Philippine Islands Ethnological Survey Publications

Date: 1905

Show Summary




In all of Igorot culture the most apparent and strikingly noteworthy fact is its agriculture. In agriculture the Igorot has reached his highest development. On agriculture hangs his claim to the rank of barbarian—without it he would be a savage.

Igorot agriculture is unique in Luzon, and, so far as known, throughout the Archipelago, in its mountain terraces and irrigation.

There are three possible explanations of the origin of Philippine rice terraces. First, that they (and those of other islands peopled by primitive and modern Malayans, and those of Japan and China) are indigenous—the product of the mountain lands of each isolated area; second, that most of them are due to cultural influences from one center, or possibly more than one center, to the north of Luzon—as influences from China or Japan spreading southward from island to island; third, that they, especially all those of the islands—excluding only China—are due to influences originating south of the Philippines, spreading northward from island to island.

Terracing may be indigenous to many isolated areas where it is found, and doubtless is to some; it is found more or less marked wherever irrigation is or was practised in ancient or modern agriculture. However, it is believed not to be an original production of the Philippines. Certain it is that it is not a Negrito art, nor does it belong to the Moro or to the so-called Christian people.

Different sections of China have rice terraces, and as early as the thirteenth century Chinese merchants traded with the Philippines, yet there is no record that they traded north of Manila—where terracing is alone found. Besides, the Chinese record of the early commerce with the Islands—written by Chao Jukua about 1250 it is claimed—specifically states that the natives of the Islands were the merchants, taking the goods from the shore and trading them even to other islands; the Chinese did not pass inland. Even though the Chinaman brought phases of his culture to the Islands, it would not have been agriculture, since he did not practice it here. Moreover, whatever culture he did leave would not be found in the mountains three or four days inland, while the people with whom he traded were without the art. The same arguments hold against the Japanese as the inspirers of Igorot terraces. There is no record that they traded in the islands as early as did the Chinese, and it is safe to say, no matter when they were along the coasts of Luzon, that they never penetrated several days into the mountains, among a wild, head-hunting people, for what the agricultural Igorot had to sell.

The historic cultural movements in Malaysia have been not from the north southward but from Sumatra and Java to the north and east; they have followed the migrations of the people. It is believed that the terrace-building culture of the Asiatic islands for the production of mountain rice by irrigation during the dry season has drawn its inspiration from one source, and that such terraces where found today in Java, Lombok, Luzon, Formosa, and Japan are a survival of a very early culture which spread from the nest of the primitive Malayan stock and left its marks along the way—doubtless in other islands besides these cited. If Japan, as has Formosa, had an early Malayan culture, as will probably be proved in due time, one should not be surprised to find old rice terraces in the mountains of Batanes Islands and the Leo Choo Islands which lie between Luzon and Japan.


It must be noted here that all Bontoc agricultural labors, from the building of the sementera to the storing of the gathered harvest, are accompanied by religious ceremonials. They are often elaborate, and some occupy a week’s time. These ceremonials are left out of this chapter to avoid detail; they appear in the later chapter on religion.

There are two varieties of sementeras—garden patches, called "payyö"—in the Bontoc area, the irrigated and the unirrigated. The irrigated sementeras grow two crops annually, one of rice by irrigation during the dry season and the other of camotes, "sweet potatoes," grown in the rainy season without irrigation. The unirrigated sementera is of two kinds. One is the mountain or side hill plat of earth, in which camotes, millet, beans, maize, etc., are planted, and the other is the horizontal plat (probably once an irrigated sementera), usually built with low terraces, sometimes lying in the pueblo among the houses, from which shoots are taken for transplanting in the distant sementeras and where camotes are grown for the pigs. Sometimes they are along old water courses which no longer flow during the dry season; such are often employed for rice during the rainy season.

The unirrigated mountain-side sementera, called "fo-ag’," is built by simply clearing the trees and brush from a mountain plat. No effort is made to level it and no dike walls are built. Now and then one is hemmed in by a low boundary wall.

