Proto Malayo-Polynesian Research Articles

South Borneo as an ancient Sprachbund area

In South and Central Kalimantan (southern Borneo) there are some unusual linguistic features shared among languages which are adjacent but do not belong to the same genetic linguistic subgroups. These languages are predominantly Banjar Malay (a Malayic language), Ngaju (a West Barito language), and Ma’anyan (a Southeast Barito language). The same features also appear to some degree in Malagasy, a Southeast Barito language in East Africa. The shared linguistic features are the following ones: a grammaticalized form of the originally Malay noun buah ‘fruit’ expressing affectedness, nasal spreading in which N - not only nasalizes the onset of the first syllable but also a *y in the next syllable, a non-volitional marker derived from the Banjar Malay prefix combination ta-pa- (related to Indonesian t ə r - + p ə r -), and the change from Proto Malayo-Polynesian *s to h (or Malagasy Ø). These features have their origins in the various members of the language configuration outlined above and form a Sprachbund or “Linguistic Area”. The concept of Linguistic Area is weak and difficult to define. Lyle Campbell (2002) considers it little else than borrowing or diffusion and writes it off as “no more than [a] post hoc attempt [..] to impose geographical order on varied conglomerations of [….] borrowings”. While mindful of its shortcomings, the current author still uses the concept as a useful tool to distinguish between inherited and borrowed commonalities. In the configuration of languages currently under discussion it also provides a better understanding of the linguistic situation in South Borneo at a time prior to the Malagasy migrations to East Africa (some thirteen centuries ago).

Comment on Blust "The Resurrection of Proto-Philippines"

Comment on Blust "The Resurrection of Proto-Philippines" Malcolm Ross I thank the editors of Oceanic Linguistics for inviting me to comment on Blust's paper arguing for the integrity of a Philippine subgroup within Malayo-Polynesian (MP). As Blust critiques the rejection of Proto-Philippines (PPh) in Ross (2005), l address that paper briefly first. Accepting that all Austronesian languages outside Taiwan belong to a single branch of Austronesian that has no members in Taiwan, in 2005 it was a reasonable geography-based inference that Proto-Malayo-Polynesian (PMP) might have come into being in the Batanes Islands and, if so, that the Batanic languages were a first-order offshoot of MP. This was Ross's (2005) "History 1." The alternative hypothesis, "History 2," was that the Batanic languages were clearly MP but not a first-order offshoot and reached their current location at a later stage. The paper reached no firm conclusion but commented, as Blust notes, that "there is a greater likelihood that History 1 is correct," because the Batanic languages appear very conservative. However, "the evidence for this conclusion is circumstantial, and does not have the same status as subgrouping evidence based on shared innovations." History 1 would entail that there was no Philippine subgroup, or, if there was, that the Batanic languages did not belong to it. If I had thought a Philippine subgroup had been well established, I could not have been proposed History 1. At the time of writing, the evidence for a Philippine subgroup seemed insufficient to rule out History 1. The published evidence in 2005 consisted of the lexical innovations proposed by Zorc (1986) in response to Reid's (1982) rejection of a pan-Philippine subgroup. Since that time further data and discussion have appeared in the form of a shortish discussion in Blust (2005:34–37) with a substantial appendix of lexical innovations, and, now in Blust (2020) detailed argumentation for the integrity of PPh, including discussion of individual lexical innovations and a note of the distribution of the reflexes of each, as well as a single phonological innovation. This brings me to the question: Fifteen years on, would I write something like Ross (2005) now? I would not. The new evidence strengthens the case for a Philippine grouping of some kind with Batanic as a somewhat peripheral [End Page 366] member subgroup—but it does not unequivocally support the reconstruction of PPh in the sense that a protolanguage is usually understood. Blust proposes that PPh is now supported by a single phonological innovation, the merger of PMP *z and *d. The remainder of its innovations is lexical, and in support of PPh Blust appeals to their very large number: 1,222 in appendix 1, plus the 37 in tables 2 and 3. I have paid particular attention to the latter, as Blust labels them as "the best" or "strong" replacement innovations (henceforth, "the strong innovations"), and have also looked at the distribution of all 1,259 across his microgroups. Together with supporting data on line in Blust and Trussel (ongoing), this is a wonderful collection of reconstructions. Despite this quantity of evidence, the argument for PPh seems to me to be weak in three interrelated respects. One is that the innovations are not distributed as one would expect of a proper subgroup, that is, the exclusive descendants of a single language; the second is the putative position of Philippine languages within the MP tree; and the third is the difficulty inherent in using lexical innovations. The distribution of the innovations cited in support of PPh does not attest that the Philippine languages form a proper subgroup. A proper subgroup is ideally defined by one or more innovations shared by all its members and exclusive to the languages of that group. Such innovations must satisfy two conditions, namely (1) that the innovation has not occurred independently in various members of the subgroup and (2) that it has not been borrowed across language boundaries. This defines an ideal, and one that is rarely achieved for large language groups. Sometimes an innovation is obliterated by further innovation in some members of a group. At times...

