Volume 53 1944 > Volume 53, No. 4 > O le Fale o le Fe'e, by J. D. Freeman, p 121-144
                                                                                             Previous | Next   

- 121

ARCHAEOLOGICAL sites are rare in Samoa. Other than house-platforms, pigeon-netting platforms, monumental cairns, earth-mounds, and the remains of old fortifications, there is little to be found. One ruin in particular, however, has excited wide interest and comment, and has been instanced by some writers as the relic of an ancient megalithic civilization which was antecedent to the existing Samoan culture. This ruin is known as 'O le Fale o le Fe'e—literally, The House of the Cuttle-fish (or Octopus). 1 Macmillan Brown (Maori and Polynesian, London, 1907, p. 4) has referred to the Fale o le Fe'e as “an ellipse of giant stone columns no mean rival to our stone henge”, and has included it as an integral part of “a definite megalithic track across the old world from the Atlantic to the Pacific”, by means of which he claims to be able to trace migrations; while Stair (Old Samoa, London, 1897, p. 218), writes of “a mysterious building of the distant past known as the Fale o le Fe'e” and suggests a connection with “an earlier, but long since extinct race”.

The purpose of the present paper is to describe accurately the conditions of the ruins as they are to-day; to make a survey of previous accounts by other visitors; and finally, to attempt an assessment of the true significance of the Fale o le Fe'e.

The first recorded visit to the Fale o le Fe'e by Europeans was made in 1845 by the Rev. John B. Stair of the London Missionary Society, and J. C. Williams, Esq., then British Consul at the Port of Apia. Stair did not, however, publish an account of his observations until forty-nine year later (“O le Fale o le Fe'e: or, Ruins of an Old Samoan Temple”, Journal of the Polynesian Society, vol. 3 [1894] 239-44). W. T. Pritchard, Esq., H.M. Consul, who was in Samoa from 1848 to 1858, was the next visitor. His impressions were published as a part of his reminiscences in 1866 (Polynesian Reminiscences; or, Life in the South Pacific Islands, London, 1866, pp. 117-21). William B. Churchward, Acting British Consul, and Deputy Commissioner for the Western Pacific followed. He has published a graphic description of an excursion he made to the Fale o le Fe'e in 1882 (My Consulate in Samoa, London, 1887, pp. 173-83). Te Rangi Hiroa visited the site in 1928 (Samoan Material Culture, Bernice P. Bishop Museum, Bulletin 75, 1930, pp. 324-28).

I propose first to record briefly the principal relevant observations of these earlier visitors, before proceeding to a consideration of the Fale o le Fe'e as it stands to-day.

Stairs states that the Fale o le Fe'e had been 48 feet in length by 45 feet in breadth. He found that “the house had been built of the

- 122
- 123

usual round or elliptical shape, but that the builders, whoever they were, had substituted slabs of basalt for the wooden posts usually placed to support the eaves”. The party discovered “twelve or thirteen of the smaller stone posts still standing, but the large centre slabs lay broken in the middle of the circle. The outer posts, which were still standing, were about 4 feet out of the ground, whilst the centre slabs appear to have been originally about 12 or 13 feet in length, 15 or 18 inches in width, and 7 or 8 inches thick”. The ends had been inserted in the ground, and Stair imagined that, when placed upright, another slab had been laid horizontally upon them, from which other slabs or posts were raised to support the roof. Several of Stair's party had seen the centre slabs standing not long before, and could thus testify to their appearance. It was said that “lately some young fellows, hunting wild pigs, had passed the spot, and amused themselves by pelting the slabs and throwing them down”. It is also recorded that “one portion of the floor had been covered with a pavement of neatly placed slabs of stone; but these had begun to be displaced”. Stair states that the “huge slabs of stone” had evidently been quarried from the side of a precipice which was close at hand, and where masses of the same kind of basaltic rock were exposed. He then presents the theory that the natives had split the rock by kindling a fire upon it in the direction they wished the fracture to run, and then, when the rock was sufficiently heated, they dashed cold water over the heated surface and so accomplished their purpose. The moving of the “huge masses of stone” is accounted for by Stair, by the use of rollers or skids upon which they were shifted or dragged. He mentions “another interesting relic of the old times”—a stone platform or seat, slightly removed from the main ruins, with a sloping stone support; and also records that his Samoan guides drew his attention to what they believed to be large blocks of coral which were lying in the vicinity.

Pritchard states, “the ground plan is an elliptic, say 40 feet in the narrowest, and 50 feet in the widest part, precisely the model and proportions after which the best Samoan houses are still built”. He found that there were 18 pillare standing in the ellipse. They measured 3 feet in height, and were 9 inches thick one way and 6 inches the other. Each pillar, according to Pritchard, had “a notch or shoulder on the inner side, for supporting the roof”. Many pieces of stone were lying on the ground between the pillars which were still standing, “in such positions as to show that the whole of the side pillars were once in their places in the building”. One centre pillar was standing. It measured 5 feet in height, and 12 inches thick one way and 9 inches the other. The top was broken off and lay at its base. A second centre pillar was lying on the ground near the place it had once occupied. It measured 7 feet in length, exclusive of the broken pieces still lying in the same line, which would have increased the length to 13 feet. A block of stone 6 feet long and 6 inches square was lying on the ground near the centre, and Pritchard suggests that “this was evidently intended to rest on the top of the centre pillar and there to form the upper angle of the roof”. He continues, “rafters lie scattered about, some inside amongst the ruins, others at the base of the hill whence they were cut, showing the house was never completed. They are in lengths of 12 and 6 feet, and are 4 inches square. Allowing for the outward curve given by Samoan builders to the rafters of houses after this model, and for the relative heights of the centre and side posts, two of the 12 feet and one of the 6 feet lengths would be required to span the arch from the wall plate to the ridge pole, that is to form one complete rafter”. Pritchard also makes mention of the - 124 stone slab in the form of a seat with a sloping support; of the bluff from which the rafters were cut; and of “immense coral boulders”.

Churchward's account of his visit is less detailed than that of either Stair or Pritchard. He mentions “upright stones about 6 or 7 feet high and of irregular thickness” forming the outer walls of a “house of the usual Samoan shape, rounded at both ends”. He states that the pillars “bore no traces of dressing, but seemed to have been split from the strata of basaltic rock in the cliffs close by”. He suggests the use of fire and water. In the centre of the outlined house was a stone rising much higher than the outside ones and fractured at the top. Churchward adds that there was nothing left to show what sort of roof the house had, if it ever carried any at all; and says that “the probability is if it was covered, that it was not with stone, as amongst the debris scattered about, no segments of anything approaching an arch-curve are to be found: besides that, there is no indication of sufficient strength of outside wall to sustain the thrust of a stone arch”. He suggests that it is very likely that there never was any roof at all, but “merely a collection of stones similar to the Stonehenge or other Druidical circles”.

Te Rangi Hiroa visited the Fale o le Fe'e on 18 February, 1928. He took a series of measurements, and has recorded fully his observations and conclusions.

The present writer, in the course of a study of the ruins (during the period 1940-43) visited the Fale o le Fe'e on six separate occasions. Considerable difficulty was experienced at first in locating the site, for none of the previous visitors has recorded its exact position; some conversance with the source-waters of the Vaisingano river had been acquired before the ruins were finally found.

The Vaisingano river, which flows into the Apia harbour has a complex system of tributaries. The first important confluence occurs three miles inland, just above Vailima, at which point the Puao river enters the main river from the west. A little over one half of a mile further on is the confluence of the Puale'ile'i river and the Soanga stream. The Puale'ile'i is the main branch, while the Soanga stream is merely an alia 2 (dry water-course). At the confluence with the Puale'ile'i—that is the Vaisingano river, proper, the Soanga stream contains no water at all (except in time of flood), but about two miles further upstream there is a considerable flow of water; all of this disappears underground however, the bed of the water-course being formed by a lava flow.