The irrigated sementeras are built with much care and labor. The earth is first cleared; the soil is carefully removed and placed in a pile; the rocks are dug out; the ground shaped, being excavated and filled until a level results. This task for a man whose only tools are sticks is no slight one. A huge boulder in the ground means hours—often days—of patient, animal-like digging and prying with hands and sticks before it is finally dislodged. When the ground is leveled the soil is put back over the plat, and very often is supplemented with other rich soil. These irrigated sementeras are built along water courses or in such places as can be reached by turning running water to them. Inasmuch as the water must flow from one to another, there are practically no two semen-texas on the same level which are irrigated from the same water course. The result is that every plat is upheld on its lower side, and usually on one or both ends, by a terrace wall. Much of the mountain land is well supplied with boulders and there is an endless water-worn supply in the beds of all streams. All terrace walls are built of these undressed stones piled together without cement or earth. These walls are called "fa-nĭng’." These are from 1 to 20 and 30 feet high and from a foot to 18 inches wide at the top. The upper surface of the top layer of stones is quite flat and becomes the path among the sementeras. The toiler ascends and descends among the terraces on stone steps made by single rocks projecting from the outside of the wall at regular intervals and at an angle easy of ascent and descent.

These stone walls are usually weeded perfectly clean at least once each year, generally at the time the sementera is prepared for transplanting. This work falls to the women, who commonly perform it entirely nude. At times a scanty front and back apron of leaves is worn tucked under the girdle.

In the Banawi [Ifugao] district, south of the Bontoc area, there are terrace walls certainly 75 feet in height, though many of these are not stones, since the earth is of such a nature that it does not readily crumble.

It is safe to say that nine-tenths of the available water supply of the dry season in the Bontoc area is utilized for irrigation. In some areas, as about Bontoc pueblo, there is practically not a gallon of unused water where there is space for a sementera.

A single area consisting of several thousand acres of mountain side is frequently devoted to sementeras, and I have yet to behold a more beautiful view of cultivated land than such an area of Igorot rice terraces. Winding in and out, following every projection, dipping into every pocket of the mountain, the walls ramble along like running things alive. Like giant stairways the terraces lead up and down the mountain side, and, whether the levels are empty, dirt-colored areas, fresh, green-carpeted stairs, or patches of ripening, yellow grain, the beholder is struck with the beauty of the artificial landscape and marvels at the industry of an otherwise savage people.


By irrigation is meant the purposeful distribution of water over soil by man by means of diverting streams or by the use of canals in the shape of ditches or troughs for conveying and directing part of a water supply, or by means of some other man-directed power to raise water to the required level.

The Igorot employ three methods of irrigation: One, the simplest and most natural, is to build sementeras along a small stream which is turned into the upper cementera and passes from one to another, falling from terrace to terrace until all the water is absorbed, evaporated, or all available or desired land is irrigated. Usually such streams are diverted from their courses, and they are often carried long distances out of their natural way. The second method is to divert a part of a river by means of a stone dam. The third method is still more artificial than the preceding—the water is lifted by direct human power from below the sementera and poured to run over the surface.

The first method is the most common, since the mountains in Igorot land are full of small, usually perpetual, streams. There are practically no streams within reach of suitable pueblo sites which are not exhausted by the Igorot agriculturist. Everywhere small streams are carefully guarded and turned wherever there is a square yard of earth that may be made into a rice sementera. Small streams in some cases have been wound for miles around the sides of a mountain, passing deep gullies and rivers in wooden troughs or tubes.

Much land along the river valleys is irrigated by means of dams, called by the Igorot "lung-ud’." During the season of 1903 there was one dam across the entire river at Bontoc, throwing all the water which did not leak through the stones into a large canal on the Bontoc side of the valley. Half a mile above this was another dam diverting one-half the stream to the same valley, only onto higher ground. Immediately below the main dam were two low piles of stones jutting into the shallow stream from the Bontoc side, and each gathering sufficient water for a few sementeras. Within a quarter of a mile below the main dam were three other loose, open weirs of rocks, two of which began on a shallow island, throwing water to the Samoki side of the river. In the stream a short distance farther down a shallow row Of rocks and gravel turned water into three new sementeras constructed early in the year on a gravel island in the river.

The main dam is about 12 feet high, 2 feet broad at the top, 8 or 10 at the bottom, and is about 300 feet long. It is built each year during November and December, and requires the labor of fifteen or twenty men about six weeks. It is constructed of river-worn boulders piled together without adhesive. The top stones are flat on the upper surface, and the dam is a pathway across the river for the people from the time of its completion until its destruction by the freshets of June or July.

