Volume 71 1962 > Memoir No.34. Polynesian Navigation, edited by Jack Golson > The settlement of Oceania: An examination of the accidental voyage theory, by G. S. Parsonson, p 11-63
                                                                                             Previous | Next   

- 11

THE HYPOTHESIS that the world's remoter islands were initially discovered and settled by accident finds its intellectual parent perhaps in the ‘Christian Topography’ of a 5th century Greco-Egyptian merchant, Cocmas Indicopleustes, who held that only the great gulfs of the ocean were designed by Providence for navigation, that the ocean itself was impassable. 2 It dates more especially from the 14th and 15th centuries when European seamen, ignorant or forgetful of the achievements of the Arabs and Vikings and still sceptical of the newly-recovered Ptolemaic geography, first began to grope their way out of the Mediterranean into the Atlantic in search of far Cathay.

It received perhaps its earliest statement in English in the volumes of the Rev. Samuel Purchas, Hakluyt's continuator, published in 1625. According to him, the art of navigation had been reserved to the children of God. In ancient times, Solomon had been “the first Founder of Long and Farre Navigation, and Discoveries”. 3 But, “barbarous Empires [had] never growne to such glory . . . because Learning had not fitted them for Sea-attempts, nor wisdome furnished them with Navigation”. 4 The compass was still unknown. Thus the “Starres and Coasts were then Guides, and without those Stilts and Stages, Navigation durst not adventure . . . to go into the Maine”. If storms should arise or the stars be obscured, the mariner must infallibly be lost. “No great Discoveries was otherwise by the Art of the Ancients - 12 performed; nor durst any repeat that Lesson by Art, which Tempest had occasionally taught him, farre from the Coast, against his will.” 5

The first fully developed and indeed still classic statement of the idea, however, dates from 1597. In a memorial addressed to the Viceroy of Peru, the Portuguese navigator Pedro Fernandez de Quiros described how his Captain General, Alvaro de Mendaña, when he was sailing in 1595 towards the Solomon Islands, discovered four small islands together, the Marquesas,

“inhabited by so good a people, that there is no account of any other having been discovered that can be reckoned equal to them . . . These islands are in the latitude of 9 deg. and 10 deg., distant 1,000 leagues from LIMA, 650 from the nearest coast of NEW-SPAIN, and another 1,000 from NEW-GUINEA. The winds there are easterly, for which reason, to go from these islands to PERU and NEW-SPAIN, it is necessary to stand to the northward, or southward, to seek without the tropic those winds which are called general; and for this are required instruments of navigation and vessels of burthen, which are two things (besides others very necessary) these people are destitute of. For these, and for other reasons . . . it remains very obvious, that it never has been possible, in any time, to communicate with the two provinces above mentioned, nor less with NEW-GUINEA and the PHILIPINAS since, from these parts to the said islands, there is no navigating, on account of the winds being east and contrary. From the four islands no other land was visible; the embarkations of the natives are adapted for short voyages. For which reason it is to be sought, what could be believed to be the manner how they could go to distant parts: and it is the most likely, that when they sail from a place from whence they can see no other land, they go, taking their marks by the land they are leaving, till they lose sight of it; and then, when they can see it no longer, they get sight of the other whither they are bound: for in losing altogether the land, as well as that from whence they sailed, as that which they are in search of, it becomes necessary to understand at least the compass, which they have not. Not to mention the contrary winds, currents, and other things, which may make them lose their right way. And this is of the more weight, as the most experienced pilots, furnished with all that these people want, in losing sight for two or four days of the land, do not know, nor can determine their situation . . . the instruments of navigation of these Indians are their own eyes, or their guess of small distances. Since to what might be alledged that they direct themselves by the sun, moon, and stars? It may be replied, that the sun is not visible by night, the mutability of the moon is well known, and, in short they are not always present, nor in the same place, nor without clouds before stars, moon, and sun. But were all this possible (which it is not) yet their voyages must be so short as has been said.
“And though it is true, that the most stupid can go in their embarkations from a small island, to seek a large country, if it be near, since if they do not hit one part, they will another; yet not for this can it be admitted, that from a large or small country, they can, without art, seek small and far distant islands.
“Amongst these Indians were some mulattoes, which diversity of colour argues a communication with other people. Besides these four - 13 islands being small, it is to be attended to, that, in the large countries men are scarce to be contained, but ever go in quest of others, where they may live with more convenience, besides those who leave them on dissentions, or because they cannot submit to masters, or because they wish to be such. Thus it may be religiously believed, that there are to the SE, S, SW, and more westward, other islands which lye in a chain, or a continent running along, till it joins NEW-GUINEA, or approaches the PHILIPINAS or the SOUTHERN-LAND of the strait of MAGALHANES, since no other places are known, whereby they who inhabit those islands could have entered them, unless by miracle.
“If it goes toward the one, or to the other part, or towards both, it is likely there are many islands, or a continent, the antipodes to the greater part of EUROPE, AFRICA, and ASIA, where from 20 deg. to 60 deg. God has made man so useful.” 6

It was Quiros's conviction that the inhabitants of this “fourth part of the globe”, whether of a continent or a mere series of islands—and his doubts on this point are noteworthy—must be, like those of Europe, “addicted to letters and arms, and so expert in everything polished . . .”.

“But leaving many other reasons which might be brought as proof of what is affirmed, it is to be observed that not one of all the islands hitherto discovered in the wide ocean in all the seas of the world was inhabited, but desert, and without men to tread upon them, except the LADRONES; these, it is certainly known from a chain which adjoins to Japan, at some small distances from one another, for in all the voyages made in that quarter, from the PHILIPINAS to NEW-SPAIN, they have fallen in with islands. For example, the TERCERAS, the island of MADEIRA, those of CAPE DE VERDE, and others of the Atlantic, which being in the main ocean, were found uninhabited: on the contrary, the CANARY islands being situated in sight of the main land of AFRICA, were found with people, as is known.
“Since these, notwithstanding their being so near the lands of EUROPE and AFRICA, wherein the knowledge of navigation is so ancient, remained for so long unknown and were discovered and peopled by accident; what can be said of the four now discovered in so wide and extensive an ocean, inhabited by a people so ignorant, and all those of these parts as much without art as them? Not to mention the islands of the Mediterranean, and all the others which adjoin to the five provinces of EUROPE, AFRICA, ASIA, NEW-SPAIN, and PERU, which being in sight, or near at hand, the inhabitants entered.” 7


The logic of any argument is one thing; the truth of its premises and therefore of its conclusion is another. Quiros was given his chance and he failed. But the accidental voyage thesis on which he had built so much was nonetheless destined to survive and to exert, some 150 years later, a notable influence upon English thinking on the problem of the South Seas.

- 14

In this the moving spirit was Alexander Dalrymple, Quiros's re-discoverer and tireless advocate, who not merely cherished a fierce and obstinate belief in the existence of the Great South Land but, as a fellow of the Royal Society and later the Admiralty's first hydrographer as well as an acknowledged expert on the East Indies and the history of oceanic navigation, long enjoyed a powerful ascendancy in both naval and learned circles.

As a scientist, he naturally preferred such recondite arguments as the need for a similar balance between land and water in both hemispheres and the observed variability of Pacific winds, but he thought it improper to

“omit entirely [Quiros's] argument of a continent to the South, from the fair-haired people found in the islands, because nothing appears to be a more conclusive proof of the existence of the Continent than this fact, which is entirely contrary to the common circumstance within the tropic . . .” 8

And his championship of the Portuguese did not fail in its effect. Thus in 1766, in deciding to prosecute discoveries in the South Pacific rather than continue the search for the North-West Passage, the Admiralty secretly instructed Wallis in words all too obviously borrowed from Quiros's memorial to search for the “Land or Islands supposed to lie in that part [between Cape Horn and New Zealand] of the Southern Hemisphere”. 9 Wallis's master, George Robertson, in 1776, employing language even more remarkably like that of Quiros, came to the conclusion that the Tuamotuans must make their way to their infertile islands at

“certain seasons of the year and then retire to some larger place . . . which in my opinion cannot be very far distant from this place—as the Largest of their craft is not fit for going on long Voyages in, neather do I think they are as Expert navigators as to conduct their Vessels to any great Distance . . . but at any rate the Original of the people which we saw must certainly come from some large Body of Land, the same as those who Inhabited the Bahama Islands which the Great Discoverer Collumbus first found out.” 10

At Tahiti, he could only regret the circumstances which compelled Wallis to order him to steer westward instead of towards the south and a continent which it seemed they had actually in sight:

“for I think their is the greatest reason to believe that the large fleet of Vessels that Came round the SW point of Port Royal Came from that Country—
“The inhabitants of the place at least would find no difficulty in supplying themselves from their neighbours' country which lies so convenient, that they can Sail from their own Island, at Sun rise, with the common trade Wind and reach their by sun set and the same trade wind will Bring them back next day if they want to come.” 11

- 15

Then, as Hawkesworth put it, Cook “subverted the theoretical arguments on which the Southern Continent had been based for upon this principle what we have already proved to be water, would render the Southern hemisphere too light”. 12 And, for the moment at least, the accidental voyage theory also went down in the wreck. In 1768, indeed, Bougainville, the heir of a less dogmatic tradition, had convinced himself that the Polynesians were capable of voyages of 300 leagues. 13 In 1769, Cook set down the new doctrine even more unequivocally. The Raiateans, he thought, could sail in their

“Proes or Pahee's . . . from Island to Island for several hundred Leagues, the Sun serving them for a compass by day and the Moon and Stars by night. When this comes to be prov'd—[after 'prov'd' he writes, and then deletes, the parenthesis ‘which I have not the least doubt of’] we Shall be no longer at a loss to know how the Islands lying in those Seas came to be peopled, for if the inhabitants of Ulietea [Raiatea] have been at islands laying 2 to 300 Leagues to the westward of them, it cannot be doubted but that the inhabitants of those Western Islands may have been at others as far to the westward of them and so we may trace them from Island to Island quite to the East Indies.” 14

For at Tahiti, as Cook said, the winds did not blow constantly from the east but were subject to variations, frequently blowing a fresh gale from the south-west for two or three days together, but very seldom from the north-west. Tupaia had also informed him that in November, December and January the winds blew constantly from the west with rain and that “they knew very well to take advantage of these in their navigations”. 15 Moreover, even in the trade wind season, sailing in a contrary direction was not out of the question; he himself had experienced variable winds for two or three days together.

Cook had, however, gone too far. He had disproved the existence of the Great South Land and none but the irate arch-theorist Dalrymple could deny him. But the accidental voyage thesis was quite another matter. The learned and aristocratic circles in which the newly promoted commander had moved since the successful outcome of his first voyage obviously found it difficult, thanks no doubt to Dalrymple's intervention, to agree with him that ignorant savages had long ago achieved that mastery of the sea which the navigators of the Old World had scarcely yet attained. He, therefore, soon came to adopt conclusions wholly at variance with his previous observations. 16

His discovery on 3 April, 1777, at Atiu, 200 leagues distant from their homeland, of three Tahitian castaways, the sole survivors of a party of 20 men and women, “an immense ocean between, with such wretched sea-boats as the people are known to make use of fit only for a passage where sight of land is scarcely ever lost” naturally clinched the issue. 17 For the incident at once seemed to him to explain - 16

“better than a thousand conjectures of speculative reasoners, how the detached parts of the earth, may have first been peopled: especially those that lie remote from any inhabited continent, or from each other.”

Thus, as Anderson, Cook's surgeon and his trusted adviser in such esoteric matters, put it, the Tahitians' knowledge of other distant islands must be traditional and had in fact been

“communicated to them by the natives of those islands, driven accidentally upon their coasts, who, besides giving them their names, could easily inform them of the direction in which the places lie from whence they came and of the number of days they had been upon the sea . . .”. 18

It was in this way that the extensive knowledge “attributed by the gentlemen of the Endeavour, to Tupia, in such matters”, might be accounted for. 19

And this opinion prevailed. Twenty years later, the Tahitians themselves apparently confirmed it. For according to Wilson of the London Missionary Society's ship Duff, writing on 14 July, 1797:

“At this time Pommare and his retinue particularly regretted their want of ships, and knowledge to conduct them to foreign countries; and, addressing himself to me, said in a tone of concern, that they were able to go no further than Ulietea [Raiatea] or Huaheine; and that at the risk of being driven they knew not whither, to perish; whereas we could sail for many moons, and in the darkest nights and strongest gales, and after all could come exactly to Otaheite.”

The reply must have been disconcerting; the wily and ambitious chief sought aid, not sympathy. But the missionaries, speaking as they believed with a deeper insight than Cook's into the true source of Polynesian backwardness, gravely replied that the British were once as they were and knew nothing but that “good men brought the speaking paper into our country, and taught us to understand it, by means of which we learnt to know the true God, to build and conduct ships, and to make axes, scissors and the various things which he saw we possessed”. 20 It is doubtful how Pomare's ancestors would have received this strange doctrine. But undoubtedly Quiros and with him Cosmas and Purchas had triumphed.


Quiros's theory had not of course been reinstated merely to salve the wounded pride of Dalrymple or the doubts and prejudices of learned men and missionaries. If it had not existed it must have been invented. For as Cook himself had been led to think, it appeared to cast a great deal of light on the problem of Polynesian origins.

Dumont d'Urville's thesis of a sunken continent which made the Polynesians a race of refugees who had fled to the mountain tops as the - 17 waters closed in and thereafter been irrevocably cut off from further commerce with each other was not calculated to impress Europeans. The Polynesians were far too homogeneous a race for that. It seemed obvious that whether or not the Polynesians were indeed the lost ten tribes of Israel, or Celts with close affinities with the peasants of Brittany, or whether they had come out of India, China, Indonesia, America or even New Zealand, they must have traversed enormous distances and that at no very distant date.

The facts of Pacific geography, in particular the distribution of winds and currents, had early suggested an eastern origin, and this thesis was adopted by de Zuniga in 1803. In 1829, William Ellis, the author of Polynesian Researches, also came down at least partially in favour of it. According to him,

“If [Polynesia] were peopled from the Malayan Islands, they must have possessed better vessels, and more accurate knowledge of navigation, than they now exhibit, to have made their way against the constant trade-winds prevailing within the tropics, and blowing regularly, with but transient and uncertain interruptions, from east to west.”

It was clear from the “monuments or vestiges of former populations found in these islands” that they were uncivilized, and incapable of voyages of 6-7,000 miles. It was easy, however, “to imagine how they could have proceeded from the east. The winds would favour their passage, and the incipient stages of civilization in which they were found, would resemble the condition of the aborigines of America, far more than that of the Asiatics”. Various accidental voyages which had come to his notice tended to show that the Polynesians had indeed proceeded from east to west,

“for it is a fact, that every such voyage related in the accounts of the voyagers, or preserved in the traditions of the natives, has invariably been from east to west, directly opposite to that in which it must have been, had their population been altogether derived from the Malayan archipelago.” 21

The view that prevailing winds barred the path of eastward bound migrants had, however, long been brushed aside, by Cook, as has been seen, in 1769 and again by La Pérouse in 1785. According to La Pérouse,

“Westerly winds are at least as frequent as those from the eastward, in the vicinity of the equator, in a zone of seven or eight degrees north and south, and they are so variable that it is very little more difficult to make a voyage to the eastward than the westward.” 22

The bulk of the theorists had in any event long since opted for a western homeland. Cook had rejected the American theory as early as 1773, 23 Andia y Varela in 1774-5 and George Forster in 1778. The “Phoenicians of the Pacific”, according to them, had come from the west, against the prevailing wind.

- 18

In 1837, John Williams, a practical man with a wide experience of Pacific sailing, thought that the migration of the Polynesians from the Malay archipelago presented no difficulties. The distance of 7,000 miles could be covered in short stages, the longest being that between the Samoan and the Cook groups, a mere 700 miles. The Hawaiian Islands might be reached by way of the Marquesas and the voyage of 2,500 miles reduced by 600 to 800 miles. The passage from Tongatapu to New Zealand would present few problems; if the wind happened to be from the north-east it could be performed in a few days. As for the winds, every two months there were westerly gales for several days and in February there were what the natives called the westerly twins when the wind blew from the west a number of days together, then veered round the compass, and in the course of 24 hours came from that point again. Williams had frequently seen it continue for eight and ten days. As for the unfitness of native canoes, the progenitors of the South Sea Islanders would not have migrated in the paltry canoes they now used but in vessels similar to those in which the king of Achian attacked and sank Portuguese frigates in the straits of Malacca in 1573. Besides, there was good evidence that formerly the Tahitians and Society Islanders had canoes far superior to those now in use, in which they performed some extraordinary voyages. He could therefore see no reason for countenancing the American theory. He was moreover so convinced of the practicability of sailing from Sumatra to Tahiti in one of the large native canoes, that, if an object of sufficient magnitude could be accomplished by it, he would feel no hesitation in undertaking the task. 24

Williams's, however, was a lone voice. For the rest of the western school firmly held that such voyages could only have been involuntary. In 1797, for example, Wilson of the Duff asserted that the

“migrations of the fairer race from the Philippines to the Carolines and farther eastward, have almost certainly been occasioned by stress of weather, which drove their canoes from island to island, and from one group to another, that had not before been peopled. Frequent incidents of this nature have been ascertained, and some of them have been specified in an account of the islands connected to Otaheite. The population of islands so widely scattered, cannot for the greater part, be otherwise explained, either upon the ground of established fact, or upon that of probable conjecture.” 25

Beechey in 1831, on the basis of a single involuntary voyage in the Tuamotu, was disposed to allow only involuntary voyages as the explanation of the settlement of the Pacific, but in the two or three months of the year when the trade wind was replaced by the SW monsoon, that is to say, in the most dangerous sailing period in the year when wise men eschewed the sea. 26

“The temporary obstruction of the trade wind in these seas, by the westerly monsoons, has not been duly considered by those who represent the difficulties as insurmountable.”
- 19 In 1841, F. D. Bennett maintained 27 that
“these unsophisticated voyagers upon entering a contrary and boisterous wind are unable to make their intended haven . . . and may thus be driven, in distress, to any remote land to which chance may direct them. The occasional occurrence of such accidents, during a lapse of centuries, together with an adventurous spirit and other causes, which often actuate these islanders of both sexes, to proceed in quest of uncertain lands, may be deemed to account for a gradual advance of population to the easternmost islands of the Pacific, or even to the shores of America.”

