Haddon, Alfred Cort The outriggers of Indonesian canoes.







A. C.


(Rogaf $nt(Jtopofcgteaf .Institute of


Q&ntatn an&

GIV . 1136692 .

by a more exhaustive treatment of the literature on in the libraries of certain learned societies the subject. Double Canoes ""* * .69 THE OUTRIGGERS OF INDONESIAN CANOES. of which I now present a brief and general account. C.88. Mixed Direct Attachment : 83.124 ^ B. By A.. Halmaheran Mixed Attachments 1.. Mixed Direct and Balinese 2.. though I am fully by observations in the aware of the imperfection of my data. Direct 1. and by a study of photographs . (6) Attachment 1. Mixed Rod and Halmaheran 2.. 5. Mixed Direct and Rod 3. Notes on the Characteristic Outrigger Canoes of the Main Districts of Indonesia Notes on the Native Names for Canoes and Outriggers General Conclusions Bibliography 97 116 119 131 FOR several years I have been accumulating material dealing with the outrigger canoes of Indonesia. Moluccan C. : 83 their Distribution : Inserted . 87. 126 Rod S Pike 3 - - 87 79.. The The The The The .. 7*7 Distribution of Single and Double Outriggers Number of the Outrigger-Booms Attachment Float of the Booms to the Hull Attachments between the Booms and the Float and A. Indirect (a) Attachment inserted into the Float : L 2.128 39 12& . CONTENTS. 92 92 94 95. Material Terminology . 127 4.> 2 i^j ' Lashed 3. Mixed Direct and Halmaheran 4. Much more remains to be done field. : 90 J20 84.. Y-shaped Stick Board Balinese tied to the Float : gy 127 87> 12g 73. HADDON.. o.. 6.. Stick 84..

and I Of far greater value is the explicit have endeavoured in all cases to record such statements. they often do not appreciate those details all is frequently no description at illustration that vouchsafed. Forest London.A. Hill. HADDON. and we have then to rely solely on any may be supplied. S. Oxford.). Rotterdam. and Art. as well as of specimens preserved in numerous museums in various parts of the world. and his knowledge of the details of construction. I hope. however. and museums. Rijks Ethnographisch Museum. Museum Halifax. Bankfield and Ethnology. The Outriggers of Indonesian Canoes. of the Several of the following museums are referred to by the names towns in which they are situated : Amsterdam. but it is not conclusive that it is the usual or common form there. A word of caution is necessary with regard to the available data. the Natura Artis Magistra. or when different photographs agree in this respect. MATERIAL. London. It is only when a photograph shows numerous examples statement of a traveller. and the pride of the artificer. C.S. Bearing in mind these limitations. and probably. in many technical skill in cases.70 A. Peabody Museum. The model-making which is characteristic of these peoples. Pitt-Rivers Museum. Ethnologisch en Maritiem Museum. The model may be carelessly made and certain details may be slurred over or even wrongly constructed to save trouble. except But a further source of that the common type is far more likely to be represented. British Ethnographical Museum of Museum. give one confidence in accepting models as trustworthy. good evidence that a certain form was then present in a given spot. of the same type that it becomes authoritative. Salem (Mass. . the models as a rule seem to be made by those thoroughly conversant with the vessels. Museum. is A photograph.E. however. error awaits one. are made by those who actually build boats. even though the various parts may not be made exact as regards their relative proportions. I have not shrunk from accepting models as good evidence. Horniman Museum. U. that this essay will help towards a more systematic con- sideration of the details of construction and of the problems of cultural distribution. The unsatisfactory character of illustrations holds good for models. or an illustration made from a photograph. On the other hand. as specialist . In a considerable number of cases the written descriptions by travellers are far from which are of interest to a satisfactory. of Archaeology of Science Cambridge. Museum Edinburgh.. his keenness of eyesight. as any canoe may be a chance visitor. travellers a great deal In drawings or sketches made by depends on the skill of the draughtsman. Koninklijk Zoologisch Genootschap. Leiden.

a 2 .

. the transverse poles of an outrigger are outrigger-booms (or simply booms). Dr. . I take this opportunity of thanking various friends and colleagues who have helped me with information. Ceram Ternate . Sorong (a village just south of Geelvink Bay. Tidor Halmahera Weda Bay : (the southern Gulf of Xulla (Sulla) Islands. . Ake Selaka. Zebu (Cebu). Nicobars. : and Nusa Laut . Baba (Babber) Tenimber (Tanimbar) or Timor Laut (Timorlao) Islands. Bali . Waigiu (Waigeu) . For convenience islands of reference I shall. Celebes Makassar and Gowa (Goa). Saparua Serwatti . Nias. Lombok Sumbawa Sumba (Sandalwood Island) Savu Lomblen Ombai (Allor) Timor Rotti Wetta (Wetter or Eetar) . Banda Kei (Ke) Islands. HADDON. . Ceram Laut : . H. (Bali to Madura. G. Various methods of indirect attachment will be sufficiently described in the course of this paper. To avoid ambiguity I propose to adopt the following terms as here defined The outrigger is a balancing apparatus that extends transversely across the hull : of the canoe . ." . Talisse. south shore of Waigiu point of ) New Guinea). Guillemard. 71. . Solor . Nanusa Islands. Talaut Islands (including the large island of Sangir). p. (Seran) . Butong (Butung). They consist in the main of bent ratan in the Moluccan attachment. Sulu Mindanao (Zamboanga at its south-west point). or of a stick or sticks. F. Obi (Ombi) . Goram . The Lesser Sunda Islands . Bawean : Islands (north of Madura). Friederici. Borneo Sarawak. their free extremities may be attached directly to the^oa^. (Gilolo) Batjan Halmahera). Mentawei Sumatra : Islands. : . : mention the several and places in the following order Andamans. Misol . Saonek (an islet off the the extreme north-west TERMINOLOGY. Tulur (Tulaur or Salibabu) : Islands. Aru Islands. . Salaier. The Outriggers of Indonesian Canoes.) . H. as a general rule. and would more particularly like to mention Henry Balfour. H. the outrigger-spar or attachment-spar 1 " Frequently authors speak of the float as an outrigger. Dr. Banka. and Galela (in the north of Halmahera). H. Minabassa (the north-east point of Celebes) with the towns of Kema and Menado. a raj on the north coast. Palembang (in the south-east of Sumatra). north-west Celebes). a rod. Ling Roth. Java. The Moluccas (Banda to Halmahera) Amboina (Ambon) Amblau The Uliasser Islands Buru Haruku. . Philippines : : the north side of the Onin Peninsula .72 A. or of a variously shaped piece of wood. North-west area of New Guinea Skroe on the south and Sekar on Archipelago. H. Tobelo. in all cases this is spoken of as the attachment. . Juynboll. and Engano (islands to the west of Sumatra). . and Manila. Tontoli (Toli-Toli. Jenkins. Gulf of Boni (the southern Gulf of Celebes). and the island of Limbe. . (See Map. Evans. Ivor H. Tenimber) Flores . 1 or indirectly by various methods . the Gagayanes (Cagayan) group. Dr. C. Lawrence W. . .

(Fig. Micronesia a spar may pass diagonally from the hull to the end apparatus this I In of the outrigger term a stay spar. 126) to the boom. that These may is. it may not connected with the float by means of a regular attachbe tied to it by means of a long lashing this may be termed a Occasionally. as they are longitudinal to. Occasionally there longitudinal spars but is is a central outrigger-boom which is lashed to the various ment. The Outriggers of Irufon> . this may be called a complex In many outriggers there are one or more poles connecting the booms and be termed usually lashed above them." but as the booms themselves are transverse as regard to the canoe and they are at right angles to the it seems preferable to adhere to my former term. C.<>ex. of This somewhat cumbersome term may be applied to a type boom is proper is that appears to be confined to the East Java Lombok area. attachment are employed on the same boom attachment. the length of the canoe. The boom short and thick. more than one false-boom may occur. 2 D) or sticks (p. parallel with. HADDON. but sometimes beneath them. I prefer to describe it as a boom-spar. (or may pass from the attachment-spar There are outriggers in which one or more of the booms attachment and the other or others another kind these . and the published illustrations known to me It may be suggested that the boom was made thick in are not clear on this point. but in that case it would obviously be too heavy to form an efficient outrigger to remedy this the projecting portions might be thinned down . though fake-boom. may have one kind of may be termed mixed two kinds of attachments. order to strengthen the hull. Boom-prolongation. They might " therefore with justice be termed transverse spars. one end of which rests on the roof of a shelter of a Mentawei . make the term applicable to all its variations. The sedek of the Balinese (p. the hull and the attachments this may 1 be termed the central longitudinal spar the outer and inner longitudinal spars run respectively 6 B. someboom. but a thinner spar is attached to it. Cases are known to me. but the exact method of the junction in actual vessels has been described only in the case of the sedek. outside our area. simply spar). though transverse as regards the length of the booms themselves.A. thus merely one form of a boom-prolongation. as the boom itself. but not in Indonesia. Sometimes there lie is of the booms. (which possibly may occur) or a more slender spar attachment of wood 88) or is bamboo might be fixed on to it. into the In small models of canoes from this area the prolongation inserted boom. where . 17). immediately on the outer and inner aspects of the attachment (Figs. a spar running more or less midway between . A boom-spar. There may be above a times being as long thus giving in order to rise to boom another spar which varies greatly in length. A thin spar. the free end of which it either inserted into or lashed to the float or connected with by means is of a rod- attachment. bracing spar. in which case it might be regarded as a an accessory or double boom parallel to the boom proper but . longitudinal spars.

D. 1. HADDON. 4) gives a drawing of a of a sailing canoe from Wetta FIG. 2. 65. MENTAWEI ISLANDS (AFTER ROSENBERG.. ZAMBOANGA figs.is figured by Rosenberg model (1888. StTLU . 1912. ZEBU . C. The Outriggers of Indonesian Canoes. 57. Fig. SANGIR (FROM MULLER. B. 56.. war-vessel. MORO CANOE. PL XVIII. MORO CANOE. . Fig. MODEL OF A WAR VESSEL. C. SECTIONS OF MODELS OF CANOES : A.O D FIG. l). 9). 1888) X .74 A. Baessler (PL VIII. knabat bogolu (Fig. . THE RIGGING OF THE VESSEL IS OMITTED. knabat bogolu. 58).

WITH TRIPOD MAST.. >. * . above which are equally long straight boom-spars. but in this case there its is a are lashed to boom and boom-spar U-Moluccan attachment. C.^just men- tioned . HADDON. the upper limbs of which In a model canoe from Tenimber in (Fig. .A. U-MOLUCCAN ATTACHMENT. 3). *. BAESSLER). these evidently correspond with tho.-. . WKTTA (FROM . WKH A BOOM . The Outrigger* of Indonesian Canoe*. 3 3 -MODEL STRAIGHT BOOMS AND BOOM SPAR* OK A SAILING SHIP. (Wetter) Islands with a double outrigger of two straight booms.

. without touching the water. 91). (G. he terms " : it a weather platform. p.. He adds We have. beneath which the . which usually are moie than two in is boom which number. obviously these elements are lashed together. and the next step was to knock off the outrigger may log [float] on the other side. January. and attachment-spar are strengthened all by a bracing spar North Celebes 115. when a flooring has been laid across the booms. D. in the Asiatic Archipelago. 4).n. out the points (1875. . laid across the vessel amidships. the boom-spar. Fig.. it lies over a longi- tudinal spar which rests on the two booms and and its end is lashed to the upper iork of a Halmaheran attachment attached directly to the (Fig. p. It does not support but is attached by lashings to the unusually thin and fragile outrigger- thus strengthened by the elasticity of the bowed boom-spar. These double weather-platform boats . on the level of the gunwales. 192 G. Nos. H ADDON. 245) describes the trans- formation of supernumerary boom-spars into curved crescentic ornaments in the Sangir and Sulu Islands but these can no longer be termed boom-spars. . No. enabled them to carry a sail of enormous size. and boom-spar double float is is the long stem of a J_-shaped attachment-spar. the Sulu Islands (Guillemard. or have but a single outrigger . is The Outriggers of Indonesian Canoes. Savage Landor. II. according as the Gale fresher or abates. 430) practical utility of a single outlayer for canoes which " outriggers. Figs. C. . 482). may become bowed and have carved ends. 56. Zamboanga and Zebu is in the Philippines (Miiller. . which weighted with men. A weather platform had been found sufficient to balance the vessel on one side. Miiller (p. I. 2. . 57). are attached directly lashed to the float. . boom. but merely an outlager. p. 12). and " Lane Fox (Pitt-Rivers) extending several feet beyond the sides (f. 232). the upwardly curved boom-spar greatly reduced in length and the float.. Munday (1848. 2. the two platforms projecting one on each side of the vessel. 244) describes a (fore aft) model of a canoe from Sangir in which the two outer float. the outrigger-booms. " " In describing the of Borneo. the People run out and in on the Outlayers. Miiller. Beeckman says Outlayers. pp. D. . their boom-spars are presumably by a lashing the inner booms have the arrangement shown in Fig. 1912. thereby converting the outrigger platform into a weather platform. and from the weather partly platform. 55. booms and . p. above each of these booms is an equally long and attached to the ends of the boom slender boom-spar which turns up at its ends . As previously mentioned. and when is it each end being run into a large blows hard. 4. quoting from " The small sailing boats [at Palette. Brooke's Journal of by R. Celebes] had outriggers of wood. 52). the Leiden Museum a thin boom-spar above each true boom . p. Miiller (1912. . p. Flying Praws To prevent their oversetting. they fix two long Poles or Outlayers one across the '' : Fore-part and another at the After-part of the Boat Bamboe . as they are not ." Folkard says that the canoes at Manila have no pole. Evidently this is what is referred to [Raja] J. to keep the Boat upright " (1718. 1840 : Gulf of Boni.76 A.. necessarily connected with outrigger-booms. 1889. At Kema. fastened . a con- trivance which be said to be derived partly from the double outrigger.

ii. the ends of which are pegged on to the floats. III).. OUTLAY1TB.. Miiller suggests that this idea of a double canoe may be due to This sakit canoe .. but as in the Philip. MODEL OF CANOE WITH A DIRECT LASHED ATTACHMENT AND AN TALAUT ISLANDS (AMSTERDAM).la. the outlayer single or double... and queries whether the outrigger traceable to a double canoe (1912. the canoes have a double outrigger (when they have any at all).A. I. Southwest India. HADDON. in the 10..m. p. 1905. B^. The only instance known to me by Hickson (1889. outlayere Guines (1796) and at Manila by Du. m Outriggers of Inlonenan 77 in inland waters. may not universally be Double canoes.i'hili|.. and we may assume that the double canoes of Oceania had their origin in Indonesia. .. may FIG.. both have equally elontogether Micronesian influence.. of the That double canoes once occurred in Indonesia is extremely probable. The importance double canoe in the early navigation of the Pacific is well known. PI. An outlayer be defined as a pole or a simple framework balancing apparatus.. Washington..l*.' He and in the Census of fie Philippine p^ ' . The outlayer . p. SAlfOIK.<! Indonesia generally. 239). p. Fig. m>. 5.s .. 6. an were also found more convenient elsewhere. C. may may therefore be single When formed. and the third from Ceylon." another are three models of double canoes . there does not seem to be any necessity for the intermediate stage of a single outrigger I. for conveying cattle and bulky goods across the rivers from the account . by Edye these appear Oxford Museum In the to be temporary arrangements (Edye. are employed on the rivers of the Malabar coast. postulated by Pitt-Rivers for this area. similarly may be DOUBLE CANOES.....s \. from Mirzapore. " one is from India. janjar.npier (KiH(i). on the Ganges. 164) : of an Indonesian double canoe model used is that illustrated this is a small for ceremonial purposes Nanusa Islands to the north-east is of Celebes composed of two canoes close gated upturned ends and are crossed by three booms. a platform of boards or closely-laid bamboos or poles is a be termed weather which may platform. 5) or double. and (Fig. in the canals in Manila. very well be evolved from an outrigger. . beyond the Salibabu Islands. also quotes the accounts of in th. In a carelessly-made model or sketch an outlayer might very well be mistaken for an outrigger.

Canoes with double outriggers and two booms occur on the coast of Eastern In Madagascar the Africa. p. single outriggers from various parts of Indonesia by De Bry and by Valentijn. There is a model in the Amsterdam Museum of a canoe from Sangir with a single outrigger (Fig. An aberrant type from the north coast Pitt-Rivers (Lane Fox) mentions a single of middle Java has been noted by Hornell. double-outrigger coasting . They also extended to the . a double float and apparently a direct attachment (Fig. can hardly be termed a proper boom-spar there (p. 1). the " " two booms of which project on the other side of the hull to form an outlayer : the downwardly curved ends of the booms are lashed to the float by a direct attachment. while rack on the windward side they sometimes place a canoe and everything on board that is movable. down from is the roof of the cabin. De Bry is PL c A). with the exceptions noted below (see Map. Fig. that in this paper I omitted the record by Miiller in Madagascar of small. 429) . B). PL XVIII. which is single outrigger on the starboard side . as nearly all the boats are drawn in side-view. 19) figures a model of a warship. knabat bogolu. on a kind of outrigger in Borneo. but there are other canoes in these islands which are without an outrigger. depicted. a coracora (V. C. A) and in the Nicobars (Fig. it has two main booms which slope it down -to the float and each has an accessory boom. are figured Numerous canoes with Nicolas. HADDON. and describes (p. XIV). which also figured by Nicolas (II. this evidence cannot be fully relied on . to larger canoes appear side of the hull to have but one float. for toy canoes with a single outrigger from the Sarawak coast see Fig. but does not give any reference (1875.78 A. and North-western Madagascar. 1918. carcolle. 5). Tab. Wherever outriggers occur in Indonesia they are double. We are probably justified in assuming that the outrigger was double in all the Moluccan craft in which one was present. No. H. they have sailing boats with single outriggers on the leeward side. 11. at Madura. I regret form an outlayer (Haddon. p. The Outriggers of Indonesian Canoes. p. but Nicolas figures (PL 14) a small war-vessel. but as the illustrations given by the old authors do not appear to be always accurate. who calls it coracora or carcolle of Banda. von Rosenberg (1888. with one. 32. To the west of Indonesia canoes with a single outrigger and two booms are found in Ceylon and parts of Southern India as early as 1599 Lintscotus figured the type occurring at Goa and Cochin (II. though the booms project beyond the other 29). 10. Folkard states 485) that on the north coast of Java. 71) : Single outriggers alone are found in the Andamans (Fig. from the Mentawei Islands (west of Sumatra) which appears to have but a outrigger. only one outrigger could be gives only one illustration of a double outrigger. PL 9). Maldives. 19) the carcolle of the King of Ternate as having a double outrigger. THE DISTRIBUTION OF SINGLE AND DOUBLE OUTRIGGERS. 481). the Comoro Islands. which as slopes . The Sumatran sailing craft known as jellore has sometimes only one then alternately to windward and leeward (Folkard.

