Sex-based grammatical gender marking is rare in Austronesian languages, although it has emerged independently a handful of times in the family. Gorontalo, a Philippine-type language of northern Sulawesi, exhibits an innovative gender distinction in its case markers. Although common nouns do not always exhibit overt case marking, personal names are marked with obligatory case markers that precede the noun. These personal name case markers encode a sex-based gender distinction, agreeing in gender with their nouns. In addition to the default personal name markers ti 'pivot' (< PAN *si) and li 'non-pivot' (< PAN *ni), which are reflexes of the inherited Proto-Austronesian markers, Gorontalo also employs the innovative masculine personal name markers te 'pivot' and le 'non-pivot', inviting diachronic explanation.
A central issue in Austronesian higher-order subgrouping concerns the position of the Philippine languages. Despite considerable debate, it remains unsettled whether these languages represent multiple intersecting MalayoPolynesian primary branches forming a linkage, or whether they descend from a single shared ancestor distinct from Proto-Malayo-Polynesian. We present three new lines of evidence supporting the former view. First, the absence of a genuine *d/*z merger in Central Luzon and Minahasan languages undermines the sole phonological argument for a Proto-Philippine branch. Second, the geographical distribution of the purported Proto-Philippine-defining lexicon favors a diffusion-based explanation. Third, the distribution of an underexplored morphological innovation across Philippine languages suggests frequent contact among major island groups, excluding interior and peripheral areas. These findings not only highlight the absence of defining innovations for a cohesive Proto-Philippine but also point to extensive horizontal transmission across the archipelago, extending beyond lexical exchange. We further propose that the *d/*z merger is better understood as an areal drift rather than a diagnostic innovation. We conclude that Philippine languages are best analyzed as intersecting Malayo-Polynesian primary branches, in line with recent perspectives. Moreover, the high number of proposed lexical innovations attributed to Proto-Philippine likely reflects contact-driven processes—diffusion, borrowing, and linkage histories—rather than retention from a unified ancestor.
Based on over 240 firsthand data, this paper provides the first formal analysis of empirical generalizations regarding the suprasegmental adaptation of Japanese loanwords in Isbukun Bunun, an Austronesian language spoken in Taiwan. This study reveals that during the loanword adaptation process, the position of the Japanese accent is entirely disregarded, and stress in Japanese loanwords is generally assigned according to native phonological rules. However, a noteworthy exception is observed: closed syllables, traditionally considered light in native phonology, unexpectedly attract stress in loanwords. Interestingly, closed syllables do not always maintain their heaviness in loanwords and may become light when preceded by a bimoraic syllable. This unexpected weight contribution from the coda suggests that Japanese loanwords do not adhere to the same phonological constraints as native words, and the variation in the weight of closed syllables indicates a flexible ranking between a constraint favoring moraic codas and a constraint penalizing unstressed heavy syllables.
This article reconstructs the gender vocalics of the South Bird’s Head family and compares them with those reconstructed for two other Trans New Guinea families, Anim and Ok. Although chance cannot be convincingly ruled out, the close correspondences between the gender vocalics of the Proto-South Bird’s Head family and those of Proto-Anim and Proto-Ok suggest the possibility of a shared innovation that replaced the third singular pronoun *[y]a/ *ua of Proto-Trans New Guinea by two gendered demonstrative pronouns, *e ‘that. M’ and *u ‘that. F’. The demonstrative vocalics formed the starting point for grammaticalization processes that resulted in similar systems of elaborate gender vocalism in the three families.
