SE-LFG14 (10/05/2014, SOAS)

14th South of England LFG Meeting

We are planning the 14th South of England LFG meeting, a student-oriented meeting for presentations and discussion of various topics from an LFG perspective. It is planned for Saturday, 10 May 2014, at SOAS, London. Please feel free to attend if you are interested, or if you would like more information please get in touch with Mary Dalrymple.

Meeting details:

Saturday, 10 May 2014, Room 4426 (4th floor), SOAS main building, Russell Square.

For directions to SOAS see here: http://www.soas.ac.uk/visitors/location/maps/ (also see map opposite). To view any planned engineering works affecting your journey within London, click here.

Sign in at the reception when you arrive to get a visitor sticker that allows you to enter the SOAS main building.

Meeting agenda:

11:00-12:00: Peter Austin (SOAS): Argument coding in linked clauses in Australian Aboriginal languages: towards a typology

When clauses are combined in complex constructions, it is often the case that argument coding differs from what is found in main (unlinked) clauses. For example, particular grammatical functions may not be directly expressed by NPs (eg. in functionally controlled complements or adjuncts) or may be coded in non-canonical case forms (eg. in non-argument cases like allative or ablative). This paper explores argument coding across a range of Australian Aboriginal languages with the goal of identifying a typology of coding strategies in linked clauses.

12:00-12:45:Louise Mycock (University of Oxford) and John Lowe (University of Oxford): Defining Discourse Functions

Dalrymple & Nikolaeva (2011) present an approach to information structure within LFG which makes critical use of an s-structure feature DF. The value of DF (TOPIC, FOCUS, BACKGROUND or COMPLETIVE) specifies the discourse function of a referent, which in turn determines the role that the relevant meaning constructor will bear at the level of i-structure. As formulated, this approach treats information structure categories as atomic. While this is unproblematic when analysing a language in which some or all of those discourse functions are associated with a particular syntactic position (e.g. Hindi/Urdu; Butt & King 1996), such an approach obscures generalizations about all the ways in which information structure can be encoded. We propose to address this important issue by replacing DF with a set of s-structure attributes that more accurately define discourse functions on the basis of principled decompositions of the notions of relative salience and newness, concepts which are often vaguely defined in the wider literature on information structure but which are understood to be fundamental to its analysis. Under our approach, s-structure attributes are available to associate a specific c-structure position not only with a feature ‘bundle’ that defines one of the discourse functions used in Dalrymple & Nikolaeva (2011) (as in a discourse-configurational language with an identifiable Topic position such as Hungarian), but also with specific attribute-value pairs to capture other equally important facts about the syntax–information structure relationship. This more refined analysis improves the descriptive adequacy of LFG’s model of information structure, permitting it to account for a wider range of data and generalizations over the different information structure categories attested cross-linguistically.

12:45-14.15: lunch

14.15-15.00: John Lowe (University of Oxford) and Louise Mycock (University of Oxford): Representing Information Structure

Building on the preceding talk, in this paper we discuss previous approaches to the formal representation of i-structure within the LFG architecture, and propose a new model of i-structure in LFG. Our formal conception of i-structure owes much to that of Dalrymple & Nikolaeva (2011), but differs from it in several important respects. We share the intuition that the elements categorized at i-structure are meaning constructors, but conceive of the categorization in terms of set membership somewhat differently from Dalrymple & Nikolaeva. We also argue that the status of i-structure within the grammatical architecture is not the same as that of the other major projections: i-structure is best understood as a level of structuring or organization imposed upon meanings as part of the mapping from semantic structure to the ‘model’.

15.00-15.30: Charlotte Hemmings (SOAS): Voice Alternations in the Kelabit Language of Northern Sarawak

Kelabit is a Western Austronesian language of North Sarawak and like many of the other languages of North Borneo, it has a three-way system of symmetrical voice alternations which variously select the agent-like; patient-like and instrumental argument as 'subject'. This talk explores syntactic properties of the three voice alternations and discusses some potential semantic and discourse factors in determining the selection of one voice over another. In particular, it focuses on the question of pronoun alternations; differences in word-order; transitivity and relative frequency.

15.30-15.45: break

15.45-16.15: Liselotte Snijders (University of Oxford): Warlpiri Non-Reflexive Binding

In this talk I will discuss some issues relating to non-reflexive binding in Warlpiri. Warlpiri data as presented by Simpson (1991) suggest that in non-reflexive binding, linear precedence and the functional hierarchy do not play a role. This leads to an f-command only analysis, excluding f-precedence or the functional hierarchy, which is a surprising find. The talk will give an overview of the data and explain the analysis in more detail, also contrasting it with some reflexive data from Warlpiri.

16:15: Discussion and planning for next time