The indigenous languages of South America are those whose origin dates back to the pre-Columbian era. The subcontinent has great linguistic diversity, but, as the number of speakers of indigenous languages is diminishing, it is estimated that it could become one of the least linguistically diverse regions of the planet.
About 600 indigenous languages are known from South America, Central America, and the Antilles (see List of indigenous languages of South America), although the actual number of languages that existed in the past may have been substantially higher.
Origins The indigenous languages of South America, Central America and the Antilles completely covered the subcontinent and the Antilles at the beginning of the 16th century. The estimates of the total population are very imprecise, ranging between ten and twenty million inhabitants. At the beginning of 1980, there were about 16 million speakers of indigenous languages; three quarters of them lived in the Central Andes.
The number of existing tribes and ethnic groups is around 1500, although some authors have suggested it might reach 2000. However, it is uncertain if each of these groups has a unique language, so the figures likely indicate an upper bound for the actual number of languages spoken. For many of the known historical groups, there is no record of their language and many of them are extinct today. There is only one record of some 550 or 600 languages, with some 180 of them totally extinct today. In many cases, the fragmented records do not allow us to decide if they have to do with different languages or of divergent but mutually intelligible dialects of the same language.
Relationships with North America and Mesoamerica Since the indigenous people of South America are historically from North America the problem of the origin involves looking for genetic and linguistic relationships with the indigenous groups of North America and Mesoamerica. Currently the only linguistic family of South America that shows a relationship with languages situated outside the region are the Chibchan languages, for which some evidence has shown that they are related to the Misumalpan and Lencan languages of Central America. In the 1970s, it was proposed that the Uru-Chipaya languages of Bolivia could be related to the Mayan languages of Mesoamerica, but while this proposal was initially accepted, Campbell argued very strongly against it.
Study of the South American languages
The first grammar of a South American language was that of classical Quechua published by Domingo de Santo Tomás in 1560. The missionaries of the seventeenth and first half of the eighteenth centuries carried out an intense activity of data collection, grammar writing (usually called language arts), dictionaries, and catechisms, in order to evangelize the indigenous populations. A good amount of linguistic records also appear in chronicles and official records. Much of the information from this period was summarized by Lorenzo Hervás y Panduro in his work Idea dell'universo (1778–87), and in the work by Johann Christoph Adelung and Johann Severin Vater Mithridates (1806–17). Subsequently, most of the information collected first hand was compiled by ethnographers in the first half of the twentieth century. Despite the magnitude and fundamental nature of the works of this period, its technical quality is below that achieved in other parts of the world, which is why South America, along with New Guinea, was one of the worst-known parts from a linguistic point of view.
Since the 1940s, the number of works about the languages of South America has grown significantly, carried out fundamentally by linguists and missionaries well trained in linguistics. However, there still exist many important gaps that affect the descriptive level, and few languages have been widely described. That has hurt comparative, historic, and typological work on the languages of South America. The descriptive work has had problems due to the shortage of linguists and the rapid extinction of many languages, often situated in remote and difficult to reach areas that require urgent study before they disappear completely. These languages produce scientific interest since their vocabulary reflects the traditional culture and contains important data about the fauna, flora, and local history of little-known regions. In addition, indigenous people have a right to bilingual education in some countries and for that education to be successful, it is important to rely on good descriptive material of indigenous languages.
Urheimat of some linguistic families For some of the main linguistic families of South America, homeland groups have been proposed, or Urheimats, grouping them from where they originated. For example, it seems quite clear that Tupian languages expanded from Rondônia, which is an area of greater diversification; in fact, almost all Tupian languages outside of Rondônia belong to only one branch of the nine branches that form the Tupian family. This principle that the original area of a linguistic family is usually also the most diversified has been questioned by some authors, although it is one of the main tools to propose an area of original expansion for most language families. Recently, the phylogenetic relationship between Chibchan languages, Lencan languages, and Misumalpan languages has been demonstrated, which suggests that these languages originated in a region near the south of the Mesoamerican area and therefore, Chibchan languages will have expanded from the north to Panama and northeast of Colombia. Similarly, it is conjectured that the Caribbean languages would have expanded from east to west, and from there some groups moved to the eastern Amazon and others to the Caribbean islands, where they were expanding in the fifteenth century upon the arrival of the Europeans to America.
Jolkesky (2016) speculates that from around 2,200 B.C. to 1,800 B.C., the banks of the Amazon River had been occupied by a continuous chain of language families. They are listed below in order, beginning from the mouth of the Amazon River to the upper Amazon and Marañón Rivers.Jolkesky, Marcelo Pinho De Valhery. 2016. Estudo arqueo-ecolinguístico das terras tropicais sul-americanas. Ph.D. dissertation, University of Brasília.
