Papuan Dogs – the first companions of man

D. Early migrations to New Guinea and Australia

In the wake of the uncallenged prehistoric expansions of human populations beyond continental boundaries dogs conquered planet Earth as well: Native American dogs originate from (multiple) Old World lineages of dogs that accompanied late Pleistocene humans (F19) across the Bering Strait (LEONARD et al. 2002). Further mitochondrial DNA research has indicated several maternal origins of dogs from the wild species wolf, Canis lupus. But like in human evolution there is just one focal area of origin from where this global dispersion happened: more than 95 % of the tested sequences (with represented all major dog populations worldwide) belong to three phylogenetic groups only, universally represented at similar frequencies, suggesting a common origin from a single gene pool for all dog populations (SAVOLAINEN et al. 2002). The same team of researchers found a larger genetic variation in East Asia than in other regions: this and the pattern of phylogeographic variation suggest an East Asian origin for the domestic dog, approx. 15.000 years ago (cf. also SAVOLAINEN et al. 2004).

The nearest recent occurrence of wolves to this suggested area of origin for the domestic dog is that of the Indian wolf (Canis lupus pallipes). The distribution-range of the extinct sub-species C.l. variabilis extended further east, to China. Both were smaller than the nominal form. Another even smaller, jackal-like, still existing sub-species is C. l. arabs in the south-estern region of the Arabian Peninsula (mostly Oman).

It should be noted that remains of the sub-species variabilis were discovered in the 400.000 years old cave sites of Homo erectus pekinensis at Zhoukoudian and other locations in China (OLSEN & OLSEN 1977). I am not aware, however, of any DNA analysis of these and other variabilisbone remains. This extinct sub-species deserves much higher atttention, though. According to KOLER – MATZNICK et al. 2003 „it has been mentioned as a C. familiaris ancestral candidate“ by LAWRENCE and BOSSERT 1967 and by OLSEN 1984. The earliest remains of domestic dog so far excavated in China are dated 9.000 BP (CRAWFORD 1992).

In this context I wish to point to the presentation by VILÀ et al. (1997) that the dog separated from the wolf some 135.000 years ago. Consequently, the first stages of domestication should have happened much earlier than commonly suggested in literature (see above). Even if one assumes that first a diversion of lineages occured and only afterwards man selected the still „semi-wild“ dog lineage(s) for domestication in gradual steps. In ruling terminology the „wolf“ is classified as a wild species and the „dog“ that originated from this wild species, is considered a taxon whose genetic differenciation evolved under the selective influence of association with humans (Homo spec.).

The Indian wolf (Canis lupus pallipes) was commonly considered as the wild ancestor of south and eastern Asia domestic dogs – until SHARMA et al. 2003 and 2004 provided convincing evidence by their comparative DNS research that the wolf of the (lowland) Indian Peninsula is an ancient, isolated lineage which did not extend to native domestic dogs. This sub-species and the Himalayan Canis lupus chanco are genetically different from any other wolves and from dogs which together form the so called „wolf-dog clade“. None of the dogs tested (both Indian and foreign samples) share any significant DNA sequence with either the Indian or the Himalayan wolf; they are rather genetically linked to the widespread wolf-dog clade found beyond the Indian lowlands and Himalayan territories of the two isolated sub-species. The researchers note: „ Our results suggest that ‚Bhutia, ‚Twang, Tibetan Mastiff and local „pariah“ dog breeds were brought into the Himalayas and peninsular India by humans….It seems likely that South Asia is not the region of origin for the domestic dog.“

The surviving wolves in India do not live in forests, but in open plains with occasional hills (SHAHI 1982). Also from other sources (F20) we learn that the Indian wolf is adapted to semi-arid, hot areas (shrublands, grasslands, semi-arid pastoral enironments). Only the eastern population of Indian wolves found in Orissa, Bihar and parts of West Bengal is an exception and occurs in moister forest habitats. These observations are significant in view of the comparable habitat of the feral dingo and of the feral populations of the highland strain of Papuan dogs, even as it was established by the research just referred to above that there was no direct line from Indian wolves to domestic dogs. It should be seen rather as an ability on the species level (Canis lupus ssp.) to adopt to equally large territories and diverse ecological conditions as the human species Homo sapiens.

