Volume 44 1935 > Volume 44, No. 173 > The chronology of Australian watercraft, by D. S. Davidson, p 1-16
                                                                                             Previous | Next   

- 1

AUSTRALIAN watercraft is a subject which has received only scanty attention. Except for a valuable and well-documented paper in 1905 by N. W. Thomas, 2 who was the first to consider this question from a continental point of view by summarizing the descriptive material available at that time, our knowledge of this important aspect of Australian ethnology has been limited to the scattered references and brief descriptions by the early explorers and colonists, and to the work of Brough Smyth, 3 which deals primarily with Victoria.

Since 1905, however, our information has been considerably expanded, thanks to the detailed reports of Roth 4 and Tindale 5 and to the shorter but nevertheless important descriptions by Basedow, 6 Love, 7 and others, all of whom have confined their attention to the craft found in the local areas they investigated but who have not attempted to correlate their data with those of other regions. It would seem, therefore, that the time has come when it should be profitable to assemble this additional information with that already brought together, and, on the basis of the whole, to discuss the chronological aspects of the - 2 different types of watercraft used on the Australian continent. This problem seems to have been generally ignored by the few individuals who have concerned themselves with watercraft in this part of the world. Such a study has a dual importance. It should increase our knowledge of the process involved in the diffusions of specialized types of culture traits, and also suggest what kind of marine equipment may have been used by the early occupiers of the Australian continent.

The question of the diffusion of watercraft in Australia is one which involves many unusual factors and conditions which are generally lacking in most other parts of the world, and which do not, as a rule, enter into problems concerned with the diffusion of other traits in Australia. These are of prime importance to our problem. It seems hardly necessary to point out that watercraft are naturally limited in use to the sea-coast or to those areas in which waters, in the form of streams or lakes, are of sufficient depth to allow the clearance of craft and of sufficient breadth or danger to make their use desirable. It does not follow that all the regions which have these requirements necessarily have watercraft in use, but, on the other hand, it is obvious that watercraft cannot be utilized in places which do not fulfill these conditions. Because of these natural limitations to the extension of the idea of watercraft, it is important to note that at least one-half and possibly two-thirds of the interior of Australia can be eliminated immediately from our consideration. A glance at any map of Australia is all that is needed to convince one of the validity of this statement, for, as is well known, Australia is almost entirely lacking in natural inland waterways in the sense used for the other great land-masses. There are no large permanent lakes at the present time worthy of the name, and the only great river is the Murray-Darling system in the south-eastern portion of the continent. Even this great stream periodically suffers from droughts which, in certain parts of its long course, lower the river-level into a series of almost if not entirely disconnected reaches and pools. The rest of inland Australia is, generally speaking, practically worthless from the point of view of navigation. By necessity, therefore, watercraft are restricted in use to the coasts and the coastal plains. Most - 3 of these are well supplied with small rivers and creeks which seldom extend inland more than three hundred miles and which, in most cases, do not reach half so far. The practice of navigation, however, seems to have been more restricted than even these limiting conditions would indicate, for, as a general rule, I believe it is true that most of the rivers are not navigated many miles inland from their mouths. There is little positive information to indicate that they have been so used and, furthermore, the upper reaches and tributaries of most of Australia's streams are hardly conducive to intensive navigation, if any at all, except perhaps during the wet season in the northern part of the continent or during the freshet season in the south.

It is thus apparent that the use of watercraft in Australia has been limited by natural conditions to the 11,000 miles of ocean coasts with their bays and sounds, to the lower portions of the innumerable short streams which flow across the coastal plains, and to the great inland basin drained by the Murray and Darling rivers. Beyond these areas watercraft cannot be used. It is important to note, however, that not all of the waters made available for navigation by nature have been utilized. The explanation of the absence in these regions, therefore, must be sought on cultural grounds. Since the explanations may vary in accordance with the different conditions in different localities it is necessary to consider each area by itself.

