History and Discussion of Native American Languages

From: http://www.indians.org/welker/americas.htm

At the time of first European contact, probably close to 1,000 American Indian languages were spoken in North, Central, and South America. Although the number of languages in daily use has steadily declined because of persecution and pressures on the Indians to adopt English, Spanish, and other originally European languages, well over 700 different American Indian--or, as they are sometimes called, Amerindian or Native American--languages are spoken today.

In the United States many of the most famous linguists of the early 20th century--among them Franz BOAS, Leonard BLOOMFIELD, and Edward SAPIR--transcribed and analyzed North American Indian languages. Many descriptions of Indian languages are important in the literature of the linguistic school known as American structuralism.

Today interest in Native American Indian languages is increasing, and Americanists, as those who study the languages are called, hold regular meetings to report on their findings. Current research on the native languages of the Americas is published in several periodicals, notably the International Journal of American Linguistics.

The great diversity of Indian languages, however, has thus far prevented proof of common origin, and most Americanists favor more conservative classifications of the languages into a number of distinct groups.

Only a few Native American Indian languages have a written history; therefore, comparative study must be based upon quite recent sources. Following the traditional principles of historical linguistics, words from Indian languages believed to be related are subjected to minute comparison, in a search for regular correspondences of sound and meaning.

Regularity is the key: thus, while Luiseno paa-la, Papago wa-, and Aztec a-tl, all meaning "water," do not immediately appear similar, the words are seen to be cognate (derived from the same word in the ancestor language) when other sets such as Luiseno pe-t, Papago woog, and Aztec o-tli, all meaning "road," are considered, since Luiseno initial p and Papago initial w regularly correspond to the lack of any initial consonant sound in Aztec.

When such correspondences are discovered, the languages being compared are judged to have a historical connection, either genetic--because of descent from a common ancestor--or through language contact and the consequent "borrowing" of words. As genetic relationships are discovered, languages are grouped into families, which then are often compared themselves. Related families can be classified in turn into larger groups called phyla (singular, phylum) or stocks, or into even broader groupings known as macrophyla or superstocks.

On the basis of the Luiseno, Papago, and Aztec words cited above, linguists have proposed the reconstruction of initial p sound in the words for "water" and "road" in the Proto-Uto-Aztecan ancestor of the three languages in question. The sounds systems and vocabulary of the ancestors of a number of different American Indian language families have been partially reconstructed through similarly detailed analysis by linguists. Comparison of these reconstructed protolanguages leads to more informed conjecture about earlier connections between the ancestor languages and the peoples who spoke them.

Language Names

Names for Native American Indian languages can be confusing. Some names are chosen politically rather than linguistically: for instance, Creek and Seminole are mutually intelligible Muskogean languages but are traditionally treated as separate because the tribes who use them are different. Many Native American Indian tribes use the word for "people" as the meaning of their nation.

Often Indian groups come to be known by a foreign term, such as the English names Dogrib and Yellowknife for Athabascan tribes in the Northwest or the naming of most Coastal California languages for the nearest Spanish mission (Luiseno was the Uto-Aztecan language spoken around Mission San Luis Rey, for example, and the Chumash language Obispeno was named for Mission San Luis Obispo). Some other designations, occasionally derogatory, originated with other Indians--the name Comanche, for example, is from Southern Paiute kimantsi, "stranger." Both languages are Uto-Aztecan.

In some cases the same name has been used for two or more distinct languages. For instance, there are two languages in Central America called "Chontal," one Hokan and one Mayan.

The names of linguistic families and stocks are usually coined by linguists, often by adding -an to the name of a representative language. The Yuman family, for example, is named for the language Yuma.

Languages of Canada and the United States

Perhaps 300 languages were spoken in Canada and the United States when the first Europeans arrived, and about 200 are still spoken by some 300,000 people. The American explorer and ethnologist John Wesley POWELL presented the first comprehensive classification of the languages north of Mexico in 1891, dividing them into 58 families. Various scholars have subsequently proposed consolidation of Powell's families into a smaller number of phyla, with the most influential of these classifications credited to Edward Sapir. C.F. and F.M. Voegelin introduced the most widely accepted modern classification of American Indian languages, grouping most of the languages of the United States and Canada into seven macrophyla, with a few families and language isolates left unclassified (Table 1).

