Voyaging into Polynesia's Past
[Note: Originally published in From Sea to Space (Palmerston North: Massey University, 1992, 5-65). Figures referred to in the article appear in original publication only.]
Ben Finney pioneered the reconstruction and sailing of Polynesian voyaging canoes. He first began dreaming about building a canoe and sailing it to Tahiti while studying at the University of Hawai'i in 1958. In the mid-1960s he built Nalehia, a replica of a Hawaiian double canoe that provided the basic information on sailing performance that went into planning Hokule'a's initial voyage to Tahiti. In 1973 he co-founded the Polynesian Voyaging Society, and served as its first president. He sailed aboard Hokule'a during the first voyage to Tahiti in 1976, the 1985 voyage to Aotearoa, and the 1992 voyage to Rarotonga, and also covered the 1995 voyage from the Marquesas to Hawai'i from Hokule'a's escort vessel. He has taught anthropology at the University of Hawai'i since 1970, and has published a number of books and articles on voyaging and exploration, including Hokule'a, the Way to Tahiti (1979), Voyage of Rediscovery: A Cultural Odyssey through Polynesia (1994) and "Traditional Navigation and Nautical Cartography in Oceania" (1998).]
The Founding of the Polynesian Voyaging Society
Since European explorers first chanced upon the islands of Polynesia and their handsome inhabitants in the 16th century, the issue of how these islands were discovered and settled has been one of the most fascinating puzzles of prehistory. Early navigators from Spain, Holland and England were largely mystified how stone age people could have found their way to these mid-ocean islands. These mariners from another world had only recently developed the technology to cross the oceans, yet on island after island they found people already living there-people who lacked ships, the compass or any of the other devices so vital to European oceanic expansion.
A number of these puzzled seafarers refused to recognize the possibility that the ancestors of the people they found living on the islands could themselves have sailed so far into the Pacific, and instead sought to explain their presence by other means. Consider, for example, the first sustained encounter between Polynesians and Europeans which occurred in 1595 when ships of Mendana's second expedition into the Pacific chanced upon the Polynesian archipelago they called Las Marquesas de Mendonca, now commonlyknown as simply "The Marquesas." The expedition's navigator, Pedro Fernandez de Quiros, was not at all impressed by the sailing canoes of the Marquesans or their claims that they could navigate far out of sight of land. Because they had no ships or navigational instruments, he judged that they lacked the "skill or the possibility of sailing to distant parts." In fact, Quiros seized upon this apparent dilemma of the presence of a people in the middle of the Pacific without the means to have sailed that far into the ocean, to support his own search for a rich "Southern Continent," the Terra Australis that many European cosmographers of that day thought must lie in the Southern Hemisphere. These islanders, he proposed, had been able to employ their primitive canoes and rudimentary ways of navigating to sail to the Marquesas from a continent lying not far to the south, or from a chain of closely-spaced islands located there, which stretched all the way to Asia and had provided the stepping-stones that enabled these primitive seafarers to expand so far into the Pacific.1
On Easter Sunday in 1722, the Dutch navigator Jacob Roggeveen happened across a lonely island in the eastern Pacific he named Paaschen Eilandt (Easter Island), but which the islanders now call Rapa Nui. Roggeveen was even more hard-pressed than Quiros had been to explain this human presence in the middle of the ocean. Whereas the Marquesans had large, double-hull sailing canoes, because of the deforestation of their island the Rapa Nui people had only miserable little outrigger canoes pieced together from scraps of native timber and driftwood. In his journal, Roggeveen records his struggle to comprehend how these people lacking the means for ocean voyaging came to be on the island. First, he asked himself whether the Spanish might have brought them, only to reject that notion because of the apparent lack of any Spanish influence on the island. He then considered that the islanders might be direct "descendants of Adam" who had "bred there naturally from generation to generation," but finally decided that "the ability of human understanding is powerless to comprehend" how these people ever reached their island.2
The common element in these and other such speculations as to the origins of the people European explorers kept finding on the mid-Pacific islands they chanced upon was the assumption that the slim canoes of the islanders, and whatever means they had of navigating without instruments, were simply not up to the task of exploring the Pacific and colonizing the many islands found there. It therefore seemed logical that the solution to the puzzle of how these islands had originally been settled must lie elsewhere than in the seemingly primitive nautical technology and abilities of the islanders themselves.