of three to p. 401. 315).n. VII) gives a photograph dug-out called " chd-rigma (p. 296. usually is selected which have a natural thin branch projecting at right angles this laid over the boom and tied firmly to it (Fr. 271. I have recently come across a book by L. " Folkard (p.). rarely more. p. but Man it Mouat also supposed that but the type of attachment entirely close to that of the Nicobarese shall see. Nos. 267) four booms may occur at Misol and Weda Judging from photographs (G.. 1913). and Ceram the outriggers almost invariably have but two booms. was adopted from the Ceylonese As we negatives this wild suggestion There are only two booms in the Nicobarese canoes.. Halmahera. Edinburgh. float of and Sakalava sailing boat. Horniman. on 6 (drawn in p.. the which has superfluous. with or without the branch) appear at all events at Ansus (Guillemard. No. Cambridge. pp. C. overloaded terminal enlargements that resemble the Miiller adds. 249. " spike attachment. pp. of a large ends of the Hawaiian double canoe. 272) denies craft. The Outriggers of Indonesian Canoe*. 239). This is the earliest record known to me of this kind . Both kinds of spikes to be used indiscriminately and separately. Augustin Bay on the south-west part To the in Oceania. like the double-outrigger type. lakkajiara. 460). 404 G. Both kinds of canoes have a new type of attachment which consists of a nail-like wooden spike or stick which passes from above through the end of the boom and is driven vertically into the float. has three or four to a dozen booms. however. 442). 402. was a recent introduction. Brown MS.e. very canoes. HADDON. of craft in Madagascar. is frequently present THE NUMBER OF THE OUTRIGGER BOOMS. Spikes are . 302. II. but within this area the single outrigger predominates and. 1595). this.. In the Andamans the smaller canoes are seven booms (A. (1598) who gives a plate. and. PI. 116). of two dugouts with a double outrigger of two booms and a vertical stick attachment the float appears to be a long thin log. 127). at its eastern entrance. p. canoes. whereas to the north and east of it they have usually four. R.A. 252-254). Models in the British.. which passes north of Borneo. Geelvink common type Guinea they begin to appear in Geelvink Bay and continue 22. This may be termed a (p." An inner longitudinal spar . {i. The double outrigger is found throughout New Bay and ceases at Cape D'Urville. and it adds to the interest to find that of the island. the stick attachment is. As the early writers did not it mention an outrigger. Halifax. and Oxford Museums have three. /. . perhaps is evidence that Malays voyaged to Madagascar in double canoes (1912. Mouat thought that (XI. Celebes. 4:59. fitted Mouat with a single outrigger consisting as does figures four (p. which has three. it occurred in St. east of Indonesia canoes with a single outrigger are the and in 1 down the coast (p. . and Haddon. with a single outrigger. Man of a small (XII. 7 ). sometimes three and occasionally only South of a line two booms (Map.

p. 292. but frequently there is " a central false boom. 242). but usually there are three (Fig. may have but two booms p. AND DIRECT LASHED ATTACHMENT. at Buton. In spite of these exceptions the canoes of Celebes appear predominantly to have but two booms. Nos. J. p. and a similar vessel at the Kanari Islands. of these the fore booms with a direct attachment to each float . Nos. p. No. booms off Batjan p.. 7) (Wilkes. 110). (1893. 172).. II. V. or the outrigger may be absent . There is a model of a large plank boat from Gowa.80 A.. SULU (Photo G.. all in Halmahera. an island off the south-east point of Celebes (Friedenci. II. while the three middle float." the end of which is connected only by a lashing to the centra FIG. 192 . and Bali. 114). with a double outrigger. Patani. No. Burbidge. p. Bay (G. C. 7. 6. 142. 180. CANOE WITH DOUBLE OUTRIGGER. THREE BOOMS. north-east of Celebes they have either two or three booms which are pegged 1 on to the floats.. WEDA BAY. Professor Hickson has presented to the Cambridge Museum several small models of ceremonial sakit canoes from Nanusa Island. No. FOUR BOOMS. 1 I do not regard the pegs as being significant they are probably merely a labour-saving device.. p. FIG. More than two booms usually occur In the Sulu Islands small canoes p. 333 . 244). G. ones are lashed to the underside of the It would not be surprising if this were a There are usually three booms model of a craft coming from the Sulu area (p. CANOE WITH DOUBLE OUTRIGGER. in the Leiden five Museum. . but usually in Weda Bay. 217). 235). and aft booms are inserted into the float. This island is just on the above-mentioned border line. The Outriggers of Indonesian Canoes. of the float (Friederici. 82) with three Forrest figures a (cf. 1889. p. No. 206 . G. AND HALMAHERAN ATTACHMENT. 15.. 4). South Celebes. 115. 290). (Guillemard. 4. 292). . " Molucca Corocoro " (pi. G. HALMAHERA (Photo G. p. S. Fig. 225 . in the Sangir Islands (Miiller. 1889. 4) or four (Guillemard. HADDON. Fig. north-west of Misol (pi. p. the outrigger has two booms with a Halmaheran attachment. 6)...

183). 81 180. 112.. characteristic of the Marshall is which 6 . (Fig.way a straight." but gives no further particulars it has three outrigger booms and what may be a Moluccan attachment. which had at least three booms to their outriggers. seas given by old Dutch voyagers there of longi- massive outriggers possessing several on or under which are fastened a variable number and to the ancient Indo-Javanese tudinal planks (i. with but few exceptions. Two straight booms are found in the Pelew. on which three men sit and paddle. stick which is inserted into the a the end straight supported by float and apparently also into the boom. the canoes have two outriggerbooms.-rfe of a Outrigger. model hose canoes are he fore and downwardly P-rate craft off north point of Borneo "piratical prahu. pp. C. 211 . 7) a double " Boot von den Molukken. 383). 378. of Indonetian Cane. on each side of the booms are three curved the Caroline Groups of each of . it was probably a Sulu FIG.. In the Philippines three booms may occur at Manila and on Lake Lanao in Mindanao (Vojnich. and typically in Group are two straight booms. 8. aft i LL In with a double downwardly curved booms outri^r co vessel. Marianne. whereas the central curved at their ends Pritchett (p." dapang. Throughout Polynesia. PIRATE CRAFT OFF NORTH POINT OF BORNEO (AFTER PRITCHETT. parallel to the hull) which are used as seats I shall refer later to these craft vessels by the paddlers. 192. Weule figures (PI.A. Edinburgh Museum) booms are a. 8) . In numerous pictures of the East Indian are depicted large sailing war-vessels with (usually three) booms. 7).e. HADDON. Fig. of the booms midway between the hull and each float. a board is lashed to the underside . 183 (Fig. p.

It not infrequently happens that a length of sapling or bamboo is lashed to the upper border of the dug-out and the booms rest on this gunwale spar. case recorded (p. The object of this spar is obviously to protect the edges of the hull .82 A. It end of which is Museum a toy canoe. and a gunwale spar is fastened over them along the length of the canoe but Svoboda (VI. Equally characteristic is the presence of three (or four) out-riggerbooms in Melanesia. or the gunwale may be the upper edges of these locally raised at the spots which support the booms . D). however. the morticed into the float. Salem Museum is from Gilbert shown in variant Nonuti. He informs me that boys frequently use The men's canoes have no outrigger. B). 11. The only instances known to me in Indonesia of but one outrigger-boom are those associated with a single outrigger in a toy boat from Borneo (Fig. PI. pp. 10. the booms rest on the gunwales (Fig.. In the Nicobars. 30. Figs. in those cases where the outrigger is double there are but two booms. Fig. The play-boat figured by Kramer (1906. In the Andamans. A. or three booms. 79. HADDON. 122 . group something preceding attachment. a . the boom rests in the inner one 295. there are in a notch I. p. C. 1) figures the ordinary arrangement. two. Muller (1917. see also p. canoes with one. especially when the hull consists solely of a dug-out. is 1). two horizontal convergent sticks from the fore and aft quarters of the hull two central parallel short ridges on the canoe-shaped float. Caroline Group. Mass. of a canoe from Ruk. from Yap in the same is it similar to is an obscure indirect the but there one. has a single outrigger and only one boom. 29 pi. p. Alexander. R. The Outriggers of Indonesian Canoes. 36." There a model in the Peabody Museum. Islands appears to be very similar. booms. as it may be termed. In the majority of cases the booms rest on the gunwales. 805. and in the by Hornell from North Java (1919. the booms pass through holes in both sides of the dug-out (Fig. 806). In all of is these areas there a single outrigger. THE ATTACHMENT OF THE BOOMS TO THE HULL. old called people say that outrigger canoes were formerly used extensively " one man them is rafts.. Fig. the ends of which are lashed to the float (Model. Salem. tatareg. . and is inserted into a hole in the outer one. according to A. Professor J. Brown MS. but the . 291) from the Gilbert illustration of a 199) gives an toy sailing canoe. protuberances may be notched. 32). 31) in which the outrigger double. Stanley Gardiner has given to the Cambridge abbuodi. Fig. with the exceptions noted on pp. from the Maldive Islands. but in New Guinea there is considerable diversity. . with a long single boom on one side only which is strengthened by . and the model from Manila 114. 124. Islands. Sometimes the booms lodge in notches in the gunwale. it is widely distributed in Oceania.

in Sumatra in this case each float consists of two bamboos between which is a small peg which projects on secure) the under and upper surfaces of the boom. in which case it may be termed a float is single. . no description ) freshwater creek of Totoat. in Indonesia. C. Leiden) Baba. II. p. 329. XXIIXA). actually does occur in practice. According to Folkard's drawing (p. Celebes (P. 13. 5. HADDON. is of all the booms are lashed to the float (Figs. The Outriggers of Indonesian Canoes. p. 1894. 336) Lake Galela in North Halmahera (KUkenthal. . PI. 480) the Sumatran jellore and ballellang have booms with . . Direct. but the booms Buru (Forbes. 62 . 32) have this method. 52. Fig. No. Kei Islands (Langen. . When not otherwise stated or more bamboos (or but occasionally two of are pieces wood) employed. The Balinese attachment 2. 124. (p. . Amsterdam model. when made . XLV. 1 Map p 71. 88) really belongs here. Fig. downwardly curved ends which appear to be inserted into the floats unfortunately the drawings in his book are on too small a scale for the details of attachment to be reliable. Pis. . The main groups 1. it is THE ATTACHMENTS BETWEEN THE BOOMS AND THE FLOAT AND THEIR DISTRIBUTION. Leiden. p. The distribution outside of Indonesia is given later. Lashed. 356) . PI.A. no description) Gowa in South Celebes (model No... . . (L. 136) Makassar (Valentijn. -The ends of both if it booms are inserted into the float. DCI. Sarasin. p. no description .. has been noted at Mentawei . 15 Kiikenthal. Inserted. has its ends cut square. of bamboo. obscure. . Paloppo and Libukang in Leiden) curved booms) Ternate . Java I. may have PI. and F. p. 1 There are various methods in the attachment between the booms and the float. PI. west (Pfluger. The float. p. The float usually consists of a it single log of wood or piece of bamboo. 172) the Gulf of Boni. . Amsterdam) Tenimber of Timor (model. on a dyukun canoe) . . 235. may be assumed that the double. no description) Sumba or . Martin. Engano (model. (L. poor figure. (Rosenberg. It 1. Batjan (G. K. between Flores and Sandalwood Island (model. 23. . treble. are : A. . Lomblen. Lintschotus illustrates a sailing vessel apparently of this type with the inscription " Navium quibus Bantam 405. II. been inserted into the float Lake Wakollo or Wakoholo. utunter " (Tertia pars. or multiple float. This appears to be very rare. but when made of wood fore usually trimmed the fore end only or both ends may be pointed. 1. XVIII. 9) model. p. 83 THE FLOAT. 7). evidently to render the lashing more Bali (Fr. with three slightly 14. 2. Leiden) Palembang Madura (model. or the end and occasionally both ends may have an upward curve. 145. The toy boats The ends of Sarawak (Fig. Fig. 7. This type widely spread throughout Indonesia. p. p.. Amsterdam) .

1889. 62. 89). 228) 333 . 217) G. p. South float. 113. 192. while the aft boom is short and straight and has inserted into each end a straight spar with a downwardly FIG. which the fore boom is straight with its ends lashed to the floats. . they have a zigzag appearance and support planks for paddlers. north-east end of the Gulf of Boni and Paloppo (Grubauer. 9. 114). sampan. and if so this is an example the fore mixed direct is and Balinese think in is all these cases it is boom which lashed to and the boom which inserted into the float. 38. V. The Outriggers of Indonesian Canoes. 53 . 108) Kema. 1. In a large canoe at Amboina figured by Valentijn (No. has five booms of which the fore and aft booms are inserted into the while the three middle ones are lashed to the underside of the float (pp. Guillemard. 180. . Figs. Dresden. 10) Sulu Islands (Wilkes.84 A. aft I is inserted into the float . . Sangir (model. 8) (Pritchett. " or sakit Meyer and Eichter. Celebes. HADDON. 57) here. North Celebes (G. but the illustration is not convincing. one of which is straight while the other is downwardly curved and inserted into the float. A model of a fishing boat. JAVA (AMSTERDAM). . these booms of a are probably intended to (p. Mixed Direct Attachment. (Vojnich. 383) Zamboanga and Zebu (Miiller. Fig. from Panarukan. B. (a) Indirect. I. (Savage Landor. single outrigger . Fig. 142. in the float. pp. II. two. 244. Fig. Fig. 225 .. 114. Talaut Islands (" Geisterkahn . 9). Vol. . Manila (p. p. 3. 12 Burbidge. Fig. Fig. 2. Figs. C. . and tied to the Rotterdam Museum. 63) . In the Leiden Museum two models from Madura. Attachments which consist lashed to the more sticks. I. A model of a large plank boat in the Leiden Museum from Gowa. of one. p. p. 4. 183) appears to belong .. 5) canoe. Amsterdam. 244. Madura Strait. 39. p. 378. curved end which represent a sedek attachment. p. II. p. . XXX. 1905. 124) the booms seem to be lashed directly to the float. Attachment inserted into the Float. 106. at the . is Practically the same arrangement shown in a model from are " Java " in the Amin sterdam Museum (Fig. 62) . . Miiller. 15. has two booms. 30) " " Pirate* craft off north point of Borneo the (Fig. MODEL WITH DOUBLE OUTRIGGER AND MIXED DIRECT ATTACHMENT. 206 Cagayanes Group of Lake Mindanao and district Manila Lanao. Ussu on the Malili River and Lake Matanna. or is Stick Attachment. No. with two booms and an outlay er. Pasig Eiver. Nos. pp. 55 Savage Landor. one end of each stick is boom while the other inserted into the float. 211. 80.. PI. pp.

A good illustration of a model of a sailing canoe. The sticks may may may be vertical. or the crossing may take place over the boom. The native names are taken from Svoboda. MODEL (EDINBURGH). ONLY TWO OF THE SIX STICKS ARE SHOWN F. D) . in such a way that two sticks generally cross each other below the boom. later (XV. The two sets of three sticks diverge is line of the float (Fig. p. are tied to an adjacent short peg into the float. REPRESENTS THE ARRANC D. 193) of a toy canoe which both booms pass hull and rest on the float and wh. or there be two or more sticks. pp. U . due. F). F). deia due. E). R. Sometimes there also a central inserted vertically between these two. nor are his figures at all ment an who does not (VI. Figs.) C. and may be fore boom (Fig. 10. undercrossed. STICK ATTACHMENTS. clear of the sticks. in which case the boom typically rests on the crossing. 17) . MENT OF THE IN A. 10. and lashed to the boom. from the median pair of undercrossed sticks. as in D_ O Q C A-C FIG. while the third or aft of the may be vertical or oblique. 10. he also giv in illustration (V. PI. In the outrigger canoe. : (A. which are inserted into the float. heneme. hentahn.A. A). is given by Man (XI. converge over the boom .. which shows details of the sticks. 10. alludes to Nicobarese canoes without saying anything and two excellent plate. describe the arra 11. The Outriggers of Indonesian Canoe*. XXIV). BROWN MS. PL I. overcrossed. in oblique parallel pairs or a pair of oblique sticks may a pair of sticks may cross under the boom. 10. of the Nicobars the attachment consists typically of a double set of three sticks. which is 10. dp. MODELS (OXFO . 436 4 about outrigg. models in the Oxford Museum In a model in the Edinburgh parallel Museum one set consists of two sticks which are almost the and converge slightly over boom and a third oblique stick (Fig. (Fig. HADDON. . occasionally a pair of sticks converge over the ' boom. in which case there may be only a single stick. either on one or on both sides of the boom the stick be irregularly oblique. C. NICOBARS STICKS . C. E. the third being more oblique (Fig. 53 79) gives short accounts of the canoes.

In a model in the Horniman boom and a Museum booms each have an attachment of only two sticks which converge over the Models in the British Museum. R. Andaman (A. two on boom and one on the other. p. B). InsL. D FIG. 1877. Anthrop. chd-rigma. gi lyanga. while the three two attachments two sticks cross over the third vertical stick present (Fig. consists The Andamanese attachment crossed sticks and one vertical stick. 11. Museum. VI. of a large canoe. LITTLE ANDAMAN MODEL (CAMBRIDGE) E. Oxford. : (A. E). cf. E) or a single one. . D). A. Journ. p. 116. B. C. D. however. in the Halifax one the arrangement in the various models. B. most frequently of one set which may be on either side of two under- of the boom This arrangement occurs in one attachment in a model in the Cambridge in the other is Museum. D. ANDAMANS A. STICK ATTACHMENTS. and Edinburgh Museums. the latter shows the attachments very well.) there are three vertical sticks. and of the larger kind without outriggers. roko. 11. 209. 11. 11. A good photograph of an outrigger canoe. Brown MS. PL VII). MODEL (BRITISH MUSEUM). p. HADDON. 11. (Fig. BROWN In the Little side of the MS. is given by Man (XII. a model vertical stick. as does the plate. C. has three booms the attachment of two is typical. 345. have three booms with a single pair of oblique sticks which cross under the boom (Fig. some variation in for example. while in the third the boom passes between a pair of crossed sticks and the but this is probably due to careless workmanship. There is. GREAT ANDAMAN. even in the attachments of the several in the booms same model .86 A.) . The Outriggers of Indonesian Canoes. of ratan from the boom to the float to render the attachment more secure. boom. . There is frequently in addition a couple of lashings (Fig.

II. East African canoes (pp.) THIS IS A MADURA WAR SHIP. Pis. "THE KARKOLLA OF AND NICOLAS. It consists of a short. though. That it may have To avoid found in some mixed types (pp. a stick attachment of this kind has not been recorded for Indonesia. 125). XII. 127). confusion with the above-mentioned stick attachment this may conveniently be possible. recorded only from the north coast of Middle been has No. has not hitherto been recorded from Indonesia as an attachment for occurred there " is of the booms of a canoe. 14. 128 Haddon. be related. as Rod Attachment. Y-shaped Stick Attachment (p. however. pi. 1918. Pis. 17) on trading. The Outriggers of Indonesian Canoes. It seems to more stick-like attachment of the .and war-vessels at Ternate (p. This attachment. we shall see (p. Board Attachment. V. 9. 29)." WITH B. to an attachment figured by Nicolas (1601. Valentijn FIG ' OF TWO PLANKS THE DOUBLE OUTRIGOEB HAS A FLOAT CONSISTING ATTACHMB1 Y-BOARD A THE KING OF TERNATE. on similar craft from the Moluccas. as it is termed a rod attachment. At that time the larger boats of the XVI) Moluccas had a double outrigger of three stout straight booms. No." .A. 55). 96). where it is associated with a single outrigger and a single boom. p." Spike Attachment (pp. 79. 116. 15. (DE BRY 12 -" CORACORA A. HADDON. This attachment Java. I propose to refer to this type as a Y -board attachment. . which consists of a more or less vertical rod all or stick inserted into the float at its lower end. 110) and by De Bry (1601. its lower end is inserted into the float broad and relatively thin piece of wood while the boom through its upper end in these two respects it resembles the " board " may . C. 87 With these exceptions. each of which was board which was apparently inserted into the supported in a deep notch in a short float (Fig. 92. 12 B). passes . 79. 127). Hornell has recently described a form of attachment to which the term be applied (1919. it is very common in Oceania. 1601.

