Current Anthropology Volume 51, Number 2, April 2010
223
Farming and Language in
Island Southeast Asia
Reframing Austronesian History
by Mark Donohue and Tim Denham
Current portrayals of Island Southeast Asia (ISEA) over the past 5,000 years are dominated by
discussion of the Austronesian “farming/language dispersal,” with associated linguistic replacement,
genetic clines, Neolithic “packages,” and social transformations. The alternative framework that we
present improves our understanding of the nature of the Austronesian language dispersal from Taiwan
and better accords with the population genetics, archaeological evidence, and crop domestication
histories for ISEA. Genetic studies do not demonstrate that the dispersal of Austronesian languages
through ISEA was associated with large-scale displacement, replacement, or absorption of preexisting
populations. Linguistic phylogenies for Austronesian languages do not support staged movement
from Taiwan through the Philippines into Indo-Malaysia; in addition, the lexical and grammatical
structure of many Austronesian languages suggests significant interaction with pre-Austronesian
languages and cultures of the region. Archaeological evidence, including domestication histories for
major food plants, indicates that ISEA was a zone of considerable maritime interaction before the
appearance of Austronesian languages. Material culture dispersed through ISEA from multiple sources
along a mosaic of regional networks. The archaeological evidence helps us to shape a new interpretative
framework of the social and historical processes that more parsimoniously accounts for apparent
discrepancies between genetic phylogenies and linguistic distributions and allows for more nuanced
models of the dispersal of technologies and societies without reference to the farming/language
dispersal hypothesis.
The Current View of Austronesian
Dispersal across Island Southeast Asia
Current portrayals of Island Southeast Asia (ISEA) over the
past 5,000 years are dominated by variants of the Austronesian
dispersal hypothesis (e.g., Bellwood 1984–1985, 1997, 2005;
Bellwood, Fox, and Tryon 1995; Blust 1995; Diamond 2001;
Diamond and Bellwood 2003; Shutler and Marck 1975). The
Austronesian dispersal is cited as an archetypal example of a
purportedly global phenomenon, the farming/language dispersal (Bellwood 2005; Bellwood and Renfrew 2002; Diamond
and Bellwood 2003). Other panregional examples include the
spread of Indo-European-language speakers across Eurasia
Mark Donohue is Research Fellow in the Department of Linguistics
at the Research School of Pacific and Asian Studies of Australian
National University (Canberra, Australian Capital Territory 0200,
Australia [[email protected]]). Tim Denham is Monash Research
Fellow at the School of Geography and Environmental Science of
Monash University (Melbourne, Victoria 3800, Australia). This paper
was submitted 24 VII 08 and accepted 12 XII 08.
(Ammerman and Cavalli-Sforza 1984; Renfrew 1987) and that
of Bantu-language speakers across sub-Saharan Africa (Ehret
1998). In each case, an inclusive historical interpretation of
demic expansion, or the movement of farming peoples together with their languages and cultures, accounts for and
links together present-day language distributions and human
population genetics and the transformation and diffusion of
material culture through time. Although the consonance of
languages, genes, and cultures is increasingly qualified (Bellwood 2005, 2007), general correspondences among multidisciplinary data remain a central tenet of the farming/language
dispersal hypothesis and underpin its explanatory power.
According to conventional wisdom, Austronesian languages, along with their speakers and associated technologies,
dispersed from Taiwan into the Philippines and Indo-Malaysia
and, after largely bypassing New Guinea, moved into the Pacific. The dispersal of these peoples is presumed to have been
accompanied and enabled by agricultural technologies, primarily based on rice but including other plant domesticates
and a suite of domesticated animals (pigs, dogs, and chickens),
and other “Neolithic” cultural traits, including pottery and
䉷 2010 by The Wenner-Gren Foundation for Anthropological Research. All rights reserved. 0011-3204/2010/5102-0004$10.00. DOI: 10.1086/650991
224
polished stone adzes (Bellwood 1997, 2005; Diamond 2001;
Diamond and Bellwood 2003). Rice-based agriculture and
other technologies are considered to have enabled the Austronesians to colonize and, to varying degrees, replace preexisting hunter-gatherer populations occupying ISEA (Bellwood 1997, 2005; Diamond 2001). Linguistic reconstructions
have been matched to archaeological chronologies to variously
argue for staged and sequential migration from Taiwan to the
Philippines beginning ca. 4,500–4,000 years ago, to eastern
Indonesia by ca. 4,000–3,500 years ago, across the north of
New Guinea to the Bismarck Archipelago (but largely excluding mainland New Guinea) by ca. 3,500–3,300 years ago,
and to the Solomons and Vanuatu by ca. 3,100–3,000 years
ago (Bedford 2006; Bellwood 2005; Blust 1995; Spriggs 2003,
2007). After this expansion, Austronesian speakers colonized
uninhabited islands over a vast area, including all of Micronesia and Polynesia in the Pacific (Kirch 2000), areas that
show no clear evidence of prior human settlement.
Detractors of the Austronesian hypothesis of regional prehistory as it applies to ISEA have speculated on alternatives
but have been unable to address all aspects of the evidence
usually cited to support that hypothesis or to provide a viable
interpretation that sufficiently accounts for the multidisciplinary lines of evidence (e.g., Meacham 1984–1985; Oppenheimer 2004; Oppenheimer and Richards 2001b; Solheim
1984–1985; Szabó and O’Connor 2004; Terrell 2004; Terrell,
Hunt, and Gosden 1997; Terrell, Kelly, and Rainbird 2001).
Generally, the standard view of Austronesian dispersal is taken
as writ; researchers across a range of fields continue to presuppose this historical account in their interpretations of data
from ISEA and the Pacific (e.g., Lansing et al. 2007).
In this paper, we examine multiple lines of historical evidence—including genetics, linguistics, archaeology, and plant
domestication histories—separately. The noncongruence of
the multidisciplinary lines of evidence offers a strong critique
of the farming/language dispersal hypothesis and similar
models of Neolithic spread (Spriggs 2003, 2007) as they apply
to the past five millennia of ISEA history. We find that the
evidence, not only in aggregate but also when examined discipline by discipline, points to a mid- to late-Holocene history
for ISEA that is very different from that portrayed by the
conventional Austronesian dispersal hypothesis. These findings are of global importance because Austronesian dispersal
has been promoted as an archetype of the farming/language
dispersal hypothesis; the lack of fit of the farming/language
dispersal hypothesis to ISEA severely undermines its relevance
as a global model to account for major language groupings
and the diffusion of agriculture.
In the latter part of the paper a new historical framework
is proposed to account for the observed multidisciplinary
phenomena in ISEA during the Holocene. This alternative
historical interpretation is a radical departure from conventional portrayals. It emphasizes the mosaic of regional networks and the social processes prevalent in ISEA before and
Current Anthropology Volume 51, Number 2, April 2010
during the spread of Austronesian languages and cultural influences from Taiwan.
Human Population Genetics
Bellwood (2002:17) asserts that “early farmers . . . frequently
colonized outwards from homeland regions . . . in the process
spreading foundation trails of material culture, language, and
genetic distinctiveness.” To what extent is there a distinctive
genetic trail among the current human populations of ISEA
indicating that an Austronesian, or “out-of-Taiwan,” colonization event occurred? This is not a simple question, since
different genetic markers carry their own tales, and there are
even questions about the extent to which we might expect a
genetic trail, or cline, to persist (e.g., Excoffier and Ray 2008).
Care must be taken in the evaluation of genetic studies,
particularly to ensure that more than one marker is used, that
the historical implications of different genetic markers are
properly understood (Hurles 2002), and that sampling coverage of regional populations is sufficient as a basis for the
conclusions (e.g., consider the limited representation of samples from ISEA in Friedlaender et al. 2008). Under the Austronesian dispersal hypothesis, which invokes the migration
of colonizing farmers tracing their origins to Taiwan, the human genetics of ISEA populations might be anticipated
to reflect a Taiwanese genetic inheritance, potentially as a
distance-decay effect, or cline, away from that island.
Although some of the genetic variation among human populations in ISEA can be attributed to Taiwanese influence, the
proportion does not by any means represent the wholesale
replacement or absorption of preexisting populations (Oppenheimer 2004; Oppenheimer and Richards 2002). Maximally 20% of Y chromosome variation (Capelli et al. 2001)
and at most 20% of the mitochondrial DNA (mtDNA) variation (Hill et al. 2007) in ISEA populations can be explained
by an out-of-Taiwan hypothesis. At the same time, approximately the same number of genetic markers in these same
ISEA populations, representing most of the diversity found,
are best explained as having been present in situ since the
colonization of the area by modern humans during the Late
Pleistocene (Soares et al. 2008). Hill et al. (2007:38–39) consider it likely, on the one hand, that “about a fifth of the
modern inhabitants can trace their maternal ancestry back to
the first anatomically modern settlers of ISEA” and, on the
other, that “if a mid-Holocene migration did occur, it was—
on the maternal side at least—demographically minor, contributing, at most, only a fifth of the modern ISEA mtDNAs.”
Furthermore, the degree of genetic inheritance does not
exhibit a clear pattern from Taiwan across ISEA and beyond.
For example, a paternal genetic signal does persist from (eastern) Taiwan to Polynesia (e.g., Kayser et al. 2008), the easternmost limit of Austronesian migration, showing that at least
some genetic markers were not lost in migratory “surfing”
(following Excoffier and Ray 2008). However, this same genetic signal is not prominent in ISEA.
Donohue and Denham Farming and Language in Island Southeast Asia
Even more problematic for hypotheses involving a largescale human migration is that Taiwan represents an endpoint,
rather than a source, of numerous genetic phylogenies (Soares
et al. 2008). The evidence from several genetic markers suggests that the net flow of genes has been from south to north
(e.g., Cavalli-Sforza et al. 1988; Hill and Serjeantson 1989;
Oppenheimer 2004; Su et al. 1999), again perhaps reflecting
more ancient population dispersals of modern Homo sapiens
than those assumed under most accounts of Austronesian
dispersal. It is likely that some of the approximately 20% of
Y chromosome or mtDNA variation will turn out to have
spread before the mid-Holocene, reducing their values as
markers of an Austronesian dispersal.
Furthermore, Austronesian language–speaking populations
tend to mimic, genetically, any non-Austronesian populations
with which they are in contact, rather than presenting starkly
contrasting genetic histories. This is true of all areas in which
there are both Austronesian-speaking and non-Austronesianspeaking populations, including Madagascar, Southeast Asia,
and Melanesia (Chow et al. 2005; Hurles et al. 2005; Li et al.
2008). This implies that the genetic history of Austronesian
speakers, in whichever region they are found, implicates the
preexisting populations of these regions as much as any putative migration out of Taiwan. The arrival of Taiwanese genes,
rather than replacing earlier populations, was associated with
gradual blending and gene flow between the populations.
In sum, multiple studies demonstrate that the Taiwanese
contribution to the genetics of ISEA populations, both Austronesian and non-Austronesian, is relatively minor. There is
no genetic evidence of a mass migration of farmers out of
Taiwan who replaced or absorbed populations across ISEA.
The majority of genetic markers, both paternal and maternal,
in ISEA populations reflect Pleistocene colonization movements by modern humans and “dispersals across the region
of Austronesia throughout the early to mid Holocene” (Hill
et al. 2007:40). Conversely, events from the mid-Holocene
onward play a relatively minor role in the genetic history of
most of ISEA. The failure of most genetic studies, especially
those focused on paternally inherited markers (Friedlaender
et al. 2008; Kayser et al. 2008), to differentiate between the
speakers of what are distinct linguistic groupings indicates
that “biological genetic features have geographically based
rather than linguistically based distributions” (Nichols and
Peterson 1998:612). There is at best a weak correlation between language and genes in ISEA (pace Cavalli-Sforza et al.
1988 but in accordance with Sokal 1988 and numerous other
studies). There is no “genetic distinctiveness” in ISEA that
suggests that a mid-Holocene out-of-Taiwan population
movement is especially privileged in determining the genetic
makeup of the present populations in the area. Rather, we
find a complex range of signals, with no single origin dominating the genetic history of ISEA.
225
Reassessing Austronesian Linguistics
Bellwood (1984–1985:109) states, “The question of Austronesian origins is basically a linguistic question,” and while the
question is no longer exclusively a linguistic one, “Austronesian” is still essentially a linguistic construct. The Austronesian family comprises more than 1,000 languages spread
over a vast area between Madagascar and Easter Island. Overwhelming linguistic evidence shows an origin for the Austronesian languages on Taiwan (Blust 1995), and we do not
dispute this. On the other hand, we question the nature of
the linguistic “dispersal” out of Taiwan and into ISEA. We
should note that the “Austronesian dispersal” might better be
termed a “Malayo-Polynesian dispersal,” since nine of the 10
primary subgroups of Austronesian are attested to only on
Taiwan and only the Malayo-Polynesian branch has members
outside Taiwan (and none on mainland Taiwan). Therefore,
we hereafter refer to “Malayo-Polynesian” rather than “Austronesian” where the former is more appropriate.
The standard version of the Austronesian linguistic phylogeny is very hierarchical, with bifurcations corresponding
to inferred movements from Taiwan (the Proto-Austronesian
[PAN] homeland, where nine of the 10 first-order subgroups
are found) through ISEA (the various languages designated
as Western Malayo-Polynesian, including groups that have
since moved to the Southeast Asian mainland and Madagascar) and eastern Indonesia (Central Malayo-Polynesian),
across northern New Guinea (the South Halmahera–West
New Guinea branch of Eastern Malayo-Polynesian), and finally into the Pacific (Oceanic, including Polynesian and Micronesian; e.g., Blust 1995; Tryon 1995; see fig. 1A).
In recent years, this Austronesian phylogeny has been
shown to be flatter at multiple levels (fig. 1B). Linguistic
subgrouping and the consequent construction of a layered
hierarchy rely on the sharing of innovations in the inherited
linguistic signal to define phylogenetic subgroups. For instance, the Malayo-Polynesian subgroup, comprising all of
the Austronesian languages spoken outside Taiwan, can be
defined as a valid subgroup on the basis of a number of shared
innovations, both regular and irregular (see table 1, abbreviating material in Blust 2001). However, it is recognized that
the same is not true of Western Malayo-Polynesian, in which
languages show the Malayo-Polynesian innovations but nothing unique relative to Malayo-Polynesian languages to the
east, namely, those assigned to Central-Eastern Malayo-Polynesian (e.g., Ross 1995). The Central-Eastern Malayo-Polynesian branch contains the Central Malayo-Polynesian and
Eastern Malayo-Polynesian subgroups and comprises the Austronesian languages of eastern Indonesia and northwestern
New Guinea, as well as those of Oceania. Problematically, the
evidence for the Central Malayo-Polynesian and Central-Eastern Malayo-Polynesian subgroupings is not conclusive, since
many of the innovations that have been proposed for each
of these subgroups (e.g., Blust 1993) are present in languages
in the Western Malayo-Polynesian area and, in some cases,
226
Current Anthropology Volume 51, Number 2, April 2010
Figure 1. Representations of the Austronesian phylogeny. A, Earlier phylogeny (from Pawley 2007, after Blust 1995); B, revised phylogeny (following Donohue and Grimes 2008 and Ross, Pawley, and Osmond 2008).
Earlier representations of the Austronesian family (A) have a very hierarchical structure, but recent revisions (B) show a flatter structure with
multiple starbursts.
even as far north as in Taiwan (Donohue and Grimes 2008).
As a result, while we can group the “extra-Formosan” languages together as Malayo-Polynesian against those groups
that did not migrate from Taiwan, we cannot justify any large
subgroupings that would link the languages of the Philippines
and western Indonesia together, as opposed to the languages
spoken near and east of New Guinea. This fact represents a
major challenge to computational models that claim success
in replicating large subgroups within this nonexistent clade
(e.g., Gray and Jordan 2000) and weakens their conclusion
that linguistic evidence supports the so-called express-train
model of a rapid Austronesian dispersal.
Further, there are very few large interisland subgroups in
the Malayo-Polynesian area west of New Guinea; most subgroups are confined to single islands or adjacent islands (fig.
2A, based on Adelaar 2005a and Ross 1995). Notable exceptions are Malagasy, which migrated relatively recently from
southeastern Borneo, and Malayo-Sumbawan, which includes
Malay and the languages of several islands east of Java (Adelaar 2005b). Since there is no known structured phylogeny
among the Western Malayo-Polynesian languages, we cannot
say that the northern Malayo-Polynesian groups represent
higher branches on the tree and that the southern groups are
farther (phylogenetically) from the source. On the contrary,
Donohue and Denham Farming and Language in Island Southeast Asia
227
Table 1. Innovations defining Malayo-Polynesian (selection only)
Proto-Austronesian
-mu “2pl genitive”
∗
panudaN “pandanus”
∗
S
∗
Siwa “nine”
∗
-eS
∗
C ( ∗t
∗
N ( ∗n
Proto-Malayo-Polynesian
-mu “2sg genitive”
pandan “pandanus”
∗
h
∗
siwa “nine”
∗
-ah
∗
C and ∗t merge as ∗t
∗
N and ∗n merge as ∗n
∗
there is no evidence suggesting that, for instance, Nias in the
southwest is any closer to or, importantly, any farther from,
cladistically, Proto-Malayo-Polynesian (PMP) than are any of
the languages of the Philippines. Consequently, the impression
of a graduated dispersal of Malayo-Polynesian languages south
from Taiwan does not hold up and must be rethought. If we
restrict our inquiry to the linguistic evidence alone, there was
a rapid, multidirectional, and multimodal propagation of
Malayo-Polynesian languages across most of ISEA, from southern Indonesia to the Batanes Islands in the north, without
direction or hierarchy (see fig. 2B; Ross 2005). Can we ascertain
a center for the dispersal? On purely linguistic grounds, the
answer must be no. It is true that the presence of nine of the
primary subgroups of Austronesian in Taiwan unambiguously
indicates that Taiwan is the homeland for the language family
and that, ultimately, the Malayo-Polynesian languages are the
southern branch of the higher Austronesian node, but this says
nothing about the internal relationships, dispersals, and later
history of Malayo-Polynesian. From the presence and location
of the proposed Central-Eastern Malayo-Polynesian subgroup,
which is seen as the eastern split of Malayo-Polynesian, as well
as the presence of more proposed subgroups in the south than
in the north, we can deduce that the center of dispersal was
in central or eastern Indonesia, using the standard historical
linguistic principle of “least moves” (e.g., Anttila 1972; Joseph
and Janda 2003). Indeed, on linguistic grounds, Blust (1995)
considers a location not far from the Sulu Sea, on the Philippines/eastern Indonesia border area, to be the logical homeland for Malayo-Polynesian, but in the same paper he later
disregards this conclusion on the basis of arguments from archaeological data. Since all the evidence suggests that the Philippines has seen significant leveling and loss of earlier diversity
in the past two millennia (Blust 1991, 2000a, 2005), we might
logically move the “center of gravity” farther north, but there
is no surviving linguistic evidence with which to test this
hypothesis.
These conclusions follow a reexamination of the data based
on the existing, “standard” view of the Austronesian phylogeny. If we follow the revisions proposed by Donohue and
Grimes (2008), which eliminate the Central-Eastern MalayoPolynesian and Central Malayo-Polynesian nodes, we are left
with a “center” that is even farther to the southeast. To this
we should add the fact that the evidence for the Eastern
Malayo-Polynesian subgroup, the western boundaries of
Comments
Irregular development
Irregular development
Regular change
Irregular development
∗
e and ∗a are otherwise distinct
Regular merger
Regular merger
which are to the west of New Guinea, is tenuous at best.
Adelaar (2005a:26) discusses the evidence for the claim, noting that “there is essentially no phonological evidence to support the claim that together they form an exclusive higherlevel subgroup.” Furthermore, Ross (1995:84) notes that the
“putative lexical innovations” that have been advanced “may
be exclusively shared inherited features,” indicating that their
origin might derive not from local innovation but from the
retention of older forms that have been lost elsewhere. If true,
this would eliminate the use of these lexical innovations to
define an Eastern Malayo-Polynesian subgroup. Eventually,
we might eliminate this node; the resulting phylogenetic tree
would have no hierarchical structure between PMP and
Proto-Oceanic. If this is the case, then the range of possible
dispersal centers for Malayo-Polynesian, based solely on linguistic evidence, would extend as far east as the Pacific. Consequently, it is clear that there is no linguistic evidence for an
orderly north-to-south dispersal of Malayo-Polynesian languages from Taiwan.
The linguistic evidence for Malayo-Polynesian presents us
with additional methodological challenges. The lexical conservatism of the family is remarkable (Blust 2000b). Outside
a Melanesian area, both west and east of New Guinea, where
various kinds of “aberrancy” are prominent (Pawley 2006),
the languages retain a very high proportion of the reconstructed vocabulary of PMP (fig. 3). The overt similarities
between languages are so striking that relationships between
far-flung members of the family were recognized 300 years
ago (Reland 1708), long before the Indo-European languages
were seen as being related. When we compare Austronesian
with other language families, it is apparent that the amount
of lexical change found in Austronesian is consistent with
either a much younger or a much smaller language family
(Joseph and Janda 2003; Peiros 2000; Wichmann, forthcoming). Smaller language families tend to be more compact geographically and show less change because of continued contact
between the different members of the family. Young language
families, on the other hand, have in the main not had the
time required to spread and diversify. The Austronesian family, however, is neither small nor seemingly recent, and these
discrepancies must be addressed.