Such ethnocentric thinking was common during the first age of European exploration when explorers were seeking to develop new routes to the riches of Asia, or new lands for exploitation such as the hypothesized Southern Continent. They were little interested in the people they met along the way, much less in giving them any credit for great maritime achievements. This attitude began to change with the coming into the Pacific of Captain James Cook. His three voyages inaugurated the extension to the Pacific of Europe's second Age of Exploration, the era when, according to the historian Ferdinand Braudel, European maritime nations began sending out expeditions "to obtain new information about geography, the natural world, and the mores of different peoples," as well as for geopolitical and commercial advantage. In fact, Cook was not only the first Pacific explorer to make a concerted effort to understand the people he encountered, but he was also the first to consider seriously how their ancestors might have actively explored and settled this island world.3
During his three voyages Cook criss-crossed the Pacific, touching on the extreme points of the Polynesian triangle, Hawai'i, Rapa Nui and Aorearoa, and many of the islands within. Because of the similarity of language and custom among the inhabitants of all these islands, he recognized the islanders to be members of the same great "nation," the first realization of the existence of the great cultural province we now call Polynesia. Although Cook did not survive the third voyage to return to England where he might have found time to write at length on his ideas about these islanders, in the journal of his first voyage-that made to Tahiti to observe the transit of Venus across the face of the sun-he sketched a theory that has formed the basis for thinking about Polynesian origins ever since.
Cook actually learned some Tahitian, and used his rudimentary linguistic skills to inquire into Tahitian nautical matters. His primary informant was Tupa'ia, a learned Tahitian who told Cook how they sailed their canoes and navigated by reference to the stars, moon and sun, and gave him sailing directions to islands as far away as the Marquesas to the northwest, the Australs to the South and at least as far west as Såmoa, Fiji and Rotuma. Cook was apparently impressed enough with the practical seamanship and navigational skills of the Tahitians, and their wide geographical knowledge, to propose what had been unthinkable to Quiros, Roggeveen and other early European explorers: that the ancestors of these islanders could have sailed into the Pacific on their own, discovering and settling the many islands on which he found their descendants.
Cook thought the islanders (whom he called "Indians" or "South Sea Islanders," for the term "Polynesian" had yet to be applied specifically to them) had worked their way eastward across the Pacific in their canoes, which he calls "proes" from the Malay prahu, or "Pahee's" from the Tahitian pahi:4
Cook saw only one obstacle to accepting the linguistic evidence, supplied to him by his chief scientist Joseph Banks, pointing to the "East Indias," or roughly the archipelago of Indonesia, as the starting point for this migration: the route would have taken canoes eastward in the face of the trade winds that blow from the east-southeast. He evidently had doubts about the ability of the islanders' canoes to sail directly into the trade winds, and quizzed Tupa'ia accordingly. The Tahitian, whom he called "Tupia," had a ready answer that supplied Cook with the information he needed to complete the picture:5
So, with his seaman's eye and eminently good sense, Cook proposed that the islanders came from the west, originally from the East Indies where related languages were spoken, and that they employed their sailing canoes, non-instrument navigational ability, and skill at utilizing westerly wind shifts to work their way eastward, from island to island, against the direction of the prevailing trade winds.
Cook's remarks and the reasoning behind them formed the basis for what might be called the orthodox view of intentional Polynesian settlement from the Asian side of the Pacific that was to be further developed in the decades that followed by a succession of navigators, scientists and other scholars. For example, in 1828 the French navigator Dumont d'Urville precisely drew the cultural and geographic boundaries of Polynesia and gave the region that name. Horatio Hale, the linguist aboard the U.S.S. Exploring Expedition that cruised the Pacific between 1839-1842, systematically traced linguistic relationships within Polynesia and from there to island Southeast Asia, as well as confirmed how westerly wind shifts can be used to sail from west to east to and across Polynesia. Abraham Fornander of Hawai'i, New Zealander S. Percy Smith and other amateur scholars working in the latter half of the 18th century and the first decades of this one collected and analyzed the voyaging traditions of the Polynesians to trace their migrations within Polynesia and to there from the western side of the ocean.6
Not all those who pondered how the islands of the Pacific came to be settled accepted this orthodoxy, however. Prominent among dissenting theories were those proposed by Joaquin Martinez de Zuniga, a Spanish priest stationed in the Phillipines, and John Lang, a Presbyterian minister living in Australia.