The upper dam is a new piece of primitive engineering. It, with its canal, has been in mind for at least two years; but it was completed only in 1903. The dam is small, extending only half way across the river, and beginning on an island. This dam turns water into a canal averaging three feet wide and carrying about five inches of water. The canal, called "a’-lak," is about 3000 feet long from the dam to the place of discharge into the level area. For about 530 feet of this distance it was impossible for the primitive engineer to construct a canal in the earth, as the solid rock of the mountain dips vertically into the river. About fifty sections of large pine trees were brought and hollowed into troughs, called "ta-la’-kan," which have been secured above the water by means of buttresses, by wooden scaffolding, called "to-kod’," and by attachment to the overhanging rocks, until there is now a continuous artificial waterway from the dam to the tract of irrigated land.

Considerable engineering sense has been shown and no small amount of labor expended in the construction of this last irrigating scheme. The pine logs are a foot or more in diameter, and have a waterway dug in them about ten or twelve inches deep and wide. These trees were felled and the troughs dug with the wasay, a short-handled tool with an iron blade only an inch or an inch and a half wide, and convertible alike into ax and adz.

There seems to be a fall of about twenty-two feet between the upper dam and the discharge from the troughs. This fall in a distance of about 3000 feet seems needlessly great; however, the primitive engineer has shown excellent judgment in the matter. First, by putting the dam (upper dam) where it is, only half the stream had to be built across. Second, there is a rapids immediately below the dam, and, had the Igorot built his dam below the rapids, a dam of the same height would have raised the water to a much lower level; this would have necessitated a canal probably ten or twelve feet deep instead of three. Third, the height of the water at the upper dam has enabled him to lay the log section of the waterway above the high-water mark of the river, thus probably insuring more or less permanence. Had the dam been built much lower down the stream the troughs would have been near the surface of the river and been torn away annually by the freshets, or the people would be obliged each year to tear down and reconstruct that part of the canal. As it now is it is probable that only the short dam will need to be rebuilt each year.

All dams and irrigating canals are built directly by or at the expense of the persons benefited by the water. Water is never rented to persons with sementeras along an artificial waterway. If a person refuses to bear his share of the labor of construction and maintenance his semen-teras must lie idle for lack of water.

All sementera owners along a waterway, whether it is natural or artificial, meet and agree in regard to the division of the water. If there is an abundance, all open and close their sluice gates when they please. When there is not sufficient water for this, a division is made—usually each person takes all the water during a certain period of time. This scheme is supposed to be the best, since the flow should be sufficient fully to flood the entire plat—a 100-gallon flow in two hours is considered much better than an equal flow in two days.

During the irrigating season, if there is lack of water, it becomes necessary for each sementera owner to guard his water rights against other persons on the same creek or canal. If a man sleeps in his house during the period in which his sementeras are supposed to receive water, it is pretty certain that his supply will be stolen, and, since he was not on guard, he has no redress. But should sleep chance to overtake him in his tiresome watch at the sementeras, and should some one turn off and steal his water, the thief will get clubbed if caught, and will forfeit his own share of water when his next period arrives.

The third method of irrigation—lifting the water by direct human power—is not much employed by the Igorot. In the vicinity of Bontoc pueblo there are a few sementeras which were never in a position to be irrigated by running water. They are called "pay-yo’ a kao-u’-chan," and, when planted with rice in the dry season, need to be constantly tended by toilers who bring water to them in pots from the river, creek, or canals. On the Samoki side of the valley during a week or so of the driest weather in May, 1903, there were four "well sweeps," each with a five-gallon kerosene-oil can attached, operating nearly all day, pouring water from a canal into sementeras through sixty or eighty feet of small, wooden troughs.


Related Resources

None available for this document.

Download Options

Title: Philippine Islands Ethnological Survey Publications

Select an option:

*Note: A download may not start for up to 60 seconds.

Email Options

Title: Philippine Islands Ethnological Survey Publications

Select an option:

Email addres:

*Note: It may take up to 60 seconds for for the email to be generated.

Chicago: Philippine Islands Ethnological Survey Publications in Source Book in Anthropology, ed. Kroeber, Alfred L., 1876-1960, and Waterman, T. T. (Berkeley, CA: University of California Press, 1920), Original Sources, accessed December 5, 2023,

MLA: . Philippine Islands Ethnological Survey Publications, Vol. 1, in Source Book in Anthropology, edited by Kroeber, Alfred L., 1876-1960, and Waterman, T. T., Berkeley, CA, University of California Press, 1920, Original Sources. 5 Dec. 2023.

Harvard: , Philippine Islands Ethnological Survey Publications. cited in 1920, Source Book in Anthropology, ed. , University of California Press, Berkeley, CA. Original Sources, retrieved 5 December 2023, from