A contemporary Presbyterian minister in Sydney, John Dunmore Lang, held a similar view. For him 28

“one of those accidents that have doubtless been of constant occurrence throughout the vast Pacific Ocean for three or four thousand years past, and that have served from time to time during the long period to people the multitude of the isles, a mere handful of Polynesians—fishing, perhaps, off the coast of Easter Island in the Southern Pacific—had been suddenly caught by a violent gale of westerly wind, and carried across the intervening tract of ocean to America, landing somewhere near Copiapo in the state of Chili, in South America; and that the descendants of these Polynesian unfortunates, carrying with them the long extinct, but comparatively high civilization of the South Sea Islands in long ages past, had, in the course of many succeeding generations, progressively settled the whole of the continent of America from Cape Horn to Labrador.”

A. G. Findlay in his Directory for the Navigation of the Pacific Ocean added the weight of his authority to the theory, though with reservations: 29

“Numerous instances are upon record of canoes being drifted out of their course, even several hundred miles, by currents and westerly winds; few narratives of voyages in the Pacific are without a notice of them; and they materially assist in explaining how remote, and perhaps very small islands, may have been peopled from the West against the direction of the prevailing wind.”

J. J. Jarves, writing in 1843, also thought that there was no difficulty in accounting for the islands being peopled; for it had been found that

“frail canoes and boats, either by accident or design, have performed voyages of sufficient extent to have arrived at the most distant lands . . . Canoes, crowded with occupants of both sexes, are annually picked up at sea, long distances from their places of departure, and drifting about at the mercy of the weather. The continent of Asia, from the numerous intervening islands, affords more facilities for reaching Polynesia in this manner, than America, though stragglers from the latter have doubtless from time to time added to the population . . .”

But the probabilities were greatly in favour of Asia, “both for certain affinities of tongue and from striking resemblances in manners, idols, clothing and physical conformation”. 30

- 20

The accidental voyage theorists had, however, now given rather too many hostages to their opponents. In 1861, W. T. Pritchard agreed that 31

“it cannot be doubted that the early migrations of the ancestors of these islands were involuntary rather than the result of roving dispositions, or of the pressure of limited and overpopulated homes; that, in fact, they were blown away from their earlier homes in their frail canoes . . . that, wherever their first home may have been, the races have been passing involuntarily from group to group, from island to island, through many ages . . . It is, however, remarkable that in all the many instances of authenticated driftings the course of drifted canoes has been from east to west before the prevailing trade winds, and not from west to east before the westerly winds which, though occurring less frequently, usually blow with greater fury than the trades. The natives do not usually venture out on their fishing or travelling expeditions in their canoes during a westerly wind, excepting always the voyage from Fiji to Tonga, when the weather is most carefully observed for some weeks before starting.”

It seemed obvious that if as Beechey said—and his view was generally accepted at this period—the Polynesians had been swept into the Pacific by occasional westerlies during the hurricane months from December to March, they must also have been swept back, and far more thoroughly, by the trades during the remaining nine months. The whole argument had become untenable. The only theory which would admit of the involuntary settlement of the islands of the Pacific was the American. And since no one was prepared to agree that the Polynesians had come from America, it became necessary to accept a theory of planned migrations.

Those newest scientists, the anthropologists, led the way. In 1860-1, in a lengthy debate in the Paris Anthropological Society, it was argued that the evidence, after all, did not rule out the possibility of migration, that is to say, of deliberate voyaging. If, indeed, windtossed castaways had been able to survive passages of hundreds and even thousands of miles and still make land, might not well-equipped expeditions have done likewise? Accidental voyaging had certainly played some part, but not the whole, in the settlement of the Pacific. The theorists had built too much on too little; too much evidence that was significant had been passed over. That which had recently been brought to light supported the possibility of migration either way. Both legend and modern observation showed that the natives waited for the wind they wanted. The Tahitians, for example, used a SW wind to go to Anaa, the Raiateans a W or a NW wind for Tahiti, the Samoans a SW wind for Penrhyn, the Tongans a SW or a S wind for Samoa; and, of course, the corresponding contrary winds to bring them back. 32

- 21

The Polynesians, too, had a much better idea of astronomy than was commonly supposed. Tupaia's sense of direction had never failed throughout the Endeavour's long voyage in many different latitudes and longitudes. The so-called frail barks of the Polynesians, it was now discovered, had been veritable little ships, capable of holding together in heavy seas and large enough to carry more than a hundred passengers. The Polynesians, moreover, were experienced and handy navigators, accustomed to making frequent and distant voyages. 33 The evidence seemed overwhelming. The isles of Polynesia and Micronesia had in short been settled deliberately, by migration. 34

The scientists, however, opted for no exclusive theory, nor can it even be said that they formed a school. The French merely asserted that accidental voyaging offered an incomplete explanation of the settlement of the Pacific. The later English-speaking protagonists of the deliberate settlement theory in their turn fully recognized the role of accidental settlement.

Thus S. Percy Smith, perhaps the most notable “traditionalist” and an adherent of the Indian origin of the Polynesians and of a western entry into the Pacific, readily admitted that “many of their expeditions, no doubt, failed in the end they sought and disappeared for ever”. 35

Elsdon Best in his monograph on Polynesian canoes thought it

“clear that two great causes have led to the settlement of the Polynesians over such a vast area of the Pacific Ocean—viz., voyages of exploration and colonization, and drift voyages. According to tradition the first settlers in the isles of New Zealand were castaways who drifted to these shores many centuries ago.” 36
“It is unquestionably a fact that drift voyages have been a very important factor in the settlement of the far-spread isles of the Pacific, and a considerable amount of evidence under that head has been collected.” 37

Though Buck dismissed a theory that had “been advanced that some islands were peopled by deep sea fishermen who were blown away while engaged in their lawful avocation” because women did not go out fishing in canoes with their men-folk, his views were thoroughly conservative. It was certain, he thought, that the first settlers set out on purposive voyages with a certain amount of food and water. Nevertheless, though they might have missed a near objective and been carried afar, their range of travel was limited by their supply of food and water and by human endurance. 38 The occupation of the Pacific was, for him, a gradual process marked by short voyages from island to island over some centuries of time.

And Hornell, excusably perhaps one of the most unabashed of the exponents of deliberate voyaging, who believed, with the knowledge we possess of some of the long voyages successfully undertaken by - 22 Polynesians between the 10th-14th centuries—Tahiti to Hawaii, 2,400 miles, and central Polynesia to Easter Island—that the voyage to the South American coast was easily within the power of adventurers manning a well-found and well-provisioned double canoe of adequate size, could also think that some of the most important discoveries were probably made by crews caught by violent adverse winds, blown off their course with no option but to run before the gale and sailing in a direction never contemplated. 39


The most recent version of the thesis has then every appearance of antiquity though it is undeniably the most extreme statement of the case that has so far been offered. According to this: 40

“Man originated as a land animal and presumably graduated through river craft to coastal vessels. The early off-shore voyages were in the land-locked seas of the Old World to extended coast-lines on the other side . . . For these journeys . . . navigation by the sun, stars, winds and currents was good enough, for when the voyagers came in sight of the coast, which they could not miss, they could pick their way along it.
“The islands which were within the land-locked seas, including the North Sea and the Arabian Gulf, and the chains of islands which were within easy distance of the continent, did not involve any difficulty . . .
“Ocean islands which were more than 300 miles from the continents, or from chains of islands extending from the continents, were in an entirely different class. In their case, there was no way of finding they existed by crossing between coast-lines, as in the land locked seas.” 41

And indeed, all but the farther Pacific Islands, Madagascar, and the Nicobars, were uninhabited at the time of later discovery—“the Falklands, Juan Fernandez, and scores of others besides the nearer and more desirable ones which remained unsettled for so long near the ancient maritime haunts of man”. The “mystery” of the first peopling of the farther islands is thus a “mystery of the Pacific and a few islands in the Indian ocean”. 42

Polynesia itself 43

“comprised a number of little worlds, inaccessible except by accidental migration. The limit of effective navigability was the distance that could be achieved in off-shore voyages of several days between islands where the conditions of wind and current were convenient for such journeys. The only islands which were in deliberate contact with others at a distance of more than 100 miles without intervening islands were Rotuma and the Ellice Islands (200 miles), the main western archipelago and Rotuma (300 miles), Fiji and Tonga (220 - 23 miles), Tonga and Samoa (150 miles to the Niuatobutabu cluster in between, and 140 miles thence to Savaii), Tahiti and the Tuamotus (170 to 230 miles), and Rarotonga and Atiu (116 miles). It was the circumstances of wind and current that determined the maritime achievements, rather than the gross distance. Areas in which lesser contacts occurred were Hawaii, the Marquesas, the Tokelaus, Manihiki-Rakahanga, and probably Rimatara-Rurutu. These journeys were the product of a long evolution and were highly impressive.
“All these separate worlds were settled by one-way voyages of isolated canoes. They should not be described as drift voyages . . . Within each area settlement was for the most part the result of accidental contacts in the first place. The development of the contact areas was the result of later deliberate voyaging within the range of effective navigation. The same fundamental facts applied to all isolated ocean islands throughout the world, including the Carolines and other North Pacific groups, and those in the Atlantic and Indian Oceans.
“. . . on occasion voluntary exiles, or exiles driven out to sea, were conveyed to other islands on one-way voyages in which precise navigation played no part.”

The Polynesian settlers of the furthermost islands of the Pacific were thus all accidental voyagers, arriving in small groups, sometimes merely of males who lived out the rest of their lives in celibacy or cast themselves adrift and were heard of no more; sometimes, though at extended intervals, of both men and women, either with or without plants and animals, but all involuntarily carried from their home shores and swept away before winds and currents to a new life on an unknown shore, or to extinction in the great spaces of the sea. On occasion, it is true, there were one-way voyages of exiles who pushed off with livestock, plants and implements. Such voyagers retained command of their vessels, and many of them would have sailed with the wind on their beam or quarter, thereby tending to go in the direction of Hawaii, New Zealand and other peripheral islands. 44 When land was sighted, they could make for it. This was exploration, although the discoveries, like all discoveries of new land, were not deliberate. Deliberate navigation to and from remote ocean islands was impossible in the days before plotting courses with precision instruments. 45 Moreover,

“accidental settlement by exiles who had made preparations and set out in the hope of finding new land, and by waifs of storms, could have achieved the peopling of the peripheral islands of Eastern Polynesia within a few centuries of the peopling of the central islands, since the winds and currents are relatively favourable to such voyages on the outward legs, although not on the return one”.

Deliberate voyaging is thus a “needless concept”, cherished for “unconscious emotional rather than scientific reasons”. 46

It follows that in prehistoric times neither the Polynesians nor any other people in the Pacific performed the feats of deliberate long voyaging and colonization with which they have been credited. The deliberate settlement of Hawaii and New Zealand by the Tahitians or other Poly- - 24 nesians is untenable. The “Great Migration” is a myth. Nobody could have arrived in New Zealand other than by accident, carried away by storm. The Chathams were also settled in the same way—for the most part from New Zealand—as New Zealand itself was settled from Polynesia. There is no evidence that any voyager ever got back from the Chathams to New Zealand in pre-European times. The notion that the Chathams could have been colonized from New Zealand becomes a flight of fancy in the light of local meteorology and the copious evidence of the inability of Polynesian vessels to make sustained voyages against variable or adverse winds. 47 The inhabitants of the Polynesian outliers in Melanesia, too, are all accidental settlers. No progress can be expected in the solving of the riddle of the extraordinary jumble of cultures and peoples in those islands until this fact is realized. 48

If the theory of accidental settlement does not throw any light upon whether the Polynesians came from some point or points in the New Guinea-Solomons-New Hebrides area, or from some point or points in Micronesia—it cannot indeed be assumed that any definite course of migration from island to island group into Polynesian ever occurred—it is clear that the Polynesians were dispersed from a suitable local centre. “Nor does one have far to look . . . Tonga and Samoa are just across the threshold of Polynesia from the approaches from the west and linguistic evidence shows the Tonga-Samoa area to have been an early ancestral Polynesian speech area.” In early times, accidental migrants, in whom brown straighter-haired people apparently predominated, came into the Tonga-Samoa area and set up the Polynesian language, culture and people, some of whom were carried progressively by accidental voyages to Eastern Polynesia and the farther Polynesian islands. It was only because Western Polynesia formed a bottleneck between the approaches to Polynesia and the eastern area, and was itself highly suited to the development of the Polynesian people and culture, that their broad uniformity occurred.” 49


Andrew Sharp's book, Ancient Voyagers in the Pacific, has undeniably made a deep impression, no doubt chiefly because it was thought to be new and therefore up-to-date. But like the earlier statements of the case, this one owes far more to its appeal to the “common sense” of the plain, wayfaring man, and to ancient prejudices which seem now to be deep-rooted in the European psyche, than it does to any genuine understanding of Oceanic life and culture, or, it should be said, to sound scholarship.

For one thing the argument is thoroughly tendentious in tone—the conclusion is stated in the first chapter and the reader is thereafter constantly plied with facts which, unless he is intimately acquainted with the sources, he cannot possibly hope to evaluate.

- 25

This latest statement of the case for accidental voyages omits and misinterprets material evidence, selects whatever suits the argument and leaves out or explains away what does not. For example, it quite inexcusably denies, no doubt in the interests of uniformity, the most irrefragable testimony concerning long distance voyaging in other seas in medieval times—the exploits of Irish monks in crossing 500-700 miles of sea to Iceland, Scandinavia and Spain, the great voyages of the Arabs and Chinese in the Indian ocean and the Chinese and Arafura seas in pursuit of a far-reaching commerce, and above all the achievements of the Vikings during more than 400 years in shaping their traditional star-guided courses between Norway and Greenland over 1,500 miles of north Atlantic ocean, 500 miles to the Faeroes and then more than 1,000 miles far out of sight of land. 50 It dismisses Polynesian tradition outright on the ground that it must be restored to its “pristine form”—whatever that might mean—before it can be of any use, though in fact traditions contain a great deal not only about deliberate voyaging but also about accidental voyaging to New Zealand and other parts of the Pacific which there seems no good reason to reject. 51 It is, as the sequel will show, far from impartial with more “reliable” sources, with Quiros or Cook or John Williams and the other early European navigators and missionaries upon whom the scholar must so greatly depend for his knowledge of a now vanished society. And it pays mere lip service to a departed generation of expert scholars whose like we shall probably never see again, so as to suggest to the unwary at least that they agree with its thesis rather than otherwise. It reposes a touching faith in “authorities” who support its views, Cook, Anderson, Dibble, Lang, Hale, Turner, Pritchard—the majority of whom are no authorities at all—and above all a miserable handful of dumb castaways whose misfortunes apparently speak louder than words. 52

Not only this, its basic assumptions are often quite untenable, if not wholly illogical. The Atlantic islands, it claims, though many of them were desirable and near at hand, were all uninhabited before being discovered by accident or by systematic exploration in and after the 15th century, and this presumably because early European seamen blown off course could always find their way back to their own extended coast. 53

Why the ancestors of the Polynesians did not also seek safety in such circumstances in the land from which they had originally come, is, however, far from clear. The only reason offered is that primitive Oceanic vessels were “ideal subjects for accidental voyages”. 54 There is - 26 in any case no evidence of long distance voyaging in the Pacific. Sharp has searched the literature and found none. It is of no use appealing to the undocumented past. Any theory that suggests that voyagers were going back and forth in some primitive era must explain why they were not in later. If it had happened by intent at that period, then the farther areas would have merged in a common contact zone as the knowledge of the courses and local conditions improved. Indeed, “if in fact the Polynesians ranged the ocean in earlier times, but stopped doing so in later, they went through a development which was the precise opposite of that of all other maritime peoples”. 55 Neither Micronesia nor Polynesia was settled by early miracles of deliberate navigation which no less miraculously stopped short in later times with complete unanimity from one end of the Pacific to the other, and which never occurred at all in the case of the rest of the world's islands. A voyaging thesis too must fit all the facts. It is useless to work out hypothetical courses for 90 per cent of the supposed deliberate colonists and leave 10 per cent floating about in the stormy expanse of the South Pacific without guidance or reliable winds. 56 This is perhaps why an indubitable case of a deliberate two-way voyage between the Carolines and the Marshalls is swept under the carpet as accidental both ways. 57 Sharp insists, too, on a conception of deliberateness in voyaging involving two-way journeys with three distinct passages which vitiates the whole argument from the start. In other words, though he is content to allow Bligh a single passage to Timor, he requires that the Polynesians must traipse backwards and forwards across the Pacific and have their comings and goings recorded by reliable European referees not once but several times. He thus dismisses all those who followed in deliberate one-way passages the long, low ragged flight of the birds to a certain landfall.