In 1803 Martinez de Zuniga published a history of the Philippines in which he asserted that the people of Polynesia and many other Pacific Islands, including the Philippines, spoke languages closely related to those of South America, and that because the steady easterly trade winds of the tropical Pacific would have prevented canoes from sailing eastward, the Pacific Islanders must have come from the Americas, blown by the trade winds from island to island west across the Pacific. Lang, writing later in the 19th century, shared Martinez de Zuniga's idea that people were forced by the wind across Pacific, but reversed the direction of migration. Noting that the easterly trade winds were seasonally interrupted by monsoon winds from the west "which often blow in heavy gales," Lang proposed that Polynesia and other Pacific Islands had been settled by a long series of maritime misadventures when hapless voyagers had been blown eastward by violent westerly winds.7
Although they differed in direction of settlement, Martinez de Zuniga and Lang shared a dim view of islanders' nautical abilities, and, hence, the belief that the islands of the Pacific could only have been settled by voyagers pushed out into the ocean by the winds, be they steady trades or westerly gales. Their theories did not gain wide credence, however, and were submerged beneath a broad consensus that Polynesian canoes, navigational methods and seamanship had been well adapted to the exploration of the Pacific and the settlement of far-flung islands, and that the ancestral Polynesians had intentionally set out from the western edge of the Pacific to explore the ocean and settle the islands they found there. Although within this orthodoxy there were conflicting opinions as to whether the ancestral Polynesians had sailed through Melanesia or Micronesia to reach the mid-Pacific, and other details of the migration, this consensus was not seriously challenged until the middle of this century when a Norwegian adventurer and a New Zealand historian burst into the then quiet waters of Polynesian scholarship to revive the heresies of Martinez de Zuniga and Lang.
One day in 1947 a raft crashed upon the reef of Raro'ia Atoll in the Tuamotus after drifting and sailing before wind and current for 101 days after leaving Peru. The expedition's leader, Thor Heyerdahl, was out to demonstrate how South American Indians could have settled Polynesia by raft. Although Heyerdahl made much of supposed linguistic and other cultural parallels between the American Indians and the Polynesians, the linchpin of his theory was the same as that of Martinez de Zuniga: Heyerdahl asserted that the "permanent tradewinds and forceful companion currents of the enormous Southern Hemisphere" would have prevented canoe voyagers from settling Polynesia directly from the west, while promoting colonization from the Americas by voyagers pushed westward by wind and current.8
A decade later, Andrew Sharp, a New Zealand civil servant turned historian, published a bombastically polemical book called Ancient Voyagers in the Pacific, in which, while accepting the orthodox view that settlement had been from the west, he resurrected Lang's theory that the islands of Polynesia had been settled accidentally by hapless canoe voyagers driven randomly across the sea by stormy westerly winds. The Polynesians could not have intentionally explored and settled the Pacific, claimed Sharp, because their canoes were not seaworthy and weatherly enough, and their navigation system not accurate enough, to have enabled them to set out on long, navigated voyages of exploration and colonization. Instead, said Sharp, Polynesia had been settled over a long period of time by the survivors of maritime accidents. He proposed that canoes were periodically lost at sea when, while sailing along the coast of an island or between closely-spaced islands, they were blown out of sight of land by storms, or simply went off course because of cloudy weather or navigational incompetence. Wherever one of these lost canoes, or others containing people forced to flee their home islands because of war, famine or overpopulation, were randomly pushed by wind and current onto the shore of an uninhabited island a new Polynesian colony would result. This accidental process, multiplied many times over, and nothing more, said Sharp, accounted for the immense oceanic dispersion of the Polynesian nation.9
Although Heyerdahl gained wide popular support for his theory of "American Indians in the Pacific," because his ideas contradicted the linguistic and cultural evidence of an ultimate Southeast Asian origin of the Polynesians, professional prehistorians thought little of them. Nonetheless, despite scholarly protestations, it soon became apparent that Heyerdahl had pointed out a major weakness in orthodox thinking about Polynesian settlement. Relationships evident in language and cultural traits that pointed to a Polynesian derivation from the west, were not matched by island-by-island archaeological excavations demonstrating that the ancestors of the Polynesians had in fact migrated eastward into the mid-Pacific.
Accordingly, archaeologists directed their efforts toward discovering the migration trail to Polynesia, as well as to unravelling the sequence and timing of settlement within Polynesia. The results have supported a derivation of the Polynesians from the western side of the Pacific, not their migration from the Americas as Heyerdahl claimed.