Wichmann (forthcoming) offers a metric for comparing
internal diversity by evaluating the degree of lexical diversification in different language families, thereby removing sub-
Figure 2. The distribution of Austronesian languages of ISEA. A, Subgroupings within Malayo-Polynesian (MP) languages from figure 1A. B,
Major subgroups of MP, from figure 1B. The subgroup of Austronesian
languages south of Taiwan (i.e., MP) shows 30–40 different primary
subgroups in A, with the only claimed subgroupings being found in
eastern Indonesia and East Timor (Central MP and Eastern MP) and in
the Pacific (Oceanic group). Non-Austronesian languages within the area
of the Austronesian dispersal are shown in yellow; their substratal effects
are obvious in many southern Austronesian languages (Capell 1975; Donohue 2007b; see fig. 5). There is no Austronesian presence anywhere in
Australia, and mainland Asia has only peripheral and recent Austronesian-speaking populations.
Donohue and Denham Farming and Language in Island Southeast Asia
229
Figure 3. Proportion of Proto-Malayo-Polynesian vocabulary retained in
individual languages. Austronesian languages show remarkable conservatism in their lexicons, except near New Guinea (especially eastern New
Guinea), where the languages display neither the grammatical structures
(fig. 5) nor, overwhelmingly, the lexicon of their more conservative relatives in, for example, western Indonesia or the Philippines. Only in the
area enclosed by the dashed line is the average retention rate less than
30%; the area enclosed by a solid line has a less than 20% average retention
rate.
jective judgements from assessments of “conservatism.” The
mean lexical cognacy found between modern languages in a
family can be determined and then evaluated in terms of the
“minimum centuries” (mc) that would be expected to have
elapsed to result in this level of diversification, assuming that
2% of the “basic” lexicon of a language will change per century, a value that is taken as standard (e.g., Swadesh 1950,
1952, 1955; for Austronesian, Dyen 1965, though see Blust
2000b).1 When this metric is applied to families for which
documentation is adequate (Wichmann’s results are summarized in table 2), large language families, those with more
than 100 member languages, show an average mc value of
93.5, implying that we expect the initial divergence to have
taken place ca. 9,000 years ago (a clearly inflated date, but
we are interested in relative values, not absolute ones). The
only exception in this group is Austronesian, with an mc value
of 35. Regardless of the faith we place in glottochronological
methods, this low figure indicates that Austronesian shows
the profile of a family with fewer than 50 languages, such as
1. Note that this method does not ascribe a glottochronological age
to a family but merely evaluates the internal diversity of the family in
terms of the lexicon.
Iroquoian (10 languages), Na-Dene (47), Plateau Penutian
(4), Mixe-Zoquean (16), or Káriban (29), rather than that of
a family with more than 1,000 languages. From the degree of
retention of common vocabulary, we can state that Austronesian, at least the Malayo-Polynesian branch that has migrated beyond Taiwan, does not exhibit the characteristics
expected of a large, ancient language family.
The lexical conservatism of Austronesian languages, when
examined in detail, does not uniformly extend to reconstructions of PAN. There is no evidence of a Taiwanese origin for
much of the basic vocabulary that is cited as characteristic of
the Austronesian dispersal, such as terms for maritime technology and many of the food plants that have been associated
with the Austronesians (cf. Blust 1984–1985; Pawley 2007).
Rather, the etymologies for the majority of the plants and
maritime technologies that are found in the Pacific can be
found in PMP, the generic ISEA entity, and not in PAN, the
Taiwan-anchored, reconstructed protolanguage. The linguistic
evidence suggests that, having left Taiwan, the Austronesians
were exposed to an influx of new material-culture items that
were already present in ISEA and were adopted into the expanding Malayo-Polynesian culture(s). This explains why
230
Current Anthropology Volume 51, Number 2, April 2010
Table 2. Measures of lexical diversity in families (from
Wichmann, forthcoming)
Language family
Yeniseian
Xincan
Plateau Penutian
Caddoan
South Caucasian (Kartvelian)
Chukotka-Kamchatka
Moseten-Chon
Barbakóan
Iroquoian
Totonakan
Eskimo-Aleut
Mixe-Zoquean
Salishan
Hokan
Káriban
Khoisan
Hmong-Mien (Miao-Yao)
North Caucasian
Uralic
Algic
Quechuan
Na-Dene
Uto-Aztekan
Altaic
Mayan
Tai-Kadai (Daic)
Tupı́an
Dravidian
Otomanguean
Nilo-Saharan
Australian
Sino-Tibetan
Afro-Asiatic
Indo-European
Trans-New Guinea
Austronesian
Niger-Congo
No.
languages
Minimum
centuries (mc)
2
4
4
5
5
5
5
7
10
11
11
16
27
28
29
29
32
34
38
40
46
47
62
65
69
70
70
75
172
199
258
365
372
443
552
1,262
1,489
5
10
42
33
40
40
51
33
35
26
30
36
45
88
37
111
40
60
60
30
15
35
48
77
42
30
55
40
60
150
95
60
113
70
100
35
100
Note. The correlation between number of languages and internal lexical
diversity, measured in mc, is significant (r p 0.35), but it increases to
r p 0.50 if Austronesian is omitted.
these terms are reconstructible to PMP and not to PAN (Pawley 2007). Figure 3 shows the percentage of reconstructed
PMP vocabulary that is retained in a number of modern
languages (from Greenhill, Blust, and Gray 2008). The percentage of retained vocabulary is calculated by comparing the
standard word list for PMP with the same word list for the
modern languages; word lists and calculations of retained
vocabulary follow Blust (1981). The original vocabulary does
not decay with distance from Taiwan; rather, it is clear that
proximity to New Guinea and its non-Austronesian societies
is the best predictor of vocabulary loss. The map shows the
areas in which the average retention rate is below 30% and
below 20%, with both clustering around New Guinea and the
islands close to (Papuan-dominated) Bougainville.
In contrast to lexical conservatism, we find a “plethora of
different systems” of grammar in different modern Austronesian languages outside Taiwan (Ross 1995:64; recent quantifications in Wichmann and Kamholz 2008). No grammatical
traits are common to all Austronesian languages (figs. 4, 5);
further, grammatical interrelationships between Austronesian
language groups follow geographical rather than phylogenetic-subgroup boundaries (Donohue 2007b). A remarkable
diversity characterizes the Malayo-Polynesian languages at all
levels of their grammar, including phonology, morphology,
and syntax; this diversity almost certainly reflects early contacts with pre-Austronesian languages (Adelaar 1995; Capell
1975; Donohue 2004, 2005, 2007b). These contacts show their
strongest effects south of the Philippines. Figure 6 shows the
locations of different isoglosses defined by contact-induced
change in Austronesian languages. Note that, despite the clear
boundaries that emerge in this examination of structural similarities, there are no major phylogenetic breaks between the
northern languages and the southern ones or between the
eastern languages and the western ones. The early recognition
of more “Papuan” characteristics in the east of the Indonesian
archipelago was stated clearly in the nineteenth century (Brandes 1884:187):
One can distinguish a western and an eastern division in
the Malayo-Polynesian languages in the archipelago. The
border between the two is a line drawn between Sawu and
Rote, Flores and the Solor islands, east of Buton, west of
the Sula islands, east of Minahasa, the Sangir and Talaud
islands, and (east of) the Philippines.2
Brandes’s division, which closely approximates the heavy
dashed line in figure 2A, is based on the then-current understanding of the relative order of the possessor and possessum in a construction such as “the man’s house.” In this
phrase, “man” represents the possessor (or “Genitive”), and
“house” the possessum (or “N,” in the literature). This English
example shows a GenN order, the same order that dominates
in Melanesia and surrounding “eastern” areas in Indonesia
and East Timor. West of the line, an NGen order is found,
as in the Indonesian translation of the same phrase, shown
in figure 7. Figure 6 marks the distribution of this typological
feature with the green line and shows additional features that
evidence the Papuan influence on Austronesian languages in
the approach to New Guinea and (less obvious in this map)
the Austroasiatic influence on the southwestern languages, as
well as the existence of a significant north-south divide in the
typology of the languages that cannot be explained phylogenetically.
Contact-induced changes such as these, resulting in the geo2. In the original, “Bij de Maleisch-Polynesische talen in den Archipel
kan men eene Westelijke en eene Oostelijke afdeeling onderscheiden. De
grens tusschen beiden is een lijn, getrokken tusschen Sawu en Rotti, Flores
en de Solor-eilanden, beoosten Buton, bewesten de Sulla-eilanden, beoosten de Minahasa, de Sangi-, de Talaur-eilanden, en de Philippijnen.”
Donohue and Denham Farming and Language in Island Southeast Asia
231
Figure 4. Locations of languages listed in figure 5. 1, Tagalog (Schachter
and Otanes 1972); 2, Indonesian (authors’ knowledge); 3, Tukang Besi
(Donohue 1999); 4, Palu’e (authors’ knowledge); 5, Ambai (Silzer 1983;
D. S. Price, T. R. Price, P. J. Silzer, C. Silzer, and N. Merasi, unpublished
manuscript); 6, Tobati (Donohue 2002); 7, Manam (Lichtenberk 1984);
8, Mekeo (Jones 1998); 9, Maori (Krupa 1973).
graphic (as opposed to phylogenetic) clustering of grammatical
properties, as well as the presence of numerous linguistic isolates within the Austronesian area (Donohue 2007a, 2007b;
Pawley 2002), are often associated with early interactions between two populations that came into contact. One linguistic
entity has vanished, in the conventional historical sense, but
has left traces in the changes the surviving language has acquired
(e.g., Nichols 1997). The data we can see in figure 6 suggest
that the Malayo-Polynesian spread in ISEA was a social overlay
involving cultural interaction with preexisting populations
rather than the simple imposition of a new linguistic phylogeny
across space through population replacement (Dutton 1995;
Ross 1994; Thomason and Kaufman 1988).
With varying degrees of linguistic shift, resulting in Austronesian conservatism in some parts of the grammar and
some parts of the ISEA area and more innovative traits
(as viewed from a Taiwanese perspective) in others, a new
Malayo-Polynesian lexicon was in many areas grafted onto a
disparate collection of at best distantly related language groups
that predated the arrival of Malayo-Polynesian languages
across the region. In many regions, especially in the south
and east of ISEA and in the Pacific, the Austronesian language
family does not display the characteristics of a phylogenetic
entity with the time depth, or historical profile, that has generally been ascribed to it. In addition to the significant break
between PAN and PMP, a significant number of culturally
significant words predate the Malayo-Polynesian dispersal
(Mahdi 1994a, 1994b), suggesting that significant pre-
Austronesian language elements form substrates across wide
regions in areas now dominated by Malayo-Polynesian languages, primarily near New Guinea (fig. 1) but also including
Borneo, the Malay Peninsula, and Sumatera in the west (Adelaar 1995). The acquisition of Malayo-Polynesian languages
by Negritos in the Philippines may provide a model for the
adoption of Malayo-Polynesian languages across much of
ISEA (e.g., Reid 1994).
Language families, like bacteria or bdelloid rotifers (Arkhipova and Meselson 2000), can form because of the convergence
of originally unrelated taxa (e.g., Dagan and Martin 2007). As
people abandon their original language and adopt a new one,
they preserve some aspects of their original linguistic code as
a substrate, creating innovations in different aspects of the
grammar of their new language (Bakker 2000; Capell 1975;
Dutton 1995; Nettle 2000; Ross 1994; Thurston 1987). During
the initial expansion of Malayo-Polynesian, linguistic affiliation
derived as much from recruitment of speakers of other languages as it did from “normal” linguistic diachrony involving
organic changes and splits. For those languages that owe their
Malayo-Polynesian affiliation to shift rather than to generic
linguistic evolution, we can identify numerous substratal traces
of their earlier linguistic affiliations in their phonologies, such
as the presence of implosive stops or gaps for /g/. Such a process
might be likened to the development of regional varieties of a
creole, although we do not suggest that this can account for
all of the divergence in Austronesian languages, even among
the so-called aberrant languages (Pawley 2006). However, the
232
Current Anthropology Volume 51, Number 2, April 2010
Figure 5. Malayo-Polynesian linguistic diversity (grammatical) and similarity (lexical). Nine Austronesian languages spoken between the Philippines and Polynesia show significantly different grammatical structures
for the same sentence (see sources in fig. 4). The verb, subject, and object
reconstruct to Proto-Malayo-Polynesian: ∗inum, “drink”; ∗(ba-)b!in1ahi,
“woman”; and ∗wahiR, “water.” Noncognates are shown in italics (the
Tobati and Manam forms reflect ∗danum, “water,” an alternative MalayoPolynesian reconstruction). At the same time, the grammatical structures
in which these Austronesian words are embedded differ greatly. The verb
“drink” in each example is shown in black; red indicates the word for
“woman” and blue that for “water.” The grammatical morphemes case
and agreement are shown in green and violet, respectively, with other
grammatical morphemes in brown. The right-hand column compares
the order of verb, subject, and object (V, S, and O, respectively), an
important typological variable (e.g., Dryer 1992). Dedicated morphological voice systems are found in Tagalog, Indonesian, and Maori, with
other voice systems being present in Tukang Besi, Palu’e, and Ambai,
while no such diathesis is reported for Tobati, Manam, and Mekeo. ISEA:
Island Southeast Asia; NG: New Guinea.
great numbers of typologically disparate Austronesian languages
are consistent with the mechanisms of language shift and abnormal transmission.
Archaeology of the “Austronesians”
We have seen that the genetic and linguistic phylogenies for
ISEA are demonstrably discordant. There is insufficient evidence to suggest large-scale replacement, punctuated or otherwise, of preexisting hunting and gathering populations by
farming-voyaging immigrants from Taiwan or anywhere else.
Although a clearly defined cline may not be anticipated for
human dispersal across an archipelago, as opposed to across
a continental land mass, multiple molecular markers provide
highly variable evidence of the Taiwanese contribution to the
genetic makeup of the present-day inhabitants of ISEA. There
is no genetic evidence for large-scale population replacement,
displacement, or absorption such as that originally postulated
for hunter-gatherers by farmers across ISEA (after Bellwood
1997); rather, the evidence is more suggestive of prolonged
interaction and mixing among populations across ISEA. Furthermore, there is no clear directionality in the frequency and
nature of molecular markers through ISEA away from Taiwan.
The same region, however, has witnessed massive language
replacement, effected by the dispersal of Malayo-Polynesian
languages originating on Taiwan seemingly within the past
few thousand years. Archaeological data can potentially shed
light on the processes that gave rise to the genetic and linguistic discordance across ISEA, and they will be considered
with specific reference to the farming/language dispersal hypothesis proposed for the region (Bellwood 1997, 2005; Diamond and Bellwood 2003).
An initial caveat in considerations of the archaeology of
ISEA during the mid-to late Holocene concerns radiometric
dating of sites and derived materials. Despite efforts at chronometric hygiene, namely, a method to restrict historical interpretation to those sites and artifacts associated with more
Donohue and Denham Farming and Language in Island Southeast Asia
233
Figure 6. Substratal effects in Austronesian languages. The numbered
lines show the southern limits of prenominal relative clauses (1), prenominal adjectives (2), prenominal demonstratives (3), and preverbal
subjects (4). Exceptions to limit 4 are discussed in Donohue (2005,
2007b). Colored lines show the western limits of prenominal possessors
(green), postnominal numerals (blue), contrastive inalienable possession
(red), dedicated morphological voice systems (yellow), and verbal agreement (brown; also found in much of Sumatera). Sources: Dryer (2005a,
2005b, 2005c, 2005d, 2005e, 2005f ) and Nichols and Bickel (2005); discussion in Donohue (2007b). The lines mark the limits of the features
concerned, in both Austronesian and non-Austronesian languages. The
appearance of Papuan languages just at the area delineated by these
different features and the presence both of Austroasiatic languages close
to Sumatera and of ample substratal and toponymic evidence for a prior
Austroasiatic presence in Sumatera strongly suggest that, away from the
areas of strong empire building over the past two millennia, significant
numbers of pre-Austronesian linguistic elements survived, even though
the languages spoken are now (lexically) members of the Austronesian
family.
reliable dates (Spriggs 1989), the chronologies of ISEA archaeology are severely limited. Even the most systematic attempts to develop robust chronologies for the occurrence and
dispersal of material-culture items, such as pottery (Spriggs
2003, 2007), are plagued by chronological uncertainties. These
include a high reliance on material types that can be highly
problematic for radiocarbon dating, including bone, marine
shell, and artifact residues; evidence of disturbance and mixing; and selective arguments for the rejection or retention of
dating evidence (Spriggs 2007). The resultant chronologies
for the dispersal of material-culture items across ISEA are
problematic and fragile; indeed, they are bedeviled by many
of the same uncertainties that have discredited claims for preLapita pig and pottery at sites along the north coast of New
Guinea (Spriggs 1996).
There is extremely limited evidence supporting the idea
that East Asian farming practices from Taiwan diffused across
ISEA approximately 4,000–3,500 years ago, that is, during the
initial period of purported Austronesian expansion. Very little
suitably aged and taxonomically specific archaeobotanical evidence exists for the domesticated plants that are hypothesized
to have dispersed with farming from Taiwan, especially rice
(Oryza sativa) and foxtail millet (Setaria italica), or for associated farming technologies and landscape transformations.
There is an almost complete lack of domesticated rice dating to 4,000–3,000 years ago in archaeological contexts associated with food processing and consumption across the
whole region, except as inclusions in pottery:
All in all, in Island Southeast Asia, there are 23 sites where
archaeobotanical sampling has been applied. Only one, Gua
Sireh, has reported rice phytoliths from the matrix of the
234
Current Anthropology Volume 51, Number 2, April 2010
Figure 7. Expression of the genitive (i.e., order of elements in possessive
phrases) in English and six representative languages of Indonesia. In each
case “man” is shown in red and “house” in black. The English phrase
codes the possessor before the noun (N) and so shows a GenN order,
the same order that dominates in Melanesia and surrounding “eastern”
areas. West of Brandes’s line, an NGen order prevails, as in the Indonesian
translation of the phrase rumah orang, literally “house-man.” Note that
in addition to the overwhelming majority of Papuan languages (such as
the unrelated Lani, Skou, and Kanum), eastern Austronesian languages,
such as Palu’e, also show the preposed genitive. In addition to the order
of the words, different languages also employ other coding strategies,
such as marking a noun as being the possessor (green; English ’s, Tukang
Besi nu, Kanum -ne) or marking features of the possessor on the possessed
noun (Palu’e -n, Skou -kéke), approximately signaling “his.”
site. There are at least 36 sites with reported rice inclusions
in pottery fabrics; one in Luzon, one in Negros island, one
in Sabah (Bukit Tengkorak) and the rest in Sarawak. (Paz
2002:279)
The situation has changed little since Paz’s review, except for
recently noted rice remains at Ulu Leang on southern Sulawesi, which may date to ca. 4000 cal BP (Paz 2005:112, 113).
The basis for this age has not been fully reported, but it may
help resolve chronological uncertainties regarding other putatively early rice remains at this site (Glover and Higham
1996:436–437; Paz 2002:277, 2005:112). At present, the two
most reliable and best-dated sites for early rice in ISEA are a
charred rice grain inclusion in a pottery sherd at Gua Sireh,
western Sarawak, dated to before 4000 cal BP (Paz 2002:277),
and organic materials in association with pottery at Andarayan, Luzon, that likely date to ca. 3700–3500 cal BP (based
on information in Paz 2002:277).
Rice cultivation was well established on Taiwan by at least
4,000 years ago (Bellwood 2005), and numerous words for
the plant and associated processing and storage practices have
been reconstructed to PAN (Blust 1984–1985; Pawley 2007).
Rice cultivation cannot, however, be heavily implicated in the
early phases of Malayo-Polynesian language dispersal as conventionally portrayed. Only two sites in ISEA provide potentially early evidence for rice, aside from inclusions in pottery,
and only one of these, Gua Sireh, is reliably dated. On the
contrary, rice seems more clearly associated with pottery than
with agriculture, implying that many of its archaeological occurrences could be derived more from extralocal trade than
from local food production (Paz 2002). Significantly, the age
of the Gua Sireh finds and the site’s location on Borneo
suggest that domesticated rice may have initially entered ISEA
from mainland Asia, before its putative dispersal from Taiwan.
Furthermore, and arguably, rice became an important staple
crop across ISEA only after the advent of open-field farming
there, which occurred in the past 2,000–1,500 years (Anshari,
Kershaw, and van der Kaars 2001; see Barton and Denham,
forthcoming) and potentially much later on some islands.
Similarly, there is a near absence of relevant archaeobotanical evidence for the dispersal of other crop plants likely to
have been domesticated on the Asian mainland. An exception
is a reported grain of foxtail millet (S. italica), noted as postdating 3000 cal BP, from Timor (Ian Glover’s research in
Bellwood 1997:231). Even taking into account the low level
of confidence for this identification (Paz 2002:279), this cereal
grain postdates the putative period of early Austronesian expansion into this region; it certainly does not provide supporting evidence for the claimed introduction of an agricultural economy to Timor approximately 4,000 years ago
(Bellwood 1997:231).
Evidence for the dispersal of agricultural technology with
early Malayo-Polynesian-speaking voyagers is similarly weak.
There are examples of farming implements in ISEA with Taiwanese or Chinese affiliations, such as the waisted hoe at
Donohue and Denham Farming and Language in Island Southeast Asia
Torongan cave and other stone hoe fragments from the Batanes Islands (Bellwood and Dizon 2005:8, 11). However, the
ages of these artifacts are not clearly determined; as reported
for the Batanes materials, some are likely to be much more
recent on the basis of inferred associations with Iron Age sites
on Taiwan (Bellwood and Dizon 2005:11).