Among the hill tribes of northern Thailand (Hmong, Karen, Lisu, Akha) I observed in the late 1980s frequently village dogs which resembled in body appearance (proportions, ears and tail postures) and colouration (including the white tip of the tail) the traditional dog strain of the New Guinea coastal areas (cf. chapter E). The Akha people are known for their habit of eating dog meat, as much as the Minahasa people (Menado) on Sulawesi Island.

In a documentary shown by ZDF German Television, about the native people of Siberut Island west of Sumatra, I noticed dogs which also resemble the Papuan lowland strain: mostly fawn brown colouration, black face, white belly and paws, white tip of the tail which is held while walking in a almost horizonal position, slightly bent upwards.

Also for the Telomian Dog from Malaysia a white tail tip is characteristic (F21).

Two types of domestic dogs have been found in archaeological sites on the island of Borneo, an earlier smaller dog and a later dingo – like dog (CLUTTON-BROCK 1959 and MEDWAY 1977). These distinct strains occurred on in the upper site layers, dated to 2.000 BP and more recently.

The Punan forest people of Borneo keep dogs whose body stature reminds of the highland strain on New Guinea (cf.photo in CARPENTER 1963), and – as the author points out – are also Basenji-like. It should be noted that Borneo (Dayak) domestic dogs „ as in Thailand“ breed only once a year, according to the same source, like the NGSD. They are only raised for hunting and are normally barkless,

There exist several hypotheses as to how the early human migrants reached Australia. They moved in stages either via the Indonesian islands and New Guinea (by „island-hopping“ or travelling over Pleistocene land-bridges) or on a direct sea route from southern Asia (India) via Timor to northern Australia. In any case it has been established that all dingoes originate from only one single small founder population of domestic dogs, perhaps just a few animals (SAVOLAINEN et al. 2004). But according to SCHULTZ (1969) the dingo presented a great variation in colouration (various shades of brown, from fawn to reddish, black, whitish and spotted white – black) already prior to cross-breeding with introduced dogs. The following distinct populations or strains have been pointed out for more recent times (TRUMLER 1881): a steppe-type (large, long legs, slim); a Cape York-strain (largest size, short hair, head reminds of Basenji); an alpine-strain ( remarkably dense, dark yellow coat, difficult to tame) and a Northern (Darwin)-type (resembles the New Guinea highland strain). Such vage descriptions may not be suffienty representative, but they indicate the fact that evolution in the new homeland of Australia contined and the dingo diversified in its new habitats (cf. the remarks in chapter C) Of course we have no evidence how the first dogs introduced to Australia looked like. And if early seafarers introduced new lineages of dogs to coastal areas of northern Australia (cf. below).

We may assume southern China to be the place of origin of the humans and their dogs which travelled together step by step via Taiwan to the Indonesian Archipelago where they split into an westward and an eastward direction and reached Timor Island, according to SAVOLAINEN et al. (2004) approx. 4.000 years ago. The oldest dog remains on Timor are dated 3.500 years BP.

It is generally accepted that several waves of immigration reached New Guinea, of Austonesians and earlier ethnic groups. No data are available as yet for when these migrations actually reached the New Guinea mainland. The earliest lowland human setttlement site in New Guinea, Bobobongara, was dated to between 60.000 and 40.000 BP (SPRIGGS 1997).The site of Lachitu, a coastal rockshelter with midden deposits was occupied from ca. 35.000 BP (GOSDEN 1995). Therefore, the first settlers may have arrived around that time or even earlier. But there is no evidence for any dogs that early. Autochthonous agriculture of taro tubers in the eastern New Guinea highlands was established 10.000 years ago (DENHAM 2006). It is this ethnic region of several distinct Papuan language groups where in recent times the highland strain of dogs was distributed, party feral and partly in human care. The evidence of sophisticated agricultural practizes is dated much earlier than the assumed migration of the people taking their dogs along (cf. the reference to Timor above). The oldest domestic dog remains excavated in (northern) New Guinea, in the coastal area between the Sepik and the Ramu Rivers were dated at about 5.500 BP (S. Bulmer in KOLER-MATZNICK et al. 2003).