Dual *kita in the History of East Barito Languages

In many Philippine, northern Sulawesi, and northern Bornean languages, Proto Austronesian *kita 'first-person inclusive plural' became a first-person inclusive dual pronoun. Robert Blust and Hsiu-chuan Liao attribute this semantic change to drift (a change happening in various related languages independently). However, Lawrence Reid contends that it had already happened in Proto Malayo-Polynesian, and that the ensuing gap in the pronominal system of this ancestral language had been filled by the formation of a new first-person inclusive plural pronoun, which was based on *kita combined with a pronominal clitic (or "extender") *=mu. The latter was a second-person plural pronoun in Proto Austronesian, but after it had lost its plural meaning in Proto Malayo-Polynesian, it was often combined with or replaced by other pronominal extenders.

The Case-Markers of Proto-Austronesian

In striking contrast to the voice-marking affixes, for which there is universal agreement, the case-markers of Proto-Austronesian (PA n ), and later protolanguages such as Proto-Philippines (PP h ), have long resisted detailed reconstruction. Four sets of case-markers (nominative, genitive, oblique, locative) can be posited for PA n , each distinguished by its application to personal names vs. common nouns, and singular vs. plural reference in the former category. Through standard use of the comparative method, it is shown that PA n must have had at least the genitives *ni, *na, *nu, and nominatives *si, *sa, where *i correlates with singular personal names, *a with plural personal names, and *u with common nouns. For PP h , application of the comparative method allows the nominatives to be expanded to *si, *sa, *su. Although the oblique and locative sets show much greater functional variability, the case-markers *ki, *ka, *ku, and *di, *da must be reconstructed for PA n (plus *du for PP h ), thus presenting the analyst with a glossing challenge for these forms that is not found with *ni, *na, *nu, or *si, *sa, (*su). By a process of elimination, *ki and *ku can be posited for PA n , Proto–Malayo-Polynesian (PMP), and PP h as the oblique of singular personal names and of common nouns, respectively. Although PP h *ka can be added to this set as the oblique of plural personal names, this interpretation can be applied to PMP and PA n only by generalizing the semantic values of the *i/a/u vowel template to cases where they are not directly attested. For the locatives, *di can be reconstructed without further specification, but *da and *du present additional challenges that are ultimately met by allowing the boundary between obliques and locatives to be permeable. The result is four tripartite sets of case-markers for PP h (nominatives *si, *sa, *su, genitives *ni, *na, *nu, obliques *ki, *ka, *ku, locatives *di, *da, *du), of which all but *su and *du can be attributed to PA n , and all but *du to PMP, suggesting replacement of these terms in all Formosan languages. A fifth set of case-markers *i, *a, *u may have existed, but details regarding its function remain elusive.