Furthermore, there is no clear paleoecological signal of the
environmental transformations that might be anticipated to
accompany the dispersal of agricultural populations through
ISEA between 4000 and 3500 cal BP. Such signals have been
discerned for the emergence and transformation of agriculture
in the highlands of New Guinea (Haberle 2003). Across much
of ISEA, large-scale anthropic disturbance of vegetation is more
recent (e.g., Dam et al. 2001; Suparan et al. 2001; van der Kaars
et al. 2001) or much earlier, including from ca. 6500 cal BP at
Niah on Borneo (Hunt and Rushworth 2005) and from at least
7000 cal BP on Sumatera (Flenley 1988). The interpretation of
the human contribution to paleoecological data over the past
5,000 years, however, is complicated by the intensification of
El Niño–Southern Oscillation (ENSO) events during this period
(e.g., Anshari et al. 2004). Increased ENSO variability, rather
than the expansion of agricultural economies, has been cited
as the primary cause of increased burning, as well as of the
maintenance of the diversity of tropical rain forests, in Indonesia and New Guinea over the past 5,000 years (Haberle, Hope,
and van der Kaars 2001).
Given that there is no solid evidence that either farmers or
farming spread across ISEA in toto from Taiwan approximately 4,000–3,500 years ago, there is similarly no good reason to assume that the history of the dispersal and adoption
of Malayo-Polynesian languages between islands, along coastlines, and later penetrating inland corresponds to the spread
and adoption of an associated “cultural package” of new technologies (Denham 2004). Over the past 30 years, archaeologists and linguists have repeatedly revised down, or “unpacked,” the size of the material-culture package that
purportedly spread from Taiwan with the dispersal of MalayoPolynesian languages (compare claims in Pawley and Green
1973 with those in Bellwood 2005 or Pawley 2007). Recent
incarnations of the Austronesian-dispersal orthodoxy restrict
the Taiwanese material-culture package to pottery, polished
stone adzes, shell artifacts, tattooing chisels, and domesticated
animals (pigs, dogs, and chickens; Bellwood 2005). Importantly, none of these material-culture types are necessary or
unique markers of farming or, perhaps more accurately, horticulture in ISEA. From the emerging archaeological record,
however, even this smaller set of remaining elements is not
a secure and exclusive archaeological marker of Taiwanese
influence or Malayo-Polynesian language dispersal (Anderson
and O’Connor 2008).
The earliest records of pottery across ISEA are suggestive
of multiple incursions into the region by at least two different
styles and traditions, one from mainland Southeast Asia and
at least one from Taiwan. Current portrayals suggest that a
red-slipped pottery tradition originating on Taiwan dispersed
235
south of the Philippines between ca. 4,000 and 3,600 years
ago and then spread eastward to eastern Indonesia (Bellwood
2005; Shutler 1999; Solheim 1964; Spriggs 1989, 2003, 2007).
An earlier “cord or basketry-wrapped paddle impressed” pottery tradition appears to have spread from mainland Southeast
Asia to Borneo, and potentially other islands, before Taiwanese influences are discerned (see Spriggs 2007). Sparse archaeological evidence and chronological uncertainties suggest
that ISEA was a region into which multiple pottery traditions
were introduced at different times, although the archaeological record does not yet provide sufficiently high-resolution
geographical and temporal information to map the spread
and transformation of each pottery tradition through time.
In the absence of historical detail, however, red-slipped pottery is still presumed to be a reliable and verifiable chronological marker that provides a likely maximum age for the
influence of Malayo-Polynesian speakers in the region
(Spriggs 2007) as well as the precursor of the Lapita pottery
tradition in Melanesia from ca. 3,500 years ago (Kirch 2000).
In terms of other claimed components of a Taiwanese cultural package, a range of shell artifact types from Wallacea
and Melanesia, including fishhooks, shell beads, and shell
adzes—and presumably the technologies required for their
manufacture—predate the period of presumed Malayo-Polynesian expansion into those regions (O’Connor 2007;
O’Connor and Veth 2005; Swadling et al. 1989; Szabó 2004;
Szabó, Brumm, and Bellwood 2007). Of particular significance are Tridacne shell adzes, together with edge-ground
stone tools and a fully ground triangular stone “chisel,” from
contexts dated to at least 7200–5000 cal BP at the Pamwak
rock shelter on Manus Island in Melanesia (Fredericksen,
Spriggs, and Ambrose 1993; Golson 2005). Recent genetic
characterizations of pig (Sus spp.) populations indicate a
spread from mainland Southeast Asia, a region that has been
proposed as a center of pig domestication (Hongo et al. 2002),
to eastern ISEA before and independent of the arrival of
Malayo-Polynesian speakers (Dobney, Cucchi, and Larson
2008; Larson et al. 2007). Similar trajectories of eastward diffusion were followed at different times by at least one pottery
tradition (Spriggs 2003) and, potentially, by rice (see above)
and the chicken from their center of domestication in mainland Southeast Asia (Dobney, Cucchi, and Larson 2008, after
Liu, Zhu, and Yao 2006).
On the basis of the archaeological evidence, there was no
“cultural package” of technologies that originated on Taiwan
and was exclusively—or can necessarily be—associated with
the dispersal of Malayo-Polynesian languages in ISEA. Most
elements that have been presumed to have diffused with
Malayo-Polynesian speakers were already present in ISEA before the appearance of uniquely Taiwanese influences in the
archaeological record of that region. Certainly, some cultural
items can be sourced to Taiwan, such as nephrite (jade) artifacts, and they are a reliable marker of that island’s incorporation into trade networks to the Philippines from possibly
ca. 4000 cal BP, with a much later and wider distribution
236
through ISEA from ca. 2500 to 1500 cal BP (Hung et al.
2007). However, different material-culture items and traditions have singular histories of emergence, dispersal, and
transformation that do not conflate into a unitary, or even
semiporous, cultural package. Consequently, there is no historical imperative to invoke a material-culture package, or
Neolithic cultural complex, that was exclusively associated
with the dispersal of Austronesian languages from Taiwan.
ISEA was already integrated into exchange networks with
neighboring regions before Taiwanese influences, which have
been presumed to mark Austronesian language dispersal, become apparent in the archaeological record. Indeed, there is
no reason to presume that early Taiwanese influences in the
archaeological record of ISEA are necessarily associated with
Malayo-Polynesian language dispersal; they may simply represent the incorporation of Taiwan, like parts of mainland
Southeast Asia and Melanesia, into a mosaic of exchange networks that stretched across ISEA. The trajectories of MalayoPolynesian language dispersal may have a separate temporality. Rather than representing the introduction of a new
cultural package, the arrival of Malayo-Polynesian speakers
in ISEA was more likely to have been associated with, either
suddenly or gradually, a new social context that revolutionized
the ways in which the mosaic of interaction across the archipelagos operated. Through time, preexisting exchange networks were intensified and expanded, bringing about, in turn,
an increasing standardization of traded items and the semblance of a cultural package.
Crop Domestication Histories
In addition to rice (discussed above), the history of Austronesian language–speaking peoples has been intimately associated with a range of plant domesticates. In a seminal paper,
Pawley and Green (1973) considered a whole suite of domesticated crop plants to be reconstructible to PAN, including
bananas, breadfruit, coconut, rice, sago palm, taro, and yam.
However, more recently (Blust 1984–1985; Pawley 2007), the
list reconstructed for a Taiwanese PAN has been revised down
to include only rice, sugarcane, and giant taro (Alocasia indica), reflecting an increasingly sophisticated understanding
of the Austronesian phylogeny. These linguistic revisions mirror those unfolding from genetic research into the histories
of domesticated crop plants.
Evaluating the data from studies of plant genetics suggests
that several staple crops underwent initial, or separate, domestication in the New Guinea or eastern Indonesian region,
including bananas (Musa spp.; Carreel et al. 2002), sugarcane
(Saccharum officinarum; Grivet et al. 2004), taro (Colocasia
esculenta; Lebot et al. 2004), greater yam (Dioscorea alata;
Malapa et al. 2005), and sago (Metroxylon sagu; Kjær et al.
2004). The early and mid-Holocene processes of use and domestication for these plants remain largely unknown (Denham 2005; Denham et al. 2003; Fullagar et al. 2006), but their
histories of domestication and diffusion can be reconstructed
Current Anthropology Volume 51, Number 2, April 2010
in broad terms from a combination of plant genetics, the
limited archaeobotanical evidence available to us, and historical linguistics (Kennedy and Clarke 2004).
The unusual genetics of bananas—whereby chloroplast DNA
(cpDNA) and mtDNA are inherited maternally and paternally,
respectively—enable the lineages and contributions of subspecies and species to be easily tracked for banana cultivar groups,
shedding light on the history of their domestication and spread
(Carreel et al. 2002; De Langhe and de Maret 1999; Kennedy
2008). For example, several different groups of bananas, including Pacific plantains (Lebot 1999), Western and Central
African plantains (AAB), and an East African AAA cultivar
(Carreel et al. 2002), trace part of their ancestry to the New
Guinean Musa acuminata ssp. banksii. Significantly, the two
groups of African bananas show different temporal-geographic
domestication pathways (Carreel et al. 2002), with the diffusion
of plantains to West Africa documented to at least 2,500 years
ago (Mbida et al. 2001) and that of a banana to East Africa to
ca. 5,000 years ago (Lejju, Robertshaw, and Taylor 2006). Although the archaeobotanical evidence for the identification of
bananas in Africa is contested by some (Vansina 2003; cf. Mbida
et al. 2004), these findings clearly demonstrate a pre-MalayoPolynesian time depth for the westward dispersal into ISEA of
bananas from the New Guinea region (Denham and Donohue
2009; Kennedy 2008).
A similar scenario, initial domestication in New Guinea
with subsequent diffusion and multiple interspecific hybridizations on mainland Asia, has been invoked for sugarcane
(Grivet et al. 2004). A term for sugarcane, ∗CebuS, reconstructs to PAN (Blust 1984–1985; Pawley 2007), and, like rice,
the crop was seemingly on Taiwan before the languages dispersed ca. 4,000 years ago. If the linguistic reconstructions
are valid and sufficiently specific chronologically, then we
must suppose that sugarcane also moved westward from its
home in New Guinea before the Malayo-Polynesian language
dispersal into ISEA. This can have been possible only if some
pre-Malayo-Polynesian seafarers brought the sugarcane north
and west to Taiwan, that is, if the dispersal of food crops
across ISEA was not reliant on Austronesian speakers.
In sum, the initial stages of banana and sugarcane domestication, with subsequent diffusion westward, occurred before
the appearance of Malayo-Polynesian languages in ISEA.
Other relevant food plants of the region reflect varying historical processes despite the absence of great amounts of archaeobotanical information. The genetic characterizations of
taro suggest separate domestication events in mainland Southeast Asia and Melanesia, with only limited gene flow between
the two regions until recently (Lebot et al. 2004), and so do
not provide any evidence for a dominating spread from either
the west or the east (and certainly none for a spread from
the north). By contrast, the greater yam is thought to have
undergone domestication in the New Guinea region, with
subsequent widespread anthropic diffusion of cloned cultivars
through ISEA, mainland Asia, and Africa (Malapa et al. 2005).
Sago palm (M. sagu) is similarly thought to have originated
Donohue and Denham Farming and Language in Island Southeast Asia
and been domesticated in Melanesia, with a subsequent westward spread (Kjær et al. 2004).
Importantly, all these staples underwent initial domestication in New Guinea. As with different types of material
culture, the variability in the geography of subsequent diffusion indicates that they did not spread together, along the
same paths, or at the same time. Specifically, these domesticates did not spread westward from New Guinea across ISEA
as part of a single agricultural or cultural package. The successful onward movement of domesticates was dependant on
social groups along the relevant exchange networks having
some knowledge of the plant-specific husbandry, cultivation,
and processing techniques. As plants such as bananas and
sugarcane were exchanged and moved across ISEA, they were
transformed through intra- and interspecific hybridization
and through incorporation into cultivation practices that were
widespread, albeit variable, across the region before MalayoPolynesian language dispersal.
Disconnecting Genes, Language, and
Material Culture
A recurrent failing of successive portrayals of Austronesian
dispersal or origins has been the presumption that the histories of genetics, language, and material culture should to
any large degree correlate and follow the same trajectory (as
discussed in Terrell 2000b). The consonance of these lines of
evidence for ISEA over the past 4,000 years is, at best, unclear;
in some cases, there are clear disparities (i.e., between genetic
and linguistic evidence) and asynchronies (i.e., for different
types of material culture and many domesticated plants). Even
the value of red-slipped pottery as a marker of Austronesian
language dispersal across ISEA is, on the basis of current
evidence (Spriggs 2007), a poorly corroborated claim. Despite
several attempts, the isochrones of pottery dispersal are yet
to be fixed across ISEA. Even after the chronology of diffusion
is determined, red-slipped pottery may mark only the incorporation of Taiwanese influences into the maritime exchange
networks of ISEA rather than a large-scale demic, farming,
language, or Neolithic dispersal.
In contrast to theories of Austronesian farming/language
dispersal, as well as critiques of (Oppenheimer and Richards
2001b) and commentaries on (Anderson and O’Connor 2008)
them, we suggest that there was not a single homeland, a
single migratory route, a single cultural package, or a single
mode of language transmission that spread through ISEA.
Equally, complexity cannot be ascribed to two homelands,
two migratory routes, or two cultural packages. A radically
different type of hypothesis must be considered to account
for the observed phenomena.
The linguistic and archaeological records, as well as plant
domestication histories, are suggestive of considerable complexity in parts of ISEA before and during the dispersal of
Malayo-Polynesian languages from Taiwan; the directionality
and chronology of diffusions are unclear or contradictory.
237
Although general regions of origin are largely known, most
types of artifact, animal, plant, and technology demonstrate
distinctive histories and geographies of diffusion. Consequently, diffusion, which refers to the cumulative movement
of ideas and things, must be assumed to have been multidirectional across space, and this diffusion yielded net eastward and westward movements just as much as net southward
ones (fig. 8). An interpretation other than one that seeks to
identify the homeland is now needed to understand the social
and historical processes that account for the widespread adoption of an intrusive language family by, and the variable appearance of material-culture elements among, the widespread
and diverse cultures that must have existed in ISEA before
the appearance of the Malayo-Polynesian-speaking peoples.
Maritime Interaction in ISEA
during the Holocene
A long history of maritime interaction can now be reconstructed for ISEA and adjoining regions, including the Asian
mainland and Melanesia. Patterns and continuity of interaction similar to those now emerging for ISEA during the
early to mid-Holocene are well attested to for the variable
interisland movement of people, marsupials, plants, obsidian,
stemmed tools, and the enigmatic “mortar, pestle, and figurine complex” in the circum–New Guinea region during the
Pleistocene and Holocene (e.g., Araho, Torrence, and White
2002; Summerhayes 2003; Summerhayes and Allen 1993;
Swadling and Hide 2005; Torrence and Swadling 2008; White
2004). Given the absence of wholesale population movements
(as suggested by the lack of any genetic evidence for replacement), the diffusion of ideas and things must have occurred
through socially mediated exchange networks over predominantly short (terrestrial and maritime) and less frequently
long maritime distances (after Hughes 1977; Irwin 2008).
These exchange networks were probably not continuously operational in either time or space, but they certainly existed in
varying forms before, during, and after Austronesian language
dispersal from Taiwan (Solheim 1984–1985).
Across ISEA, different peoples had been exchanging, adopting, and transforming various plants, animals, and technologies since the early Holocene, including (1) localized translocations of the Sulawesi warty pig (Sus celebensis) to Flores
(Dobney, Cucchi, and Larson 2008) and of a marsupial (Phalanger orientalis) from New Guinea to East Timor (White
2004); (2) widespread diffusion of the domesticated pig (Sus
scrofa) from mainland Southeast Asia to eastern Indonesia
(Dobney, Cucchi, and Larson 2008); (3) widespread diffusion
of bananas and sugarcane from the New Guinea region westward to mainland Asia (Denham and Donohue 2009; Kennedy 2008); and (4) potential diffusion of pottery from mainland Southeast Asia to Borneo (Spriggs 2003, 2007). Some
movements were idiosyncratic, others more general. Some
things, such as taro and some yams (e.g., Dioscorea bulbifera;
Lebot 1999), did not move, presumably because they were
238
Current Anthropology Volume 51, Number 2, April 2010
Figure 8. Places of origin for selected cultural items entering exchange
networks in Island Southeast Asia during the mid-Holocene. The eastward
and westward movements from mainland Southeast Asia and Melanesia
occurred before the southward movement from Taiwan.
relatively ubiquitous resources, were subject to varying local
forms of management and domestication, and did not make
highly prized trade items.
Two related topics are keys to understanding how MalayoPolynesian speakers and some elements of their distinctive
material culture became incorporated into preexisting networks of maritime-facilitated exchange in ISEA: the origins
of voyaging technology and the rapid appearance of redslipped pottery across the northeast fringe of the region between 4,000 and 3,600 years ago.
There is nothing uniquely Taiwanese about the outrigger
canoe, the technological artifact currently most famously associated with Austronesian dispersal. Linguistically, the term
does not reconstruct to PAN (Blust 1984–1985; Pawley 2007),
and the vessel has long been presumed to have originated in
Wallacea or, potentially, Melanesia (e.g., Horridge 1995).
There were maritime movements of material culture, plants,
animals, and technologies to, within, and from ISEA and
Melanesia before Austronesian language expansion from Taiwan, and it is not unreasonable to assume that the canoes,
or their precursors, were used in this exchange. Maritime
exchanges also occurred between Taiwan, China, and nearby
islands from potentially 4,500 years ago, thought to have been
facilitated by bamboo rafts (Rolett, Chen, and Sinton 2000).
The rapid spread of red-slipped pottery, a manufacturing
style considered to have spread through ISEA southward from
Taiwan, clearly indicates the incorporation of Taiwan into
some ISEA exchange networks from ca. 4,000 years ago
(Spriggs 2003, 2007). These preexisting exchange networks
indirectly linked vast areas and may well have initially remained under the control of non-Austronesian peoples after
contacts between ISEA and Taiwan. Without a significant cultural disruption, the existing networks would have enabled
the widespread distribution of any new prized item, such as
a new pottery style.
Through time, perhaps millennia, more and more parts of
these previous maritime networks came to be increasingly
dominated by Malayo-Polynesian languages, probably initially
in scattered coastal enclaves. Just as different technologies
were variously adopted and adapted in ISEA, so too did different peoples in different locales take up an intruder language, to different degrees, as the larger societies in which
they participated were increasingly dominated by MalayoPolynesian speakers.
On the basis of DNA evidence and sexual roles in modern
societies, we propose that a small group (or groups) emigrated
from Taiwan and intermixed with indigenous populations,
leaving an inconsistent genetic trace. The domination that
these immigrant groups enjoyed over the preexisting populations was neither numerical nor accompanied by large-scale
population movements or replacement. It is unclear why
Malayo-Polynesian speakers came to dominate, but an “elite
dominance model” (e.g., Best 2002; Spriggs 2003), whereby
small groups of politically important men leave a greater genetic signature than their numbers would suggest, can be
invoked. Social dominance, as well as the continuation, intensification, and extension of trade networks through time,
is linked to control over new trade goods in the network,
Donohue and Denham Farming and Language in Island Southeast Asia
perhaps such as red-slipped pottery and some types of polished stone adze, as well as over the knowledge to make them.
The continuation and intensification of exchange through
maritime networks and the subsequent ubiquity of items give
the illusion of a standardized “Neolithic” cultural package at
archaeological sites, despite the disparate sources of many of
the items implicated in this “package.” After decades or centuries of increasing exchange and social interaction, numerous
items, practices, and kinds of knowledge that may have been
locally restricted within or around ISEA became more common, both across space and in their co-occurrence. Thus,
dispersed occurrences of “early” artifacts in ISEA and neighboring regions are not archaeological anomalies that require
methodological, as opposed to historical, explanation. Similarly, a Taiwanese cultural package, whether Austronesian or
Neolithic, is illusory for ISEA as a whole; it did not originate
in one place and disperse outward but formed through the
increasing co-occurrence of used and traded items by people
across ISEA, who were increasingly connected through an
expanding network of exchanges.
Social and Historical Contexts
of Exchange
Trade is not just about use or exchange value; it occurs between people and establishes or reinforces social relationships
(e.g., Mauss 1990). The arrival of people with new material
culture, language, technologies, and worldviews can have dramatic impacts on societies; under the right conditions, these
impacts can confer a high status on the new arrivals. History
is replete with examples of powerful groups dominating larger
indigenous populations, with language transfer and the variable adoption of cultural items and practices. The period of
European colonial expansion over the past 500 years is simply
the most recent and, perhaps, most visible example.
The nature of some of the social processes involved can be
illustrated with reference to a historically well-documented
example: the arrival of Europeans in highland New Guinea
in the 1930s (Connolly and Anderson 1987; Leahy 1991).