JAVA (EDINBURGH). the other being inserted into the float and reinforced by lashing (1919. The Outriggers of Indonesian Canoes. No. short. presumably to keep longitudinal spar or board ran across the ends In one coracom in place. C. " van Titaway (Fig. was apparently a straight Halmaheran attachment near each end of Generally one or two planks are fastened transversely to " In an illustration entitled De Cora-cora the booms upon which men sit to paddle. thick. (p. We may adopt Hornell's term for the type of attachment which consists of a straight or slightly curved spar one end of which is spliced. a plank on which eight men sit to paddle replaces the longitudinal spar. who gives the Bali names FIG. XLII. 2) . . p. BALI (FRIEDERICI. E. 2) (Fig. many Balinese Attachment. II. 14 A). 13. 184). A thick them there of the booms.88 A. 235. fig. Fig. there are four other similar planks on the lowermost of which are also paddlers. p. this was frequently utilized as a seat for paddlers. 363) shows these boards as also lashed to the floats on Moluccan vessels. MODEL FROM BANJUWANGJ. This type was first described for and figured by Friederici (II. 55). pegged and lashed to each end of both of the short straight booms. " DE CORA-CORA VAN TITAWAY " (VALENTIJN. 13) there are six outrigger booms which are supported by as FIG. No. the longitudinal spar. 14. B. squared bars of wood which appear to be inserted into the float. BALINESE ATTACHMENT : A. HADDON.

p. Nos. Pis. The typical Moluccan attachment ratan. and the Pfluger. and Hornell gives a good photograph (1919. FROM la). Bali. p. 239.. dukong. 235). p. 1894. K. II.. on aft side of all the booms) 161)^. Kei Group (model. 240. Ill. Amboina . 27a III." If the sedek were a form of Halmaheran attachment one would expect it to be only lashed to the float.. 242. II.. p. : A. Baessler. 3. 232. 14 B). C. 15).. 331. 161 BATJAN B. XII. north of Ceram 239) . 73). G. 240.. 89 boom." III. As the sedek is virtually an integral part of the boom an added piece. II. Ombi. p. PI.Fig. in the Edinburgh Museum with this attachment (Fig. . (K. Leiden. C. Figs. Fig. p. Pis. p. brayunan. . p. Ceram Uliassers Ill. is and the base is lashed to the float (Figs. a device for extending the boom downwards so as to save the necessity for searching for a naturally-bent piece of wood with which to Hornell suggests that this is construct an outrigger-boom of the required form.. Fr. 3. 8 Fr. FT. XLIII. and figures details of the sedek from Boleleng. 237 . MOLUCCAN ATTACHMENTS PHOTOGRAPHS. Ceram. Leiden). PL in the south-east islands of the " . ." should be described as a speaking (6) Attachment tied to the Float. pp. form the float. (A. (Fr. . 11. Fig. II. float. usual Galela. if it whereas it is inserted into it. " Tobelo anc G No 327) northern Halmahera (" occasionally at Ake-Selaka. . . on the Bali Strait. II. . 237. No. 12 . 232. II. the " sedek may be regarded as one form of the boom-prolongation. 11.II. Buru from model. 237 Banda (Fr. AMBOINA B. 161). here and 7a. Fig. Martin. .. at Amboina (Fr. Fig." sedek. 78. p. The Outriggers of Indonesian Canoes. AFTER FRIEDERICI. p. as it appropriately would be is were derived from a direct inserted attachment the upper end and not a spar lashed on to " is strictly it. predominates . . . 161) Buton (Fr. 5.. Moluccan. BANDA fig...Pl. p. A) of a Lombok canoe. kater. 161. 8. from Banjuwangi. C. Batu mera (Riedel. . I Fig. 328. 15. 1894. Java. Ternate (Kukenthal. G. XXVIII. intermediate piece. Batjan (Fr. II. 334. (Fr. " HADDON. p. 1 attachment which is the common form at . model. FIG. There is a model of a sailing boat. p." greatlylpredominates' 336). 4) Fig. II. the horns of which are lashed to one side of formed by a U -shaped piece of the free end of each of the booms.. Amsterdam) there. As I have stated (p. Martin. 15) " " . p. It has been noted from models at Wetta (Riedel. PI.. p. " " coast with the exception of a few places on the north (Fr.. VII. G. . with three booms and three bamboos to attachment which predomu Other varieties of this type are the O-shaped 131 models. Fig. G. Fr. this form of attachment boom it indirect. XLI.A." but if it be regarded merely as a prolongation of the " direct inserted attachment.

. 5. 6. canoe (Fig. 16. north coast of of the central Bum (K. AND D. Leiden). 1894. 22. Martin. 336). 17.. 76 and Fig. B." even when it is a simple " " " " or elbow-stanchion type of Hornell. . may be vertical. HADDON. The simplest condition consists of a straight spar which Batjan canoe (Eiedel. 4. The rod or stick. HALMAHERAN ATTACHMENTS A. 16. D) may be a variant. The attachment The spar may p. AND AN INNER AND AN OUTER LONGITUDINAL SPAR. p." Molukken-Verbindung." was lashed above to a boom and below introduced by Friederici (II. 17). This attachment consists of a variously shaped spar which is The Moluccan attachments . 336) and in a sailing vessel. 328. I assume that the spar tied to and not inserted into FIG. Fig. bero. Nos. C. DAM) F. PL XXVIII. Martin. 82. 16). p.. 17. 4. MODEL FROM TALAUT (AMSTER. booms of the Sangir XIV) and Banda (G. Fig. PI. FIG. 4 . WED A BAY. The Outriggers of Indonesian Canoes. 239). This is the oblique " " like that of term Halmaheira-Verbindung. : .. 4. 267) or oblique. Bay of Bara. BUTON HALMAHERA (FRIEDERICI II. BATCHAN (PHOTO G. North Ceram (K. as in a 16 F . (Fig. Misol (G. p." and has three booms. 235 ff. Batjan (G. 331). the float in this case) . I propose spar. 27) AKKE SELAKA. 331). 1903.. as in a canoe on the PI. of boats between Selang and Batjan figured by " Molucca CoroForrest (pi. . E. lyre tanjong coro. which is inserted " element to as a refer this to into the float. 2. tripod Halmaheran. to the float (Figs. figs. CANOE WITH DOUBLE OUTRIGGER. 86) are not very clear the former is a mast and a a sail. HALMAHERAN ATTACHMENT. or oblique. No. C. usually it is more or less vertical. BATJAN (FROM PHOTO G. G. 346).. 9. pi. pp.90 A. at the is Tenimber Islands XXVII.. " " stick In order to prevent confusion with the attachment. be forked and practically horizontal (model from Tenimber.

(" common. Fig. and can other parts of New Guinea and undoubtedly So far in Halmahera (Fig. 99) II. The Outriggers of Indonesian Canoes. XXVII.. Pogoyama .. 16 . 267) Hornell. I. . II. 242) .. 232. 6). 9. but no description) River. 5) Menado PI. 29) . . PI. photo. 9 . 99) 233 . p. angled spar would result. 17. 85). Tobelo and Galela (" greatly predominates. HADDON.. (G. Patani and Buli (two booms. p. G. . and indistinct east it is < photograph . sometimes two varieties may PI. " type Tidor (K. model. since nearly every variety occurs data go. .e. XLVI. in the and elsewhere several varieties are found same spot . . Nos. B. River. 147) Tenimber (Riedel. No. . and Kruyt. PI.. p. p." Fr. p. Batjan . p. (Pfluger. "common 331. 1894. but one type of outrigger of Geelvink Bay." Hornell. which may be so slight as to be almost with the ends pointing upwards as (Figs. Ake Selaka (Fr. Limbe Island (Guillemard. Martin. 99) Banda and Ambon (" occasionally. Hornell.. Leiden) (Hornell. Xulla. p. p. 242). p. No.A. Fig.. p. in Krieger. 171). Fig. Celebes : Makassar (R. 91 If Most frequently the spars are angled. 240. be found on a single canoe. so far as A and E). Minahassa (Dumont d'Urville L' Astrolabe. Martin. 243. Chap. II. Figs. 232). Fig. p. 3 models. (p.. p. No. G. 1903. pp. Martin. 161 1894. 376. and the upper part bowed (Fig. 1894. 99) Timor . Ill. Gulf of Boni (Grubauer. do not appear to be significant as regards on the coasts of or on the islands immediately adjacent to Celebes. 325) Pis. 239) Ombi (Obi) (Fr. PI. . 336) . 239 . The stem may be is straight may have a slight sigmoid flexure. and Fig. Gowa (model. 27). p. p. the main stem of a forked spar were cut off above fork an the immediately 16). Misol (G. or bent in various ways (Fig. II. Aru Islands PI. Fig. on the the two the double outrigger has four booms. 192. elbowed. 233) . 239. as in a in a Buton canoe (Fr. p. South-east Celebes (Sarasin. No.. pp. 240. north coast (K. . (" a few. judging Kampauer.. (0. . p. 10) . with the ends pointing downwards. . 239 . often with a central false-boom. where and to Manukwari (Dorei) at the westerly opening other attachments. 242). 16 Occasionally the spar sharply bent.) . Fr. Leiden) Halmahera : Weda Bay (Fr. North-west Tontoli (Toli-Toli). the outrigger Although they are nominally beyond to as they differ from those of canoes of North-west New Guinea must be alluded are of direct Indonesian origin. Ceram. It occurs at Lombok . 242. Martin. Leiden) Dobbo. ? Baba . Celebes (Pfluger. 233) Ternate (model. p. 22. Konaweha . 99) .. II. my form or distribution. 235. Bum (K. at the north-east corner of the head of the Gulf of Gorontalo or Tomini : Malili 8. Hornell.. Leiden) . be perfectly matched by a type from Weda Bay extends from Skroe to Waigiu and posa as I can gather. Warburg photo. p. i.. model. 6) . Figs.. . 239.. 204). PI. or model from Talaut in the Amsterdam Museum These variations. Gulf of Tomini (Adriani Atlas-. 328. XIV Hornell. pp. XXVIII. 4). p." Fr. p. 23). the scope of this paper. (photo. 1894. Martin.. Fr. Buton (Fr." Fr. Fr. 1894. Todjo. . 235. or Sula (K. C. 4) . p. Kema (G. replaced by in 1899 on the north shore At Skroe (a port founded by the Dutch from Pfliiger's sma south side of Onin Peninsula). the spar straight.

Fig. lies the unshipped triangle-mast and on the other the rolled-up sail (Fr. from Probolingo. p. sometimes it is an elbowed spar an inner longitudinal spar passes its aft end. 427. over the four booms and underneath the angle of the spars. I the booms of a now proceed to give examples of mixed attachments. Prichett (p.. Hornell. 248). C. 385). a Halmaheran attachment of the Waigiu type. this be lashed to the boom or connected with the float by means of may pass through it the lower end appears to be inserted . there and the two is aft of which are nearer together than are the two central ones . is . Apparently. Waigiu (Waigeil). double outriggers.. The larger have a platform with side on one side and an atap On the booms. like sledge-runners. p. The four booms the outrigger stretch across the wash-strakes..92 A. roof. 248. Figs. HADDON. i. The Outriggers of Indonesian Canoes. who removed there in 1865 from Sorong. Nos. as in Tahiti.. is The same applies to the aft boom except that the boom-prolongation is thicker and with a swollen end. In the interior of the hull are transverse struts and lateral longitudinal spars as at at Halmahera to the west. Manukwari and Japeri ( Jobi) in Geelvink Bay to the east and The sides of the larger canoes are heightened by superof of imposed wash-strakes. on the north side of Onin Peninsula Kreiger says they have very high boats with outriggers (p. 30 . p. p. according to Pfliiger's photographs (pp. the last attachment spar being near a Halmaheran attachment of a long fairly straight spar with a bent end. 84). all In the foregoing accounts the attachments are similar on canoe. direct attachment (p. 28 . shows the same type of canoe as that recorded by Friederici (II. on the mainland of New Guinea The dug-out or plank-built canoes have Hornell. II. and. upwardly curved a rod which may . p. " or patnati). but in small canoes they rest directly on the edges floats The two There may be shorter than the hull. Dr. Guillemard's photographs (G. as in Sorong. 99). 99) at Saonek. and 1889. 373) of canoes in Chabrol Bay. . to which they are lashed by means ledges (" Leisten of the dug-out. run far forward.e. sukung. which have a crescentic or other form . a sailing canoe. the instance craft is float in this composed of two bamboos instead rails of the usual single one. 248. 174. and not by gabbagabba as in further east. 29. 175) a similar type occurs at Sekar (Segaar) on the south shore of Telok Berow (Berou) or McClure Inlet. The mixed been described and mixed direct and Balinese attachment have already 175) gives a drawing (Fig. Friederici figures an attachment at Saonek with an additional outer longitudinal spar. C. 305. p. Mixed Attachments. there are usually forked supports for gear. Madura Strait. but its diameter markedly varies is in parts. or it may be that the boom is in one piece. II. or Kaap Noi. or. in which the fore boom appears to be a boom-prolongation which is lashed to the float. on both sides of the canoe. a village near Cape Spencer. an islet near the south coast of Waigiu and among the Sorong (Soron) people on the island of Dom. 18) Mixed oi Direct and Rod Attachment. (Fr.

Both of the short booms have a boom-prolongation. AND ROD ATTACHMENT. Madura seems to clear up the points that are doubtful in Pritchett's drawing. The one forward outrigger supports being low down and FIG. but the drawing is not decisive as regards these two point*. 19.A. 175). the fore one is lashed to the float. C. 19) of a sekong in the Rotterdam Museum from Pasuruan. FISHING BOAT. a lashing also connects boom-prolongation and one is boom and the float at this spot. MODEL OF A Sekong WITH MIXED DIRECT E. sukung. 18. JAVA (ROTTERDAM). The Outriggers of Indonesian Canoes. " to allow the wash to pass under freely when the in vessel is at her high speed (p. p. 174). . 93 into the float. WITH MIXED DIRECT AND ROD ATTACHMENT. HADDON. (Fig. a rod passes expanded and decorated with the into is inserted float. A model Strait. the FIG. PAStfRAN. MADURA STRAIT (PRITCHETT. PROBOLINGO. free end of the aft through this the fret carving. that aft curving up pronouncedly. In " " " referring to the he says.

main points the outrigger agrees with that of the Probolingo boat. p. 23). steamer. it must be made of bent ratan. 20. a single piece of wood. as his sketch indicates. The same author gives a sketch (Fig. as it is inconceivable that the whole apparatus could be made out of FIG. it therefore suppose that we In either case the form requires explanation consists of the characteristically upwardly curved . 244). fig. In the region embracing North Celebes. each divergent limb being connected by means of a vertical T- FIG. the fore . . 21. 1912. C. Thus in all the . being floats almost U-shaped. the Talaut (including Sangir) and Tulur Groups canoes have two outriggers and many of them have but two booms. 21) of a canoe from the Bawean Islands the fore . If. boom is straight. boom is a doubly bent yoke-shaped bar.94 A. boom he made . Fig. Banka. it falls in apparently inserted into the Miiller states that previous examples. how it is attached to the float is not evident. SKETCH OF A MIXED DIRECT AND BOD ATTACHMENT. The Outriggers of Indo'nesian Canoes. 22) of a canoe from Madura (Fig. 244. the ends of which are lashed to the the aft boom. if this be with the these sketches whilst on board a Mixed Direct and Halmaheran Attachment. 1912. or boom-prolongation. The author states that at the stern the bamboo floats are suspended without spars (Auslegerstange) from an elastic bent rod which is concave above (p. 20). 22). In a sketch by Miiller (1912. HADDON. fig. this rod is all in one piece. shaped rod with the float the transverse upper end of the rod is presumably lashed to the boom and probably its lower end is inserted into the float. but it is a direct attachment. is strongly curved upwardly. the lower end so. SKETCH OF A MIXED DIRECT AND ROD ATTACHMENT. may boom or boom-prolongation to each of which is end of which a vertical rod is fastened. BAWEAN ISLANDS (FROM MULLER. MADURA (FROM MULLER.

as Dumont Halmaheran attachment " (III. directly to the float. 161). It may be a spar of this kind to which Graafland the two outer (fore and aft) In Sangir. p. According to the pp. No. but Figs. boom is Moluccan attachment.A. In some dug-outs at Kema. booms of the double outrigger and their boom-spars bend downwards and are . 1. the drawings (Atlas. p. Bugi. 234. and the shorter straight aft float by a bent Halmaheran spar (G. hndi. The aft boom is straight and its ends are connected with the floats FIG.n. of Indonesian Canoes. The fore boom is curved and its ends 2) prove. Makassar. boom is attached to the A model canoe from the Talaut Islands in the Amsterdam Museum has two straight booms.. 244). 22.-TAe : Outrigger. 67). AND HALMAHERAN ATTACHMENT TALISSE CELEBES (HICKSON). of which the fore one has a direct tied attachment and the aft one has a to the bent or bowed spar attachment. SAILING CANOE. (Fig. f. 405) this might be considered as a d'Urville's semi. but in this case the free ends of the spar are lashed boom and the bend to the float. 22). in Tahiti and at the Marquesas-is fastened connected by means of an S-shaped spar with the lescription of Graafland (Minahassa. 16 E). according to Miiller (1912. Hickson gives a figure of a model of a sailing dug-out. II. HADDON. N. PI. C. the fore-boom is downwardly curved with a direct lashed attachment. slightly to the south of Banka Strait. Friederici says In Minahassa there are outrigger boats whose fore boom is curved and-^xactly as in Bali. p. in the strongly are lashed to the floats which are generally made of two or three pieces of thick bamboo firmly lashed together. while the aft straight float.. WITH MIXED DIRECT ISLAND. from Talisse Banka Strait (Fig. 22). 1. it thus bears some resemblance to the U Moluccan attachment refers. by means of an S-shaped Halmaheran attachment (Fig. londi. 404. at the Mariannes. it is really.E.

floats. shows a canoe with attachments which. 77. the other two short and terminal. I do not know of any other example of two floats on each side each with its own attachments. Mixed Rod and Halmaheran Attachment. one long and low down The outer attached to the brow tine of a deer's antler). The mast of is stepped in a socket hinder consisting of a section of fore of bamboo. the lower part of the bow is produced into a long spur which rises in a gentle curve. on which an equally short upwardly curved spar. i. consisting of a short straight or boom-spar. by two vertical sticks. There are two floats on each side. Maclear of the Challenger. or possibly from the Nanusa Islands as the ends of the canoe resemble those of the sakit canoes described by Hickson (cf. the two fore booms lie very do the two float is aft booms. the ends of which are kept in place by passing below projections left in the inner sides of the hull. attachment. figs. the double. (PHOTO G. No. attachment. CELEBES 232). The Outriggers of Indonesian Canoes... 2 D.E. HADDON.. the one attached to the inner float a spar rests on the lowest branch of the fork and is lashed in its middle to the socket of the mast. C. and have a J_ -shaped attachment end of which is lashed to the double floats. the upper end is which is lashed to the it. 23. N. 83. are distinct At both ends is canoe there is boom. whereas the central booms have the structure shown spar. The inner boom is secured by two lashings within . treble if they were from North Celebes came the the can be no doubt There that specimen simple area. the lower connected directly with the in Fig. the upper is part being prolonged into a short upwardly slanting beak. 1. each of which has the ordinary pair of booms close together. The its aft fore booms have a direct lashed attachment. possibly from the Talaut Group. The fore forks are lashed to the stouter and hinder of the two fore booms." The dug-out has notched ends. while the lower end steadied by a cross-bar passing through the ends which abut against the inner sides of the hull.96 A. are treated as floats. There Capt. 15 B). (like boom with a bowed Halmaheran while the inner is attached to its aft boom by a ^-shaped Halmaheran The forks on the booms have three branches. MIXED ROD AND HALMAHERAN ATTACHMENT KEMA. is At what is presumably the its end of the canoe there is a vertical rod which fastened to the ends of boom FIG. from any other known to me. the hull of the canoe to a cross-bar. as . pp. a framework Minahassa. these are braced fore A photograph of the (Gr. which was collected by " P. . North-east Celebes.e. D. a very remarkable model in the Oxford Museum. or multiple floats previously noted (p. 2. 232) at Kema. J. boom. 113).

