The Boundless Sea

by David Abulafia

  The pottery evidence is so remarkable because it shows clearly that this was a single culture with regional variations. The pottery was handmade, without the use of a wheel, and without kilns, meaning it was probably fired out in the open. Here we have a common ‘dentate’ style, where pots were often stamped with a tooth-shaped instrument, and intricate patterns were created with great artistry. These patterns have been seen as a sort of vocabulary, conveying messages now lost; there were also local variations in the pottery styles, and the most striking fragments to have survived show incised human faces, or at least features such as eyes. Possibly these represent gods or ancestors, and the designs may have been similar to those used in tattooing, which was widespread (tattooing instruments have turned up in excavations). The spread of this pottery through ‘Remote Oceania’ provides vital clues about the arrival of the first humans on islands deep within the Pacific. The inhabitants of the Bismarck archipelago were making Lapita-style pots around 1500 BC. Over the next century or so the pottery reached ‘Near Oceania’ (Vanuatu, Kiribati and neighbouring island chains). By 1200 BC it was being produced in Samoa. Interestingly, only the oldest pottery from Fiji shows such concern for intricate decoration. Was this art lost over a generation or two? Did the decoration lose its significance, particularly in new societies which were not yet part of networks of reciprocal exchange? Oddly, as the Lapita people moved still further eastwards they brought their plants and animals, and their knowledge of navigation, but eventually lost interest in pottery entirely.34

  There was a single culture; but was there a shared culture? Chemical analysis of the clay proves that pots were carried from island to island, though no doubt some were moved around the Pacific simply as utensils containing the food navigators needed; many undecorated pots would have been suitable for use as containers for sago flour, which kept well and provided ideal nutrition for navigators. Care needs to be taken with the assumption that the movement of these goods and other items such as obsidian and chert (the class of rock that includes flint) adds up to ‘trade’. Trade might be defined as the systematic exchange of goods, for which a notional though generally variable value is set. In Pacific island societies, as the great ethnographer Bronisław Malinowski showed, the exchange of goods was not simply concerned with commercial acquisition; reciprocal exchange was a means by which individuals established their place in the social and political pecking order, a way of establishing claims to leadership and of emphasizing who was a client to whom.35 This would be even truer of societies that experienced plenty, as these island communities generally did; and yet there were certainly foodstuffs and tools, most notably cutting implements and adzes, that were not to be found on coral atolls and that needed to be obtained over the water. The closer one looks at this world, the more connected it appears to be.

  An example from the western end of the Lapita world provides rich evidence. Talepakemalai lies on the northern edge of the Bismarck archipelago. The history of this village can be traced over five or even seven centuries, beginning halfway through the second millennium BC. At that point, early in Lapita history, obsidian arrived from islands not far away, as well as adzes and chert for making tools, plus pottery from twelve sources, not all identifiable but all distinct in the composition of their clay. Meanwhile the islanders were adept at making fish hooks, and also decorative jewellery consisting of beads, rings and other objects created out of shell. Archaeologists therefore speculate that some sort of exchange network linked Talepakemalai to a series of island communities in western Near Oceania. Yet by the first millennium the early expansion had slowed, a process mirrored in the contraction (or ‘regionalization’) of this part of the Lapita world. This could reflect a greater degree of autarky, that is to say, less need to rely on neighbours for certain types of goods, which could now be produced locally. The local economy perhaps strengthened, but what the archaeologists tend to see is less evidence for external links, which gives the illusion of Lapita decline. This may have some bearing on a phenomenon to be observed in a moment, the long interval between the Lapita expansion and a new phase of exploration and settlement in the first millennium AD.36

  We know very little about Lapita boats. One or two rock carvings offer clues to the shape of sails (including an interesting ‘claw’ shape, with a roughly triangular profile, but with a concave top line); but much depends on the Austronesian words reconstructed by philologists, because nothing of the original boats survives in the archaeological record. Broadly, we can conceive of sailing vessels with outriggers, similar to those used in later centuries; some may have been catamarans, though double canoes of this sort seem to have developed mainly in Remote Oceania, around Fiji. By modern times the variety of boats was considerable, but they conform to a common type: sailing vessels whose builders paid close attention to their stability.37 It was understood that a single hull did not suit small boats in high seas. Polynesian boats were hard to topple; and those that set out for new lands must have been large enough to carry men, women, supplies of food and water (often stored in bamboo tubes), domestic animals and seeds or tubers ready for planting in new lands. Those heading for familiar territory evidently carried goods to be exchanged, such as ceramics, local produce of the soil, and tools or blocks of stone for making into tools. No doubt there was great variety, though some features, such as the use of vegetable fibres to tie together the components, were probably standard. These bonds, made of coconut fibre, were strong and resilient, and rendered the hull more secure because of the flexibility they offered.

