Classification schemes for indigenous languages of the Americas

An early attempt at North American language classification was attempted by A. A. Albert Gallatin published in 1826, 1836, and 1848. Gallatin's classifications are missing several languages which are later recorded in the classifications by Daniel G. Brinton and John Wesley Powell. (Gallatin supported the assimilation of indigenous peoples to Euro-American culture.)

(Current terminology is indicated parenthetically in italics.)

Families

1. Algonkin-Lenape  (=Algonquian)
2. Athapascas  (=Athabaskan)
3. Catawban  (=Catawba + Woccons)
4. Eskimaux  (=Eskimoan)
5. Iroquois  (=Northern Iroquoian)
6. Cherokees  (=Southern Iroquoian)
7. Muskogee  (=Eastern Muskogean)
8. Chahtas  (=Western Muskogean)
9. Sioux  (=Siouan)

Languages

Families

1. Algonquian languages
2. Athabaskan languages
3. Catawban languages
4. Eskimoan languages
5. Iroquoian languages (Northern)
6. Iroquoian languages (Southern)
7. Muskogean languages
8. Siouan languages

Languages

John Wesley Powell, an explorer who served as director of the Bureau of American Ethnology, published a classification of 58 "stocks" that is the "cornerstone" of genetic classifications in North America. Powell's classification was influenced by Gallatin to a large extent.

John Wesley Powell was in a race with Daniel G. Brinton to publish the first comprehensive classification of North America languages (although Brinton's classification also covered South and Central America). As a result of this competition, Brinton was not allowed access to the linguistic data collected by Powell's fieldworkers.

(More current names are indicated parenthetically.)

Paul Rivet (1924) lists a total of 46 independent language families in North and Central America. Olive and Janambre are extinct languages of Tamaulipas, Mexico.[1]

Below is Edward Sapir's (1929) famous Encyclopædia Britannica classification. Note that Sapir's classification was controversial at the time and it additionally was an original proposal (unusual for general encyclopedias). Sapir was part of a "lumper" movement in Native American language classification. Sapir himself writes of his classification: "A more far-reaching scheme than Powell's [1891 classification], suggestive but not demonstrable in all its features at the present time" (Sapir 1929: 139). Sapir's classifies all the languages in North America into only 6 families: Eskimo–Aleut, Algonkin–Wakashan, Nadene, Penutian, Hokan–Siouan, and Aztec–Tanoan. Sapir's classification (or something derivative) is still commonly used in general languages-of-the-world type surveys. (Note that the question marks that appear in Sapir's list below are present in the original article.)

"Proposed Classification of American Indian Languages North of Mexico (and Certain Languages of Mexico and Central America)"

The Voegelin & Voegelin (1965) classification was the result of a conference of Americanist linguists held at Indiana University in 1964. This classification identifies 16 main genetic units.

Chumashan, Comecrudan, and Coahuiltecan included in Hokan with "reservations". Esselen is included in Hokan with "strong reservations". Tsimshian and Zuni are included in Penutian with reservations.

Campbell & Mithun's 1979 is a more conservative classification where they insist on more rigorous demonstration of genetic relationship before grouping. Thus, many of the speculative phyla of previous authors are "split".

(preliminary)

Subtiaba–Tlapanec is likely part of Otomanguean (Rensch 1977, Oltrogge 1977).

Aztec–Tanoan is "undemonstrated"; Mosan is a Sprachbund.

Glottolog 4.1 (2019) recognizes 42 independent families and 31 isolates in North America (73 total).[2] The vast majority are (or were) spoken in the United States, with 26 families and 26 isolates (52 total).

North American languages families proposed in Glottolog 4.1

Paul Rivet (1924) lists 77 independent language families of South America.[1]

Classification of South American languages by J. Alden Mason (1950):[7]

Čestmír Loukotka (1968) proposed a total of 117 indigenous language families (called stocks by Loukotka) and isolates of South America.[8]

Terrence Kaufman's classification is meant to be a rather conservative genetic grouping of the languages of South America (and a few in Central America). He has 118 genetic units. Kaufman believes for these 118 units "that there is little likelihood that any of the groups recognized here will be broken apart". Kaufman uses more specific terminology than only language family, such as language area, emergent area, and language complex, where he recognizes issues such as partial mutual intelligibility and dialect continuums. The list below collapses these into simply families. Kaufman's list is numbered and grouped by "geolinguistic region". The list below is presented in alphabetic order. Kaufman uses an anglicized orthography for his genetic units, which is mostly used only by himself. His spellings have been retained below.[14]

In addition to his conservative list, Kaufman list several larger "stocks" which he evaluates. The names of the stocks are often obvious hyphenations of two members; for instance, the Páes-Barbakóa stock consists of the Páesan and Barbakóan families. If the composition is not obvious, it is indicated parenthetically. Kaufman puts question marks by Kechumara and Mosetén-Chon stocks.