418) other authors do not appear to mention them either. 79. however. 83 . The first of these attachments somewhat resembles the aft attachment of the Bawean canoe This Kema type can easily be (Fig. as has occurred in the craft of the harbours Kei Islands. eloha. except perhaps for its attachment to the float. " These are long narrow dug-outs [cobara-eloha]. 206). The main point to remember is that in both groups there is an inserted stick attachment. . the usual figure being as if . man with arms and legs extended in-coiiu or that of a frog to keep apart the sides of the canoe grotesque figures the specimens of these quaint Engano being swallowed by a snake are amongst . while in the Nicobars has but two. figured by Kosenberg VIII. Tfa Outriggers of Indonesian Canoes. with single . which appears to be a true Halmaheran attachment. The Mentawei war-ship. According to Modigliani. resolved into the Sangir type (Fig. 80. Fig. from Mentawei.-shaped spar of the Sangir type. but in the Andamans the it single outrigger has several booms. canoes are very rare in Nias. Giglioli. collected by Modigliani. two booms and a direct lashed attachment. 2. 86. says or seats are often two with made from a single tree-trunk. which have supplanted the older types. The NOTES ON THE CHARACTERISTIC OUTRIGGER CANOES OF THE MAIN DISTRICTS OP INDONESIA. and negative evidence is full of pitfalls. PI. I. a marked absence of outriggers from the greater part of Sumatra and Java and from the whole of Borneo. C. a Eosenberg (1878.A. I feel considerable hesitation. 97 the rod to be inserted into the float. outrigger. p. H ADDON. of Engano they Modigliani describes the simple but seaworthy canoes. 21). D) if the booms and upwardly curved boom-spars of the latter were shortened and the oblique sticks or stays were greatly placed is and curved boom-spar and appears straight rod also corresponds fairly closely with the J. which is evidently due to a knowledge of the There is art of building sea-going plank boats and ships. (1888. from the in describing the specimens hull with a direct tied attachment. 85. and he does not mention an outrigger (1890. outriggers . At the other end replaced by a bent spar. p. in the notches four in rests which gunwale a are dug-outs with double outrigger deep booms and floats far away two shows Museum Leiden in the A model (1894. it has apparently a single outrigger which on the starboard and a direct lashed attachment. 176) figures a canoe. side 78. abak. vertically. as it is lashed to the float. as I have been able to find very little positive evidence one way or the other. and rivers The general absence of outriggers from the small of Borneo is not so easy of explanation. is referred to on is pp. in making definite statements concerning the distribution of outrigger canoes in certain areas. 74. the cross-pieces : that of a beautifully carved. The outrigger canoes of the Andamans and Nicobars have been sufficiently dealt with on pp. Tcnabat bogolu. p. 9).


A. C.


The Outriggers of Indonesian Canoes.


sails are

and the paddles

are plain


of the usual



(1893, p. 131).

At the present day,
(II, p.

outriggers appear to be scarce in Sumatra.


235) states that outriggers have almost disappeared in the Singapore,


and Biliton


narrates in his Voyages that the Nicobar canoe in which

he sailed to Achin had

good outlayers lashed very

and firm on each

side the

So that while these continued firm, the vessel vessel, being made of strong poles. could not overset ... we were therefore much beholden to our Achinese companions
for this contrivance



(quoted from Kloss, pp. 267, 268). This was in 1688 that the were true outriggers and not outlayers (p. 76) seems probable, as in


his description of his visit to the Nicobars, Dampier speaks of the " of the local canoes (loc. cit., p. 260), and, as outlayers on one side

"small slight we have seen

78), the

Nicobarese canoes have single -outriggers


none at


refers to the long narrow jellore and ballelang of Sumatra which are fitted (p. 480) with double outriggers, which stand out a considerable distance from the sides." He figures a jellore with the sail partly furled, the ends of the two booms curve down-


wardly and apparently are inserted into the floats. Folkard does not say where he saw these craft, probably it was on the coast of the Palembang district he adds,

Other sailing vessels without jellores have sometimes only one outrigger." In the Amsterdam Museum there outriggers are the panchallang and the bantang. is a model of a canoe from a with double outrigger, two booms and a Palembang
direct tied attachment.
Giglioli (1893, p. 116



Fig. 8) describes the



dug-outs with a wash-strake but no outrigger


bow and

stern decorations are

noteworthy, they consist of sticks with tufts of hair and a central phallus the bow in addition has a wooden carved and painted buffalo head. Brenner (1894, p. 284)
refers to

two kinds

of keel-less dug-outs


the Batak of Lake Toba, the solu

ratsaran or fishing canoe,

and the

solu bolon for trade

and war



of the latter,

he says, are provided with an outrigger, but he gives neither an illustration nor a Fischer (p. 114) refers to a model in the Leiden Museum of a sail boat, description.

from Padang, with a double outrigger of two feeble bowed booms


Dr. Juynboll informs the


that these are half inserted and attached


a nail


Hornell's observation that North-Central Java


a locality where the outrigger
of properly built boats (1919,

pattern of canoe has long
p. 98)

been discarded in favour

supports Friederici's remark that outriggers have almost disappeared from.the

Javan coasts

References have been made (pp. 82, 87) to a degenerate (II, p. 235). canoe from North Java described by Hornell. There is a distinct type of attachment in the Eastern Javan area (Madura

Madura, and the Bawean Islands) which has already been described (pp. 92-94, Miiller remarks (p. 244) that the Bawean and Madura boats are quite Figs. 18-21).

isolated in the western archipelago,

and only

find affinities in the

extreme east in the

A. C.


The Outriggers of Indonesian Canoes.
In both areas the fore


South Philippine
while the aft
Fig. 2,

local group.


has a direct attachment

the Sangir model as a parent type from which the Eastern Javan and South Philippine types have diverged, and appears to regard the upwardly curved aft booms of the


has an indirect attachment.

Miiller takes



" boats as the equivalent of the " upwardly curved Biigel (boom-spars, p. 73) of the South Philippine area It seems to me much more reasonable (Figs. 2, A, and 7). to regard them as true booms, or boom-prolongations, which are recurved, and not as reduced boom-spars of the Sangir model, of which the outrigger-boom and oblique stays have disappeared. In my opinion a much closer analogy is to be met with in the Northern Celeban types illustrated by Figs. 22 and 23, the main difference that in the latter the attachment spar is tied to the float as in the being ordinary Halmaheran attachment, whereas in the Eastern Javan type the attachment rod appears to be always inserted into the float. The rod-attachment of the Kema (North Celebes) canoe is associated, as we have seen with a Halmaheran
(p. 96),


The Eastern Javan area


that of the Balinese attachment

(p. 88).

contiguous to and partly overlapped by Juynboll describes (p. 37) a model in the

paduwang from East Java) with a double outrigger of two booms which are tied to the float. In Madura and Bali the attachment may be direct and lashed In (p. 83).



of a trading vessel (Madura,


Strait the straight fore-boom


be lashed to and the curved aft


i.e., a mixed direct attachment (p. 84). Typical of Bali the Balinese attachment, which consists of a spar rigidly fastened by one end to the boom, the other end being inserted into the float it also occurs at Lombok

inserted into the float


(p. 88).

A mixed direct lashed and

Balinese attachment occurs at


(p. 84).

A mixed direct lashed and rod-attachment occurs in Madura Strait and at the Bawean
Islands, the rod being lashed to or inserted through the

boom and

inserted into

the float

(p. 92).

We may

thus define an Eastern Javanese area which includes the extreme

eastern end of that island, Madura, Bawean, and Bali, and


by the

occurrence of attachments which form a gradation of types



In Bali both booms

have the







lashed directly to the float

the aft


(a) inserted into the float,

a prolongation of

the sedek,


inserted into the float,

and inserted into or lashed to the boom, or (d) the It is always risky to suggest an evolutionary aft boom is lashed directly to the float. series, but it looks as if an inserted direct attachment, here usually under the modified form of the sedek, might have been the earliest form for both booms. The greatest
a rod

inserted into the float

strain in

an outrigger
it will


at its fore end,

adopted to counteract
the aft


and a lashed attachment might have been while the assumed primitive form would be retained for

provided for the sedek for greater The rod attachment is probably a modification of the sedek. It would security. and the float to convert only require that the rod should be lashed to both the boom c 2



be noted that a lashing



The Outriggers of Indonesian Canoes.


into a simple form of

Halmaheran attachment.

The lashing would give at the

same time greater strength and elasticity and probably less liability to fracture. The colonization of Java from India, according to Ha veil, was probably a sequence of the final collapse of the Saka power in India at the beginning of the fifth century,

when the kingdom
the Saka

of Sarashtra or

Kathiawar, which had been ruled for centuries by

dynasty, of foreign origin,

was conquered by Chandra-gupta

II (Vikra-

maditya) between A.D. 388 and 401. This great monarch, although tolerant of " After Buddhism and Jainism, was himself an orthodox Hindu (V. A. Smith, p. 292).
that Brahmanism supplanted


as the principal State religion of India, the

Buddhist art traditions went with the Saka immigrants to Java, where they reached " their highest expression in the magnificent sculptures of Borobudur (Ha veil,
p. 113).




having been


say the [Javanese] chronicles,


to a king of Kuj'rat,

or Gujerat, that his kingdom would decay and go to ruin altogether, the Prince and embarked him with about five thousand resolved to send his son to Java



followers for that island.

of a

these followers were people skilled in agriculture,


learned in medicine, able writers, and military men.


sailed in

six large ships

and upwards





in A.D. 603.

Later on

From this period,' continue the a reinforcement was sent of two thousand people.] an extensive commerce chronicles, Java was known and celebrated as a kingdom

was carried on with Gujerat and other countries. During the sovereignty of the Prince and his two immediate successors, the country advanced in fame and artists especially in stone and metals arrived from distant countries prosperity







and at Borobudur in Kedu during these periods and temples were constructed artists India.' The from building of the splendid shrine of Borobudur, the by most magnificent monument of Buddhist art in the whole of Asia, is ascribed to







circa A.D.

must have spread over several centuries. It was not in fact entirely completed before the Buddhist faith in Java was superseded " by orthodox Brahmanism as the State religion, about the tenth century (Havell,
750 to 800, but the decoration of

pp. Ill, 112).

Brawidjaja of

Leemans, however, states that, according to the annals of Java, Kalinga founded the Empire of Mendang Kamoulan in Java. This

prince arrived in the year 525 of Saka (A.D. 603).


transferred the religion of

Hindustan to Java was not war, these were not conquests, it was commerce and navigation (p. 541). The fall of the Empire of Borobudur took place towards the end
of the tenth


(p. 537).

Assuming that there was a
A.D. 603, it presupposes a

large organized expedition
of the island

from India to Java in





knowledge an undetermined time previously there must have been voyages to and fro. events, we can date the sculptures of the ships at latest within the eighth and


of its suitability for colonization,

tenth centuries and the types of the ships may have been common much earlier. These carvings are of especial value in the present connection as they are the earliest

27) the straight booms pass FIG. 24. other five ships bipedal mast. : below the single longitudinal spar. Four have two masts. The tripod have outriggers which we may suppose were double. 101 records of outriggers. and in a fourth (24) they appear to pass in front of both elements. THBEE STRAIGHT BOOMS WHICH PASS BETWEEN THE TWO LONGITUDINAL SPARS AND OVER THE DOUBLE FLOAT. In two a single series of rungs project from the single or both masts. the other (26) has a single one one (24) clearly has a double or and in others it is . pi. one having a simple mast and the other a one. as four show a port outrigger. there may pass below or over a gunwale board. and also in the aft mast of another (27). ciii. it is possible they may have been connected with it. another (27) has four booms. which usually pass over the gunwale board and under . possible that be bipedal or tripod masts. but in this ship the foremast has two rungs. of which several varieties were fitted to these ocean-faring plank-built ships. 27) the ends of these booms lie well above the float. INDO-JAVANESE SHIP TWO BIPED MASTS. In four ships. In one ship (28) the rungs to abut appear against rope. in association with the straight booms are an equal number of downwardly curved booms. The Outriggers of Indonesian Canoes. while the three booms of the fifth All these booms (28) do not appear to be straight. C. 25) they pass between the two longitudinal spars. 176). and one a starboard outrigger. two of which are without outriggers. In two ships (26. but the details are obscure. THREE CURVED BOOMS WHICH PASS OVER THE TWO LONGITUDINAL SPARS AND OVER THE DOUBLE FLOAT (LEEMANS. HADDON.A. in one (26) they appear to pass behind the inner element of the double float. Three ships have three straight outrigger-booms. Representations of seven ships are given in Leemans' atlas of " Boro-Boedoer " (some of which have been copied by Radhakumud Mookerji). in two (24. In two ships (25.


216). FOUB CUBVED BOOMS WHICH PASS OVEB THE LONGITUDINAL SPAB AND BETWEEN THE TWO ELEMENTS OF THE DOUBLE FLOAT AND PROJECT BEYOND THE OUTER OF THESE (LEEMANS. 27) the curved booms pass between the two elements in the other.. The Outriggers of Indonesian Canoes. 106). .A. THEIR ENDS ARE PRESUMABLY LASHED TO THE DOUBLE FLOAT (LEEMANS. When one longitudinal spar if there are (in front of) it. HADDON. 28. : FIG. 103 FIG. 27. another board or rail. of the outer element of the double float and curl under and more or less to the front . 27) they over both in one longitudinal spars they curve is and doubtfully over the inner and behind the outer (i. Ixviii.e. THBEE rNDO-JAVANESE SHIP BOOMS WHICH ABE PBOBABLY SLIGHTLY CURVED. : pi. FOUB STRAIGHT BOOMS WHICH PASS BELOW THE LONGITUDINAL SPAB. or two curve over present (26. cxxiii. C. they pass between them) In two cases (25. INDO-JAVANESE SHIP TWO MASTS WITH RUNGS. pi. TWO MASTS WITH BUNGS.

If they come out of the vessel they may be regarded as true outrigger- booms. or as the curved booms of Pritchett's " pirate craft " (Fig. in two others they appear to reach the double float. there were technical difficulties to its be overcome in representing so complex an object as a two-masted sailing ship with gear in relatively low relief. the straight booms being absent. Owing to the kindness of L. then the have been intended simply to strengthen the outrigger. Probably they were artists who had no practical knowledge of sea-craft . on the other hand. this spar is further supported by two false booms to which it is tied. C. of the latter. and a central lashing passes between this spar and the float. Salem. which evidently were supplied to them. but this is not so markedly the case in the fourth case (Fig. but if this be so the straight booms straight ones for. and it is amazing they did it so well.104 A. and longitudinal spars of the curved booms to the float. bears a superficial resemblance to Eosenberg's model of a sailing canoe from the (Fig. further. 24. curved booms. We may must remember that these craft were ocean-going sailing vessels. of the Peabody Museum. contrivances frequently persist embracing both elements in Indonesian canoes. two ships resemble the booms of existing boats with a Halmaheran attachment. . The Outriggers of Indonesian Canoes. If the curved booms are the essential booms. 8).. and the ends booms. between the two elements of the float. in three cases they seem to come out of the side of the vessel. 1). One gets the impression that the sculptors did not really understand the details of the working drawings. It is evident that the straight must have been lashed The together. but whether they were actually lashed to it is uncertain. What is the nature of the curved booms ? At first sight they appear to be fairly typical Halmaheran attachments. here the curved booms apparently meet the straight booms on the float . W. In one (26) of the two ships with a single internal longitudinal spar. HADDON. the double float of which is apparently lashed to one boom out coming horizontally from the gunwale of the canoe and to another slanting down " " from the roof of the cabin . but. In the fifth ship (28) the longitudinal spar is absent and the ends of the slightly curved three booms pass to the outer side of the double float. The most problematical of these straight booms in is it Fig. the model lacks the longitudinal spars which I regard as being present in the Indo-Javan vessel. but these usually are not indicated. the two elements of which are shown as being tied together I regard these three booms as the equivalents of the curved booms. Jenkins. These It is very difficult to arrive at a satisfactory conclusion as to the construction of 'these vessels and to the real nature and structure of the outrigger. in the fourth in another (26) they simply pass (24) they pass in front of both elements. it is possible that this slanting boom or boom spar Mentawei Islands may correspond with the curved boom of the carving . and consequently have to be accounted the outrigger had to be very strong. Mass. 24). somewhat analogous to the central booms of the Sulu canoes.

29. " The other spars on the curved booms form a platform. have already suggested how a Halmaheran attachment might arise from an inserted rod-attachment (p. 126) the two straight booms are supported by inserted sticks. I when the size of the vessel was increased. In the Marshall Group (pp. The Outriggers of Indonesian Canoes. FIG. The earliest evidence on ? . as appears to be the case in some of these carvings. 81. of a straight With the increase in the size of the vessel and the assumed addition boom to strengthen the outrigger we Indoget the apparently typical If lighter vessels were built the straight boom might Javanese arrangement.A. One other alternative presents itself : it is possible that the Halmaheran attachment may be derived from a curved outrigger-boom." This model supports the suggestion that the straight booms of the Indo-Javanese craft merely supported the curved booms. 33). HADDON. Granting a pre-existing Halmaheran attachment. 29) a canoe model from Nonuti (Nanouti) Island. 99). NONUTI (NANOUTI) GILBERT GROUP (SALEM). which has a single outrigger of three curved booms. Whence did these sea-faring vessels which bore Indian immigrants to Java obtain their outriggers solely of At the present time the South Indian outrigger is composed this two booms with a direct tied attachment. it of the curved boom be retained persist and only the curved terminal portion . which are attached to the float in the same manner as that of the Funafuti canoe There are (Fig. C. 105 I am able to illustrate (Fig. which appear to have been the essential ones. its upper part of the attachment may have been prolonged so as to enable it to be secured firmly to the hull of the vessel. WITH SKETCH FKOM A PHOTOGRAPH OF A MODEL OF A CANOE WITH THREE CURVED BOOMS A DIRECT LASHED ATTACHMENT AND TWO STRAIGHT BOOMS. There is also a weather platform. in addition two straight booms which are lashed on to the short longitudinal spar. would then become the means of connecting the straight boom with the float. Gilbert Group.

from Goa and Cochin with a single outrigger of two booms which on the gunwales and on the . 146). lonco war vessels small. are lashed to the underside of the float (p. an inserted so far as I stick . (I. South Celebes. Bali has already been dealt with. be. in attachment occurs in Sumba and Lomblen. : sailing cathar tied. the Javan craft consisted of two-masted merchant ships. judging from an indistinct photograph like it is. HADDON. and a Moluccan at Batu merah. Hornell states (p. 99) that a Halmaheran attach- ment which is common occurs. know. . This negative evidence ships be true outriggers. C. parao. probably was well known to the if the outrigger was already established in Indonesia through centuries of trading intercourse. or that it originated in Indonesia and was adopted by Hindus great expedition of A.106 A. Hickson has given it A direct lashed me a photograph of Coupang. In 1596. which was sometimes connected with the float by a lashing. with or without a sail. the bent booms of the Javan booms then they might be compared with the modern South Indian except for their number and the curled lower ends supporting the float. in the south-east. flat method of fastening is not clear it may upper surface of the large float. in Lombok. the 603 surely indicates that the country to which they voyaged leaders. XIV. probably sail attachment. for vessels sailing in this region. An additional argument for Indonesian origin is supplied by the presence of the longitudinal spar. Timor. A direct lashed attachment has been seen in a fresh water creek at Totoat. according to L. So far Javan vessels would is as I am aware. Moluccan at Wetta. The Lesser Sunda I have been able to gather but very little information as regards these islands. point I have been able to discover PI. while they carried a large which gave them marvellous speed. in which case those of the is would pass through undercut If holes. p. by no means conclusive. and it would not be surprising that the local Hindus adopted so practical a device to render their large craft more sea- worthy. 35). D 3) rest of a boat is the drawing given by Lintscotus (II Pars. The Outriggers of Indonesian Canoes. unless the vertical is spar of Riedel's model from Tenimber inserted into as well as lashed to the float. The be by means of stapler. a unique record for this part of Indonesia. and large . difficulty as the larger size of the The number presents no necessitate this. and small fishing canoes with a double outrigger and direct.. XCIX. (2) it follows either that this method was invented in India. 84). A typical Halmaheran . but Professor S. the only instance I know of recorded from the whole group. but more probably float by sinnet lashings. If (1) the curved booms be an enlarged Halmaheran attachment. Pfliigf r (p. and the are the direct lashed. As previously noted.D. 4). but a true Halmaheran attachment does occur there (Fig. Kei Islands. J. the arrangement by which the float supported by the ends of the curved booms is unique. so. The Kei and Aru Islands. At Baba (Babber) there attachment I cannot Halmaheran ? and. by make out the something that looks but if it details. Islands. though the three central booms of a model from Gowa.