Small, disparate groups of newcomers were able to dominate
much larger indigenous populations through the manipulation of now greatly expanded trading networks and worldviews, reinforced by the use of violence (Swadling 1996). Several items of European introduction to Melanesia had
differentially spread through local exchange networks ahead
of direct contact, in some cases by centuries, including items
of high value and prestige, such as steel axes (Hughes 1977)
and a widely adopted staple crop, the sweet potato (Ipomoea
batatas; see papers in Ballard et al. 2005). Some intermixing
occurred between indigenous populations, Europeans, and
accompanying lowland Melanesians. People soon began to
learn Tok Pisin, a largely English-lexified creole language that
serves as a lingua franca in much of Papua New Guinea (Baker
and Mühlhäusler 1996; Laycock 1982; Mühlhäusler 1996),
where fewer than six million people still speak more than 800
239
languages (Lewis 2009). Through time, some local languages
have become less important and increasingly replaced by what
had previously been a secondary language. Indeed, the increasing use of this language of wider communication in
Papua New Guinea has compromised the future of many
languages, which are now endangered, prompting educational
support in community schools and academic efforts to document them.
Social change in the New Guinea highlands over the past
75 years is not directly comparable to the maritime landscape
of ISEA in the mid-Holocene, but the comparison is useful
insofar as it highlights some of the social processes that may
have occurred in ISEA during the Holocene. In both cases, a
small incoming group effected major linguistic and cultural
changes without leaving a major genetic signal. Genetically,
most populations remained in place, except for a minor overlay derived from newcomers and intermixing reflecting new
movements, social interactions, and lifestyles. For instance,
the expanded role of Malayo-Polynesian speakers in the trade
networks of ISEA may have initiated or intensified exchanges
among other social groups and may have been accompanied
by a new sociocultural milieu. During the mid- to late Holocene, new material-culture items and associated technologies diffused differentially through exchange networks and
were variably incorporated into preexisting practices. Through
time, exchanged material-culture items became more widespread and co-occurred with greater frequency, producing the
semblance of a cultural package. Also through time, the original languages of ISEA were increasingly replaced by versions
of the Malayo-Polynesian lingue franche that were variably
adopted by local populations in order to participate in a new
social world. Language replacement was effected not by population replacement or absorption but by changing social contexts and practices (Mufwene 2001).
The model of maritime interaction and social transformation in ISEA advanced here accords with the observed data
better than other hypotheses, including that of Austronesian
farming/language dispersal. Further, we have shown that the
various lines of evidence, both individually and taken together,
require us to suppose a series of viable networks connected
by maritime technology, trading a range of commodities and
ideas. Given that we must posit these technologies and innovative exchanges in ISEA before the advent of the MalayoPolynesian-speaking peoples, considerations of parsimony require us to question the need to posit a Taiwan-initiated
advance in food, material, and maritime technologies. The
continued transformation in social contexts during the late
Holocene furthered linguistic and cultural homogenization
through the maintenance and intensification of maritime exchange networks, the spread of Indic civilizations, and later
Islamization and European colonialism (e.g., Reid 1995; Supomo 1995).
The “Austronesian dispersal” cannot be considered to have
been a single expansionary process from Taiwan through the
islands to its south, let alone east to Polynesia (contra Dia-
240
mond 2001). We have proposed an alternative historical
framework that focuses on the ways that Taiwanese, mainland
Southeast Asian, and Melanesian influences, as well as those
from within ISEA, were transformed through time and across
space. The social dynamics underlying Malayo-Polynesian
language dispersal through previously inhabited regions of
ISEA and Melanesia (Green 2000) were almost certainly fundamentally different from those driving the colonization of
uninhabited islands in the Pacific.
Of broader significance, the Austronesian dispersal has been
promoted as an archetype of the farming/language dispersal
hypothesis. Our discussion demonstrates the inability of that
hypothesis to account for the distribution of genes, languages,
and material culture across ISEA over the past 4,000 years.
We have proposed an alternative, more complex hypothesis
drawing on the distinctive histories of each line of evidence
and the social contexts of exchange across ISEA. Consequently, the presumed consonance of genes, language, and
material culture, and the suitability of the farming/language
dispersal hypothesis to other historical and geographical contexts, such as the Bantu expansion in sub-Saharan Africa,
should be progressively challenged.
Acknowledgments
Thanks are due to Huw Barton, Geoff Hope, Peter Kershaw,
David Mitchell, Andrew Pawley, Victor Paz, Martin Richards,
Malcolm Ross, and Matthew Spriggs for answering queries
and to the Centre of Southeast Asian Studies at Monash University for providing an opportunity to present our research.
We also thank four Current Anthropology referees for constructive criticism of earlier drafts and Kara Rasmanis, Uri
Gilad, and Kay Dancey for their help preparing our figures.
Comments
Peter Bellwood
School of Archaeology and Anthropology, Australian National
University, Canberra, Australian Capital Territory 0200,
Australia ([email protected]). 21 V 09
This “nuanced” but misinformed paper denigrates the farming/language dispersal hypothesis by associating it with imagined concepts of “wholesale replacement,” “large-scale human
migration,” and “mass migration,” concepts not required by
the clinal models of population dispersal that I prefer (Bellwood 1997, 2005; the authors quote both books but appear
to have read neither with due attention). The genetics discussion is dominated by haploid molecular clock–based assumptions, ignoring current criticisms (Cox 2005, 2008). The
statement that only 20% of modern mtDNA and NRY (nonrecombining Y chromosome) variation reflects Austronesian
dispersal in ISEA is guesswork in the absence of ancient-DNA
Current Anthropology Volume 51, Number 2, April 2010
analysis and any reliable molecular dating method for the
Holocene (Ho et al. 2007). Austronesians do not “mimic,
genetically, any non-Austronesian populations with which
they are in contact.” There are no such populations beyond
peripheral regions of Vietnam, peninsular Malaysia, Near
Oceania (especially New Guinea), and Madagascar.
In terms of linguistics, the Malayo-Polynesian (MP) “tree”
has long been observed to be a rake (Pawley 1999), indicative
of rapid multidirectional dispersal, thus explaining why ProtoMP cannot be sourced to any specific extra-Taiwan region.
The authors offer no new insight here. Subsequent high levels
of interaction between MP-speaking populations in ISEA have
led to high levels of lexical retention. Austronesian beyond
Taiwan results from a recent, rapid, and multidirectional dispersal, mostly within the past 3,500 years, or Wichmann’s
(forthcoming) 35 minimum centuries. Why should it be expected to “exhibit the characteristics expected of a large, ancient language family,” when it is not ancient at all?
The suggestion that MP languages spread by “varying degrees of linguistic shift,” with new lexicon grafted onto preexisting non-Austronesian languages, is grossly overdone.
Why are there not far more non-Austronesian enclaves surviving in ISEA, as there are in western Island Melanesia, where
such a language shift is implied by the higher levels of lexical
diversity? “Elite dominance” simply does not work and lacks
historical analogy. The recent spread of Tok Pisin in Papua
New Guinea is due to imposition of a language in colonial
and modern circumstances in which hundreds of mutually
unintelligible languages are spoken. At least two millennia of
precolonial trade and interaction in New Guinea, famous to
anthropologists since Malinowski and before, led to no significant lingua franca formation. Were language shift to be
the true explanation, then surely all the peoples of New
Guinea would have shifted to MP languages 3,000 years ago.
The archaeological records for Taiwan and the northern
Philippines, unquoted by these authors, are rich in artifactual
(but not yet economic) detail, assisted by large numbers of
14
C dates (Bellwood and Dizon 2008; Bellwood et al., forthcoming; Hung 2008). There is no network of preceramic interaction hiding here or in the Moluccas, Borneo, or Sulawesi.
Excepting claims for Timor Leste, there is no good evidence
(and the authors offer none) that cord-marked pottery, shell
fishhooks, cut-shell beads, and shell adzes predate MalayoPolynesian arrival in ISEA, given the widespread use of old
shell for making artifacts and the propensity of these artifacts
to migrate through or be cached within cave strata (Bellwood
2009). Pigs did not spread from mainland Southeast Asia
before they spread from Taiwan into the northern Philippines
(Piper et al. 2009). The authors lack knowledge of the archaeologies of Vietnam, Borneo, and certainly Taiwan, where
the gradual development of red-slipped pottery as a dominant
style was well established between 4500 and 4000 BP, long
before such pottery appeared to the south (Hung 2008). The
Taiwan “cultural package” is certainly not illusory in the
northern Philippines (Bellwood and Dizon 2008).
Donohue and Denham Farming and Language in Island Southeast Asia
Bemoaning the “lack of domesticated rice dating to
4,000–3,000 years ago,” the authors fail to note the lack of
searching with appropriate techniques. Little phytolith or
starch research has yet been done in ISEA, and the archaeology
is dominated by caves, reluctant locales for food production.
Coastal Neolithic village sites in ISEA are likely to be buried
beneath many meters of redeposited sediment (Bellwood et
al. 2008). Lack of evidence in circumstances lacking suitable
recovery techniques cannot be proof of absence.
The Austronesian cognate vocabulary for food production
and increasing numbers of reports of rice in pottery and as
phytoliths, even when identifiable charcoal macroremains are
absent, leave little room for any nonfarming argument for
early MP-speaking colonists. Suggestions that bananas, sugarcane, greater yam, and sago might have been domesticated
in Melanesia, especially New Guinea, are not here disputed,
but the only “archaeological evidence” presented for their
early domesticated occurrence in ISEA consists of a controversial claim for banana phytoliths in a site in Uganda.
The paper finishes by denigrating the ability of the farming/
language dispersal hypothesis to account for the distribution
of genes, languages, and material culture across ISEA over the
past 4,000 years. But this hypothesis remains the best explanation for the observed situation (Bellwood 2009). The MP
language spread occurred mainly as a result of population
movement (Ross 2008), not elite dominance.
Murray P. Cox and J. Stephen Lansing
Institute of Molecular BioSciences, Massey University,
Palmerston North 4442, New Zealand ([email protected]
.nz)/Department of Anthropology, University of Arizona,
Tucson, Arizona 85721, U.S.A. ([email protected]). 28 V
09
As the influential statistician George Box once noted, “all
models are wrong, but some are useful” (Box 1979:202). This
statement is not mere semantics: models are by necessity simplifications of the real world; they can be considered useful
only if they tell us something new. While Donohue and Denham rightly warn about the dangers of entrenched views, their
own model of Indo-Pacific prehistory is neither new nor simple, nor, more importantly, is it particularly useful.
Despite claiming differences of opinion, Donohue and
Denham’s idea that “material culture [and language and
genes] dispersed through ISEA from multiple sources along
a mosaic of regional networks” is remarkably similar to earlier
themes. As far back as 1988, John Terrell advocated a related
outlook: that Indo-Pacific prehistory revolved around “interlocking, expanding, sometimes contracting and ever-changing
set of social, political, and economic subfields” (Terrell 1988:
647). Such general concepts have understandable appeal: they
capture the complexity of human society in a way that simple
models do not. Unfortunately, such generalities are not nec-
241
essarily right, and, more importantly, they are not often very
useful.
A good model should be simple (i.e., reduce the complexity
of the world), explicit (i.e., use stable, well-defined concepts),
and, most importantly, testable (i.e., have clearly defined outcomes). These ideals are not readily apparent in Donohue
and Denham’s work, primarily because concepts like a “mosaic of regional networks” are not sufficiently concrete. Did
Indo-Pacific populations interact equally across geography
and time? (Greenhill and Gray [2005:34] would call this a
“maximally interconnected network.”) Or, as we suspect Donohue and Denham intend instead, did Indo-Pacific populations interact in very specific ways at very specific times in
very specific places? If so, when exactly were those times, and
what exactly were those places?
Herein lies the rub. The out-of-Taiwan model is so easy to
criticize precisely because it is relatively simple, explicit, and
testable. It has clear expectations: we can distinguish when
data fit the model and when they do not. Conversely, Donohue and Denham’s model is sufficiently imprecise that it
cannot readily be evaluated. The charge leveled against Terrell’s earlier incarnation applies: such an “anything goes”
model risks being “vague to the point of uselessness” (Lum’s
comment in Terrell, Kelly, and Rainbird 2001:116).
By way of example, the out-of-Taiwan model is commonly
criticized for its insistence on correspondences between interdisciplinary data sets (such as genetics and language).
Amusingly, the initial stimulus behind this now-lengthy debate (Terrell and Fagan 1975) was the finding of just such a
statistical association between g-immunoglobulin diversity
and language affiliation in New Guinea (Giles, Ogan, and
Steinberg 1965). However, this discovery was no mere fluke:
over the past 40 years, similar associations have been found
on geographical scales both large (e.g., Cavalli-Sforza et al.
1988) and small (e.g., Lansing et al. 2007).
The latter case is particularly informative: the study examined genetic and linguistic diversity in 532 men from eight
small communities on Sumba, an island in east Indonesia less
than 200 km across. Under the out-of-Taiwan model, we
might expect a rapid Austronesian expansion to have left
specific signals in modern communities: they should carry
some Asian contribution and have a mid-Holocene origin,
and we might also observe remnants of shared linguistic/
genetic heritage. This is exactly what we find. Around a quarter of Sumbanese men carry Asian Y chromosome lineages;
Bayesian coalescent inference, a particularly robust form of
molecular dating, indicates that their villages are less than
5,000 years old; and the proportion of male Asian ancestry
correlates significantly with villages’ retention of protoAustronesian cognates. Expectations under Donohue and
Denham’s model are much less clear, but it seems unlikely
that long-standing regional networks could produce a combination of patterns quite like these.
For those who dislike the tenacity of the out-of-Taiwan
model, the reason seems obvious: there is no meaningful al-
242
ternative. Clearly, some early ideas were wrong (any expansion
did not result in complete replacement), but equally clearly,
some ideas were right (several genetic markers do trace back
to Taiwan). Rather than yet another personal take on Pacific
prehistory, what we need now are good alternative models:
simple, explicit, and testable. What expectations do these produce for Sumba? How do expectations differ between models?
No matter if these new models turn out to be wrong; in the
meantime, it would at least be very helpful if they were useful.
Waruno Mahdi
Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, D-14195 Berlin, Germany ([email protected]
.de). 9 VI 09
Donohue and Denham present a challenging reappraisal of
the Malayo-Polynesian dispersal, also reexamining the role of
non-Austronesians in ISEA culture history and reinspecting
the farming/language dispersal hypothesis. The MalayoPolynesian dispersal hypothesis adapted repeatedly to newer
insights: a successively developing “out-of-Indochina” model
(Heine-Geldern 1932; Kern 1889; Schmidt 1906) gave way to
the “out-of-Taiwan” (Chang, Grace, and Solheim 1964) model
currently under discussion. The dispersal is a baffling subject,
having been linear in neither its linguistic, genetic, nor cultural
aspect. However, if the manifestations and consequences of
that nonlinearity previously remained inadequately appreciated, the present treatment perhaps oversimplifies them.
Modern humans inhabited ISEA long before the Holocene,
but the rising sea (14,000–10,000 BP; Dunn 1970) caused
migrations from inundated lowlands. Besides increasing population diversity and density (encouraging plant domestication) at both ends, Indochina and New Guinea, retreating
populations would often be trapped on temporary islands;
only those who learned to cross the sea survived (Mahdi
1988). The authors thus rightly stress early to mid-Holocene
development of maritime interaction and the resulting diffusion of cultural artifacts and irregular distribution of genetic
markers. Equatorial populations of inundated regions also
reached Taiwan, implying that the later Malayo-Polynesian
dispersal included their descendants. Whether genetic traces
of this southward movement are discernible before the background of their earlier northward migration is unclear.
In terms of former racial concepts, ethnologists distinguished darker-skinned “Proto-Malays” and lighter “DeuteroMalays,” as having, respectively, a greater and a lesser equatorial admixture (cf. Bylmer 1943). A similar distinction was
made by the Malayo-Polynesians, with two distinct protoforms for “person” (Mahdi 1994b). Intensive contacts emanating from western ISEA since 2000 BP could explain why
distribution of genetic markers linking Polynesia with Taiwan
“is not prominent in ISEA.” Nevertheless, indigenous Taiwanese are reported to be genetically closer to the population
Current Anthropology Volume 51, Number 2, April 2010
of ISEA and Oceania than to that of the mainland (Lin et al.
2005).
Language subgrouping based on exclusively shared innovations requires that the considered cognate sets include only
inherited forms (not borrowings). But Malayo-Polynesians
are notorious for long-distance contacts, and there was, moreover, a systematic source of noise. Malay-speaking seamen
spanned ISEA since 200 BCE–200 CE (Solheim 1980)—transporting spices from Maluku to the Malayan Peninsula, India,
and China—and sailed through the Philippines to China before 300 CE (Mahdi 1994a, 1999a, 1999b). This resulted in
borrowings across group boundaries and in misleading secondary sound correspondences, with fateful consequences for
eliciting a structured phylogeny of Western and Central
Malayo-Polynesian. Until this is conclusively resolved, it inevitably creates a misleading impression of “propagation of
Malayo-Polynesian languages . . . without direction or
hierarchy.”
Borrowing also camouflages the separate status of CentralEastern Malayo-Polynesian: several “almost-exclusive” innovations have cognates only in Sulawesi and the Philippines,
but their distribution suggests that they represent a remnant
substratum of the former southward movement of CentralEastern Malayo-Polynesians (Mahdi 1994b). Malayo-Polynesian is only one of several highest-level branches of Austronesian, and a “significant break between PAN and PMP”
is relevant only if demonstrated to be substantially greater
than those between PAN and Atayalic or between PAN and
Proto-Tai-Kadai (grouped under PAN; Sagart 2005). So altogether, the apparent inconsistencies in the linguistic picture
do not disprove an out-of-Taiwan dispersal.
With regard to the material-culture record, equatorial climatic conditions required significant adaptations. The daily
photoperiod was too short for japonica rice (Grist 1959), while
Zimmermann (1992) concluded that rice-cultivating “ProtoMalays” settled the highlands because of the more suitable
soil (Mahdi 1994b). Adapting the rice variety to the tropics
and littoral immigrants to a highland habitat required time.
Malayo-Polynesians of the first wave depended on locally
available staples, learning from the indigens they encountered.
The authors correctly stress the role of the latter in domestication and distribution of the cultigens but perhaps underestimate the adaptive ability of Malayo-Polynesians.
Farming by Malayo-Polynesians as well as by indigens must
have been the decisive factor underlying maritime contact
and exchange. Thus, stressing the role of indigens and the
importance of contact and exchange does not dismantle the
farming/language dispersal hypothesis altogether but modifies
it, by appreciating the reciprocal role of indigenous farming.
Watercraft likewise gained central importance, but the outrigger canoe was not the original Malayo-Polynesian watercraft. That was the double canoe (Doran 1981; Mahdi 1999a;
also Mahdi 1988, 1994b). A distribution area, beginning in
China (Mahdi 1994b, 1999a; Needham, Wang, and Lu 1971),
stretches over ISEA to Oceania and also to India and Sri Lanka
Donohue and Denham Farming and Language in Island Southeast Asia
(among others, Doran 1981, fig. 40; Mahdi 1999a, fig. 5.4).
But this too apparently involved exchange, even of actor identities. Thus, seafaring Sama-Bajau and Oranglaut were typically Negritos, and Malay speakers were described in Chinese
sources as “black and naked” (Mahdi 1999a; Pelliot 1925).
Meanwhile, there are hunter-gatherers of distinct Mongoloid
appearance in the interior of Kalimantan (Sellato 1994; Sercombe and Sellato 2007). Therefore, Donohue and Denham
rightly modify the farming/language dispersal hypothesis in
a second way, by allowing for interchange of immigrants and
indigens in various roles.
The linguistic data do not contradict a Malayo-Polynesian
language dispersal from Taiwan, but the authors’ genetic and
culture arguments indeed strongly indicate a more active role
of indigens in that process.
Stephen J. Oppenheimer
Institute of Cognitive and Evolutionary Anthropology,
School of Anthropology, Oxford University, 64 Banbury
Road, Oxford OX2 6PN, United Kingdom (stephen
[email protected]). 5 VI 09
Paradigms, especially old ones, die harder than Bruce Willis.
(James Adovasio 1999)
The “express train to Polynesia from Taiwan” paradigm has
a contentious 34-year history (Shutler and Marck 1975). Donohue and Denham, like Shutler and Marck a linguistarchaeologist team, without directly challenging the arguments for an Austronesian linguistic homeland in Taiwan,
question its whole relevance to Southeast Asian and Oceanic
demography, culture, and prehistory. What is new, and what
will hopefully have most impact in shifting the logjam in the
field, is that now there is a linguist leading the challenge.
Why should a single discipline matter so in what is necessarily a multidisciplinary field? Simply because the archaeologist paradigm-champions made it so. Donohue and Denham aptly quote Bellwood’s tautology “the question of
Austronesian origins is basically a linguistic question”; this is
more malapropism than tautology, based on the conviction
that languages, genes, and culture move hand in hand to
invade and replace (Diamond and Bellwood 2003). Combined
with the oft-stated stricture that only linguists and likeminded archaeologists are qualified to comment on Austronesian (AN) linguistics, this idea has locked the issue up from
objective review by anyone but an inner circle for well over
a generation. For these arcane reasons, voices from the humanities guilds of linguists and archaeologists doubting the
adequacy of the emperor’s old clothes will resonate better
than any number of raggedy dissenters from other disciplines,
however objective and informed their critiques.
