Special Issue on Fluid Construction Grammar in Constructions and Frames

Eds. Luc Steels and Katrien Beuls

Despite the growing influence of Construction Grammar in many different areas of language research, we still lack a widely used computational platform for defining and testing construction grammars and for studying constructional processing in parsing, producing and learning grammars. Such a platform is needed because otherwise we cannot hope to develop and test construction grammars with wide empirical coverage and we cannot use construction grammars as the core of natural language processing applications. Various groups are currently making significant progress in developing operational platforms. This special issue of the journal Constructions and Frames focuses on one of them, namely Fluid Construction Grammar (FCG).

Each of the contributions published in the special issue is supplemented with a web demonstration, that allows the reader to test the operationalized grammar fragment by hand. The following contributions have been received (click on the title to go to the web demo):

  1. Basics of Fluid Construction Grammar, by Luc Steels.
    The first contribution `Basics of Fluid Construction Grammar’ by Luc Steels explains the core ideas of FCG, in particular how constructions are represented and processed. FCG uses many mechanisms already familiar to formal and computational grammarians, in particular, feature structures, unification, and a problem solving architecture supporting search. But these devices are used in novel ways in order to implement the major tenets of construction grammar, namely the tight integration of syntax and semantics, the smooth continuum between lexicon and grammar and the use of multiple perspectives (phrase structure, functional structure, case structure, information structure, etc.).
  2. Robust Processing of the Dutch Verb phrase, by Paul Van Eecke.
    The Dutch verb phrase is a big challenge for formal models because of  three unusual properties: (i) modal auxiliaries can be stacked, so you can say `hij moet kunnen komen’ (literally: he must can come) to express `he must be able to come’, (ii) perfect aspect is expressed by adding a perfect auxiliary, as in `Hij zal moeten hebben gesprongen’ (lit: he will must have jumped) to express `he will have to have jumped.’, and (iii) there is variation in word order of the auxiliaries, as in `hij zal moeten hebben gesprongen’ (he must have jumped) which can also be expresed as `hij zal moeten gesprongen hebben’ or `hij zal gesprongen moeten hebben’. The paper not only provides a fully functioning operational account of the Dutch verb phrase using Fluid Construction Grammar, but also explores how grammatical processing can become more robust using re-entrance. Re-entrance means that in case of problems in parsing an utterance due to ungrammaticalities, missing linguistic knowledge, etc., the listener first reconstructs as much of the meaning of an ungrammatical phrase as possible by flexible parsing and then re-produces a new utterance from this partial meaning. This technique has important applications for building language tutoring systems or in insight language learning.
  3. A Fluid Construction Grammar Account of the English Auxiliaries and Verb Phrase, by Remi van Trijp.
    This contribution discusses the long ranging debate in linguistics between proponents of the main-verb analysis, in which auxiliaries are considered to be main verbs and the rest of the VP is taken as a complement, and the lexical verb analysis in which auxiliaries are helpers that express functions like tense, aspect, modality and voice of the lexical verb. van Trijp shows how a constructional analysis can resolve this dichotomy. He proposes a `subscription design pattern’ inspired by Fillmore’s proposal to see the slots in a construction as different offices requiring various qualifications.
  4. An Open-ended Construction-based Grammar for the Spanish Verb Phrase, by Katrien Beuls.
    The contribution on the Spanish verb phrase by Katrien Beuls demonstrates how a multi-dimensional morphological system, with over 90 possible conjugated forms for a single verb can be captured by means of constructions of different types, that each combine information from multiple linguistic levels. Morphological constructions use functional and syntactic information to pick the right morpheme for a given base form. Phonological constructions convert stems or suffixes according to the stress pattern or the conjugation class. Grammatical constructions handle the temporal and aspectual conceptualizations of the speaker and translate them in syntactic categorizations that are in turn handled by morphological constructions. The grammar that is presented in the paper covers all Spanish verb forms and includes strategies to handle new verbs.
  5. Russian verbs of motion and their aspectual partners in Fluid Construction Grammar, by Katrien Beuls, Yana Knight, and Michael Spranger.
    The final article by Katrien Beuls, Yana Knight and Michael Spranger tackles a subset of Russian verbs, namely verbs of motion. Russian motion verbs are typically considered as exceptional but it turns out that there is at the core a highly complex regular aspectual system based on stem directionality. Beuls et al. show how Russian motion verbs and their aspectual partners can be implemented and processed with Fluid Construction Grammar by relying on the cooperation of a number of constructions; lexical and morphological constructions but also directionality constructions, perfective/imperfective constructions, aktionsart constructions, trajectory constructions and phrasal constructions.