Pre-Proto-Cariban (from the Xingu to Tapajós confluence) Pre-Proto-Nambikwaran (Tapajós confluence) Pre-Proto-Peba-Yagua (Madeira confluence) Pre-Proto-Kwaza (Madeira confluence) Pre-Proto-Puinave-Nadahup (Lower Madeira River) Pre-Proto-Jivaroan (Rio Negro confluence) Pre-Proto-Yanomami (along the Lower Rio Negro) Pre-Proto-Arawan (Purus confluence) Pre-Proto-Harakmbut-Katukinan (Tefé confluence) Pre-Proto-Bora-Muinane (Japurá, Jutaí, Juruá confluences) Proto-Pano-Takanan (Içá and Jutaí confluences) Pre-Proto-Witotoan (upper Amazon River) Pre-Proto-Zaparoan (Napo confluence) Pre-Proto-Urarina (Ucayali and Marañón confluence) Pre-Proto-Cahuapanan (Huallaga and Marañón confluences)
Urheimat of other language families: Pre-Proto-Chibchan: Lower Magdalena River Valley Pre-Proto-Warao: Orinoco Delta Proto-Duho: Upper Inírida, Upper Vaupés, and Upper Apaporis basins Pre-Proto-Tukanoan: Andean foothills and Upper Caquetá basin Pre-Proto-Barbacoan: Ecuadorian Andes Pre-Proto-Mochica: Daule River basin Pre-Proto-Cholon-Hibito: Upper Santiago and Upano basins Pre-Proto-Macro-Arawakan: Middle and Upper Ucayali basins Pre-Proto-Candoshi-Shapra: Upper Huallaga basin in the Peruvian Highlands Pre-Proto-Quechuan: Lake Junín region Pre-Proto-Jaqi: middle Mantaro basin in the Peruvian Highlands Pre-Proto-Mapudungun: coast of southern Peru Pre-Proto-Kunza: sources of the Urubamba and Apurimac Rivers Pre-Proto-Uru-Chipaya: shores of Lake Titicaca Proto-Tupian: Roosevelt and Upper Aripuanã basins Proto-Macro-Jê : Mato Grosso, from the São Lourenço River to the Rio das Mortes Pre-Proto-Macro-Mataguayo-Guaykuru: Pantanal
Classification Although some of the classifications are based on geographic, ethnographic, or cultural criteria, these methods are not valid from a linguistic point of view, even though on some occasions there are correlations between them and a genuine phylogenetic relationship. The previous correlation is supported only in branches or subgroups, but within the more extensive and diversified linguistic families, the correlation between cultural similarity and linguistic relationship is notably reduced, becoming random and arbitrary. Although families like Cariban languages or Tupian languages are formed by villages with typical cultures of the tropical jungle, there are villages that speak Tupian languages, like Aché and Sirionó, who have very different cultures from those of the tropical jungle. And equally, the villages of a homogeneous cultural area like the eastern slopes of the Andes in fact, belong to unrelated linguistic families. In the same manner, isolated languages, or less diversified families, tend to concentrate in marginal areas, while Quechua, that is a family of few languages not very diversified, occupies a very prominent place both from a historical and a demographic point of view.
Problems with the classification The major part of the proper linguistic classification of the languages of South America has been done on the basis of vocabulary lists and some grammatical traits. That procedure, although it leads to the clear recognition of the top-level genetic groups, does not distinguish well between the coincidences and the lexical loans of the words retained from the common proto-language. Also the glottochronology, which is more criticizable to find relationships of higher levels, has been used extensively. Just recently the comparative method has been applied carefully and patiently to find relationships between the identifiable phylogenetic subgroups from a simple list of vocabulary. For that reason, the phylogenetic classifications of the languages of America are far from definitive, and the best of them, in the best of cases, are only an approximation of the real relationships. For that reason, many proposed families like Macro-Arawakan or Macro-Chibchan are questionable because evidence that exists in their favor is sporadic, debatable, and not very solid.
The number of dead languages with scarce records is also very high among the languages of South America. Those languages are frequently labeled as unclassified languages, when in reality they are unclassifiable because the recorded material relies on unverifiable philological interpretations and extremely scarce data with which it is not possible to establish a relationship unequivocally, as is the case with living languages for which data can be collected in sufficient numbers to decide if they are related to other languages or are genuine isolated languages.