S. BULMER (2000) presents further details on early dog finds in New Guinea and wider Melanesia: Dogs were widespread in prehistoric times; there is evidence from dog bone fragments and teeths found at rockshelters, caves and un-sheltered dwelling and midden sites.

The „Lapita cultural complex“ which is commonly considered as the first evidence of Austronesian language speakers in the Bismarck Archipelago and wider Island Melanesia is related to finds of „the first appearance of three Pacific domesticates, the pig, the dog, and chicken and marks the beginnings of Pacific animal husbandry“ (SPRIGGS 1997). There are no datings presented by Bulmer for Island Melanesia, but a dog burial near Port Moresby in south-estern mainland New Guinea which belongs to an „Oceanic DNA lineage““ (that is likely to occur also in the Bismarck Archipelago), was found in association with a Lapita-like pottery estimated to be 2.500 – 2.000 years old. Dog remains were discovered in several more Austronesian sites along the New Guinea coast and on Yule Isand, of approx. the same age.

For the New Guinea highlands archaeological records of dog are sparse. Dog bone was secured from 6 of the 13 rockshelters and caves investigated and from none of the 4 open settlement sites.The rockshelters and cave sites altogether cover the period of 25.000 BP to recent times. The evidence of dog, in connection with burials and temporary shelters in the highlands, is confined to a period of 3.000 years BP only, but it may be assumed that the dog reached the highlands „by at least 5.000 BP if not earlier, even though none of the archaeological evidence so far suppports that early a date“ (S.BULMER 2000). Therefore findings of dog in the highlands so far are not as old as the ones discovered in lowland New Guinea.

In reference to Susan Bulmer’s presentation several major differences of opinions should be pointed out:

  • The author distinguishes (loc. cit.) between the „wild dog of New Guinea, the Singing Dog, and … the semi-domestic dogs of the New Guinea Highlands“, according to „biological and behavioral evidence“. In my understanding, such differentiation is not justified: the indigenous highland dog should be seen as a distinct strain, with some regional differentiation, of Papuan dogs which either is presented in feral or in semi-domestic or in domestic populations. It should not be seen as a wild animal which arrived on New Guinea either as captive wild or tamed individuals. There is no indication which would support such speculation, nor is there evidence for a separate dog strain (the „semi-domestic dogs of the New Guinea Highlands“ referred to above) prior to the invasion of foreign dogs imported from Australia and elsewhere.
  • According to the same author Singing Dogs may have once populated all of New Guinea: in contrast it is our opinion that the type of dog which was brought to New Guinea (and from there on to Australia) partly retained inherited features and partly differentiated (further) on New Guinea in response to environmental conditions and methods of husbandry to the various types which will be discussed in the following chapter – one of those evolved on site being the recent highland strain. It may well be the case that different immigrants introduced different dogs to New Guinea. However, the uniformity of native strains is not noticeably dependent on ethnic relations but on environmental factors, (e.g. altitudinal zones like the lowlands and the highlands: cf. the following chapter).
  • Instead of noting the existence of several recent strains of equally valid native Papuan dogs, all focus is directed to the Singers only, and that sole representative of the ancient lineage on New Guinea was supposed to have roamed all the island in prehistoric times (cf. also F1); then retreated to the highlands – a very unlikely scenario which postulates the former existence of a „Lapita“ – Singing Dog whose remains have been found in excavations and which then disappeared again from most parts of the island, presumably except for the highlands , leaving a vacuum behind which was filled by some kind of pariah dogs (the lowland dogs) of no identified origin (F22)
  • Once again: there exists no reason to assume that New Guinea was populated by the descendents of dogs imported to New Guinea as wild or „virtually wild“ (captive or tamed) founders, and not by already domesticated dogs. One has to consider the circumstances of those early travels: to carry on top of the burden related to these pioneer voyages some animals which had no emotional traditional associaton with their home places from where these migrants came from, is as unrealistic as constructed from an unlikely point of view. One also should consider the likeliness that these voyages did not happen in subsequent phases; there might have been distinct stop overs on their way of considerable length; should those „wild or virtually wild“ individuals have been kept confined and fed all that length of time…?