Through a distinctively decorated pottery called Lapita, and associated artifacts, archaeologists have been able to trace the migration of the immediate ancestors of the Polynesians from the Bismarck Archipelago off the northeast coast of New Guinea across Melanesia to the oceanic archipelagos of Fiji, Tonga and Såmoa where they arrived around 1500 B.C. These mid-Pacific islands, and not any distant continental shore, have emerged as the long-sought homeland of the Polynesians, for excavations show that it was in this oceanic setting that ancestral Polynesian culture evolved from its Lapita roots.10
From this homeland region, now called West Polynesia, the trail of artifacts leads to the archipelagos of central East Polynesia-the Cooks, Societies, Australs, Tuamotus and Marquesas, and then from there to the distant islands of Hawai'i, Rapa Nui and Aotearoa. In all the hundreds of excavations conducted throughout Polynesia, no prehistoric pottery or other ancient artifacts that can be directly traced to either North or South America have been uncovered. Although the pre-European cultivation by Polynesians of the sweet potato, a plant of South American origin, indicates that there must have been some communication between the Americas and Polynesia, the archaeological record demonstrates that Polynesians are descended from seafarers who moved eastward across the Pacific from the western edge of the ocean.11
In contrast to the immediate and widespread opposition to Heyerdahl's thesis, Sharp's accidental settlement hypothesis found a degree of acceptance among many historians, cultural anthropologists and prehistorians. Not only was it congruent with orthodox thinking that migration into the Pacific had been from west to east, but it appealed to those scholars who, like the European explorers cited earlier, could not comprehend how people who had only slim canoes, and who lacked the compass and other navigational instruments, could have intentionally explored and settled Polynesia. Furthermore, they saw it as a welcome correction to overblown and ill-founded accounts of Polynesian seafaring and migration, one that offered a simple explanation of Polynesian settlement based on random processes rather than a complicated one based upon seafaring feats that were difficult to imagine.12
Sharp's thesis had a special appeal to some prehistorians who were then introducing modern archaeological methods into Polynesia. They had just come from archaeological centers in Europe and the United States where the focus on migrations that had long dominated the study of prehistory was being replaced by one centered on internal processes of adaptation and change within each cultural or social unit investigated. By embracing the idea of accidental settlement, pre-historians could rid themselves of scenarios-developed largely from oral traditions-involving long voyages of exploration, colonization, and subsequent two-way communication, and of all the complications these would bring to their efforts to comprehend the development of individual island cultures. Instead, they had only to assume initial settlement of an island or archipelago by the random arrival of a canoe, and then, in the centuries that followed, cultural development in isolation from all but the nearest of neighboring islands, perhaps broken only occasionally by the landing of another drifting canoe from a distant island.13
Not everyone, however, embraced Sharp's accidental settlement thesis, particularly the negative assessment of seafaring capabilities upon which it was based. The Polynesians, Sharp claimed, could not possibly have intentionally set out to explore and settle their island realm because their canoes were too flimsy and unseaworthy, their navigation methods too imprecise and their seamanship skills too rudimentary for the task. Unfortunately, however ethnocentric and ill-informed Sharp's assessment may have seemed to those who took exception to it, efforts to refute Sharp foundered on the lack of exact information about the seaworthiness and windward ability the Polynesians canoes, the accuracy of their navigation system, and the quality of their seamanship. The great canoes and their navigators had long since disappeared from Polynesian waters, and the descriptions of canoes and navigation in the explorer's journals were too imprecise and contradictory to settle the question definitively. Without the necessary information, the debate between Sharp and his supporters on the one hand, and champions of the idea that the Polynesians and their ancestors had played an active, seafaring role in the discovery and settlement of their island world, quickly reached an impasse marked more by polemics than insight.14
Initiatives developed to breakout of this stalemate included efforts by nautically-minded investigators to study canoe navigation in the central Caroline Islands of Micronesia, the only place in the Pacific where traditional techniques were still being widely used to guide canoes from island to island, and a massive computer simulation study designed to test whether the pattern of winds and current prevailing in the Pacific would have been conducive to the settlement of the islands by drifting canoes. By detailing how it was possible to navigate without instruments, and how unlikely it was that the movement from West to East Polynesia and from there to Hawai'i, Rapa Nui and Aotearoa could have been accomplished by drifting canoes, these studies went a long way toward undermining Sharp's thesis. Nonetheless, however enlightening was the ability of Micronesian navigators to guide their canoes across the relatively short inter-island gaps of Micronesia, and the statistical unlikelihood that canoes could have drifted over the major inter-archipelago gaps of Polynesia, these studies did not supply the missing information on how canoes could have been intentionally sailed over the long seaways of Polynesia.15