Malagasy Personal Pronouns: A Lexical History

This paper traces the history of pronouns in various regional forms of Malagasy and proposes a reconstruction of Proto-Malagasy pronouns. Four sets of pronouns are reconstructed for Proto-Malagasy: a default nominative set marked with Ø, a topicalized nominative set in which 1st person pronouns are marked with a form *i, a genitive set marked with *=n-, and an oblique set marked with *an=. The development of some pronouns is shown to provide clues for the internal classification of Malagasy varieties. The Proto-Malagasy pronouns are also compared with external references and higher-order reconstructions, namely pronouns from the closely related Southeast Barito languages in Borneo and Proto–Malayo-Polynesian. Finally, an attempt is made to reconstruct Proto-Southeast Barito pronouns.

Dobel Historical Phonology

The Aru Islands of the southern Moluccas in eastern Indonesia are linguistically among the most neglected regions in the Austronesian world. This paper draws on a published vocabulary of about 1,000 words for Dobel, spoken in east-central Aru, and outlines its historical development from Proto–Malayo-Polynesian. Extensive sound change has concealed the origins of a number of forms that become apparent through application of the comparative method. One of the most salient features of Dobel and some other languages of the Moluccas is glide fortition, a development that has been misunderstood by some historical linguists, who have appealed to it to propose Proto-Austronesian reconstructions that are clearly misconceived.

The Proto—Malayo-Polynesian Multiplicative Ligature *ŋa: A Reply to Reid

In seeking a plausible morphological source for the "intrusive velar nasal" of Palauan, Reid has argued that the multiplicative ligature *ŋa posited by Blust is invalid, and that the shape of this morpheme in Proto—Malayo-Polynesian vowel-final numerals from '20' to '90' was *=m, from earlier *na. However, evidence extending from the northern Philippines to eastern Polynesia supports *ŋa, and shows that Reid has confused two distinct morphemes, the attributive ligature *na, and the multiplicative ligature *ŋa. His further attempts to account for the nonetymological velar nasal of Palauan from an independent ligature *ŋ that ultimately reflects *na are unconvincing because they appeal to ad hoc sound changes, because they are largely a prioristic statements of possibility rather than controlled efforts to track evidence in a single language, and because Palauan has a regular reflex of *na in the form of the ligature əl , which closely matches the functions of Tagalog na and the similar attributive ligature in other Philippine-type languages.

Marking-Reversals and Their Relevance for Prehistory: An Eastern Indonesian Case

In languages of central Flores (eastern Indonesia), domestic goats are labeled with reflexes of the Proto-Malayo-Polynesian (PMP) term for 'deer' ( Cervus timorensis ). At the same time, deer are named with the same lexeme plus a modifier, being designated in effect as 'forest goats'. This development is best explained as the product of a marking-reversal. Like most large mammals, both goats and deer were brought to Flores by humans, yet the present naming of goats with reflexes of a PMP term for deer indicates that deer were introduced before goats. While some zooarchaeological evidence suggests deer were first brought to Flores by Europeans in the sixteenth century, other evidence indicates the presence of goats on the neighboring island of Timor for a very much longer period. In this way, particulars of Flores animal nomenclature inform a discussion of the faunal prehistory of the region.

A Proto-Oceanic Passive? Evidence from Bola and Natügu

Though morphological passives are said to be rare in Oceanic languages, we report on such passives in Bola (Western Oceanic), spoken in Papua New Guinea, and Natügu (Temotu Oceanic), spoken in the Solomon Islands. These languages have passive morphemes that are cognate with Proto–Malayo-Polynesian (PMP) *-in- both in form and function, which leads us to conclude that a morphological passive should be reconstructed for Proto-Oceanic (POC).

