Log in / Register
Home arrow Business & Finance arrow Handbook of Business Communication

Structure of specialized dictionaries

Information provided by a lexicographic resource is governed by a set of predefined structuring principles. Between three and seven different types of structure are differentiated in the literature. Common to all approaches are the three main structure types: the medio-, macro-, and microstructure. Each of them operates on its own level. The mediostructure (Wiegand 1989), also called megastructure (Svensen 2009) or hyperstructure (Gelpi 2004), describes the overall organization of the resource, that is, the system of cross-referencing different components of a dictionary. For instance, the mediostructure is concerned with relations between text in the main lexical articles and sub-articles of the main word-list or external sources. The macrostructure systemizes the order of the main articles across a dictionary, while the microstructure deals with the data organization within a specific article.

In addition, the macrostructure encompasses criteria for selecting headwords. Once that is done, its major task is to order them, the most common ordering being alphabetical. The macrostructure also affects the ordering of lexemes, but more in terms of internal organization, that is, syntagmatic and paradigmatic decisions (Svensen 2009: 78). When articles are ordered alphabetically, the decision on how to present compound terms, for example, noun first, is taken on the macro level. Another way of ordering articles is by semantic relationships between lexemes, as exemplified by the lemma amortization in Figure 23.2. This approach is more frequently adopted by specialized dictionaries such as the EcoLexicon ( index.htm) than by general language products.

Macrostructure based on semantic relationships (of WordNet data) Source

Figure 23.2: Macrostructure based on semantic relationships (of WordNet data) Source:

Decisions at the microstructural level are concerned with the number of information categories to be included in each article, as well as their types. These depend on the lexical categories the design team opts to incorporate, for instance definitions, phonetic transcriptions, or synonyms. This decision is, of course, governed by the user profile and the intended function. Important microstructural elements are definitions and, for multilingual resources, language equivalents. Definitions can either be lexical - i.e., they list synonyms of the headword - or conceptual, providing delimiting and/or essential characteristics of a lexeme.

In general, specialized lexicographers seem still to believe in the existence of simple one-to-one correspondences between expressions in different languages. However, in specialized contexts, too, it is possible to identify differences between semantically corresponding words in terms of synonymy, polysemy, homonymy, historical contexts, regional settings, etc., albeit to a lesser extent than in general language. The degree of equivalence has been classified as full, partial, and zero depending on the semantic coverage of two lexemes (Bergenholtz and Tarp 1995: 109). Full equivalence refers to the situation where two expressions are formally and semantically identical. Partial equivalence can occur on the pragmatic level, that is, differences in use, or the semantic level, that is, differences in lexical sense. Finally, zero equivalence refers to the absence of an equivalent in the target language(s). In the last case, the lexicographer can either construct a translation or limit the entry to a definition in the target language without any translation.

Few approaches have dealt with criteria for the selection of equivalents to avoid endless, and not very helpful, lists as in example (1) in Section 3.3. The most comprehensive set is offered by Werner (1999), who gives a number of factors that determine the selection of equivalents, as follows:

a. a dictionary’s function;

b. evaluation of expertise and knowledge of intended users;

c. criteria for selecting specialized lexical units;

d. degree of parallel structures across languages;

d. the decision as to whether one or both languages should be restricted to defined or standardized lexemes.

Of course, these criteria are interdependent. A dictionary’s function depends on its users’ expertise and knowledge, which in turn provide criteria for the lemma selection process. The degree of parallel structures refers to the extent to which full and partial equivalences exist in a specific domain and for specific language pairs. In combination with the general characteristics introduced in Section 2.2., these criteria determine the number of equivalents given in a specialized dictionary. In particular, a prescriptive dictionary will provide fewer equivalents than a descriptive one.

Examples of specialized expressions in a dictionary article are lexicographic features on a microstructural level. They show how a lexical unit is used, which means they are always tied to a specific context. If well selected, they can introduce the user to a typical context of the expression concerned. In contrast to many other lexicographic features, the information provided by examples is of an implicit nature. In fact, “well-chosen examples provide implicit encyclopedic and linguistic information, including collocational and grammatical construction possibilities” (Bergenholtz and Tarp 1995:140).

The mediostructure can cross-reference between microstructural elements, for instance by relating compounds of a headword to individual senses, on a macrostructural level between articles, or to external resources (Gouws and Prinsloo 1998). The last type of reference is a common and effective method of enriching a dictionary’s content without using up much space. It is particularly common among multilingual specialized dictionaries in an electronic format, but has also been used for printed resources.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science