One phylum, American Arctic-Paleosiberian, includes both Eskimo-Aleut, spoken from Alaska to Greenland, and the Chukchi-Kamchatkan family of Siberia. This phylum is the only American language family to have an accepted connection with a non-American language group.

Meso-American Languages

Recent estimates place the number of Meso American Indian languages at about 70, with at least 5 million speakers. Of course, language boundaries and political boundaries do not coincide. The Hokan and Aztec-Tanoan phyla of North America also include a number of Central or Meso-American languages, and some South American groups have outlying representatives in Central America. Many of the groupings in Table 2 are still highly controversial.

South American Languages

Linguistic diversity is greatest in South America, where many languages spoken in remote jungle and mountain regions remain unrecorded and unclassified. There are probably over 500 different languages still spoken, with perhaps 14 million speakers. The various languages of the Quechua group alone have 5 million speakers.

Broader classifications of the more than 80 South American language families into a smaller number of macrophyla have been proposed by Joseph Greenberg, Morris Swadesh, Cestmir Loukotka, and others. Because these South American stocks have not as yet been fully documented with lists of cognate sets, they are not accepted by all specialists.

Recent Controversy

Current scholarly approaches to Native American Indian language classification are polarized. Most Americanists accept only certain parts of the Voegelin classification, while rejecting others, with the Macro-Penutian and Hokan phyla of North America receiving most challenges. Joseph Greenberg recently proposed a new classification, with just three groups of languages: Eskimo-Aleut, Na-Dene, and a third stock, Amerind, which includes all the other languages of North, Central, and South America. Although some mainstream Americanists find this proposal intriguing, they have criticized Greenberg's research for its methodology and data, and the theory is not widely accepted.

Grammatical structure

The grammatical structure--phonology, or sound system; morphology, or word structure; and syntax, or sentence structure--of Native American Indian languages varies considerably, but none of the languages can be called primitive.

Phonology Though some Indian languages have a simple phonological structure (the Arawakan language Campa, for instance, has only 17 contrastive speech sounds, or phonemes), the phonology of others is very complex. Certain sounds, many of which are articulated toward the back of the vocal tract, have been cited as characteristic of the American Indian languages, but none of these occur in all the languages. The glottal stop, made by briefly closing the vocal cords, as in the middle of the English word uh-oh, is a common sound. Many languages have glottalized consonants, made with a glottal stop produced simultaneously with another consonant sound. For instance, Navajo ts'in, meaning "bone," has a glottalized ts sound (represented by ts'), while tsin, "tree" has a plain ts. Another common sound is a back k sound, normally written q, articulated not at the velum, as is English k, but rather in the postvelar or uvular region. Many languages contrast k and q in words like Cahuilla (Uto-Aztecan) neki, "my house," versus neqi, "by myself."

Vowel systems also vary considerably. Quite a few American Indian languages have nasalized vowels. Nasalization is represented by a tilde symbol in Chickasaw, for example. The use of pitch accent or tonal systems (as in Chinese) to differentiate words is more common in the Americas than the use of contrastive stress like that found, for example, in English import, pronounced im-port' as a verb and im'-port as a noun.

Morphology and Syntax

The most commonly cited trait of American Indian languages is polysynthesis--the expression of complicated ideas within a single word containing many separate meaningful elements, or morphemes. The use of verbs with attached subject and object indicators (most often prefixes) is common; in many languages adverbial and other elements may also be attached to the verb, forming complex single-word sentences, like the Lakota (Siouan) wica-yuzaza-ma-ya-khiya-pi-kte, "you all will make me wash them," which includes the component morphemes them + wash + me + you + make + plural + future.

