Volume 57, Issue 3

In this paper concepts from cognitive linguistics are combined with assumptions and methods from corpus linguistics in order to study categorisation in everyday life. The notion of categorisation, although ambiguous, will cover both the act of setting up categories and the act of identifying instances of a category. The presentation and discussion of current linguistic thinking in “categorisation theory” will entail a reassessment of the notion of binarism as popularised in structural linguistics, preferably following Langacker’s A ® B principle.

The paper first discusses the criteria by means of which affixes have been or may be classified and sets up ten parameters: language origin, linguistic form, position, word class, base structure, meaning, effect, productivity, valency and closure. It then looks at studies investigating the combinability of affixes in Modern English and assesses to what extent and in which hierarchies these parameters have been taken into account and how this affects the adequacy of description

English grammar as presented to schoolchildren, university students, and the general public is in a state resembling what biology might be like if teachers had paid no attention at all to On the Origin of Species (1859) or anything that followed. In fact that may be an understatement: English grammar has been drifting along unrevised not just since Darwin published his magnum opus, but since before he was born. The main outlines of English grammar are presented in 2009 just as they would have been in 1909, when Jespersen produced the first volume of his magnificent but mostly ignored 7-volume grammar, or in 1809, the year of Darwin’s birth. In this short article I lay out a small amount of the evidence for this claim, concentrating on the very basic topic of the ‘notional’ definitions of grammatical categories or ‘parts of speech’.

Linguistics and lexicography are two different disciplines. One discipline can use terms from another discipline, but quite often it has to make its own terms or give certain terms a certain meaning in the discipline. Of historical reasons lexicography is often seen as a subdiscipline of linguistics, metalexicographers therefore quite often use linguistic terms exactly in the same way as in linguistics. This is not appropriate for a lot of reasons. The most important one is that normally - also within one and the same discipline - different terminological systems are needed for different aims: the only criteria by which a proposed scheme of classification can be judged is that of its effectiveness in attaining a certain fixed aim. This thesis is the starting and also ending point of this paper about word class and word combination categories in linguistics and lexicography.

Homonymy and polysemy are concepts developed in relation to linguistic theory. How should they be treated by lexicography and dealt with in dictionaries? Should they be taken over as they are; should they be transformed and modified; or should they be completely rejected as useless and irrelevant? The answers to these questions depend basically on the idea that each scholar has of the relationship between linguistics and lexicography. The contribution discusses this relationship and argues that lexicography should be considered an independent discipline which should develop its own theory, concepts, and methods. On this basis, it analyses the treatment of homonymy and polysemy in dictionaries and proposes a functional approach to the problem.