It is at the source of the Soanga stream, on a tongue of flat land between the main stream of the Soanga and a tributary known as the Vai o le Fe'e (The Stream of the Cuttle-fish), and about seven miles from the mouth of the Vaisingano river, that the Fale o le Fe'e lies. The height above sea-level at this point is approximately 1,500 feet. The position of the ruins is clearly shown on the accompanying maps (figs. 1 and 2). There are two possible routes to the site; one via Vailima, and the other via Mangiangi, both of which have been shown on the maps. The time required to reach the Fale o le Fe'e from Apia, by either route is from four to five hours. The conditions of travel are fairly arduous.

My first three trips were chiefly by way of reconnaissance; on my fourth visit, however, in April, 1943, I was accompanied by a party of seventeen young men from the Samoan Teachers' Training School,

- 125
- 126

Apia, who had volunteered to assist in the work of clearing the site. My previous visits had made the necessity of this course of action evident if a thorough examination of the ruins was to be made. It is doubtful if the site has ever been completely cleared before. Stair, Pritchard, and Churchward all make mention of the dense undergrowth which was presnt, largely obscuring the ruins; and Te Rangi Hiroa in his account, says: “The place was so overgrown, and interfered with in parts by the roots and stem of of a growing tree that it was impossible in the time available to clear the ground sufficiently to make a thorough examination of the floor of the site”. It took eighteen of us over three hourse work before the site was finally cleared. Two very large mamalava (Planchonella samensis) trees encroach upon the site; these trees must be of great age. They were mentioned by Stair as being present in 1845.

It was found that there is no clearly defined paepae or base, and that although the base, such as it is, is roughly ellipitical in shape, it is impossible to arrive at any more than an arbitary set of dimensions.

- 127

This accounts for the variation in the measurements recorded by previous visitors. The figures given by Te Rangi Hiroa: 50 feet by 45 feet, agreed with my own measurements.

It is difficult to determine to just what extent the ruins have been meddled with by sightseers. Since Stair's visit in 1845, it is certain that considerable interference has occurred, 3 and Stair himself, relates that he was informed of the fact that some of the slabs had been thrown down by a party of Samoans not long before his own trip. Nevertheless, I have recorded the approximate disposition of the pillars as I found them in April, 1943 (fig. 3). In all about sixty coloumns varying from 2 feet to 5 feet in length lying scattered about, were counted. In addition to these there were many smaller fragments. Only the larger pillars have been included in the diagram.

Some of the pillars were standing, but none of them were firmly embedded in the ground in the manner one would expect of permanent fixtures. It was evident that many of them had been recently erected—some were merely propped up, and so no attempt was made to make a separate record of their positions. The majority of the pillars were fallen and some were well nigh buried.

The twenty largest of the columns present at the site were measured. The columns are for the most part slightly irregular in shape, but the following table of measurements gives a good indication of their character:—

5ft. 9ins. 7½ins. 6ins.
5ft. 2ins. 12ins. 9ins.
5ft. 2ins. 6½ins. 6ins.
4ft. 10ins. 6ins. 5½ins.
4ft. 6ins. 8ins. 7ins.
4ft. 10ins. 5ins.
3ft. 10ins. 14ins. 5ins.
3ft. 10ins. 11ins. 6ins.
3ft. 10ins. 8½ins. 8ins.
3ft. 9ins. 10ins. 7ins.
3ft. 8ins. 9ins. 5ins.
3ft. 4ins. 8ins. 7½ins.
3ft. 4ins. 6ins. 4ins.
3ft. 1in. 9ins. 5ins.
3ft. 8ins. 7ins.
3ft. 8ins. 5ins.
2ft. 9ins. 9ins. 5ins.
2ft. 8ins. 8½ins. 3ins.
2ft. 7ins. 7ins. 3ins.
2ft. 6ins. 8ins. 4ins.

The quarry from which the columns were obtained lies about 50 feet to the south of the Fale o le Fe'e. It is in the form of an almost perpendicular cliff. The exposed face of the cliff is approached by a steep slope of volcanic earth. Many columns are still visible—some partially detached. All of the columns run in vertical strata. The greatest exposed face is approximately 15 feet in height. A number of fallen columns are strewn over the slope at the base of the cliff. The distance from the cliff face to the site of the Fale o le Fe'e is slight; and any of the columns at present to be found among the ruins could easily have been moved there by manual labour alone.

A section of one of the columns from the Fale o le Fe'e was submitted to Dr. C. O. Hutton, of the Geological Survey Department, - 128 Wellington, N.Z. Dr. Hutton's examination of this specimen proved it to be an analcite dolerite. The columns are of natural formation and have not been worked in any way. The cliff from which the columns were obtained is also a natural formation—such columnar formations are not rare in the volcanic islands of the Pacific. The geological evidence disposes of the theory postulated by both Stair and Churchward, that the columns were “split” from the cliff face by the use of fire and water. Moreover, fire and water would not have the effect of producing columns of the dimensions of those to be found at the Fale o le Fe'e.

Both Stair and Pritchard make mention of a large centre slab lying broken on the ground. They both estimated its length to have been about 12 or 13 feet. In the course of my examination of the ruins I discovered that what appeared to be one of these centre slabs was still present. The slab in question was in three sections, lying end to end and in the same line, and firmly embedded in the earth. Only the upper surface of each section was visible, and it appeared that the three separate sections were in fact formerly a single slab about 12 feet in length. This would certainly have been the conclusion of a casual observer, and it is reasonably probable that this was the slab referred to by Stair and Pritchard. It was decided to excavate the three sections with a view to determining their exact dimensions. They were accordingly unearthed and measured. The results were as follow:—

First section 5ft. 10ins. 7½ins.
Second section: 3ft. 10ins. 13ins. 3½ins.
Third section: 2ft. 3ins. 9ins. 6ins.

From a comparison of the width and depth measurements of the three sections it will be seen that any possibility of their once having been a single slab is immediately ruled out; moreover the lines of fracture did not correspond with one another.

To the south of the main ruins, at a distance of about 9 feet, is the stone ‘seat’ with a sloping support mentioned by all four writers—Stair, Pritchard, Churchward, and Te Rangi Hiroa. The name given to this arrangement of basaltic stone by the Samoans is 'O le nofoa 'o le Fe'e (The Seat of the Cuttle-fish). 4 The stone slab forming the main portion of the seat is the most massive present in the ruins. The sloping stone support is about 6 feet in length and is firmly embedded in the ground. The arrangement of the stones is clearly shown in plate 1—a photograph taken over thirty years ago showing Dr. E. Schultz (who succeeded Dr. Solf as Governor of German Samoa), seated upon the nofoa o le Fe'e. 5 Stair has suggested that the nofoa o le Fe'e was used as a seat by a priest or that it may have served as a “coronation seat, or post of honour, at the inauguration ceremonies of a chief's installation”. Te Rangi Hiroa, however, writes, “Though said to be

- i
- ii
- 129

a seat, it appears more like a monumental tia raised over a grave” (Samoan Material Culture, Hawaii, 1930, p. 328). With a view to testing the validity of this opinion, the main stones were removed and the area beneath excavated to a dept of 3 feet to 4 feet. Compressed earth of clayey composition, which bore no signs of having been previously disturbed, was encountered. No trace of any burial was found.

In the course of my investigations I was able to collect much legendary information concerning the Fale o le Fe'e from the people of Vaimaunga. My principal informants were from the villages of Tanungamanono and Mangiangi; I may mention in particular the names of Ne'emia of the former village, and Ti'auli of the latter. I also obtained valuable information from Eti, Chief Inspector of Samoan Schools in Western Samoa. All of my inquiries were carried on in the Samoan language.

