Language typology

from Wikipedia, the free encyclopedia

The language typology is a research field of linguistics that deals with the comparison and classification of languages based on structural properties. The typological classification differs from the genetic classification, which classifies languages into language families according to their primary etymological origins, i.e. according to their original languages , and from the geographical classification, which groups languages into linguistic groups based on similarities resulting from persistent language contact . A typological class is called a language type. There are different approaches to language typology.

Morphological language typology

Forerunner and founder of the language typological approach

In 1767 Nicolas Beauzée's main work was published, the two-volume Grammaire générale ou exposition raisonnée des éléments nécessaires du langage, pour servir de fondement à l'étude de toutes les langues , which was based in part on Beauzée's contributions to the Encyclopédie ou Dictionnaire raisonné des arts, des arts et des métiers based. It was a universal grammar in the spirit of Antoine Arnauld and Claude Lancelot . Beauzée can be seen as the founder of the language typological approach. When justifying his theory, Beauzée tried, as far as he was able, to have a sound empirical basis.

Classic morphological typology

One of the earliest typologies is that of August Wilhelm Schlegel and Wilhelm von Humboldt . They divided the languages into synthetic and analytical languages based on morphological criteria.

The modern parameter-based morphological typology

Although the classical classification is still widely used today, some weak points of the system have been criticized in the recent past: The biggest drawback is that the classical morphological typology postulates a number of rigid language types that represent prototypes at best and only very much in their pure form are rarely found. For example, a language can have mostly agglutinating affixes, but also some fusional elements. For this reason, an alternative classification system has been proposed in the last few decades that does not work with prefabricated types but with two parameters on which languages ​​move with fluid transitions.

  • The first parameter is the morpheme-per-word rate, i.e. the criterion is the number of morphemes per word . Extreme cases that mark the endpoints of the scale (but are not the only possibilities) would be, on the one hand, completely isolating languages (typically exactly one morpheme per word), on the other hand, polysynthetic languages ​​(typically potentially very many morphemes per word) .
  • The second parameter is the degree of fusion, i.e. the extent to which the grammatical morphemes can be segmented . Extreme cases here would be languages ​​that merge to a high degree (with low segmentability and high morphophonological variance of the morphemes) and agglutination (segmentability and invariance of the morphemes).

By combining the two parameters, a great many languages ​​in the world can be satisfactorily characterized.

Statements such as “ Turkish is an agglutinating language”, in which only an indication of the language type is given, relate to the classical morphological typology; if two statements are given, the more modern variant is usually implied as the underlying. The statement " Nahuatl is an agglutinating, polysynthetic language" (cf. the corresponding article) should be read in such a way that it is a language with many morphemes per word (polysynthetic), whereby these are mostly segmentable (agglutinating).

Language typology using the means of statistics

Aware that languages ​​have properties such as “isolating”, “agglutinating” or “inflecting” to varying degrees, Greenberg developed a total of 10 measures for morphological and syntactic properties that allow the degree to which a language has a certain property has to measure accurately. The best-known measure is the so-called "synthesis index", in which the number of morphemes in a text is related to the number of words in which these morphemes occur. The result is a characteristic for a considered language, which consists of 10 measured values ​​and enables exact comparisons with any other languages. This concept was further developed by Altmann and Lehfeldt, in which they discussed the theoretical principles and showed that there are correlations between the indices (measures). They also showed how one can come to a typological classification of languages on this basis with the help of numerical taxonomy and with what result. A continuation of these approaches can be found in Silnitzki, u. a. tests another language measure and includes other languages ​​in its investigations.