common is substratum with the Alfurs of the Moluccas and Minahassa. Friederici construction which do not concern us here. Archipelago for their skill in shipbuilding. have two In Banda the sides are often heightened by planks. p. 92. almost (2) Outrigger-canoes similar stem- and stern-posts. as seen from the side. All floats at the Some Banda of canoes have the high beak of the orembai. II. The natives are undoubtedly Papuans. An important element in the Kei Islands was formed by fugitives from Banda. have a stem-post. Van Hoe veil have little to do with those dialects. as they. but settlements from Indonesia have modified some The Moluccas. but in general reveal the Indonesian original type. 183-186. ethnography. end two booms of the thickness athwart the hull. like bahasa tanah. as has occurred in North-west Indonesia. : (1) the plank boats. Wallace gives a description of the construction of boats by the Kei Islanders and states that they are the best boat-builders in Indonesia (II. and mythology show that the natives of the Aru and Kei Islands have a great historical connection with Melanesia. thinks that the dialects of the Aru Islands appear to be closely akin to bahasa tanah. pp. 159). Friederici does not find it so. this latter persists among the fishing population and in form and construction resembles the mon of Buka in the Solomon Islands. 169- . 235-243. orembai. C. Ellat and Fehr in Great Kei (Fr. language. some extent. (1) the outrigger canoe. The Outriggers of Indonesian Canoes. three types of craft in Banda and Amboina (3) the dug-out. 1 The following account is largely taken from Friederici. Ill. it forms an economic between New Guinea and Indonesia. the float and boom are attached by pieces of ratan The Moluccan attachment of the thumb and lashed with spliced liana or thin ratan. but they have been so influenced by cultures coming in from the Archipelago that they may now appropriately be considered as The Kei Islanders are noted throughout the essentially Indonesian. and The beautiful large plank boats. on the other hand. pp. pp. in the (2) main. From an as they still are racially to ethnological point of view the Kei Islands were originally Papuan. 1 There are. 107 attachment occurs at Dobbo (Pulo Wamar). but in Amboina with a bamboo gunwale.. 43). Am Islands. which. 9). of the coastal groups. sometimes with additions. The orembai of Amboina and Banda have been modified by European influence. and HI. HADDON. gives further details of their consist of a dug-out with moderately pointed. as to a proved by the characteristics and language of the folk at Eli.A. and the dialects of the Kei Islands. The little western island of Dobbo has an certain extent annual market which link is frequented by traders from all parts . see also also pp. is broader and lower than the stern-post. which have a direct culture But. Doubtless the great effect development of plank-built boats in this region has had an inhibiting on outrigger craft. Langen gives an account of their boat-building and says that they supply boats to the natives of the Aru and Banda Islands (p.

is described on p. 83. 15). These poles and the booms are tied with lashing of ratan or arenga to ledges projecting are laid over the edge of the hull. paddles. 12. The Amboina terms are also much Malayized. 86). sapu and sapou. 5. with a direct tied attachment (p. common Polynesian baler. 89. (3) In Banda and Amboina there are also very rude. which are of light wood. to a beneath them and lashed with are pole placed arenga-string strength they between the sides of the canoe. instead of the customary two.108 A. talalo. or made of the sheath of a palm-leaf which has the form of the (II. Buru the sarua is used in Kajeli Bay in the east : 258) . side by The boom is usually quite end. Kolekole is the same word as korcik ra is (p. a vessel of the same construction as a large orembai. A direct lashed attachment occurs on Lake Wakollo in Buru. in general use. so Friederici adopts the Melanesian term of patnati for them. The The Outriggers of Indonesian Canoes. 358) the prau as seen on Lake Wakollo. Friederici 238) states that the technical nautical terms collected all by him at Banda are nearly Malay Corresponding terms in the language of the old inhabitants of Banda perhaps survive in a single village in the Kei Islands inhabited by descendants of the Banda folk. p. 67. slightly affected villages refers to three kinds of craft at Martin describes an orembai seen by him in Piru Bay. These craft are especially known in the history of the Moluccas and of Western fleets. in part trough-like. pis. of The old true Alfuran terms occur only in the remote and Amboina and the neighbouring islands. at Amboina aft of one side. 10. of Tifu (p. or Malayized. HADDON. dugouts called kolekole. Amboina and the neighbourhood are prau (prahu) and h&ka. from the keel and side planks through the eyes there is no special name in the Moluccas for these gear. 23. which are also called prau when a clumsy double outrigger is added. the double outrigger consists of three slightly curved booms. floats at Banda consist of bamboo. West Ceram (1894. 117). The Moluccan korra-korra but with a double outrigger. in which the lashing is passed holed ledges. all the The commonest names for outrigger-canoe in Banda. or two lengths of gabbagabba (the mid-rib of the leaf Both kinds are bent up in front like sledge-runners. of the sago palm). with a quadrangular sail set obliquely. talo. Buru. and (p. but in the country-speech of Amboina and the Uliassers are also found tola. near to the aft end of the the fore boom is further from its fore The sometimes for booms. the fakatora is the boat of the Galelaese who live with the Sulanese in different kampongs on the south coast west . and still among the primitive fishers of parts are sewn and tied together. New Guinea as the constituent part of the notorious Hongi is In Banda and Amboina a tripod-mast. 4. and other . is unusual. 329). The first and best description and illustrations of these was given by Forrest (pp. C. The baler is the concave portion of a large sea-shell. Sometimes Y -shaped wooden forks are lashed to the booms to carry fish-spears. . dial. float. 356). Originally. Banda and Ambon. p. he describes the rig and method of sailing (p.

427) states that the fauna is Papuan and that the inhabitants of the interior are true Papuans. HADDON. on Amboina. are physically distinct from the Malay and Malayized population of Mongoloid affinity now occupying 3). p. The Outriggers of Indonesian Canoes. while at usually consists of one or On Amboina the attachment is almost Banda the 6 -type predominates." (1878. taller. of outrigger is precisely like that characteristic of the north-west area of New Guinea. Martin " are closer I have no doubt that the highlanders of Buru and Seran [Ceram] says. No. (III. New Guinea. The simple Moluccan attachment is confined to Indonesia. 79. 129). p. p. p. with fine yet strong limbs and joints. They are darker than the latter. 109 as the predominant attachment on Banda. more powerful. . Guillemard (1894. especially Laut and Nusa from mixed. pp. " and without the flat noses of the Malays. Four booms may occur but the type at Misol (G. a distinction met with all the larger islands of the South Seas. or exclusive . Following Friederici. strongly von Kosenberg has found in Ceram " name implies are bushmen. only persisting now in the interior of Ceram and Buru. the Uliassers. to Melanesians than to Malaysians " (1894. Ambon. 26). which Friederici believes was carried thither by his Alfuren migration. Misol. 1-4) of the modern abuses of that word. 288). I do not know whether this is generally the case. : " the greatest part of the East Indian Archipelago and the Philippines. The Alfurs of Ceram. or at all events for early inhabitants of the Moluccas. 119. but on the coast are a mixed Malayo-Papuan race. Moluccas and North-east Celebes are not somatically uniform and their languages over Eastern Indonesia. as in a large measure they appear . Amboina and the Uliassers on Ceram. Tagals and others (III. Friederici speaks of it as . 267) . The coast population of Buru. a term which merely indicates the ruder inland hill-people as contrasted with the more advanced coast-dwellers. C. Amblau has received immigrants is end. 161). and " He also points out that. II. sporadically diffused throughout the Northern Moluccas and as being found wherever the bahasa tanah is spoken always the O-type. The U -Moluccan attachment occurs here and there on Buru form of . At Amboina the float usually two lengths of gabbagabba (midrib of sago palm-leaf). but a crossed double variety occurs sporadically in the West Pacific (p.A. 150). Some still live on the coasts in various places as though the majority as their since In Buru and Amblau the population has undergone a considerable mixture at the Kajeh the arrival of the Europeans.. these AJfurs of the belong rather to a linguistic family. with the exceptions of most places on the north coast Obi. p. and at Obi it is more in evidence than the Moluccan attachment. He enters into a lengthy discussion (III. positive historically they Melanesian the to contributed have population of New Guinea and the islands further east" (III. Some form of Halmaheran attachment is known from Banda. Ethnoiogically they form a fairly uniform layer of an older evolutionary or colonization period than the layer of the coast people to form a and also unity. Buru. we may adopt the term Alfur as the historical name for the aborigines. pp. He adds and Buru. cf. the north coast of Ceram.

two booms and direct attachment fishing canoes with outrigger. 7) states that the so-called is split into very numerous dialects on different islands. " a Paradise and 2. the largest had a fighting-platform. Ceram Laut. Fugitives from Banda have mixed with the inhabitants of Eastern Ceram. bahasa tanah. 93 Fr. Amboina and the and till the arrival of the Portuguese formed an ethnological and somatological whole. 239). Perry." cymbe.110 A. with only dialectic differences. p. For further information on and for a suggestion as to its origin. it is Everywhere here the trade language is Molucca-Malay. The U-Moluccan attachment predominates at Batjan. though others had two booms usually with paddlers. tied) gondola. Ambonese language. Pis. Ill. of the old Of the dialects of Western Ceram that Huamual (Little Ceram) appears to have been the mother. II. a " attachment.. and on Lake Galela in North Halmahera. J. Western Ceram. times a second is tied beside the first. II. C. The Alfurs of Ceram comprise the Patasiwa in the west and the Patalima in the east. Huamual was one great garden. and Goram. carcolle. three booms with the usual Y-board attachment and longitudinal planks for paddlers. As I have previously . too it is much reliance should not be placed safe to 1 assume that the outrigger was double in this dual division on these early engravings. Language and tradition show that the Alfurs of Western Ceram and the original inhabitants of Uliassers were one tribe. among the Javanese). p. but the Alfurs of the interior have largely preserved their speech <Fr. with a few cases of the Halmaheran attachment the float is almost always of bamboo as in Banda. 46. . A direct lashed attachment occurs at Batjan. At Ternate the U -Moluccan attachment greatly predominates (Fr. or at least understood. . Friederici (III. pp. 9). in which case the paddlers sat on the float stated. p.000 inhabitants in which no human beings live (Kiedel. a kind of tongue of several others. Halmahera and Neighbouring Islands.. some. or Ancient Javanese. including that of Amboina sacred language (as is the kawi. over which was a longitudinal plank for . p. on earth." with 11. but I think all vessels we know it was in some. 92. and of Western Ceram. 8). p.. The Outriggers of Indonesian Canoes.. At the time when the Dutch first arrived. 9). The dialect of Eastern Ceram reaches also to G-oram. Ternate. ships with an outrigger of three booms and a Y-board attachment . 14. .vessels. Ill. The Halmaheran attachment occurs at Tidor. Ill. This condition 1 formerly obtained in the Uliassers and in Amboina. now it is an awful wilderness No trace remains of the original inhabitants of the Obi (Ombi) Islands (Fr.000 warriors. p. the Uliassers. two booms and direct attachment. with outrigger. 15. . a double outrigger. or a direct (? Y -boards. 17) there were at Ternate two-masted merchant war. HADDON. In 1599 (Nicolas. is manifestly somewhat more remote. it was the classical. thus the bahasa tanah of Amboina. see W..

bent upwards the forked mast and gear holders runner type is prevalent. In Weda Bay. p. 112. with two notches on floats at Tobelo. and to this is attached the mast Below. sail. 18) was the first to draw The upper ends of three attention to the merits of the very practical tripod mast. Boot von den Molukken " with a double outrigger of three booms. have a crutch grip. Paddles. Friederici noted that the Halmaheran attachment greatly predominated in the north of Halmahera at Tobelo and Galela.A. nor the date of the craft. and the booms lateral by arenga or by coconut-fibre string. and Kau are fashioned out of its free border in which the booms The bamboo poles slightly bent up The forks on in front. As a matter of fact. but it would not be safe to base any definite argument on this Weule (PI. which the mast. whereas the outer longitudinal spai projects much with arenga or ratan lashing and its ends are bound tightly to the ends of the falsewhile a third lashing passes from the crossing of the In Weda Bay and boom and outer longitudinal spar to the middle of the float. outrigger canoes. negative evidence all.7 ) figures a " Moluccan attachment. bamboo poles of this very light mast are fastened close together hinge-wise in such a way that only one pole projects beyond the others. but in some a false-boom is added extends from lies beneath the knee-bend of the attachments) merely . bamboo. and other gear are laid may be made of a natural branch. or of a branched stick inserted into the hollow of a piece of bamboo these are strongly lashed to tre booms. Friederici did not notice . but he does not give his authority. a carved piece of wood. p. into the hole in the feet of the two masts. The Outriggers of Indonesian Canoes. in of the carcolle type. the sledgerather high.. though occasionally the U -Moluccan attachment occurs. to the other. or of thin wooden rods shaped like the runners of a sledge. Patani. Licuala leaf or are cut out of a piece of bucket-shaped little basket made out of a . there is a Fig. but in Tobelo and Galela the stems of spikes bent at a right angle are fastened to cross-bars. None of the old . Galela. occasion demands (Fr. II. in Buli Patani the floats are usually of bamboo. 9. and the inwardly projecting spike is inserted Forrest (pp. 240). through these holes is passed a transverse rod which itself passes through and is supported by two short stanchions which are lashed to a cross-bar . A short central wash-strake is added to some rest. float. 241). books I have consulted indicate a Moluccan attachment. beneath the middle of which is a longitudinal board for paddlers. but cross-bars are placed beneath are lashed to the cross-bars feet of the triangle-mast them athwart the hull. canoes have two booms with a Halmaheran in this case the inner longitudinal attachment. one outrigger further boom beyond the booms. and Buli. HADDON. The two have their bases perforated. C. Wooden similar anchors are widely spar (which employed in Indonesia. as iron has not come into use there (loc. In Halmahera the small ledges on the inner side of the hull of the canoe are not eyed.. as everywhere else Balers at Halmahera and Ternate consist of a in the Moluccas. the middle foot is the longest and it can be tilted at any angle as rope. anchors weighted with stones are used universally in Tobelo and Galela. cit. Ill and probably Presumably the direct attachment was tied.

The triton-shell trumpet p. (6) an attachment which recalls the Halmaheran he saw no outrigger boats. however. sporadically. also occurs. of the sledge -runner type. II. But the bahasa tandh and the dialects of Minahassa have a so that originally the Alfurs of common linguistic substratum Ceram and its neighbourhood were akin to those of Minahassa" (Fr. The Outriggers of Indonesian Canoes. hull. these are left open further east. remote from the Malayo-Polynesian languages and apparently unconnected with any hitherto known Papuan languages. leptilepa. Ill. When mast and is The rectangular sail. dug-out without outriggers occurs (loc. 544. which. quoted by Fr. 16. 16 C). 11). 9). (a) with usually In Makassar three booms. often rough. " There are at least five dialects among the Alfurs of Minahassa.. more obtuse form of Halmaheran attachment is found and here and there the U-Moluccan attachment. at Ake-Selaka (Fig.112 A. Everything else is as in other places Halmahera (loc. which. A simple. The aboriginal inhabitants Celebes . and the Togian Islands) are further removed from Melanesian languages. p. The more westerly and southern languages of Celebes (those of Gorontalo. 235). Ethnologically the Alfurs of Halmahera form a great group with the Alfurs of the Moluccas and those of North-east Celebes. 10. of the Sulla Islands belong ethnologically to (Fr. cit. p. and Halmahera differed appreciably in appearance from other Indonesians. tit. 9.. but. but according to models in the museum at Weltevreden double outriggers occur. according to Kern. Celebes and Neighbouring Islands. occur at Buli.. 10). the U -Moluccan attachment and three bamboos bound together to form (Fig. The attap cabin in the middle of the craft had closed sides. in set obliquely. they are mostly crescentic. the edges of the dug-outs were heightened amidships by the addition of strips of gabbagabba. as in the (or wash-strake) added to the north of Halmahera. C. p. reach to Formosa. Boegi. Middle the coastal folk are a very mixed crowd There is a Halma- heran attachment. and in none he note features or skin-colour approaching those (III. Baree. Neither did he see a plank in Weda Bay and Patani. The tripod-mast is present. The floats are mostly but the bamboo form with a slightly upward curve in front gear forks occur. A shorter. Friederici points out that the peoples of Halmahera and neighbouring islands form a separate linguistic group. 242).. A). the float. Ternate. as usual.. is used. and in the main probably anthropologically of the places that of the so-called as well. The dialects of the Alfurs of Minahassa and of the Sangir and Talaut Islands belong to the great group of Philippine languages. Ill. He could not see that the natives of Tidor. pp. HADDON. he visited did " Aryan race " pp. Friederici (2) saw three kinds of craft in Buton : (1) boats with a double outrigger with two kinds of attachment a boat without outrigger. In Bugi also the lepailepa is found (Matthes. 243). p. .

." : The examples known to me (p. where the Sarasins illustrate one canoe with a float composed of four bamboos it also recurs at . separated by the whole length of the close in detail. The Halmaheran attachment. sakit or spirit of sickness (pp. may be placed at the aft end of the canoe (Guillemard. in both the fore boom is lashed directly to the but in the Javan area the indirect attachment whereas in the Celeban area There is. 1889. 84. while the aft boom A direct lashed attachment occurs at Sangir and a mixed lashed direct and Minia- straight vertical Halmaheran attachment. the Talaut (including Sangir).. 96. so (Fig. are described on pp. 106. 71) have a triangle mast. 100. eastern Javanese area " . p. show some resemblances. however. at Kema in Minahassa a mixed rod and Halmaheran apparently attachment in which the fore boom has a rod attachment which is has a typical Halmaheran attachment There was much going to and fro in Indonesia in former times. the model from Gowa just mentioned invariably being very exceptional and probably represents a Sulu craft and not a local one.A. Kema in the extreme north-east of the island. a model from the latter place with five booms and a mixed direct attachment is' referred to on pp. . for short. Sulu Archipelago. The Outriggers of Indonesian Canoes. In the Sulu archipelago small canoes may have but two booms and the outrigger 206 G. According to the illustrations of D'Urville (II. OD Matanna Lake. 113 In Celebes a direct lashed attachment has been recorded at Makassar and Gowa. Minahassa. 77. 114 it is found on the Malili River (at the north-east of the Gulf of Boni. are simple dug-outs with square boarded ends. and Tulur groups is characterised by the occurrence of a mixed direct lashed and Halmaheran attachment this may. 180. of the aft it. 217). large island of Celebes.. 98). HADDON. no surprise need be felt if these two regions. 76. The region embracing Minahassa. affinities of Miiller. whereas the central booms are downwardly curved d . 80. pi. but these are not inserted into the float. booms which are simply pegged on to the float . 3) and Pfliiger (pp. Fig. In these canoes the fore and aft Usually there are three or four booms (p. boom is inserted into the float. Banka. the ture sakit canoes from Nanusa Island (to the north-east of Celebes) have two or three . p. as these are merely ceremonial models some sakit canoes are without outriggers. 80). The lowered to the outlayers of sailing boats in the Gulf of Boni are referred to on p. they are employed in the exorcism of the and Hickson. C. 103) the canoes on Lake Tondano. be termed the " north-eastern Celeban area. The " praus in roadstead " of Makassar figured by Pfliiger (p. mode of attachment has no signifi- cance. the yard may be boom or the boom raised to the yard. The Celeban canoes almost have two outrigger booms. 23). it is lashed to being a typical Halmaheran attachment. XXV. 96. in various forms. Nos. at booms are always straight. further inland. predominates throughout Celebes. 95. however. 290). More than two booms are present. as we have seen draws attention to the " the mixed attachment in this area with that of the float. 1893. at Paloppo (on the west side of the head of the Gulf of Boni and at Libukang in the same gulf.