According to my informants, the Fe'e came from Fiji, 6 where he was a powerful god. On his journey he is said to have stayed for some time in Pu'apu'a, Savai'i, where there is a large river. He finally arrived at Apia, however, where he made his home. It is in this way that the formation of Apia harbour is explained. At Apia, the Fe'e met a beautiful young woman called Sina-alalafa, 7 the daughter of Fasavalu, a chief of Falelatai, A'ana, whom he courted and later married. The Fe'e decided that if he was going to stay in Vaimaunga (the district of which Apia is a part), he would live far inland beyond the areas inhabited by men, and with this in view he retired to the source-waters of the Vaisingano river. Sina-alalafa, his wife, lived at Tanumaleto, near Vailima. She is said to have had several children of whom the Fe'e was the father. The exact number is uncertain, but the following names of four of the children were quoted: Mualeoā and Samalaulu, both girls; and Longovā and So'oalo, 8 both boys. The Fe'e resolved to build himself a house and he called together from all over Samoa a company of tufunga, 9 or house builders. All of these tufunga were aitu or sauali'i—that is, they were gods, like the Fe'e himself. The Samoans do not attribute the Fale o le Fe'e to any human agency; rather, they actively deny that men had anything to do with its construction. Furthermore none of my informants believed that the Fale o le Fe'e was ever actually completed. There are extant several legendary explanations for this state of affairs. One of the legends tells that after the work had been in progress for some days the head tufunga asked one of his assistants to fetch him some drinking water. The assistant proceeded to the nearby stream—the Vai o le Fe'e, to get the water, but returned to report that the stream was discoloured with blood. He was ordered to return to the stream once more, this time with another of the minor aitu present. The two of them soon returned, however, to announce that the water was still tainted. At this, the head tufunga appointed a third from amongst those present to accompany the other two, and instructed all three of them to go upstream until they discovered the source of the pollution. The three - 130 aitu proceeded up the bed of the stream until they reached a large pool, where they found a woman. 10 She was washing some ngatu (Pratt, Grammar and Dictionary of the Samoan Language, Samoa, 1911, p. 151, Ngatu, s., an old siapo, 11 used as a wrapper, or as old rags in sickness). Upon seeing the three aitu the woman fled into the forest and escaped. The searchers returned and related what they had seen. Upon hearing their story the head tufunga was greatly angered and he ordered that all work upon the Fale o le Fe'e should immediately cease, and that all those present should disperse, each returning to his own district—and this was done. Another version which I recorded, is to the effect that the work was stopped because the company of tufunga was displeased at the unseemly noise created by So'oalo (the son of the Fe'e) who was living at Tanumaleto. Pritchard (Polynesian Reminiscences, London, 1866, pp. 117-18), relates a legend gathered from a Vaimaunga orator (tulafale), which accounts for the unfinished state of the Fale o le Fe'e in yet another manner. According to Pritchard's informant, the Fe'e in order to test the loyalty and obedience of his subject gods, commanded them to build a stone house. Some carried coral from the reef up to the mountains, and others cut stones from the hills for pillars or rafters. No song, or word, or sound was heard as the work progressed—all was silence. The labour was hard and harassing. The Fe'e lounged arrogantly on a stone slab, watching the operations and drinking kava. He thought himself the greatest of gods, unmatched in power, unrivalled in prowess, unequalled in magnificance. Suddenly, in the midst of the labourers, there stood a god who proclaimed himself the rival of the Fe'e and challenged him in battle. The Fe'e, pushing aside the kava-bowl, proudly arose; and the two gods wrestled together. The Fe'e was vanquished, and fled in shame to the sea from whence he has never returned; and the labourers, freed from his yoke, abandoned their work. The Fale o le Fe'e was left unfinished—left as it now stands.

However fanciful these legends may appear to Europeans, they at least offer a quite satisfactory explanation of the unfinished state of the Fale o le Fe'e to the Samoans, and credence is still placed in them by many of the people. The fact that there are these varying explanations would suggest, however, that the Samoans actually have no reliable knowledge of the origin of the Fale o le Fe'e, and that the legends have been developed merely to take the place of the authentic information which is lacking. 12

Further evidence of the Fale pou ma'a having been the abode of the Fe'e was pointed out to me by my informants. As already mentioned there are several limestone formations 13 in the vicinity of - 131 the Fale o le Fe'e; and the Samoans are still adamant in their belief that what in fact are pieces of travertine, are clumps of coral brought from the far-distant reef by the Fe'e. To them this is the quite obvious solution. In extenuation let it be recorded that more than one European visitor has been similarly mistaken. I was informed also that until quite recently a number of small, blue, coral fish known as mamo were to be found in the Vai o le Fe'e. These fish, which are to be seen in almost every lagoon in Samoa, live in clumps of coral, and they too are claimed to have been brought up to the source-waters of the Soanga stream by the Fe'e.

It is well established that the Fe'e (cuttle-fish) was once the war-god of the Vaimaunga district. Kraemer (Die Samoa-Inseln, Stuttgart, 1902, p. 229), Stuebel (Samoanische Texte, Berlin, 1896, p. 76), Turner (Samoa a Hundred Years Ago and Long Before, London, 1884, p. 31), and others, make mention of the fact; and further, it is still well recollected by the chiefs (ali'i) and orators (tulafale) of Vaimaunga themselves.

In ancient Samoa the war-gods occupied a position of great importance, and were held in high regard. Some such as the goddess Nafanua gained especial prominence and received recognition throughout the group; but in addition to these major figures each had its own war-god to which it paid homage. Turner (Samoa a Hundred Years Ago. . . , London, 1884, pp.23-66) has given numerous examples. In many instances these war-gods were considered by the Samoans as being incarnate in animal form. Thus the owl (lulu), the heron (matu'u), the kingfisher (ti'otala), the rail (ve'a), the swamp-hen (manuali'i), the sea-eel (pusi), and the flying fox (pe'a) were all believed to be incarnations of certain war-gods, each in its own district. The cuttle-fish (fe'e), was similarly regarded, and apparently by other districts as well as Vaimaunga. Turner (Samoa a Hundred Years Ago. . . , London, 1884, pp. 28-32) gives descriptions of the varying ways in which the cuttle-fish was worshipped as a war-god; and Stair (Old Samoa, London, 1897, p. 218) states that the cuttlefish was also the war-god of A'ana. 14

The Fe'e, according to my informants was the war-god of the whole of the Vaimaunga district and was revered by all within its boundaries. The district of Vaimaunga extends from the village of Lauli'i to the village of Alamangoto; that is, roughly, the area between the two mountains which appear on the Admiralty Chart of the Navigator islands (Jan., 1916 edition) as Vaea Berg and Letogo Berg. This fact offers an explanation of the origin of the name Vaimaunga, which literally means—The space between the mountains (Pratt, Grammar . . . Samoan Language, Samoa, 1911, p. 345, va, s. a space between; p. 183, maunga, s. a hill, a mountain). The Vaimaunga district is made up as follows:—

- 132
Lauli'i Tuisamau
Vailele Saleupolu
Mangiangi Fuaiupolu
Apia Sangaunga
Alamangoto 17  

The poulation of Vaimaunga in former times would probably have been about 3,000 people.

Although it is definitely known that the Fe'e was the war-god of Vaimaunga, there is no direct evidence that the Fale o le Fe'e was ever used as a place of worship. All of my informants denied that it had ever been so used. Any claim that the Fale o le Fe'e was once used as a temple (malumalu) is, moreover, not supported by our knowledge of religious practice in ancient Samoa, for it is well known that the Samoans did not have especially-prepared temples for the worship of their gods. Stair (Old Samoa, London, 1897, p. 226), for example, in discussing the dwellings (called fale-aitu: spirit-houses; or, malumalu-o-le-aitu: the temple of the aitu) in which the war-gods were commonly worshipped states:—

“These spirit-houses were built in the usual shape and style with nothing in their build or finish to distinguish them from other dwellings, being at times mere huts.”

According to my informants the Fe'e was invoked by the priests (taula aitu) 18 of Vaimaunga in time of impending war with a view to gaining a premonition as to the course the conflict would take. If a prolonged roaring noise was heard to travel up the course of the Vaisingano river and then up that of the Soanga stream toward the Fale o le Fe'e, this was known to portend disaster, and the projected war was abandoned; if, however, the noise was heard proceeding downstream from the mountains were the Fe'e dwelt, toward the sea, this was considered a favourable omen and the war was prosecuted with all the assurance that the knowledge of supernatural approval brings. 19

The exact nature of the roaring noise (which many of my informants claimed to have heard in earlier days), was described in varying ways. One informant used the term u, which means, to emit a hollow sound or to roar, as the waves on the reef, or the tramp of troops (Pratt, Grammar . . . Samoan Language, Samoa, 1911, p. 57); another - 133 used the word, ngongolo, which means a rushing sound, as of waves, wind or thunder (Pratt, Ibidem, p. 152); while a third employed the term ta'alili, which means resounding, sonorous, as waves, thunder or a conch shell trumpet (Pratt, Ibidem, p. 278).

