Language family

A language family is a group of languages with a common historical origin and phylogenetically related, that is, they seem to derive from an older language or protolanguage (mother tongue).

Families originate when a language, called the «protolanguage of the family», gives rise to different languages through a process of dialectal diversification. The languages of a family are usually mutually unintelligible, although in most cases they retain phonetic and grammatical similarities. When the similarities between the languages are clear, it is possible to reconstruct their common origin, and even the proto-language from which they derive, using the methods of historical linguistics.

Families of languages in the world

Kinship recognition

Systematic comparison of the world's languages using the methods of historical linguistics has made it possible to prove that most languages are not phylogenetically isolated languages, but rather form groups or families among themselves. Often within a family it is possible to reliably reconstruct the common origin or "mother protolanguage" of that family. The systematic study of many families has made it possible to reconstruct the various proto-languages or ancestral languages that, due to diversification, would have given rise to various families. Said reconstruction is based on the similarities observed between the languages of the same family and tries to determine which words or grammatical characteristics are the result of the inheritance of the common linguistic ancestor or protolanguage.

Protolanguage or common ancestor

The Indo-European linguistic family, which has various branches, is the most widely disseminated linguistic family worldwide. Until the discovery of America his presence was confined to Western Eurasia:

Hellenic (Greek) Italic (Romantic) Indoiranio Celtic German

Armenian Baltic Slave Albanés Non-indo-European languages

Targeted areas indicate multilingual areas.

The common ancestor of most families is not known or is only known directly on a few occasions, since the historical record of most languages is very short. However, it is possible to recover many of the features of the common ancestor of related languages by applying the comparative method —a reconstruction procedure developed in the s. XIX by the neogrammar school in which the linguist August Schleicher stands out. Language families can be subdivided into smaller units, usually called "branches" (the history of a language family is often represented as a phylogenetic tree).

The common ancestor of a family (or branch) is known as a "protolanguage". For example, the reconstructed proto-language of the well-known Indo-European family is called Proto-Indo-European (of which no written remains survive, since it was used before the invention of writing). Sometimes a protolanguage can be identified with a known language. Thus, the provincial dialects of Latin ("Vulgar Latin") were the origin of the modern Romance languages. That is, the "proto-Romance" language is more or less identical to Latin (although not identical to the learned Latin of the classical writers). The Old Norse dialects are the protolanguage of Norwegian, Swedish, Danish, and Icelandic. Hence, these proto-languages result from a "reconstruction" that linguists attempt from known data.

Lexicostatistics

Tentatively, when such good data on a language do not exist, various systems of linguistic comparison based especially on lists of words are preliminarily used, through statistical treatment. Among these techniques are glottochronology or more generally lexicostatistics. In this last line, work is being done on the ASJP systematic comparison project, which can suggest ways of systematic comparison between poorly documented languages, as well as clarify the internal structure of language families and provide an approximation of the global phylogenetic tree with more than 4,000 languages. languages.

Isolating and unclassified languages

Languages that cannot be safely classified into any family are called language isolates. There are several reasons why a language is classified as isolate:

1. Geographically nearby languages do not seem to have sufficient lexical coincidences to establish beyond doubt the existence of a common ancestor, between the language in question and other languages. This is more or less the situation for burushaski.
2. Languages related to the language in question disappeared before the language documentation of the language documents allowed to recognize kinship. This is the accepted situation for the Basque-Aquitaine, whose next relatives, if they had existed, could have disappeared due to the indo-European population of Europe.
3. Documentation on those and other languages in the same region is poor. This is the case of many Amazonian languages on which documentation does not allow to conclude whether or not they are related to other geographically nearby languages. When the evidence is very little some authors use in that case, preferably the term unclassified language.

A language isolate is actually a language family with only one member. Lyle Campbell puts the number of known language isolates, including some extinct, at 129.

Antiquity of the families

Glottochronological estimates have shown that most of the well-established language families have diversified in the last 50 centuries. A family is understood to be well established when there is consensus that these languages form a group and it has been possible to reasonably reconstruct the original protolanguage. While the less established families, or even controversial ones, and the macrofamilies seem to have longer diversification times. That reflects two facts:

1. The families of languages with less time depth show greater kinship, which makes it easier to apply the comparative method and, therefore, rebuild the protolengue.
2. Unestablished families could represent both families of great temporary depth, in which the passage of time had made kinship less recognizable, as groups of languages that are not really related, but show phenomena of linguistic contact and typical diffusion of linguistic area.