Donohue and Denham commence their linguistic reassessment with the old dissenters’ truism that “the ‘Austronesian dispersal’ might better be termed a ‘Malayo-Polynesian
243
[MP] dispersal,’ since nine of the 10 primary subgroups of
Austronesian are attested to only on Taiwan and only the
Malayo-Polynesian branch has members outside Taiwan (and
none on mainland Taiwan)” and then proceed, with evidence,
to collapse the hierarchical MP tree, also removing its northsouth, west-east directional nature, questioning even the validity of the Central-Eastern- and Eastern-MP branches. The
MP tree thus increases from 25 to 30–40 primary branches,
of which Proto-Oceanic may be one, and ends up looking
less like a tree than a scatter bomb with no identifiable point
of dispersal, although their evidence points geographically to
eastern Indonesia (pace Dyen 1965). By showing that “the
impression of a graduated dispersal of Malayo-Polynesian languages south from Taiwan does not hold up and must be
rethought,” they kick out the language support from the archaeolinguistic orthodoxy and puncture recent lexicostatistical claims to reproduce and date those stages in a multitiered
hierarchy (Gray, Drummond, and Greenhill 2009).
Taking another dissenters’ leaf, they point out the fact that
“there is no evidence of a Taiwanese origin for much of the
basic vocabulary that is cited as characteristic of the Austronesian dispersal, such as terms for maritime technology and
many of the food plants that have been associated with the
Austronesians.” This observation is not new, also applies to
faunal terms, and long ago was shown as a hole (e.g., Oppenheimer and Richards 2001a) in the cognate lists of Blust’s
seminal homeland paper (1984–1985), where less than a third
of any PMP cognates reconstruct in Taiwan. This back-tofront linguistic picture fits with their cited archaeological and
genetic evidence that the same items are prior Southeast Asian
innovations, again already reviewed in the literature (e.g., Oppenheimer 2006). However, their linguistic perspective also
implies a massive influx (up to 68%) of borrowed lexicon in
Island Southeast Asia if Taiwan was the true homeland. This
perspective is not taken account of in their characterization
of Austronesian as a lexically conservative family, although
they have made this estimation for the effect of similar nonAN borrowing in Melanesia. For instance, of Blust’s
(1984–1985) cognate set, only 32% of those that reconstruct
in Taiwan survive in Proto-Oceanic, by contrast with the 56%
survival for PMP cognates (88% in Blust [1993:245]), making
only PMP conservative.
A couple of flies buzz around this excellent paper. The
authors claim that their reconstructive approach to the issue
is new, which it is in respect of the illuminating linguistic
perspectives, but their claim that dissenters “have been unable
to address all aspects” is disingenuous, since most of the rest
of their archaeological and genetic arguments, though sound,
are extensively discussed and prefigured in their cited dissenting literature (see also Oppenheimer 2004, 2006; Oppenheimer and Richards 2001a, 2002). They claim novelty for
several insights, such as the importance of maritime networking, effectively omitting full acknowledgement of Solheim’s seminal work (2006), and the complexity of interactions, previously stressed by Terrell (1988) and caricatured by
244
his opponents as “confused.” Even the claim that dissenters
were too concerned with the relevance of language and homelands is overstated (e.g., Oppenheimer 2004; Oppenheimer
and Richards 2002). That was the orthodox camp. Finally,
they seem ambivalent about the linguistic significance of redslipped pottery, acknowledging the absence of a clear link but
sometimes relying on it as evidence.
I thank Martin Richards, John Terrell, Mark Donohue, and
Tim Denham for helpful comments on mine.
Victor Paz
Archaeological Studies Program, University of the
Philippines, Diliman, Quezon City, Philippines
([email protected]). 1 VI 09
Donohue and Denham address head-on an ongoing discourse
on early Island Southeast Asian (ISEA) history and propose
an alternative way of looking at the state of knowledge in
three fields. The paper engages the dominant farming and
language dispersal hypothesis (FLDH) as represented in the
Austronesian dispersal, which explains the processes involved
in the transformation of culture within a time depth of about
5,000 years by using data from linguistics, archaeology, and
genetics. The authors review the latest studies and find the
Austronesian expansion hypothesis wanting in answering the
complexities of past human cultures around the mid-Holocene in ISEA. My own views on the topic have close affinity
with the spirit behind the article. I agree with the authors on
some points, differ on others, and have more sympathy for
the Austronesian-spread hypothesis than they do.
There are strong arguments in the paper. It presents how
complicated are the dynamics between the various data sets
used to answer questions about the human past—clearly
showing that these data sets are more often than not hard to
whip into cadence—while still underscoring the importance
of pursuing multiple disciplinary lines of inquiry. I also agree
that the default thinking for understanding cultural history
in ISEA should not be that cultures developed in isolation
before interacting; it should rather be that cultures developed
while interacting with other cultures from the very beginning.
It is significant to me that the authors did not strongly
challenge the historical linguistic conclusion on the Austronesian homeland, concentrating instead on the nature of the
spread and eventual dominance of this language family across
ISEA and the Pacific, which they then use to question the
reasoning of the FLDH. This tells me that it is still a relevant
challenge to know the sequence of language formation, especially within the Malayo-Polynesian languages. The genetics
data review highlights the significance of the small percentage
of Taiwan-based markers found in the current ISEA population. It is, however, not clear to me whether this conclusion
was due to sampling strategies or to methodological limitations of modern population genetics in revealing fine-grained
images of past populations. I also have reservations about the
Current Anthropology Volume 51, Number 2, April 2010
view that interaction without population movement is sufficient to change languages. In the Philippine experience, for
example, the power and prestige in the hands of a few native
speakers of Spanish or English did not change the languages
of the Philippines from Austronesian to Indo-European in
the past 400 years; this may have something to do with the
small population size of native speakers. If we assume that
cultures were already interacting at the very beginning, the
possible advantage of the Austronesian-speaking cultures over
others may be based on a suite of cultural and technological
know-how that island populations saw fit to adopt at various
points in a long series of interaction.
The seeming simplicity of the FLDH may be misleading to
critics. It has been expounded on and transformed by advocates, especially in its ISEA application, in response to new
data and in engagement with basic research. For example,
from a main hypothesis that cereal agricultural was the engine
for population movement and culture spread, Bellwood
(1995:108–110) has postulated at least seven interrelated explanations as to why people moved, including a possible role
for an early form of marcher culture.
The claimed “new historical framework” for understanding
the nature of culture change in the mid-Holocene in ISEA is
not necessarily new. The emphasis on “the mosaic of regional
networks and the social processes” has already been articulated in a way very close to the description presented here
(see Meacham 1984–1985, 1995; Solheim 1969, 1990, 2006).
The approach is fresh, however, in that it works within the
same methodological frame in which the dominant FLDH is
based and not just in a dominantly single-discipline data set.
I see the discourse on the nature of mid-Holocene ISEA
culture as married to the very problematic archaeological period label “Neolithic.” A serious assessment of what “Neolithic” means in ISEA is much needed, and this work is a
contribution in that direction. For starters, I think that Austronesian-speaking cultures did not simply encounter small
bands of hunter-gatherers as they moved through the landscape. There is a small but growing data set from my own
research and that of colleagues that seriously suggests the
possibility of a pan-ISEA existence of much more complex
societies that, at least, were maritime, had a developed suite
of shell material culture, perhaps had non-cereal-based agriculture that had existed before and had interacted with Austronesian-speaking cultures.
I posit that the early spread of Austronesian-based cultures
may not necessarily be well represented in observed ISEA
cultures. I am also looking at the possibility that the more
significant pattern coming from the twinning of data sets
reflects much later interactions and confluences, perhaps from
around the time of population movements of Austronesians
with developed rice agriculture, metallurgy, and intricate pottery production. Much is yet to be reflected on and learned,
making basic research in our region ever more exciting.
Donohue and Denham Farming and Language in Island Southeast Asia
Matthew Spriggs
School of Archaeology and Anthropology, A. D. Hope
Building, Australian National University, Canberra,
Australian Capital Territory 0200, Australia
([email protected]). 12 VI 09
The core of the paper is a new and stimulating treatment of
the linguistics of ISEA. The treatments of genetics and archaeology seem built on much flimsier foundations. In part,
this is because trying to catch a “moment” in historical genetics is almost impossible, given that there are frequent revisions of interpretation and continuing problems of inadequate sampling of relevant populations.
Historical genetics presents a very immature disciplinary
profile compared to both archaeology and linguistics for the
region. Thus, we can have the “origin of the Lapita complex”
solved by a study of people in the East Sepik region on the
north coast of Papua New Guinea (Vilar et al. 2008), a population that almost certainly derives from a westward Austronesian expansion that, on archaeological evidence, arrived
only sometime around 1,700–1,500 years ago! Similarly, Friedlaender et al. (2008) “contribute to a resolution of the debates
over Polynesian origins and their past interactions with Melanesians” by choosing, almost without exception, populations
in the Bismarcks from areas where there is either no evidence
of Lapita occupation or clear evidence of post-Lapita catastrophic vulcanism and subsequent population replacement.
The modern populations are thus almost completely irrelevant in consideration of the problem they are seeking to resolve: the ISEA contribution to the settlement of the Pacific.
Problems of interpretation do not stop there. Donohue and
Denham quote a most interesting paper by Excoffier and Ray
(2008) to explain the apparent link between indigenous Taiwanese and Polynesians as being due to genetic “surfing,”
where originally low-frequency alleles can reach high frequencies during rapid population expansion. But the paper
makes clear that there is a major equifinality problem in the
interpretation of genetic data in historical terms. The equally
plausible explanation for the link is, of course, that the original
“signal” of the expansion remains clear in both Taiwan and
Polynesia but has been obscured by later population replacement or mixing in areas in between.
This is, in fact, exactly what seems to have happened linguistically as well, and again current situations may be poor
guides to past ones. But Donohue and Denham do not seem
to draw the obvious conclusion. There may be much mixing
and substratal influence in ISEA Austronesian languages, but
this has likely occurred over the past 3,000-plus years, after
the population wave had passed through to the east en route
to Polynesia. Delineating such influence is interesting, but it
is again irrelevant to the problem they are seeking to address,
which is primarily about the conditions under which Austronesian languages and putatively associated sets of material
culture spread across the region in the second millennium
245
BC. Their view could be tested by comparing the structures
of reconstructed protolanguages for slightly later-settled areas
to the east, such as Proto-Oceanic (POc) or later subgroups
in the island Pacific beyond any possible direct substratal
influence. There is also another underutilized resource: the
two remotest ISEA Austronesian languages, those of the Marianas and Palau, settled from ISEA simultaneously with or
even slightly earlier than Lapita in the Bismarcks and presumably not subject to a great deal of later Southeast Asian
linguistic influence. Do they show the required substratal influences? I doubt it. So genetic support is fickle, and the early
Malayo-Polynesian (MP) languages that were involved in subsequent spread into the Pacific were neither near-creoles
nor MP-lexically-but-something-else-structurally, as Donohue and Denham portray them.
For the archaeology, the immediately pre-Neolithic period
in ISEA is critical, as they suggest, but at present the evidence
of the long-distance preadapted exchange networks that their
model requires is lacking. The diffusion of crops of New
Guinean origin to the west in pre-MP times doubtless took
place, but it may have done so over a long period and with
initially no great implications for the societies involved. It was
the combination of plants and domestic animals that was
revolutionary, along with a developed boat technology that
we know must have existed in ISEA for the 3500 BP or earlier
open-ocean crossing to the Marianas. Add in the pottery,
polished stone adzes, distinctive shell artifacts, and tattooing
chisels noted by Donohue and Denham—and one could add
other features, such as the jar burial complex (Bedford and
Spriggs 2007)—and it still looks like a “package” to me.
There are many positive aspects of the paper, too. The
flattening of MP linguistic subgroups between Proto-MP and
POc better explains the almost instantaneous archaeological
dates for Neolithic spread across much of ISEA around 3800
BP than does Blust’s (1995) earlier subgrouping: I wish I had
had this paper in front of me when I last considered the
relationship between Austronesian spread and archaeology
across its range in a paper written in 2006 (Spriggs, forthcoming). Contrary to their hope, however, there may still be
a need for an Eastern MP or at least some node between PMP
and POc, as there is a pause between the earliest Neolithic
dates for the Philippines, Sulawesi, and the Lesser Sundas, at
about 3800 BP, and dates for the earliest Lapita sites in the
Bismarcks, at around 3350 BP.
John Edward Terrell
Field Museum of Natural History, Chicago, Illinois 60605,
U.S.A. ([email protected]). 30 V 09
If a generation is taken to be 25 years long, then two generations of archaeologists working in the Pacific have been
working with, or under, the same interpretative paradigm.
This abiding source of inspiration has been comparative historical linguistics, a field of intellectual expertise with a lengthy
246
pedigree in Western scholarship that effectively reached the
university scene in Hawai‘i, New Zealand, and Australia after
World War II, concurrently with the arrival of academically
trained archaeologists at the same universities.
There is no doubt that research done on Pacific prehistory
over the past 50 years, inspired by the formulae and logical
procedures of comparative historical linguistics, helped jumpstart modern archaeological work in Oceania after the war.
It is no secret, however, that I think this paradigm outlived
its effectiveness years ago.
Instead, I have been preaching (I am sure this is how some
have seen my own contributions to this journal, written with
several of my colleagues) that the ancient Pacific, seen in
human terms, is an “interlocking, expanding, sometimes contracting and ever-changing set of social, political, and economic subfields” (Terrell 1988:647), a playing field, so to
speak, where people have taken up different positions, at different times, traveling different distances, and perhaps from
different directions to play often similar although somewhat
different games.
Critical to this perspective is the idea that isolation between
settlements, islands, and even archipelagoes has rarely been
so absolute that new rules, discoveries, fashions, inventions,
or genetic traits could not be shared, passed along, traded
back and forth, or, for that matter, stolen if thought worthy,
wonderful, pleasing, or desirable. Also important is the idea
that the people involved in each instance could have come
from down the road, across the bay or valley, or from just
over the hill. They did not have to be people who lived far
away in some unknown homeland in southern China or
Taiwan.
The metaphor of a playing field only partly captures this
interactive perspective on the past, which is why more than
20 years ago I also suggested, borrowing a metaphor from
Darwin, that Pacific prehistory could be seen as an “entangled
bank” (Terrell 1988), a suggestion since then widely misunderstood and often maligned (Terrell 2000a). Therefore, while
it may be just wishful thinking, I am delighted to find Donohue and Denham saying much the same thing when they
write about “prolonged interaction and mixing among populations across ISEA,” and emphasize “the mosaic of regional
networks and the social processes prevalent in ISEA before
and during the spread of Austronesian languages and cultural
influences from Taiwan.”
I would, however, quibble somewhat about their choice of
wording. A “mosaic” is a tessellated picture or pavement.
Opting for this metaphor could be read as saying that human
diversity is a jigsaw puzzle of different entities (“societies,”
“cultures,” and the like) rather than “an interlocking, expanding, sometimes contracting and ever-changing set of social, political, and economic subfields.”
Nor do I think that using the word “mimic” to describe
why it is that neighboring Austronesian and non-Austronesian
speaking communities are often genetically similar is likely to
be helpful. This word choice makes it sound as if what has
Current Anthropology Volume 51, Number 2, April 2010
been at play, genetically speaking, is some form of Batesian
mimicry rather than love and marriage or the commonplace
but less sanctioned—although still highly interactive—alternatives to either or both.
I also have a substantive quibble. Was Taiwan the ProtoAustronesian homeland because this is “where nine of the 10
first-order subgroups are found”? Taiwan is a beautiful island,
and not a small one as islands go. But until someone explains
why there are so many top-notch subgroups there (and why
not all of them, pray tell?), I will be skeptical. Why did this
degree of diversity develop there? Why has it been maintained
there for so many millennia? Why should this diversity, however highfalutin, be seen solely as a sign of great age rather
than as a sign also that there is something noteworthy about
Taiwan as a social, ecological, and political realm?
Nonetheless, Donohue and Denham have given us a datarich, compelling, and radical reinterpretation of prehistory in
island Southeast Asia. I never thought I would live to see the
day when a linguist and an archaeologist working in the Pacific would team up to write jointly, for example, “language
families, like bacteria or bdelloid rotifers, can form because
of the convergence of originally unrelated taxa.” I do not think
that even George Grace would go this far, and shockingly few
Pacific archaeologists have ever seriously challenged the historical truth value of the methodological simplifications that
are built into the comparative method and have for far too
long provided too many with easy and compelling but implausible reconstructions of Pacific prehistory.
Robin Torrence
Anthropology Department, Australian Museum, 6 College
Street, Sydney, New South Wales 2010, Australia
([email protected]). 20 V 09
First impressions gained during European expansion have
heavily influenced research agendas for Pacific and ISEA archaeology that underlie models for the spread of Austronesian
languages. This welcome and convincing critique of the pervasive Austronesian farming-language dispersal model overcomes many, but not all, of these biases. First, Cook (and
successors) was especially taken by the Polynesians, with their
fair skin and hierarchical society (“just like us”), but was
horrified by the unruly Melanesians. Since then, scholars have
been obsessed with human colonization of the remote reaches
of Polynesia and uncomfortable with imagining how the
seemingly less sophisticated Melanesian cultures could have
contributed to the achievements of their eastern cousins. This
focus also meant that archaeologists have sought a single explanation for Polynesia, one that involves simple, directional
change, and have been less concerned whether it was equally
applicable to the enormous and highly diverse region encompassed by Austronesian languages. Several commentators
noted that the “Polynesian tail wagging the dog” (e.g., Allen
1991) prevented new ideas about the peopling of the Pacific
Donohue and Denham Farming and Language in Island Southeast Asia
and the spread of Austronesian languages, but the farminglanguage dispersal model (“one size fits all”) has continued
to dominate research and thinking, even for ISEA.
The second consequence of European exploration is an
assumption that since New Guinea was always separate culturally from its ISEA neighbors, the movement of Austronesian language/ceramics/pigs/etc. required a singular, oneway event. This misguided view also derives from historical
circumstances. After Europeans took control of the Spice Islands, long-term contacts to the west were significantly disrupted, creating an unusual period of isolation of ISEA from
New Guinea and beyond. As the Donohue and Denham results now make brilliantly clear, modern international boundaries did not operate in the past (cf. Swadling 1996).
Two of the authors’ approaches are integral for their successful overhaul of the language-farming hypothesis. First,
independent studies of the genetic, linguistic, and archaeological data using distinct and specific methods relevant to
each field of study have efficiently unpacked the model. Since
their analyses show that genes, languages, agriculture, and
technologies moved at different speeds, directions, and timings, it is clear that social processes were the key motor. This
is an important theoretical shift from the vague notion of
“progress” underlying agriculture as a prime mover. Second,
they note that seemingly directional changes can accrue over
long time periods. Their paper makes it abundantly clear that
a supra-island social entity fueled by frequent communication
and exchange was built up slowly. This diffuse nest of social
networks was not coterminous with a single language/genetic
group. Its existence has been overlooked because of the assumption that cultural groups were isolated and because it
operated on a greater spatial scale than is generally studied
by linguists and anthropologists.
Donohue and Denham’s effective criticism of the language/
farming dispersal model encourages reinterpretations of ISEA
and Melanesian prehistory that consider supraisland social
networks. Their use of Australian colonization of the New
Guinea highlands as an analogy for the spread of Austronesian
language and the Neolithic cultural package, however, shows
that there is still a need to shed Eurocentric biases. Their
thinking is flawed because it uses landlocked groups, like the
original Renfrew dispersal theory they dispute, in contrast to
the maritime universe that they stress fits ISEA. The Donohue
and Denham model also suffers because it is fundamentally
based on notions of power relations between visitor and local
communities derived from recent colonial society in New
Guinea. In this highly risky and unstable environment, where
natural disasters (volcanoes, tsunamis) are frequent and widespread, the maintenance of external links was essential for
continual reproduction of small groups. Since transport was
easy and quick but potentially perilous, people could achieve
prestige by returning safely from beyond the horizon with
mates, resources, and “wealth” (Renfrew 1993) that might
stimulate the adoption of novel ideas/languages/objects. Over
247
the long run, a semblance of homogeneity could be generated
across a region.
A brief example will highlight the potential for new interpretations of prehistory that follow Donohue and Denham’s
identification of a supraisland social entity. My focus on people actively seeking innovations to bring back home might
explain the rapid distribution of red-slipped pottery within
ISEA and the Bismarck Archipelago. It echoes Allen and
White’s (1989:142) proposal that “the craft of pot making
was first acquired by some aceramic voyagers” rather than
introduced by foreigners. More recently, Torrence and Swadling (2008; cf. Torrence et al. 2009) argued that wide-scale
social networks of the sort implied by Donohue and Denham
provided the structure through which the practice of redslipped ceramics could have spread initially. The concept of
“adventurers,” driven by competition for status, who return
novelties to their home communities rather than impose them
on others, adds the missing process from previous ideas. With
Eurocentric biases removed, prehistory in ISEA and the Pacific
has a bright future.
Reply
The comments on our paper are thoughtful, diverse, and well
received; they variably offer criticism, lend support, and suggest new lines of inquiry. Inevitably, our response focuses on
criticisms. However, we acknowledge the predominantly positive reaction of most commentators and look forward to
pursuing some of their thought-provoking suggestions.
Foremost, we do not claim to be the first either to critique
the farming/language dispersal hypothesis in the Austronesian
context or to propose an alternative metanarrative focused
on maritime social networks within ISEA. Our arguments
follow and augment but are different from those made previously by Solheim, Meacham, Terrell, Oppenheimer, and
Richards. As Paz notes, ours is a “head-on” multidisciplinary
assault on and revision of ISEA history during the Holocene.
The strengths of the argument derive from our examination—
in disciplinary isolation and in multidisciplinary combination—of the historical implications of genetic, linguistic,
archaeological, and crop plant data for ISEA during the Holocene. Ours is a new synthesis intended to provoke a new
challenge.