The power to presage the future was apparently considered the major function of the war-god in old Samoa; and many similar invocations to the one related above have been recorded. It the pe'a (flying fox), a war-god of Savai'i, for instance, was seen flying before the warriors all was well; but if it flew in the opposite direction it was a sign of defeat, and a warning to go back. Another war-god, known as Taema, was considered to be present in a bundle of sharks' teeth done up in a piece of tapa-cloth, and before going into battle was always consulted; if the bundle felt heavy it was a bad augury, but if it felt light it was a prediction of success. Many further examples are recorded by Turner (Samoa a Hundred Years Ago . . . , London, 1884, pp. 16-77). Especially worthy of note is the example he gives connected with the cuttle-fish itself (Turner, Ibidem, p. 28); the district is not specified however. Turner states that before going into battle all assembled on the malae of the village concerned and one of the priests prayed as follows:—

Le Fe'e e! fa'afofonga mai ia,
'O a'u, 'o Fale, o le ā tula'i atu nei;
Le Fe'e e, 'au mai ia ou mūmū fua,
Se'i tau a'i le taua nei.
O Fe'e! listen,
I am Fale, 20 who now stands up;
O Fe'e! give us your red flaming rage,
With which to fight this battle.

Turner says that all listened carefully to the enunciation of this prayer by the priest, for if he was observed to stutter a single word it was a bad omen.

The following superstitions, common amongst the people of Vaimaunga, concerning the Fe'e are also worthy of mention. It is strictly forbidden for anyone to urinate in either the Vai o le Fe'e, of the Soanga stream, for such an act arouses the anger of the Fe'e, and before long the stream involved will be in violent flood. Some of my informants averred that the flooding occurred even though no rain was observed to fall. The superstition is still firmly believed by most, and I was frequently warned of the situation during the period I was conducting my investigations. Other informants told me that the presence of the Fe'e far inland is always indicated when the waters of the Vaisingano river are badly discoloured (by the Soanga stream) and “run white”, as it was put. Some of my informants claimed that the Fe'e still dwelt in the mountains behind Vaimaunga.

I was able to find no evidence of there ever having been a tapu (sa) on the eating of the octopus as food in Vaimaunga. There is at least no such prohibition existing to-day, nor has there been one within existing memory.

In former times the Samoans attached particular significance to the Fe'e; it is a name that is encountered repeatedly in Samoan mythology. The Samoan underworld, or hades was known as Sa le Fe'e—literally, The Clan of the Cuttle-fish. According to a myth from Manu'a, the Fe'e was discovered floating on a piece of coral by the god Tangaloa, and upon stating that he had no parents, was taken - 134 to Manu'a, where he became the father of the demon-like beings Sinasa'umani and Sasasa'umani (Kraemer, Die Samoa-Inseln, Stuttgart, 1902, p. 45). Pratt (Some Folk-songs and Myths from Samoa, Journal of the Royal Society of New South Wales, 1891, pp. 70-73), has recorded a legend from Manu'a telling of Sa'umani and the Fe'e engaging in a fishing competition. And Kraemer (Ibidem, p. 231) makes mention of the Fe'e being killed by Ti'eti'e-i-talanga of Upolu.

These, and other legends, however, would seem to refer to an ancient Samoan god called Fe'e, who is no way related to the war-gods of Vaimaunga and A'ana. Williamson (The Social and Political Systems of Central Polynesia, Cambridge, 1924, vol. 1, p. 95), writes:—

“The interest of this matter for my present purpose is that Fe'e was a Samoan god of the dead, the souls of whom plunged into the sea at the western extremity of Savai'i; that after his departure his place was, it is said taken by Savea Si'uleo, another Samoan god of the dead; that this last mentioned god was the same as Hikuleo, the Tongan god of the dead; that in Samoa he was primarily a Savai'i god, more especially connected with the Tonumaipe'a family, whom I regard as having been of Manu'an (Tangaroan) origin, and who claimed him as their divine ancestor (the father of the ancestral war-goddess Nafanua); and that according to Samoan beliefs, Fe'e was specially connected with Sa le Fe'e, their hades below the earth, whereas Puloto, their paradise, believed by some to be an island away to the west, was presided over by Si'uleo. I am inclined to think that Fe'e was one of the oldest gods of Samoa, and that he was probably a god of the dead of the pre-Tangaroans, having taken the place of Mau'i, a still older god of an ancient volcanic cult.” 21

Williamson (Ibidem, p. 230) stresses the danger of confusing this major figure with the minor war and household gods who were considered to be incarnate in the fe'e (cuttle-fish). Moso, for instance, was considered in one district to be incarnate in the cuttle-fish (Turner, Samoa a Hundred Years Ago . . . , London, 1884, p. 37); and the same remark also applies to another general village or household god, Taisuamalie (Turner, Ibidem, p. 59). The household god Samani, included the cuttle-fish among its incarnations (Turner, Ibidem, p. 72); so did the household-god Soesai (Turner, Ibidem, p. 74); and the god Tuiali'i (Turner, Ibidem, p. 76). The war-gods of Vaimaunga and A'ana may also be added to the list.

While it is true that all of these minor gods were considered to be incarnate in the form of the cuttle-fish, none of them must for this reason be associated with the god of the underworld, Fe'e. The word for cuttle-fish is the same throughout all Polynesia, subject to the differences that arise in accordance with the rules for the interchange of consonants and it is not unusual to find cuttle-fish gods worshipped. The cuttle-fish was apparently one of the more common forms of incarnation in Western Polynesia in ancient times, 22 and especially was this the case in ancient Samoa. There are, however, no grounds at all for assuming that the Fale o le Fe'e and the legends concerning it refer to any other than the war-god of Vaimaunga. There is - 135 certainly no evidence that the Fale o le Fe'e was in any way connected with the worship of Williamson's pre-Tangaroan god, Fe'e.

We are now in a position to make an analysis of the data which has been assembled. Both Stair and Pritchard, it has been seen, inferred that the Fale o le Fe'e had once been a structure with the typical framework of an ordinary Samoan fale tele (round house) but that the framework had been made of stone pillars. As the circumference of the base was marked by a number of upright stone pillars it was judged that these had formed the wall-posts (pou lalo) of the house; and it was thought that the centre slabs (only one of which was ever seen standing) were the main supporting-posts (pou tū). It was suggested that other stone slabs rested on the top of the centre pillars, there to form the upper angle of the roof, and so on. Upon examination, this theory of Stair's and Pritchard's is found, I believe, to be untenable. The base of the Fale o le Fe'e measures 50 feet by 45 feet. This area is larger than that occupied by the average Samoan fale tele. Te Rangi Hiroa (Samoan Material Culture, Hawaii, 1930, p. 24), gives the dimensions of an average-sized fale tele at Fitiuta, Manu'a as being 31 feet 7 inches by 27 feet 6 inches. The base-measurements of an exceptionally large house at Iva, Savai'i were 54 feet 8 inches by 45 feet 6 inches. The main posts (pou tū) of this house at Iva were 32 feet above floor-level, which in turn was over 6 feet above ground-level, so including the portion sunk in the ground, these posts must have been over 40 feet in length. The base dimensions of the house at Iva approximate to those of the Fale o le Fe'e, but the longest centre-pillar seen standing by Pritchard was only 5 feet high. Both Stair and Pritchard suggested that the centre pillars had once been 12 feet or 13 feet high, but this still falls far short of the height required—about 40 feet, for a fale with such a large base. An average-sized fale tele, examined at Sa'anapu, Savai'i had thirty-two wall-posts (pou lalo) and the length of these posts averaged over 6 feet. Pritchard saw only eighteen pillars standing, and gave their average height as 3 feet. There are certainly not sufficient pillars of the length required for wall-posts to be found at the ruins as they are to-day. Pritchard stated that all of the wall-posts had a “notch or shoulder on the inner side for supporting the roof”. Only three or four notched pillars were seen; furthermore, the wall-posts (pou lalo) of a Samoan fale are always notched on the outer side to support the wall-plates (amo pou), and not on the inner side as Pritchard suggests. There is the further point that in the construction of a Samoan fale the wall-posts are the last portion of the house to be erected and not the first. Pritchard calculated the length of the stone rafters required to span the arch from the wall-plate to the ridge-pole as being 30 feet, made up of two 12 feet lengths and one 6 feet length. Pillars of this length are not to be found among the ruins as they exist to-day, and when one bears in mind the great number of rafters (fatunga), purlins (amoamo), collar beams (so'a), and arches (fau) required in the construction of a fale tele it is seen that the supply of stone pillars present at the Fale o le Fe'e is altogether inadequate for the construction of a complete house. Furthermore, how were the three separate sections mentioned by Pritchard ever joined? Pritchard speaks of allowing for the outward curve given by Samoan builders to the rafters of houses. How could such a curve be attained with three pieces of rigid material? In addition there is the even more impossible task of providing the arched purlins which form the rounded end (tala) of a Samoan fale. From a purely technical point of view the theory that the Fale o le Fe'e was ever a completed (or partially completed) - 136 stone structure must be dismissed. The problems to be faced in its construction are insuperable, in terms of Samoan house-building technique; and these technical problems must be disposed of before a structure in stone can be seriously considered. The theory of Stair and Pritchard is not, moreover, supported by Samoan opinion. All of my informants were quite definite in their assertion that there never was a completed structure; and such legends as do exist are not in agreement with the theory of a finished stone house. I discussed the question with a number of Samoan tufunga (house-builders); they were all most decided in their opinion that a stone structure as suggested by Stair and Pritchard would be impossible of achievement.

