The Comparison about Macro-structure and Micro-structure Between Two Dictionaries

    李英杰

     As is often the case, the dictionary is consulted rather than read in that the reader’s intention about

    consultation to the dictionary is to shed light on a particular piece of information as quickly as possible, such as one word’s pronunciation, spelling, usage, and so on. The dictionary is never consulted in its entirety. Therefore, the form and content of the necessary about how to compile well in a systematical and orderly way can never be over-emphasized. Here is a question about how to offer the most efficient way of helping the reader to get what they have expected from the dictionary. In general, the contents must be predictable to a certain extent, that is to say, there must be a program of information that is repeated at each entry, or the users will never know what to expect when they look up a word. Given the objective analysis about some relative materials and some experiences from the crystallization about their extraordinarily consistent and painstaking efforts about the dictionary research, we can realize that we should highly attach importance to two aspects: macro-structure and micro-structure, if we want the dictionary popular with the people.

     The macrostructure is the result of a selection of words through the use of various criteria. As we have seen, the word-list must correspond to a set of lexical units which, in a way, exists somewhere else, if only in the minds of the users. By and large, the structural nature of the macrostructure is less clear. As every lexicographer knows, it is always possible to add or subtract an entry-word without destroying the quality of the word-list, whose unity is not clearly definable: no lexical set is defined with enough precision to forbid this. Anyway, it is totally impossible to imagine a dictionary without a macrostructure: all dictionaries necessarily have a list of entry-words, and if the list is not structured at all the book is clearly not a dictionary. No dictionary can avoid giving, at the very least, two pieces of information: the spelling of the entry-word; the entry-word belongs to the lexical set that the dictionary purports to represent. The microstructure is a rigid structure: all modern dictionaries are characterized by the uniformity of their entries, both in terms of content and in terms of layout. In modern dictionaries, some information may be missing in some entries, but this can always be interpreted by the users; information not explicitly given means normal information in any particular field of the entry; unknown information (for example an unknown etymology) is normally indicated as such. The differentiating characteristic of the dictionary lies in the fact that its two structures interact: all the entry-words that make up the macrostructure receive microstructural information, and all the words used in the microstructure must normally be included and treated in the macrostructure. As the old saying goes, ―Action speaks louder than words.‖ On the basis of the theories above, we

    can give an elaborate analysis about two different dictionaries — Collins COBUILD English Language Dictionary

    and Longman Dictionary of Contemporary English

    COB is characterized by the following aspects about the arrangement with regarda the macrostructure and microstructure:

    Firstly, tt is based on an entirely new, 7,3-million-word corpus, the COBUILD corpus, which is the property of Collins and of the University of Birmingham. All the examples provided by COB were extracted from that corpus, with only minor adaptations. This ensures a total authenticity of the language, but it also has unfortunate consequences. For example, when taken out of their original contexts, sometimes make no sense at all. For example, one meaning of every is illustrated by :‖ One woman out of every two hundred is a sufferer….‖

    Secondly, the syntactic patterns of verbs, nouns, nouns, and adjectives are indicated in coded form in a special column to the right of the main text, so that the two types of information, semantic and syntactic, are clearly separated.

    Thirdly, the words are defined in an entirely new way. Each word is inserted in a complete, grammatical sentence instead of the traditional, formulaic definition. The word is used in a typical syntactic pattern and accompanied by its typical collocates. For example, ―bemoan‖ is defined as ― If you bemoan something, you

express sorrow or dissatisfaction about it‖; the definition for careworn is ―A person who looks careworn looks

    worried, tired and unhappy‖, etc. The new style is very often convincing, but fails in places. Some definitions are long, inelegant, and sometimes even ambiguous.

    Fourthly, the macrostructure of COB is extremely synthetic, as if the dictionary were meant to be used only for the decoding of isolated, context-free forms. This is taken to extreme lengths, unknown in English-language lexicography. COB has been very successful; many learners of English find that, after all, its defining technique works quite well. A simplified version, aimed at British schoolchildren.

    LDOCE has the following different characteristics in comparison with COB:

    Firstly； it used a coding system not only for verbal construction but also for nouns and adjectives. This led to an even larger quantity of codes that the users had to memorize, or at least understand. It has a strict alphabetical ordering of all the entry-words. This sometimes makes it difficult to locate an entry. For following entry-words are treated in separate entries in the following order: put over, put over on, putrefaction, putrefactive, putrefy, putrescent, putrid, etc. with the sort of ordering, there is very little grouping of words and phrases that are semantically related.

    Secondly, it tends to distinguish as many senses as possible in polysemous entries, sometimes to absurd lengths. The most often quoted example is the verb walk: LDOCE distinguishes seven different meanings, the first two definitions being introduced by, respectively. Above all, it uses a defining vocabulary of 2,000 words, given in a list in one of the Appendices. All the definitions in the dictionary are written with 2, 000 words only. The idea has been first used in the New Method Dictionary where the 24,000 entry-words has been defined with 1,490 words only.

    Thirdly, as is known, the difficulty in the use of a controlled defining vocabulary is the precise identification of the words that are used. In the case of LDOCE, many—not all –words on the list of 2,000 are polysemous and

    frequent. Although there are only about 2166 words in the controlled vocabulary, over 24 000 of the 74000 senses defined in LDOCE are senses of these words.

    Fourthly, LDOCE’s defining vocabulary has been extensively discussed. It has simplified its coding system, adopted a less mechanical use of the alphabet, and modified slightly its definition. The approaches of the ordering of the entries, the quality of the definitions, and the excessive analyticity of the treatment of polysemous words are welcome and quite different from the other dictionaries.