It may also be that now and then ambitious younger sons of chiefs, discontented at not being able to attain a higher rank within the community, organized expeditions and left home in order to acquire new lands and there found their own chiefdoms (Akerblom 1968:92-93).

So far, I have reviewed some of the sources of evidence that relate, however shakily, to the prehistory of hierarchy in Austronesia. I am inclined to push back some form of hierarchy, probably hereditary, to at least Proto-Oceanic and perhaps even Proto-Malayo-Polynesian times. Proto-Austronesian, however, still seems to elude the comparativist on such matters. It is now necessary to ask if, and how, founder-focused ideologies of rank, especially through principles of founder rank enhancement, might have influenced the course of Austronesian expansion.

Reasons for individual episodes of Austronesian colonization would, of course, have been many. One thing is certain, however; inter-island colonizations over large distances must have been intentional if they were to be successful, owing to the need to carry a human population with a capacity for reproduction and viable stocks of domestic animals and plants. Current reconstructions of Oceanic canoe and navigational technologies (Gladwin 1970; Lewis 1972; Feinberg 1988; Irwin 1989, 1992) render an undue reliance on unplanned drifting rather unnecessary.

The initial movements of pre-Austronesian-speakers, along the coastlines of southern China and into Taiwan, might have been simply the result of a gradual increase of population consequent upon the development of rice cultivation in southern China between 8500 and 5000 years ago. The Austronesian language family, like those of most other major populations with long histories of agriculture (Papuan, Austroasiatic, Sino-Tibetan, Thai-Kadai (Tai Kadai or Daic), Elamo-Dravidian, Afro-Asiatic, Indo-European, Nilo-Saharan, Niger-Kordofanian) thus commenced its expansion, according to linguistic reconstructions based on patterns of diversity, in a region where agriculture developed in a primary sense from a previous foraging baseline (i.e. without diffusion from an external source). The gradient of population density between cultivators and foragers, even if only slight, would have been sufficient to commence the process, as I have discussed elsewhere (Bellwood 1991).

As the process of expansion continued, however, and as Austronesians moved towards increasing opportunities for isolation in the island worlds of Southeast Asia and Oceania, the process of founder rank enhancement would have come increasingly into play. It matters little whether Initial Austronesians had hereditary or non-hereditary systems of leadership before the expansion process began. The founder rank enhancement process itself would have been reinforced by constant repetition as colonizers moved ever onwards, and in return it would have stimulated, by direct feedback, more colonization.

What were the main advantages in being a founder? We can see that founders in many Austronesian societies were (and often still are) revered by their descendants and provided with enduring fame. Their descendants have also fared well; they occupy positions of high rank in many aspects of social life and clearly have some major advantages in the overall strategies of acquiring land and material goods. We also need to consider that most founders would know of the successes of predecessors, and the more recent the founders the more the successes behind them. In other words, the initial settlers of New Zealand about 1000 years ago would have been heirs to perhaps 3000 years of successful Austronesian expansion, much doubtless recorded in detail in their traditions.

On the material side, the first founder and his/her followers to reach a new territory or island would have had free access to all resources. They would have been able to choose the best dwelling, agricultural and fishing locations, and they would have had moral rights to claim and mark these against the encroachments of later comers (who presumably would also have been much smaller in numbers in most cases). Naturally, any founder would wish his/her descendants to maintain such rights and privileges, and one obvious way to ensure this, especially in a period of rapid follow-up founder competition (such as might have occurred frequently during the process of Lapita expansion), would be to promote an ideology whereby the offices that gave the needed control over resources should be transmitted within a family or lineage, rather than simply laid open to free contestation in big man style. Enter, no doubt gradually and not without some resistance, the ideology of primogenitural inheritance?^[10]

From a more diachronic viewpoint it is not difficult to see how continuous founder events, stacked one upon the other into the depths of Austronesian memory and enshrined in ritual chants, would reinforce the worldly status of founders and their direct descendants. They would also reinforce their access to the best of those material resources (including prestige goods) which theory dictates they should have monopolized in order to enhance or aggrandize their statuses. In short, founders had a head start whenever they discovered a new territory, and their descendants knew how to keep the head start within the family, at least until other forces out of their control (fluctuations in population numbers or successful invasion, for instance) overwhelmed them.

Some care, however, is needed with the founder rank enhancement principle as offered so far. In the discussion I have been careful to stress the concept of new territory, rather than simple fission within an already-settled area. Most founder movements would have been highly local events, like those described in the quotation from Michael Young for Goodenough Island given above at p.26. In these local circumstances founders might have found it hard to enhance rank, especially if they belonged, as many surely did, to junior lines in the parent community. The aim for a really ambitious young man of a junior line might perhaps have been to remove himself and his followers as far as possible from his home settlement in order to convert himself and his descendants into a senior line without interference. What better way to do this than to find another island, or at the very least a piece of good land far removed from home? Spatial separation thus becomes a factor of importance in the equation.^[11]

It is when the concept of moving to another island comes in that another variable enters the equation — the sea-going canoe capable of carrying viable populations of humans, plants and animals. As Hayden (1983) has stressed, the construction and manning of a large canoe needs manpower, and this may mean that founders who successfully undertook long voyages, such as many of those necessary to colonize Oceania, would already have belonged to a stratum of the homeland society which might well have had élite privileges. In other words, the need to build and man a canoe would have selected for founders those of relatively high rank, and it is rather irrelevant here if that rank in early stages was ascribed or achieved. A strong chain of authority on a voyaging canoe could have meant the difference between success and failure, as anyone will realise after reading Finney’s entertaining account (Finney 1979) of the experimental Hokule’s voyage from the Hawaiian Islands to Tahiti in 1976.

If we imagine the founder process occurring into relative or complete isolation time and time again over a period of four millennia, with each major ocean crossing selecting for a leader with at least a better than average access to wealth and manpower and giving that leader a pristine new laboratory in which to enhance his own rank and that of his descendants, then it may not be hard to understand why so many anthropologists have commented on the general west to east gradient in the occurrence of ascribed ranking in Oceania. The big and great men in western Melanesian Austronesian societies may reflect some degree of Papuan assimilation, and if Austronesians had not undergone this experience we might expect ascribed rank to have been more common in this region. But this may not be the only reason for the gradient; the ancestors of the Polynesians and Micronesians, after all, presumably went through more founder enhancing events than anyone else.

In addition, it is worth reflecting that there might also have been a similar gradient, now masked by Indic and Islamic influences, from Taiwan, the Philippines, Borneo, and Sulawesi down into the Malay world, the Sunda islands of Java and Bali, and Nusa Tenggara. This was also an axis of Austronesian colonization, and my overall reading of the literature suggests to me that traditional societies tended to rely more on ascribed ranking in the south (e.g. Java, Bali and Nusa Tenggara) than in many parts of the Philippines and Borneo, where many societies have probably retained stable forms of non-hereditary leadership since initial Austronesian colonization occurred. Clearly, one would not wish to explain the differences between Maori and Hawaiian society, or between the Kayan, the Iban and the Balinese, purely on the basis of the number of decisive founder events in their respective histories. Factors of ecology, production, circumscription (Kirch 1988a) and even perhaps free-thinking ideology surely had roles to play as well, as in all societies, not to mention the influences of India and Islam. But there is something attractive about the concept of a founder event that leads to a complete separation between a homeland and a daughter settlement. After all, this is precisely how many biologists would explain the process of speciation in living organisms. Perhaps small human founder groups in isolation can produce massive cultural changes, just as, if given far longer spans of time, their animal cousins can create species.