Further the book is quite unreliable on nautical matters: a weakness which reflects its preoccupation with theory rather than with fact, and overmuch reliance on secondary sources. It thus constantly exaggerates the difficulties involved in oceanic sailing. The habitual blue of the Pacific sky and its myriad and familiar bright stars, the constant delight of the Pacific visitor, are here too often obscured. Continuous sight either of the sun or the stars on long journeys is apparently a somewhat phenomenal event. 58 The currents are too baffling and too little susceptible of detection, the winds too changeable and violent. The birds fly unnaturally high and swiftly and are strangely silent; they can therefore give the navigator no help. 59 The low mournful note of the conch too soon recedes from the listening ear. The ocean is too deep for anchors! Hypothetical Polynesian squadrons could not have managed to keep together in colonizing expeditions as anyone might discover for himself merely by looking out of his back - 27 door on a squally or overcast night. 60 The Polynesian canoe, moreover, is oddly sluggish, unseaworthy and unweatherly. 61 The navigator cannot know exactly where an island lies or how to find it. The ocean is thus full of accidental voyagers and of their piteous if wholly imaginary fates. 62 It becomes hard in the end to believe that the Polynesians could have negotiated even the limited distances allotted to them. 63 Indeed, evidence to the contrary notwithstanding, the voyages of the Tongans to Fiji in historic times can only have been infrequent, mere hit and miss affairs, in which otherwise accomplished deep sea navigators apparently deliberately courted disaster on reefs and shoals in order to shorten the distance they must travel out of sight of land. 64

Nor is the book any safe guide in various other important matters. It is too prone to stress the differences in material culture, the absence of harpooning in one area, of fish-hooks in another, and to explain these as the result of one-way settlement and the localization of culture traits. It is quite obtuse on the question of adze types and the significance of typological studies, the implications of which are fatal to its thesis—as indeed its thesis is to systematic scholarship. 65 It is misleading on botanical and zoological questions, in particular on the manner of the transmission of animals and plants across the Pacific. It has pigs and dogs leaping aboard unoccupied canoes and disembarking on distant islands. 66 It asserts that weeds could only have been transported by accident; useful plants by contrast were “dribbled” along piecemeal over 3,000 years from group to group and from island to island. 67 And if it is ready to admit that the seedless breadfruit and banana must be propagated by shoots, that the coconut, the sweet potato and the taro all require very careful cultivation if they are to survive, as well as careful transportation, that as Cook has recorded, pigs were difficult to carry over long distances because of their susceptibility to cold, 68 it is much too ready to argue that their often haphazard distribution might be accounted for by accidental dispersal, even without human intervention, - 28 rather than by local conditions, the inadequacy of shallow atoll soils, the availability of other staple foods and by the custom of the natives to dispense with their pigs and dogs when food was low. 69


In the final analysis, the accidental voyage thesis owes its survival less perhaps to its inherent plausibility than to such factors as the haphazardness and incompetence of the early explorers as observers of native life, to the sudden and almost complete destruction of the ancient Oceanic culture and even of the Polynesian and Micronesian race itself in the great epidemics and civil wars of the 19th century, and to the comparative infancy, even yet, of Pacific studies, notably of archaeology, which might eventually lead to the solution of the problem: to the relative absence, in short, of any evidence by which it might be tested. The scattered facts that remain are nonetheless very impressive.

The inhabitants of Oceania, it is clear, had been from time immemorial denizens of the sea. They passed their days on heaving canoe platforms, or on small islands hemmed in by closely circumscribed clan boundaries or narrow atoll shores, surrounded by a vast ocean upon which they were almost entirely dependent. They thus viewed it with an indifference born of familiarity and of complete acceptance.

There can be no gainsaying their mastery of their environment. They excelled as swimmers and divers. Distances of 40 miles were well within their strength; 70 women with infants clasped about their necks passed effortlessly through even the most mountainous surf.

They were adept in the handling of canoes, both large and small. The sea was their highway and they naturally spent much of their time paddling or sailing about in the lagoon or within the reef, engaging in ceremonial processions, or visiting kinsmen at a distance. Where there was no reef the men normally fished out in deep water. The search for bonito might take them far out of sight of land and engage numerous fleets of double canoes equipped with small outriggers as “chasers”. On occasion, whole families might set forth on short inter-island passages to pay homage to a chief, to attend a feast or, as at Rakahanga, to take up temporary residence on a neighbouring island when food supplies had been exhausted.

They also crossed the seas in search of trade. The Tongans visited Fiji for sandalwood. 71 The Tahitians undertook annual voyages to Tahaa and Borabora to exchange tapa for bamboos filled with coconut oil. 72 The Tuamotuans visited neighbouring uninhabited islands in search of eggs. 73 The Marquesans and Tahitians and the Tongans made - 29 lengthy voyages for the sake of red feathers used in the cult of Oro. 74 In the 19th century, the people of the Carolines resumed their ancient expeditions to the Marianas, bringing with them shells, tapa, wooden vessels and cordage in exchange for pieces of iron, copper, nails and knives. 75 The Tongans carried huge stones by water and, according to legend, might venture considerable distances to look for material for royal tombs. 76 The natives of Yap in Micronesia went 300 miles to the Palau Islands for “stone money”, great calcite disks, six to twelve feet in diameter. 77 The atoll dwellers of Polynesia visited far-off high islands in search of suitable stones for adze-making. 78

Many sought fame or fortune in other lands. Some went out of sheer curiosity, shedding a few tears perhaps as Tupaia did in 1769, and then “showing no further signs of seriousness or concern”. 79 Cook describes a native named Hete-hete of Borabora who in 1774 visited Tahiti to see the sights and subsequently accompanied the expedition to Tonga, New Zealand, Easter Island and the Marquesas. 80 In 1797, Wilson set a Tahitian ashore at the Marquesas and was plagued by several Marquesans who wanted to go to Tahiti. 81 Others again, like Autourou, Omai and the prince of Peleliu, did the rounds of foreign cities, London, Paris, Boston, New York, or visited the new colonies in the antipodes. Many never settled down. Lucatt (Lucett) says of his “chancellor”, a Hawaiian, typical of many of his race at the time, 82 that “from his childhood he has been sailing about in whale ships . . . At the close of voyages he has been set ashore at different islands, and in this way I came to pick him up at Rapa.”

Casual voyagers might spread their sails in the confidence that a more propitious wind would eventually bring them home. Mariner describes the adventures of a Tongan Kau Moala whom he saw return from Fiji after an absence of fourteen years on an “excursion”. After two years' absence he had set sail for home. But the wind proved unfavourable at Vavau and the sea being high he ran for Samoa. There he encountered another heavy gale which drifted him down to Futuna where the natives, as custom decreed, stripped him of all his possessions. His canoe was broken up, his sandalwood destroyed. He spent the next twelve months building a new canoe, the people setting him up again with goods and provisions and directing his course to Fiji where he proposed to lay in another cargo of sandalwood. His crew now consisted of 35 of his own people with 14 or 15 women and four Futunese men who wanted to go to distant lands. At Rotuma, he took on three more women and again set sail for Fiji where he eventually arrived safe and sound. Here, at Viti Levu, he took up residence with a chief and - 30 engaged in wars with other islands; later he joined forces with a chief of Bau. At length, having taken part in a gargantuan feast in honour of victory in which there were heaped up 200 human bodies, 200 pigs, 200 baskets of yams and 200 fowls, he sailed for home to a hero's welcome. 83

Others, younger members of chiefly families, like the sons of the medieval Norman knight, Tancred de Hauteville, saw no chance of advancement at home, organized expeditions and set out on long exploratory voyages. Diapea refers to the voyages of Tongans 84

“which they sometimes make to the extent of 600 or 800 or even 1,000 miles—being not infrequently absent a year or two from home, wandering and gadding from island to island, going all through Samoa, Fiji, and all the ‘Friendly Islands’, not omitting even the more distant ones of Wallis' Island, Futuna, Nieuafou, Nieuatobutabu, as well as the three nearer groups of Tongatabu, Vavao and Hapai.”

Atoll dwellers in particular seem to have been continually on the move. Lucatt (Lucett) notes 85 that

“the inhabitants of nearly all the Paumotu Islands are of roving migratory habits; they wander from island to island in their large double canoes, so that at times an island will appear to be thickly peopled, and at others scarcely an individual is to be found. The latter was the case with Katin in the present instance, and was more than once so with Bow or Hao Island during my stay there; and unless a person dwelt amongst them long enough to be able to recognize features, he would be apt to return a very false estimate of the island population.”

The missionaries fulminated in vain against these tendencies. Bishop Bataillon of Uvea, for example, complained of the migrations of the Uveans that they were “at present the principal wound of these missions, and we force ourselves by all possible means to combat and cure it”. 86

From time to time, the ocean was filled with refugees from natural disasters, volcanic disturbances, tidal waves, drought and famine. Lucatt (Lucett) describes several large double canoes at Rotuma which were used in former times when the population exceeded the means of support or when this seemed likely. Oracles were consulted and at their instigation a party set out to look for new land. 87 In the Marquesas, droughts often lasted several years together. “It was in such times that canoes ventured out in search of other islands and that priests went from house to house and in a squeaking affected voice informed the people that their friends had found a land abounding in breadfruit, hogs, coconuts” and invited others to follow them. 88 In this way, the Marquesans came to be indefatigable land seekers, frequently sailing - 31 away in large well-equipped fleets, well-stocked with food and water, pigs, poultry and young plants. According to an eye-witness, in only a very few years “more than eight hundred men, women and children . . . left this and the other islands of the Washington and the Marquesas Group . . .”. 89 Most of these were never heard of again. It is perhaps significant that the Marquesans “imagined [Cook's ships] came from some country where provisions had failed; and that our visit to them was merely for the purpose of filling our bellies . . . a circumstance which puzzled them exceedingly, our having no women with us . . .”. 90 Such a long voyage, it might be argued, was not beyond the capacity of a well-equipped Marquesan expedition of the time. 91

Quarrels and disagreements took others abroad. Some went for religion's sake. The followers of Tane, according to Buck, left Tahiti rather than bow to Oro and settled in the outer fringes of the Cooks. 92 In the 19th century, the heathen often fled from Christian islands. Diapea describes a large Tongan canoe manned by eighty young blades which had been absent on a twelve months' cruise to all the outlying islands. 93 Most of the time had been spent at Samoa,

“a place they used to go for the purpose of cheating Tuekanokopolu's law by having the much-coveted breeches tattooed on their persons, and which heathen custom had been, some time previous to this, interdicted by this said king, by the advice of the missionaries throughout the Friendly Islands, and so on their return they easily submitted to the fine, knowing that the breeches could not be pulled off again, whereas if they had remained at home they could never have been put on, and then they would have run a terrible risk of being estimated in the eyes of the fair sex about the same as if emasculated”.

Many were set adrift in rafts or small canoes for crimes and peccadillos of various sorts, murder, adultery, insults, breaches of etiquette, even juvenile incorrigibility or mischievousness. 94 Men of abnormal physical strength or influence of whom people were jealous or afraid might be exiled. 95

On occasion, a considerable number of people might be deported together. Some time before Cook's arrival in the Society group in 1769, the chiefs of Tahiti and of a number of adjacent islands banished thieves and other criminals not deserving of death to Borabora

“which before the commencement of the law, was almost barren and uninhabitable, which practice continued several years. In process of time, however, their numbers so greatly increased, that the island was insufficient for settlement. Being men of desperate fortunes they made themselves canoes, turned pirates . . . One of them Opoone [sic] made himself their chief or king and by frequent acquisition of prisoners, - 32 eventually became strong enough to impose his rule on neighbouring islands, amongst them Ulietea [Raiatea].” 96

Warfare also contributed largely to the volume of marine activity. The Tongarevans constantly crossed their lagoon to raid one another's coconut plantations. 97 The natives of Anaa periodically descended upon their neighbours, ravaging all the islands of the Tuamotu, burning, looting, and slaying. 98 Bougainville describes the natives of Tahiti as constantly at war with the inhabitants of the nearby islands; 99 Cook records an immense Tahitian armada manned by 7,760 warriors in motion against Eimeo. 100

In the 17th and 18th centuries the Tongans scoured the seas to Fiji, Samoa, Uvea, Niue, the Tokelau, the Gilbert and Ellice Islands and even as far west as Rotuma, the Banks and Duff groups and Tikopia and Sikaiana. 101 In the 19th century, following the great civil war of 1797 which these activities had helped to foment, they concentrated upon Fiji, conquering and settling all the windward islands, and eventually, under the notorious Ma'afu, dominated affairs in the greater islands as well, feasting, brawling, robbing, plundering and murdering much as the ancient Vikings had done in the British Isles and in Carolingian Gaul. 102 The Tongan archipelago was meanwhile wracked by incessant conflict, great fleets descended upon various islands in pursuit of vengeance or lordship.

The native histories of the Pacific are full of references to the flight of defeated parties before their enemies. 103 D'Urville states that Samoa and the Fiji islands were often invaded by whole populations of fugitives. 104 Many never returned. Their coconut and breadfruit trees were cut down and their villages razed. They might, moreover, be remorselessly hunted down and eventually driven to far distant places. On occasion, those who feared defeat might make elaborate arrangements for flight. Porter describes the preparations made by a chief of Nukuhiva in 1811 to convey himself and his people to another island. The threat of war, however, disappeared and the many large canoes which had been specially constructed for the anticipated event were dismantled and stored away for possible future use. 105

- 33

Inevitably, a proportion of this vast maritime traffic met disaster or misadventure. The sea has always been a hard mistress. The legends speak of considerable numbers of fishermen swept away singly, or in twos and threes, in sudden storms, in some places more frequently than in others. The annals of Tubuai record only four victims of accidents while fishing, three from recklessness, the other from weakness. 106 But Burrows describes boatloads of Futunan fishermen borne out to sea and never seen again. 107 There are many accounts, too, of interisland expeditions being carried away, of Rakahangans who somehow failed to make Manihiki and drifted down to Pukapuka or Nassau. 108

Some of these unfortunates drifted about for many days and were propelled immense distances across the Pacific. Père Paul Clain records in 1697 the arrival of 29 people at Samar in the Philippines after a voyage of 70 days from the Palau islands. One of their number, we are told, had formerly been on Mindanao, but “found the way back to his country”. 109 Chamisso records a voyage of 2,300 miles from Yap in the Carolines to Aur in the Rataks. 110 Kotzebue records a voyage lasting “eight moons” made by a native named Kadou and three companions between Ulea in the south-east Marianas and Aur in the Ratak group, 1,500 miles east. 111 Hersheim records unintentional voyages of 1,200 miles made by the Marshall islanders and one by two Gilbertese in which they covered 680 miles in 10 days. 112 Towards the end of the 19th century, three Gilbertese drifted ashore at Urak—the modern Mokil in the Carolines—after a voyage of six or seven weeks. 113 Woodford describes Gilbert and Ellice islanders at Ontong Java, 800 and 1,000 miles distant respectively, and also castaways from much further away, from Ocean Island. 114 And as recently as 1951, seven Marshallese in a small engineless schooner from Kwajalein, their compass put out of action by atomic radiation, landed in the New Hebrides, 1,800 miles away, after 100 days at sea. 115 In the South Pacific, Peter Dillon mentions an incident in which ten Aitutakians bound for Rarotonga finally made Tonga after five months at sea. 116 Natives of Rotuma often drifted down to Samoa, 600 miles to the leeward. 117 Others drifted in the opposite direction, to Tikopia. 118 In 1820, a canoe from Rurutu fetched up at Maurua, 20 miles west of Borabora, a voyage of 800 miles accomplished in 15 days with the aid of a strong W or SW wind 119 - 34 Pritchard records drifts of at least 800 miles between Tongareva and Samoa. 120 John Williams attributes the evangelization of Atiu in the Cooks to a party of native teachers, lost on their return to Tahiti, who finally reached that island six weeks later. 121 He also describes the arrival in 1832 at Manua of a native of Raivavae, 2,000 miles away, after three months at sea. 122 The missionary John Inglis records the wreck at Aneityum in the New Hebrides in December 1866 of a boat which had set out on a voyage from Futuna to Somosomo in Fiji and, missing that group, had come sailing on its original course, apparently with the aid of a compass, a distance of 700 miles in all. 123 And in 1884, eleven natives of Mangareva who had left their island in consequence of a disagreement with the missionaries sailed for Fiji and finally made Sikaiana, 3,700 miles away, a feat, it might be thought, which deserves to rank with Bligh's as one of the greatest deliberate one-way voyages in the history of Pacific navigation. 124