  Navigators had to face strong challenges. The most obvious were the easterly winds. Colonization of the Pacific occurred in the face of the wind rather than as a result of happy accidents as sailors were caught in the wind and carried to unknown islands. The trade winds and the currents all point westwards; the trade winds cross the Lapita area of settlement from south-east to north-west, forming a coherent band that matches quite well the Lapita area. The Pacific currents consist of four main trans-Pacific movements: a southern current lying away from all the islands; the South Equatorial Current heading westwards with a slight southward inclination; and above the Equator two contrary currents that separate Hawai’i from the rest of the Polynesian world. Looking at the South Equatorial Current, as with the winds, the broad shape of movement from Samoa westwards very roughly coincides with the zone of Lapita settlement. Evidently, Polynesian navigators perfected the art of sailing against the wind; they needed to ensure that they could return from their explorations, and the best way to do that was to challenge the winds and currents, tacking back and forth, moving slowly but securely.

  As they developed these techniques over many centuries, they also learned the art of dead reckoning, judging distance as they sailed to gain some sense of longitude; they appear to have found this easier to do than European sailors, who had to await the invention of the chronometer in the eighteenth century to be sure of their longitude. Tupaia, a Polynesian navigator who accompanied Captain Cook, astonished Cook’s companions by his almost instinctive awareness of where the ship stood, without instruments or written records. The Polynesian navigators proved that one can solve some challenging problems without any technology at all, just the super-computer of the human brain.38 As for latitude, much easier to judge, they observed the stars: ‘to travel between the south of the main Solomons and the Santa Cruz group was as simple as following a zenith star path – east or west – with the seasonal winds.’39 Knowing the stars was the key to successful navigation. This was not casual knowledge but a science learned during a long apprenticeship, through practical experience and by way of an elaborate oral tradition; it was a secret science, intended for carefully chosen initiates who would be able to navigate the boats while the rest of the crew performed more humdrum tasks.

  Even in the 1930s these methods were taught to boys, beginning at the age of five, as the story of a celebrated sailor from the Carolinas named Piailug reveals. Once his grandfather decided that the bo
y should become a navigator, he had to spend his time listening to stories of the sea and acquiring information about the science of navigation. His grandfather assured him that as a navigator he would be better than a chieftain, would eat better food than others and would be respected throughout society. By the age of twelve he was travelling the ocean with his grandfather, and he began to master the secrets of the sea – the movement of birds, the changing map of the stars, but also magical lore. All this was committed to memory, leading to a full initiation around the age of sixteen which involved a month of seclusion during which his teachers bombarded him with the knowledge he needed. He had no use for written texts, but he made models out of sticks and stones that he could memorize and rebuild, when the time came to instruct the next generation in the art of navigation.40 In the Carolinas, navigators would prepare a sidereal compass, a chart of the key points in the night sky, which in modern times they greatly preferred to a magnetic compass; in other parts of the Pacific similar compass-type charts were constructed out of sticks and stones to show wind direction or the movement of the sun across the sky.41

  The Polynesians did not necessarily require a compass of any sort. There is the story of a schooner captain who lost his compass overboard and confessed to his Polynesian crew that he was lost. They told him not to worry, and took him where he wanted to go. Puzzled by the ease with which they had achieved this, he asked how they knew where the island was. ‘Why,’ they replied, ‘it has always been there.’42 The extraordinary confidence Polynesian navigators possessed in their methods can also be judged from an interview with a navigator from the Marshall Islands conducted in 1962: ‘we older Marshallese people navigate our boats both by feel and by sight, but I think it is knowing the feel of the vessel that is the most important.’ He explained that a practised navigator would have no difficulty sailing in daytime or at night, and that it was important to take proper account of the movement of the waves:

  by the boat motion and the wave pattern a Marshallese sailor who has been trained in this kind of navigation may know if he is thirty miles, twenty or ten, or even closer, to an atoll or island. He also knows if he has lost his way, and by looking for a certain joining of the waves, he will be able to get back on course.43

  In cloudy weather, any break in the clouds at night-time had to be exploited immediately; but there were other signs, such as the swell of the sea, that a practised navigator could use to identify where the boat was heading. There were plenty of additional signs, various combinations of which made a landfall a certainty. Land could be detected through the flight of birds such as terns coming out to sea to feed. Their range from land was known; the direction they came in the morning and returned at night was the best possible clue to where land lay. Other signs included cloud formations, which might change colour, reflecting the land that lay underneath (coral atolls would cast an opal tinge on the cloud above). Phosphorescent patches in the sea were a further sign that land was near. Increasing amounts of flotsam generally suggested land.44 The very smell of the sea air would help guide a sailor to a known haven.45 It was important to compensate for currents and wind, making use of the sun by day and the stars by night to adjust course as appropriate. One of the most extraordinary methods of navigation was what might be called the Polynesian Theory of Relativity, a system known in the Carolina Islands as etak. Here, the assumption was that the boat remained still and the rest of the world moved. Judgements therefore had to be made about how the position of the islands altered in relation to the boat – a relationship not just between the boat and its destination, but between the destination and another island in the vicinity; the method depended on placing this third point accurately in relation to the stars. This was not perhaps Einstein; but it involved some powerful mental geometry, not to mention an astonishingly detailed, carefully memorized, mobile map of the heavens.46

