How and when was Japan settled by speakers of Japanese? Exploring the clues to Japanese prehistory preserved in old dialect divisions

Project Information

JapPrehistMigration

Grant agreement ID: 677317

DOI

10.3030/677317

Project closed

EC signature date 9 March 2016

Start date 1 May 2016

End date 31 October 2021

Funded under

• EXCELLENT SCIENCE - European Research Council (ERC)

Total cost

€ 1 394 040,00

EU contribution

€ 1 394 040,00

Coordinated by

RUHR-UNIVERSITAET BOCHUM
Germany

Periodic Reporting for period 4 - JapPrehistMigration (How and when was Japan settled by speakers of Japanese? Exploring the clues to Japanese prehistory preserved in old dialect divisions)

Reporting period: 2020-11-01 to 2021-10-31

Summary of the context and overall objectives of the project

The Japanese language is thought to have been introduced to Japan from the Korean peninsula alongside wet rice agriculture around 900 BCE. Despite the indigenous populations likely speaking many languages, only two languages are now represented in Japan: Japanese and Ainu. This indicates the replacement of the languages of hunter-gatherers with the language of a newly introduced agricultural society.
The project set out to shed light on some of the migratory routes that spread the Japanese language through the islands by studying the tone systems of the Japanese dialects. The research project was prompted by the observation that one of the tone systems in Japan is the result of innovations that did not occur in other tone systems, and that this tone system (the so-called Gairin or ‘outer circle’ sub-type of the Tokyo type tone systems) is found in four widely separated areas in the periphery of Japan. Apart from these innovations, the Gairin type tone system also seemed to be set apart by different tone rules for compound nouns (the first element deciding the tone of the compound, while in the other dialects in Japan this was thought to be decided by the second element of the compound).
The analysis of the Gairin tonal type as innovative follows from the reconstruction of the tone marks found in manuscripts from the Middle Japanese period (11th-14th century) proposed by Robert Ramsey in 1979. This reconstruction is the exact opposite on the one that is standard in Japan: In the 1930s, the tone marks found in the old manuscripts – mostly from the former capital of Kyoto – were deliberately given tonal values to align them with modern Kyoto’s tonal system, but this was a mistake. The Kyoto type tone system is a relatively recent innovation, limited to a large area around Kyoto. Exactly reversing the tonal values of the Middle Japanese marks aligns the old tone system with the majority of the tone systems in Japan, which are of the so-called Tokyo type.
The Gairin tone system is one of the sub-types of the Tokyo type tone systems. The things that set it apart could mean it represented a separate lineage that developed once, then spread to distant areas through migration.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Extensive fieldwork was conducted in the Gairin areas, the tones were analyzed, entered into a database and cross-referenced with the tones of the standard dialect of Tokyo and the tone dot material of Middle Japanese. Of each compound the tone class of the first and second member in Middle Japanese was marked.
It turned out that the rules for compounds in the modern dialects are fundamentally different than previously thought. In all dialects the role of the first element is crucial. If the first element belongs to a class that started with low tone in Middle Japanese, the compound contains no phonological high tone. If the first element belongs to a class that started with high tone in Middle Japanese the compound contains a phonological high tone somewhere in the word. The location of this high tone in the word largely depends on the tone class of the second element.
The link between the tone class of the first element and the presence or absence of high tone had not been noticed in the majority of tone systems in Japan, because in these modern tone systems the high-starting tone classes of Middle Japanese often merged with the low-starting tone classes of Middle Japanese. What had been regarded as unpredictable irregularities turned out to be linked to the tone class that the first element of the compound had belonged to in Middle Japanese.
In the Gairin type tone systems on the other hand, the merger pattern between the tone classes is different; the low starting tone classes of Middle Japanese have merged with each other, and so the link between the tone of the first element and the tone of the compound was overt. Coincidental circumstances had created the illusion of a dichotomy in the tone rules for noun compounds in the dialects of Japan.
Some regularizations in compound tone form regional patterns; in both Izumo and adjacent north Kyushu longer compounds have started regularizing the tone that characterized compounds with high-starting first elements in Middle Japanese. This reveals that the word-melody systems of northwest Kyushu developed from this type, as they display a similar pattern. Thus proves that they are not genetically close to the word-melody system of Kagoshima in southern Kyushu.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The modern compound tone rules of Japanese are usually seen as recent innovations, without any link to the tonal history of Japanese. Since the Middle Japanese tone system is not reconstructed correctly, there is no resemblance between compound tone in Middle Japanese and compound tone in the modern dialects. The outcomes of the project show that there is, in fact, a close link between the tone of compounds in Middle Japanese (and by extension proto-Japanese) and the tone of compounds in the modern dialects, but only if the Middle Japanese tone system is reconstructed in accordance with Ramsey's ideas. This proves beyond any doubt that the upside-down reconstruction of Japanese tonal history that is still prevalent in Japan needs to be abandoned.
The insight that a Gairin type tone system is the result of an innovation, but need not be seen as belonging to a separate lineage, based on its rules for compound tone, has solved a puzzle concerning three dialect areas along the Sea of Japan coast that has intrigued researchers since the 1950s. The widely separated dialects of Izumo, the Noto peninsula and the northeast of Japan share a very distinct segmental phonology. There are strong resemblances between the Gairin type tone systems of Izumo and the northeast, but the tone system of the Noto peninsula in-between the two is very different. Following my reconstruction of Japanese tonal history based on Ramsey’s theory, the Noto tone system is archaic, while the tone systems of Izumo and the northeast are innovative.
These three areas can now be regarded as representing different stages in a single dialect, separated through different waves of migration. The first migration took place in the 1st century CE, and brought a dialect with a distinct segmental phonology and an archaic tone system from the ancient kingdom of Izumo to the Noto peninsula. After innovations resulted in a Gairin-type tone system in Izumo, a second migration in the 6th century CE by-passed the Noto peninsula and brought rice farming culture and the distinct Gairin type tone system and segmental phonology of Izumo to the northeast of Honshu. This sequence of events explains the resemblances in segmental phonology between the three areas, and the archaic nature of the Noto tone system. It also explains why the resemblance in tone between Izumo and northeast Honshu is much closer.
Distinctive resemblances in the shape of burial tombs as well as mythology offer supporting evidence for the first migration and resemblances in DNA and folk music for the second. The most complete overview of the result of the project will be published open access in:
De Boer, Elisabeth M. 2022. “What tonal data can tell us about the history of the Japanese language: Concentrating on evidence from the Tokyo type tone systems.” Current Issues in Linguistics Theory, John Benjamins Publishing Company.

scan-20211230-5.jpg