Distribution and number of speakers

Geographic distribution

The geographical distribution of families is a reflection of the historical evolution of their speakers. Thus, most of the large language families seem to have expanded thanks to agricultural or technological "revolutions" of some other kind. The Neolithic revolution caused the spread of Afroasiatic languages in Africa and the Middle East, the Sino-Tibetan languages in the Far East, and, according to Renfrew's theory, the spread of Indo-European languages in western Eurasia.

Improved navigation techniques allowed speakers of Austronesian languages to spread from the island of Taiwan throughout Oceania, even reaching Madagascar, off the African continent. European imperialism brought the Indo-European languages to the Americas and to many areas of Africa, Oceania and, to a lesser extent, Asia. Apparently, the use of iron and other technologies would have allowed the Niger-Congo languages to prevail in Africa, displacing the speakers of other now small families, such as the Nilo-Saharan languages or the Khoisan languages.

Currently, the two largest language families by number of speakers, the Indo-European languages and the Sino-Tibetan languages, together add up to a number of speakers equivalent to 75% of humanity. On the other hand, among indigenous languages In America, for example, many language families barely exceed a few thousand speakers.

Families by number of speakers

The following table summarizes the main families in the world according to the number of speakers:

The number of languages on which the above figures are based is 6,533, including 310 extinct languages (no current speakers), 71 pidgins and creoles totaling about 2 million of speakers. In addition, some 75 sign languages are known that are not counted because they are not oral languages attached to any family on the previous list.

Natural (oral) languages

They are grouped geographically without regard to relationships between families. In the list below, each item is a known language family. The geographical denominations of the titles are only to maintain order and facilitate reading. Geographic relationships are convenient for that purpose, but they do not represent any kind of attempt to create phylogenetic "superfamilies."

Families from Africa and the Near East

Languages of Africa.

Since ancient times, it was recognized that the Semitic languages of the Near East and the Arabian Peninsula were genetically related, something that was explained in semi-mythical terms, judging that the Semitic peoples were common descendants of Sem. Extensive classification of the languages of sub-Saharan Africa began in the 19th century on the basis of linguistic data and ethnographic data, dominated by a racist and supremacist vision of the human variety. The fully scientific study free of racist prejudices began properly in the XX century. The work of Joseph Greenberg, now generally accepted by most Africanists, groups the languages of the African continent into four large macrofamilies:

• Afro-Asian languages, previously classified as camitosemitic (North Africa and Middle East).
• Niger-Congo languages (Sub-Saharan Africa), sometimes called the Níger-cordophan family.
• Nile-Saharian languages (Africa Middle East)
• Jordanian languages o khoisan (Austral Africa)

Minor discrepancies persist about the internal groupings of these groups, and about the classification of some Cordovan languages. Significant comparative work has been done on most subfamilies within the Afroasiatic languages and the Niger-Congo languages. Several reasonably complete reconstructions of Proto-Afroasiatic, Protonilo-Saharan have been proposed, although the phonological system reconstructed by various authors may differ considerably, indicating that further clarification is still needed on the kinship and characteristics of these languages.

Families from Europe and North, West and South Asia

During the European Middle Ages, the obvious relationship between some languages was recognized (Romance languages, Celtic languages, Germanic languages, Slavic languages, etc.). However, all of these groups were not suspected of being ultimately related until much later. Towards the end of the 18th century British judge Williams Jones seriously advanced the idea that Latin, Greek, Germanic, Celtic, Sanskrit, and Persian were related. This was the first identification of the Indo-European family in a form close to how we know it today. During the XIX century, the development of the comparative method made it possible to identify other families.