  • In her discussion of south-east Asian native dogs Bulmer distinguishes between „modern village dogs, commonly referred to as ‚Pariah‘ or ‚Pye‘ dogs“ and the older „Dingo/Singing Dog early domestic dogs“. The author notes in reference to HIGHAM et. al. (1984): „that the domestic dog in Thailand had by 5.500 changed morphologically half way towards the modern pariah dogI“. Such generalising view is not justified, though: CARPENTER (1963) refers to distinct native types of dogs in modern Thailand and presents the recent Punan dog of Borneo being Basenji – (and Singer -) like (photo No…). There is no significant categorical distinction justified as yet in separating older and more recent lines. I rather follow the view of the so called „general“ or „primitive“ or Shensi or Pariah dogs forming a common lineage which is represented today by the various strains of so called Pariah dogs, the Papuan dogs (including the Singers) and the dingo. My assessment is in accordance with TITCOMB 1969 and GROVES 1995.

S. BULMER (loc. cit.) suggests an arrival of dogs (domesticated or „wild“) in New Guinea during the period of 14.000 and 10.000 BP (cf. also OLSEN 1985 and CLUTTON – BROCK 1995). But it should be noted that there were no dog remains found so far older than 5.500 years. The land-bridge at present Torres Straits between southern New Guinea and Australia existed until 7.000 – 8.000 BP (F20). But throughout the glacial periods during the Pleistocene, the open sea passage across Wallace Line (at least 50 km wide) remained a barrier that only by skilled nagivation could be crossed. The passage of human migration from Asia to New Guinea has probably touched also Borneo, as assumed from the typology of native dogs on this very large island (see above).

Following my own comparison of canid phenotypes on New Guinea as well as in Australia a migration to the continent via New Guinea is the most likely routing in prehistoric times.

Our photographs and observations in the field show a significant conformity of the strains of endemic New Guinea (=Papuan) dogs with the dingo – to a varying degree, though (cf. chapter E). The recent DNA analyses by SAVOLAINEN et al. (2004) support this hypothesis for the highland strain: The „so called New Guinea singing dogs are feral and show some morphological and behavioral similarities to dingos. A common origin and some gene flow between the two populations is therefore possible.“ A later gene flow seems very unlikely, though,due to the geographical separation and the fact that different dog strains are sandwiched in between (cf. chapter E).

A recent study that utilizes newly developed DNA typing methods for human leucocyte antigens (HLA) provides new information about the peopling of New Guinea (MAIN 2001). According to this source, Melanesians are likely to have evolved largely from the same ancestral stock as Aboriginal Australians but to have since differentiated. The highland Papuans are likely to be descendants of earlier migrations who have been isolated for a long period of time. I am reluctant to follow this interpretaton, though, because Aborigines are considered as a very ancient ethnic group which passed through New Guinea before the Austronesians arrived there. SAVOLAINEN et al. (2004) suggest that the dingo descended from dogs which Austronesians brought „into Island Southeast Asia“. But why not these earlier migrants referred to above which might have reached New Guinea with the ancestors of Papuan dogs, and then eventually went on to Australia with the ancestors of recent dingoes?