Word order typology

A more recent approach is the universals research of Joseph Greenberg , which searches for general structural laws in the languages ​​of the world. One example of this is word order typology, which is based on syntactic criteria. It classifies languages ​​according to the order of subject , object and verb in an unmarked sentence. The respective language types of a class are often simply called the "type of language":

  • SVO subject-verb-object , e.g. B. English, Chinese, French, Spanish, Russian
  • SOV subject-object verb , e.g. B. Turkish, Japanese, Persian, Latin (there the word order is basically free; however, there is a strong tendency to SOV or OSV)
  • VSO verb-subject-object , e.g. B. Gaelic, Welsh, Aramaic, Tagalog, Standard Arabic

In almost all languages, however, the subject precedes the object, so that the following three types only appear very sporadically:

In German and Dutch , this classification is made more difficult by the fact that the "analytically" compound verb (but according to fixed rules) is distributed in several parts over the sentence and subject , all direct and indirect - including genitive - objects and all place, time and or modal information etc. can also be placed in between and in front of the rules in accordance with the rules, for example: “I saw a fox in the forest” or “I saw a fox in the forest”, or “We should have got rid of this fox long ago”. These languages are therefore often referred to as V2 classified -Sprachen because the conjugate part of the verb regardless of the position of subject, object and the other parts of the sentence in any case in the second place, while the remaining parts of the verb is always at the end of the main clause are . More often, however, the order used in the subordinate clause is used as the basic word order (in the subordinate clause the conjugated part of the verb is always at the end of the sentence), in this example "that I saw a fox in the forest", so that German and Dutch are classified as SOV accordingly become.

Some languages, especially strongly inflected ones , cause particular problems when classifying them in this system, since they basically allow any order of verb and object. Examples are Latin and the Polish language . However, this is more due to the syntactic analysis approach, which does not help here. On the other hand, a pragmatic approach seems to help, such as the one provided by Simon C. Dik's Functional Grammar, which roughly distinguishes between topic (the well-known actant about which something is said) and focus (the most important element of the utterance). This approach also helps in even more inflected ancient Greek, as H. Dik has shown in two books on Herodotus and the language of tragedy from 1995 and 2007. However, such pragmatic analysis approaches put the largely syntactic word order typologies into perspective.

Theo Vennemann and Winfred P. Lehmann have reduced the six basic types to two by removing the subject ( VO and OV ). The far-reaching consequences, especially of a linguistic historical nature, that they derive from it are, however, controversial in the professional world

Relational typology (morphosyntactic alignment)

The relational typology classifies languages ​​according to their morphosyntactic expression of the fundamental grammatical relations (see accusative , active and ergative language ).

Phonological language typology

Depending on the research interest, typological considerations can be based on criteria from all sub-disciplines of linguistics . From phonological perspective can languages for example in accent scoring, moren counting and syllables counting divide.

See also


Web links

Wiktionary: Language typology  - explanations of meanings, word origins, synonyms, translations

Individual evidence

  1. ^ Georg Bossong: The beginnings of typological thinking in European rationalism. P. 7. (PDF; 187.93 kB).
  2. Script Uni Heidelberg SS07 ( Memento from June 10, 2007 in the Internet Archive )
  3. Archived copy ( Memento of March 2, 2008 in the Internet Archive )
  4. see Bernard Comrie: Language Universals and Linguistic Typology. Chicago University Press, Chicago 1989, especially Chapter 2.3.
  5. ^ Joseph H. Greenberg: A quantitative approach to the morphological typology of languages . In: International Journal of American Linguistics . Volume 26, 1960, pp. 178-194.
  6. ^ Gabriel Altmann, Werner Lehfeldt: Allgemeine Sprachtypologie . Fink, Munich 1973, ISBN 3-7705-0891-2 .
  7. George Silnitsky: typological indices and Language classes. A quantitative study. In: Gabriel Altmann (Ed.): Glottometrika . Volume 14, Wissenschaftlicher Verlag Trier, Trier 1993, ISBN 3-88476-081-5 , pp. 139-160.
  8. ^ Contra: Bernard Comrie: Language Universals and Linguistic Typology. Chicago University Press, Chicago 1989 (English); careful per: Larry Trask: Historical Linguistics. Hodder Arnold, London 1996, 8.3, ISBN 0-340-60758-0 , 8.8 (English).