An examination of the legendary data concerning the Fale o le Fe'e does not lead to any decisive conclusion. Tradition indicates that the ruins were in some way once associated with the war-god of Vaimaunga, who was incarnate in the fe'e; but there is no existing record of the Fale o le Fe'e ever having been used as a temple or place of worship, and there is no reliable information concerning its origin.

Although the geological evidence proves that the columns are of natural formation and were never worked, it is certain that some human agency was responsible for the Fale o le Fe'e—even if we assume nothing more elaborate than the ruins discovered by Stair. The conflicting legends of its origin seem to be attempts to account for a phenomenon concerning which the Samoans have no definite information. This at least suggests that the ruins are of considerable antiquity. The Samoans have been in Samoa long enough, however, for the actual circumstances of the origin of the Fale o le Fe'e to have been forgotten, and for legendary explanations to have arisen in their place in the natural process of time.

In face of the lack of any real evidence, we are in a position to do no more than offer tentative hypotheses concerning the Fale o le Fe'e. Te Rangi Hiroa (Samoan Material Culture, Hawaii, 1930, p. 327) has written:—

“As I visualize the circumstances, the site between the fork of the streams with the basaltic cliff at the back intrigued the imagination in early times. The Polynesian is more readily influenced by freaks of nature or by the uncommon that most people imagine. The basaltic columns were talked about by fowlers and others who had seen them. The site became a camping ground and a place of interest that was visited. People played with the fallen blocks and let the imagination evolve around the idea of using the pillars as house material.”

This is a quite credible explanation. The possibility that the Fale o le Fe'e may once have been a place for the worship of the Vaimaunga war-god, the Fe'e, must also be borne in mind, however. A series of posts set up in the same manner as those of a Samoan fale may have provided the setting for kava ceremonies; and the massive slab known as the nofoa o le Fe'e, together with the travertine which was taken as being coral, may conceivably have been associated with the worship of the Fe'e as a war-god.

Finally, certain definite conclusions may be drawn from the material, which has been presented. These conclusions—although largely negative in character—will at least serve the purpose of stripping the Fale o le Fe'e of a little of the “mystery and wonder” with which it has become enshrouded, and of enabling those interested in Polynesian research to regard the ruins in a more realistic light.

- 137
  • 1. The stone columns which were considered by some earlier writers to have been “wrought by the natives” are of natural formation.
  • 2. The Fale o le Fe'e must be considered as a product of the existing Samoan culture; there is no reason to assume that the Fale o le Fe'e is the relic of a pre-Samoan megalithic civilization.
  • 3. Macmillan Brown's reference to the Fale o le Fe'e as “an ellipse of giant stone columns no mean rival to our own stone-henge” is not justified by the facts; and further, there are no grounds for considering the Fale o le Fe'e as an integral part of the “megalithic track” upon which Macmillan Brown bases his theories.
  • 4. The theory put forward by Stair and Pritchard that the Fale o le Fe'e was once a stone structure with the typical framework of a Samoan fale must be dismissed.
  • 5. There is no evidence to suggest that the Fale o le Fe'e was in any way connected with Fe'e, the Samoan god of the underworld.

I am indebted to Dr. C. O. Hutton for the examination of geological specimens; to Mr. J. Radford for assistance in mapping the headwaters of the Vaisingano river; and to Mr. J. T. Salmon and Mr. A. J. Tattersall for photographs.

