic project we will approach the question of temporality from three different perspectives, each of them focusing on a different linguistic interface and relating the research question to a specific linguistic subdiscipline.


An overview of the composition of the research team

Table : The composition of the research team

Understanding linguistic categories: Temporality at the interfaces Principal investigator

Prof. dr hab. Joanna Błaszczak Coordination Supervision of subprojects SP-1, SP-2, SP-3 Synthesizing study
Subproject Researcher Relevant domain Interface Subdiscipline
SP-1 Dr Dorota Klimek-Jankowska Intra-sentential

domain ;

« main clauses »

Syntax-semantics Linguistic typology (including pidgin and creole languages)
SP-2 Dr Krzysztof Migdalski Syntax-phonology Diachronic linguistics
SP-3 Dr Patrycja Jabłońska Inter-sentential domain;

relation between main and subordinate clause

Syntax-morphology Psycholinguistics

(experimental research, especially ERP studies)

Project cohesion

The three subprojects approach the question of temporality from three different perspectives. In this sense the subprojects clearly have their own focus, so researchers can carve out their own empirical and theoretical domains. This notwithstanding, the subprojects are tightly connected to each other. Given these close interdependencies the researchers will be able to operate as a team. This in turn will enable us to gain insights from different perspectives to answer the central question of the project, namely how humans linguistically categorize temporal relations.