Sprachbund

History

In a 1904 paper, Jan Baudouin de Courtenay emphasised the need to distinguish between language similarities arising from a genetic relationship and those arising from convergence due to language contact. The term Sprachbund, a calque of the Russian term языковой союз (yazykovoy soyuz; "language union"), was introduced by Nikolai Trubetzkoy in an article in 1923. In a paper presented to the 1st International Congress of Linguists in 1928, Trubetzkoy defined a sprachbund as a group of languages with similarities in syntax, morphological structure, cultural vocabulary and sound systems, but without systematic sound correspondences, shared basic morphology or shared basic vocabulary. Later workers, starting with Trubetzkoy's colleague Roman Jakobson, have relaxed the requirement of similarities in all four of the areas stipulated by Trubetzkoy.

In contrast, a sprachraum (from German, "language area"), also known as a dialect continuum, describes a group of genetically related dialects spoken across a geographical area, differing in their genetic relationship only slightly between areas that are geographically close, and gradually decreasing in mutual intelligibility as distances increase.

The Balkans

The idea of areal convergence is commonly attributed to Jernej Kopitar's description in 1830 of Albanian, Bulgarian and Romanian as giving the impression of "nur eine Sprachform ... mit dreierlei Sprachmaterie", which has been rendered by Victor Friedman as "one grammar with the three lexicons". The Balkan sprachbund comprises Albanian, Romanian, the South Slavic languages of the southern Balkans (Bulgarian, Macedonian and to a lesser degree Serbo-Croatian), Greek, Balkan Turkish, and Romani. All but one of these are Indo-European languages but from very divergent branches, and Turkish is a Turkic language. Yet they have exhibited several signs of grammatical convergence, such as avoidance of the infinitive, future tense formation, and others. The same features are not found in other languages that are otherwise closely related, such as the other Romance languages in relation to Romanian, and the other Slavic languages such as Polish in relation to Bulgaro-Macedonian.

Indian subcontinent

In a classic 1956 paper titled "India as a Linguistic Area", Murray Emeneau laid the groundwork for the general acceptance of the concept of a sprachbund. In the paper, Emeneau observed that the subcontinent's Dravidian and Indo-Aryan languages shared a number of features that were not inherited from a common source, but were areal features, the result of diffusion during sustained contact.

Emeneau specified the tools to establish that language and culture had fused for centuries on the Indian soil to produce an integrated mosaic of structural convergence of four distinct language families: Indo-Aryan, Dravidian, Munda and Tibeto-Burman. This concept provided scholarly substance for explaining the underlying Indian-ness of apparently divergent cultural and linguistic patterns. With his further contributions, this area has now become a major field of research in language contact and convergence.

Southeast Asia

The Mainland Southeast Asia linguistic area is one of the most dramatic of linguistic areas in terms of the surface similarity of the languages involved, to the extent that early linguists tended to group them all into a single family, although the modern consensus places them into numerous unrelated families. The area stretches from Thailand to China and is home to speakers of languages of the Sino-Tibetan, Hmong–Mien (or Miao–Yao), Tai-Kadai, Austronesian (represented by Chamic) and Mon–Khmer families.

Neighbouring languages across these families, though presumed unrelated, often have similar features, which are believed to have spread by diffusion. A well-known example is the similar tone systems in Sinitic languages (Sino-Tibetan), Hmong–Mien, Tai languages (Kadai) and Vietnamese (Mon–Khmer). Most of these languages passed through an earlier stage with three tones on most syllables (but no tonal distinctions on checked syllables ending in a stop consonant), which was followed by a tone split where the distinction between voiced and voiceless consonants disappeared but in compensation the number of tones doubled. These parallels led to confusion over the classification of these languages, until Haudricourt showed in 1954 that tone was not an invariant feature, by demonstrating that Vietnamese tones corresponded to certain final consonants in other languages of the Mon–Khmer family, and proposed that tone in the other languages had a similar origin. Similarly, the unrelated Khmer (Mon–Khmer), Cham (Austronesian) and Lao (Kadai) languages have almost identical vowel systems. Many languages in the region are of the isolating (or analytic) type, with mostly monosyllabic morphemes and little use of inflection or affixes, though a number of Mon–Khmer languages have derivational morphology. Shared syntactic features include classifiers, object–verb order and topic–comment structure, though in each case there are exceptions in branches of one or more families.

Northeast Asia

Some linguists, such as Matthias Castrén, G. J. Ramstedt, Nicholas Poppe and Pentti Aalto supported the idea that the Mongolic, Turkic, and Tungusic families of Asia (and some small parts of Europe) are genetically related, in a controversial group they call Altaic. Koreanic and Japonic languages, which are also hypothetically related according to some scholars like William George Aston, Shōsaburō Kanazawa, Samuel Martin and Sergei Starostin, are sometimes included as part of the purported Altaic family. This latter hypothesis was supported by people including Roy Andrew Miller, John C. Street and Karl Heinrich Menges.

Gerard Clauson, Gerhard Doerfer, Juha Janhunen, Stefan Georg and others dispute or reject this. A common alternative explanation for similarities between said-Altaic languages such as vowel harmony and agglutination is that they are due to areal diffusion.

Others