We do not question the hypothesis of a Taiwanese origin
for the Austronesian languages; indeed, recent work adds further credence to the idea that this island was the most likely
homeland of Austronesian languages (Ross 2009). We question the extent to which the dispersal of Austronesian languages from Taiwan can be linked to the demic dispersal of
farming or a “Neolithic revolution” across ISEA. Explicit or
implicit assumptions of linkages between genetics, linguistics,
and archaeology are pervasive in “Austronesian culture his-
248
tory” and continually lead to problems of interdisciplinary
circularity.
Some commentators demonstrate a continual slippage between archaeological evidence and linguistic interpretation to
present an inclusive historical metanarrative. The two lines
of evidence are continually deployed in self-referential and
interchangeable ways for mutual support. There is no explanation of why agriculture, the purported “Neolithic” package,
and Malayo-Polynesian languages necessarily spread together;
they are merely assumed to have spread together. Our argument, by contrast, has been carefully crafted to ensure that
different lines of evidence are neither conflated nor used interchangeably. We consider each line of evidence separately
and then compare them to assess accordance or discordance.
Illustrating the problem of slippage, Spriggs suggests that
an Eastern-MP subgrouping should be retained because of “a
pause between the earliest Neolithic dates” for ISEA and the
earliest Lapita sites in the Bismarcks. His use of archaeological
evidence to justify a linguistic interpretation is predicated on
a major assumption, namely, that an archaeological artifact
(effectively, red-slipped pottery) can be used to closely track
language dispersal in the past. The veracities of this particular
link, as well as of overarching conceptions of farming/language dispersal and Neolithic spread, are left unexamined. In
our article, we open these assumptions up to examination
and find them wanting.
As several commentators note, there are major problems
with the application of historical genetics in ISEA. These include the absence of “any reliable molecular dating method
for the Holocene” (Bellwood, contra Cox and Lansing), the
“methodological limitations of modern population genetics
in revealing fine-grained images of past populations” (Paz),
and disciplinary immaturity and equifinality (Spriggs); all
conspire against a unified picture. Our review highlights the
lack of clear and significant genetic clines from Taiwan across
ISEA. Indeed, no unequivocal reconstructions of human migrations within ISEA during the Holocene emerge from the
human genetics of contemporary populations; various studies
detect different signals and reach different conclusions, depending on geographical scale of analysis, locale, molecular
marker, and interpretative stance. At best, genetic signals
across ISEA are “fickle” (Spriggs).
There is general agreement on the need to revise current
interpretations of the phylogeny of Austronesian languages
(see Mahdi, Oppenheimer, and Spriggs), although several minor linguistic issues are raised. Spriggs suggests that the nonTaiwanese features of the grammars of Malayo-Polynesian
languages may be due to borrowing rather than substratal
influences. We note that substratal properties might reflect
borrowing early in the history of a language or language shift.
Given the nature of the properties we discuss and the massive
replacement of vocabulary (see our discussion of Wichmann’s
measure and the replacement rates highlighted in Oppenheimer’s response), we tend toward the language-shift hypothesis. Following Spriggs’s comments, we confirm that Cha-
Current Anthropology Volume 51, Number 2, April 2010
morro and Palauan, the Austronesian languages of the
Marianas and Palau, show evidence of substratal influences
that must have been acquired before their migration east from
the Philippines area (Donohue 2007b). Similarly, while we
agree with Spriggs that Proto-Oceanic is well-established
within Malayo-Polynesian, it is defined by a series of irregular
changes and has a strongly “Papuanized” grammar, indicating
nonorganic developments in its history.
We reiterate that the genetics of Austronesian-speaking
populations need not be different from those of non-Austronesian-speaking populations in regions of contact (Mona et
al. 2009). Bellwood circumlocutiously concurs by implying
that such populations occur in several peripheral regions,
namely, all of those places in which Austronesian and nonAustronesian speaking populations are juxtaposed. As Terrell
notes, the word “mimic” implies an intentional act, whereas
we intended to characterize unforeseen, cumulative effects.
Cox and Lansing reaffirm the purported correspondence
between genetics and language in ISEA at both large and small
scales. Given the problems outlined above, it is not surprising
that there are few others who still maintain that there are
significant correspondences between Austronesian languages
and Taiwanese genetic signals at the large scale in ISEA (e.g.,
Capelli et al. 2001), with similar reservations in other regions
(e.g., Heggarty 2007). At the small scale, Cox and Lansing
throw down the Sumba gauntlet, even though, a priori, a
pattern between language and a genetic marker on one island
cannot be extrapolated to be symptomatic of a whole region.
On Sumba itself, we note the discordance between linguistic
and genetic data sets for the island (based on information in
Lansing et al. 2007): the linguistic “homeland” of Austronesian language dispersal on the island shows the lowest rates
of occurrence of the Y chromosome haplogroup (O) that is
claimed to be diagnostic of Austronesian farmers.
How do we account for the geographical variability in genetic markers and linguistic associations across ISEA? It seems
improbable that a “simple” model (following Cox and Lansing), albeit one that is ”explicit and testable,” will do; ease
of utility is no measure of veracity. At the same time, we assert
that the mosaic of social interactions that we are proposing
as a conceptual framework for understanding ISEA is no less
explanatory or testable than a seemingly unitary hypothesis.
To quote Bellwood (2002:27), “No claim is being made that
only farmers ever dispersed through already-inhabited areas
in prehistoric times, or that only farmers ever spread large
language families. . . . No claim is being made that all early
farmers were obliged to undergo expansion” (emphasis in
original). Campbell (2002) points out that language families
associated with agriculture show the same frequency of
spreading as those that are not. In short, the farming/language
dispersal hypothesis does not provide a testable model for
ISEA or elsewhere, because ultimately, the key factors in determining spread of language families are not the presence or
absence of agriculture but rather “historical contingency and
opportunity” (Bellwood 2002:27).
Donohue and Denham Farming and Language in Island Southeast Asia
Archaeology is concerned with elucidating the complexities
of human (pre)history through observed traces of material
culture. As such, it is ideally situated to shed light on the
social and historical complexities that underlie the geographical variability of genetic-linguistic accordance/discordance in
ISEA. As has long been realized, spatial correlations are often
methodologically problematic and, even when valid, require
explanation (Gould 1970). In this case, we must elucidate the
(pre)histories of social exchange, interaction, and practice
within ISEA that account for the observed genetic, linguistic,
and archaeological phenomena.
Of greatest significance is that the farming/language dispersal hypothesis is devoid of archaeological and paleoecological evidence to indicate the spread of farming across ISEA
from Taiwan. The earliest Malayo-Polynesian speakers were
agricultural; this conclusion is apparent from historical linguistic evidence and accords with archaeological evidence
from Taiwan. Malayo-Polynesian speakers contributed to the
range of plant exploitation practices within ISEA, as did the
people living there and those from mainland Asia and Melanesia. The chicken, dog, and pig were initially domesticated
in Asia and then brought into ISEA, but the archaeozoological
evidence does not indicate that they were all brought at a
single time or along a single route, namely, via Taiwan.
None of the commentators present evidence to bolster an
argument for the spread of farming from Taiwan. The lack
of archaeobotanical evidence for rice is claimed by Bellwood
to result from a lack of microfossil (phytolith and starch
analyses) applications in ISEA contexts. This is a sleight of
hand (see Paz 2005). Rice is frequently detected from macrobotanical remains at archaeological sites in mainland East
Asia and yet is seldom found, except as inclusions in pottery,
at purportedly early “Austronesian” or “Neolithic” sites in
ISEA. Although the absence of evidence is not evidence of
absence, it certainly cannot be used in support of something,
namely, farming.
Most commentators agree with, and Mahdi and Paz elaborate on, our notion of pre-Malayo-Polynesian cultivation in
ISEA, which is implied from the dispersal of vegetatively propagated plants. Our argument for this is not founded solely
on the controversial evidence of banana from Uganda (Lejju,
Robertshaw, and Taylor 2006), as Bellwood claims (see papers in Denham, De Langhe, and Vrydaghs 2009). Spriggs
concludes that these forms of horticulture had “no great implications for the societies involved.” He considers the combination of plants and domestic animals carried by MalayoPolynesian-speaking, “Neolithic” farmer-voyagers to be the
source of truly revolutionary change. There are two problems
with Spriggs’s argument: conceptual and evidential. Conceptually, his interpretation carries echoes of a “people without
history” (Wolf 1982), whereby not much happened before the
arrival of Malayo-Polynesian speakers; real (pre)history seemingly begins with the “Neolithic.” Such views exhibit a strong
Eurocentric inheritance (raised by Torrence), and, further, Paz
249
problematizes the application of the concept “Neolithic” in
the ISEA context.
Evidentially, there is no multidisciplinary evidence of revolutionary socioeconomic or cultural changes across ISEA at
the supposed time of Malayo-Polynesian language dispersal,
whether fast train, slow boat, or pulse paused. If the plants
and domestic animals carried by these early farmer-voyagers
were such important drivers of change, why is archaeological
and paleoecological evidence of these revolutionary transformations so elusive? More dramatic evidence of agricultural
expansion, as well as more ubiquitous cultural remains, occur
millennia later (as Paz notes). Even if “early” rice and domesticated animal remains are found at the occasional site in
ISEA, such fragmentary evidence is consistent with introduction via multidirectional exchange networks and adoption
within, as well as transformation of, preexisting practices and
social forms. As Mahdi eloquently states, greater consideration
is needed of the “reciprocal role of indigenous farming.”
Preceramic maritime interactions within ISEA, as well as
with and within neighboring Melanesia, are clear (Bulbeck
2008; Torrence and Swadling 2008; White 2004). An increasing body of evidence from archaeology and historical linguistics indicates that several elements of the putatively
Malayo-Polynesian cultural package either were already in
ISEA before Taiwanese influences or were introduced from
elsewhere; these traits include voyaging technology, some domestic animals, at least one pottery style, numerous crop
plants, non-cereal-based agriculture, and shell material culture
(see Mahdi’s and Paz’s comments). Taiwanese influences were
probably strongest in the northern Philippines, just as Melanesian influences were strongest in eastern Indonesia; these
patterns are products of distance-decay effects across socially
constructed space. Although distinctively regional material
cultures persisted, a seeming material-culture “package” consisting of originally indigenous and exotic traits formed in
ISEA through a partial homogenizing process of social interaction, exchange, adoption, and transformation. Items were
assimilated into Malayo-Polynesian languages within ISEA,
and Malayo-Polynesian speakers took these words and things
to, as well as exchanged them with, places outside ISEA.
We have not questioned a Taiwanese origin for red-slipped
pottery, but as Torrence has noted, it need not have dispersed
solely with Malayo-Polynesian people or languages. To allay
any ambivalence on this matter (noted by Oppenheimer), the
spread of red-slipped pottery should not automatically be
equated with or used as a proxy for the spread of farming, a
language family, or people in ISEA. There are no presumptive
correspondences: farming ( red-slipped pottery ( language
( people. Each element has its own historicogeographic expression; sometimes these elements co-occur, and sometimes
they do not.
Torrence takes us to task for using the recent colonial experience of New Guinea as an analogue of processes on islands
in ISEA during the Holocene. The analogy was designed to
be a heuristic illustrative of “some of the social processes”
250
involved and should not be read literally. Her criticism is
founded in terms of our use of a landlocked example, whereby
we fail to capture the appropriate maritime qualities needed
to understand ISEA during the Holocene. We accept these
criticisms.
To sum up, we are not proposing an alternative hypothesis,
simple or otherwise, to “model” human interaction over the
Holocene period in a pancontinental region; we are seeking
to recast the whole multidisciplinary narrative. None of the
comments cause us to question our primary conclusions. Various incarnations of the farming/language dispersal hypothesis
fail to provide an historical narrative to account for the distribution of genes, languages, and material culture across ISEA
over the past 5,000 years. “Austronesian dispersal,” whether
fast, slow, or pulse paused, was not a single expansionary
process from Taiwan through ISEA and eventually onward to
Polynesia. We propose an alternative interpretative framework
that focuses on the social dynamics through which Taiwanese,
mainland Southeast Asia, and Melanesian influences, as well
as those from within ISEA, were transformed through time
and across social space.
—Tim Denham and Mark Donohue
References Cited
Adelaar, Alexander. 2005a. The Austronesian languages of Asia and
Madagascar: a historical perspective. In The Austronesian languages
of Asia and Madagascar. Alexander Adelaar and Nikolaus P. Himmelmann, eds. Pp. 1–42. London: Routledge.
———. 2005b. Malayo-Sumbawan. Oceanic Linguistics 44(2):
357–388.
Adelaar, K. Alexander. 1995. Borneo as a cross-roads for comparative
Austronesian linguistics. In The Austronesians: historical and comparative perspectives. Peter Bellwood, James J. Fox, and Darrell
Tryon, eds. Pp. 81–102. Canberra: Australian National University.
Adovasio, James M. 1999. Paradigm-death and gunfights. In Special
Report “Monte Verde revisited.” Scientific American Discovering
Archaeology 1(6):20. [SJO]
Allen, Jim. 1991. Introduction. In Report of the Lapita Homeland
Project. Jim Allen and Chris Gosden, eds. Pp. 1–8. Occasional
Papers in Prehistory 20. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University.
[RT]
Allen, Jim, and J. Peter White. 1989. The Lapita homeland: some
new data and an interpretation. Journal of the Polynesian Society
98(2):129–146. [RT]
Ammerman, Albert J., and Luigi Luca Cavalli-Sforza. 1984. The Neolithic transition and the genetics of populations in Europe. Princeton,
NJ: Princeton University Press.
Anderson, Atholl, and Sue O’Connor. 2008. Indo-Pacific migration
and colonization: introduction. Asian Perspectives 47(1):2–11.
Anshari, Gusti, A. Peter Kershaw, and Sander van der Kaars. 2001.
A late Pleistocene and Holocene pollen and charcoal record from
peat swamp forest, Lake Sentarum Wildlife Reserve, West Kalimantan, Indonesia. Palaeogeography, Palaeoclimatology, Palaeoecology 171(3–4):213–228.
Anshari, Gusti, A. Peter Kershaw, Sander van der Kaars, and Geraldine Jacobsen. 2004. Environmental change and peatland forest
dynamics in the Lake Sentarum area, West Kalimantan, Indonesia.
Journal of Quaternary Science 19(7):637–655.
Current Anthropology Volume 51, Number 2, April 2010
Anttila, Raimo. 1972. An introduction to historical and comparative
linguistics. New York: Macmillan.
Araho, Nick, Robin Torrence, and J. Peter White. 2002. Valuable and
useful: Mid-Holocene stemmed obsidian artefacts from West New
Britain, Papua New Guinea. Proceedings of the Prehistoric Society
68:61–81.
Arkhipova, Irina, and Matthew Meselson. 2000. Transposable elements in sexual and ancient asexual taxa. Proceedings of the National Academy of Sciences of the USA 97(26):14473–14477.
Baker, Philip, and Peter Mühlhäusler. 1996. The origins and diffusion
of Pidgin English in the Pacific. In Atlas of languages of intercultural
communication in the Pacific, Asia, and the Americas, vol. II.1.
Stephen A. Wurm, Peter Mühlhäusler, and Darrell T. Tryon, eds.
Pp. 551–594. Berlin: Mouton de Gruyter.
Bakker, Peter. 2000. Rapid language change: creolization, intertwining, convergence. In Time depth in historical linguistics. Colin Renfrew, April McMahon, and Larry Trask, eds. Pp. 585–620. Cambridge: McDonald Institute for Archaeological Research.
Ballard, Chris, Paula Brown, R. Michael Bourke, and Tracey Harwood, eds. 2005. The sweet potato in Oceania: a reappraisal. Oceania
Monograph 56. Sydney: University of Sydney; Pittsburgh: Department of Anthropology, University of Pittsburgh.
Barton, Huw, and Tim P. Denham. Forthcoming. Prehistoric vegeculture and social life in Island Southeast Asia and Melanesia. In
Why cultivate? anthropological and archaeological approaches to
foraging-farming transitions in Southeast Asia. Graeme Barker and
Monica Janowski, eds. Leiden: KITLV Press.
Bedford, Stuart. 2006. Pieces of the Vanuatu puzzle: archaeology of the
north, south and centre. Terra Australis 23. Canberra: Australian
National University.
Bedford, Stuart, and Matthew Spriggs. 2007. Birds on the rim: a
unique Lapita carinated vessel in its wider context. Archaeology in
Oceania 42(1):12–21. [MS]
Bellwood, Peter. 1984–1985. A hypothesis for Austronesian origins.
Asian Perspectives 26(1):107–117.
———. 1995. Austronesian prehistory in Southeast Asia: homeland,
expansion and transformation. In The Austronesians: historical and
comparative perspectives. Peter Bellwood, James J. Fox, and Darrell
Tryon, eds. Pp. 103–118. Canberra: Australian National University.
[VP]
———. 1997. Prehistory of the Indo-Malaysian Archipelago. Honolulu: University of Hawaii Press.
———. 2002. Farmers, foragers, languages, genes: the genesis of
agricultural societies. In Examining the farming-language dispersal
hypothesis. Peter Bellwood and Colin Renfrew, eds. Pp. 17–28.
Cambridge: McDonald Institute for Archaeological Research.
———. 2005. First farmers: the origins of agricultural societies. Oxford:
Blackwell.
———. 2007. Overview. In Review feature: First farmers: the origins
of agricultural societies. Cambridge Archaeological Journal 17:88–91.
———. 2009. Holocene population history in the Pacific region as
a model for world-wide food producer dispersals. Paper presented
at the Wenner-Gren Symposium “The Beginnings of Agriculture:
New Data, New Ideas,” Hacienda Temozón Sur, Mexico, March
6–13. [PB]
Bellwood, Peter, Geoffrey Chambers, Malcolm Ross, and Hsiao-chun
Hung. Forthcoming. Are “cultures” inherited? multidisciplinary
perspectives on the origins and migrations of Austronesian speaking peoples prior to 1000 BC. In Investigating archaeological cultures: material culture, variability and transmission. Ben Roberts
and Marc van der Linden, eds. Dordrecht: Springer. [PB]
Bellwood, Peter, and Eusebio Dizon. 2005. The Batanes Archaeological Project and the “out of Taiwan” hypothesis for Austronesian
dispersal. Journal of Austronesian Studies 1(1):1–31.
———. 2008. Austronesian cultural origins: out of Taiwan, via the
Batanes Islands, and onwards to western Polynesia. In Past human
Donohue and Denham Farming and Language in Island Southeast Asia
migrations in East Asia: matching archaeology, linguistics and genetics. Alicia Sanchez-Mazas, Roger Blench, Malcolm D. Ross, Ilia
Peiros, and Marie Lin, eds. Pp. 23–39. London: Routledge. [PB]
Bellwood, Peter, James J. Fox, and Darrell Tryon. 1995. The Austronesians in history: common origins and diverse transformations.
In The Austronesians: historical and comparative perspectives. Peter
Bellwood, James J. Fox, and Darrell Tryon, eds. Pp. 1–16. Canberra:
Australian National University.
Bellwood, Peter, and Colin Renfrew, eds. 2002. Examining the farming-language dispersal hypothesis. Cambridge: McDonald Institute
for Archaeological Research.
Bellwood, Peter, Janelle Stevenson, Eusebio Dizon, Armand Mijares,
Gay Lacsina, and Emil Robles. 2008. Where are the Neolithic landscapes of Ilocos Norte? Hukay (Manila) 13:25–38. [PB]
Best, Simon. 2002. Lapita: a view from the east. Auckland: New Zealand Archaeological Association.
Blust, Robert. 1981. Variation in retention rate among Austronesian
languages. Paper presented at the Third International Conference
on Austronesian Linguistics, Denpasar, Bali, Indonesia, January
1–11.
———. 1984–1985. The Austronesian homeland: a linguistic perspective. Asian Perspectives 26(1):45–67.
———. 1991. The Greater Central Philippines hypothesis. Oceanic
Linguistics 30(2):73–129.
———. 1993. Central and Central-Eastern Malayo-Polynesian. Oceanic Linguistics 32(2):241–293.
———. 1995. The prehistory of the Austronesian-speaking peoples:
a view from language. Journal of World Prehistory 9(4):453–510.
———. 2000a. Chamorro historical phonology. Oceanic Linguistics
39(1):83–122.
———. 2000b. Why lexicostatistics doesn’t work. In Time depth in
historical linguistics. Colin Renfrew, April McMahon, and Larry
Trask, eds. Pp. 311–331. Cambridge: McDonald Institute for Archaeological Research.
———. 2001. Malayo-Polynesian: new stones in the wall. Oceanic
Linguistics 40(1):151–155.
———. 2005. The linguistic macrohistory of the Philippines: some
speculations. In Current issues in Philippine linguistics and anthropology: parangal kay Lawrence A. Reid. Hsiu-chuan Liao and Carl
R. Galvez Rubino, eds. Pp. 31–68. Manila: Linguistic Society of
the Philippines and SIL Philippines.
Box, George E. P. 1979. Robustness in the strategy of scientific model
building. In Robustness in statistics. Robert L. Launer and Graham
N. Wilkinson, eds. Pp. 201–236. New York: Academic Press. [MPC/
JSL]
Brandes, Jan A. 1884. Bijdragen tot de Vergelijkende Klankleer der
Westersche afdeeling van de Maleis-Polynesische taalfamilie. Leiden:
van de Weijer.
Bulbeck, David. 2008. An integrated perspective on the Austronesian
diaspora: the switch from cereal agriculture to maritime foraging
in the colonisation of Island Southeast Asia. Australian Archaeology
67:31–51.
