The investigations reported in the chapters of this book were motivated by the
discovery that the Salawati dialect of the Austronesian language Maya features a
hybrid word prosodic system with both contrastive lexical stress and lexical tone
(see Chapter II). No other language has been reported to feature such a word
prosodic system. This discovery raised two questions:
How did Maya develop a word prosodic system with both stress and tone?
An answer to this question could reveal why such a hybrid system is so
What are the word prosodic systems of closely related languages like, and
what kind of dialectal variation do we find between Maya dialects in terms
of their word prosodic systems? Investigations into these related word
prosodic systems could produce more interesting data.
Question I was answered in Chapter VI, where Maya words were compared with
cognates in related languages in order to find evidence of segmental processes that
have implications for the word prosodic system. The main finding of this
investigation was that, unlike most other Asian tone languages (e.g. Thai, Chinese),
tonogenesis in Maya did not go together with a trend towards a more monosyllabic
vocabulary. As a consequence, the stress feature which Maya had inherited from its
ancestor language was retained in the lexicon. I have hypothesized that the reason
why tonogenesis in Maya was not accompanied by a trend towards monosyllabism
has to do with contact situation between Austronesian and Papuan languages (see §
As for Question II, the study of the word prosodic systems of other dialects of
Maya and other related languages began with a survey of the language situation of
the Raja Ampat archipelago (Chapter II). Apart from being welcome in its own
right, this survey revealed that three out of five Maya dialects feature lexical tone,
and that there is one other tone language in the Raja Ampat archipelago: Matbat.
The dialectal variation in the tone system of Maya and the Matbat tone system were
investigated in Chapter IV and V, respectively.
The discovery of a five-toneme contrast for Matbat is surprising, since it is only
the fifteenth tone language out of 1,236 Austronesian languages. This discovery, in
combination with evidence of Papuan influence in the lexicon, suggests that Matbat
developed this feature through contact with non-Austronesian tone language. It also
constitutes indirect support for the hypothesis that Maya developed its tone system
through contact with a Papuan language.
The existence of three Maya dialects with lexical tone provided an opportunity to
gain insight in how tone languages change over time. Since the differences between
the tone systems of the three dialects were relatively small, it was possible to
determine the origin of changes by means of between-dialect comparison. This study
revealed that the Misool dialect has undergone a push-chain tone shift, and that the
Laganyan dialect had developed a boundary tone. Both of these phenomena can be
traced back to synchronic variation in Maya before the dialect split, since evidence
of this synchronic variation has been retained in some dialects.