  It is therefore completely wrong to conclude that without writing there can be no exact science, even if the navigational science of the Polynesians had a good share of incantations, magical practices and invocation of the gods. The extraordinary understanding that Polynesian sailors developed of the sea and its whims, and the increasing evidence that they settled the islands in the face of the winds rather than by being blown towards landfalls, has significant implications in explaining the voyages. Much ink has been spilled contesting the apparently plausible views of Andrew Sharp, whose book Ancient Voyagers in the Pacific, first presented to the Polynesian Society in 1956, insisted that those who found new land by and large did so by accident, when they were blown off course or otherwise lost. Sharp did not challenge the argument that Polynesian sailors were exceptionally skilled, but he did underestimate their remarkable abilities.47 What Sharp really demonstrated was something else: we still do not really know why the Lapita people and their successors, including the Māoris, settled one territory after another across the vast expanse of ocean. We can say with some confidence how it was done, and more or less at what periods it was done (though, even there, there is forceful disagreement). Why they kept on the move is a topic for speculation.

  The period of rapid Lapita expansion reached a climax around 1000 BC when Vanuatu and the Fijian islands were settled. This involved ambitious journeys well out of sight of land, particularly to reach Fiji; there were a few more stepping stones on the route eastwards that led the Polynesians to Samoa and Tonga. The Lapita had reached the limits of their expansion, and had created a series of networks across about 4,500 kilometres of the Pacific, in a great arc from New Guinea to Tonga.48 Just as great a mystery as the origins of Lapita expansion is its cessation for up to a millennium. Was this because Polynesian boats were unable to venture on to the vast tracts of open ocean separating Lapita lands from Hawai’i, New Zealand and Easter Island? The problem with this argument is that Lapita sailors had already managed to reach Fiji and Samoa, well out in the ocean.49 These were very skilled and imaginative navigators, and it is hard to believe that they were incapable of adapting their already impressively durable boats to face stormier seas. Overpopulation was apparently not a strain on them. A fine ecological balance had been achieved, despite the radical replanting of large tracts of soil on the islands. The problem with materialistic explanations is that migration in many parts of the world has often been stimulated by religious beliefs that are impossible to recover so far back in time. Supposing for the moment that the Polynesian explorers were guided by a religious imperative to seek the rising sun (an argument for which, admittedly, there is hardly even circumstantial evidence), then cultural fashions could have changed as religious ideas altered. Once the cult of local ancestors developed strongly, a greater sense of being rooted in the island where one lived would act as a brake on further expansion – though not, as it turned out, indefinitely.

  2

  Songs of the Navigators

  I

  Pacific navigation revived by the fifth century AD. Why this should have happened is as uncertain as why it had ceased in the late Lapita period. A link to the so-called Little Climatic Optimum has been suggested, but this does not quite fit the chronology, which suggests a revival of navigation at least a couple of centuries earlier. Rising sea levels during warmer weather might have made life difficult on low-lying islands with plantations by the shore, stimulating migration.1 Only in the millennium after about AD 300 did settlement expand north and south, and much further west, into areas of varied climate and resources, as far as Hawai’i in the north and New Zealand in the south. During this phase Tahiti and the Society Islands were one focus of settlement, beginning around AD 600, if evidence from domesticated coconuts on the island of Mo’orea is given full credit; however, the earliest inhabited sites on these islands that have actually been discovered date from somewhere between 800 and 1200, although many earlier sites may be under water as the coastline has altered.2

  The journey northwards from Tahiti or the Marquesas to Hawai’i may have taken three to four weeks, and winds blowing in various
directions had to be managed: first east to west, then west to east, and finally east to west again. Attempts to mimic the Polynesian voyages during the 1970s, led by Ben Finney in the canoe Hokule’a, showed that the journey was manageable. Finney, along with the New Zealander Jeff Evans, has been a pioneer in reconstructing traditional boat types and in encouraging the Polynesians to take an interest in their ancient navigational skills, and his experimental voyages are taken seriously.3 A more difficult question is whether the Hawai’i archipelago was settled immediately after discovery by a group of migrants bringing plants and animals, including pigs and dogs, as in the island chains further south. It cannot be assumed that the settlement of Hawai’i, a whole group of islands, was a single event; different islands may have been colonized at different times, sometimes from neighbouring islands in the archipelago, and sometimes from the island chains much further south, around Tahiti and Samoa.