Lexical Evidence for Early Contact between Indonesian Languages and Japanese

Abstract

Forty-one pairs of words with CVCVC structure selected from Old Japanese and Old Javanese dictionaries are presented. It is claimed that these are the result of borrowing into an antecedent of Old Japanese from an Indonesian source. Semantic relationships are discussed, and sound correspondences are specified within a discussion of the segmental phonology and phonotactics of the two languages. The agreement in phonological form is shown to be extensive, applying in some cases to up to five segments in each word pair, and to also make sense, given the phonotactic restrictions of the recipient language. The semantic agreement is often also of comparably high specificity, showing moreover a further level of structure in its partial resolution into semantic fields, including some that resonate with nonlinguistic findings related to ritual and rice cultivation in the Yayoi period of early Japanese history. The amount of agreement in semantic and phonological form is shown statistically to be greater than chance. The argument is further strengthened by several additional independent levels of agreement in the data that are discussed within a Bayesian framework. The phonological correspondences map unidirectionally from Old Javanese to Old Japanese, and a search for cognates in all Austronesian languages covered by the major comparative dictionaries reveals that the lexical items are localized to the Indonesian subarea of Malayo-Polynesian. This points to one or more Indonesian languages as the source of the borrowings. The agreement between semantic and archaeological evidence on material and spiritual culture dates the contact to the Yayoi period. The semantic evidence further suggests that, contrary to the received view, important Yayoi innovations are likely to have been introduced into Japan from the south, and not from China or Korea as usually supposed.

1. Introduction¹

1.1 Aim.

We wish in this paper to present some linguistic data that are indicative of an early interaction between Japanese and one or more of the so-called "Western Malayo-Polynesian" languages, most probably of Indonesian origin.

1.2 Background.

Nonlinguistic evidence from rice genetics and rituals relating to kingship (Kumar 1992), and from biological anthropology and mitochondrial DNA (Kumar 1998), has provided considerable support for a Western Malayo-Polynesian, and most probably specifically Indonesian, influence on early Japan. The first author has suggested that this took place in the Yayoi period (early centuries B.C. and A.D.). An attempt was therefore made to find out whether any linguistic evidence also existed to support this hypothesis. It was assumed that if such evidence existed, it would, of course, be locatable in historical varieties of Japanese, and in some Western Malayo-Polynesian (WMP) languages.

The results of the investigation showed that there was indeed lexical influence, and this is described in detail below. We first present background information on the two languages we compared—Old Japanese and Old Javanese—and give reasons why the latter was selected as our WMP language. Next we describe the procedure we used to compare lexical material in Old Japanese and Old Javanese. We then present and discuss the results. The second author also addresses the question of whether the high degree of phonological and semantic similarity demonstrated can be shown probabilistically to be indicative of nonfortuitous, bona-fide borrowing.

Attempts to link Japanese genetically with other languages abound. We would like to emphasize that we make no claim in this paper for genetic relationship.

2. Languages Compared

2.1 Old Japanese.

The history of Japanese is divided into a number of different periods. The first, Proto-Japanese, may be dated to the period 300-400 B.C. to 300-400 A.D., a time span corresponding to the Yayoi culture, possibly continuing into the succeeding Kofun period. It is generally agreed that Old Japanese is the language of the oldest extant written records and of eighth-century Nara literary works such as the Kojiki and Man'yōshū, but scholars differ as to when it ends. Some scholars consider Old Japanese to go up to the tenth century at the latest, while Miller (1967:37-39) distinguishes a "late Old Japanese" that includes...