  • 1. Photograph of a part of the Fale o le Fe'e taken by Mr. A. J. Tattersall, of Apia, Western Samoa. The photograph shows Dr. E. Schultz, Governor of German Samoa from 1910-1914 seated upon the nofoa o le fe'e.
  • 2. Photograph of a section of a column collected at the Fale o le Fe'e in April, 1943. The dimensions of the section shown are: Length, 6½ins.; width, 2¾ins.; depth, 2ins. Otago Museum, N.Z., D.44.11. Photograph by Mr. J. T. Salmon, Dominion Museum, Wellington, N.Z.
  • 3. Photograph of a piece of travertine collected at the Fale o le Fe'e in April, 1943. Length, 5½ins. Otago Museum, N.Z., D.44.12. Photograph by Mr. J. T. Salmon, Dominion Museum, Wellington, N.Z.
  • 1. Map of the island of Upolu, Samoa, showing the location of the Fale o le Fe'e, and of Vaimaunga district.
  • 2. Map of the source-waters of the Vaisingano river, Upolu, Samoa, showing the location of the Fale o le Fe'e.
  • 3. Sketch plan of the Fale o le Fe'e made in April, 1943.
  • 1. Beauclerc, G. A. F. W., “The Meaning of Kalou and the Origin of Fijian Temples”, Transactions of the Fiji Society, for the year 1914, Suva, 1915.
  • 2. Brown, George, Melanesians and Polynesians, London, 1910.
  • 3. Brown, J. Macmillan, Maori and Polynesian, London, 1907.
  • 4. Burrows, Edwin G., “Western Polynesia: A Study in Cultural Differentiation”, Etnologiska Studier, vol. 7, Gothenburg, 1938.
- 138
  • 5. Churchill, Llewella Pierce, Samoa 'Uma, New York, 1902.
  • 6. Churchill, W., “A Samoan Legend”, Journal of the Polynesian Society, vol 24 (1915) 118.
  • 7. Churchward, William B., My Consulate in Samoa, London, 1887.
  • 8. Collocot, E. E. V., “Notes on Tongan Religion”, Journal of the Polynesian Society, vol. 30 (1921) 231.
  • 9. Fornander, Abraham, An Account of the Polynesian Race, London, 1890.
  • 10. Frazer, Sir James George, The Belief in Immortality and the Worship of the Dead, vol. 2, London, 1922.
  • 11. Gill, Rev. William Wyatt, Life in the Southern Islands, London, 1876.
  • 12. Gurr, E. W., New Zealand Samoa Guardian, vol. 1, no. 20, 1929.
  • 13. Handy, E. S. C., “Polynesian Religion”, Bernice P. Bishop Museum, Bulletin 34, Hawaii, 1927.
  • 14. Handy, E. S. C. and Willowdean Chatterson Handy, “Samoan Housebuilding, Cooking and Tattooing”, Bernice P. Bishop Museum, Bulletin 15, Hawaii, 1924.
  • 15. Heycock, E. N., “Megalithic Remains in the Pacific, with Especial Reference to the Images of Easter Island”, Transactions of the Fiji Society, for the year 1921, pp. 7-13, 1922.
  • 16. Kennedy, Donald Gilbert, “Field Notes on the Culture of Vaitupu, Ellice Islands, Memoirs of the Polynesian Society, no. 9, New Plymouth, New Zealand, 1931.
  • 17. Kraemer, Augustin, Die Samoa Inseln, Stuttgart, 1902.
  • 18. Lister, J. J., “Notes on the Natives of Fakaofu (Bowditch Island), Union Group”, Journal of the Anthropological Institute, vol. 21, pp. 43-63, 1892.
  • 19. Mackenzie, Donald A., Myths and Traditions of the South Sea Islands, London.
  • 20. McKern, W. C., “Archaeology of Tonga”, Bernice P. Bishop Museum, Bulletin 60, Hawaii, 1929.
  • 21. Pratt, George, Grammar and Dictionary of the Samoan Language, Samoa, 1911.
  • 22. Pratt, George and Fraser, John, “Sa'umani and the Fe'e”, Journal and Proceedings of the Royal Society of New South Wales, pp. 70-73, 1891.
  • 23. Pritchard, W. T., Polynesian Reminiscences, or Life in the South Pacific Islands, London, 1866.
  • 24. Rivers, W. H. R., The History of Melanesian Society, London, 1914.
  • 25. Smith, S. Percy, Hawaiki: The Original Home of the Maori, Wellington, 1910.
  • 26. Stair, Rev. John B., “'O le Fale o le Fe'e; or the Ruins of an Old Samoan Temple”, Journal of the Polynesian Society, vol. 3, pp. 239-244, Wellington, 1894.
  • 27. Stair, Rev. John B., Old Samoa, or Flotsam and Jetsam from the Pacific Ocean, London, 1897.
  • 28. St. Johnston, T. R., The Islanders of the Pacific, London, 1921.
  • 29. Stuebel, O., Samoanische Texte, Berlin, 1896.
  • 30. Te Rangi Hiroa, “Samoan Material Culture”, Bernice P. Bishop Museum, Bulletin 75, Hawaii, 1930.
- 139
  • 31. Tischner, A., Die Verbreitung der Hausformen in Oceanien, Leipzig, 1934.
  • 32. Tregear, Edward, The Maori-Polynesian Comparative Dictionary, Wellington, 1891.
  • 33. Turner, Rev. George, Samoa a Hundred Years Ago and Long Before, London, 1884.
  • 34. Violette, P. L., “Notes d'un Missionaire sur l'Archipel de Samoa, Les Missions Catholiques, Paris, 1870.
  • 35. Watson, Robert Mackenzie, History of Samoa, Wellington, 1918.
  • 36. Williamson, R. W., The Social and Political Systems of Central Polynesia, 3 vols., Cambridge, 1924.
  • 37. Williamson, R. W., Religious and Cosmic Beliefs of Central Polynesia, 2 vols., Cambridge, 1933.
  • 38. Williamson, R. W., Religion and Social Organisation in Central Polynesia, Cambridge, 1937.

In volume 1 of the Journal of the Polynesian Society, 1892, (pp. 62-64) there appeared a note entitled “Prehistoric Remains in Samoa”.

The note was accompanied by the following prefatory remarks by the Editors: “The ‘Asiatic Quarterly Review’ of October, 1890, contains an article written by Mr. R. A. Sterndale, and based on material found in the note-books of the late Handley Bathurst Sterndale, so well known as a competent and enthusiastic Polynesian explorer. We transcribe a quotation from one of the note-books in the hope that some of our members resident in Samoa may be stimulated to make enquiry and research with a view to further discoveries. The burying-place described in the quotation is situated in the mountain ranges of Upolu.”

Then follows Handley Bathurst Sterndale's description:—

“There was no path, although in places I could perceive that there had in former times been one, several crevasses being artificially bridged over with causeways of rude construction.

“By previous observation I had determined the position of a lofty spur (or radius from a great volcanic centre) which, on undertaking the journey I had proposed to myself to ascend, in the hope of thereby reaching the summit of the great interior at a point much to the eastward of where it had been accustomed to be crossed by the natives. Looking in that direction I perceived this ridge separated from me by a broad and dangerous-looking ravine with a narrow canon (or chasm with perpendicular sides) in the bottom. Hazardous as was the appearance of this valley I had to attempt it, and scrambling to the brink of the crevasse which constituted its most inaccessible feature, I found, after some search, a fallen tree, whereby I effected the passage. Beneath me was a torrent flowing in darkness over a bed of black lava as smooth as glass. I knew this to be one of the head waters of a river called the Vai-vasa, which presents the singular phenomenon of exhibiting some miles inland a volume of water more - 140 than double in quantity to that which is visible in its bed where it disgorges itself into the sea, the remainder being absorbed by subterranean channels.

“About 200 feet above me on the opposite side I observed the mouth of a rift or gully opening towards me, and seeming by its aspect to have been produced by an earthquake or some such cause. Having with great labour and with some risk succeeded in reaching the crown of the ridge at some distance below that point, I soon came to the edge of the strange looking crack. There was no way of crossing it except by sliding over fallen boulders to the bottom, and in the same manner ascending the opposite side, where was an opening between the rocks, just wide enough for a man to pass through. As I believed that the end of this gully, which ran at right angles to the direction of the range, might afford me a prospect of the next valley to the eastward, I proceeded in that direction along the bottom, but had not gone far when I perceived to my surprise that it was not a natural fissure, as I had supposed, but a great fosse formed by the hands of man, being in some places excavated, in others built up at the sides; and that which was farthest from me (or next to the rise of the hill) had been still more heightened by a parapet wall. At the far end was nothing to be seen but a perpendicular cliff, and the inaccesible face of the opposite mountain. Returning to the spot at which I entered, I climbed up the other side of the gully, and passed through the narrow gap I had previously noticed, when my astonishment increased on beholding before me upon a level space of limited area, a truncated conical structure of ‘Heidenmauer’ of such huge dimensions as must have required the labour of a great multitude to construct. So little did I expect in this neighbourhood to meet with any example of human architecture, and so rudely monstrous was the appearance of this cyclopean building, that from its peculiar form, and from the vegetation with which it was overgrown, I might have passed it by, supposing it to have been a volcanic hillock, had not my attention been attracted by the stone-work of the fosse. I hastened to ascend it. It was about twenty feet high by one hundred in diameter. It was circular with straight sides; the lower tiers of stone were very large, they were lava blocks, some of which would weigh at least a ton, which must have been rolled or moved on skids to their places. They were laid in courses; and in two places near the top seemed to have been entrances to the inside, as in one appeared a low cave choked with rocks and tree roots. If there had indeed been chambers within, they were probably narrow and still existing, as there was no sign of depression on the crown of the work, which was flat, and covered with flat stones, among which grew both trees and shrubs. It is likely that it was not in itself intended as a place of defence, but rather as a base or platform upon which some building of importance, perhaps of timber, had been erected, no doubt in the centre of a village, as many foundations of a few feet high were near it. The fosse, when unbroken and its inner wall entire, was probably crossed by a foot-bridge, to be withdrawn on the approach of an enemy; and the little gap by which I had entered closed, so that this must have been a place of great security. The Samoan natives, as far as I have been able to learn, have no tradition of what people inhabited this mountain fastness. At the upper end of the plateau was a broken reservoir, which had been fed from springs by a stone channel. I followed the course of the brook for a few hundred yards until I found it to disappear in a sheet of spray over the edge of a frightful precipice. No food-bearing trees were to be found here. There could not have been more than a few acres (perhaps twenty) in the whole plateau. The mystery was what the people could have lived upon. They could not have been at peace with their neighbours, or whence the necessity for these strong defences. They must have been numerous, from their works which remain.