It is difficult to be sure how many “accidental” voyagers ever reached safety or survived to found a new settlement. Lucatt (Lucett) describes a father and a young family, a son and three daughters, caught in a gale in a small canoe and swept away to Hereheretue in the Tuamotu, an unoccupied and infertile islet where for lack of suitable timber they were obliged to remain. 125 As the children reached puberty, the son took his sisters to wife and raised a family by them all. Eventually, the settlement was overrun and absorbed, the men however dying soon after apparently from jealousy. Others reached inhabited islands where they intermarried. A warrior leader like Sualo of Samoa who drifted ashore at Tongoa in the New Hebrides about 1825 and subsequently settled down on Efate after the death of most of his crew from fever might succeed in establishing some sort of hegemony in a new land. 126

But the mortality must often have been high, even on the shortest passages. 127 On one occasion, a party of Rakahangans, two men and seven women, struck on the reef at Manihiki. Two women drowned from exhaustion; the two men were later killed in a fight. 128 Castaways were rarely made welcome and were as often as not massacred or at best banished to a remote part of the island. William Diapea states that the Cikobians of Fiji once slaughtered some 30-40 Tongans who had drifted to their island. 129 The Vanikoroans are said to have killed 35 of a party of 50 Tongans; the rest escaped to sea. 130 In 1889, a European, William Harris, and a party of Nauruan boatmen who had consumed too much gin were washed up in the Solomons. Harris died - 35 the day before the boat arrived; the natives killed the survivors except for one woman whom a trader sheltered. 131

Many died at sea of hunger or thirst. In February 1858, the whaler Mercury picked up 13 natives, men, women and children, 190 miles distant from the nearest of the Line Islands, “left to the mercy of God on the waves, without a paddle or an oar to help themselves with, and nothing eatable or drinkable on board their little vessel”. 132 They must otherwise have perished. Captain Worth of the Rambler describes another party of 27 natives without either sail or paddle, still floating about in mid-ocean after three moons. Half of their number were already dead. 133 Many, too, went mad and fought or threw themselves over the side. 134 Others drowned. West describes the loss in 1835 of 70 Niuatoputapu people aboard one of two canoes caught in a tempest which raged for six days, the survivors being picked up by the other canoe which had nonetheless managed to remain by it throughout. 135 Occasionally, whole fleets were swallowed up in disasters which must have brought voyaging in various parts of the Pacific to an end for years at a time. In 1830, 100 Micronesian canoes were overwhelmed and only one escaped. In 1860, 35 Marshallese canoes set out from Jaluit for Kili, 150 miles away, and were all utterly lost; and another 22 canoes disappeared in the course of a passage from Majuro to Jaluit. In 1885, 10 canoes on a voyage of 300 miles also went missing. 136 Something of their fate may perhaps be guessed from the horrible catastrophe described by Diapea when a Tongan canoe in which he was travelling broke up in a hurricane, his native companions being eaten by sharks and he himself deposited more dead than alive on one of the Lau islands. 137 And many succumbed to malaria, contracted in the unhealthy islands of the west, and left their companions stranded without hope of return.

It is obviously dangerous to read, as some have done, too much into the meagre examples of accidental voyaging that have come down to us. Even a perfunctory examination of the evidence shows that they involve for the most part tiny outrigger canoes or European boats and small groups of people often quite unversed in the arts of navigation and unprepared to cope with the exigencies of a long involuntary sea passage.

The Pacific islanders in fact were skilled seamen and shipbuilders. In the 18th and 19th centuries, the Micronesians crossed the seas in the flying proa. The Polynesians voyaged in double canoes of two fundamentally different types. 138 In the first, the hulls were equal or nearly equal in size, usually with a mast stepped forward of midships. The heads and sterns were dissimilar. These predominated in Hawaii, - 36 the Tuamotu, the Society Islands, Manihiki, the Marquesas, the Cook Islands, New Zealand, and until the 18th century at least, also in Western Polynesia. In the second type, the Fijian, which later became predominant in the western islands, there was a marked disparity in the size and form of the hulls.

Some of the more primitive ships, the Raiatean va‘a, for example, were provided with only two strakes—carvel-built as was the case throughout Oceania—so that, in the event of baffling or contrary winds, they could be paddled more effectively than was the case with some other types. The storage room of the Raiatean va‘ was in consequence comparatively limited. It has been computed that it could hold no more than twenty days' supply of provisions at a time. The Samoan va‘a tele was a much larger canoe with a strong deck and a thatched house surmounted by a small platform on which sat the chief. It was designed to carry two va‘a alo or fishing canoes. A va‘a tele figured by Schouten shows a characteristic cock motif on its sail. The Tahitian pahi was a large two-masted double canoe, rigged with crab claw sails very much like those used in Hawaii, and fitted with high curved sterns. The Tahitians also built war canoes, mere fighting platforms, which in combat were laid side by side. The great state canoe of the Tahitian chief, the tipairua, which seems to have been that most frequently employed in deep sea voyaging, was characterized by a low head and high upturned stern. It could be propelled both by paddle and by sail. The Tongans sailed in tongiaki, equal-hulled canoes. Perhaps the finest canoe-builders were the Tuamotuans. For lack of timber of adequate dimensions they built their ships of short planks, sewn patchwise together.

But the foremost achievement of the Pacific canoe-builder's art was the Fijian drua, the model of the later Tongan kalia, clearly Micronesian in conception, in which the second hull was little more than an outrigger and which was so rigged as to permit of the mast being raked at either end at will, so that it could sail both ways. Like all the other double canoes that have been mentioned, it was steered by a great oar which in rough weather was reinforced by another on the opposite side.

It has been well observed that the word “canoe” is a misnomer which has served to prejudice the European attitude to the whole problem of Pacific settlement. 139 The flying proa often measured as much as 75 feet, the tipairua and the tongiaki 67-70 feet. The largest Marquesan canoes accommodated 80-100 warriors. 140 And the canoes even of the isolated atoll of Tongareva might attain 60-70 feet and carry a complement of 100-200. 141

Many Polynesian vessels achieved immense proportions, comparable with those of at least some contemporary Europeans' ships. Cook's Endeavour, for example, measured 106 feet in length and was 29 feet 3 inches in extreme breadth. There are several records of Polynesian ships measuring upwards of 100 feet. Fornander describes the remains of a Hawaiian double canoe 108 feet long. 142 Cook has a reference to a - 37 double canoe at Tahiti whose hulls were each 108 feet long. 143 The Fijian camakau often attained 100 feet: the missionary Thomas Williams mentions a drua which was 118 feet long and 50 feet wide, with a mast 68 feet high and yards measuring 90 feet. 144 Wilkes records a camakau of 102 feet. 145 A member of the crew of the Glide saw several double canoes at Fiji, each as long as his ship and carrying from 300 to 400 men. 146

Something of the capacity of such a ship might be gauged from a statement by Wall that a whole pig could be roasted in the open cooking place on the deck and that a man could easily walk in the hold without bumping his head on the superstructure. 147 Forty to 100 men were required to manage its sails. 148 On one occasion, we are told, a canoe of this sort carried twelve head of cattle in her holds for a distance of 120 miles and another a cargo of bagged maize sufficient to load the Alarm and the Xerifu, two ketches of 30 and 20 tons burthen respectively. 149

In 1774, Cook saw at Tahiti no fewer than 160 large war canoes and 170 sailing canoes; in 1777, 210 war canoes and a vast attendant navy of tipairua only slightly smaller in dimensions. 150 Nor were these large canoes built only in the 18th century. Father Martin de Munilla, the Father Commissary on Quiros's voyage of 1606, describes several very large canoes in the Santa Cruz group, obviously Tongan in pattern, one measuring 12 baras, about 66 feet in length, and equipped with a deck with room for 60 persons and a storehouse, like the store of a mill, where provisions were put when setting out on a long voyage “which sometimes lasts 24 to 30 days”. 151 Such a ship carried as much sail as the Spanish launch. Its water supplies, according to Prado, were kept in great bamboos containing ten gallons each, strapped to the underside of the platform. 152

The sailing qualities of many of these vessels appear to have been little short of marvellous. Robertson of the Dolphin thought the Tuamotuan canoes sailed “about one-fourth faster nor our ship when we made all sail we could conveniently carre”. 153 Cook describes Tongan canoes “everyone of which outruns the ships considerably”. 154 Wilkes speaks of the velocity of the camakau as almost inconceivable. 155 In favourable weather it was capable of 12 to 14 knots. A captain Berry - 38 once estimated the speed of a drua at 15 knots within three points of the wind. 156 Thomas West describes a journey he took in a Tongan kalia: 157

“Up went the huge sail, down went the great steer oars, splashing into the sea, and away we shot like a racehorse. The breeze was strong. Every timber of the canoe creaked again, while the mast bent like a reed and cracked in its socket as if it would split the deck in two . . . the sea was like a hissing cauldron on either side of our course, and the vessel, instead of having time to mount over the smaller waves, cut its way right through them.”

The ship covered a distance of 38 miles in three hours. Grimble once logged a Gilbertese baurua at 18 sea miles in five minutes over the hour. 158 A. F. Ellis records that a large Gilbertese sailing canoe was capable of 16 knots or more. 159 The Micronesian flying proa is said by several witnesses to have been capable of 22 knots for sustained periods. 160

Nor is this all. There is abundant evidence to show that the ships of Polynesia and Micronesia were much more weatherly than contemporary European vessels. Bougainville remarks of some Samoan canoes that “though we ran seven or eight knots at this time, yet the piraguas sailed round us with the same ease as if we had been at anchor”. 161 Cook describes double canoes at Nomuka “with a large sail, that carried between forty and fifty men each. These sailed round us apparently with the same ease as if we had been at anchor”. 162 At Tonga,

“while we were plying up to the harbour, to which the natives directed us, the king kept sailing round us in his canoe. There were at the same time a great many small canoes about the ships. Two of these, which could not get out of the way of his royal vessel, he run quite over with as little concern as if they had been bits of wood.” 163

The much-maligned tongiaki, according to Hornell, 164 was

“so clumsy and ill-designed that it could not beat to windward; when a favourable wind failed, there was no alternative but to change course or drift with the sail down, with one of two results—either the crew perished or they landed as castaways on some distant island.”

- 39

It nonetheless served the Tongans well enough in many long voyages. But the drua which the Tongans finally adopted in place of the tongiaki was normally able to continue a given course, “even against a headwind”; 165 some at least could also beat up against the wind with the greatest of ease. According to West, the drua could sail within three points of the wind; Wall adds that it could beat to windward at four knots. 166 Hornell, however, remarks 167 that

“the one course the ndrua could not sail on was with the wind directly aft; the great weight of the sail, held down at the tack to the head of the canoe, combined with the pressure of the wind upon it, caused the canoe to run under instantly when set square before the wind.”

Gilbertese and Caroline ships were even more handy. Anson noted 168

“how dextrously they are fitted for ranging this collection of Islands called the Ladrones: since as these Islands bear nearly N and S of each other, and are all within the limits of the trade winds, the Proas, by sailing most excellently on a wind, and with either end foremost, can run from one of these Islands to the other and back again, only by shifting the sail, without even putting about; and by the flatness of their lee side, and their small breadth, they are capable of lying much nearer the wind than any vessel hitherto known, and thereby have an advantage, which no vessels that go large can ever pretend to: The advantage I mean is that of running with a velocity nearly as great, and perhaps sometimes greater than what the wind blows with.”

Chamisso was once astonished to see the brig Rurik on which he was a passenger obliged to tack while “the canoes of the Ratak islanders kept a straight course, outran the European ship and then let their sails down to wait for her”. 169 The master of the Morning Star saw in 1860 two large canoes which had beat up 100 miles dead to the windward to Kusaie in five days and which, after waiting for a fortnight for a fair wind, finally started off another 300 miles “with the fresh trade wind blowing directly ahead”. And Woodford has described how a European sailing ship in which he made a voyage to the Gilberts took nine days to make Apamama from Kuria, a distance of only 15 miles to windward. In the event, because of a strong current running to the west, it proved necessary to run from Kuria in 30″ N to about 4° N. The morning on which the ship eventually reached Apamama, nine large canoes, one of them over 70 feet long, arrived from Kuria, having beaten up the night before against the wind and current. 170 As Cook said, “They were vessels fit for distant navigation.” 171

- 40

The navigational skill of the people of Oceania has won universal if not always whole-hearted respect. Basically, it might seem to have been founded on that almost uncanny sense of direction which early man perhaps shared with birds and other animals whose existence depends upon their capacity to traverse long distances to and fro, winter and summer, across apparently featureless wastes of forest, desert or ocean and which civilized man has so largely lost. In so far as it was a cognitive process, directed to the solution of the specific problem of crossing and re-crossing a great new sea, it was probably only slow. The first explorers no doubt came into the Pacific in pursuit of fish such as bonito and in particular of migratory sea-mammals, notably whales, which in the nature of things find their chief sustenance in the vast plankton-rich zones lying across the Pacific in the region of the Line and to a greater or lesser extent in the neighbourhood of the islands of Micronesia and Polynesia. 172 Thus Maui literally fished up numerous lands, among them New Zealand, where the sperm whale converged upon their fertile waters in search of food or shelter. In this early period the discovery of islands was perhaps merely secondary. Even in historic times small groups of fishermen still visited uninhabited islets, moving on as the fish migrated elsewhere and leaving behind them those few scant traces of their brief occupancy which have excited so much speculation. 173 In the subsequent search for land which their progressive penetration of Oceania and their new found dependence upon agriculture made necessary they must have relied to an increasing extent upon migratory birds, in particular the long-tailed cuckoo, the sooty shearwater, the golden plover and possibly also the godwit, whose flight, spread out over several weeks, they had long studied. 174 In this way, they gradually traversed the Pacific, consolidating their knowledge as they went. They turned to good account, too, their observations of the colour and temperature of the water, the apparent direction of currents and the leeway of the ship which they knew how to measure, the odour and the loom of the land, the reflection of the shallow waters of an atoll in the sky above, the interruption of the swell in the neighbourhood of an island, the red glow of an active volcano at night, and the habits of sea-birds. 175

The Marshall Islanders made a painstaking study of currents, tabulating their accumulated knowledge of the run of ocean swells on - 41 elaborate charts fashioned from twigs bound together with sennit. Curved strips indicated the altered direction taken by swells when deflected by an island, their intersection the nodes where these met in a confused sea, one of the most valuable indications of the whereabouts of the voyager. Currents were shown by short, straight strips; long strips indicated the direction in which certain islands were to be found. The islands themselves were located by small cowry shells tied to the framework. Such charts, of course, were not made to scale and were little more than mnemonic devices intelligible only to their individual owners. They might, however, serve to dispose of the legend that the Pacific islanders were unable to judge the set and run of the sea. 176

The evidence shows, too, that the earliest navigators relied heavily on the wind. Like the Arabs, the Polynesians divided the horizon into a greater or lesser number of points, the Tahitians into 16 parts, the Cook islanders 32, to each of which there corresponded a wind. According to Andia y Varela, 177

“when setting out from port the helmsman reckons with the horizon thus partitioned counting from E or the point where the sun rises; he knows the direction in which his destination bears: he sees, also, whether he has the wind aft or on one or other beam, or on the quarter, or is close-hauled: he knows, further, whether there is a following sea, a head sea, or if it is on the bow or on the quarter. He proceeds out of port with a knowledge of these [conditions], heads his vessel according to his calculation, and aided by the signs the sea and wind afford him, does his best to keep steadily on his course. This task becomes more difficult if the day be cloudy, because of having no mark to count from for dividing out the horizon. Should the night be cloudy as well, they regulate their course by the same signs; and, since the wind is apt to vary in direction more than the swell does, they have their pennants, [made] of feathers and palmetto bark, to watch its changes by and trim sail, always taking their cue from a knowledge of the course from the indication the sea affords them.”

With experience born of a growing acquaintance with the great ocean and perhaps of a more settled life than the pioneers had known, both the Polynesians and the Micronesians came to rely increasingly on the stars. The entire Polynesian economy indeed depended on a knowledge of astronomy. The Polynesians knew a great number of stars by name so that “the cleverest among them could tell in what part of the sky they were to be seen in any month when they were above the horizon”. 178 Forster noted of Tupaia 179 that he

“was so skilled in this [knowledge of the stars and points of the compass] that wherever they came with the ship during a navigation of nearly a year, previous to the arrival of the ship at Batavia, he could always point out the direction in which Taheitee was situated.”

According to Wilson, 180

“the men are excellent judges of the weather from the appearance of the sky and wind, and can often fortel [sic] a change some days before - 42 it takes place. When they are going to any distant island and lose sight of land they steer by the sun, moon and stars, as true as we do by compass. They have names for the fixed stars and know their time of rising and setting with considerable precision . . .”

Pritchard remarks 181 that even “involuntary” migrants

“invariably preserved correctly the direction of their lost homes; the trade wind, and the rising and the setting of the sun and moon, being their unerring indicators.”