Europe is currently the least diverse continent from a linguistic point of view. This is basically due to the Indo-European migrations during the Neolithic, and the subsequent formation of empires whose speakers used Indo-European languages, which ended most of the pre-Indo-European languages on the continent, with exceptions such as Basque, Finnish, Hungarian and various other languages. from the Caucasus. The Caucasus, on the other hand, is a much more diverse mountainous region than the rest of Europe. On the other hand, Central Eurasia is a much more diverse region than Europe (except the Caucasus), with several large language families. Generally accepted families today include:

Families from the Far East (Asia), Micronesia, East Melanesia and Polynesia (Oceania)

The Far East is home to the second family by number of speakers (Sino-Tibetan macrofamily) and the largest family in the world by number of languages (Austronesian family). A list of the region's generally accepted familis is as follows:

Australian Families

Pama-Ñung Division of Australian Languages.

Many authors assume that ultimately all Australian languages might be remotely related given Australia's relative isolation from other regions. Research on Australian languages recognizes between 228 and 262 languages. Most of them, around 160, belong to a well-identified phylogenetic family (pama-ñung family), the rest of the languages belong to more or less a dozen small families, half of which are in fact language isolates. There is no established evidence of the parentage of all Australian languages, since the methods of historical linguistics only work well for times of diversification of a few millennia. Whether in fact all Australian languages are ultimately emerentate is not known for sure, no matter how reasonable some consider such a conjecture. The basic classification of Australian languages is therefore:

• Pama-ñungana languages
• Non-Pama-ñungana languages (various non-related families)

Families from New Guinea and Western Melanesia

New Guinea is the most linguistically diverse region on the planet, since between 750 and 900 languages are spoken in its territory. There is no complete agreement on the number of different families that exist, and comparative work reveals that the number of valid phylogenetic units identified is enormous, so that the Papuan languages cannot be considered to constitute a single family. And although various authors have suggested that all Papuan languages are ultimately related, this has not been satisfactorily proven due to the enormous diversity of these languages.

If we look at the number of incontrovertibly identified or well-established families, there would be more than 60 linguistic families in New Guinea (see for example WALS classification); Of these, about a dozen would actually be isolated languages and the other 50 families themselves. It is possible that the number of well-established families could be reduced to a lower number, as Ross and Wurm's tentative classifications suggest, but at present those classifications are tentative. Greenberg considers the Papuan families to constitute a genetically valid subgroup within the Indo-Pacific languages, but such a classification is highly speculative and has been widely criticized.

Families of America

The classification of the languages of the Americas is probably one of the most controversial areas of the phylogenetic classification of languages. Proposals range from more than seven dozen phylogenetic units, to just three units for the entire continent.

North and Central American Families

North America.

The first sufficiently ambitious work to classify the languages of North America was carried out by John Wesley Powell, who recognized more than fifty phylogenetic groups. Edward Sapir made some controversial proposals to reduce the number of groups by trying to identify relationships between Powell's groups. Some of Sapir's proposals have gained acceptance, but most have been scrapped. A Powell-style "conservative" classification based on currently available evidence recognizes the following groups:

Comparative work on these languages could lead to reducing the number of groups, although such work of looking for distant relationships between well-established families is difficult, because the above groups appear only remotely related to one another, and possible relationships are in general very debatable and insecure.

Families from South America

The main families of South America (except Quechua, Aymara and Mapudungun).

In South America there are six large families by number of languages: the Tupi family (76 languages), the Arawak family (64 languages), the hypothetical macrofamily Macro-Gê (32 languages), the Caribbean family (32 languages), the Pano-Tacana macrofamily (33 languages), the Chibcha family (25 languages) and the Tucana family (22 languages). By number of speakers, there are important families or macrolanguages made up of a reduced number of languages that reached great diffusion: Quechua, Aymara, Guarani and Mapuche. A more or less complete list of the rest of small language families is the following:

Proposed superfamilies

Usually the language family name is reserved for a group of languages whose historical kinship and common origin is not controversial. In general, when linguistic work is advanced to the point of having reconstructed a few thousand terms of the original protolanguage, it is considered beyond any doubt that the languages form a family. Archaeological comparisons and linguistic and statistical speculations suggest that the reconstructed ancestors do not go back more than 5,000 or 7,000 years. For longer periods of differentiation, linguistic change is so profound that genetic relationships can hardly be found or demonstrated (in the same way that tests of genetic relationships between biological families of people can only be applied to very closely related people).