“The path was paved, and plainly visible. Beyond the spring the ridge became steep and narrow for a distance, and then widened out into another flat. Here were a great number of cairns of stone, - 141 apparently graves disposed in rows among huge trees, the unlifting roots of which had overturned and destroyed very many of them. There was one great Banyan tree which I approached, and, perceiving a cavity entered. The darkness was profound. Tall creepers, which twined themselves about the columned trunks, and lay in masses upon the summit of this giant tree, trailed in waving festoons on every side, and excluded even the faintest glimmer of the feeble twilight which prevailed in the sombre forest. I kindled a flame and explored the interior. Some large bats flew out from an inner chamber, or cell, about ten feet square. The floor was of flat stones, the walls of enormous blocks of the same placed on end; the roof, of intertwisted trunks of the Banyan, which had grown together into a solid arch. In the centre was a cairn, or a cromlech, about four feet high, formed of several stones, arranged in a triangle with a great flat slab on the top. Upon it was what appeared to be another small stone, but on examination turned out to be a great conch shell, white with age, and incrusted with moss and dead animalculae. The atmosphere of this vault was heavy and oppressive, the light burned with difficulty, and the smoke was unable to rise, but rolled low down out of the entrance in a dense serpentine volume. A great koviu, or land crab (Birgus latro), sat perched upon an angle of the wall, regarding me sideways with a look of great malignity as he struck his bony claws with the sound of a hammer on the stone, like some sinister spirit-rapper holding communion with the manes of the departed.

“And his eyes had all the seeming
Of a demon what was dreaming,
And the lamplight o'er him streaming
Cast his shadow on the floor.

“Now, what manner of men could have inhabited the stronghold below, and have been laid to rest in this woodland necropolis? For the reception of what noble corpse had they constructed this ancient sepulchre? Its antiquity was manifestly great, from the Banyan having grown around and over it. The enclosure had first been erected without a roof, the tree (perhaps purposely planted), whose age was beyond estimation, had afterwards enveloped and preserved it. Nay, it would even have altogether and for ever enclosed it in its hollow base had it not been that several of the great slabs which formed the entrance had been forced together at the top, and so retained a passage. (I have seen idol temples in the East so grown over by Banyan trees which are said to be older than the Mohomedan conquest) That this was the tomb of a man of authority among his tribe there could be no doubt, for they had not interred him under a simple cairn like his fellows—there had been art, and much labour in the manner of his burial. I am well convinced that these remains were the work of a people anterior to the existing race of Samoans. Their origin, like that of many other remarkable relics and ruins in the Pacific, is a part of the great mystery of the Isles—i.e., of the early distribution of man throughout the Polynesian archipelagos. I much regretted that I had neither leisure nor appliances to dig in this place for sculls, so as to have them submitted for examination to some man of science (perhaps some future traveller may act upon this suggestion). Being the first civilised man who had been privileged to examine this singular mausoleum, I inscribed my name (as is the custom of les touristes anglais) upon a conspicuous place; and paying my respects to the great crab, who like a guardian gnome, still kept his sullen vigil, I returned to the outer world.

“Dark as was the cave from whence I had emerged, the forest was scarcely more cheerful in its aspect. All the light which prevailed was a sort of gloaming, dying away into the obscurity of a ‘pillared shade’, but of which the hoary trunk of some great maridi or mamala tree stood forth here and there like a dungeon column

“Massy and grey,
Dim with a dull imprisoned ray.
- 142

“And I stumbled among graves, some huge tumuli, others but three or four stones. Here were, doubtless, the bones of many generations. Whatsoever had been their deeds, the very knowledge of them was lost. With them indeed was ‘no remembrance of the wise man any more than the fool for ever’. King and counsellor, spearman and slinger, friend and foe, all alike had gone to eternal oblivion.

“Hic motus animarum, atque haec certamina tanta
Pulveris exigui jactu compressa quiescent”.

The writer was resident in Apia, Western Samoa, from 1940 to 1943 and was able to make an attempt to re-discover the “remains” so vividly described by Sterndale. Unfortunately Sterndale does not give specific information regarding the situation of the “remains”. It became necessary, therefore, to make certain deductions from his description. The one geographical feature which Sterndale mentions by name is the Vaivase river; 23 and from this, it would seem clear that the “great volcanic cone” refers to Le Pue, a peak of 3,600 feet lying inland from Vaimaunga, while the “point . . . accustomed to be crossed by the natives” would be the Tiavi pass, over which the track from Apia to Safata and Siumu crosses. The area thus demarcated is extensive, and to traverse it in its entirety would be a considerable undertaking. It was determined, however, to make a traverse of the Vaivase river. This was commenced at the mouth of the river. The terrain proved difficult—a series of formidable gorges and falls being encountered—and four separate excursions, the last of which involved benightment, were necessary before the traverse was completed. A constant watch was kept for the features enumerated by Sterndale—but without success. Subsequently ascents were made of several of the ridges which converge upon Le Pue and traverses made of the Fangali'i and Soanga streams, but once more no trace was found of Sterndale's “ruins”. Direct search having failed it was decided to conduct inquiries among the natives of Vaimaunga district, many of whom have become well conversant with the area concerned in the course of hunting wild pigs. Many Samoans, some of whom displayed remarkably accurate knowledge, were questioned, but the only thing mentioned which in any way resembled the “remains” described by Sterndale was the Fale o le Fe'e.

At this juncture one was compelled to raise the question of the authenticity of Sterndale's material, and an attempt was made to discover if Sterndale had published any other statements upon Samoan archaeology. Investigation disclosed that another statement had been made by Sterndale in an article entitled “A Lone Land and They Who Lived On It”, which appeared in the Monthly Review, Wellington, of 1890. The relevant passage reads as follows:—

“Remains of like description, but of ruder construction, exist in the interior of the Samoan Islands; there, also, upon the Island of Upolu, is a Druidical circle of a corresponding form to those of Britain and Scandinavia. Behind the settlement of Apia, and at a distance inland of about seven miles in a straight line, is a stupendous waterfall consisting of the whole body of a large stream called Singano, tumbling over the face of a sheer precipice of 200 feet in height. This forms to mariners a very conspicuous landmark, being visible to a great distance from the coast. The summit of the cascade is about 2,000 feet above the level of the sea. It is accessible by a laborious pathway leading up the face of a cliff; formerly well frequented by the Natives, but now choked with brambles and almost obliterated. At a short distance above the fall the river forks into two branches, leaving between them a space of level land overgrown with grass and - 143 shrubs. There are some breadfruit trees, which, wherever found, may be accepted as proof that there has been at one time a human habitation. In the midst is the monument to which I refer. The diameter of the circle is about 40 feet; the pillars are of basalt, very ancient, no doubt, as some have fallen from the perpendicular, and all have been partially buried by the accumulation round them of vegetable mould. They appear to be about 12 feet long, and to have been connected by others laid horizontally upon their tops, in the orthodox Druidical manner. In the centre were two others, taller than the rest. According to Native tradition it was the intention of the builders to have filled in the spaces between the upright pillars with blocks of wrought coral, for which purpose a quantity of that stone was carried up to this mountain solitude from the sea-shore, but, a quarrel arising among the artificers, they abandoned that intent, and threw all the coral into the river close by. Such a tale seems absolutely incredible, forasmuch as between this place and the nearest point where the coral stone could be procured are at least seven miles of frightful ravines, torrents, and dense forests, as also a precipice to be scaled of 200 feet high; nevertheless the coral was conveyed there, and was thrown into the river as the tradition affirms, since there stands upon the bank one mass of it about the height of a man, and four feet thick, and in the clear bed of the stream below are to be seen some tons of the remainder. Of course the story goes that the work was performed by supernatural agency, by order of a powerful wizard; and the sequence of its incidents resembles very circumstantially the Scandinavian story of Esbern Snare, who built a church in Lund, by the labour of a ‘troll’ wherein should be so many pillars each one of stone, and concerning the which service to be faithfully performed ‘they made a paction between them two,’ to the effect that should Snare, on the completion of the work, have failed to discover the name of his assistant, he should pay the forfeit of his heart and eyes to be torn out and disposed of at the pleasure of the malignant ‘troll’. But it came to pass that before the stipulated time Esbern Snare having accidently obtained the needful information imprudently made display of his knowledge, whereupon the ‘troll’ struck work and would carry no more stone, and so the edifice remains unfinished to this day. This legend with a change of names and some increase in the number of characters is almost identical with the Samoan account of the building of this singular monument which they call Fale-o-le-Feē, ie. The House of the Spirit. The occurence of the word feē involves a coincidence not unworthy of attention. It is found with the same sound and meaning in several languages of the Old World. It is not too much to say that a mind skilled like that of Max Muller in the unravelling of ethnological enigmas might deduce important data from the comparison of many apparently absurd and barbarous Polynesian myths with those of the earliest progenitors of the people of Europe.”