While most languages have accusative case systems like that of English (opposing grammatical categories of subject and object), active systems in which the same morpheme is used to indicate the object of a transitive verb and the subject of a stative verb are not uncommon. For example, the prefix ma-, "me" in the Lakota example just presented means "I" in a sentence like ma-s'amna, "I stink. "

Many languages use unmarked verbs for the third person. Thus Chickasaw hita can mean either "to dance" or "he dances." Possessive and locational indicators are often attached to nouns, as in Yup'ik Eskimo anya-a-ni (boat + his + in), which means "in his boat." Gender distinctions like those of the Indo-European languages are found in only a few languages, such as Garifuna (Arawakan), in which halau, "chair," is masculine, but muna, "house," feminine. More languages make a grammatically comparable distinction between animate, or living, and inanimate nouns. Alienable possession or ownership is often indicated differently from inalienable possession of items such as kinship terms and body parts. Reduplication--the doubling of all or part of a word, usually to indicate plurality or intensity--is common, as in Barbareno Chumash ma, "jackrabbit," ma ma, "jackrabbits."

The arrangement of words into sentences also varies from language to language. While the most common basic word order is Subject-Object-Verb, Subject-Verb-Object is used in many languages, and the rarer word orders Verb-Subject-Object, Verb-Object-Subject, and Object-Verb-Subject are also found.

Many Native American Indian languages make use of special syntactic patterns to distinguish among third-person participants in a sentence. Obviation (in the Algonquian languages) and the use of the so-called fourth person (in Athabascan) allow one participant to be coded as more important or interesting than another. Switch-reference is the name given to an unusual grammatical device that allows a speaker to specify whether the subject of one clause is the same as or different from that of another clause. The English sentence "he knows he's fat" is ambiguous. If the first "he" is known to refer to Tom, for instance, the sentence has one meaning. If the second "he" also refers to Tom ("Tom is fat and he knows it") and another if the second "he" refers to, say, Bill ("Bill is fat and Tom knows it"). Although the Mojave (Yuman) sentences isay-k suupaw-pc (fat + same know + perfective) and isay-m suupaw-pc (fat + different know + perfective) both translate as "he knows he's fat," they are not ambiguous: the first implies that the knower is fat, while the second means that someone else is.

The Whorfian Hypothesis

Because of different cultural needs, American Indian vocabulary structure varies greatly, and some of the semantic concepts and sentence patterns often seem unfamiliar to those who have not grown up speaking the languages. The American linguist Benjamin Lee WHORF argued that the differences in semantic and syntactic organization of languages as diverse as English and Hopi were correlated with differences in thought processes. The so-called Whorfian (sometimes Whorf-Sapir) hypothesis that grammatical structure reflects cognitive structure is not widely accepted among linguists but has been influential in other social sciences.

Language contact

Unrelated languages whose speakers are in daily contact often come to share various grammatical traits, which can then be called areal features of the region. In the Pacific Northwest, for instance, there are several unrelated genetic groups with strikingly similar, unusually complex consonant systems. Many languages of the Tupian family of South America have nasalization as an attribute, not just of vowels or consonants, but of whole syllables, and this feature has been borrowed by some unrelated neighboring languages.

Loan words can reveal the prior history of a linguistic group. Alaskan languages and some as far south as California have Russian loans, for instance, dating from the time of extensive trade with Russia, and borrowings from Spanish are common throughout California, the Southwest, and, of course, Latin America. Borrowed words are often changed to fit the structure of the borrowing language--Spanish caballo ("horse") was borrowed into Tubatulabal (Uto-Aztecan) as kawaayu, for example, because all Tubatulabal words have final stress and the language has no bilabial v or b sound. Indian words have also been borrowed into English and and other European languages. The words moccasin, squash, squaw, and toboggan, like the majority of Indian loans into English, are from Algonquian languages; chocolate, from Aztec, tobacco, from Taino, and condor, from Quechua, are examples of words that were borrowed first into Spanish and then into English. The names of thousands of places throughout the Americas are of Indian origin.