Since the ancient voyaging canoes and their navigators had disappeared from Polynesian waters, the obvious course was to experiment, to recreate the voyaging canoes and ways of navigating without instruments and then try them out at sea. In other words, the situation called for a nautical application of experimental archaeology, that branch of prehistory concerned with the reconstruction and testing of ancient artifacts and techniques. This experimental effort got underway in the mid-1960s, when David Lewis navigated his catamaran from Tahiti to New Zealand without instruments, and when my students and I built a replica of a 40-foot long Hawaiian double-canoe which we used in a series of instrumented trials that showed that such a craft sailed well downwind and across the wind, and could be tacked slowly to windward.16
On the basis of these experiments, I proposed that it would be feasible to sail a reconstructed voyaging canoe over the legendary route from Hawai'i to Tahiti and return, and to navigate all the way by traditional methods. Hawai'i and Tahiti are separated by some 2,250 nautical miles of open ocean, yet Sharp had claimed that it had been impossible for Polynesians to make intentionally navigated round-trip voyages between islands separated by more than 300 nautical miles. To complete such a crossing in a reconstructed voyaging canoe would therefore challenge a key tenet of Sharp's theory.17
In 1973 a group of us from Hawai'i started the Polynesian Voyaging Society to raise funds and then build a large canoe for the Tahiti voyage. The design and construction of a craft that would represent a voyaging canoe of many centuries ago posed a number of problems, for we could not start with an archaeologically-excavated specimen, and then copy it using all the tools and materials of the original builders-as is recommended for experimental archaeology projects. Aside from a few bits of canoes recovered from swamp sites and burial caves, we had no archaeological specimens to guide us, certainly nothing like the virtually complete craft that our colleagues in the Mediterranean and northern European waters have recovered and in some cases copied. Unlike ancient Mediterranean sailing vessels, unballasted Polynesian canoes do not sink. Nor had the Polynesians been so obliging to future archaeologists as had the Vikings, who buried their chiefs in their long boats. However, we did have abundant drawings and descriptions of Polynesian canoes in use during the European contact period, and we used these to develop a "common denominator" design to represent an archaic voyaging canoe ancestral to these local types.
We would have preferred to build our canoe using stone adzes, miles and miles of coconut fiber sennit line, and other features of traditional craftsmanship. But, beyond lashing some components of the canoe with sennit made for us on remote atolls where the old men still knew how to manufacture this cordage, and making an experimental sail out of strips of pandanus matting woven specially for us on the Polynesian Outlier of Kapingamarangi, we did not attempt to build the canoe with traditional materials and methods, for we knew that to try and recreate ancient tools and lost arts would have interminably delayed our project. Instead, we used some modern tools and materials, fabricating our hulls, for example, out of frames covered with layers of plywood strips, and then lashing the hulls, decking and other structures together mostly with modern line. However, we constantly strove to make our canoe in shape and weight a "performance accurate" replica of a traditional voyaging craft that would tell us much about how ancient canoes sailed. For example, despite numerous suggestions that we should widen the stance of the hulls to enable the canoe to carry more sail, add keel fins to the hulls to enhance their ability to resist leeway, and adopt a modern sail rig for greater speed, we stuck to traditional precedents of a narrow separation between hulls, a semi-rounded hull shape and the inverted-triangle sprit sail so that our canoe would sail no better than her ancient predecessors.
We assembled the components of our canoe-two hulls each 62 feet in length, eight crossbeams, decking, rails and two masts-at Kualoa on the north shore of O'ahu Island. The completed canoe was launched in 1975 and christened Hokule'a, Hawaiian for Arcturus, the bright star which passes directly over the island of Hawai'i. Although sea trials were not without mishaps, when properly sailed and maintained the craft proved to be stable and seakindly. Driven by two Polynesian sprit sails, and steered by long steering oars, Hokule'a could make 10 knots or more sailing on a broad reach before strong trades. Her speed of course dropped off when sailing to windward, but trials showed that in sailing full and by against brisk trades she could easily make at least 4 or 5 knots, an adequate windward performance, we felt, for the long voyage to Tahiti.
While we might have fallen somewhat short of the ideals of experimental archaeology in the construction of our canoe, we intended to follow a much more rigorous experimental protocol on our voyage than had been carried out on previous ocean crossings made in reconstructed craft, such as Heyerdahl's 1947 voyage from Peru to the Tuamotus Archipelago aboard the raft Kon-Tiki, and Magnusson's 1893 voyage from Norway to North America in a reconstruction of a Viking longship. Both these crossings had been one-way only, and had been navigated with magnetic compass, charts and other modern aids. In contrast, we intended to make a round-trip voyage between Hawai'i and Tahiti, replicating the two-way voyages celebrated in Hawaiian legend, and, furthermore, we planned to navigate by traditional, non-instrument methods.18
1. Quiros 1904, 2:152; Quiros in Kelly 1966, 2:309.