主講人：Professor Peter Harder（哥本哈根大學）
講座題目: Functional and procedural aspects of syntax in cognitive linguistics
Cognitive Linguistics arose as an alternative to Generative Linguistics. Naturally enough, focal interest in Cognitive Linguistics has been on those properties of language that are not focal in generative linguistics, and hence are not well-covered in Chomskyan linguistic theory.
Thus, just as autonomy is central to generative theory, the dependence of language on general cognitive capacities (those that are not specific to language) has been equally central in cognitive linguistics. Further, the sharp boundaries that are characteristic of generative linguistics also internally in language have motivated a skepticism among cognitive linguists against all sharp boundaries. These include the separation between semantics and pragmatics, and between semantics and syntax, both of which are central from a generative point of view, and have been downplayed in Cognitive Linguistics. If language as a whole is an integral part of cognition as a whole, it is natural that also internal divisions between subcompartments of the language ability have blurry outlines.
This ‘continuism’ extends also to the classical dichotomy between lexicon and grammar. For generative grammar, this division is central because the lexicon has obvious relations with the world of experience and cannot plausibly be understood as autonomous in the same sense as syntactic categories arguably can. While the existence of empty subjects in sentences like it is raining entails that the subject category can plausibly be kept apart from concrete experiential content, this is not the case for semantic categories. Cognitive Linguistics, in which a foundational assumption is that all linguistic units have a semantic as well as a phonological pole, however, does not recognize the existence of a special domain of syntax in which elements are purely formal, and therefore linguistic units are not assumed to be rigidly separable into two subcategories, grammatical vs. lexical.
While I align myself with this general orientation on the part of Cognitive Linguistics, I think it has had the unintended consequence of under-addressing the question of what is the special nature of syntactic phenomena, thus leaving this to some extent as the turf of generative grammar. This makes syntax a potential chink in the armour of cognitive linguistics, given that the existence of syntactic structure is generally recognized to be a design feature of human language. Therefore I think it is worth while for Cognitive Linguistics to upgrade its picture of what syntax means for human languages.
I am going to address two aspects of this issue:
(1) The general features of syntax in a cognitive-functional perspective – with particular focus on the role of ‘content syntax’ (understood as the combinatorial relations between meanings), cf Harder (2010, ch. 6).
(2) The special linguistic and neurocognitive characteristics of grammatical (as opposed to lexical) items - with particular focus on the procedural (as opposed to declarative) aspects of grammar. This includes the issue of grammaticalization (Boye & Harder 2012): what is it that happens when a lexical element becomes grammatical?
I argue that these aspects are underemphasized in the constructional approach to grammar, which has been the most widespread foundation for syntactic theory in Cognitive Linguistics. In the proposed alternative, I draw on two main sources, one linguistic and one neuro-cognitive: the European structural-functional tradition (e.g. Engberg-Pedersen et al. 1995), and Ullman’s research on the declarative and the procedural systems in the brain, as explored especially by Ullman (e.g., 2016).
皮特?哈德教授出席過一系列的功能語法國際會議(1988, 1990, 1992, 1994, 1998, 2000, 2004, 2006)和認知語言學國際會議(1993, 1995, 1997, 1999, 2001, 2003, 2007, 2011, 2013, 2015, 2017)。Peter Harder作為主旨發言人在諸多國際會議上發言，如1997年第4屆國際認知語言學會議，2010年3月在朗多舉辦的第10屆LAUD會議，2010年10月在俄羅斯坦波夫舉辦的俄羅斯認知語言學家協會國際認知語言學大會，以及2017年6月法國認知語言學協會舉辦的會議。