Bylmer, Hendricus J. T. 1943. Newer data on blood groups from the
Indian Archipelago, and their classification in the triangle of
Streng. Genetica 23:257–276. [WM]
Campbell, Lyle. 2002. What drives linguistic diversification and language spread? In Examining the farming/language dispersal hypothesis. Peter Bellwood and Colin Renfrew, eds. Pp. 49–63. Cambridge: McDonald Institute for Archaeological Research.
Capell, Arthur. 1975. The “West Papuan phylum”: general, and Timor
and areas further west. In Papuan languages and the New Guinea
linguistic scene, vol. 1 of New Guinea area languages and language
study. Stephen A. Wurm, ed. Pp. 667–716. Canberra: Pacific
Linguistics.
Capelli, Cristian, James F. Wilson, Martin Richards, Michael P. H.
Stumpf, Fiona Gratrix, Stephen Oppenheimer, Peter Underhill,
251
Vincenzo L. Pascali, Tsang-Ming Ko, and David B. Goldstein. 2001.
A predominantly indigenous paternal heritage for the Austronesian-speaking peoples of insular Southeast Asia and Oceania.
American Journal of Human Genetics 68(2):432–443.
Carreel, Françoise, Diego González de León, Pierre Lagoda, Claire
Lanaud, Christophe Jenny, Jean-Pierre Horry, and Hugues Tézenas
du Montcel. 2002. Ascertaining maternal and paternal lineage
within Musa chloroplast and mitochondrial DNA RFLP analyses.
Genome 45(4):679–692.
Cavalli-Sforza, Luigi Luca, Alberto Piazza, Paolo Menozzi, and
Joanna Mountain. 1988. Reconstruction of human evolution:
bringing together genetic, archaeological, and linguistic data. Proceedings of the National Academy of Sciences of the USA 85(16):
6002–6006.
Chang, Kwang-chih, George W. Grace, and Wilhelm G. Solheim II.
1964. Movement of the Malayo-Polynesians: 1500 B.C. to A.D.
500. Current Anthropology 5(5):359–406. [WM]
Chow, Rachel A., Jose L. Caeiro, Shu-Juo Chen, Ralph L. GarciaBertrand, and Rene J. Herrera. 2005. Genetic characterization of
four Austronesian-speaking populations. Journal of Human Genetics 50(11):550–559.
Connolly, Bob, and Robin Anderson. 1987. First contact: New
Guinea’s highlanders encounter the outside world. New York: Viking.
Cox, Murray P. 2005. Indonesian mitochondrial DNA and its opposition to Pleistocene era origin of Proto-Polynesians in Island
Southeast Asia. Human Biology 77(2):179–188. [PB]
———. 2008. Accuracy of molecular clock dating with the Rho
statistic: deviations from coalescent expectations under a range of
demographic models. Human Biology 80(4):335–357. [PB]
Dagan, Tal, and William Martin. 2007. Ancestral genome sizes specify
the minimum rate of lateral gene transfer during prokaryote evolution. Proceedings of the National Academy of Sciences of the USA
104(3):870–875.
Dam, Rien A. C., Jennie Fluin, Papay Suparan, and Sander van der
Kaars. 2001. Palaeoenvironmental developments in the Lake Tondano area (N. Sulawesi, Indonesia) since 33,000 yr BP. Palaeogeography, Palaeoclimatology, Palaeoecology 171(3–4):147–183.
De Langhe, Edmond, and Pierre de Maret. 1999. Tracking the banana:
its significance in early agriculture. In The prehistory of food: appetites for change. Chris Gosden and Jon Hather, eds. Pp. 377–396.
London: Routledge.
Denham, Tim. 2004. The roots of agriculture and arboriculture in
New Guinea: looking beyond Austronesian expansion, Neolithic
packages and indigenous origins. World Archaeology 36(4):610–
620.
———. 2005. Envisaging early agriculture in the highlands of New
Guinea: landscapes, plants and practices. World Archaeology 37(2):
290–306.
Denham, Tim, Edmond De Langhe, and Luc Vrydaghs, eds. 2009.
History of banana domestication. Special issue, Ethnobotany Research and Applications 7:164–380.
Denham, Tim, and Mark Donohue. 2009. Pre-Austronesian dispersal
of banana cultivars west from New Guinea: linguistic relics from
eastern Indonesia. Archaeology in Oceania 44:18–28.
Denham, Tim, Simon G. Haberle, Carol Lentfer, Richard Fullagar,
Judith Field, Michael Therin, Nick Porch, and Barbara Winsborough. 2003. Origins of agriculture at Kuk Swamp in the highlands
of New Guinea. Science 301:189–193.
Diamond, Jared. 2001. Polynesian origins: slow boat to Melanesia?
(reply) Nature 410(6825):167.
Diamond, Jared, and Peter Bellwood. 2003. Farmers and their languages: the first expansions. Science 300:597–603.
Dobney, Keith, Thomas Cucchi, and Gregor Larson. 2008. The pigs
of Island Southeast Asia and the Pacific: new evidence for taxonomic status and human-mediated dispersal. Asian Perspectives 47:
59–74.
252
Donohue, Mark. 1999. A grammar of Tukang Besi. Berlin: Mouton
de Gruyter.
———. 2002. Tobati. In The Oceanic languages. John Lynch, Malcolm
Ross, and Terry Crowley, eds. Pp. 186–203. London: Curzon.
———. 2004. Typology and linguistic areas. Oceanic Linguistics 43(1):
221–239.
———. 2005. Word order in New Guinea: dispelling a myth. Oceanic
Linguistics 44(2):527–536.
———. 2007a. The Papuan language of Tambora. Oceanic Linguistics
46(2):520–537.
———. 2007b. Word order in Austronesian from north to south
and west to east. Linguistic Typology 11(2):351–393.
Donohue, Mark, and Charles E. Grimes. 2008. Yet more on the
position of the languages of eastern Indonesia and East Timor.
Oceanic Linguistics 47(1):115–158.
Doran, Edwin, Jr. 1981. Wangka: Austronesian canoe origins. College
Station: Texas A&M University Press. [WM]
Dryer, Matthew S. 1992. The Greenbergian word order correlations.
Language 68:81–138.
———. 2005a. Order of adjective and noun. In World atlas of language structures. Martin Haspelmath, Matthew S. Dryer, David Gil,
and Bernard Comrie, eds. Pp. 354–357. Oxford: Oxford University
Press.
———. 2005b. Order of demonstrative and noun. In World atlas of
language structures. Martin Haspelmath, Matthew S. Dryer, David
Gil, and Bernard Comrie, eds. Pp. 358–361. Oxford: Oxford University Press.
———. 2005c. Order of genitive and noun. In World atlas of language
structures. Martin Haspelmath, Matthew S. Dryer, David Gil, and
Bernard Comrie, eds. Pp. 350–353. Oxford: Oxford University
Press.
———. 2005d. Order of numeral and noun. In World atlas of language structures. Martin Haspelmath, Matthew S. Dryer, David Gil,
and Bernard Comrie, eds. Pp. 362–365. Oxford: Oxford University
Press.
———. 2005e. Order of relative clause and noun. In World atlas of
language structures. Martin Haspelmath, Matthew S. Dryer, David
Gil, and Bernard Comrie, eds. Pp. 366–369. Oxford: Oxford University Press.
———. 2005f. Order of subject and verb. In World atlas of language
structures. Martin Haspelmath, Matthew S. Dryer, David Gil, and
Bernard Comrie, eds. Pp. 334–337. Oxford: Oxford University
Press.
Dunn, Frederick L. 1970. Cultural evolution in the Late Pleistocene
and Holocene of Southeast Asia. American Anthropologist 72(5):
1041–1054. [WM]
Dutton, Tom. 1995. Language contact and change in Melanesia. In
The Austronesians: historical and comparative perspectives. Peter
Bellwood, James J. Fox, and Darrell Tryon, eds. Pp. 207–228. Canberra: Australian National University.
Dyen, Isidore. 1965. A lexicostatistical classification of the Austronesian
languages. International Journal of American Linguistics Memoir
19. Baltimore: Waverly.
Ehret, Christopher. 1998. An African classical age: eastern and southern
Africa in world history, 1000 BC to AD 400. Charlottesville: University of Virginia Press.
Excoffier, Laurent, and Nicolas Ray. 2008. Surfing during population
expansions promotes genetic revolutions and structuration. Trends
in Ecology & Evolution 23(7):347–351.
Flenley, John R. 1988. Palynological evidence for land use changes
in South-East Asia. Journal of Biogeography 15(1):185–197.
Fredericksen, Clayton F. K., Matthew Spriggs, and Wallace Ambrose.
1993. Pamwak rockshelter: a Pleistocene site on Manus Island,
Papua New Guinea. In Sahul in review: Pleistocene archaeology in
Australia, New Guinea and Island Melanesia. Michael A. Smith,
Matthew Spriggs, and Barry Fankhauser, eds. Pp. 144–152. Oc-
Current Anthropology Volume 51, Number 2, April 2010
casional Papers in Prehistory 24. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National
University.
Friedlaender, Jonathan S., Françoise R. Friedlaender, Floyd A. Reed,
Kenneth K. Kidd, Judith R. Kidd, Geoffrey K. Chambers, Rodney
A. Lea, et al. 2008. The genetic structure of Pacific Islanders. PLoS
Genetics 4(1):e19.
Fullagar, Richard, Judith Field, Tim Denham, and Carol Lentfer.
2006. Early and mid Holocene tool-use and processing of taro
(Colocasia esculenta), yam (Dioscorea sp.) and other plants at Kuk
Swamp in the highlands of Papua New Guinea. Journal of Archaeological Science 33(5):595–614.
Giles, Eugene, Eugene Ogan, and Arthur G. Steinberg. 1965. Gammaglobulin factors (Gm and Inv) in New Guinea: anthropological
significance. Science 150:1158–1160. [MPC/JSL]
Glover, Ian C., and Charles F. W. Higham. 1996. New evidence for
early rice cultivation in South, Southeast and East Asia. In The
origins and spread of agriculture and pastoralism in Eurasia. David
R. Harris, ed. Pp. 413–441. London: University College London
Press.
Golson, Jack. 2005. The middle reaches of New Guinea prehistory.
In Papuan pasts: cultural, linguistic and biological histories of
Papuan-speaking peoples. Andrew Pawley, Robert Attenborough,
Jack Golson, and Robin Hide, eds. Pp. 451–491. Pacific Linguistics
572. Canberra: Research School of Pacific and Asian Studies, Australian National University.
Gould, Peter. 1970. Is statistix inferens the geographical name for a
wild goose? Economic Geography 46(suppl.):439–448.
Gray, Russell D., Alexei J. Drummond, and Simon J. Greenhill. 2009.
Language phylogenies reveal expansion pulses and pauses in Pacific
settlement. Science 323:479–483. [SJO]
Gray, Russell D., and Fiona M. Jordan. 2000. Language trees support
the express-train sequence of Austronesian expansion. Nature
405(6790):1052–1055.
Green, Roger. 2000. Lapita and the cultural model for intrusion. In
Australian archaeologist: collected papers in honour of Jim Allen.
Atholl Anderson and Tim Murray, eds. Pp. 372–392. Canberra:
Australian National University.
Greenhill, Simon J., Robert Blust, and Russell D. Gray. 2008. The
Austronesian basic vocabulary database: from bioinformatics to
lexomics. Evolutionary Bioinformatics 4:271–283.
Greenhill, Simon J., and Russell D. Gray. 2005. Testing population
dispersal hypotheses: Pacific settlement, phylogenetic trees and
Austronesian languages. In The evolution of cultural diversity: phylogenetic approaches. Ruth Mace, Clare J. Holden, and Stephen
Shennan, eds. Pp. 31–52. London: University College London
Press. [MPC/JSL]
Grist, Donald H. 1959. Rice. 3rd edition. London: Longmans. [WM]
Grivet, Laurent, Christian Daniels, Jean-Christophe Glaszmann, and
Angelique D’Hont. 2004. A review of recent molecular genetics
evidence for sugarcane evolution and domestication. Ethnobotany
Research and Applications 2:9–17.
Haberle, Simon G. 2003. The emergence of an agricultural landscape
in the highlands of New Guinea. Archaeology in Oceania 38:
149–158.
Haberle, Simon G., Geoff S. Hope, and Sander van der Kaars. 2001.
Biomass burning in Indonesia and Papua New Guinea: natural
and human induced fire events in the fossil record. Palaeogeography, Palaeoclimatology, Palaeoecology 171(3–4):259–268.
Heggarty, Paul. 2007. Linguistics for archaeologists: principles, methods and the case of the Incas. Cambridge Archaeological Journal
17(3):311–340.
Heine-Geldern, Robert. 1932. Urheimat und früheste Wanderunngen
der Austronesier. Anthropos 27:543–619. [WM]
Hill, Adrian V. S., and Susan W. Serjeantson, eds. 1989. The colonization of the Pacific: a genetic trail. Oxford: Clarendon.
Donohue and Denham Farming and Language in Island Southeast Asia
Hill, Catherine, Pedro Soares, Maru Mormina, Vincent Macaulay,
Dougie Clarke, Petya B. Blumbach, Matthieu Vizuete-Forster, et
al. 2007. A mitochondrial stratigraphy for Island Southeast Asia.
American Journal of Human Genetics 80(1):29–43.
Ho, Simon Y. W., Beth Shapiro, Matthew J. Phillips, Alan Cooper,
and Alexei J. Drummond. 2007. Evidence for time dependency of
molecular rate estimates. Systematic Biology 56(3):515–522. [PB]
Hongo, Hitomi, Naotaka Ishiguro, Takuma Watanobe, Nobuo Shigehara, Tomoko Anezaki, Vu The Long, Dang Vu Binh, Nguyen
Trong Tien, and Nguyen Huu Nam. 2002. Variation in mitochondrial DNA of Vietnamese pigs: relationships with Asian domestic pigs and Ryukyu wild boars. Zoological Science 19:
1329–1335.
Horridge, Adrian. 1995. The Austronesian conquest of the sea: upwind. In The Austronesians: historical and comparative perspectives.
Peter Bellwood, James J. Fox, and Darrell Tryon, eds. Pp. 143–160.
Canberra: Australian National University.
Hughes, Ian. 1977. New Guinea Stone Age trade: the geography and
ecology of traffic in the interior. Terra Australis 3. Canberra: Australian National University.
Hung, Hsiao-chun. 2008. Migration and cultural interaction in
southern coastal China, Taiwan and the northern Philippines, 3000
BC to AD 1. PhD thesis, Australian National University, Canberra.
[PB]
Hung, Hsiao-chun, Yoshiyuki Iizuka, Peter Bellwood, Kim Dung
Nguyen, Bérénice Bellina, Praon Silapanth, Eusebio Dizon, Rey
Santiago, Ipoi Datan, and Jonathan H. Manton. 2007. Ancient
jades map 3000 years of prehistoric exchange in Southeast Asia.
Proceedings of the National Academy of Sciences of the USA 104(50):
19745–19750.
Hunt, Christopher O., and Garry Rushworth. 2005. Cultivation and
human impact at 6000 cal yr B.P. in tropical lowland forest at
Niah, Sarawak, Malaysian Borneo. Quaternary Research 64(3):
460–468.
Hurles, Matthew E. 2002. Can the hypothesis of language/agriculture
co-dispersal be tested with archaeogenetics? In Examining the farming/language dispersal hypothesis. Peter Bellwood and Colin Renfrew, eds. Pp. 299–309. Cambridge: McDonald Institute of Archaeological Research.
Hurles, Matthew E., Bryan C. Sykes, Mark A. Jobling, and Peter
Forster. 2005. The dual origin of the Malagasy in Island Southeast
Asia and East Africa: evidence from maternal and paternal lineages.
American Journal of Human Genetics 76(5):894–901.
Irwin, Geoffrey. 2008. Pacific seascapes, canoe performance, and a
review of Lapita voyaging with regard to theories of migration.
Asian Perspectives 47(1):12–27.
Jones, Alan A. 1998. Towards a lexicogrammar of Mekeo. Canberra:
Pacific Linguistics.
Joseph, Brian D., and Richard D. Janda, eds. 2003. The handbook of
historical linguistics. London: Blackwell.
Kayser, Manfred, Ying Choi, Mannis van Oven, Stefano Mona, Silke
Brauer, Ronald J. Trent, Dagwin Suarkia, Wulf Schiefenhövel, and
Mark Stoneking. 2008. The impact of the Austronesian expansion:
evidence from mtDNA and Y chromosome diversity in the Admiralty Islands of Melanesia. Molecular Biology and Evolution
25(7):1362–1374.
Kennedy, Jean. 2008. Pacific bananas: complex origins, multiple dispersals? Asian Perspectives 47(1):75–94.
Kennedy, Jean, and William Clarke. 2004. Cultivated landscapes of
the southwest Pacific. Research Management in the Asia-Pacific
(RMAP) Working Paper 50. Canberra: RMAP, Australian National
University.
Kern, Hendrik. 1889. Taalkundige gegevens ter bepaling van het
stamland der Maleisch-Polynesische Volken. Verslagen en Mededeelingen der Koninklijke Akademie van Wetenschappen, Afdeeling
Letterkunde, 3e reeks, 6:270–287. [WM]
253
Kirch, Patrick Vinton. 2000. On the road of the winds: an archaeological history of the Pacific islands before European contact. Berkeley:
University of California Press.
Kjær, Anders, Anders S. Barfod, Conny B. Asmussen, and Ole Seberg.
2004. Investigation of genetic and morphological variation in the
sago palm (Metroxylon sagu; Arecaceae) in Papua New Guinea.
Annals of Botany 94(1):109–117.
Krupa, Viktor. 1973. Polynesian languages: a survey of research. The
Hague: Mouton.
Lansing, J. Stephen, Murray P. Cox, Sean S. Downey, Brandon M.
Gabler, Brian Hallmark, Tatiana M. Karafet, Peter Norquest, et al.
2007. Coevolution of languages and genes on the island of Sumba,
eastern Indonesia. Proceedings of the National Academy of Sciences
of the USA 104(41):16022–16026.
Larson, Gregor, Thomas Cucchi, Masakatsu Fujita, Elizabeth Matisoo-Smith, Judith Robins, Atholl Anderson, Barry Rolett, et al.
2007. Phylogeny and ancient DNA of Sus provides insights into
Neolithic expansion in Island Southeast Asia and Oceania. Proceedings of the National Academy of Sciences of the USA 104(12):
4834–4839.
Laycock, Donald C. 1982. Tok Pisin: a Melanesian solution to the
problem of Melanesian linguistic diversity. In Melanesia beyond
diversity. Ronald J. May and Hank Nelson, eds. Pp. 263–272. Canberra: Research School of Pacific Studies, Australian National
University.
Leahy, Michael J. 1991. Explorations into highland New Guinea:
1930–1935. Douglas E. Jones, ed. Tuscaloosa: University of Alabama Press.
Lebot, Vincent. 1999. Biomolecular evidence for plant domestication
in Sahul. Genetic Resources and Crop Evolution 46(6):619–628.
Lebot, Vincent, Made Sri Prana, Nelleke Kreike, Herman van Heck,
José Pardales, Tom Okpul, T. Gendua, et al. 2004. Characterisation
of taro (Colocasia esculenta (L.) Schott) genetic resources in Southeast Asia and Oceania. Genetic Resources and Crop Evolution 51(4):
381–392.
Lejju, B. Julius, Peter Robertshaw, and David Taylor. 2006. Africa’s
earliest bananas? Journal of Archaeological Science 33(1):102–113.
Lewis, Paul M., ed. 2009. Ethnologue: languages of the world. 16th
edition. Dallas: SIL International. http://www.ethnologue.com/.
Li, Hui, Bo Wen, Shu-Juo Chen, Bing Su, Patcharin Pramoonjago,
Yangfan Liu, Shangling Pan, et al. 2008. Paternal genetic affinity
between western Austronesians and Daic populations. BMC Evolutionary Biology 8:146.
Lichtenberk, Frantisek. 1984. A grammar of Manam. Honolulu: University of Hawaii Press.
Lin, Marie, Chen-Chung Chu, Richard E. Broadberry, Lung-Chih
Yu, Jun-Hun Loo, and Jean A. Trejaut. 2005. Genetic diversity of
Taiwan’s indigenous peoples: possible relationship with insular
Southeast Asia. In The peopling of East Asia: putting together archaeology, linguistics and genetics. Laurent Sagart, Roger Blench,
and Alicia Sanchez-Mazas, eds. Pp. 230–247. London: RoutledgeCurzon. [WM]
Liu, Yi-Ping Liu, Qing Zhu, and Yong-Gang Yao. 2006. Genetic relationship of Chinese and Japanese gamecocks revealed by mtDNA
sequence variation. Biochemical Genetics 44(1–2):18–28.
Mahdi, Waruno. 1988. Nachtrag. In Morphophonologische Besonderheiten und historische Phonologie des Malagasy. Waruno Mahdi, ed.
Pp. 347–423. Veröffentlichungen des Seminars für Indonesische
und Südseesprachen der Universität Hamburg 20. Berlin: Reimer.
[WM]
———. 1994a. Some Austronesian maverick protoforms with
culture-historical implications. I. Oceanic Linguistics 33(1):167–
229.
———. 1994b. Some Austronesian maverick protoforms with
culture-historical implications. II. Oceanic Linguistics 33(2):431–
490.
