Genealogical Classification of Languages

a classification based on the genetic principle—that is, the grouping of genetically related languages in a language family. The genealogical classification of languages became possible only after the emergence of the concept of linguistic relationship and the acceptance of the principle of the historical method in linguistic research in the 19th century. Classifications are made as the result of the study of languages, using the comparative and historical methods. Being historical and genetic, the genealogical classification of languages—in contrast to the multiplicity of typological and areal classifications—is in the form of a single scheme. Being linguistic, it does not coincide with anthropological classification, and, in particular, it does not presuppose that peoples who speak related languages are also members of the same race. The existence of systemic tendencies in language development is used to prove the genetic relationship of languages. The presence of systematic correlations—regular sound correspondences in the indigenous language materials (vocabulary and grammatical elements)—serves as the specific criterion for this purpose. However, the absence of systematic correlations between comparable languages does not prove the absence of a relationship between them, since the relationship might be too remote for any systematic correlations to be revealed in the linguistic data.

Although language families are continuously evolving, their formation generally predates the appearance of class society. The factor of linguistic differentiation plays a leading role in the formation of linguistic families, given the phenomenon of the parallel and convergent development of languages. Language families are usually made up of smaller groups that unite languages that have closer genetic relationships. Many of these groups originated in a much later period—for example, the Slavic, Germanic, Italic (from which the Romance languages developed), Celtic, and Indo-Iranian groups of the Indo-European language family. Modern genealogical linguistic classification does not provide any grounds to support the once-popular linguistic concept of the monogenesis of the languages of the world.

The best-known language families of Eurasia and Oceania include Indo-European, Uralic, Turkic, Mongolian, Manchu-Tungusic, Chukchi-Kamchadal, Sino-Tibetan, Mon-Khmer, Malayo-Polynesian, Dravidian, and Munda. Four large language families are recognized in Africa—Semito-Hamitic, or Afro-Asiatic (also spoken in the adjacent territory of Asia), Nilo-Saharan, Congo-Kordofanian, and Khoisan. Less adequately developed are the genealogical classifications of the aboriginal languages of America and Australia, where genealogical classification is not yet delimited from typological classifications. (In particular, E. Sapir’s classification of North American languages into six families has yet to be verified.) In view of the difficulty involved in differentiating distantly related languages and unrelated languages, purely hypothetical constructs are encountered in a number of instances: compare the Altaic family (embracing the Turkic, Mongolian, Manchu-Tungusic, and sometimes Korean, languages), the Caucasian family (including the Abkhazo-Adygeian, Kartvelian, and Nakhian-Dagestanian languages), and the Nostratic family (consisting of several large families of Eurasian languages). The so-called mixed languages also find a place within the well-known language families. (For example, almost all the Creole languages are placed in the Indo-European family.) There are isolated languages, genetically unrelated to any other language, which can be regarded as the sole representatives of separate families—for example, Basque in Europe; Ket, Burushaski, Nivkh, and Ainu in Asia; and Kutenai, Zuñi, and Keres in America.