p. usually they are attached to the booms (when they may be called " " of Miiller) and frequently they bear. " the is inhabited chiefly by Island Bajaus or sea-gipsies. two booms. is unique. p. the arms of which in the sectional view (V. Mr. the forked ends of which are Guillemard states that the southern part of Sulu lashed to the floats (pp. 31) of one of two similar models of canoes used on the Pasig River. Manila the specimens were collected by the late Admiral Maclear and presented by Mrs. 30. Maclear to the Pitt. 244) says that the canoes of Sulu. a people quite distinct from " the Sulus. their ends (Fig. Zamboanga. more upwardly curved bars are generally fastened transversely across the " boomcanoe. HADDON. a crutch or forked stick to support gear . which are not more than and fragile. p. 2. Rivers Museum. In all cases there is a direct lashed attachment. . Judging from an illustration given by A. near each end Biigel spars. Henry Balfour has very kindly made drawings for me (Figs. the fore and aft straight into. while the middle three are lashed to it (p. They are fitted with a double outrigger which. 76. p. 378. 7). 383). Miiller (p. Oxford. and of a much lower type (1894. but in Wilkes' figure (V. booms are inserted into the float. and Cebu are " Biigel in the main two similar. the other end of the float is lashed to a thwart which projects slightly beyond the sides of the hull. The Outriggers of Indonesian Canoes. 113). which pirate craft Probably belonging to this region is the difference in form of the Sulu booms. where the straight booms are tied to. 81. however. 228). but this would be the exact opposite to the arrangement at Madura. each of the three booms supports a semicircular spar with forked ends. The models are furnished with an awning. and one has three strongly curved outrigger-booms on each side. 84). so far as I am aware. 333) all three booms are downwardly curved this is probably incorrect. an arrangement which. 84). The outlayers of Philippine boats have been noted on p. It is possible that the latter may throw some is light on the somewhat perplexing tempted to suggest that the outer straight Sulu booms were originally inserted into the float. 90). H. is provided with but one boom which is lashed to the float. A C) but are attached directly to the in number and are unusually thin booms [by a lashed attachment]. Vojnich figures three booms at Manila and on Lake Lanao in Mindanao (pp.114 A. Philippines. and a direct lashed attachment occurs in the Cagayan Group. The floats are not attached to the " (Fig. a canoe with a double outrigger. Savage Landor. the float (p. in the canoe figured by Wilkes a lashing. II. 332) are shown as connected with the boom by is An analogous Madura combination of straight and partially curved booms found in the district." they are the or of the Two boom. " " of Pritchett. C. and the curved ones inserted In a model from South Celebes. and need strengthening by the elas" [" boom-spars "] ticity of the concave " Biigel which are half the length of the booms. and the remark made by Folkard raises the question as to whether the structures of the Oxford models may not really be outlayers. Mindoro Sea (I.

their almost total absence is rather surprising. and is Philippines Malayan elements. I. p. Ill.A. p. but this absence. SAME AS FIG. the booms of which pass through the the solid hull and are inserted laterally into the floats. 30. PASIG RIVER. important. BALFOUR. (II. seems of a rough model McDougall give an illustration (copied from Ling Roth. MODEL. HADDON. 31. which is in the from removed far Leiden Museum. a district made of pith of a small canoe the Malanau country. 133). Malanau in the bayoh ceremony for the casting out (22 . portions of both Sumatra and Java. 11). p. Apparently a similar imitation canoe. and in North Borneo (Fr. as in the western and northern FIG. FIG. Fig. Basilan. BALFOUR. The Outriggers of Indonesian Canoes. jong. 144) with two outriggers. MANILA (SKETCH BY H. OXFORD). 115 best known languages of the southern " that of the so-called " tribes of Mindanao and Sulu. C. OXFORD).. As from Borneo outrigger canoes are such a feature in Indonesia. in the Sulu and Tawi-Tawi Archipelagoes. OF A CANOE WITH A DOUBLE OUTRIGGER AND ONE BOOM. 30 SEEN FROM ABOVE AND WITHOUT THE AWNING (SKETCH BY H. Hose and to be due to secondary causes. spoken Borneo. came from South-east Borneo. Palawan. Balabec. pirate It is a dialect or form of Bisaya with a strong infiltration of The most widely spread. and is by the Moros of Mindanao. The model. is employed by of diseases 84.

most of whom have recently become converted to Islam. p. SARAWAK (FROM SKETCHES BY* A. OF MALANAU AND MALAY CHILDREN. . The Outriggers of Indonesian Canoes. of Perak. a mixed group of Bornean peoples. They while state that the the coast regions of all of Malanau are Klemantans. [chd-rigma]. while the other has only a in both cases the booms are inserted into holes in the side of the float. pelempong. tena. thus the canoe and its name may be an importation. This was the only occasion FIO. Bali. we find in Sumba.. The outlayers described by Beeckman have been noted on p. Gomes also Hose and (pp. 249) McDougall (I. Middle and Eastern Flores : Sikka. them have been much influenced by Malays (II. TOY CANOES. Terms of outrigger canoes added by me are placed within [ ]. tenna. 129). Sumatra ballelang]. but he had not very much to do with " The Bajaus were great pirates in the old days the villages near the coast. . Confined to a restricted area of the Lesser Sunda Islands. drawn from memory. some Bajaus outrigger of temporarily fitted a small low dug-out. on which he had seen an outrigger in Borneo. Solor. and might have picked up the idea in any of the islands further east.n. of toy " schooners. . Jong. 144. A. pp. C. 283) he states. single boom .116 A. may be indigenous to the area as a whole. tenah . of Sarawak. 55. Evans. the names for boats without outriggers are placed within ( ). 132. sailed by Malanau and Malay children along the Sarawak coast. sampan in Malay. for informing me that when two travelling in the Tempassuk District of British North Borneo. and for ceremonial boats (I. 76. for straight him with a double booms which were lashed to bamboo floats. Mentawei [abak]. gobang. . 246-254). 49-51) p. of Sarawak. For a description of the usual type of craft in Borneo see Ling Eoth (II. Lawrence. The following terms 'for outrigger canoes are given by Friederici. LAWRENCE). would seem at one time Sumatra was supplied with boats from Borneo (II." jong. 32." The word gobang or is the Sulu guban (infra). 32). I am also indebted to Ivor H. " and Nieuenhuis. dyukun. [due. pp. HADDON. doe' and doai\. both have a single outrigger. " It f. 166). and Maumeri. E. kindly gave me two sketches (Fig. pp. E. He adds. Nicobars [jellore. 56. tena. one with two straight booms. NOTES ON THE NATIVE NAMES FOR CANOES AND OUTRIGGERS. Andamans Engano teneh [eloha].

hftka. . prau when outrigger added but called prau belan. Valentijn. etc. korakora. or otil. dawong . Also cf. alal. \ 17 Flores. and Luzon. 159) that the Moluccan korakora corresponds to the " " (Folkard. Mota (Banks Islands) kora San Cristoval (South Solomons). Batjan. Halmahera Otili : . It is more than doubtful. Tagal. p. [The Achinese have a sailing boat called kolay or The Malay prahu. throughout the Moluccas. Bicol. Tobelo. 244-246). [Ray points out that the term laka. In Saonek wari (Doreh) wa or wai all in New Guinea. Sulu. which so widely and Sorong. coracora. an places for craft with or without is outrigger. Uliassers are prau (prahn) and hakd but also tala. the Uliassere Sumbawa. whether the word korakora of the Moluccas really goes back to the Arabic qorqor and thence to the Portuguese carraca. yel\ is Buli. nyon. tala. 481)]. lepalepa. (here the lepalepa a dug-out without . who gives also gives the variants of these terms that occur in Melanesia. Malay Peninsula. Bisaya and . is derives it when he from the kura-kura. and therefore has no special canoe may be termed prahu semdn. the Malay djokong . Amboma and the (II. jong] Diadokii or diadukv the name outrigger-booms at Tobelo (Halmahera). Celebes. boat. due] Batjan. ora. adyong is Hoko. and it also extends to Flores. kolek (a small fishing boat) Amboina. djungkung : [Nicobar. . korakora. significance. deru . sapu (orembai). asserts that this equation it. . Philippine baranguay. but these do not concern us here. Ake Selaka. perahu. in Halmahera and Western New Guinea. but that it is also a true Malayocarraca . who treats the word as an indigenous one. for the . alal. Friede: in the . He are taken mainly from Friederici (II. wot. C . prahn semdn. talalo. pp. Galela.' Weda Bay and Patani. p. . Ternate. oti or dti. korakora Salaier. 159). talalo. korakora. prau. p. rungku. but never de Morga not only says expressly that it is an ancient and indigenous word among the Tagals of Mindoro. and Manukspread in the South Seas as to be universal. 106). [Borneo. carcoa.. . Javanese djukung. p. says Friederici (III. oti. which term is current in Makassar. orembai). In the oldest Portucertainly wrong Malay guese and Spanish accounts of the Moluccas we find caracora. : . guban.A. ptlXn. Polynesian word kolekole rici . any form eastwards . He adds (III. Amboina. ionco (p. sapu and sapou are to be found in the country speech talo. HADDON. . Marinduque. nyon daon in the Philippines Tontemboan. ship and appears to be used indiscriminately in a general term for canoe. The most customary names for an outrigger canoe in Banda. of the Philippines. The following notes full references. outrigger) lopi. Banda. for not merely accidental as there is other support and he has shown that most of the boat-names of Eastern Indonesia recur in is Melanesia. 237). Kolekole is clearly the same word as kor&kora. lepa or prahti lepalepa. The Outriggers of Indonesian Canoes. ngotirl. thus in Amboina an is outrigger The MM of Amboina is the Malayo-Polynesian wangka. of Ambon and (kolekole. lopie or lopi.rahu. The dyuktin It is of Bali does not occur in . prahu korakora kolekole. prau.

Later ofak. berok. lepa. from the Kei Islands.. lettej.. Moluccas. abak is the va'a or vaka of Polynesia (1899. Ternate. p. 160. Malay and . II. of terms for booms given by Friederici is insufficient to enable The Moluccan term seems to be the uramon or dram&n Several terms Banda. Tololiku. 53). The beautiful. talo. In Halmahera and the neighbouring islands we find 8ti or oti. Sava (between gobun. . does not come thence but probably of Malay origin. Malay wangkang. Sulla Islands of south-east point Ceram . C. p. which becomes bairtimm in Bat j an and brayunan in Bali. 235).. waag. Friederici also groups together loju. a junk . Tidor Isam. lopi prahu lepalepa (Fr. be equated kowa. Obviously connected with peldn of Buli in Halmahera are : bero. lupek. Friederici (III. With the term guban of the Sumba and Timor) goba in Malay. Bongu. HADDON. waga. p. oti. belo.. : Another interesting word from the point . 236). 116]. In Ambon and the Uliassers are dial. ngotlrl. which is apparently confined to Weda Bay (Halmahera). rembaja are : other variants in the country-speech of Amboina arobail. (p. 1918. f. bero. Roni. also that Malagasy and Tagal are more closely related than either of them to the language of the islands between" (Haddon. a vessel gives is : .118 A. 235. kolek in Malayo-Polynesian languages (Fr. Ill. also ordnbdi or oranbaik. . Keane points out that the Mentawei . An orembai with is outriggers is called korakora. lete'ie. p. Aru Islands. Tagal series] baranay. Christian (p. but their is name of orembai or orembai. is also a related word [among the Bajau of British North Borneo we find gobang.. lopie or lopi . it is possible that ySl. appears to him that the Dayak top. large plank-boats without outriggers of Ambon and Banda come . a canoe . . . a pirate craft. lopeh. Wetta and Tenimber . 245). tola. 159). a small ship. p. a word which Martin friend is found as " (or kolekole. pp. at Bima and Sangar in Sumbawa. p. prahu .w. 229) Philippine Islands. Tobelo (Halmahera). Rotti (south of Timor) 280) adds that the guban of the Sulu Islands occurs as Sulu Islands may . Kern equates the lotu.. in the Philippines Fiji balanay. arubai. Apparently restricted to Galela and Loda is the word deru. Ibu . 244 of 1920 f. talal. . p. north-west of Triton Bay. II. Timor . The Outriggers of Indonesian Canoes. velo-velo of [which belongs to this with Biduq. 86) says that the name orembai a contraction for orang baik ! " good " servant ") and comes from the era of the Hongi voyages of the The number us to arrive at any conclusions.. Gesir. is no doubt the Melanesian waka. 241) Philippines. lopek .n. II. leptilepa. Madole. p. belongs here. Amboina beri-beri : Namatote Island. for Amboina. . banka. and as arumbae in Gesir (Fr. biluq Tagal (Fr.n. etc. talalo . and it . berok. variants are it is also employed at Makassar Malay Peninsula. . Iloko. lapis. arumbai. Solor . a dug-out without an outrigger . prahti bel&n or belang. Tabaru the word : .] view of migrations to the Western Pacific is the lepalepa of the Bugi [originally a people of the Boni district of Celebes].. Ake-Selaka notili. " used in Madagascar for the canoe with a double outrigger. ofa. Salaier Islands. otili and otil Tobelo. lapes. Waioli ngootili. Waigiu. arubaillo. /. New Guinea .

[The number of names available so limited as to be inconclusive. to construct la banca (the Philippine canoe). or some variant the Malayo-Polynesian .sawa means the old Javanese sama. companion. ungerti hera is : . at the same time like. is so widespread it among peoples that the investigator should note whenever does not occur . indeed has been stated. whether of animals or be taken as a general principle in distributions. sah&mmang. Roni.] As Friederici points out (II. that the general term for outrigger-float is derived " also Sumba. with. Thus we have : : Banda and Ambon. one which the ntldyu-w'idyu of Ake Selaka extends as be expected to Ternate and reaches Amboina. so widely distributed in Melanesia and Polynesia [The Nicobar name is deia due. scdtk . contrary to Kern. but as Friederici points out. less advanced." " Tagal. of 119 are employed in Halmahera. . somdn . that the outrigger boat has been derived from " " samasama. samd and semdsemd Malay also variants as in Sangir. types are generally men. . which he has nothing to say. to accompany and companion : . semHn Batjan. " . samaco. at the same time. somln in Tobelo. . . pagu Halma- Buli. hamand Galela. and the kdter or kdtir of Bali.] is which Friederici gives the following names . C. If we should the area be an archipelago like Indonesia more primitive conditions on its fore expect to find outskirts or in the interior . Ake Selaka. and Ternate and Amboina are again linked together. Amboina. Mindanao." Bali.A." The only other terms Friederici notes for float are the 6am of Buli. sumd . Halmahera Weda Bay (and Patani) zonuln.. This appears very significant and North Celebes. together. Ternate. HADDON. Ake Selaka. : Bali. it 'For the float is in reality a miniature boat. a fitted-up boat " and " to prepare or embellish. s$k. pdgupdgu Tobelo. . an outrigger boom. like. from an ethnological point of view. concerning : katig. tuddtudd Ternate. for outrigger attachments . 246. t%te. a small counterpart of a boat. the terms an of Buli and sesd of Galela (both in Halmahera) there is eri. it is its variants . sama samasama . 235-243). katik when we recall the correspondence between . which is quite obviously and Bikol. He from " thinks. that is to say or undesirable locaht be found on the margins of an area or in the less accessible It may within the area." etc. similar the double-boat. that the more primitive. resembling. of it for the float of an outrigger. The Nicobar term is heneme. tdlntene. Concerning nothing to say. katig Lake Magindanao. where nadyi/n also occurs. samd . sama. certain craft of Bali and those of the area between the Philippines [The Nicobar name is hentaha. Batjan. Banda. The sedek of Bali may be connected with the seke of Ake Selaka.] GENERAL CONCLUSIONS. So also have we in Malay " " " " and in sama. the term sdmdn. pp. that need comparison. the ditiddktt of Tobelo seems to be the might origin of the for word kiato. The Outriggers of Indonesian Canoes. Tara (Baree) katigi Bisaya. like the servant together.

the first to utilise the general type of canoe. but the details he gives are very meagre. and Rivers have done great Graebner (1905. 1913) has attempted to map out certain cultural streams from Indonesia into the Pacific. I believe.120 A. and the less early Pre-Dravidians. HADDON. so the converse equally true. 1908) has presented a mass of technological and linguistic evidence cultures concerning canoes. and however. view Graebner. and eventually we may hope to discover what kinds of boats they employed. which. necessary A very brief summary be out of place. as well as the presence or absence of outriggers and most obvious features of their attachments. as evidence for the differentiation of various and culture-strata in Oceania. My main object in this paper is to prepare the way for more extended historical studies on these lines. at all events somewhat admixed with southern Mongoloids. Rivers (1914) has utilised similar evidence in his is of the greatest dissection of the cultural strata in Melanesia. that the relative age of ethnographical data in Indonesia will ultimately be elucidated. It is from a consideration of such samples in these Oceanic areas and a co-ordination of the linked cultural elements they contain. were the migrations from somewhere in Southern Possibly some of Asia of the dolichocephalic Indonesians. travellers tell us that as a rule the latter is of the larger islands. Finally. that a disentangling of the chronological sequence of the cultures undertaking. But for my immediate purpose I find minology to make a few remarks upon certain distributions of canoes in Oceania. Friederici. perhaps not There seems sufficient evidence to assume that much later than of the probable racial history of the area will the original occupation of a part or the whole of Indonesia by Negritos. which importance for the study of these problems. each always associated with definite characteristic cultural elements. has bearing on the present problem. and so many of this internal movements have taken is place. Cultures have been so crowded into Indonesia at various times. From different points of service in this direction. and history it paper for this purpose canoes in Oceania are beyond will be necessary to make a more thorough of the types of canoes and outriggers throughout the Pacific. already them were. a definite terit have to be generally adopted. It is recognised that migrations of variable extent take place everywhere. We know that similar migrations from Indonesia have taken place at intervals to various regions of the Pacific which transported samples of the then stages of culture. C. and many of the indeed the case. in their later migrations. 1909) was. furthermore. there being many varieties will which have to be accounted for. and later (III. The Outriggers of Indonesian Canoes. The problems the scope of this analysis of of the distribution . Papuans. . As a knowledge ethnography is of Indonesia is necessary for a study of that of Oceania. All movements between different islands in Indonesia and those thence to Oceania must have been accomplished by boats of some description or other. the types of outriggers and varieties of attachments. Friederici (II. a very difficult A valuable little attempt kind has been made by Perry (1918). and.

Apart from the East of African area. of the priority of the single or double outrigger earliest is of some im- So far as historical data go. it would follow that the single double outrigger was the primitive type. orou. 35). where he found a nomadic We know of Indian colonies in people. ne Outriggers of Indonesian Canoes. who crossed over from Menangkabau in Sumatra thence at the close of the thirteenth century they spread over the East Indian archipelago. though in this case there is no structure from which an outrigger could be naturally developed. Bali. an evolution which took place solely hypothesis be correct. HADDON. the That the canoe logs of was evolved from the central log have persisted as the in Indonesia. missionary Fa-Hien. The Javanese Babads tell of an Indian prince who came to Java about A. If two outermost which floats of first the a double outrigger. of the sixteenth century the Portuguese made settlements.. for the float of a single canoe. as for example at Mailu.D. 100).A. the record we have is that of the Indo-Javanese double outrigger ships of twelve centuries ago. but the second hypothesis would make the initial boat. But to this other long previously peoples had secured a foothold in Java and elsewhere. bark canoes. H. In double canoes one of them often is smaller than the other and in some places the smaller canoe bears the same name as the float. in British New Guinea. Sumatra and Java in the third century (Fritsch. visited Java from India in the Chinese infiltration The Chinese Buddhist fifth century A. p. Arabian traders voyaged to the East Indian archipelago long before the time of Muhammad. . I have already pp. ]_>i of brachycephalic southern Mongoloids who conveniently be termed Proto-Malays. some of whom may have been crossed previously with other stocks. 78). and where the term larima is used for the smaller element (2) of a double canoe.D. 1911. or Hsien. vaona. may not have begun after 220 B. But if we assume a dug-out to be the it is or has been used in practically every part of the world (rafts. and were continued for a long period. 21). double outriggers outside . Orang Malayu. but Islam changed the Arab trader into a teacher of the new doctrine. The Malay Peninsula was first occupied in the twelfth century A. and outrigger the original form. by the true Malays. and were followed later by other European peoples (Haddon. The question portance. referred to a later colonisation (p. there does not seem to be any reason why either form of outrigger should be the earlier. when South China was from the conquered aboriginal population and a seaboard acquired. At the beginning it is Two hypotheses are current concerning the origin of the outrigger (1) That derived from a double canoe.D. perhaps they have never ceased. Somewhat later perhaps came swarms may . to which I have already alluded (p. but commercial relations existed with Java and other islands in the fifth century A. On the whole they have dominated the true Indonesian but the bulk of the peoples of the population archipelago consists of various blends of these two stocks. From the first century of our era there were migrations from India. 78 or 120. of a float or raft. and skin boats are another story).. one of the canoes having degenerated into the float : of the outrigger. C.C.D.

