Amerindian languages

from Wikipedia, the free encyclopedia
  • The Amerindic macro family is shown in light blue

    • Amerindisch , also Amerind or Amerindische Sprachen , is a macro family proposed by Joseph Greenberg , which he describes and justifies in detail in his work Language in the Americas from 1987. Amerindic, as defined by Greenberg, includes all indigenous American languages , except for the Eskimo-Aleut languages and the Na-Dené languages .

    The Amerindian hypothesis is not accepted by the majority of Americanists. On the contrary, it was actually fought against in sometimes aggressive posts.

    The following representations are largely based on the considerations of Joseph Greenberg and must therefore be viewed as controversial.

    The tripartite division of the American languages ​​according to Greenberg

    After the successful classification of the African languages , Greenberg examined the indigenous languages ​​of America, which, according to the majority of relevant research, are broken down into hundreds of genetic units and isolated languages . His result, published in 1987, is the division of all American languages into only three genetic groups:

    This tripartite division is supported by human genetic studies by Cavalli-Sforza (cf. Cavalli-Sforza 1996) and by archaeological research which shows that these three groups immigrated from Siberia to America via the Bering Strait at different times , first the carriers of the Amerindian languages , then the Na Dené peoples, and finally the Eskimos.

    Greenberg's method and its criticism

    While Eskimo-Aleut and, in principle, Na-Dené as well, had long been recognized as genetic units, the concept of the Amerindian languages ​​found no support among most Americanists. The presentation by Lyle Campbell , American Indian Languages (1997) with well over 200 separate genetic groups and many isolated languages is typical of the current state of American studies (see below the comparison of Greenberg's American Indian with the genetic units of Campbell).

    Lexical mass comparison

    The very massive Americanist criticism was not only directed at Greenberg's classification result, but above all his method of lexical mass comparison , in which the classification results from the comparison of words and morphemes from a very large group of languages ​​(in the case of Amerindian, almost all indigenous languages America). Here are word equations established and derived from these classifications; the establishment of sound laws and the reconstruction of proto-languages ​​is then a second step that confirms, refines or refutes the results of the previous classification hypothesis. (Greenberg usually left this second step to others.) Greenberg had already used the method of mass comparison in his now largely accepted, but also in many points questioned, classification of African languages. According to Greenberg's supporters, it is ultimately the method with which the researchers of the early 19th century recognized the genetic unit and the essentially correct structure of Indo-European or Finno-Ugric, long before phonetic laws were established or proto- languages ​​were reconstructed. Many critics of Greenberg - according to his defenders - overlooked the fact that the rigid historical-comparative linguistics also has preliminary stages that correspond exactly to the inductive-heuristic methods of Greenberg. But William Poser and Lyle Campbell criticize this view in an essay and explain why, in their opinion , there are decisive differences between Greenberg's and the traditional historical-comparative method, as developed primarily in the context of Indo-European studies in the 19th century.

    Incorrect data

    In addition, Greenberg was accused of numerous errors in his data material (see Lyle Campbell 1997), such as incorrect or non-existent words, use of distorted or overstretched meanings, words that were assigned to the wrong languages, incorrect decomposition of the word material into prefixes, word kernel and Suffixes. Although Greenberg defended his method in several essays and also showed that many allegations were incorrect (these essays are summarized in Greenberg 2005), the verifiable errors in Greenberg's argument have contributed significantly to the lack of acceptance of the Amerindian.

    Success of the smaller units

    Only medium-sized and smaller genetic units of his classification (most of which had already been proposed by Edward Sapir at the beginning of the 20th century) could be confirmed by further research, which is already a great step forward given today's fragmented linguistic landscape in America. The demonstration or final refutation of the relationship of larger language groups in America will certainly require a few more decades of intensive linguistic field work and comparative research, if this is not made prematurely impossible by the alarmingly rapid extinction of Indian languages ​​that can already be observed today. As a work program, Greenberg's “Amerind” concept is the largest reduction in the number of separate language families in America to date.

    Belonging of subfamilies to other language families

    In some cases parts of the Amerind have been assigned to other language families, for example Holst (2005) shows that the Eskimo-Aleut languages ​​are closely related to the Wakash languages .

    Some language families are also used by specialists in Na-Dene.

    Such relationships - if they can be confirmed - cross the line drawn by Greenberg between the three genetic language groups and weaken the Amerind hypothesis as a whole.