The one great region in Australia in which navigation might be practised but in which it is completely lacking is the sea-coast between the Murray river in South Australia and Shark bay on the western coast of the continent. Along this 1,600 mile stretch no watercraft of any description are to be found. Indeed, the rivers which flow across this coastal country are also devoid of water-craft with the exception of the Albany-Esperance region of south-western Australia, in which Mrs. Hassell 8 reports the use of a log as an aid in crossing streams. This is similar to that used at the mouth of the Gascoyne river, which flows into Shark bay in Western Australia, where - 4 Austin in 1851 found a crude “one-log” sort of raft. In the words of Thomas

“It was a light log, 11 feet (3.3m.) long and 10 inches (25cm.) in diameter. At one end it was curved to an angle of 160 degrees, and pegs were driven in on each side of this end, on which were two layers of small twigs bound up with bark, forming a basket like a dish, about half the length of the raft. A portion of a similar one, 6 feet (1.8m.) long, was found by Phillips on Babbage Island at the mouth of the Gascoyne in 1855.” 9

From the point of view of a European it may seem strange that watercraft were not used along the south-western coasts of Australia. Off the west coast of South Australia between cape Catastrophe and the head of the Great Australian bight there are numerous islands and bays where navigation in all probability would be profitable. The same conclusion would seem to hold for the Recherche Archipelago between Israelite bay and Esperance bay in southern Western Australia and for the many groups of islands, and the bays and sounds, on the western coast of the continent. None of these islands seem to ever have been reached by the natives. According to Mr. Glauert of the Western Australia Museum 10 there are no archaeological evidences of any aboriginal occupation of Rottnest island, yet it can be seen distinctly from Fremantle. This conclusion is also indicated by the famous French explorer, Peron, who visited the island before 1809. He says:

“This island is uninhabited, and it did not appear that any of the natives of the continent had ever found their way thither.” 11
Jukes 12 indicates the same condition at the time of his visit (about 1846) although he refers possibly to Garden island.

But, on second thought, is it so strange that these coasts were not utilized for navigation as we ourselves would do? Is it any more strange that the natives of this particular part of the world should not have watercraft at the time of their discovery than that we did not have aircraft a century ago? In our case we have evidence of the historical process which has resulted in our use of - 5 advanced forms of transportation and we realize that without a well-developed technical background to which all the industrial nations of the world have contributed, our present facilities would have been impossible to attain. The case of the natives of south-western Australia is possibly quite similar. Living in a region to which the influences of watercraft-users have not penetrated, it is not surprising that they have not felt the urge to take to the sea but have been content to linger, in so far as watercraft are concerned, in that state of mental lethargy which has characterized every people of the earth at various times.

The possibility that diffusion has not yet brought ideas of watercraft into this huge coastal area, therefore, must be regarded as an explanation which may have a very important bearing on the case. The proof of this contention may be sought in the distributions of the various types of craft employed in the other parts of Australia. If we will find that watercraft gradually become more primitive as the area of negative appearance is approached it should be obvious that the lack of watercraft on the south-western coasts of the continent is due, in all probability, to the lack of diffusion of watercraft and the ideas associated with them. If, on the other hand, we find that the types used in other regions stop abruptly at the border of the negative area it will be more likely that resisting forces have been at play to deter these influences from crossing the border.

Another factor which we must take into consideration is the type of watercraft which may bound the area of negative appearance. It is quite possible that the types used in the neighbouring localities may be not suited to the conditions of the south-western and western coasts. This would be a very good reason why diffusion may not have carried watercraft into the negative area. However, since natural conditions of this region differ completely in the various localities from temperate to tropical climate, from wind-swept, surf-beaten coasts to calm, snug harbours and bays, it would seem strange that no influence had been able to penetrate past the borders into the many receptive areas, if the forces of diffusion had been intently pressing for any considerable period of time. Although it is easy to see why the crude craft, unsuited for ocean use, did - 6 not spread to the south-western and western coasts, this does not explain why these primitive types are not found on the rivers of this region, if it can be shown that it has been possible for diffusion to have introduced them. Here again, the distributions of the positive traits should indicate an answer to this problem.