And Coote describes seeing the elder of three small boys from Santa Cruz aboard the missionary schooner Southern Cross

“teaching the names of various stars to his younger companions, and was surprised at the number he knew. Moreover, at any time of night and day, in whatever direction we might happen to be steering, these boys, even the youngest of the three, a lad of ten or twelve, would be able to point to where his home lay; this I have found them able to do many hundreds of miles to the south of the Santa Cruz group.” 182

Traditional evidence indicates how extensively astral navigation was practised even in the earliest times. Ancient Hawaiian sailing directions inform the navigator bound for Tahiti that he will discover new constellations and strange stars over the deep ocean. When he arrives at Piko a Weka he will lose sight of Hokupaa (apparently the North Pole star) and then Newe—the Southern Cross—will be the guiding star and the constellation of Humu will stand as a guide above him. 183 Other instructions specify Canopus, Orion's Belt, Sirius, Venus, the Pleiades and the Milky Way. The later directions were much more explicit. Eighteenth century sailing directions for the voyage from Niuatoputapu to Tonga, for example, laid it down that Haliatoloa, Luaae-tangata and Taipa should be followed. In sailing north again, the voyager was advised to steer by Faipa, Aloua (two-oared boat), Telengehue, Fatanalua, Tupa and various other stars. 184

Instructions of so detailed a nature might seem to suggest a considerable advance in technique so that navigation rapidly became a specialized profession. In Polynesia, two experts accompanied canoes on long voyages and passed the night in examining the stars. 185 These faatere were competent to navigate over long distances, steering by the rhumb of the particular star known to rise or set over the island to which they were going.

In Micronesia, a single expert might suffice. The master of the Morning Star describes the arrival at one of the Marshall islands in 1860 of upwards of 40 persons, men, women and children, without a compass or nautical instrument of any kind. He

“lent them a compass, and Dr Gulick who was somewhat acquainted - 43 with their language, explained its use, at which they were very much astonished, but pointed out an old bushy-headed veteran, whose eyes looked out from under his shaggy locks, saying that his head was ‘all same as compass’.” 186

Little is known about the manner in which they learned their craft. In the more democratic Gilberts, the Marshalls and the Carolines, youths were drilled until they had become word perfect in long lists of stars by which courses might be steered to various lands. They then graduated to navigational schools conducted in the great house of the village, the roof with its purlins serving as an imaginary sky on which the position of the constellations was marked. 187 Every constellation was thus allotted its place in the thatch according to its angular distance from Reigel and its declination north or south of that star, which the pupils, clustered round the central pole of the building, must learn line by line. Only then were the neophytes taken to the eastern beach and introduced to the open heavens. 188 Hawaiian youths were taught with the aid of a gourd. 189

In essence, the system depended upon a concept of a series of domelike heavens resting on the earth and forming concentric horizons on its surface. It consisted in laying a course direct to a given destination by keeping the bow of the vessel pointed towards a star near the horizon whose bearing corresponded to the direction of the destination, that is to say, along a great circle route. 190 In addition to these navigating or “horizon-grazing” stars, other stars were used to determine latitude, in the words of Kepelino, a Hawaiian scholar, stars “which are suspended in turn over each land as Hoku-lei over Hawaii”. The mariner thus knew when he had arrived at the latitude of his destination by noting where a particular star passed overhead during the night and by studying the directions of the wind and of the waves whether his objective lay east or west. 191 It was customary to begin a voyage in the early evening when it was dark enough to see the brighter stars, but still sufficiently light to discern the landmarks against the sky. 192 The missionary, John Williams, describes the procedure by which he at last found Rarotonga. 193 The chief of Atiu, Romatana, who informed Williams of its direction

“at one time pointed in one direction, and at another in quite the opposite. But this was soon explained; for the natives, in making their voyages, do not leave from any part of an island as we do, but invariably have what may be called starting points. At these places they have certain landmarks, by which they steer, until the stars become visible; and they generally contrive to set sail so as to get sight of their heavenly guides before their landmarks disappear.”

- 44

Williams adds:

“Knowing this we determined to adopt the native plan, and steered our vessel round to the starting point. Having arrived there, the chief was desired to look to to the landmarks, while the vessel was being turned gradually round; and when his marks on the shore ranged with each other, he cried out ‘That's it! That is it!’ I looked immediately at the compass, and found the course to be SW by W and it proved to be as correct as if he had been an accomplished navigator.”

Before losing sight of the landmarks, the navigator substituted two conspicuous stars for the marks, one near the horizon over the bow of the canoe, the other 180° away over the stern. 194 Noting the direction of winds and waves, he maintained this course during the day, checking it constantly by reference to the changing bearing. When night fell the ship was “aligned between the two stars in the same declinations and having the same diurnal paths as the bow and stern stars of the early morning”. 195 Thus in any given course navigation depended upon a series of stars. “No small part of the equipment of the old captains [of Tonga]” we are told 196 “was the ability, based rather on experience and judgement than on rules, to determine when to shift from one star or constellation or to set course by another group.” The Tikopians employed a similar procedure in sailing to Anuta. According to Firth, 197

“they set out at the end of the day in order to run through the night and . . . maximise their chances of landfall by daylight. They carefully chose a favourable wind. The best wind for Anuta, the Ariki Kafika told me, is the Susako, from the south-east. With this wind the canoe is steered straight on its path, and there is no need to set the bow off to make allowance for drift.”

Having taken sight, so to speak, on their objective the navigators kept their mark in sight as long as possible.

“But the major navigational guide is the Star-path, the ‘Carrier’ (Kavenga). This is a succession of stars towards which the bow of the canoe is pointed. Each is used as a guide when it is low in the heaven; as it rises up overhead it is discarded and the course is reset by the next one in the series.”

That the system was effective cannot be doubted. It could hardly otherwise have been invented. But it may be suspected that, unlike the older rough and ready, rule of thumb methods, still no doubt employed by fishermen and casual travellers, it was particularly suited to shorter distances such as might be regularly traversed by traders or to direction finding where the target was something of a “slippery stone” and accuracy was therefore vital. Its defects are obvious enough. Theoretically at least, it ruled out the possibility of cruising about unless the direction of a number of other islands was known. Furthermore, like any system of astral navigation, it might be of little use in bad - 45 weather; 198 though this was perhaps no great risk in that halcyon age of voyaging when meteorological conditions were probably far better than they are now. And like all highly developed systems, it was apt to go wrong, particularly in the hands of the inexperienced, and too vulnerable, especially so in Polynesia when in the great social changes of later times navigation, and canoe-building, came to be monopolized by a handful of guildsmen who all too often took their knowledge with them before they had had time to pass it on. 199


It is incontrovertible that the Polynesians and the Micronesians were superb seamen, that they were well versed in the arts of navigation, that their ships were adequate to undertake lengthy voyages, that they had reached both Hawaii and New Zealand in long-distance passages which only prejudice or surmise could dismiss as invariably accidental. But can it be shown, if only for the sake of argument, that they ever successfully essayed deliberate two-way voyages involving at least three separate journeys beyond the often traversed 200-300 miles which the accidental voyage theorists illogically represent as the extreme effective range of the Oceanic canoe, or that, as commonsense might suggest, navigational skills which sufficed over shorter distances might not also have served had Maui fished up his islands five or six hundred or even a thousand miles apart?

The extent of the geographical knowledge of the Polynesians and the Micronesians is perhaps eloquent enough on this point. In 1595, the Marquesans, observing a black man amongst the Spaniards, informed Mendaña that to the south of their island there was a land inhabited by “black men”—the reference is undoubtedly to the suntanned atoll-dwellers of the Tuamotu who were usually several shades darker than their compatriots on the higher islands—and that these were their enemies; that they used the bow and arrow; and that the big war canoes lying in the bay were destined for a war upon them. 200

The chief of Taumako in 1606 described a very great number of islands whose direction he pointed out together with the distance to each. 201

Wilkes mentions a native of Raraka who drew with considerable accuracy a map of all the islands of the Tuamotu he was acquainted with and their relative situation; “his knowledge of the western part of the group was quite surprising.” 202

- 46

Arago describes a celebrated pilot of the Carolines who, by means of grains of Indian corn,

“contrived to represent all the islands of his archipelago, and to mark their relative positions with wonderful ingenuity. He named every one of them; designated such as were easy of access, and told us what were the productions of each; in short he neglected nothing which could prove to us that he was acquainted with the geography of one part of the Pacific.” 203

Tupaia knew the names of above 70 islands “at most of which he himself has been”. 204 In particular, he had a considerable knowledge of the Society Islands. He could also enumerate and give directions for nearly all the islands in the Austral and Cook archipelagoes, many in the Tuamotu, and others farther afield, including Rotuma and possibly also Niue, and some which appear to belong to Fiji, a large number of which were not at this time known to Europeans but which were subsequently identified and found to correspond with Tupaia's list. 205 Even if we accept Anderson's view—and he never knew Tupaia—that the Tahitian chief had merely picked up his information from accidental voyagers, he nonetheless still knew where they were. In other words, the direction of islands could be communicated to others and recorded.

The Tongans, too, told Cook of upwards of 150 islands in the Friendly group, though he himself saw only three during the course of his third voyage. He then heard of Samoa, Vavau, and Fiji for the first time, and no European had been on any of them. 206

Various involuntary voyages in the 19th century, moreover, suggest that the Polynesians and Micronesians yet retained some capacity to engage in long distance two-way voyaging at need. Beechey, for example, records the discovery of a party of people from Anaa on Byam Martin's Island, Ahunui, in the Tuamotu, 400 miles to the east, where they had been swept by a storm during the course of a voyage to Tahiti. Seventeen of them had died of thirst. After many months' preparation they turned for home. The leader of the party, Tuwarri [sic], whom Beechey conveyed back to his home island, showed the greatest apprehension during the course of the voyage. “He could not imagine our motive for pursuing so indirect a course, and frequently enquired if we were going to his native land, and if we knew where it was, occasionally pointed in the direction of it.” 207

A. F. Ellis describes how in 1886, Dick Harris of Nauru and some natives drifted for several weeks to what they supposed to be New Ireland, 900 miles to the west. There they secured some coconuts and set out for home. They eventually made Ocean Island, 1,000 miles - 47 away, where they decided to wait until the steamer called. They had not missed their goal by very much. 208

In the period 1879-84—the actual date is uncertain—three Gilbertese drifted ashore at Urak (Mokil) in the Carolines. 209 Two survivors were subsequently returned to the Gilberts. One of them, Shekin, later made an intentional voyage from his home island with another Gilbertese, Towe, and handed over his whale boat to the missionary, apparently as a token of his appreciation of the services of the Ponape missionaries in getting him back home after his earlier venture. Weckler remarks that he thought it interesting that his Mokil informants seemed to take this second trip in a very matter-of-fact way. “When I inquired of my informants why anyone would be foolhardy enough to risk a second voyage of this length, I was told, ‘Oh Shekin had been over the route before, he knew the way’.”

Ellis records an incident in 1887 involving the recovery of a large mooring buoy at Enderbury Island in the Phoenix group. 210 It was arranged to send the company's steam launch.

“They ran their distance, but there was no sign of the island. Search was made in various directions, without success . . . One of the natives had been a sailor on Mr. Arundel's vessel, the s.s. Explorer, and he remembered the courses steered for that port, about 900 miles distant, and made the harbour without difficulty.”

The proved range of deliberate long distance two-way voyaging in both Micronesia and Polynesia can only be regarded as impressive. In Micronesia, voyages of 300 miles without intervening islands, that is to say at “extreme range”, were frequent. In Eastern Polynesia, the Tuamotuans seem frequently to have ventured considerable distances. The presence moreover of great voyaging canoes at Tongareva in the middle of the 19th century when local energies seem to have been wholly absorbed in an embittered and constant contest for local food supplies would suggest that these isolated folk had only recently abandoned long distance voyaging to Rakahanga and no doubt also to Tahiti and Samoa, a mere 800 to 1,000 miles away.

The Society Islanders knew not only the Tuamotu but had visited all the more fertile islands of the Austral group and the Cooks. They had, moreover, according to two impeccable witnesses, Autourou and Tupaia, ventured deep into Western Polynesia, upwards of 1,400 miles away, on deliberate two-way voyages. Autourou informed Bougainville that the longer voyages of the Tahitians occupied 15 days. Tupaia's account of the matter is wholly circumstantial. Cook 211 notes in 1769 that

“Since we left Ulietea Tupia hath been very desireous for us to steer to the westward and tells us that if we will but go that way we shall - 48 meet with plenty of Islands, the most of them he himself hath been at and from the description he gives of them they must be those discovered by Captain Wallice and by him call'd Boscawen and Kepple Islands, and these do not lay less than 400 Leagues to the westward of Ulietea; he says that they are 10 or 12 days in going thither and 30 or more in coming back and that there Pahees, that is their large Proes sails much faster than this ship; all this I believe to be true and therefore they may with ease sail 40 Leagues a day or more.”

And Cook was not deceived. In the 19th century various lesser navigators, John Williams, numerous whalers and even ship-wrecked sailors, all proved that it could be done, that Tupaia had been, as he said, to Samoa, Tonga and to Fiji.

Perhaps the final proof of the point, however, derives from evidence concerning the nature and extent of native sailing in eastern Melanesia in the 17th century.

The local histories of such islands as Sikaiana clearly show that the Polynesian outliers were frequently visited by canoes from Samoa and Tonga. On one occasion a Tongan chief, Waioma, led a party of 100 men to Sikaiana. Against the advice of a Samoan, a member of an earlier expedition who had remained behind, the natives supplied the visitors with food. The Samoan meanwhile sought out and killed one of the Tongans whom he knew as a bad character and a fight ensued in which the Sikaianans were defeated, most of them being killed. The Tongans sailed off soon after, taking with them Semalu, the chief's son, and many girls. They then visited Taumako “where they were at first well received, but the natives, having heard from Semalu the account of the occurrences at Sikaiana, fell upon them with bows and arrows and exterminated the whole party. Descendants of the Sikaianan girls taken to Taumako by the Tongans are said to be still living there.” 212

The Sikaianans also tell of a famous navigator, Kaidakita, who made many voyages from Sikaiana in an outrigger canoe, obviously of Gilbertese pattern. According to Woodford, he went from Sikaiana to Mala, from there to Gao on Ysabel and thence to Laina, probably on Choiseul. Here the natives broke up his canoe and he was obliged to build a new one, this too being destroyed. Finally, he managed to repair it and sailed for Ontong Java and so back home. He subsequently made other voyages from Sikaiana to Taumako, Tucopia [sic], Nukulonu (?) and Tinakula, which he correctly described as a fire island. He appears also to have visited Rennell. 213

And this story is no myth. In his Narrative, Quiros describes how the chief of Taumako, Tomai, pointed out for him the direction of more than 70 islands, including a very large one named Manicolo, lying to the SE, the SSW, the W and the NW. 214 Quiros subsequently learned from one of the natives he took at Taumako, a Sikaianan soldier-weaver who had been enslaved by the Taumakoans in a recent war, the names of these islands, among them Chicayana, Guaytopo, Pilen, Nupan, - 49 Fonofono, Mecaraylay, Manicolo, Tucopia and Pouro. Several of these are readily identifiable. Chicayana, we may agree with Codrington, was Sikaiana, Guaytopo Vaitupu of the Ellice Islands, Nupan and Pilen the Reef Islands of Santa Cruz. Mecaraylay is, I think, probably Nukulaelae of the Ellice Islands whose people Pedro vividly described, Fonofono is not Funafana, the modern Niulakita, but Funafuti which, as Quiros's informant remarked, is divided into many little islets. 215 The “large country named Pouro”, as Guppy and Woodford have argued, was indeed San Cristobal where the name describes the coast at the western extremity of the island and a small peninsula on its northern side.

The island of Manicolo has been almost unanimously identified with the modern Vanikoro, but this is plainly erroneous, as Pedro's description of it shows. A couple of islands of unequal size, the first less than thirty miles round, the other scarcely nine, and two or three islets surrounded by an expansive reef thirty miles in circumference and one or two miles wide, described by one of its inhabitants as merely a tenth of the size of Santa Cruz, could hardly have been referred to by those intrepid Taumakoan voyagers as a great land. 216 Prado argues that Tomai meant New Guinea and Torres agrees that the chief pointed to the west. 217 Their interpretation, however, naturally owed something to their subsequent discoveries. It is at least significant that such experienced mariners should have thought it credible that the Taumakoans might have been there and found their way home again. Leza says that “the great land with cows”—the reference is probably to pigs' tusks or whales' teeth—lay ten days to the south (Father Kelly informs me that Markham mistranslates at this point) and Quiros, citing Pedro as his authority, asserts that the “great land” lay to the south or south-east. According to Quiros, 218

“Pedro much extolled the magnitude, populousness, fertility and other things of that country and he and other natives went to it in one of their canoes in quest of the trunk of a large tree of the many which are in it to make a piragua and that he saw there a port and estimated it was longer and the entrance narrower than that of the bay of St. Philip and St. Jago and that he observed the bottom was sand and the shore shingle as the other I have described and that it has within four rivers and many people and that along the coast of that land they went to the westwards a greater way than from Acapulco to Mexico without seeing the end of it and returned to his island.”