    General problems with the classification of American languages

    In the case of the American languages, the amicable large-scale breakdown is made more difficult by the fact that, on the one hand, in large parts of America (especially in South America), many languages ​​have so far only been insufficiently researched and documented and, due to the dramatic rapid extinction of many languages, this can no longer be made up for. On the other hand, the diversity and multitude of languages ​​(around 1000, of which around 400 are extinct) hardly allows a single Americanist to have the detailed knowledge necessary to classify many language families. To make matters worse, English-speaking and Spanish-Portuguese researchers are not well networked.

    Time depth of proto-language

    As with all proposed macro families , a fundamental problem with Amerindian is the great depth of time of its hypothetical proto-language . It is debatable whether the settlers of America possessed a single language about 12,000 years ago. On the contrary, there are good reasons for a diversification of the later Amerindian languages ​​in Northeast Asia. A hypothetical proto-Amerind, which Greenberg and his followers never attempted to reconstruct, would then have to be at least 15,000, if not 20,000 years old. Most researchers in comparative linguistics - not just Americanists - believe that phonetic , grammatical, and lexical similarities can no longer be demonstrated after such a long time.

    "Lumper" and "Splitter"

    In American studies in particular, there is a very wide gap between the lumpers , who want to group as many languages ​​as possible in as few large families as possible, and the splinters, who are extremely critical of genetic units that are not beyond doubt , resulting in a multitude of small and very small genetic units Groups and isolated languages ​​leads. Accordingly, some linguists see only one (Amerind) or at most a dozen large genetic units, while other linguists accept up to 200 language families and isolated individual languages ​​on the American double continent. In general, it can be said that two types of errors can be made in the classification of languages: on the one hand, incorrect merging of languages ​​that are not genetically related; on the other hand, the failure to recognize genetic relationships. The second type of error is often considered to be less serious; the - mostly small, clear and easily verifiable - units of the “splitters” are at least very likely correct and form a minimum consensus, while units going beyond this remain possible, but are still awaiting proof. Convincing "Splitters" of a proposal is not impossible, but they apply stricter criteria to a proof, while "Lumpers" in the opinion of "Splitters" misinterpret areas in common as in language groups as an indication of genetic relationships, while similarities also misinterpret commonalities could well have come about through language contact . Both types of errors can play a role in the classification of American languages.

    In principle, it must also be stated that historically real relationships that go back far into the past are in many cases simply no longer clearly identifiable and can no longer be clearly delimited from commonalities that have come about through contact, which is why it is entirely possible that "Lumpers" in In certain cases, historically real connections are advocated, the unwillingness of the “splitters” to accept these if the evidence is (still) insufficient or the arguments are (yet) not good enough, but are nevertheless methodologically correct or preferable. Truth and scientific evidence are different things. In addition, it is quite possible that "lumpers" underestimate the scope of the problem of assessing suggestions about the relationship between hundreds of different languages ​​in which one is not a specialist. For example, even if the Amerind languages ​​did indeed form a language family, it would be extremely difficult to find someone who had the competence to demonstrate such a comprehensive and far-reaching relationship, or to assess an existing evidence completely and in every detail, as it was a designated one You should be familiar with at least a large number of individual languages ​​and language groups in the area of ​​the American Indian languages. The amount of data would be overwhelming. In a book chapter devoted specifically to this topic, Lyle Campbell, an avowed “splinter”, assigns impressionistically estimated probabilities (in the form of percentages) to various suggestions in the area of ​​the American Indian languages, each combined with an estimate of the reliability of this information (from the self-assessment of his competence out), to emphasize the fact that it is not about “absolute truths”, politics or psychology (mere conservatism or supposed “unwillingness” on the side of the “splinters” to recognize the “truth” for irrelevant motives) but about the quality of methods and arguments.

    Amerindian and its structure

    In the following the Amerindian and its structure according to Greenberg 1987 is shown. Greenberg divided the Amerindic languages ​​into six primary branches, namely North Amerind (North American languages), Central Amerind (Central American languages) and the South American units Chibchan-Paezan , Andisch , Equatorial-Tucano and Ge-Pano-Carib .

    In the following illustration, the genetic units according to Campbell 1997 are given in brackets . This results in a concordance between the subunits of Greenberg and the genetic units (language families and isolated languages) of Campbell, who in his work comprehensively presents the current majority opinion on the classification of American languages. The Campbell's units are also based on the web links given below for the classification of North, Central and South American languages.

    Structure of Amerindian according to Greenberg and concordance to Campbell

    The genetic units according to Campbell 1997 which correspond to the subgroups Greenbergs 1987 are given in brackets.