The History of Faunal Terms in Austronesian Languages

This paper offers an overview of reconstructed faunal terms primarily at the Proto-Austronesian, Proto--Malayo-Polynesian, and Proto--Western MalayoPolynesian levels, with some additional reconstructions at lower levels in the Austronesian family tree. The basic division is into domesticated vs. nondomesticated animals, and within the latter category there is a further breakdown into mammals (22 terms), birds (29 terms), reptiles and amphibians (14 terms), creepy-crawly creatures (68 terms), fish (97 terms), and marine invertebrates (41 terms), for a total of 277 etyma, most of which represent protolanguages spoken no later than about 4000 B. P. Among the more important new insights reported here is the discovery that the PMP word for 'bird' (* manu-manuk) almost certainly was derived from the word for 'chicken' (* manuk) by reduplication, a development that is fundamentally at odds with the life-form encoding sequence proposed by Cecil H. Brown (Language and living things: Uniformities in folk classificati on and naming, Rutgers University Press, 1984). INTRODUCTION The concept of semantic fields or domains has a fairly long history in linguistics, particularly in connection with reconstruction. Saussure (1959) credits the Celticist Adolphe Pictet with pioneering the concept of linguistic paleontology in his book, Les origines Indo-Europeennes (1859-1863), a work that makes implicit use of semantic domains. Much the same tradition in Indo European linguistics is seen in the more recent work of Benveniste (1973). More recently, Pawley and Ross (1994) have applied the label to subsets of the lexicon that exhibit semantic coherence either as a synchronic system of reference or in terms of historical change, or both. This paper is concerned with the history of faunal terms in Austronesian (AN) languages. Because faunal terminologies form a substantial portion of the lexicon, the discussions that follow will be relatively terse. The basic division is into domesticated vs. nondomesticated animals and in each of these divisions reconstructions will be made primarily at the Proto-Austronesian (PAN), Proto--Malayo-Polynesian (PMP), and Proto--Westem Malayo-Polynesian (PWMP) levels, although some additional reconstructions are made for Proto-Philippines (PPH), Proto--Central--Eastern Malayo-Polynesian (PCEMP), and Proto-Oceanic (POc). Western Malayo-Polynesian may not constitute a subgroup distinct from Malayo-Polynesian, and where a cognate set is widespread in Western Malayo-Polynesian (WMP) languages it may in fact be evidence for a PMP etymon. Reconstructions are drawn from Blust (in progress), and more recent research. A number of reconstructions for fauna in Dempwolff (1934-1938) have had to be discarded either because th e distributional evidence suggests that they were relatively late innovations in western Indonesia (*badak 'rhinoceros', *buru[UNKEYBORDABLE SYMBOL] 'bird', *tikus 'rat', *tupay 'squirrel' = 'tree shrew'), or more rarely because the etymology itself is defective (*ba[n]uy 'eagle'). As just noted, most of the data to which attention is drawn here are from non-Oceanic languages. There are several reasons for this choice. First, Proto-Oceanic had few mammal terms in relation to earlier AN protolanguages that were spoken west of the Wallace Line, where a far richer mammalian fauna exists than is true of either eastern Indonesia or the islands of the Pacific. Second, two domains of Oceanic faunal terms have already been covered with considerable thoroughness: birds (Clark 1994), and marine invertebrates (Pawley 1996). Moreover, Geraghty (1994) has done a virtually exhaustive job of documenting fish names in Proto--Central Pacific. This leaves some room for comments on fish names in Proto-Oceanic, and I have taken advantage of this opportunity in a few cases. I have not attempted to be thoroughgoing in filling this gap, however, because it presumably will be filled in the series of volumes initiated by Ross, Pawley, and Osmond (1998). …