The areas in which watercraft are partially lacking include various stretches of coast eastward from the Murray river to the coast of Queensland. Generally speaking, crude watercraft are known to most if not all of the tribes in this coastal belt, but their use is necessarily limited, as the result of their unseaworthiness, to the sheltered harbours and bays, and to the rivers. This is the condition which we would expect to find along the coast of Western Australia if diffusion had introduced a primitive type of craft. Along the coasts of northern Queensland the natives are actually able to venture into the sea but this is due partly to their more seaworthy craft and partly to the protection of the Great Barrier reef and the periodical tranquility of the ocean.

Now the appearance of the same type of watercraft in Victoria, south-eastern South Australia, and the Murray-Darling basin, would seem to represent an excellent example of how a water-trait has diffused overland (see figs. 8 and 9). Such a diffusion undoubtedly has resulted from the frequent intercourse of the natives of local neighbouring areas for ceremonial or other purposes. Certainly the crude type of bark-canoe of this region did not reach its present distribution by way of the sea, for it would be impossible to navigate this type of craft in any but the most placid waters. It is probable, however, that its diffusion has followed the coastal plain in general, the concept of this type of craft having been taken overland from river-valley to river-valley. This would be particularly plausible in so far as Victoria or South Australia individually are concerned. The major occurrences in these two regions, however, separated as they are by the relatively arid country of western Victoria and the extreme south-eastern corner of South Australia, may have been the result of a diffusion by way of the Murray river. In so far as the likewise primitive but less crude type of craft (fig. 9) found along the coast of New South Wales is - 7 concerned, it is again obvious that diffusion must have been overland in a great many instances. This craft was also too frail to put to sea in rough waters, hence its spread could have taken place only by the avoidance of the rough and rugged portions of the coast and by its idea being carried overland from river to river and from bay to bay.

In a broad general sense, however, we may speak of diffusion as having taken place along the coastal belt, keeping in mind, of course, the necessary qualifications needed for the natural features of the land and sea and for the limitations in navigation for any particular type of craft. In fact, since watercraft are completely lacking throughout the greater part of the interior of the continent, with the exception of the Murray-Darling basin, and are found only along the south-eastern, eastern, northern and north-western coasts and the rivers which flow across them, it is obvious that diffusion must have been confined to the coastal routes and the Murray-Darling river system. It is, therefore, apparent that we have a most unusual condition with which to contend, for instead of having the possibility of diffusion from a point of origin or introduction taking place outward to all points of the compass, as is theoretically possible for the ordinary culture-trait when not restricted by geographical or cultural factors, we find the diffusion of any type of watercraft in Australia specifically limited to but two directions, up and down the coast. Such a condition is more or less unique in cultural history, for, although numerous traits in the world are found only in limited coastal distributions, in most cases there have been no insuperable barriers to their diffusion into adjoining inland areas. The case of the Eskimo kayak is a good example of this point. It is limited entirely in distribution to the Arctic littoral, although, in so far as natural conditions go, there is no reason why it could not be used throughout the area occupied by the birchbark canoe.

Now the limiting of watercraft to the coasts of Australia should simplify our problem considerably, for we do not have to take into consideration the possibility that a trait may have reached a given area by other than the coastal route. It is logical to expect, therefore, that in so - 8 far as the chronological aspects in any region are concerned, they will be indicated by geographical distribution of types. This does not necessarily mean that the distribution of types as found on all the coasts of Australia are indicative of a chronological relationship among all the varieties of watercraft, for local development must be allowed for when the facts show evidence that such has taken place. However, it seems apparent that when two non-contiguous appearances of one trait are separated by the contiguous appearance of another, the age of the latter may be inferred as the lesser of the two in that particular region.

There is another possible arrangement of distribution which should indicate relative age. In the cases where it is found that the upper reaches of various rivers have a type of craft which is different from that extensively distributed along the coast and on the lower courses of the rivers, it seems logical to believe that the former type is of greater antiquity and that it has been displaced along the lower regions by the diffusion of the latter. Possible explanations of such cases suggest themselves in the easier avenue of intercourse along the coast, the unfriendly relations which may exist between interior and coastal tribes, or in the greater conservativeness of the interior natives, who, in not being so dependent upon watercraft for their economic activity, are less apt to change rapidly from one type to another.