"Good" stocks:

• Awaké–Kaliana (Arutani–Sape)
• Chibcha–Misumalpa
• Ezmeralda–Jaruro
• Jurí–Tikuna
• Kechumara (=Kechua + Haki) (good?)
• Lule–Vilela
• Mosetén–Chon (good?)
• Páes–Barbakóa
• Pano–Takana
• Sechura–Katakao
• Wamo–Chapakúra

"Probable" stocks:

• Macro-Je (=Chikitano + Boróroan + Aimoré + Rikbaktsá + Je + Jeikó + Kamakánan + Mashakalían + Purían + Fulnío + Karajá + Ofayé + Guató)
• Mura–Matanawí

"Promising" stocks:

• Kaliánan (=Awaké + Kaliana + Maku)

"Maybe" stocks:

• Bora–Witoto
• Hívaro–Kawapana
• Kunsa–Kapishaná (now abandoned)
• Pukina–Kolyawaya
• Sáparo–Yawa

Kaufman's largest groupings are what he terms clusters and networks. Clusters are equivalent to macro-families (or phyla or superfamilies). Networks are composed of clusters. Kaufman views all of these larger groupings to be hypothetical and his list is to be used as a means to identify which hypotheses most need testing.

Lyle Campbell (2012) proposed the following list of 53 uncontroversial indigenous language families and 55 isolates of South America – a total of 108 independent families and isolates.[13] Language families with 9 or more languages are highlighted in bold. The remaining language families all have 6 languages or fewer.

Campbell (2012) leaves out the classifications of these languages as uncertain.

• Boran – Brazil, Colombia
• Esmeralda – Ecuador
• Guamo – Venezuela
• Leko – Bolivia
• Mure – Bolivia
• Puinave – Colombia, Venezuela

Jolkesky (2016) lists 43 language families and 66 language isolates (and/or unclassified languages) in South America – a total of 109 independent families and isolates.[15]^:783-806

• † = extinct

Glottolog 4.1 (2019) recognizes 44 independent families and 64 isolates in South America.[2]

South American languages families proposed in Glottolog 4.1

Morris Swadesh further consolidated on Sapir's North American classification and expanded it to group all indigenous languages of the Americas in just 6 families, 5 of which were entirely based in the Americas.[16]

1. Vasco-Dene languages included the Eskimo–Aleut, Na-Dene, Wakashan and Kutenai families along with most of the languages of Eurasia.
2. Macro-Hokan roughly comprised a combination of Sapir's Hokan–Siouan and Almosan families and expanded into Central America including the Jicaque language.
3. Macro-Mayan comprising Mayan along with Sapir's Penutian and Aztec-Tanoan families, the Otomanguean languages and various languages of Central and South America including the Chibchan languages, the Paezan languages and the Tucanoan languages.
4. Macro-Quechua comprising the Zuni language, the Purépecha language and various languages of South America including Quechua, the Aymara language, the Panoan languages and most of the various other languages of Patagonia and the Andes.
5. Macro-Carib, an almost entirely South American family including the Carib languages, the Macro-Je languages and the Jirajara languages, albeit including some Caribbean languages.
6. Macro-Arawak, a family primarily confined to South America and its component families included the Arawakan languages and the Tupian languages. However, it also was proposed to include the Taíno language in the Caribbean and the Timucua language in Florida.

Joseph Greenberg's classification[17] in his 1987 book Language in the Americas is best known for the highly controversial assertion that all North, Central and South American language families other than Eskimo–Aleut and Na-Dene including Haida, are part of an Amerind macrofamily. This assertion of only three major American language macrofamiles is supported by DNA evidence,[18] although the DNA evidence does not provide support for the details of his classification.