254
———. 1999a. The dispersal of Austronesian boat forms in the
Indian Ocean. In Archaeology and language III: artefacts, languages
and texts. Roger Blench and Matthew Spriggs, eds. Pp. 144–179.
London: Routledge. [WM]
———. 1999b. Linguistic and philological data towards a chronology
of Austronesian activity in India and Sri Lanka. In Archaeology
and language IV: language change and cultural transformation.
Roger Blench and Matthew Spriggs, eds. Pp. 160–240. London:
Routledge. [WM]
Malapa, Roger, Gemma Arnau, Jean-Louis Noyer, and Vincent Lebot.
2005. Genetic diversity of the greater yam (Dioscorea alata L.) and
relatedness to D. nummularia Lam. and D. transversa Br. as revealed with AFLP markers. Genetic Resources and Crop Evolution
52(7):919–929.
Mauss, M. 1990. The gift: the form and reason for exchange in archaic
societies. W. D. Halls, trans. London: Norton.
Mbida, Christophe M., Hugues Doutrelepont, Luc Vrydaghs, Rony
L. Swennen, Rudy J. Swennen, Hans Beeckman, Edmond De
Langhe, and Pierre de Maret. 2001. First archaeological evidence
of banana cultivation in central Africa during the third millennium
before present. Vegetation History and Archaeobotany 10(1):1–6.
———. 2004. Yes, there were bananas in Cameroon more than 2000
years ago. InfoMusa 13(1):40–42.
Meacham, William. 1984–1985. On the improbability of Austronesian origins in South China. Asian Perspectives 26(1):89–106.
———. 1995. Austronesian origins and the peopling of Taiwan. In
Austronesian studies relating to Taiwan. Paul Jen-kuei Li, Chenghwa Tsang, Ying-kuei Huang, Dah-an Ho, and Chiu-yu Tseng, eds.
Pp. 227–294. Taipei: Academica Sinica. [VP]
Mona, Stefano, Katharina E. Grunz, Silke Brauer, Brigitte Pakendorf,
Loredana Castrı̀, Herawati Sudoyo, Sangkot Marzuki, et al. 2009.
Genetic admixture history of eastern Indonesia as revealed by Ychromosome and mitochondrial DNA analysis. Molecular Biology
and Evolution 26(8):1865–1877.
Mufwene, Salikoko S. 2001. The ecology of language evolution. Cambridge: Cambridge University Press.
Mühlhäusler, Peter. 1996. Development and spread of Tok Pisin 1920
to present. In Atlas of languages of intercultural communication in
the Pacific, Asia, and the Americas, vol. 1. Stephen A. Wurm, Peter
Mühlhäusler, and Darrell T. Tryon, eds. Map 50. Berlin: Walter de
Gruyter.
Needham, Joseph, Wang Ling, and Lu Gwei-djen. 1971. Civil engineering and nautics, vol. IV:3 of Science and civilisation in China.
Cambridge: Cambridge University Press. [WM]
Nettle, Daniel. 2000. Linguistic diversity, population spread and time
depth. In Time depth in historical linguistics. Colin Renfrew, April
McMahon, and Larry Trask, eds. Pp. 665–677. Cambridge:
McDonald Institute for Archaeological Research.
Nichols, Johanna. 1997. Modeling ancient population structures and
movement in linguistics. Annual Review of Anthropology 26:
359–384.
Nichols, Johanna, and Balthasar Bickel. 2005. Possessive classification. In World atlas of language structures. Martin Haspelmath,
Matthew S. Dryer, David Gil, and Bernard Comrie, eds. Pp.
242–245. Oxford: Oxford University Press.
Nichols, Johanna, and David A. Peterson. 1998. Amerind personal
pronouns: a reply to Campbell. Language 74:605–614.
O’Connor, Sue. 2007. New evidence from East Timor contributes to
our understanding of earliest modern human colonisation east of
the Sunda Shelf. Antiquity 81:523–535.
O’Connor, Sue, and Peter Veth. 2005. Early Holocene shell fish hooks
from Lene Hara Cave, East Timor establish complex fishing technology was in use in Island South East Asia five thousand years
before Austronesian settlement. Antiquity 79:249–256.
Oppenheimer, Stephen. 2004. The “express train from Taiwan to
Current Anthropology Volume 51, Number 2, April 2010
Polynesia”: on the congruence of proxy lines of evidence. World
Archaeology 36(4):591–600.
———. 2006. The “Austronesian” story and farming-language dispersals: caveats on timing and independence in proxy lines of
evidence from the Indo-European model. In Uncovering Southeast
Asia’s past: selected papers from the 10th International Conference
of the European Association of Southeast Asian Archaeologists. Elizabeth A. Bacus, Ian C. Glover, and Vincent C. Piggott, eds. Pp.
65–73. Honolulu: University of Hawaii Press. [SJO]
Oppenheimer, Stephen, and Martin Richards. 2001a. Fast trains, slow
boats, and the ancestry of the Polynesian islanders. Science Progress
84(3):157–181. [SJO]
———. 2001b. Polynesian origins: slow boat to Melanesia? Nature
410(6825):166–167.
———. 2002. Polynesians: devolved Taiwanese rice farmers or Wallacean maritime traders with fishing, foraging and horticultural
skills? In Examining the farming-language dispersal hypothesis. Peter
Bellwood and Colin Renfrew, eds. Pp. 287–297. Cambridge:
McDonald Institute for Archaeological Research.
Pawley, Andrew. 1999. Chasing rainbows: implications of the rapid
dispersal of Austronesian languages for subgrouping and reconstruction. In Selected papers from the Eighth International Conference on Austronesian Linguistics. Elizabeth Zeitoun and Paul Jenkuei Li, eds. Pp. 95–138. Taipei: Institute of Linguistics, Academia
Sinica. [PB]
———. 2002. The Austronesian dispersal: languages, technologies
and people. In Examining the farming-language dispersal hypothesis.
Peter Bellwood and Colin Renfrew, eds. Pp. 251–273. Cambridge:
McDonald Institute for Archaeological Research.
———. 2006. Explaining the aberrant Austronesian languages of
southeast Melanesia: 150 years of debate. Journal of the Polynesian
Society 116:213–256.
———. 2007. The origins of early Lapita culture: the testimony of
historical linguistics. In Oceanic explorations: Lapita and western
Pacific settlement. Stuart Bedford, Christophe Sand, and Sean P.
Connaughton, eds. Pp. 17–49. Canberra: Australian National
University.
Pawley, Andrew, and Roger Green. 1973. Dating the dispersal of the
Oceanic languages. Oceanic Linguistics 12:1–67.
Paz, Victor. 2002. Island Southeast Asia: spread or friction zone? In
Examining the farming/language dispersal hypothesis. Peter Bellwood and Colin Renfrew, eds. Pp. 275–285. Cambridge: McDonald
Institute for Archaeological Research.
———. 2005. Rock shelters, caves, and archaeobotany in Island
Southeast Asia. Asian Perspectives 44(1):107–118.
Peiros, Ilia. 2000. Family diversity and time depth. In Time depth in
historical linguistics. Colin Renfrew, April McMahon, and Larry
Trask, eds. Pp. 75–105. Cambridge: McDonald Institute for Archaeological Research.
Pelliot, Paul. 1925. Quelques textes chinois concernant l’Indochine
hindouisée. In Études asiatiques, publiées à l’occasion du vingtcinquième anniversaire de l’École Française de l’Extrême-Orient, vol.
2. Pp. 243–263. Publication de l’École Française de l’ExtrêmeOrient (EFEO) 20. Hanoi: EFEO. [WM]
Piper, Philip J., Hsiao-chun Hung, Fredeliza Z. Campos, Peter Bellwood, and Rey Santiago. 2009. A 4000 year-old introduction of
domestic pigs into the Philippine Archipelago: implications for
understanding routes of human migration through Island Southeast Asia and Wallacea. Antiquity 83(3):687–695. [PB]
Reid, Anthony. 1995. Continuity and change in the Austronesian
transition to Islam and Christianity. In The Austronesians: historical
and comparative perspectives. Peter Bellwood, James J. Fox, and
Darrell Tryon, eds. Pp. 333–350. Canberra: Australian National
University.
Reid, Lawrence A. 1994. Possible non-Austronesian lexical elements
in Philippine Negrito languages. Oceanic Linguistics 33(1):37–72.
Donohue and Denham Farming and Language in Island Southeast Asia
Reland, Adriaan. 1708. Hadriani Relandi dissertationum miscellanearum pars tertia et ultima dissertatio de linguis insularum quarundam orientalium. Trajecti ad Rhenum (Utrecht): Broedelet.
Renfrew, Colin. 1987. Archaeology and language. London: Jonathan
Cape.
———. 1993. Trade beyond the material. In Trade and exchange in
prehistoric Europe. Christopher Scarre and Frances Healy, eds. Pp.
5–16. London: Oxbow. [RT]
Rolett, Barry V., Wei-chun Chen, and John M. Sinton. 2000. Taiwan,
Neolithic seafaring and Austronesian origins. Antiquity 74:54–61.
Ross, Malcolm. 1994. Areal phonological features in north central
New Ireland. In Language contact and change in the Austronesian
world. Tom Dutton and Darrell T. Tryon, eds. Pp. 551–572. Berlin:
Mouton de Gruyter.
———. 1995. Some current issues in Austronesian linguistics. In
Comparative Austronesian dictionary: an introduction to Austronesian studies. Darrell T. Tryon, ed. Pt. 1, fasc. 1. Pp. 45–120. Berlin:
Mouton de Gruyter.
———. 2005. The Batanic languages in relation to the early history
of the Malayo-Polynesian family. Journal of Austronesian Studies
1(2):1–24.
———. 2008. The integrity of the Austronesian language family:
from Taiwan to Oceania. In Past human migrations in East Asia:
matching archaeology, linguistics and genetics. Alicia Sanchez-Mazas,
Roger Blench, Malcolm D. Ross, Ilia Peiros, and Marie Lin, eds.
Pp. 161–181. London: Routledge. [PB]
———. 2009. Proto-Austronesian verbal morphology: a reappraisal.
In Austronesian historical linguistics and culture history: a festschrift
for Robert Blust. Alexander Adelaar and Andrew Pawley, eds. Pp.
295–326. Canberra: Pacific Linguistics.
Ross, Malcolm, Andrew Pawley, and Meredith Osmond. 2008. The
lexicon of Proto-Oceanic: the culture and environment of ancestral
Oceanic society. 3: Plants. Canberra: Pacific Linguistics.
Sagart, Laurent. 2005. Tai-Kadai as a subgroup of Austronesian. In
The peopling of East Asia: putting together archaeology, linguistics
and genetics. Laurent Sagart, Roger Blench, and Alicia SanchezMazas, eds. Pp. 177–181. London: RoutledgeCurzon. [WM]
Schachter, Paul, and Fe T. Otanes. 1972. Tagalog reference grammar.
Los Angeles: University of California Press.
Schmidt, Wilhelm. 1906. Die Mon-Khmer-Völker, ein Bindeglied
zwischen Völkern Zentralasiens und Austronesiens. Archiv für Anthropologie 33:59–106. [WM]
Sellato, Bernard. 1994. Nomads of the Borneo rainforest. Stephanie
Morgan, trans. Honolulu: University of Hawaii Press. [WM]
Sercombe, Peter G., and Bernard Sellato, eds. 2007. Beyond the green
myth: Borneo’s hunter-gatherers in the twenty-first century. Copenhagen: NIAS. [WM]
Shutler, Richard, Jr. 1999. The relationship of red-slipped and limeimpressed pottery of the southern Philippines to that of Micronesia
and the Lapita of Oceania. In Le Pacifique de 5000 à 2000 avant
le présent/The Pacific from 5000 to 2000 B.P. Jean-Christophe Galipaud and Ian Lilley, eds. Pp. 521–529. Paris: Institut de Recherche
pour le Développement.
Shutler, Richard, Jr., and Jeffery Marck. 1975. On the dispersal of
the Austronesian horticulturalists. Archaeology and Physical Anthropology in Oceania 10:81–113.
Silzer, Peter. 1983. Ambai: an Austronesian language of eastern Indonesia. PhD thesis, Australian National University, Canberra.
Soares, Pedro, Jean Alain Trejaut, Jun-Hun Loo, Catherine Hill, Maru
Mormina, Chien-Liang Lee, Yao-Ming Chen, et al. 2008. Climate
change and postglacial human dispersals in Southeast Asia. Molecular Biology and Evolution 25(6):1209–1218.
Sokal, Robert R. 1988. Genetic, geographic, and linguistic distances
in Europe. Proceedings of the National Academy of Sciences of the
USA 85(6):1722–1726.
255
Solheim, Wilhelm G., II. 1964. Pottery and the Malayo-Polynesians.
Current Anthropology 5(5):360–406.
———. 1969. Reworking Southeast Asian prehistory. Paideuma 15:
125–139. [VP]
———. 1980. Neue Befunde zur späten Prähistorie Südostasiens und
ihre Interpretation. Saeculum 31(3–4):275–317, 319–344. [WM]
———. 1984–1985. The Nusantao hypothesis: the origin and spread
of Austronesian speakers. Asian Perspectives 26(1):77–88.
———. 1990. Thoughts on land and sea peoples in Southeast Asia
and their possible relationships to initial settlement of Micronesia.
Micronesica Supplement 2:241–246. [VP]
———. 2006. Archaeology and culture in Southeast Asia: unravelling
the Nusantao. Quezon City: University of Philippines Press. [SJO,
VP]
Spriggs, Matthew. 1989. The dating of the Island Southeast Asian
Neolithic: an attempt at chronometric hygiene and linguistic correlation. Antiquity 63:587–613.
———. 1996. Early agriculture and what went before in island Melanesia: continuity or intrusion? In The origins and spread of agriculture and pastoralism in Eurasia. David R. Harris, ed. Pp.
524–537. London: University College London Press.
———. 2003. Chronology of the Neolithic transition in Island
Southeast Asia and the western Pacific: a view from 2003. Review
of Archaeology 24:57–80.
———. 2007. The Neolithic and Austronesian expansion within Island Southeast Asia and into the Pacific. In From Southeast Asia
to the Pacific: archaeological perspectives on the Austronesian expansion and the Lapita cultural complex. Scarlett Chiu and Christophe Sand, eds. Pp. 104–125. Taipei: Centres for Archaeological
Studies Research and for Humanities and Social Sciences, Academica Sinica.
———. Forthcoming. “I was so much older then, I’m younger than
that now”: why the dates keep changing for the spread of Austronesian languages. In A journey through Austronesian and Papuan
linguistic and cultural space: papers in honour of Andrew Pawley.
John Bowden, Nikolaus Himmelmann, and Malcolm Ross, eds.
Canberra: Pacific Linguistics. [MS]
Su, Bing, Junhua Xiao, Peter Underhill, Ranjan Deka, Weiling Zhang,
Joshua Akey, Wei Huang, et al. 1999. Y-chromosome evidence for
a northward migration of modern humans into eastern Asia during
the last Ice Age. American Journal of Human Genetics 65(6):
1718–1723.
Summerhayes, Glenn R. 2003. The rocky road: the selection and
transport of Admiralties obsidian to Lapita communities. Australian Archaeology 57:135–142.
Summerhayes, Glenn R., and Jim Allen. 1993. The transport of Mopir
obsidian to late Pleistocene New Ireland. Archaeology in Oceania
28:145–148.
Suparan, Papay, Rien A. C. Dam, Sander van der Kaars, and Theo
E. Wong. 2001. Late Quaternary tropical lowland environments
on Halmahera, Indonesia. Palaeogeography, Palaeoclimatology, Palaeoecology 171(3–4):229–258.
Supomo, Suryohudoyo. 1995. Indic transformation: the Sanskritization of Jawa and the Javanization of the Bharata. In The Austronesians: historical and comparative perspectives. Peter Bellwood,
James J. Fox, and Darrell Tryon, eds. Pp. 309–332. Canberra: Australian National University.
Swadesh, Morris. 1950. Salish internal relationships. International
Journal of American Linguistics 16(4):157–167.
———. 1952. Lexicostatistic dating of prehistoric ethnic contacts.
Proceedings of the American Philosophical Society 96:452–463.
———. 1955. Towards greater accuracy in lexicostatistic dating. International Journal of American Linguistics 21(2):121–137.
Swadling, Pamela. 1996. Plumes of paradise: trade cycles in outer
Southeast Asia and their impact on New Guinea and nearby islands
until 1920. Honolulu: University of Hawaii Press.
256
Swadling, Pamela, John Chappell, Geoff Francis, Nick Araho, and
Baiva Ivuyo. 1989. A Late Quaternary inland sea and early pottery
in Papua New Guinea. Archaeology in Oceania 24:106–109.
Swadling, Pamela, and Robin Hide. 2005. Changing landscape and
social interaction: looking at agricultural history from a SepikRamu perspective. In Papuan pasts: cultural, linguistic and biological
histories of Papuan-speaking peoples. Andrew Pawley, Robert Attenborough, Jack Golson, and Robin Hide, eds. Pp. 289–328. Pacific Linguistics 572. Canberra: Research School of Pacific and
Asian Studies, Australian National University.
Szabó, Katherine. 2004. Technique and practice: shell-working in the
Western Pacific and Island Southwest Asia. PhD thesis, Australian
National University, Canberra.
Szabó, Katherine, Adam Brumm, and Peter Bellwood. 2007. Shell
artifact production at 32,000–28,000 BP in Island Southeast Asia:
thinking across media? Current Anthropology 48(5):701–723.
Szabó, Katherine, and Sue O’Connor. 2004. Migration and complexity in Holocene Island Southeast Asia. World Archaeology
36(4):621–628.
Terrell, John. 1988. History as a family tree, history as an entangled
bank: constructing images and interpretations of prehistory in the
South Pacific. Antiquity 62:642–657. [MPC/JSL, SJO, JET]
Terrell, John Edward. 2000a. Anthropological knowledge and scientific fact. American Anthropologist 102:808–817. [JET]
———. 2000b. A “tree” is not a “train”: mistaken analogies in Pacific
archaeology. Antiquity 74:331–333.
———. 2004. The “sleeping giant” hypothesis and New Guinea’s
place in the prehistory of Greater Near Oceania. World Archaeology
36(4):601–609.
Terrell, John Edward, and Joel L. Fagan. 1975. The savage and the
innocent: sophisticated techniques and naive theory in the study
of human population genetics in Melanesia. Yearbook of Physical
Anthropology 19:2–18. [MPC/JSL]
Terrell, John Edward, Terry L. Hunt, and Chris Gosden. 1997. The
dimensions of social life in the Pacific: human diversity and the
myth of the primitive isolate. Current Anthropology 38(2):155–195.
Terrell, John Edward, Kevin M. Kelly, and Paul Rainbird. 2001. Foregone conclusions? in search of “Papuans” and “Austronesians.”
Current Anthropology 42(1):97–124.
Thomason, Sarah Grey, and Terrence Kaufman. 1988. Language contact, creolization, and genetic linguistics. Los Angeles: University of
California Press.
Current Anthropology Volume 51, Number 2, April 2010
Thurston, William R. 1987. Processes of change in the languages of
north-western New Britain. Canberra: Pacific Linguistics.
Torrence, Robin, and Pamela Swadling. 2008. Social networks and
the spread of Lapita. Antiquity 82:600–616.
Torrence, Robin, Pamela Swadling, Nina Kononenko, Wallace Ambrose, Pip Rath, and Michael D. Glascock. 2009. Mid-Holocene
social interaction in Melanesia: new evidence from hammerdressed obsidian stemmed tools. Asian Perspectives 48(1):119–148.
[RT]
Tryon, Darrell T. 1995. Comparative Austronesian dictionary: an introduction to Austronesian studies. Berlin: Mouton de Gruyter.
van der Kaars, Sander, Dan Penny, John Tibby, Jennie Fluin, Rien
A. C. Dam, and Papay Suparan. 2001. Late Quaternary palaeoecology, palynology and palaeolimnology of a tropical lowland
swamp: Rawa Danau, West-Java, Indonesia. Palaeogeography, Palaeoclimatology, Palaeoecology 171(3–4):185–212.
Vansina, Jan. 2003. Bananas in Cameroun c. 500 BCE? not proven.
Azania 38(1):174–176.
Vilar, Miguel G., Akira Kaneko, Francis W. Hombhanje, Takahiro
Tsukahara, Ilomo Hwaihwanje, and J. Koji Lum. 2008. Reconstructing the origin of the Lapita Cultural Complex: mtDNA analyses of East Sepik Province, PNG. Journal of Human Genetics 53(8):
698–708. [MS]
White, J. Peter. 2004. Where the wild things are: prehistoric animal
translocations in the Circum New Guinea Archipelago. In Voyages
of discovery: the archaeology of islands. Scott M. Fitzpatrick, ed. Pp.
147–164. Westport, CT: Praeger/Greenwood.
Wichmann, Søren. Forthcoming. Neolithic linguistics. In Language
and prehistory of the Indo-European peoples: a cross-disciplinary
perspective, Gojko Barjamovic, Irene Elmerot, Adam Hyllested, Benedicte Nielsen, and Bjørn Okholm Skaarup, eds. Budapest: Archaeolingua.
Wichmann, Søren, and David Kamholz. 2008. A stability metric for
typological features. STUF: Language Typology and Universals
61(3):251–262.
Wolf, Eric R. 1982. Europe and the people without history. Los Angeles:
University of California Press.
Zimmermann, Gerd R. 1992. Die Besiedlung Südostasiens: eine ethnoökologische Perspektive. Nackenheim am Rhein: M.-C. Edith Zimmerman. [WM]