With these cultural peculiarities of watercraft and the geographical limitations of Australia in mind, we may turn to the consideration of the types of watercraft used in Australia and Tasmania. Generally speaking, we may classify the types into four main groups:

  • 1. Those having a hollowed-out log (dugout) as a base.
  • 2. Those made from one or more pieces of bark (bark canoe).
  • 3. Those consisting of two or more logs or rolls of bark or bundles of reeds, etc. (raft).
  • 4. Those consisting of a single log, or roll or bark or bundle of reeds, etc. (float).
- 9

This group may be subdivided, as far as we are concerned, into three classes:

  • A. The plain dugout canoe.
  • B. The dugout canoe with a single-outrigger.
  • C. The dugout canoe with a double-outrigger.

Considering the history of watercraft in general, it seems quite obvious that outriggers, either single or double, are historically more recent than the ordinary dugout on which they are dependent. Although the ordinary dugout and those with outriggers are found in Australia, their genetic relationship is not a problem for us to consider, for the ordinary dugout seems to have been introduced by one people in one area and the dugout with an outrigger or outriggers seems to have been brought into another region by a different group. In so far as Australia is concerned, therefore, the two constitute separate historical movements. There is also the possibility that there has been in addition an independent development of a dugout in Australia.


The use of the outrigger in association with a dugout canoe is one of the most widely spread of aboriginal culture-traits. In either a single or double form it is found from Easter Island on the east to Madagascar and the neighbouring east coast of Africa on the west, a distribution extending half way round the world. Although there are places where it seems to be lacking, its use may be said to be contiguous if we allow, of course, for the great expanses of ocean which necessarily separate the positive appearances. The only notable negative areas that especially concern us in this paper, are the continent of Australia, excepting the Cape York peninsula of Queensland, and the western part of the southern coast of New Guinea.

The question as to the relative age of the single-and double-outrigger is one of the most knotty problems concerning the development and diffusion of culture traits which ethnologists have attempted to solve. The trouble does not lie in a lack of data as in so many problems of this kind, for there is an abundance of information of a - 10 reliable nature collected in the field during the past century, as well as numerous accounts of the writings of the early explorers which date back to the beginning of the seventeenth century. In spite of seemingly adequate material, however, we seem to be no nearer to a satisfactory conclusion than we were when this problem was first considered, for the data indicate conflicting conclusions in so far as any of the present methods of interpreting such facts are concerned. The distributions of the uses of the single-and the double-outrigger are in no way consistent nor are those of the associated but minor traits, such as the various methods of attaching the booms to the floats.

The first intensive study of outriggers was made by Haddon, 13 who limited his investigation primarily to Indonesia. The conclusions he reached are not in accord with those of Wissler, 14 who subsequently used his data, nor with those of Dixon 15 who more recently treated the subject more extensively. The problem is still open, therefore, and since it is improbable that all the evidence needed for a satisfactory conclusion can ever be retrieved from the pre-historic horizon, this perplexing puzzle may never be answered.

The Australian appearances of the outrigger, as already mentioned, are confined to the coast of the Cape York peninsula from the Batavia river on the gulf of Carpentaria to approximately cape Grafton on the east coast, a distance of well over a thousand miles without allowing for the many indentations in the coast line (fig. 1). The double-outrigger types (figs. 2 and 3) are now found from the Batavia river to Princess Charlotte bay, and the single-outrigger (fig. 4) southward from this region to cape Grafton. We find, therefore, that the controversial problem as to the relative age of the two forms extends even into Australia where, from the continental point of view, the two types are decidedly foreign and of relatively minor significance. The Australian appearances are also concerned in the controversy regarding the chronology of the methods of attaching the booms to the floats, for three different

- 11
FIG. 1.
- 12

methods are found, two associated with the double-outrigger and one with the single-outrigger.

There can be no doubt but that the Australian uses of the outrigger have been derived from New Guinea. Cape York is only about one hundred miles from New Guinea, the two being separated by Torres strait which is studded with islands and which offers an open avenue for the diffusion of culture-influences. The same type of double-outrigger is used in both areas and in the intervening areas as well. Furthermore, according to Roth, 16 as late as 1904 hulls were traded over three routes to Torres strait and cape York.