Concerning this second account there can be no doubts: it is a garbled and highly inaccurate description of the Fale o le Fe'e. Sterndale's initial and major error is in locating his “Druidical circle” in the fork above the Afutapu fall (see foregoing map of the source-waters of the Vaisingano river). Despite the obvious nature of Sterndale's mistake it was decided in the interests of certainty to visit the position he so graphically describes. The confluence above the Afutapu fall was reached—not without difficulty—but no trace of any “monument” was discovered. Considering the magnitude of Sterndale's error it would seem probable that he never actually visited the Fale o le Fe'e (which is situated, of course, in the source-waters of the Soaga stream); but rather constructed his account from information derived from others.

To what extent then may one depend upon Sterndale's account of the “ruins” in the Vaivase valley? Is it possible that here also we - 144 are confronted with an exaggerated account of the Fale o le Fe'e? Te Rangi Hiroa (Samoan Material Culture, Hawaii, 1930, p. 328) in discussing Sterndale's report makes the following comment:—

“The traveller writing popular articles is liable to give his imagination full play and is not in the frame of mind to distinguish between the stones artificially arranged by man and the huge masses moved by nature. Identification takes place and man is credited with doing the impossible by some secret of engineering skill now supposed to be lost.”

To conclude; it may well be true that the remains described by Sterndale do exist, and it is not impossible that they have so far defied all attempts at re-discovery—it is to be hoped that others resident in Samoa may be “stimulated to make enquiry and research” —but meanwhile, there would seem to be little justification for placing too strong a credence in Sterndale's account. This much is at least certain: A megalithic culture that preceeded the Samoan cannot be seriously entertained unless supported by further data that can be checked.

1   Another name common among the Samoans is Fale pou ma'a—The House with Stone Posts; clearly merely a descriptive term. Rivers (The History of Melanesian Society, London, 1914, p. 368) gives the name as being Falipouma, which is a misspelling of the correct term, Fale pou ma'a.
Le Fale o le fe'e=Maori Te Whare o te wheke (the house of the octopus.)
2   alia=Maori aria; 1, deep water between two shoals; 2, open space among rocks; 3, pool on shore filled only at high water; 4, bay, or deep pool in a river; etc.
3   Stair reported only thirteen pillars standing, while Pritchard says he found eighteen. E. W. Gurr (N.Z. Samoa Guardian, vol. 1, no. 20, September 19, 1929) states that about 1903, Asi, a chief of Matafangatele (a village of the Vaimaunga district) brought some of the stone pillars from the Fale o le Fe'e down to his village to form permanent posts in a new house he was building. Interrogation of the chiefs of Matafangatele corroborated this story. They stated that one of the pillars had been used as the pou o le matua tala (the post in the centre of the rounded portion of a Samoan fale, and of great ceremonial importance, for it is there that the high chief of a village sits during kava ceremonies, etc. A fragment of the stone column may still be seen surmounting Asi's grave (1943).
4   Te Rangi Hiroa (Samoan Material Culture, Hawaii, 1930, p. 328), gives the name of “The Seat of Sina”; he adds, however, that he believes this name to be a “recent embellishment”. This name is incorrect. Sina the legendary figure who was pursued by an eel (tuna) from whose head grew the first cocount tree, has no connection with the Fale o le Fe'e. Te Rangi Hiroa must be considered as having been misinformed in this matter.
nofoa=Maori nohoanga
5   A photograph of the Nofoa o le Fe'e appears in Te Rangi Hiroa's work Samoan Material Culture (plate 35 C). At some time since the photograph was taken (1928) six pillars from the main ruins have been grouped about the ‘seat’ by visitors; I have shown these as I found them (1943), there being no way of determining their original position (fig. 3).
6   Turner (Samoa a Hundred Years Ago and Long Before, London, 1884, p. 28), writes of “a chief called Tapu'a'au, who swam hither from Fiji with his cuttle-fish”.
7   A variant form of this name—Sina-le-lalafa, was also recorded.
8   So'oalo later became a notorious cannibal chief. Cf. Churchill (Samoa 'Uma, New York, 1902, p. 206), “So'oalo, who nets men and women from his tia, or stone platform. . . . His body is buried on the very summit of mount Vaea, but his soul has never found rest”.
9   tufunga=Maori tohunga, an expert; tohunga-whakairo, an expert in applying ornament, as in carving, tattooing, painting, weaving; tohunga-moko, expert in tattooing; etc.; an ordinary builder was kai-hanga.
10   Some of my informants suggested that this woman may have been the Fe'e's wife.
11   Siapo is the Samoan for tapa cloth.
12   Percy Smith (Hawaiki, Wellington, 1910, p. 160) records the following Rarotongan tradition:—
“Is is clear however that Tinirau was an historical personage and the Maoris trace descent from him. He was a chief of great power and beauty and of great fame in ancient days; whilst numerous wonders were due to his action. He possessed a famous fish-pond in Upolu, and it was in Upolu also that Ari built his house of which the pillars were stone, as were the rafters, whilst a stream flowed through it. Ari has been shown to be contemporary with Tui-tarangi (circa A.D. 450) and here he is accredited with being the builder of what I believe to be Le Fale o le Fe'e, situated in the mountains behind, Apia, Upolu.”
I was unable to obtain any corroboration of this legend from the Samoans.
13   A specimen of limestone from the Fale o le Fe'e was submitted to Dr. C. O. Hutton, of the Geological Survey Department, Wellington, N.Z. He identified it as being travertine.
14   Stair (O le Fale o le Fe'e: or, Ruins of an Old Samoan Temple, J.P.S., vol. 3 [1894], 241) speaks of the Fale o le Fe'e as being the temple of the war-god of A'ana. I was able to find no support for this statement in Samoa. The district of A'ana lies at the western end of the island of Upolu at a distance of approximately twelve miles from Vaimaunga. This and other evidence I believe clearly shows Stair to have been in error on this point.
Note: lulu (owl) =M. ruru; matu'u (heron) =M. matuku (ordinarily bittern; but matuku hurepo=M. of the swamp, to distinguish it from matuku-moana, blue-heron); ti'otala (kingfisher) =M. kotare (or te kohare); ve'a (rail)=M. weka; manuali'i (swamp-hen) =M. pukeko—manu ariki would be chief of birds; pusi (sea-eel) =M. puhi, a large variety of eel; pe'a (flying-fox) =M. pekapeka, bat.
15   This is the Dorfteil of Kraemer.
16   This is the Dorfschaft of Kraemer.
17   This is a recent name. Alamangoto village was formerly known as Matafele, and was situated where the commercial centre of Apia township now lies.
18   Cf. Turner (Samoa a Hundred Years Ago . . . , London, 1884, p. 31) who states that in time of war, the war-god Fe'e “sent a branch drifting down the river as a good omen, and a sign to the people that they might go on with the war, sure of driving the enemy”.
19   taula aitu: in Maori a taura or tauira was a teacher, or skilled person, or a pupil, particularly one under instruction by a tohunga: aitu in Maori=demon.
20   Fale was the name of the priest (taula aitu).
21   Cf. Fornander (An Account of the Polynesian Race, London, 1890, vol. 1, p. 51): “The Hawaiian tradition of Pele, the dreaded goddess of volcanic fires, is analagous to the Samoan Fe'e”. (This is my foot-note, J.D.F.)
22   Gods incarnate in the form of the cuttle-fish have been recorded in the following areas of Western Polynesia: Samoa, Tonga, Ellice islands, Tokelau islands, Tongareva, and Tikopia.
23   Sterndale gives Vai-Vasa which is incorrect.