However, beyond the family level, there are some isolated similarities that allow us to speculate that many of the identified language families could be grouped into tentative macrofamilies or superfamilies, using less demanding methods than protolanguage reconstruction. Different linguists have proposed some groupings of language families into superfamilies. Many have been proposed based on the Morris Swadesh method. None are widely accepted as a phylogenetic group of languages, but some, such as the Amerindian, Papuan, Australian, Khoisan, and Paleo-Siberian families, are practical as geographical groupings of small families with common characteristics.

Creole languages, pidgins and lingua francas

• One Creole languageAlso called Creole or patois, it is a language usually born in a community composed of diverse origins that do not previously share any language, that they need to communicate and that they are forced to use a language that is not that of any of them. The result is a language that takes the lexicon (usually very deformed) of the tongue imposed and which, however, has a syntax that is more like that of other Creole languages than that of the "mother" language.
• A pidgin (pronounced [pvij'ambin]) is a language characterized by combining the syntactic, phonetic and morphological features of one language with the lexical units of another. The pidgin is not usually the mother tongue of any ethnic or social group; it is usually the language used by an immigrant in his new place of residence, or a lingua franca uses in an intense contact zone of linguistically differentiated populations, as a very active port; the pidgins They were also frequent in the colonies, mixing elements of the dominant nation's language with those of the natives and slaves introduced into it.
• French Language (o) lingua franca) is the language adopted for a common understanding between a group of several coexistenceers. Acceptance may be due by mutual agreement or political issues. In Europe during a part of antiquity, the Greek and Latin were adopted as French languages. In today's world, English acts as lingua franca especially in supranational organizations and international scientific publications.

Language isolates

A language isolate is a natural language for which no relationship to another living or dead language has been proven. Presumably, an isolate language is one that does not belong to any proper language family (ie, it is the only member of its family). Ainu, Euskera, Buruchasqui or Burushaski, and Sumerian are examples of languages frequently classified as isolates. There are also isolated languages in America, such as Mapuche (South America), Purépecha (Mexico) or Zuni (United States).

Isolated and almost isolated languages

• Pirahã. Some languages have become isolated languages in historical times after the disappearance of all other languages of the family. One example is the pirahã, the language of Brazil, the last surviving family mura de lenguas.
• Vasco. Another case of quasi-aislada is that of modern Basque. It is often said to be an isolated language, although it seems to be the modern descendant of the ancient aquitan.
• Sumerio. It is an extinct language of the Middle East, which was in fact the first language that was documented in writing.

Natural (gestural) languages

Major sign language families in the world. In green, the largest family based on the "old French sign language".

Clear historical connections can also be established between sign languages, in many cases the history is also documented. Thus, for example, modern French Sign Language, American Sign Language, and Mexican Sign Language have all evolved from variants of the same language: Old French Sign Language (used by the deaf community in Paris for the 18th century). By contrast, British Sign Language is not related to American Sign Language, even though the British and hearing Americans use variants of English. That is, a signer (user of sign language) of American Sign Language will have less difficulty in communicating with a signer of French Sign Language than with a signer of British Sign Language.

The following list includes some known families of sign languages:

• Based on LSF. Languages originated in the old language of French signs, from which the modern language of French signs (LSF), the language of American signs (ASL), the language of Mexican signs (LSM), the language of Venezuelan signs (LSV), the language of Austrian signs (ÖGSCS), the language of signs of the German Switzerland (DSGIRS), the language of Italian signs (LIS).
• Based on BSL (BANZSL family). Languages originated in the British sign language (BSL), which diversified during the centuryXIX giving rise to the Australian sign language (Auslan), the sign language of New Zealand (NZSL) and the sign language of Northern Ireland (NIRSL).
• Based on DGS. Languages originated in the German sign language (DGS), the Polish sign language (PJM) and the Israeli sign language (ISL).
• Based on KSL. Languages originated in the old sign language of Kent, used during the centuryXVII, which gave rise to the sign language used in Martha's Vineyard (Massachusetts) and which had an important influence on the American sign language (ASL).

Some sign languages are language isolates:

• Language of signs of Nicaragua, emerged in the second half of the 20th century.
• Sign Language of the North American Plains (Indian Sign Language, Great Plains Sign Language).
• Sign language adamorobe, Ghana.