The antiquity of the outrigger in Australia cannot be accurately indicated but, since it is likely that it was introduced at a time subsequent to the Melanesian invasion of the south-eastern coast of New Guinea, it is probably relatively recent.

The earliest record of the double-outrigger in the cape York peninsula, apparently, is that of Jukes 17 (1837) who described the type found at cape Direction. An earlier report by King 18 (1819) is not specific. Thomas 19 regards the “canoes” mentioned by King for the Bird Isles as double-outriggers, which they quite possibly may have been, although King himself says that they were similar to the type he saw at Endeavour river, 20 which in being like those described for the Blomfield rivulet, 21 must have been single-outriggers. Other explorers who mention the double-outrigger for the eastern coast in the cape York region include Macgillivray 22 (1852) and Jardine 23 (1867).

We have no means of determining the limits of distribution of the double-outriggers at any time during the nineteenth century. In spite of the unsatisfactory evidence, however, it seems likely that they were not used - 13 as far south on the east coast as they are at the present time.

According in information contributed by Roth 24 in 1910, the double-outrigger was used as far south as the Claremont islands, just south of cape Tribulation, and in 1928 Hale and Tindale 25 found them in common use still further south at port Stewart in Princess Charlotte bay, so there are evidences to show that a southward diffusion has been taking place during the past quarter century, if not for a longer period.

On the gulf of Carpentaria coast we have no information to indicate what has taken place. All we know is that the double-outrigger was found as far south as the Batavia river in 1910. 26 There have been no reports for this region since them.

The single-outrigger (fig. 4) is now found from Flinders island in Princess Charlotte bay to the neighbour-hood of cape Grafton. This southern limit was set by Roth 27 in 1910 and no records are available to indicate whether a change has occurred since then. In fact we cannot be definitely sure that cape Grafton was the limit in 1910, for in 1908 Roth 28 gave Hinchinbrook island as the southern boundary, and as early as 1852 Macgillivray 29 saw the outrigger as far south as the Palm isles. Incidentally he speaks of “outriggers,” but there can be no doubt, as Thomas suggests, 30 that he was referring to the single type. It is quite possible that the inconsistencies in the reports for the southernmost appearances may be due to a lack of sufficient records for the area in question in so far as Roth's statements are concerned. On the other hand, cape Grafton may be the last point where the outrigger is consistently used. If so the more southern appearances may be ascribed to either the temporary visits of the cape Grafton natives or to the possibility that diffusion has not yet done more than introduce them to the area south of - 14 cape Grafton where their use is still superficial and spasmodic. There is a possibility, of course, that their distribution has retracted since the time of Macgillivray, but I can see no reason why this should happen unless European influences are responsible, for the outrigger is certainly more efficient than the bark-canoes used in this area. Furthermore, unless the natives of this region are retiring from their off-shore activities, it would be surprising to find them giving up one type of watercraft unless they were adopting another in its place. There is no information to show that any other type of craft has been entering this area, and since it is unreasonable to suppose that these people would revert to the inefficient bark-canoes which characterized their region in the time of King 31 (1819), the only safe answer seems to be that the outriggers seen so far south were those of visiting natives, or that the present distribution extends further south than the reports would seem to indicate. Regardless of what the truth of the matter may be, it does seem true that there has not been much of a southward trend, if any at all, in this marginal locality since 1852. There do not seem to be any natural conditions which might hinder a diffusion down this coast, and if there are cultural forces which are deterring or actually barring such a diffusion they are not at all apparent.

There are evidences, however, to show that the southern boundary has moved southward during the period between 1819, the time of King's visit, and 1852. King saw only bark-canoes in use at Goold island, near Hinchinbrook island, and first encountered the single-outrigger at Blomfield rivulet on Weary bay. 32 He specifically states that this craft in

“being hollowed out of the trunk of a tree, was of very different construction to any we had before seen.”
The outrigger is described as being about two feet from the side of the dugout which measured 21 feet by 15 inches at its greatest beam. He found “similar” craft at the Endeavour river, cape Flinders, and the Bird isles. It is - 15 the latter instance which Thomas regards as a double-outrigger.