G. Forster argues that Manicolo was the modern Malekula of the New Hebrides, Cook's Mallicollo, and indeed but for its minuteness, Port Sandwich might seem likely enough. 219 It is patent, however, that Pedro's bay is Natewa Bay, a large calm-watered port forty miles long and seven to ten miles wide, the only port in all these seas which fits - 50 the description, 220 his Manicolo Vanua Levu in the Fiji group to whose windward shores the canoe-builders of Western Polynesia commonly resorted in search of large trunks of Intsia bijuga, the vesi tree, a much valued hardwood used in the construction of hulls and strakes. It can hardly be coincidental that one of the districts of Vanua Levu is named Wainikoro, that the first and last sight our voyagers had of the Fiji archipelago was Cikobia, a small island 22 miles north of Udu point, itself inhabited by Polynesian folk, their starting place Tikopia, 720 miles to the north-west. 221

The evidence thus shows that until quite recent times, Polynesian seamen were still making deliberate two-way voyages without benefit of intervening islands of from 720-840 miles—the latter the distance from Taumako to Nukulaelae—and of 1,270 miles with a single intermediate stop, from Sikaiana to Fiji, and 1,400 miles from Raiatea to Niuatoputapu, sufficient, it might be thought, to have brought all the major island groups of the Pacific, the Chathams and New Zealand not excluded, within the range of their great double canoes. And in that same period, Micronesian proas were frequently going backwards and forwards over distances of at least 1,000 miles between the Carolines and the Philippines and the Marianas. 222


Much stress has been laid on the significance of a score or so of accidental voyages in the “dark age” of Polynesian history following the coming of the European. A review of almost 70 separate instances of small boat voyaging in the Pacific undertaken for the most part by wrecked American whalemen in the 19th century and varying in length from 300-3,500 miles is therefore relevant here. 223

In the nature of things, the early European sailor in these waters was peculiarly liable to sudden disaster, shipwreck at night, dismasting, burning, mutiny, native attacks, assaults by whales. The Pacific, however, was a kindly ocean. As a contemporary put it, these events were usually not “attended with that immense loss of life which attends the loss of vessels in the Atlantic Ocean. Never were waters more rightly named than when this ocean was called PACIFIC.” 224 Generally speaking, too, the survivors, many of whom were Polynesians and Micronesians, were habituated to extended passages at sea, often of several years at a time, and to long days in open boats, cutting in whales. And above all, they knew the Pacific. It was said that “from the Southern Pacific to the Indian and Chinese seas there is scarcely an Island, a rock - 51 or a reef, which are known that they cannot put a finger on; many, indeed, are known to them alone.” 225 They were thus seldom at a loss.

Where circumstances permitted, a small vessel might be built from the wreck, often only with the roughest of tools. Thus in June 1837, the captain and the mate of the Gledstanes, wrecked on Kure in the North Pacific, set off for Hawaii, 1,460 miles distant, in a small schooner caulked with lime and seal oil. 226 In 1843, the crew of the Shaw, wrecked on Ponape in the Carolines, spent four and a half months building a small boat in which they all embarked and eventually made Guam, 1,020 miles away, in eleven days. 227 In September 1844, a small schooner put together from the remains of the Holder Borden which struck on Lisiansky reached Hawaii, 1,000 miles distant, in 23 days. 228 In May 1846, part of the crew of the Konohasset which also went to pieces on Lisiansky, reached Honolulu in 42 days in a vessel of eight tons manned by the captain and six men. 229 In April 1860, the chief and second mates of the Lexington, left behind on Kusaie in the Carolines after the rescue of their ship-mates, built a “singular craft” in which, having abandoned their original intention to make for San Francisco, they squared away with the trades for Hongkong, 39 days and 3,200 miles distant. 230 And in August 1869, the notorious Bully Hayes, after spending three months lengthening and decking his long boat from the remnants of his ship Samoa on the reef at Manihiki, set sail for Samoa with a complement of 28 men, women and children, European and native, “huddled together on a nine ton leaky ship, besides a very large Newfoundland dog and a very small poodle to go a distance of some 800 miles, besides towing the Roma's long-boat astern”. 231

The most interesting example of the kind concerns the Wild Wave, wrecked in March 1858 on Oeno. 232 After staying on the island for ten days, the captain set out in one of the boats for Pitcairn, 65 miles away, only to find that the inhabitants had removed to Norfolk. He was also unfortunate enough to lose his boat in the surf while landing. His sole remaining resource was a few rusty tools left behind by the islanders. But he and his companions set to work. For lack of saws they used axes, hewing each board from a single tree. They managed to secure a few nails by burning down a couple of houses; for the rest they used wooden pins. They thus painfully built a schooner 30 feet long with a hold three feet deep and nine feet wide and decked over. She can scarcely have looked seaworthy for three of the men refused to sail in her. But after 4½, months the John Adams was launched, the intention being to make for Tahiti by way of Oeno where 32 of the crew still - 52 remained. In the event, however, a severe gale from the west lasting three days forced the vessel so far east that it became impossible to fetch the island and the captain, still secretly clutching his bag of gold, bore away for the Marquesas, 1,300 miles distant, where he took ship for home.

More frequently, such voyages were undertaken in the ship's boats. Perhaps the most famous of these was the epic journey of the men of the Essex, rammed and sunk by an infuriated whale in November 1820 in 120° W, 00° 40′ S. 233 The crew set out in three boats, ill-advisedly enough, for South America, but with the wind blowing steadily from the E and SE they were obliged first to steer for Ducie Island in 24° 40′ S, 124° 40′ W, 1,720 miles off. Here, leaving behind three of their more timid ship-mates virtually to starve, they set out again for Valparaiso, another 3,530 miles distant. Eventually, one of the boats disappeared. In 34° S, the captain and the mate also parted, the mate being picked up only after 60 days, the captain after 90 when eight of his crew had already been sacrificed for the sake of food and only he and a boy remained alive.

In January 1835, 18 survivors of the Corsair, cast up on Tabiteuea in the Gilberts, after losing the captain, the mate and the surgeon and six men in skirmishes with the natives, set out in three boats for Guam, 2,310 miles distant, and made it after 27 days. 234

In December 1847, the Maria Helena's launch was dispatched from Starbuck Island, set out in three boats for the Society group. 235 All went well until the third night when the second mate and his entire crew fell asleep for three whole hours. The chief mate meanwhile disappeared and eventually turned up at Tahiti, 925 miles away. The remaining two boats then ran into heavy weather during which the sails had to be taken down and the oars put out as a sea-anchor. After 12 days, they made Borabora with the aid of numerous flocks of tame birds which flew west at sunset and which they followed until 10 o'clock before heaving to. They were then, however, obliged to change course because of a strong SE gale and the lack of a keel. They thus finally reached Manuae in the Cook Islands, having covered 1,400 miles in 19 days.

In December 1847, the Maria Helena's launch was dispatched from the site of the wreck off Christmas Island 1,270 miles to Honolulu, the voyage being accomplished in 22 days. 236 In December 1848, Captain Hall of the Emerald, wrecked on Fakarava, made Tahiti, 300 miles distant, after four sleepless nights and days, the rain falling continuously in torrents. 237 In February 1849, two separate parties from - 53 the Gem, wrecked on Suvarov Island, reached Samoa 500 miles away, the mate with a full set of instruments, charts and a compass, the captain, Worth, with a flotilla of three boats and 16 men, apparently with a quadrant and a compass only. 238 The crew of the Ceres wrecked in June 1849 in Torres Strait set out in four boats for the Booby Islands, 600 miles off, and reached them in 13 days. 239

In August 1849, the Speculateur was suddenly thrown on her beam ends by a squall in boisterous weather and filled, only two small, flat-bottomed boats being saved. Since pumping was hopeless, the boats were lashed together in the fashion of a double canoe and rigged with blankets. Then, the “sextant, Norrie's epitome, compass, etc.” having been obtained by diving into the cabin, the crew bore away for Caroline Island which, however, they missed, passing, they thought to leeward. They also missed Staven's, Flint's and Bauman's [sic]. They then searched without success for Suvarov Island where they hoped to pick up pieces of the wreck of the Gem to repair the boats before they should break up under the stress of the continual heavy seas. At length, they gave up and altered course for Samoa which they reached on 18 September, completely exhausted, swollen limbed and covered with sores from lack of movement, having covered 1,800 miles in 26 days. 240

In March 1853, the captain with three men of the Chatham, wrecked on Tongareva, blithely set off in a “crazy” boat for Samoa. His promising voyage, however, came to a sudden end at Manihiki, 300 miles to the east, when two of the crew refused to go any further. 241 In May 1854, a boat from the Townsend, burned in latitude 35° S, reached safety at Juan Fernandez, 660 miles distant, though with only twelve survivors, one of the boats having been swamped, the other two capsized. 242 In January 1855, 29 survivors of the Logan wrecked on Conway Reef in 21° 40′ S, 174° 40′ E, reached Fiji in three boats after a voyage of seven days. 243 In October 1855, the captain and nine men of the Julia Ann, a cargo ship with 40 passengers wrecked on uninhabited Fenuaura (Scilly) in 16° 31′ S, 154° 43′ W, landed at Tahiti, 350 miles dead to the windward after 14 days' incessant toil at the oars, night and day. 244 In May 1859, the captain and seven men of the Virginia wrecked on Baker Island proceeded to Fiji, 1,270 miles distant, in the long boat. 245 In 1861, the captain of the Norna wrecked on Oroluk in the Carolines set off in his boat with his wife and child and seven men for Guam, six days and 820 miles away. 246 Early in 1866, the mates and the cooper of the Harvest, burned by the pirate ship Shenandoah, left Ponape in the Carolines in an open whale boat and made Guam, a distance of 1,000 - 54 miles. 247 Again, in May 1866, the crew of the Hornet, burned at sea in 2° N, 135° 5′ W, set out in three boats for the American mainland. They all kept together for 19 days when they parted by agreement, the captain subsequently deciding in view of the heaviness of the weather to run for Hawaii with the wind, and eventually reached it with a half-starved crew after 48 days, having traversed at least 2,050 miles. The other boats were never seen again. 248 In March 1868, following the wreck of their vessel La Belle on Wake Island, the mate and 21 other persons, including two women, members of an English opera company, crowded together in a boat 22 feet long to undertake a voyage “of 1,525 miles, subject to equinoctial storms, calms, and a tropic sun, with short rations, and an ocean studded with hidden rocks and coral reefs”. 249 And they made Guam 13 days later though the mate's reckoning of the longitude was 6 degrees in error. And in October 1870, Lieutenant Talbot and four of the crew of the U.S.S. Saginaw wrecked on Ocean Island in 28° 20′ N, 178° 10′ W, made Hawaii, a 30 day voyage of 1,000 miles during which they encountered three severe gales and lost all their oars. 250

On 6 August, 1842, the captain, first officer and four men of the Cadmus wrecked on Morane in 23° 26′ S, 138° W, set out for Tahiti “without a chart, book or any kind of nautical instrument, excepting a boat compass; all our instruments having gone to pieces with the ship”, steering N by W in order to run out their latitude first. For two days they had a head wind. It then turned south with squalls of wind and rain. On the next three days the weather was fine with breezes from the S and NE. On 13 August they experienced strong breezes, with a heavy cross sea; at night it was so heavy that a sea anchor was put out. On the 14th the weather was again fine and they steered to the NW again and saw land. This they passed by, not knowing what it was. On the 15th, judging themselves to have run sufficiently far north, they turned west. On the 16th, having traversed 900 miles in 9 days, they again saw land to the north and a French brig at anchor. To their surprise they found that they were three degrees north of Tahiti and two degrees to the west of where they had supposed themselves to be. 251

Occasionally, the navigation of these boats was only of the sketchiest kind. In November 1832, part of the crew of the Mentor cast up on Peleliu (Palau group) in the previous May set out “in an open whale boat with provisions and water for 20 days, without any means of finding my situation at any time, only having a compass, and 600 miles from any land I could stop at with safety, Ternate being the nearest port”. Nine days later they made Tobi, 360 miles distant. 252

On 31 August, 1870, Captain Vandervord of the China tea clipper Dashing Wave wrecked in the middle of the night on Wake Island - 55 managed to secure a chart and nautical instruments but strangely enough no compass. “Every day the captain got the boat's position but when the weather was dull of course they went in all directions for want of a compass and especially on cloudy nights.” It was the master's intention to make for the Gilberts but the current intervened. A course was then steered for Ponape “and had any one been able to row they might have made it, but weak as they were, all they could do was to keep their boat before the wind with the blanket sail”. At length, after 31 days at sea, the 13 men in the long boat reached Kusaie, a matter of 750 miles. Vandervord subsequently made another attempt to reach the Gilberts but was driven back by a gale. 253

Navigation difficulties were not, however, always entirely disabling. In April 1854, for example, the crew of the Canton, wrecked on 4 March on the island to which the ship gave its name, set out in four boats for the Gilberts, 800 miles off. The boats were kept together at night and in the daytime fanned out as much as possible to look for land or for a sail. Handicapped by bad weather and the lack of correct instruments—the captain's observations were out by 4′ in latitude and 1° 17′ in longitude at Canton—they missed these islands and bore away on the same course for Guam, the four boats reaching Saipan together on the 45th day and Guam itself on the 49th after a voyage of at least 3,080 miles. 254

There are more remarkable examples of the kind. In 1793, Captain John McCluer, a voluntary settler at Peleliu in the Palau, embarked in his boat with three Malays and two of his own slaves to go to Ternate. The weather beyond the islands proving unpropitious, he decided to proceed to China instead through what is now the Balintang channel. Touching again at Peleliu for some coconuts, he stood away to the north and in ten days reached the Bashi islands. Finally, after braving very bad weather in the China sea, he reached Macao. He and his companions had no food but coconuts and water; nor had they any instruments or charts for their guidance “except a single chart of Captain Wilson's; they spoke nothing but the Pelew language during the passage, it being the only language they all understood.” 255

In November 1818, the survivors of the Resource, wrecked in 28° N, 180° E, set off in the whale boats while it was still dark. The long boat meanwhile overturned and was found next day with three men still clinging to it. Thirteen men were then allotted to each boat, the captain with the compass being in one, the mate, so it appears, with a quadrant and sextant only in the other. On 22 November the boats put away for Agrihan in the Marianas in 19° N, 146° E. That evening the captain was lost and the compass with him. When no sign of him was seen, the surviving boat put away to the SW. The weather proved pleasant all the way but the men suffered terribly from thirst, several - 56 dying. At last, after having covered 2,325 miles in 25 days, they made Agrihan, the island coming up a mere 25 miles SSW when first seen. 256

In March 1829, Captain Keating of the Dash, wrecked on Ngulu in the Carolines, shaped his course by the sun and stars alone and on the third day, admittedly after a voyage of only 220 miles, reached the Palau. Here he procured a compass from a chief and with four companions embarked for Manila, traversing the 850 miles to San Berndardino Strait in nine days. 257

That this was no accident might appear from yet another famous episode. On 23 June, 1849, while hunting whales in 3° N, 104° W, Captain Homer and five men of the bark Janet capsized. Their ship meanwhile sailed off on another course. In the effort to bail the boat, two men were lost overboard. Two of the survivors were seized with delirium. At length, after 48 hours' immersion in the water, Homer raised a bit of a sail and steered north-easterly for Isla de Coca on the Peruvian coast, 1,180 miles away. On the seventh day they cast lots as to which of their number should be sacrificed to prolong the lives of the others. On the 8th day, another member of the crew died from exhaustion. On the 9th they got a shower and a dolphin leapt aboard. Several tame birds were also secured. Finally, on 13 July, after 20 days, their desired haven came into sight. “Being without compass, Captain H. depended for guidance on his observations of the North Star, and the rolling swell of the sea from the South.” 258

In June 1849, the mate and five men of the George and Susan lost the run of their ship in 4° 30′ S, 113° 30′ W. It is not clear whether they had even a boat compass but they bore up and stood for the Marquesas and after 13 days made Eiao, 1,900 miles distant, the trades with them all the way. 259

On another occasion, an Englishman, William Diapea, sailed with some deserters from an American whaler from Futuna (Horn Is.) to Lomaloma in Fiji, 200 miles distant. “The wind shifted and blew like ‘Old Harry’”, and the sea seemed formidable. But Diapea made land as he had intended.

“ . . . the only watch or even compass we had for this precarious kind of navigation, excepting the stars, by night, and as one constellation or single star dipped below the horizon, so we took another higher up, each night, till daylight broke in the eastward. The wind and the heave of the sea were quite a helpful guide in our calculations.” 260

Much later, in 1901, Captain J. C. Voss and N. K. Luxton, a journalist who had never been to sea before, sailed across the Pacific from Vancouver in an Indian war canoe, the Tilikum, measuring only 40 feet in length, 6 feet in the beam, and 3 feet in depth. In heavy - 57 weather, they used a sea anchor as the Polynesians before them had done. They reached Tongareva in 58 days. At Niuafoou they struck on the reef but were carried safely over by the next wave. At Fiji, Luxton's place was taken by a newcomer named Begent. Begent, however, fell overboard during the night of 28 October while attending to the compass light and was never seen again. Nor was the compass, and the rest of the voyage had to be completed without it. Voss was alone. Bad weather followed and the boat had frequently to ride out heavy seas with a sea anchor. In a newspaper interview, Voss said: 261

“Having lost the binnacle, I had nothing but the stars to steer by night, and I would have managed right enough but for the gales. For many days and nights I was drifting about at the mercy of the elements, and I did not know where I was. Occasionally I picked up a star and made a good course, but then the weather became bad and all I could do was to put the drag out and let her drift. The gales burst upon me with terrific fury, and I had the drag out three or four times a day. For four days I was without any sights, and for three days I was in a violent northerly gale.”