Linguistic Evidence for Primogeniture and Ranking in Proto-Oceanic Society

Linguistic Evidence for Primogeniture and Ranking in Proto-Oceanic Society1 Per Hage The presence of a seniority distinction in Proto-Oceanic sibling terminology has been used as evidence for primogeniture and ranking in Proto-Oceanic society. Cross-culturally, however, sibling terminologies with the seniority distinction are very common, found in egalitarian as well as ranked societies. A stronger argument can be made. If Proto-Oceanic society was based on primogeniture and ranking, the term for elder sibling should be marked in Oceanic terminologies. There is both diachronic and synchronic evidence in support of this hypothesis. Kirch (1997) suggests that one explanation for the rapid and extensive Lapita (Oceanic) expansion was a ranking system based on a rule of primogeniture in which "first-born sons inherited rights to the ancestral house and its estate. In such societies junior siblings frequently adopt a strategy of seeking new lands to settle where they can found their own 'house' and lineage, assuring their own offspring access to quality resources" (Kirch 1997:65). This scenario is consistent with Bell-wood's (1995) more general theory that a "founder-focused ideology" was the motive for the earlier Malayo-Polynesian expansion. As evidence for a ranking system, Kirch cites the presence of a seniority distinction in Proto-Malayo- Polynesian (PMP) sibling terminology, in the contrast set *kaka 'elder sibling' versus *Sua(n)ji 'younger sibling', and the continuation of this distinction in a great many Oceanic (OC) terminologies. Cross-culturally, however, sibling terminologies with the seniority distinction are very common, found in egalitarian as well as ranked societies (Murdock 1968). A stronger argument is possible. Although very little research has been done on markedness in kinship terminologies, it appears that seniority, like sex, is not marked on a universal basis.2 If [End Page 366] Proto-Oceanic and Proto-Malayo-Polynesian society was based on primogeniture and ranking, the term for elder sibling, which designates successor and heir, should be marked in OC and MP terminologies. There is both diachronic and synchronic evidence in support of this hypothesis. As shown in table 1, PMP, POC, and Proto-Polynesian (PPN) sibling terminologies have the same four categories with terms for elder parallel (same sex) sibling, younger parallel sibling, man's sister, and woman's brother (Blust 1980; Pawley 1981; Marck 1996). PPN terms are reflexes of POC terms, and POC terms are derived from PMP terms. Click for larger view View full resolution TABLE 1. SIBLING TERMS IN PROTO-MALAYO-POLYNESIAN (PMP), PROTO-OCEANIC (POC), AND PROTO-POLYNESIAN (PPN)* In the evolution of PN sibling terms there is, as Clark (1975) has shown, a pattern of semantic simplification in which the seniority distinction is lost first, then the sex distinction, and, finally, the parity distinction ending with a single term for sibling. For cognate terms there is "a fairly consistent pattern of semantic survival where one term extends to a larger category and ousts another. Thus *tahina replaces *tuakana; *tuanga'ane replaces *tuafafine and *tahina replaces *tuanga'ane ..." (Clark 1975:86). Clark suggested, although he did not pursue the matter, that an explanation for this pattern "might be sought in terms of semantic markedness" (1975:86). Markedness or marking refers to a hierarchical relation between two terms of an opposition. The concept applies to all levels of language-phonological, grammatical, and lexical-semantic. Three criteria of marking, from Greenberg (1966, 1990 [1980]), are applicable to the analysis in this paper. 1. Par excellence expression. The unmarked term may represent the entire category or the opposite of the marked term. For example, in English (and in many other languages) man = 'human being' and man = 'male human being' (unmarked) versus woman = 'female human being' (marked). [End Page 367] 2. Contextual neutralization. In certain contexts only the unmarked term appears. For example, in German phonology voice is neutralized in the final position for obstruents. 3. Syncretization. When two sets of categories intersect, distinctions that are present in the unmarked category are absent or neutralized in the marked category. In the English pronominal system, a gender distinction in the singular (he, she, it) is neutralized in the plural (they). If merger...

In Defense of Dempwolff: Austronesian Diphthongs Once Again

In the six decades since Otto Dempwolff published the third and final volume of his foundation-laying Vergleichende Lautlehre des austronesischen Wortschatzes (1934-38), objections have been raised against various features of his Uraustronesisch, a reconstructed language equivalent to what is now called ProtoMalayo-Polynesian (PMP). Some of these objections appear in the early work of Isidore Dyen, who corrected fundamental problems in Dempwolff's reconstruction of the laryngeals and advocated a more practical orthography. During the I970s and I98os, John Wolff proposed that scholars jettison certain of Dempwolff's consonant distinctions, most notably Proto-Austronesian (PAN) *r, *d, *c, *z, *g, and *T (summarized in Wolff I988). Despite these proposed modifications of the consonant system, until now the vowels and diphthongs have remained essentially beyond reproach.' Clynes (1997:347), however, has recently argued against the reconstruction of diphthongs in the phoneme inventory of Proto-Austronesian: They have an extremely restricted distribution, being posited to have occurred only in absolute morpheme-final position-in for example PMP *lakaw 'walk, go', *balay 'public building', PAN *baliw 'moiety', *babuy 'pig'.... The putative diphthongs are moreover unusual: they differ in structure both from the generally accepted notion of a diphthong, and also from other interpretations, such as that of the U.S. structuralist tradition.... Indeed if they do exist, they constitute a complex segment not to my knowledge reported elsewhere in the phonological literature.2