    • North Amerind
      • Almosan-Keres
        • Almosan (Algonkin-Ritwan, Kutenai, Chimakum, Salish, Wakashan)
        • Keres group (Caddo-Pawnee, Iroquois, Keres, Sioux, Yuchi)
      • Penutally
        • Californian (Maidu, Miwok-Costano, Wintu, Yokuts)
        • Chinook (Chinook)
        • Mexican (Huave; Maya, Mixe-Zoque, Totonac-Tepehua)
        • Oregon (Alsea, Coos, Takelma-Kalapuya, Siuslaw)
        • Plateau (Cayuse, Klamath-Modoc, Molala, Sahaptin - Nez Perce)
        • Tsimshian (Tsimshian)
        • Yuki-Gulf (Atakapa, Chitimacha, Muskogee, Natchez, Tunica, Yuki-Wappo)
        • Zuni (Zuni)
      • Hoka
        • Nuclear Hoka
          • North Hoka (Karuk, Shasta, Chimariko; Yana; Pomo)
          • Washo (Washo)
          • Esselen-Yuma (Esselen, Yuma-Cochimi)
          • Salinan-Seri (Salinan, Seri, Chumash)
          • Waicuri (Guaicura)
          • Maratino (maratino)
          • Tequistlatec (Tequistlatec)
        • Coahuiltecan (Tonkawa, Coahuilteco, Comecrudo, Karankawa)
        • Subtiaba-Tlapanec ( see Oto-Mangue)
        • Jicaque (Jicaque)
        • Yurumangui (Yurumangui)
    • Central Amerind
      • Kiowa-Tano (Kiowa-Tano)
      • Oto-Mangue ( Otomangue )
      • Uto-Aztec (Uto-Aztec)
    • Chibcha-Paez
      • Macro chibcha
        • Cuitlatec (Cuitlatec)
        • Lenca (Lenca)
        • Nuclear Chibcha (Chibcha, Misumalpa)
        • Tarascan (Tarasco)
        • Xinca (Xinca)
        • Yanoama (Yanomam)
      • Macro paezan
        • Nuclear Paezan (Andaqui, Barbacoa, Choco, Paez)
        • Other Paezan (Huarpe, Atacama, Betoi, Chimu-Mochica, Itonama, Jirajara, Mura, Matanawi, Timucua, Guarao)
    • Andean
      • Aymara ( Aymara )
      • Itucale-Sabela (Urarina, Sabela)
      • Kahuapana-Zaparo (Cahuapana, Zaparo)
      • North Andean (Sechura-Tallan, Cholon-Hibito, Culle , Leco)
      • Quechua ( Quechua )
      • South Andean ( Kawesqar  [Alakaluf], Mapudungun  [ Araucanian ], Chon, Yamana)
    • Equatorial Tucano
      • Macro Tucano (Tucano; Aushiri, Kanichana, Kapixana, Katukina, Gamela, Aikana-Tubarao, Irantxe, Maku, Sape, Arutani, Koaya, Movima, Munichi, Nambiquara, Natu, Pankararu, Puinave, Makú, Xukuru, Ticuna, Yuri, Huamoe)
      • Equatorial
        • Macro Arawak (Arawak, Chapacura, Guamo, Puquina; Guajibo, Katembri, Otomaco, Tinigua)
        • Jibaro-Kandoshi (Cofan, Esmeralda, Jivaro, Kandoshi, Yaruro)
        • Kariri-Tupi (Tupi, Kariri)
        • Other equatorial (Cayubaba, Camsa, Saliva-Piaroa, Taruma, Timote, Trumai, Tuxa, Yuracare, Zamuco)
    • Ge-Pano-Carib
      • Macro-Carib (Carib; Andoque, Bora-Witoto, Kukura, Peba-Yagua)
      • Macro-Pano (Pano-Tacana; Charrua, Lengua-Mascoy, Lule-Vilela, Mataco, Guaikuru, Moseten)
      • Macro-Ge (Ge, Bororo, Botocudo, Karaja, Rikbatsa, Guato, Kamaka, Mashakali, Ofaye, Puri; Chiquitano, Oti, Jabuti)

    literature

    Web links

    Individual evidence

    1. ^ William J. Poser, Lyle Campbell: Indo-European Practice and Historical Methodology. In: Proceedings of the Annual Meeting of the Berkeley Linguistics Society. Vol. 18, 1992, ISSN  0363-2946 , pp. 214-236, ( online (PDF; 153 kB) . Accessed November 9, 2016).
    2. ^ Lyle Campbell: American Indian Languages. The Historical Linguistics of Native America (= Oxford Studies in Anthropological Linguistics. 4). Oxford University Press, New York NY et al. 2000, ISBN 0-19-514050-8 , pp. 260 ff., ( Limited preview in Google book search).