Apparently no dingoes had reached Tasmania before it was separated from Australia some 12.000 years ago. The oldest dingo finds are dated at 3.500 BP. Since then there was probably no further introduction of dogs to Australia, until the arrival of Europeans. Consequently, „the dingo presents a unique isolate of early undifferenciated dogs“ (SAVOLAINEN et al. 2004). So do the endemic Papuan strains. Some caution is due in both cases, though, because Indonesian traders from the Moluccan Islands regularly visited the coasts of northern Australia and western New Guinea in pre-European times. And the Moluccas again were visited by seafarers from further west across the wide gap of open sea (Wallace Line).

Of particular interest is the dingo population of Fraser Island 40 km off the Queensland coast. Fraser Island dingoes are said to have remained pure without mixing with European dogs (F23). They are tall and slim and match in their appearance largely the Papuan coastal strain of dogs (cf. chapter E). Genetically they form a homogenous population of characteristic dingo lineage (F24).

A dingo-type dog skeleton was excavated in Pukapuka Island of the Cook Islands with an estimated age of about 350 years (SHIGEHARA et al. 1993). LUOMALA (1960) suggested that an investigation of the breed of Polynesian dogs would provide clues to the history of migration of the Polynesians, who domesticated the dogs. As the Pukapuka dog skeleton is morphologically similar to the dingo it suggests some racial or trading interchange among Australia and Polynesian islands before contact with western civilization (F25). Dogs were not known to the New Caledonian Islanders prior to the arrival of white people. The other South Pacific islands where no dogs were kept in pre-European colonial times are : Wuwulu and Aua, probably also Ninigo, Luf and Kaniet, St. Mathias, Emirau and Tench (STERLY 1962). Also on the Polynesian satellite islands of the Solomons: Luangiua and Nukumanu the first dogs are said to have been imported by Europeans (loc. cit.).

The term for “dog” on islands of the New Hebridies, kuri, is widely spread and certainly pre – European; related terms are found, for instance, in the south – western New Guinea lowlands among the Utanata people (wurie) and Kamoro people (wuri), and as far as the Maori of New Zealand and the Tonga-tabu natives (kuri, guri: loc. cit.).

Still the puzzle of data only allows a very fragmentary interpretation of the early history of domestic dogs. The dingo and the Papuan dogs are at the very centre of these investigations. I wish to add some personal remarks of caution: Scientific data should not be mistaken for facts in real terms. Even the most sophisticated research methology only produces results that reflect the narrow range of eventualities which were conscious to and taken into account by the human operator at a particular time. When I was myself a scientific assistant to the Dean of Zoology at Munich University, Prof. Dr. Dr. Hansjochem Autrum, in the mid 1960s, I was shocked to hear from this eminent scientist: „What we cannot prove by our research instrumentarium does not exist“. So how can concepts for research programmes be consigned that consider possible but not (yet) recognized alternative factors and facts, under such limiting suppositions. And what an immensely rich reality is being ignored by such position in principle!

The largely unexplored natural and cultural diversity of New Guinea in later years, when I had made myself independent from such academic narrowness, gave me an ever inspiring insight of how adequate and even superior the knowledge and awareness of those „stone age people“ overthere acually were – and how little it is nowadays respected or even considered. It becomes clear from a quick look into the Internet that the scientific conclusions derived from an examination of just a few mitochondrial elements, for instance, over-rule all the wealth that was accumulated in tribal cultures over millenia. And all this knowledge will never get access into the Internet, because it has no more relevance nor recognized validity … fabricated and selectively promoteddata rule the modern world, regardless.

On the other hand, in order to balance my critical remarks in a fair way: without the comprehensive research of molecular scientists which are named in the relevant papers quoted above, we still would argue about the validity of vage hypotheses. It was only through their break-through DNA analyses that much speculation has been replaced by founded evidence.

But what is regarded today as the unquestionable scientific evidence, may be the antiquity of tomorrow, out of date in an ever changing scenario. This understanding of relative reality, of course also applies to the contents of this book , like of any other… It is still worth writing, though, because future evolution can only select from what exists on a broad spectrum of awareness even as this human achievement is undergoing changes all the time.