It is true, of course, that standards of appropriateness in language are not

uniform. Usage changes from generation to generation and it varies from one dialect or

style to another. What may be appropriate lexical usage in one language variety, may

be inappropriate — or even ostracised as 'vulgar', 'slang' or 'taboo" — in another.

The word height, for example, may appear in the form heighth (probably on the

analogy with width and length); in this form it is marked by the dictionaries of the

standard literary language as either 'dialect' (e.g. in Webster's Third New

48

Internationa} Dictionary, 1961) or "obsolete' {Oxford English Dictionary, 1933). The lexicographer will often find himself in a situation in which he must decide on whether an expression is current or deviant, appropriate or inappropriate, typical or abnormal for the variety of the language which he has set out to describe.

Most general dictionaries which have been published in the major worid languages deal predominantly or exclusively with the so-called standard (cultivated/educated/literary rather than broad/common/illiterate) norms of usage. Because dictionary compilers sometimes find themselves together with grammarians and language teachers — in the role of arbiter or authority on questions of usage, they willy-nilly contribute to the continuous development of linguistic standartisation. This is sometimes referred to as 'codification', a task which imposes grave responsibilities. It may be argued that the large monolingual dictionaries (Grimm's Deutsches Worierbuch, Murray's Oxford English Dictionary, Robert's Dictionnaire) have been performing important standardising functions akin to the deliberate efforts of language planners in smaller and culturally less developed communities. Some languages which extend over a munber of separate national territories can develop competing regional varieties, e.g. British/American English, German/Austrian/Swiss German, Iberian/South American Spanish, Egyptian/Iraqi Arabic, etc., and consequently may be liable to different kinds of lexical codification in the form of dictionaries.

Codifying usage is an important, but not the chief objective of dictionaries. Most lexicographers derive at least some satisfaction from the knowledge that the product of their labours can help ordinary language users in situations of communicative conflict or deficit. Defining the range of meaning of a particular word is one such source of difficulty which we have already encountered. Other problems which make people turn to dictionaries for help include uncertainties about spelling and pronunciation, curiosity about the origin of a word or expression, the search for suitable synonyms in composition, translating from or into a foreign language, etc. The final section in this chapter will take up this theme in greater detail.

3. The lexeme as the basic unit in dictionary-making

To bring order into the diffuse mass of a language's vocabulary, the lexicographer needs to refine his conceptual tools. The pre-scientific notion of the 'word', for instance, is inadequate to capture all the complexities; does it refer to a sequence of sounds (phonological word), of letters (graphic word), is it a formal unit composed of several building blocks (morphological word) or a unit expressing a basic meaning (semantic word)? Are all the inflectional and derivational variants [high, higher, highest, height, heights, heighten, heightens, heightened, high-class, highly strung) one unit or several?

Since Ferdinand de Saussure introduced his two-sided model of the sign as the basic linguistic unit, specialists in lexical and semantic analysis have adapted and modified it to their own ends. The word as a sign is said to have two related sides or aspects, (1) a formal shape or phonetic/graphic image and (2) a semantic content or meaning/sense. To take Saussure's own example, the sign expressing the notion of 'tree' is made up of (1) the acoustic or graphic form [tn:] or tree and (2) the mental concept or meaning that we have of trees in our heads.

The problem with this view of the word as a two-sided sign is that very often the relation between form and meaning is not one-to-one, i.e. there may be more than one meaning expressed by one and the same word-form, or a similar meaning may be shared by several different word-fonns. Examples of the former would be the various senses of tree ('tall woody plant', 'branch diagram', 'branch-like growth', etc.) or high ('tall', 'numerically large', 'elated' etc.), examples of the latter would be the near synonymous groups tree plan! shrub, or high elevated great tall.