The Madurese Reflexes of Proto-Malayopolynesian

1. Although Madurese' has retained some archaic features of proto-Malayopolynesian (PMP) such as the distinction between dental and alveolar stops, it has not seriously been used for the reconstruction of either the PMP phonemic system or of individual PMP morphemes. The most comprehensive work on PMP phonology, that of Dempwolff,2 mentions Madurese once (2.123); Dempwolff's comment is: 'Nirgends stdsst man auf Laute, die Anlass bieten, die Lautsystem des UIN3 zu erginzen oder zu verbessern.' Madurese is also mentioned in Dyen's work on subgrouping the MP languages.4 Madurese here shows a higher percentage with Malay (by about seven percent on a 196 word Swadesh list) than would be expected if it were a coordinate member of a group containing Javanese and Sundanese. The latter has been the standard view.' This is not unlikely since Madurese has plainly borrowed heavily from Malay. On the other hand, it has also borrowed many words from Javanese. It is necessary to keep this in mind in considering the Madurese reflexes. In this paper I would like to examine the Madurese reflexes and the evidence presented by Madurese for both the reconstruction of the PMP phonemic system and of some individual morphemes. All reconstructed forms are from Dempwolff (volume 3) except those later changed by Dyen. In some cases it will be noted that Madurese presents no evidence for a problematic part of PMP. In other cases, Madurese will be used to decide the correct reconstruction where Dempwolff or Dyen has an ambiguous form. 1 Madurese is a Malayopolynesian language spoken by about six million people in the eastern part of Java and the adjacent island of Madura. Research was carried out in Indonesia from November 1960 to March 1962 with the aid of a Ford Foundation Area Training Fellowship. Madurese forms in this paper are taken from this research and from H. Kiliaan, Madoereesch-Nederlandsch Woordenboek (Leiden, 1904). 2 0. Dempwolff, 'Vergleichende Lautlehre des austronesischen Wortschatzes', Zeitschrift fuir EingeborenenSprachen (Berlin): 1. 'Induktiver Aufbau einer indonesischen Ursprache', 15. Beiheft (1934), 2. 'Deduktive Anwendung des Urindonesischen auf austronesische Einzelsprachen', 17. Beiheft (1937), 3. 'Austronesisches Worterverzeichnis', 19. Beiheft (1938). For a list of the PMP phonemes see Dempwolff 3.9 (with d' omitted and Z later added by I. Dyen, 'ProtoMalayopolynesian *Z', Language 27.534-40 [1951]). I have used certain symbols which differ from Dempwolff's. In the following list, my symbol precedes and Dempwolff's follows the comma: c, k'; e, a; j, g'; n, A; q, h; r, 1; R, y; s, t'; w, v; y, j; z, d'; D, d; T, t. Dempwolff's 'is not transcribed at all. Dempwolff also has a brief account of the Madurese reflexes of PMP *r, *1, *d, *D, *R, and *j in 'Die 1-, r-, und d-Laute in austronesischen Sprachen', ZfES 15.278-80 (1925). In a number of places his conclusions are incorrect. 3 Urindonesisch (proto-Indonesian) which Dempwolff later identifies, as far as the sound system is concerned, with Uraustronesisch (proto-Austronesian) here called proto-Malayopolynesian. 4 I. Dyen, 'A lexicostatistical classification of the Austronesian languages' 48, IJAL Memoir 19 (Baltimore, 1965). 5 See S. J. Esser, Atlas van Tropisch Nederland map 9 (The Hague, 1938) and R. Salzner, Sprachenatlas des indopazifischen Raumes 1.11 (Wiesbaden, 1960).