Voss nonetheless made Sydney safely though not before a current had swept him 50 miles to the south. His comments on the last part of his journey are of some interest. 262

“Sailing without a compass for 1,200 miles, as I did after I lost my mate some 120 miles southward and eastward of New Caledonia, I was frequently beset by high seas, and was several times blown out of my course. In the daytime, I steered by the sun, and at night by the stars. If you are steering by the sun in the daytime, and the weather threatens to become thick, take notice which way the sea is running, and, in southern latitudes, if the sea runs towards the sun at noon, you will know that it is from the south. Therefore, if you wish to sail west, keep the sea on the left side of your boat, or if to the east, on the right side of your boat. In this way it is possible for a fairly correct course to be steered without a compass.”

The last word on all this might be left to the missionary, John Williams. Though he knew little about ship-construction or the art of navigation, he set to work and built the Messenger of Peace, 263 a little vessel of 70 or 80 tons, 60 feet long and 18 feet wide. The trees were split with wedges and the natives adzed the boards with small hatchets. For lack of nails he relied on wooden pins. The rigging he fashioned from hibiscus rope and he used matting sails. In this contrivance he sailed from Rarotonga to Tahiti, a distance of 800 miles, the wind throughout being fair and the sea as smooth as a millpond. 264 On another occasion he sailed from Rurutu to Tahiti, a distance of 350 miles, in a matter of 48 hours. 265 In 15 days, he sailed about 1,700 to 1,800 miles to the eastward, an instance, he thought, perhaps unparalleled in the history of tropical navigation. 266 Later again, he made another visit to Raiatea and in the course of five days he ran 800 miles - 58 to the west without shifting sails. 267 His opinion that the Polynesians would have found no difficulty in reaching even the most distant islands of the Pacific is clearly entitled to respect, a view which Cook's own easy passages, especially from Tahiti to New Zealand and from Tahiti to Hawaii in January 1778, light winds, fine weather and birds all the way, amply confirms. 268


The Pacific, then, is far safer than it has been made out to be. The stars do shine, the winds are reliable enough. Flotillas of even the smallest canoes could easily have kept together in the longest of passages. Lateral drift is a landsman's bogy which in practice can be surmounted and which is in any case a hazard the seaman takes for granted.

The supposed “inadequacy of all craft before the days of large sailing ships” is a myth. The Venetians traversed the Mediterranean in large sailing ships long before necessity drove them out into the Atlantic. Their successors circumnavigated the globe in ships infested with teredo and built so high fore and aft that they drifted like balloons. Modern sailors boldly strike out across the Atlantic and the Pacific single-handed in mere cockle-shells. In fact, as an expert in the handling of small boats has said, when it comes to deep sea passages, size does not matter. 269 And certainly, as has been seen, one could, if one chose, travel almost any distance in any weather in even the craziest of conveyances. The Polynesians and the Micronesians were under no such necessity and to pretend otherwise is perverse.

There remain various other objections of a more or less fundamental sort which might be generally if briefly discussed here. The appeal to the supposed shortcomings of Polynesian navigational methods will not serve. The “elementary difficulties of primitive navigation” have been invented, not forgotten, by Europeans. And this is not surprising. As Professor E. G. R. Taylor has pointed out, the fundamental sources of time-keeping and of position-finding are still the heavenly bodies; the sailor must even yet con the sky. But the ordinary human being has long been separated from any direct reliance upon the stars. He depends rather upon experts, astronomers, mathematicians, almanac-makers, instrument makers, and he thinks only in terms of clocks, maps and calendars. “And because he himself never has occasion to consult the sky, he finds it impossible to imagine how the seaman makes his confident way across the pathless and unpeopled ocean waters.” He thus easily concludes that the ancients hugged the coast, fearful of venturing out of sight of land. 270

The point might be put another way. 271 European seamen graduated from lead and line and rule of thumb into a particularly difficult - 59 school where ancient methods were useless. For their world had been invaded by scholars, by Ptolemy, Euclid, Mercator, by landsmen, that is to say, who saw the question in a very different light. Simple-minded sailors had now to solve not the comparatively simple problem of running in a general direction towards an island or a particular point on a coast—which they knew well how to do—but of establishing its position to a nicety upon a flat map in terms of the intersection of lines of latitude and longitude. They had also to learn how to plot their precise whereabouts at sea, in brief, how to navigate. This inquiry occupied the best brains of Europe for more than 300 years during which men nevertheless continued to find their way across the seas. Even the invention of the chronometer did not wholly settle matters since seamen had still to learn to sail along the hypotenuse of their navigational triangles rather than along the two sides and then to establish a practical means of sailing over a great circle route. 272 When at last this was mastered in the early 19th century, sophisticated Europeans had only arrived at a more complicated method of doing what the Polynesians and the Micronesians, the Arabs and the Vikings had long since achieved.

The key to this paradox lies in the fact that a literate society faces the peculiar difficulty that it must be able to set down on paper, in this case maps, charts and almanacs, what non-literate folk might easily read in the sky and carry in their heads. In the end, it comes to imagine that the job can be done in no other way.

The increasing complexity of the subject adds to the confusion. As A. G. Findlay has put it in his Directory, 273

“There is one misfortune attendant on the advancement of science, that by following out to minute particulars each special branch of it, the mind is more or less diverted from the simple first principles. The whole tendency of modern research . . . is to multiply the facts attendant on any department of physics; so that, instead of the plain matters of fact which our forefathers were content only to know, we have now such a multitude of phenomena to deal with, that the real question of importance is often subsidiary, or lost sight of . . .”

Nor is the accidental voyage thesis altogether satisfactory on the manner of the settlement of the Pacific. According to the theorists, Oceania might have been occupied both from east to west and from west to east. The light-skinned inhabitants of Melanesia are in this view accidental settlers from the east who could not find their way back.

I have argued elsewhere that a dominant factor in the peopling of Melanesia has been the prevalence of malaria in all that region west of 170° E longitude, Buxton's line, and the pitiful susceptibility of a non-immune maritime folk to its ravages. 274 For here Polynesian and Micronesian expansion came at length to an end. Various small groups - 60 of invaders from the east and north occasionally found safe lodgement on non-malarious islets, among them Tikopia, Rennell, Bellona and Sikaiana, or in comparatively healthy districts on a few of the larger islands. Others again took refuge on artificial islets laboriously raised in sheltered lagoons. Burutu, the earthly paradise of which the Tongans and Samoans spoke, lay out of reach, its air “infecting mortal bodies with speedy death”. 275 The pattern of settlement here owes little or nothing to accidental voyaging, a conclusion which the evidence relating to native voyaging in eastern Melanesia fully substantiates.

It is also doubtful, as a number of earlier observers have suspected, that eastern Polynesia was settled from the west, whether by accident or design. Anyone who has spent some time in the Pacific islands cannot fail to be impressed with the richness and variety of the flora and fauna of Melanesia, the paucity of those of Polynesia. It is true that oceanic barriers have not been particularly effective against such plants as sedges or against strong-flying insects such as dragon-flies, damsel-flies and certain of the larger water beetles. Various other plants and creatures have been transported farther east by birds and man. But the facts are impressive. Freshwater sponges have failed to reach any of the islands of Fiji. Stone flies are unknown in tropical Polynesia. Water bugs do not extend south or east of the Solomons. To take an even more significant example, there are 79 species of mosquitoes in the Solomons, 18 in Espiritu Santo, 9 in Aneityum, a half dozen only in Polynesia, at least four of which appear to be recent introductions. 276 Of the remaining two, the exceptionally hardy and now ubiquitous Aëdes (Stegomyia) polynesiensis Marks, the chief vector of filariasis in Polynesia, is not merely confined to Polynesia and Fiji but, on present indications at least, might well have spread across the Pacific from east to west, 277 in the same direction that is to say in which it seems to me the Polynesians themselves came.

It remains to consider whether the individual populations of Oceania could have been established in the way that the accidental voyage theorists have imagined. The great events of the prehistory of Oceania, it has been argued, were the arrivals of isolated canoes with women aboard since women did not ordinarily travel in canoes with men. These might have occurred no more frequently than once every 500 years, but then several thousand years is a long time. 278 If in 800 B.C.—the figures are chosen arbitrarily and merely by way of example—the ancestors of the Polynesians had numbered only fifty, this handful of people coming into the Pacific two or three thousand years ago could easily have propagated the total population, say 800,000, of the Polynesian triangle in historical times. Thus “one canoe-load of people, borne onto one of the islands in the early centuries of the - 61 Christian era with a rate of increase of 5 per cent to 10 per cent per generation, could account for the total population of the entire grouping in historical times”. 279

The idea is naive. Primitive man—witness the Australian aborigine, the pygmy and the Eskimo—has always been a rare creature subject in a marked degree to disease, accident and war and in particular to the narrow biological limitations imposed by a purely subsistence economy. It is thus unlikely that the earliest inhabitants of Oceania—even if their primary appearance in the supposed dispersal areas can satisfactorily be accounted for—could effectively have occupied its myriad isles in the brief space which the available radio-carbon dates would appear to allow.

The legends suggest, more reasonably, that the Pacific was occupied at a comparatively late date and within a quite short period following a far-reaching agricultural revolution which stimulated not merely a sudden growth of population and a fierce rivalry for cultivable land but the emergence of new and more complicated social and political institutions. 280 In this case, the movement cannot but have been very considerable, involving the migration of whole clans, of hundreds of people at a time, of chiefs, priests, warriors, artisans, and above all, women and children, in short, of thoroughly viable units. For, as the history of the Viking settlements in Iceland and Greenland and the early English colonies in North America perhaps shows, the establishment of tribalized communities in distant lands is a difficult business likely to succeed only when the original venture is sufficiently representative of the social structure of the homeland and supported by frequent and strong reinforcements. 281

The Marquesan evidence is instructive on this point. Even in the 19th century, it would seem, Polynesian colonizing ventures necessarily remained pretty sizeable affairs, involving the co-operation over a period of many months, indeed years, of large numbers of people and of experts of every kind, in building canoes and in laying in great quantities of preserved food-stuffs which must last the voyagers until a new harvest could be gathered in some distant land and finally, when the expedition set out, the enlistment of scores of men to manage the sails and if need be to paddle. 282

These conclusions appear to find some support in the so-called Sewall-Wright effect. 283 Put briefly, the degree of what is termed genetic drift depends on the size of any given population. When a community is small, the movement is swift and substantial changes may take place in a single generation; in four generations a particular allele - 62 may finally disappear. When it is large, the change proceeds more slowly and is slight in degree. If mankind had always been a single interbreeding group, then genetic drift would have been of minor importance. The scattered distribution of man in families over widely separated parts of the earth and the prevalence of natural catastrophes have however aggravated the trend and produced folk notably diverse in physical and other ways. Mutation and natural selection have also played their part.

The scope and degree of the changes involved are well exemplified in Melanesia in originally homogeneous populations perennially separated by differing strains and concentrations of the malarial parasite, so that even the shortest journey beyond closely circumscribed clan boundaries might involve illness or death. In Polynesia, the Negroid strain in the racial composition is strongly represented in Easter Island, the Tuamotu and New Zealand, less so elsewhere. There are, moreover, considerable differences in such features as the cephalic index; short heads predominate in Central Polynesia, in Hawaii and to some extent in Samoa and Tonga, long heads in New Zealand and Easter Island, the Marquesas and the Tuamotu. 284 It can hardly be without significance, especially when one takes into account the supposed long isolation of the more remote island groups and their manner of settlement, that there is nonetheless throughout the Polynesian triangle a still remarkable degree of similarity in physical type, that the Polynesians are not readily confused with either the American Indians or the Indonesians whom they are said most nearly to resemble, that they are much more alike in many respects than are the Melanesians.

It is hardly less notable that even in the 19th century, the inhabitants of Oceania preserved a unity of language and culture which never ceased to astonish Europeans, accustomed to the phenomenon of being unable to make themselves understood across even the 20 miles of the English channel. Tupaia might converse with the Maori, 2,000 miles off, his countryman, Omai, with the Tongans, Maori seamen with the Hawaiians. And Polynesian craftsmen, though apparently long separated by vast distances of open sea, still continued to shape their adzes, to build their canoes and houses, to carve brow ridges on their clubs, to lash and plait, to fashion their fish-hooks much as their ancestors in far-off Hawaiki had always done.

The notion that the Pacific can have been settled in any effective sense by waifs and strays drifting helplessly in some sort of cosmic lottery on barren reefs and atolls or even by occasional “voluntary exiles” who, it is held—without a shred of evidence—could not navigate their way back and forth beyond arbitrarily set limits is in short erroneous. If the legends speak of occasional voyagers who, for example, came to New Zealand as Kau Moala might have done, because of temporary stress of weather, they provide no warrant for believing that these played any notable part in the colonization of Oceania. It is in any case remarkable, to say the least, that there is no record of involuntary - 63 arrivals in either Hawaii or New Zealand in the 150 years which have elapsed since the beginning of European settlement in these seas such as the accidental voyage theorists might be required to show if they are to be believed.

The alternative might be briefly stated. Like all primitive people, the Goths, Franks, Saxons, Vandals, Huns, Arabs, Vikings, Magyars, the Bantu, the North American Indians, the Eskimos, the Australian aborigines, the early inhabitants of Oceania were essentially nomadic, traversing immense distances in accordance with the laws of their environment, of wind systems, currents and the migratory patterns of the animals upon which they were dependent, and discovering islands, not as Europeans discover them, by chance, but naturally where birds and fish led them. The scattered archipelagoes of the Pacific were settled as the “traditionalists” have supposed, initially by small groups of oceanic rovers, and then, much later, by numerous deliberate colonists who, so far as a very different environment and the inescapable tendency towards cultural divergence in virtual isolation would allow, succeeded in establishing in remote lands societies similar in most respects to those they had left behind.

The original pressure in Hawaiki having been relieved, the great migrations of the 10-13th centuries naturally soon dried up. The scattered inhabitants of Oceania meanwhile turned increasingly to the exploitation of their own lands. Adventurers, it is true, still cast themselves upon the sea to engage in long distance cruising but life, as is necessarily the case in a largely agricultural economy, had become more static. Suitable outlets for settlement had in any case everywhere receded. A land in which the coconut and breadfruit would not grow no doubt appeared unattractive to prospective migrants from equatorial lands. The expansion of the Polynesians westwards, whether from eastern or western Polynesia, was in the final analysis checked at Buxton's line.

The last stage of the drama was quickly over. A new burst of maritime activity in the 18th century, coincident with the emergence of strong native dynasties in all the major island groups, which might otherwise have ended in a struggle between the Tongans and the Society Islanders for the mastery of the central Pacific, was cut short by wasting civil wars and more especially by the appalling ravages of newly introduced diseases. The abrupt and irreversible eclipse of the canoe-maker's art and the rapid growth of foreign shipping in which the South Sea islanders found a new outlet for their wandering instincts brought to a close even the lesser voyaging of later times. The wide expanses of the Pacific were thus left at last to those whose misfortunes in tiny boats and outriggers have led theorists in Cook's day and this to underestimate the achievements of the Polynesians and the Micronesians in mastering the world's greatest sea.