In Torres Straits and the estuary of the Fly we have definite evidence that the single outrigger is ousting the double. apparently from Cape Direction S. from Batavia River in the Gulf of Carpentaria. Claremont type with a direct lashed attachment. 12 S. C. is Apparently both forms occurred at Easter single.122 A. and Gulf districts of British New Guinea. and Geelvink Waigiu and are associated with a numerous booms Bay . 14 They have been (lat. including the Sulu Islands. and formerly at the Marquesas. I that the double outrigger am came not geographically culturally remote and there by the West Pacific route. other neighbouring islands. S. West Papuo-Melanesian. we have seen (p. and provisionally we is may accept this supposition. HADDON.) (6) the . They are also to be found in North Queensland. Nicobars and reappears in Geelvink Bay. Miiller (1912. doubtfully at Samoa. Sangir Islands (occasionally).) (Haddon. in Misol. and that the double outrigger has in course of time been displaced by the Melanesian single outrigger and has stood its ground Island only in the island of Nissan. both being associated with an overcrossed stick attachment (p. temporarily in the Solomons with three booms lashed to a float consisting of a bundle of bamboos. and it seems probable that this has occurred elsewhere. 85). formerly in the Pelew Islands. Geelvink Bay. the latter The double outrigger float in order to more stable but clumsier than the has better sailing qualities. in the area They occur which includes Torres Straits and the estuary of the Fly River. 80) that the Philippine area. (a) the Batavia River type with an attachment of two vertical sticks. On the other hand. Indonesia are found with a rare and scattered distribution in Oceania. the Massini. I dated July 14. booms Very frequently associated with a single outrigger is the presence of several as in the Andamans. Although the Torres Straits area strongly of opinion that it is remote from Indonesia. 1913. A scattered marginal distribution of this kind suggests antiquity. though there is very little precise information on the subject. it is practically universal in New Guinea single outrigger is a marginal The and Oceania. 1913). Friederici says have now no doubt that the hop [the Nissan double outrigger canoe with a direct tied attachment] has been brought by a Philippine or sub-Philippine wandering : In a letter to me " stream to New Ireland and neighbourhood. but great care has to be exercised in the management of the maintain the balance of the boat. 124). p. 621). phenomenon. and also formerly at Easter Island with a direct lashed attachment (Haddon. but single outriggers also now occur there. 245) alludes to a change of double outrigger into a single as having occurred in the Pelew Islands. p. 78). 1913. The Outriggers of Indonesian Canoes." (p. Within Indonesia it occurs It is normal in the Andamans and and very rarely sporadically (p. but to a much less extent in Polynesia.) to Claremont Point (lat. they occur at Nissan with a direct lashed attachment. 13 reported in Oceania. and most parts of Melanesia. and doubtfully in the Seniavina Group of the Carolines . round Cape York to about Cape Grenville (lat.

Fig. Indonesia. II. On plank of the raft had become converted a attachment stick the other hand. p. all these records. direct inserted attachment has a less extended range than the lashed. p.. outriggers having The diffusion of canoes with double outriggers and two booms Irom Indonesia into Oceania must have taken place in very early times and (p. 194. which. however. and we know there is to this belong class. The in the p. 249 end the branch wardly slanting New Hebrides (Hedley. Figs. excluding distribution.A. while in the to be inserted in the of Atiu and Mitiaro the end of the booms is curved downwardly attachment also occurs in 127-130). I . possibly on two occasions. is therefore characterised by canoea two double with booms. one with a direct tied attachment ment. I With the exception of the somewhat aberrant Balinese attachment have not come across an unequivocal case in Indonesia of an actual canoe where all 1. three We must not. . only Melanesian record is from Eromanga a downhave booms the that Samoan the in straight Tutuila at Kramer found group of which is inserted into the float (1903. Inserted. most prevalent in the Southern Polynesian area. 1911. p. if 124) and the other with an inserted stick attach. 80). as I suspect its eastern margin. as this is practically confined to the eastern margin of Indonesia. as the Tutuil I feel some slight hesitation in accepting it looks as if it should ment is very like that which occurs at Funafuti. At present I leave it an open question whether these are relics of a more general them to be. p. I now pass to a consideration of the distribution of the main types of attachments. II. really relations between the Samoans and Ellice Islanders. have been told it occurs at Rotuma Friederici more northerly Cook Islands also records it for Mangaia and Rimatara. 415 . though have I a lashed attachment. or as borrowings from the eastern margin. the aft with a stick attachment). C. the fore with a the Paumotus (Pallander. appears to be characteristic arise until the central of various types of primitive outrigger canoes. Direct. figures a canoe with It thus appears to be crooked inserted boom. A. 287). 23a). 123 double outrigger. the association of several booms with double outriggers. the booms The are so attached. two booms. overlook the occasional occurrence of booms within the central area of Indonesia (p. and so is. 42. The Outriggers of Indonesian Canoes. raft. The direct inserted 1906. not in a position to suggest which is the older type but the float be derived from the smaller element of a double canoe an indirect I float At present am attachment would not become necessary until the had been reduced to a log or On the hypothesis of the evolution of the outrigger canoe from a a piece of bamboo. float (Fr. as the need of an indirect attachment would not into a canoe. p. the tied attachment would be the more primitive. Thus so far as Indonesia is concerned several booms to a single outrigger is a marginal characteristic. but to a less extent. HADDON. 314.

But it is also significant that it is very prevalent in the Sulu Islands and in the southern and central Philippines. in a model Alexander. 1913. a single outrigger of two booms kee). the hop of Nissan in the North Solomons (Krause (Fig.. 240) with a double outrigger (Choris. 6. Salem Museum ? Pallander.124 A. . in the eastern margin of our area. Haddon. i. 2. FUNAFUTI. HADDON. very rarely crops up in Western Oceania. 5) a canoe from Ansus. a short distance . PL 29) Leeward Group of the Society Islands and Huaheine. 14) and North-east Queensland from 13-14 It occurs S. an ancient form. 1. loc. Nonuti in the Gilberts (Fig. sometimes it is lashed to short pegs on both sides of the (Hedley. after Roth). cf.e. p. upon and is tied to the booms is . with a direct lashed but as all the Figs. Paumotus (Wilkes. lat. cit. In 1914 I saw at BunTki. Fig. ? Napuka (Wytoo. Fig. . unless It is interesting to note that it crops that it is really rivers. (Ellis. is The Outriggers of Indonesian Canoes. 21 the easternmost of the p. where it is associated with a double Friee. attachment. 101) outrigger It . derici. ELLICE ISLANDS (AFTER HEDLEY). This type very widely but sparsely distributed throughout Indoit signifies nesia. 1831. The direct lashed attachment (with a single outrigger) alone occurs in south type is India and Ceylon. . . with a single outrigger at at Easter Island . 796. and in the . Tahiti (Wilkes. p. 7 in the . 29) and universally . pi. Haddon 1913. PL. 352). 1. boom PL XV. Marquesas (model. Fig. (Claremont type. or may equally well have traversed in the reverse direction. PL X. p. ATTACHMENTS OF CANOES. in the Hawaiian Islands. C. p. but the sketch not convincing) . so much so that little can be deduced therefrom. 33) there are three straight booms FIG. I. Edge-Partington. where its up in various lakes and appearance may mean for the needs of that calmer waters this it suffices an old type. Fig. cf. p. Lashed. Ethnol. whence it may have spread into Indonesia. the tip of this branch rests on the which it is lashed. pi. 1798. 35) the pegs are inserted into the Cambridge Museum there is only one peg float. p. II.g.. 33. Album. 319. De Clercq and Schmeltz describe (p. New Guinea. 94) and figure (PL XXIV.. II. 7. At Funafuti in the Ellice Islands (Fig. " spike accounts we have from Geelvink Bay speak is solely of a " attachment we may suspect that the model inaccurate. float to which have a long oblique branch near the end. the float rests I. 61). is shown. 621) but in the " " Atlas du Voyage de la Perouse (London. 286. or simply that in these the fisher folk.

i. are attached to the float of the by two sticks . a smaU on the east side mouth of the Bamu) in the example of a canoe with a single outrigger of two booms with a direct tied attach- mentit may have been an individual occurrence of no special significance. it a flying proa taken at the Ladrone Islands. The Outriggers of Indonesian Canoes. p. all the boats seen Friederici by at Tahiti and the North and South Tuamotus the aft boom had a direct (Paumotus) inserted attachment. p. Attachment Inserted into the Float. which apparently is inserted into the float between its a similar spar on each side of the two straight booms occurred on a model from Kusaie in the Salem Museum (but one of these apic has entirely disappeared. Fig. Stick. sticks. A mixed direct inserted and stick attachment. II. 132) . . 493) gives a figure of a Guam canoe with two booms each with one . there are two longitudinal spars and two oblique stay spars Safford (PI. ment. 123. Y . p. Figs. XI. 131. The only places Andamans and here the sticks vary in number and arrangeNicobars. in which the angular (? fore) boom is inserted into the float and the straight (? aft) boom is supported by two pairs of undercrossed sticks. PI. where it becomes free from the fastened a curved spar. C. 194) figures a canoe at Anaa. 314. p. but the booms have a another Kusaie inserted and stick-. B. and direct inserted and Y-stick-attachments) and a model (Salem Museum) has two pairs of undercrossed sticks for each boom forwards beneath curved spar which is lashed to the boom and comes downwards and the normal attachment and ends . 20 LIV. LIII. p. 463) of has three booms (the central one may or may not be a false boom) with the same and attachment. to each of which. the latter hull. while the other is represented only by its distal end which how they were fastened to the hull is not shown in the is inserted into the float -stick attachment (these are thus a mixed direct photograph). HADDON. pair of undercrossed sticks and apparently a vertical lashing between them. river of the Bebea River (the northerly Western Division of British New Guinea. oblique or if in pairs overcrossed or undercrossed.e. Fig. In. (a) 1. a solitary up the Dabara arumo.. 5) a paopao canoe from Nukuor and a war canoe from Ponape in the Carolines with a similar attachment of two oblique . is has two stay spars. Indirect Attachment. they may be vertical. In a copy of an old engraving (Mager. Paumotus." sketched by Folkard. Figs.A. p. as the canoe was a very small one another canoe had the stick attachment characteristic of that district. Kubary figures (Pis. Pallander (PL. but the fore boom had sticks (Fr. on the same side both booms in a canoe Mariana or Ladrone Islands " Anson gives a most and excellent account and engraving (reproduced by Lang Roth. . 137) two double booms.. 118. in Indonesia where this occurs are the . apic. 3. p.. one above the marginal within the area other (boom and of boom spar).

6) or two on each side (model boom (Macmillan Brown. 81. 98 . A simple stick attachment characterizes the tsine canoe of Nissan. II. p. and the three northerly islands Nuguria (Thilenius. Friendly latter and the Gilberts and between the Islands (Tongan group).. a similar arrangement. 106-108). where it is associated with a There is considerable variation in the stick attachments of Northern and Southern Melanesia.. The stick attachment may certainly be regarded as an ancient type. Fig. 105. Its New Guinea. Fig. Fig. or the taller Oceanic Negroids . Fig. copied Wilkes' drawing (V. a model in the Cambridge Museum from this group has three vertical sticks. p. which is remarkably like the Nissan tsine. arise from the on either side of the booms and a mixed direct lashed 7) indicates and stick attachment) see pp. Wolea a " are boom. 1913. II. 17 S. one on each side (Kramer 1906. of two pairs of diverging overcrossed sticks arrangement double outrigger. Samoan group. and Nukumanu (Fr. or two sticks on each side. where d'Urville (I. which extends along the north-east coast QueensS. p. two sticks. PL LXI. Fr. New Guinea is also characterized by various attachment it. an island between New Ireland and Bougainville (Fr. 2 . presence in the Andamans. 6) Solomons : PL III. 499). vertical sticks on the outer side of each " slants up from the sticks to the bracing spar .. of the scattered chain to the east of the Haddon. and a two-stick attachment occurs at Batavia River. whence probably of came the Cape Bedford wangga.. 617). Kubary (PI. 103). p. II. in the two former three curved booms which are lashed to the float (i. PL F. canoe from Drummond (Tapiteuea). and another model has one pair of undercrossed sticks to each of the three booms. is Figs. 61. 49) of a I. canoe with four curved booms and an attachment of four the Fiji group the sticks consist precisely this is Fig. except that there is a double outrigger. as usual. 1902. further allude to of Geelvink preparing a memoir on this subject I need not except to say that the undercrossed type extends from the east . Gilbert or Kingsmill Islands. Union or Tokelau group. two Mayer. Fig. 2) illustrates a double In . tall vertical sticks.. HADDON. which lack of space precludes me from describing. Tauu (Fr. The furthest east that I know of it is from Nikuhiva (Marquesas). and Nieue.e. Folkard (p. 129). Thus we meet with it in Rotuma. The most common type consists of undercrossed types of stick sticks.) land from the Flinders Group in the south of Princess Charlotte Bay (lat. 300). typical of Torres Straits. LIV. and Melanesia at first sight might suggest that it was primitively associated with the pygmy. so it is not surprising that a similar form of attachment is common to them all.. There intercourse between these islands and the Polynesian Islands to the south. North-west Queensland. and photograph Amsterdam Museum) hull . 1913. II. p.126 A. 416. Figs. C. at The Outriggers of Indonesian Canoes. p. as I am Bay to the Massim District in the extreme south-east. The typical Marshall Islands canoe has two straight booms with one stick (model Salem Museum). 14 to Cape Grafton (lat. 95.) (Haddon. with three stick converging booms and one oblique on each side of each boom . the crossings of the sticks .

The general prevalence of the stick attach- ment not only in New Guinea and Melanesia. HADDON. for occurs may supported It is lashed and occurs in the Gilberts and at Nauru (Fig. 251) says that this attachment is Moluccan or Halmaheran. . 767). but also in the South Polynesian area. 34). therefore. type of stick attachment 2. 766. 127 who Papuans. stick or spike which is driven through the into the float usually a thin the boom. Its distribution.-S. Among the Negritos only the Andamaneae and the arrangement of their stick attachment points to a borrowing of possess the outrigger from the Nicobars. Indonesia. combined bowed the top of the end of the boom to which it is lashed. lat. peoples possess an outrigger of for the float. though not recorded but it appears Pacific and is characteristic of Southern Micronesia. The rod attachment appears to be confined to Indonesia (p. be boom 4. booms in the Andamans. hulls of the canoes in the A with a spike attachment. On the western border of Indonesia the stick attachment ie associated with a single outrigger of several generally. it consists of a thin . and roughly N. 3. pp. In New Guinea and Melanesia the evidence U it. the spar which passes along Paumotu Group other end being lashed to the float. from Liueniua and the Equator. The ^-shaped stick attachment. is associated with a single outrigger of two booms. occurs at Nukutavaka. and several migrations of Melanesian peoples have passed into New Guinea carrying with them special types of canoes and donesia into outriggers and their distinctive names. is due to an early supports the conclusion that its introduction into these regions Indonesian influence. A double overcrossed Y-stick what resembles the each i U in the Loyalty Islands and New Caledonia. 87). C. (Alexander. The spike attachment is somewhat similar to the foregoing and is I believe peculiar to Geelvink characteristic of and Bay in New Guinea boom and (p.E. and that the natives hang up the corridor of the turtle-roofed houses.W.A. and especially We Guinea are in the great majority of cases of Austronesian origin know that during long periods of time many migrations have spread from In- New New Guinea and the Western Pacific. and the names for canoes. are usually termed strongly against it being part of the old Papuan culture. 79) . The Outriggers of Indonesian Canoes. in Very few true Papuan any kind. p. as it is in Melanesia Although the Andamanese are certainly among the most primitive oi all existing peoples. the earlier of which gave rise to that mixed folk whom we term Melanesians. there are indications that they have borrowed certain elements In the Nicobars the same of a higher culture. Carolines the in flank eastern and Sikaiana on the the side of a boom in Liueniua (Ontong Java) recorded by Friederici the stem of the Y the Solomons. branch extends at right angles from the spike which is lashed on to to taken more pieces than the easily Friederici (II. In the Liueniua canoe in addition. of which the outrigger canoe is one. so that superficial!: immersed in the float and there is a lashing -Moluccan attachment. widely in the West and Marshall groups. In the Pelew Islands to be absent from the Mariana (Ladrone) it is prevalent in the fork of one or two of these sticks. n 10 rouj lghly between N.

and at the same time supplied a certain amount of elasticity. but with a stick or block of wood instead of a board (Haddon. is 5. A precisely similar arrangement is found on the coast of East Africa. unrecorded elsewhere Haddon." and rolled over. 34. and so far as I know. These pieces are seized to the top side of the float [and not inserted into it as in a true board attachment] The crosspieces with are braced withes. in this the boom passes through a board which inserted into the float (p. the lower end of which " is fastened to the According to his description. the float was a more secure method of fastening. unique. feature. Alexander (p. 1918. 745) figures a canoe with a single outrigger. HAMBRUCH. HADDON. the distance from the gunwale to the waterline. of which passes through a board. The board attachment is typical of Indonesia . The advantage are beached is of detaching the outrigger float from the hull occurs when the canoes able. In hull. Here it is associated with a double outrigger of two booms. 87). A somewhat similar arrangement occurs at Wukuhiva.128 A. present. Lashed indirect attachments seem to have ousted the inserted stick attachments throughout the greater part of Indonesia. . as Hornell suggests. the Santa Cruz Islands the whole outrigger apparatus may be detachable from the (6) Attachment Tied tq the Float. the separated hull being more manageis The Balinese really a direct inserted attachment. . 6. the ends of the booms is. perhaps. FIG. but the occasional unshipping of the ends of the three booms which permanently attached to the float is a new. are seized to the perpendicular pieces. each of the two booms float. 252). but an in" Hedley says (p. instead of simply inserting it into. C. No. but are sometimes made divisible. The East African outrigger canoe is universally recognised as having been derived from Indonesia. is and Sikaiana to (cf. the length of which as a rule. NAURU. The Balinese attachment it is teresting analogy to occurs in Funafuti confined to the Eastern Javan area." A is and stanchions curved brace [booms]. usually entire. probably because lashing the attachment to. 1918. . The inserted stick attachment is thus confined to marginal areas. MICRONESIA. fig. 29). or stanchions. HORNIMAN MUSEUM (cf. the Funafuti a direct are lashed one. No. ATTACHMENT OF A CANOE. before indirect attachments were lashed to the float in the latter region. New Caledonia and the Loyalty Islands. Marquesas Islands. spliced in a lock-joint [scarfed] and served. The Outriggers of Indonesian Canoes. 1915. float. but 68). . ekuo. 286) that the booms are .

now obsolete in the Tongan Group. for if this from Indonesia to Oceania must have taken place long attachment was suitable for ocean-going ships it would The there. at Witu (French and among the Nakanai of the north coast of New Britain close to the Gazelle Peninsula. but with outriggers In Indonesia sailing craft and so it was found that the outrigger was unnecessary for large of the was discarded in the orembai. at San Cristoval in the Solomons. I erroneously stated on pp. can be no under foreign influence.A. and thus the knowledge of an effective tied attachment should have been known throughout the area. There first is retained the double outrigger. The persistence and it of the double outrigger in Indonesia requires some explanation. Mayer gives a photograph of it at Vavau and (1916. which is sufficient for the general purposes of probably doubt that these at Friederici says this (II. with its single outrigger and direct lashed attachment. 1918. p. For more extensive trading voyages boats built up of planks were constructed. adaptable to vessels of large as well as of small size. 68). which looks as if it indicated a definite probably late cultural drift. 600. 1918. 2. If these ships had a Halmaheran attachment we could assert that the great voyages before A.D. as A. The Halmaheran attachment It varies it is confined to Indonesia and the north-west of New perhaps because widely spread within the area. The Outriggers of Indonesian Canoes. 1. I was misinformed that this attachment is of the western end of New New Britain. but in the case of the waga single. Perhaps the large waga it Massim district of British New Guinea were derived from vessels of this type before is the outrigger the outrigger was discarded. 25). C.) the Indo- Javanese ships traded between Java and India we may assume that they also traded in Indonesia. The Molwcan attachment has a restricted distribution within the area (in it Guinea. G. is 1 j'. is. 117. the inference being that this method developed in Indonesia after into the all the great migrations had taken place. surely have been taken into the Pacific and have persisted somewhere entire absence of double canoes from Indonesia and their occurrence in Oceania one which was contem suggests that these craft belong to an old culture stratum and with the inserted stick attachment. considerably in form and is my paper in Man. and in the Tongan Islands (Haddon. 3. and Kilenge on the north coast Islands) north of Britain. HADDON. at all events these are at present asso- poraneous ciated together in Oceania and New Guinea. a craft of the construction of a large orembai. With the exception of the last. Kobe. it is significant that no indirect attachment tied to the float has been reported from Oceania. The Hawaiian type. p. an essentially fishing community. may belong to a special migration. seems to be due to the fact that when provided with an outrigger the dug-out canoe is a very handy. as the korra-korra still do. 237). of relatively recent date. No. occurred in the The doubk or crossed \J-Moluccan attachment occurs among the Barriai. . of the Moluccan orembai type. 119 that Sulu archipelago). (The introduction of the If mon Solomon Islands however. light and stable craft.

