Ainu
Ainuic
Geographic distribution	Currently only Hokkaido; formerly also southern and central Sakhalin, the Kuril Islands and possibly northern Honshu
Linguistic classification	One of the world's primary language families
Subdivisions	Hokkaido Ainu †Sakhalin Ainu †Kuril Ainu
ISO 639-2 / 5	ain
ISO 639-3	ain
Glottolog	ainu1252
ELP	Ainu (Japan)
Map of the historical distribution of Ainu languages and dialects

The Ainu languages (/ˈaɪnuː/ EYE-noo),^[1] sometimes known as Ainuic, are a small language family, often regarded as a language isolate, historically spoken by the Ainu people of northern Japan and neighboring islands.

The primary varieties of Ainu are alternately considered a group of closely related languages^[2] or divergent dialects of a single language isolate. The only surviving variety is Hokkaido Ainu, which UNESCO lists as critically endangered. Sakhalin Ainu and Kuril Ainu are now extinct. Toponymic evidence suggests Ainu was once spoken in northern Honshu and that much of the historically attested extent of the family was due to a relatively recent expansion northward. No genealogical relationship between Ainu and any other language family has been demonstrated, despite numerous attempts.

Varieties

Recognition of the different varieties of Ainu spoken throughout northern Japan and its surrounding islands in academia varies. Shibatani (1990:9) and Piłsudski (1998:2) both speak of "Ainu languages" when comparing the varieties of language spoken in Hokkaidō and Sakhalin; however, Vovin (1993) speaks only of "dialects". Refsing (1986) says Hokkaidō and Sakhalin Ainu were not mutually intelligible. Hattori (1964) considered Ainu data from 19 regions of Hokkaidō and Sakhalin, and found the primary division to lie between the two islands.

Hokkaidō Ainu

Hokkaidō Ainu clustered into several dialects with substantial differences between them: the 'neck' of the island (Oshima County, data from Oshamambe and Yakumo); the "classical" Ainu of central Hokkaidō around Sapporo and the southern coast (Iburi and Hidaka counties, data from Horobetsu, Biratori, Nukkibetsu and Niikappu; historical records from Ishikari County and Sapporo show that these were similar); Samani (on the southeastern cape in Hidaka, but perhaps closest to the northeastern dialect); the northeast (data from Obihiro, Kushiro and Bihoro); the north-central dialect (Kamikawa County, data from Asahikawa and Nayoro) and Sōya (on the northwestern cape), which was closest of all Hokkaidō varieties to Sakhalin Ainu. Most texts and grammatical descriptions we have of Ainu cover the Central Hokkaidō dialect.

Kuril Ainu

Data on Kuril Ainu is scarce, but it is thought to have been as divergent as Sakhalin and Hokkaidō.

Sakhalin Ainu

In Sakhalin Ainu, an eastern coastal dialect of Taraika (near modern Gastello (Poronaysk)) was quite divergent from the other localities. The Raychishka dialect, on the western coast near modern Uglegorsk, is the best documented and has a dedicated grammatical description. Take Asai, the last speaker of Sakhalin Ainu, died in 1994.^[3] The Sakhalin Ainu dialects had long vowels and a final -h phoneme, which was pronounced [x].

Scant data from Western voyages at the turn of the 19th–20th century (Tamura 2000) suggest there was also great diversity in northern Sakhalin, which was not sampled by Hattori.

Classification

Vovin (1993) splits Ainu "dialects" as follows:^[4]

Proto-language

The proto-language was reconstructed twice by Alexander Vovin.^{[5] [6]}

The second reconstruction shows the voiced stops except for [b] being distinct phonemes and uses ⟨*q⟩ for the glottal stop.^{[Is this a doubtful reconstruction?]} He also tentatively proposes that there might have been a third fricative alongside *s and *h, which was voiced, its place of articulation unknown. He represents it with ⟨*H⟩.

Reconstructed Proto-Ainu numerals (1-10) and its reflexes in selected descendants are as follows:

Eight front and back vowels are reconstructed; three more central vowels are uncertain.

External relationships

No genealogical relationship between Ainu and any other language family has been demonstrated, despite numerous attempts. Thus, it is a language isolate. Ainu is sometimes grouped with the Paleosiberian languages, but this is only a geographic blanket term for several unrelated language families that were present in easternmost Siberia before the advances of Turkic and Tungusic languages there.

A study by Lee and Hasegawa of Waseda University found evidence that the Ainu language and the early Ainu-speakers originated from the Northeast Asian/Okhotsk population, which established themselves in northern Hokkaido and expanded into large parts of Honshu and the Kurils.^[8]

The Ainu languages share a noteworthy amount of vocabulary (especially fish names) with several Northeast Asian languages, including Nivkh, Tungusic, Mongolic, and Chukotko-Kamchatkan. While linguistic evidence points to an origin of these words among the Ainu languages, its spread and how these words arrived into other languages will possibly remain a mystery.^[9]

The most frequent proposals for relatives of Ainu are given below:

Altaic

John C. Street (1962) proposed linking Ainu, Korean, and Japanese in one family and Turkic, Mongolic, and Tungusic in another, with the two families linked in a common "North Asiatic" family. Street's grouping was an extension of the Altaic hypothesis, which at the time linked Turkic, Mongolic, and Tungusic, sometimes adding Korean; today Altaic sometimes includes Korean and rarely Japanese but not Ainu (Georg et al. 1999).