1   I am particularly indebted to Dr. H. D. Skinner for having first suggested this topic to me and for much subsequent encouragement and instruction, to Dr. H. Bernardelli, Dr. D. D. McCarthy and Mr. H. E. Maude for helpful criticism and advice, to Professor W. P. Morrell, Dr. Angus Ross, Mr. R. G. Lister, Mr. P. W. Gathercole, Mr. D. R. Simmons and the late Dr. D. G. Herron for their friendly interest and kindness, and to Mrs G. M. Strathern and Miss L. A. Dovey of the Hocken Library, Dunedin, for their gracious and assiduous attention to my many demands upon their attention for several years past. I am also under a special obligation to Mr. J. Golson for his expert editorial judgement and forebearance. It is a pleasure to acknowledge the generosity of the Nuffield Foundation under whose aegis part of this work was carried out. In view of the controversial nature of the topic it is only proper to say that I am solely responsible for what follows.
2   Taylor 1956:60-1.
3   Purchas 1905a:49.
4   Purchas 1905a:52.
5   Purchas 1905b:3.
6   Dalrymple 1770-1a:98-102.
7   Dalrymple 1770-1a:203-4.
8   Dalrymple 1770-1b: Appendix 2:19.
9   Robertson 1948:xxii-xxiii.
10   Robertson 1948:127.
11   Robertson 1948:234.
12   Hawkesworth 1773c:477.
13   Bougainville 1772:228.
14   Beaglehole 1955:153-4.
15   Beaglehole 1955:137. cf. also his more detailed remarks in Cook and King 1785b:144.
16   This seems to me the only possible explanation of the facts.
17   Cook and King 1785a:200-2.
18   Cook and King 1785b:141-2.
19   Cook and King 1785b:177-8.
20   Wilson 1799:203.
21   Ellis 1829b:51-2.
22   La Pérouse 1798c:66.
23   Hawkesworth 1773c:477.
24   Williams 1837:504-10.
25   Wilson 1799:lxxxvi.
26   Beechey 1831a:251-2. cf. also Wilkes 1845e:18.
27   Bennett 1840b:105-7.
28   Lang 1877:327.
29   Findlay 1884:595.
30   Jarves 1843:27. West 1865:251-2 plagiarizes Jarves.
31   Pritchard 1866:402-3.
32   Lesson 1884:71.
33   Lesson 1884:2.
34   Lesson 1884:64-5.
35   Smith 1904:131.
36   Best 1925:8.
37   Best 1925:245-6.
38   Buck 1938a:61.
39   Hornell 1945:168, 176-7.
40   Sharp 1957:33. The Pelican edition is cited throughout.
41   Sharp 1957:34.
42   Sharp 1957:37.
43   Sharp 1957:30-1.
44   Sharp 1957:72; 1961a:127.
45   Sharp 1961b:219.
46   Sharp 1961b:225.
47   Sharp 1957:182-3.
48   Sharp 1957:97.
49   Sharp 1957:94-5.
50   On all this see: Marcus 1954-5:71ff.; 1955:121; Taylor 1956:77ff.; Beazley 1949a:414ff., 469, 472; Nilakanta 1955:77, 135, 182, 212; Williams 1837:508-9.
51   Sharp 1957:122.
52   For a critical opinion of the ability of Anderson, Sharp's chief authority, as an observer, cf. Cook and King 1785c:132; cf. also Cook and King 1785b:142 for Anderson's own realistic views on the value of contemporary European observations.
53   Sharp 1957:37-8.
54   Sharp 1957:38.
55   Sharp 1957:128.
56   Sharp 1957:41.
57   Sharp 1957:53.
58   Sharp 1957:40. The information is merely second hand.
59   Sharp 1957:52. The view is at variance with the facts.
60   Sharp 1957:48. But see evidence to the contrary given below, p. 52 ff.
61   cf. Sharp 1957:205, for example, where the flying proa is represented as only 25-40 feet in length. Also 1957:207 where its speed is put at five knots. Many Polynesian double canoes could be paddled faster than this. cf. Williams 1837:412. Also Salem Gazette, 10 December, 1813, which gives the speed of Hawaiian canoes when paddled at 11 or 12 miles an hour.
62   cf. for example Sharp 1957:147.
63   Sharp 1961b:224, where apparently his voyagers are in considerable difficulty after only 100 miles out on a hypothetical long voyage.
64   Sharp 1957:54. The remark is misleading. cf. Derrick 1950:118ff., also many other sources on the great frequency of Tongan voyages to Fiji. cf. also Derrick 1950:112-7, on the great expedition of George Tupou with 30 large canoes and 2,000 warriors to Bau in March, 1855. Return voyages from Fiji proper were often made direct, 460 miles non-stop. The outward journey, however, usually took in the windward islands for purposes of recruitment.
65   Sharp 1961a:127. The subject might have been left to the experts.
66   Sharp 1957:132-3.
67   Sharp 1957:136-7.
68   Cook and King 1785b:541.
69   On this latter point see Dumont d'Urville 1833e:114, 305-6; Mariner 1817a:265.
70   Becke 1908:226.
71   Mariner 1817a:72.
72   Forster 1778:366.
73   Boston Daily Journal, 15 December, 1854. Hundreds of thousands of eggs might be taken at a time.
74   Cook and King 1785a:259. Also Forster 1778:366.
75   Arago 1823:10-13; Essex Register, 29 September, 1834.
76   McKern 1929:6; Gifford 1924:38; Burrows 1937:163.
77   Krieger 1943:37.
78   Robertson 1948:123-4; Beaglehole 1938:400-1.
79   Banks 1896:110.
80   Cook and King 1785b:87.
81   Wilson 1799:136-7.
82   Lucatt (Lucett) 1851a:235.
83   Mariner 1817a:346.
84   Diapea 1928:111-12.
85   Lucatt (Lucett) 1851a:258.
86   Burrows 1937a:48; also Patterson 1882:411 re similar tendencies amongst the Aneityumese.
87   Lucatt (Lucett) 1851a:235.
88   Porter 1823:94.
89   Porter 1823:93.
90   Cook and King 1785c:26.
91   Sharp 1957:77 concedes the possibility.
92   Buck 1938a:86.
93   Diapea 1928:218-9.
94   Aitken 1930:54.
95   Beaglehole 1938:400-1.
96   Parkinson 1784:73, note.
97   Buck 1932a:54-7.
98   Lucatt (Lucett) 1851a:260.
99   Bougainville 1772:253.
100   Cook and King 1785a:320-1. cf. below, p. 37.
101   Gifford 1929:204ff.; Burrows 1937:19ff.; Loeb 1926:29; Macgregor 1937:27; Pritchard 1866:403; Dillon 1829b:112; Codrington 1891:343; Woodford 1906:166-7. It is of interest that so distant an island as Rotuma was still paying tribute as late as 1824, cf. Dillon 1829a:295.
102   Derrick 1950:122-3.
103   Buck 1938b:11.
104   Dumont d'Urville 1833d:244.
105   Porter 1823:93.
106   Aitken 1930:45.
107   Burrows 1937:45.
108   Beaglehole 1938:393-4.
109   Burney 1817:6-9.
110   Chamisso 1821:87, cited in Heine-Geldern 1952:315.
111   Lesson 1884:57, quoting Kotzebue.
112   Numelin 1937:177-8.
113   Weckler 1943:69-70.
114   Woodford 1916:32.
115   Anthonioz 1952:295-8.
116   Dillon 1829a:272.
117   Dillon 1829a:294-5; 1829b:104.
118   Dumont d'Urville 1833e:113.
119   Lesson 1884:57, quoting Kotzebue.
120   Pritchard 1866:405.
121   Williams 1837:393-5.
122   Williams 1837:410-11.
123   Inglis 1867:255-6.
124   Woodford 1906:168.
125   Lucatt (Lucett) 1851b:40-1.
126   Turner 1861:386-92.
127   cf. Pritchard 1866:406.
128   Beaglehole 1938:393-4.
129   Diapea 1928:75.
130   Dillon 1829b:269.
131   Ellis 1937:245.
132   The Friend, Honolulu, October 1858.
133   Nantucket Inquirer, 3 March, 1832.
134   Buxton and Hopkins 1927:100-1.
135   West 1865:219.
136   Haddon and Hornell 1936:374.
137   Diapea 1928:125-36.
138   For what follows see Hornell 1945:169ff. and Haddon and Hornell 1936: passim; 1937: passim; 1938:41-2.
139   Handy 1930:22.
140   Salem Gazette, 21 March, 1817.
141   George F. Snow, cited in the Boston Courier, 21 November, 1853.
142   Fornander 1880b:9.
143   Cook 1777a:343-4, also plan.
144   Williams 1870:63.
145   Wilkes 1845c:167.
146   Best 1925:246.
147   Wall, cited in Haddon and Hornell 1936:326.
148   Beauclerc, cited in Best 1925:248.
149   Wall, cited in Haddon and Hornell 1936:326.
150   Cook and King 1785a:320-2. Vancouver 1801a:334-5 describes the entire collapse of this maritime activity by 1800.
151   Kelly, in press.
152   Stevens 1930:117.
153   Robertson 1948:140.
154   Cook and King 1785a:274.
155   Wilkes 1845c:54.
156   Best 1925:189.
157   West 1865:236.
158   Grimble 1924:125. The size of the canoe in question is not clear, but baurua commonly attained 70 feet in length. Mr. H. E. Maude kindly allows me to say that he remembers seeing at Tabiteuea in 1930 a baurua measuring 100 feet in length with two decks and carrying 106 people. Baurua were, however, found on only one island in his time and the reason for their construction, he thinks, was merely inter-village rivalry. Personal communication, 16 November, 1961, with photographs enclosed.
159   Ellis 1937:247.
160   Rogers 1712:367; Haddon and Hornell 1936:439; Burney 1816:241, citing Dampier. These vessels were once used by the Spaniards on the run between the Carolines and the Philippines because of their superior speed and weatherliness. According to Dampier, one of these ships sent express from Guam to Manila made the distance of 480 leagues in 4 days.
161   Bougainville 1772:283.
162   Cook and King 1785a:228.
163   Cook and King 1785a:276.
164   Haddon and Hornell 1936:334.
165   Haddon and Hornell 1936:334.
166   Haddon and Hornell 1936:327.
167   Haddon and Hornell 1936:327.
168   Anson 1776:342.
169   Heine-Geldern 1952:320.
170   Woodford 1916:28-9. cf. Finney 1959:36-9 for a modern demonstration of the superior qualities of Pacific canoe design.
171   Cook 1777a:215.
172   I owe the latter idea to Dr. Bernardelli and am at present elaborating it. The evidence, it seems to me, suggests that the Pacific islanders are originally descended from a “circumpolar” whale hunter folk. For ethnological support of the general thesis see Anell 1955:238-47 and Skinner in Golson 1957:288-9.
173   Tasman 1898:55; Wilson 1799:114-6; Emory 1928 passim; Macgregor 1937:30; Ellis 1937:30, 57.
174   Grimble 1924:128; Gatty 1943:6-7; Aitken 1930:54-5. The Gilbertese used tame frigate birds as land finders. Becke n.d.:118; Turner 1884:282-3 describes the use of frigate birds as mail carriers at Funafuti in the Ellice group in 1876. For the use of birds amongst the Arabs and the Vikings see Taylor 1956:60-1, 71-2; amongst the Vikings Marcus 1953:127-8.
175   cf. Gatty 1943:passim.
176   Haddon and Hornell 1936:373.
177   Corney 1915:285-6.
178   Banks 1896:162.
179   Forster 1778:509.
180   Wilson 1799:340.
181   Pritchard 1866:406.
182   Coote 1882:153-4. It is worth noting that the Bedouin still cross the Arabian desert without compass and are rarely a degree out over 400 miles. W. Thessiger in B.B.C. broadcast, 1961.
183   Best 1922:28.
184   Collocott 1922:162.
185   Makemson 1941:38.
186   The Friend, Honolulu, 1 November, 1860.
187   Grimble 1931:197; Hornell 1945:170.
188   Grimble 1924:133-4.
189   Makemson 1941:20-1. The details of the system are not recorded.
190   Makemson 1941:58. cf. also Gatty 1943:passim.
191   Makemson 1941:13-14.
192   Makemson 1941:41, apparently mistakenly, writes “dawn” for “evening”.
193   Williams 1837:96-7.
194   Makemson 1941:41. cf. also Collocott 1922:158.
195   Makemson 1941:41. Also 42-5, which reconstructs the navigational procedures which might have been used on ancient voyages from Pukapuka to Niue.
196   Collocott 1922:157.
197   Firth 1954:91.
198   Gladwin 1958:895-6; Firth 1954:91-2.
199   Collocott 1922:157 records that few Tongans are now able to point to and name more than a few stars. In many parts of the Pacific, however, notably in Micronesia but also in parts of Polynesia, mariners still prefer to use their ancient knowledge and they scorn compasses and clocks. Aitken 1930:54-5; Gladwin 1958:895. In modern Fiji, though the natives know how to navigate with the compass, they never carry one in their canoes but navigate by the sun and stars, the direction of the waves, the wind, and the great swell which always comes from the south. Neyret 1950:12.
200   Markham 1904a:152.
201   Markham 1904a:227.
202   Wilkes 1845a:330.
203   Arago 1823:13-14.
204   Banks 1896:108-9.
205   Forster 1778:511-13; Quatrefages 1866 and Smith 1898:801-6 print the chart. It suffers from the defect that the drafter mistook north for south and vice versa. cf. Hale 1846:122 and Lesson 1884:16-17.
206   Cook and King 1785a:367ff.
207   Beechey 1831a:235.
208   Ellis 1937:241-2.
209   Weckler 1943:69-70. The distance involved was about 900 miles.
210   Ellis 1937:239.
211   cf. Bougainville 1772:268-9, Beaglehole 1955:156-7; also cited but glossed over in Sharp 1957:206. Biological evidence which I am reviewing elsewhere appears to require a whole series of such voyages.
212   Woodford 1906:166-7.
213   Woodford 1906:167-8.
214   Markham 1904a:227-8.
215   Markham 1904b:495.
216   Dumont d'Urville 1833e:210-11.
217   Stevens 1930:111-13.
218   Markham 1904b:494-5.
219   Forster 1778:222.
220   Described in Derrick 1951:57.
221   See Parsonson, in Kelly, in press. I am indebted to the Hakluyt Society for permission to reproduce my conclusions on this point.
222   Haddon and Hornell 1936:440.
223   Newspaper references throughout have been abstracted from United States Central Pacific Islands Project 1950. Individual papers are cited by name.
224   New Bedford Daily Mercury, 19 March, 1845.
225   New Bedford Mercury, 30 January, 1829.
226   Sandwich Islands Gazette, 11 November, 1837. I am indebted to Mr. R. G. Lister and Mr. G. A. H. Kidd of the Geography Department of the University of Otago for measuring the distances. They are expressed in statute miles.
227   Honolulu Friend, 9 October, 1844.
228   New Bedford Daily Mercury, 19 March, 1845.
229   Honolulu Friend, 15 August, 1846.
230   Boston Daily Journal, 17 August, 1861.
231   San Francisco Chronicle, 13 November, 1869.
232   Boston Daily Advertiser, 29 October, 1858.
233   Boston Daily Advertiser, 16 June, 1821; Salem Gazette, 12 June, 1821, and various other sources.
234   Essex Register, 3 March, 1836.
235   The Polynesian, 24 October, 1840; Columbian Centinel, 28 September, 1836.
236   Boston Cultivator, 12 August, 1848.
237   Boston Semi-Weekly Advertiser, 6 May, 1843.
238   The Friend, 1 September, 1849; 15 November, 1849; New Bedford Mercury, 31 August, 1849.
239   Boston Daily Advertiser, 22 December, 1849.
240   The Polynesian, 6 April, 1850.
241   Boston Courier, 21 November, 1853.
242   Boston Morning Post, 12 August, 1854.
243   Daily Mercury, 31 October, 1855.
244   Salem Observer, 22 March, 1856.
245   The Friend, 3 September, 1859.
246   The Friend, 2 September, 1861.
247   Boston Daily Journal, 23 August, 1866.
248   Boston Daily Journal, 23 August, 1866.
249   The Friend, 1 September, 1866.
250   The Friend, 1 January, 1871.
251   Boston Semi-Weekly Advertiser, 15 February, 1843.
252   Boston Daily Advertiser and Patriot, 12 August, 1833.
253   The Friend, 1 April, 1871.
254   The Friend, 1 March, 1855; Whalemen's Shipping List, 24 April, 1855; Daily Mercury, 1 November, 1854.
255   Hockin 1803:54.
256   Essex Register, 7 June, 1820.
257   Essex Register, 29 September, 1834. The report notes that “at certain times of the year [the natives of the Caroline Islands] have a regular trade with the Ladrone Islands several hundred miles distant”.
258   Boston Daily Evening Transcript, 17 October, 1849; Boston Courier, 18 October, 1849; Essex County Freeman, 20 October, 1849.
259   The Polynesian, 4 August, 1849.
260   Diapea 1928:69.
261   Voss 1903:27.
262   Voss 1903:30.
263   Williams 1837:141-9.
264   Williams 1837:168.
265   Williams 1837:362.
266   Williams 1837:363.
267   Williams 1837:409. cf. also Williams 1837:358 for a voyage from Niue to Rarotonga, 800 miles due east covered in 7 days.
268   Cook and King 1785b:190-1.
269   Klestadt 1955:55-6.
270   Taylor 1956:3-4.
271   The view is my own.
272   cf. Vancouver 1801:34-5; Findlay 1884:889ff.
273   Findlay 1884:889.
274   cf. Parsonson, in Kelly, in press; also article in preparation. I am indebted to the Hakluyt Society for permission to make use of this argument here.
275   Mariner 1817b:109-10. The ancient debate on the situation of this supposed Polynesian ancestral homeland might now seem to have been settled.
276   Laird 1956:82-3, and passim.
277   I am indebted to Dr. D. D. McCarthy for this suggestion. We are examining the question further.
278   Sharp 1957:70.
279   Sharp 1957:57.
280   cf. Buck 1949:34-64.
281   On Iceland and Greenland cf. Marcus 1954:77-80; 1957:411ff.; on the disastrous early history of the British North American colonies see Newton 1929:136ff.
282   Porter 1823:93.
283   I owe the original reference to Dr. Bernardelli. The statistical dynamics of the process is explained by Stern 1960:717ff.
284   Buck 1938a:12ff.