and the art of building sea-going plank boats. constructed in a similar to the orembai. is of Perak. lashed to the sides ofprahus. C. and we may therefore confidently attribute the invention to them. Man. informs me that his Malay assistant told him that a simple type of double outrigger occasionally fitted temporarily to Malay boats. The early Arab navigators and others who may have come from the Bed Sea or Persian Gulf were ignorant of this contrivance. though it is equally probable that Alfur has been applied (p. We As have now to attempt to determine where the outrigger was invented.130 A. Probably later immigrants adopted the outrigger. 238). indeed.) type of built-up canoe of the Solomon Islands. Friederici points out that the mon (mona. loids (Proto-Malays) earlier brought immigrants invented the outrigger after they had established themselves in the islands. 160) says that the mon is not a good sea-going craft (and he says the same is true of the orembai) it is true the Solomon it is . seen large tree trunks. and it does not invalidate the generalization here made. In certain large areas of Indonesia outrigger canoes are extremely rare or even absent altogether. this craft highly improbable that the Southern Mongowith them. Friederici (III. something like outriggers. HADDON. me. The Outriggers of Indonesian Canoes. generally to theprahii sagor. when heavily laden." Mr. p. rather than to the Mongoloid. II. Evans. in bringing them by sea. 118). morel. H. element in the Alfurs. an art which without doubt was introduced 1 Mr. Tasman compared the plank-built mon of South New " " of Ternate (Fr. whose wood was of greater specific gravity than water. but the Orang Malayu always seem to have preferred the outrigger-less craft to which they were accustomed. Islands have been developed on special lines. and I may here add to these the unThis leaves us with the Indonesians and that early admixture of Indonesians with the first swarm of Proto-Malays to which the term progressive Pre-Dravidians. is etc. 109). as were also the Chinese navigators. mold. I have already pointed out (p. p. which is without an outrigger. but they are very careful to choose good weather. term guban Sulu archipelago name for a canoe with a double outrigger) but in the is West Pacific the double outrigger retained only in the hop (a variant of guban) of Nissan. Thus of East Indian archipelago.. and Islanders make long voyages in them. Evans has " . a small dug-out. 126) that it is very improbable that this invention was due to Negritos or Papuans. XVIII. A is the. the only possible introducers it is doubtful whether they originally employed it. Indeed it may very well be that this device was due to the inventiveness of the Indonesian. corre-corre The mon of the Solomon Ireland to the manner open to discussion whether they were derived from an outriggerless plank boat or whether the outriggers were discarded in these islands. migration and not to the Philippine migration criterion of the Philippine migration . p. The main reason for this absence seems to be a knowledge of of all the voyagers to the an outrigger were Indians. (a He is inclined to attribute its introduction to the second branch of his Alfuran (cf. it is except in the Southern Indian area. as the idea may have been borrowed from the islands. 1 outrigger canoes are at the present time absent from the mainland of Asia. 1918. This is the only record known to. I.

and thus some them may have passed over is and have reached a far destination. stick is or direct lashed also characteristic of the most central area. tl K-NotT treats. A. parts of New Guinea. Melanesia. II. double outrigger with lashed attachment. Andamans. Single outrigger with two booms and stick attachments: north coast of British New Guinea and South Polynesian area. direct Within the area : two booms and a Widely distributed. 71 suggests that the present focus of outrigger is it is legitimate to suppose that from Indonesia. C. if not actually from the Moluccas. double outrigger with two booms and a Moluccan attachment. 0* Commit^ Part xxvii.A. of the to be a probability that Moluccan the most recent.-De Barte-sprekende Toradja* ete. p. and Chinese have contributed degrees at different times . for 1901. Platen and Kruyt. BIBLIOGRAPHY. Adrian! N en van Midden-Cdebe*. migrations took place at various times. Report of tne South Sea 1902. with two booms and a direct lashed attachment Single outrigger : South Indian area and North Polynesian area. HADDON. C. and those that started last would have a more limited distribution but we . must also remember that the later swarms would be more of civilised and have a better earlier layers technical equipment. Indians.. 1S1 from without. a ancient than the single. of Series. An An canoes inspection of the in the Moluccas. despite its persistence outrigger is more If we apply i which does not fit in with the foregoing argument. map on and p. The general distribution of the : main types as follows : Marginal Double outrigger with two booms and a but the double outrigger attachment . As a general rule one might special type of expect to find that the earlier types of canoes or of outriggers were those that went furthest. The Outriggers of Indonesian Canoes. and to which Arabs. Most restricted. each with ite canoe or with some partial modification. a HalmaLess distributed. There also appears in Indonesia. double outrigger with two booms and heran attachment. . : and stick attachments NicoSingle outrigger with several booms bars. 161). in varying primitively this art appears to have come from Ancient analogous change has taken place in the Solomon Islands owing to a cultural drift from the Moluccas (Fr. of Fiah and 1 . section we are led the principles enunciated at the beginning of this an historical sequence in whicl to conclude that the above order roughly represents oldest and indirect attachments. the stick is the it seems evident that. Washington. Egypt. I^nd^S *.

etc. T. VII I. IV. B. 68. Roy. W. India Orientalis. Nos. Reichsmuseums. H. de. J.. III. 2nd Leipzig. T. Ethnologie. von. Paris. Schutzgeb. Baessler. Standford's Compendium of Geography." G.S. W. Eastern Pacific Lands : Tahiti and the Marquesas Islands. 1894. Hickson. S. Giglioli. Voyage aux Moluques et ala Nouvelle Paris.. Proc. A. 1899 Revised Edition. A. Guinee. Mikronesien Bd. J. Mitteil. Cambridge." Ibid. A. 1897. . 1868. 1920. 29. 1780. XCI. Man. I. 1601. 1834. H. W est-noordkusl Dumont Edye. 1891. S. H. F. A Naturalist in the Celebes.).. 1834. 1 Nauru 2 Halbband. (Quinta Pars. T. Besuch bei den Kannibalen Sumatras. C. Clercq." 1913. XI. Ethnogr. Unter Kopfjagern in Central-Celebes. 1748. J. O. Untersuchungen iiber eine melanesische Wanderstrasse. A 1718... Journ. C. Leiden. H.. S. Erganzungsheft Nr. the A Naturalist's Wanderings in the Eastern Archipelago. Sydney. VI. 1907. Grubauer. A. Keane ? A. III). E. W. Graafland. F. and North Queensland. P. H. 1911 . d'Urville. 1913. Edit. W. Asiat. and McDougall. Arch.. The Sailing Boat. (referred to as Fr. 1893.. H.. XXXVII. Seventeen Years among the Sea Dyaks of Borneo. 1911. 1775 and 1776. Leiden. Juynboll. 1915. Ethnogr. Hose. " III.Man. Australian Museum.The Pagan Tribes of Borneo. XIX. Bry. Indies Orientalis. A. de. E. De Minahassa. IV. 1912. Atoll of Funaf^lti. 1898. 5th Edit. 1889. J.. Soc. 1910. Folkard. Wurzburg. 1916. Fritsch. Reichsmuseums. HADDON. Graebner. Macmillan. X. The Gardens of the Sun. Antiqu. aus den Deutsch. J. 5. N.Zeitschr. 1908-1910. Indian Sculpture and Painting. The Hambruch. 2 Vols. -heft. G. Katalog Ethnogr." Essays and Studies canoes of Torres Straits outrigger to William presented Ridgeway. C. London. J. Ethnographische Beschrijving van de Leiden. Anthropos. No. Guillemard. van Nederlandsch Nieuw-Guinea. to A Voyage New Guinea and Moluccas. W. 1885. II. H. Christian. Photographs "Marchesa.. Arch. 1914. " Australasia. 1908. E. Pars V. The Cruise II.The Wanderings of Peoples. D. Sudsee Exped. H. 7 Globus..und Sprachenkunde von Deutsch-Neuguinea.. H. J. fur Ethnogr. 2nd Edit.Sumatra (Sumatra.). 1893. XVIII. 55. Brenner. 1916. Soc. London. 1891 (1893). . 1889. " Beitrage zur Volker. 43." 1881-1884. of the Marchesa. F. Brown. A Voyage Round World. ~F. Guillemard taken when on the cruise of the Cambridge. Voyage pittoresque autour du Monde. F. Voyage to and from the Island of Borneo. Katalog Ethnogr. Hedley. Havell. D. 1905. No. Camb. G. Wissenschaftliche Ergebnisse einer amtlichen Forschungsreise nach dem Bismarck. 1918. during the years 1774. Haddon. Internat. H. Man.)... Forbes. Gomes. Burbidge. 1919. the The Outriggers of Indonesian Canoes. The Memoir (N. S. 2nd Edit. Bickmore. and Schmeltz. Erg. F. f. 1920. Ellis. Friederici. B. " in the possession of Dr. 1894. Int. Beeckman. G. Hornell. C. 1912. Mittel.. Ergebnisse d. f.Archipel im Jahre 1908. Polynesian Researches. XIV. Fischer. Forrest. Hamburg. Haarlem. Travels in the East Indian Archipelago. 1779. 1831. G. 1880. J.) Francofurti. Cambridge. H. 1909.132 Anson. C. Nr. 1913. Java (II. II.. Man Pist and Present. 1901.

u. T. VII.Jahrbuch d. 1912. 1899. Andaman Islanders.. p. 1S74. contenant le vray Discours et Narration Nicolas. C. Perry. Milan. A. A.. Edge. New York. 1904. p. 1848. Bd. . Geologische Theil. 1903. 1886. R. Linschoten (Lintscotus). Sarasin. [Printer]. to Record* of Museum. 1878. Frankfurt-a-M.. u. 1914. Society. 1601. Inst. L. Anthrop. Partington. 1876.Expedition. Nookerji. Ostmikronesien und Samoa. Indies Orientalis). Savage. A. Amstelredam. Adventures and Researches among the Baessler-Arch.. Inc. I. Premier Lime. Leide. 1878. A. R. E. Wien. Munday. . Kiikenthal. early History of India. The Gems of the East. 133 In the. XI. Boro-Boedoer. 6... p. F. Neu-Ouinea. Archipel. etc. 1908-1910. . Nicola* Landor.-Eth. Kor. Bonn. Andamans and II. G." Richter. Nieuenhuis. 1896. 218. 1905.I. Melanesian Cambridge. Crozefs Voyage " and its Guam Tasmania. B.d. Oxford. Anth. des Kiinigl. Atlas du Voyage de la Ethnographical Album of the Pacific Islands. Key. E. X. S it dsee. G. 1863. Hawaii. Arch.The Megalithic Culture of Indonesia. C. Paris. " Forschungsreise in den Molukken und in Borneo. 0.oder Kii. Scucktn bergischen naturfors. H. 268 Martin. C. Museums zu Dresden. New Zealand. Anthrop. Pallander. Oesellschaft. 1918. The Outriggers of Indonesian Canoes. J. de VHistoire de la Navigation aux Index orientate. W. F. Narrative of Events in Borneo and Celebes from the Journal of James Brooke. 1899. Mager. Un Viaggio a Nias. The H. 428. H. in Bry. Leipzig. E. A. 1890. LXIII. E. historique du Voiage faict par les hiucl Navires. Miiller. etc. T. Quer durch Borneo. Manchester. d. Pars III (Tertia Pare. Meyer. Die Samoa -inseln. 1906 (1907). "North Queensland Ethnography. M. M. G. A. V.. L'Isola della Donne. Journ. Inst. London. sluik-en kroesharige rassen. J. 235. F. 1912. and Heape. H. Boro-Boudour (French translation). Journ. Stuttgart." Abhandl. India Orientalis. W." Smithsonian Report. 1882. . Leiden. and F Smith.Inseln des O. Indian Shipping. VIII. 1902. Bull. Roth. 1906. H. 1894. f. Mus. Safford. Der H. 1886. Nicobars. reprinted in Evolution Inst. I. 1904-07. von. 1894 . G. Krause. 1916 " Abhandl. W. 3rd Edit. II. 1883. Ethnograph. Manchester and London. 1890." Sydney. p. Leiden. II. A. n. Kramer. W. W. P. London. XII. ~F. Yap. Ling. W. Man. p. IV. K. W. Malay Rosenberg. J. H.. p. C. and Geisterfallen im Ostindisohen Archipele. Modigliani. Krieger. ]>. L. 1902 (1903). 1601. Radhakumud Riedel. Kloss. History of Rivers. stddl. Blalt d. Mayer.. C. 1896. 1901. The Reisen in Celebes. Miszellen. Le Monde polynesien. 1902. H. " The Islands of the Mid-Pacific. 1917. 1798. Le Second Livre Journal ou Comptoir. Perouse. Viilkerkunde zu Leipzig. Pfluger. 434. E. Lane-Fox (Pitt-Rivers). 1905. 1911. Wiesbaden. f. Langen.Pen De and Craft round the World. de. J. Francofurti. People. 1903. A. Smaragd-inseln der Siidsee'. 1910. Hamburg. Berlin. HADDON.A. Milan. Ergebnisse der Amsterdam. Berlin. 1888. 1903. van. S. Oxford. s' Gravenhage. H. Leiden.. 1903. Roth. The Log of an Island Wanderer. No. ftir Ethnogr. I. XV. Anth. Mouat." The Scientific Monthly. II. XXII. 1873 J. R. B. Ethnographische Beitr<iye zur Kentniss des Karolinen Archipelt. ische Archipel. 117 . 14.. 399 of Culture by Pitt-Rivers. etc. Journ. Zool. 1914. 1598. C. Ber. Kubary. Reisen in den Molukken. Deutsch Gea. Anthrop. etc. A. Leemans. W. . A. and Pencil Sketches of Shipping Pritchett. J. Leiden.

No. Halle. Narrative of the United States Exploring Expedition. Arch. Double outrigger with a small board (" peg stanchion ") attachments to the two floats. of Bengal." I may add that I have just come across a very badly drawn illustration of a canoe from Mohelia. all with two booms : : Direct inserted in a model of a sailing boat with an ordinary float on one side. Begunne Anno 626 . by the middle of the nineteenth century both forms occurred. the former is the longer and has more illustrations. These reached my more valuable The Origins and Ethnological Significance of Indian Boat Designs. London and Philadelphia. 1. Single outrigger with similar attachments. it serves to extend -do. Naturforscher. Vol. 1869. VII. the latter having rungs between the limbs. IV. ADDENDA. Die polynesischen Inseln a Abh. Nr. 1845. HADDON. Hornell discusses the Javanese ships of the Borobudur sculptures and gives new drawings of my Figs. A. London 1634. V. F. C. W. Berlin. LXXX. The author believes that the MalayoPolynesian migrations left Indonesia before the Hindu influence there. 1915. 1. [Herbert] Esquier. and that the Sumatrian migration to Madagascar took place in the tenth century A. and took place in the second and fourth centuries A. Printers in Ordinary to His Majesty. In the sixteenth and seventeenth centuries all the canoes appear to have had double outriggers." Man. and on the other the booms carry a reduced float which evidently serves as an outlayer. Leipzig.134 Svoboda. G. Martin's Lane. far Ethnogr. 3. [Reprinted from the Journal of the Royal Anthropological Institute. A Relation of Some Yeares Travaile.The Island World of the Budapest. C. 1920.. -Carol. 1914 Verhandl. Leop. Leitfaden der V iilkerkunde. each pair of the two divergent booms appear to be directly connected with its float. 25 and 28 which clear up certain both have biped masts. Wilkes. but only the fore boom is prolonged on the other side. Thilenius. 1920.Polynesian migrations were originally made on rafts of three beams. The Malay Archipelago. 1920. z^^ Strassburg i. L. Valentijn. St. and that canoes with double outriggers were developed from " them . 1893. 2. 1912. but in the other East African canoes it is oblique and the flat board is similarly horizontal or canted. Bd. 1. 1914. Asiatic Soc. " " Les Pirogues a balancier de Madagascar et de 1'Afrique orientale." Mem* Hornell. 1. Leipzig. p. der der Ostgrenze Melanesiens. London. H. the reduction of these to single outriggers is quite a natural process " . A. This is a very valuable monograph. Hornell rightly observes that these facts justify his previous conclusion that the outrigger canoes of are derived directly from Javanese [or at all events Indonesian] Madagascar and East Africa " He points out that In their consistently double form. Friederici. " Ethnograph. the earlier of these two papers is the as there are copious references. 1726. with a double outrigger . Akad. Malaio-Polynesische Wanderungen. Weule." Paris. 1902. Ltd. the African varieties have types. do. me after paper was in print . Calcutta. Pacific..." Nova Acta. now none but single ones occur. Vojnich. the starboard sheet. the double canoe arose from the latter by the float being increased in size and hollowed out. 1909. Els. 4. as it is at Lamu. . G. R.. 67. T. These excellent papers have appeared too late for me to refer to them.D. . Oud en Nieuw Oost-Indi:-n.D. Deutsch. " Madagascar and East Africa Canoes. 1920. 1892 VI. O. He suggests that the Malay o. Hornell shows that the following varieties occur on the west coast of Madagascar. the booms on the other side being tied to a light bamboo spar to form an outlayer.] Harrison and Sons. des XIX Deutschen Geographentages . " retained a primitive structure almost entirely lost in present-day Madagascar outriggers and that " the Lamu variety having vertical stanchions approaches most closely to the Madagascar form. La Geographie t XXXIV. Wallace. The Outriggers of Indonesian Canoes. J. 139. January-June. . Ergebnisse aus Melanesien 1. K. and the three points booms of this ship are distinctly curved and pass between the two elements of the float. p. Theil. der Kaiserl. . Int. The attachment in Madagascar is always vertical.


the stimulation Guineas. London.. all the principal Anthropological flsF- ' graphs and lanlr use of students of Anthropology.C. 1 Association I eciei ai Advancement anthii:- with . . W. OF GREAT AND This Institute ^ Anthrop promotion of the Science of Mankind A n of globe. can borrow books from the libnn of The appointment Committees to conduct - special investigation*. :asion offers. to and <f 4. reading of papers and i 2. Meetings for qir the. The rnaintena' .ROYAL ANTHROPOLOGICAL BRITAIN INSTITUTE IRELAND.loir. as wfcll F* as in London. His Majesty. Martins Lane. in the establishment i t>ntres of anthropological study and generally. in a rigorously exr- >irow upon i of man's natui- progressive dev imdr hologically. to Institute are requested to Harriion and Sow. object of tb 1 the 1. Ltd. 1 of individual and local effort! to further the objects of the Ir annual fion is Two lue on election (unless -iber) such election tak 1st of aionth of Novern :ceeding year. 50. and on the January 11 A Member may of at any time compound subscriptions by the payment 31 105. in the various branches of Anthropology. foreign investigations ?>f co-operation with individuals and institutions explorations and . mthrop-j The appointmeu collect information. Great Russell Street. operation with the British Sciei. Printer* in Ordinary Persons who wish become Fellows of the communicate with the la Secretary. The iesup Meetings. St. light It investigates. v 3.

d.]) PLEASE DO NOT REMOVE FROM THIS CARDS OR SLIPS POCKET UNIVERSITY OF TORONTO LIBRARY . 1855-1940 The outriggers of Indonesian canoes. Royal Anthropological Institute of Great Britain and Ireland ([n.2 H34 Haddon.GN 440 . Alfred Cort.

Sign up to vote on this title
UsefulNot useful