From a perspective more centered on Ainu, James Patrie (1982) adopted the same grouping, namely Ainu–Korean–Japanese and Turkic–Mongolic–Tungusic, with these two families linked in a common family, as in Street's "North Asiatic".

Joseph Greenberg (2000–2002) likewise classified Ainu with Korean and Japanese. He regarded "Korean–Japanese-Ainu" as forming a branch of his proposed Eurasiatic language family. Greenberg did not hold Korean–Japanese–Ainu to have an especially close relationship with Turkic–Mongolic–Tungusic within this family.

The Altaic hypothesis is now rejected by the scholarly mainstream.^{[10][11][12][13]}

Austroasiatic

Shafer (1965) presented evidence suggesting a distant connection with the Austroasiatic languages, which include many of the indigenous languages of Southeast Asia. Vovin (1992) presented his reconstruction of Proto-Ainu with evidence, in the form of proposed sound changes and cognates, of a relationship with Austroasiatic. In Vovin (1993), he still regarded this hypothesis as preliminary.

Language contact with the Nivkhs

The Ainu appear to have experienced intensive contact with the Nivkhs during the course of their history. It is not known to what extent this has affected the language. Linguists believe the vocabulary shared between Ainu and Nivkh (historically spoken in the northern half of Sakhalin and on the Asian mainland facing it) is due to borrowing.^[14]

Language contact with the Japanese

The Ainu came into extensive contact with the Japanese in the 14th century. Analytic grammatical constructions acquired or transformed in Ainu were probably due to contact with the Japanese language. A large number of Japanese loanwords were borrowed into Ainu and to a smaller extent vice versa.^[15] There are also a great number of loanwords from the Japanese language in various stages of its development to Hokkaidō Ainu, and a smaller number of loanwords from Ainu into Japanese, particularly animal names such as rakko ('sea otter'; Ainu rakko), tonakai ('reindeer'; Ainu tunakkay), and shishamo (a fish, Spirinchus lanceolatus; Ainu susam). Due to the low status of Ainu in Japan, many ancient loanwords may be ignored or undetected, but there is evidence of an older substrate, where older Japanese words which have no clear etymology appear related to Ainu words which do. An example is modern Japanese sake or shake, meaning 'salmon', probably from the Ainu sak ipe or shak embe for 'salmon', literally 'summer food'.

According to P. Elmer (2019), the Ainu languages are a contact language, i.e. have strong influences from various Japonic dialects/languages during different stages, suggesting early and intensive contact between them somewhere in the Tōhoku region, with Ainu borrowing a large amount of vocabulary and typological characteristics from early Japonic.^[16]

Other proposals

A small number of linguists suggested a relation between Ainu and Indo-European languages, based on racial theories regarding the origin of the Ainu people. The theory of an Indo-European—Ainu relation was popular until 1960; later linguists dismissed it and concentrated on more local language families.^[17][18]

Tambovtsev (2008) proposes that Ainu is typologically most similar to Native American languages and suggests that further research is needed to establish a genetic relationship between these languages.^[19]

Geography

Until the 20th century, Ainu languages were spoken throughout the southern half of the island of Sakhalin and by small numbers of people in the Kuril Islands. Only the Hokkaido variant survives, with the last speaker of Sakhalin Ainu having died in 1994.

Some linguists note that the Ainu language was an important lingua franca on Sakhalin. Asahi (2005) reported that the status of the Ainu language was rather high and was also used by early Russian and Japanese administrative officials to communicate with each other and with the indigenous people.^[20]

Ainu on mainland Japan

It is occasionally suggested that Ainu was the language of the indigenous Emishi people of the northern part of the main Japanese island of Honshu.^[a] The main evidence for this is the presence of place names that appear to be of Ainu origin in both locations. For example, the -betsu common to many northern Japanese place names is known to derive from the Ainu word 'pet' ("river") in Hokkaidō, and the same is suspected of similar names ending in -be in northern Honshū and Chūbu, such as the Kurobe and Oyabe rivers in Toyama Prefecture.^[21] Other place names in Kantō and Chūbu, such as Mount Ashigara (Kanagawa–Shizuoka), Musashi (modern Tokyo), Keta Shrine (Toyama), and the Noto Peninsula, have no explanation in Japanese, but do in Ainu. The traditional matagi hunters of the mountain forests of Tōhoku retain Ainu words in their hunting vocabulary.^[22][23] However, Elmer (2019) has also suggested Japonic etymologies, which supposedly got borrowed into early Ainu and lost in contemporary Japonic dialects.^[16]

The direction of influence and migration is debated. It has been proposed that at least some Jōmon period groups spoke a proto-Ainu language,^[24] and that they displaced the Okhotsk culture north from southern Hokkaido when the Ainu fled Japanese expansion into northern Honshu, with the Okhotsk ancestral to the modern Nivkh as well as a component of the modern Ainu. However, it has also been proposed that the Ainu themselves can be identified with the Okhotsk culture, and that they expanded south into northern Honshu as well as to the Kamchatka Peninsula,^[8][25] or that the Emishi spoke a Japonic language, most closely related to ancient Izumo dialect, rather than anything related to Ainu, with Ainu-speakers migrating later from Hokkaido to northern Tōhoku. The purported evidence for this are old-Japanese loanwords in the Ainu language, including basic vocabulary, as well as distinctive Japonic terms and toponyms found in Tōhoku and Hokkaido, that have been linked to the Izumo dialect.^[26]