It is apparent, therefore, that the single-outrigger was diffusing southward at least between 1819 and 1852. During this period of diffusion along the southern boundary, indeed, up until the present day, alterations were also taking place on the northern limit, for there are evidences to show that the latter has been pushed southward for a considerable distance by the southward diffusion of the double-outrigger. If we accept King's statement in regard to the use of a single-outrigger at the Bird isles (11° 50') it follows that the double-outrigger has displaced the single type along the entire coast between there and port Stewart. However, if the craft seen at the Bird isles by King was really a double-outrigger we still have the information from Roth and Hale and Tindale which shows that the double type has moved southward from the Clarmont isles to port Stewart during the past twenty-five years and, presumably, this movement has been at the expense of the single form.

In so far as the west coast of Queensland is concerned, only the double-outrigger has ever been observed. It is important to note, however, that the children at the Batavia river, the southern border of the double-outrigger in 1910, make toy sailing boats

“with a single outrigger, always on the weather side, which can be shifted from port to starboard and vice versa as the occasion requires.” 33
This practice is interesting and important for it may indicate that the single-outrigger was once used on the northernmost gulf of Carpentaria coast as well as on the eastern coast. Until detailed inquiry on this point has been made, however, there is the alternative possibility, as Roth suggests, that this peculiar appearance may be “due to civilizing influences under missionary auspices.”

A change from the double-to the single-outrigger took place temporarily in 1888 at the island of Mabuiag in Torres strait as the result of the presence of a native from the New Hebrides. Ten years later, however, Haddon found the double-outrigger still in popular use. 34

- 16

Throughout the inter-oceanic distribution of outriggers, a variety of methods of attaching the booms to the float or floats are used. These have been described in detail by Haddon and by Dixon. In Australia, three different methods are found:

  • 1. Stick-type. 35 By this method the float is secured to the boom by one or more straight sticks which are lashed at the upper end to the boom and at the lower end either inserted into the float like a peg or lashed to it (fig. 2).
FIG. 2.
Double-outrigger Canoe with “Stick”-method of Attachment. After Roth. Northern Cape York peninsula, Queensland.
  • 2. Direct-lashed-type. 36 This method, as its name implies, consists of lashing the ends of the booms, of which there are generally only two, directly to the float (fig. 3).
FIG. 3.
Double-outrigger Canoe with “Direct-lash”-method of Attachment. After Roth. Central eastern coast of Cape York peninsula, Queensland.

(To be continued.)

1   This paper represents one of the studies in Australian ethnology made under a fellowship grant by the Social Science Research Council of New York.
2   Thomas, 1905, 1906.
3   Brough Smyth, 1878.
4   Roth, W. E., 1910, Bull. 14.
5   Tindale, 1926.
6   Basedow, 1913.
7   Love, 1917.
8   Hassell, MS.
9   Austin, cited by Thomas, 1905, p. 70.
10   Glauert, L., verbal information.
11   Peron, 1809.
12   Jukes, Athenaeum, 1862, no. 1793, March 8th.
13   Haddon, 1913, 1920.
14   Wissler, 1926, pp. 27-30.
15   Dixon, 1928, pp. 70-105.
16   Roth, W. E., 1908, no. 88.
17   Jukes, 1, p. 106.
18   King, 1, p. 237.
19   Thomas, 1905, p. 67.
20   King, 1, p. 220.
21   King, 1, p. 209.
22   Macgillivray, 2, p. 16.
23   Jardine, p. 83.
24   Roth, W. E., 1910, pp. 12-13.
25   Hale and Tindale, oral information.
26   Roth, 1910, pp. 11-12.
27   Roth, 1910, p. 13.
28   Roth, 1908, no. 88.
29   Macgillivray, 1, p. 98.
30   Thomas, 1905, p. 67.
31   King, 1, p. 200.
32   King, 1, pp. 200, 209.
33   Roth, W. E., 1910, p. 16.
34   Reports of the Cambridge Anthropological Expedition to Torres straits, vol. 4, p. 210.
35   For description, distribution and discussion see Dixon, pp. 92-94; Haddon, 1920, p. 126.
36   Dixon, pp. 91-92; Haddon, 1920, p. 124.