| |
The diversity of human languages can sometimes appear quite overwhelming, but in fact languages tend to fall into distinct categories according to the underlying structural characteristics they share. Through the classification of vast numbers of languages, the limits within which languages can vary are revealed and interesting conclusions about the nature of language per se can be advanced. One approach to typology, adopted by Prague School linguists, classified languages on the basis of their sound systems. In tone languages, like Chinese or Thai, pitch contrasts help distinguish one word from another. Thus, in one dialect of Chinese, if the word-form ma is pronounced with a level tone, it means ‘mother’, whereas a falling tone would alter the meaning to ‘reprimand’. In non-tone languages, like English or Japanese, on the other hand, pitch contrasts of this kind are not distinctive and hence contribute nothing to word meaning.
More recently, typological research has focused on aspects of syntax, most notably in connection with the ordering of the basic sentence constituents, subject (S), verb (V) and object (O). The simple sentences of a given language display one of six possible configurations: SOV, SVO, OSV, OVS, VSO and VOS. By far the most frequently encountered combinations are SVO (e.g. English) and SOV (e.g. Japanese), followed by VSO (e.g. Welsh) and, lagging further behind, VOS, OVS and OSV. In fact, it is a matter of controversy whether any language actually displays the pattern OSV.
Evidently, ‘subject-first’ languages predominate, and it has been suggested that this unequal distribution reflects a themefirst principle, in which new or thematic information (normally expressed by the subject) is positioned most naturally at the beginning of a sentence. The so-called verb-object bonding principle also helps explain the distribution of word-order patterns with the observation that the verb of a simple sentence forms a more natural association with the object than it does with the subject. The prevalence of SOV and SVO languages bears witness to the rarity of the subject interceding between verb and object in the world\'s languages.
Word-order typology has revealed the existence of so-called implicational universals, whereby the presence of one feature in a language allows us to predict that certain other features will also be present. For example, it has been discovered that SVO languages prefer prepositions to post-positions (hence ‘in the White House’ rather than ‘the White House in’), and furthermore that there is a predilection for genitives after the noun rather than before (giving ‘the House of Commons’ in preference to ‘of Commons the House’). There is no obvious or necessary reason why SVO languages should prefer prepositions and genitives after the noun, although attempts are being made to find out if the connection is more than coincidental. This correlation is useful, though, for predicting the form of newly encountered SVO languages. Implicational universals have also been influential recently in comparative-historical linguistics, since they provide a valuable predictive device for the reconstruction of extinct languages. Given a bundle of structural features, evidence is only required for one of them in order to project that a language displays the entire range of relevant characteristics. However, caution needs to be exercised since implicational universals represent strong tendencies, rather than cast-iron certainties. Furthermore, many languages do not fit easily into a typological schema based on the simple ordering of S, V and O, since, in some cases, there may be no single, basic word order discernible. MS
Further reading B. Comrie, Language Universals and Linguistic Typology; , W. Croft, Typology and Universals. |
|
