In considering the native dances it must be remembered that the accompanying songs are essential elements of the entertainment: they mark the character of the dance; and equally, in considering the songs, it must be remembered that the imagination of the native never goes beyond the relation of the sexes. The Indian’s poetry is an inverted form of romanticism. Instead of seeking to give rhythmical expression to an idealisation, to find in the beauties of Nature an analogy to the realities of Life, he reverses the process. For instance, he views a ripe fruit, and it only suggests to him a pregnant woman. In all such natural phenomena as he recognises he notes but the crude, if possibly the scientific, origin. In the most ordinary conversation he refers to conditions that appear indecent in common print; they are, however, undetachable from him.

So it is that in his songs he debases idealism, does not elevate realism. His poetry is on a par with that of the music-hall comedian who conceals a mass of filth under avowedly innocent words—but the intention is very different. The Indian possesses no other verbal vehicle, knows no other source of inspiration. His imagination is bound by his vocabulary, as his vocabulary is limited by his imagination. Curiously enough the effect upon his audience is gained by the same means as those employed by the red-nosed singer in the places of entertainment south of the Bridges, and is almost identical in degree. Some of the Londoners of the County Council schools have advanced ethically but little beyond their naked brothers of the Amazonian bush.

These Indians cannot be said to love music for its own sake. The use of music in any form is almost entirely ceremonial. They neither sing nor play instruments as a rule merely for pleasure. On the occasions of their festivals and dances, though, they give evidence of the possession of voices of considerable flexibility. They also display much ingenuity in the manufacture of their instruments, and, next to their weapons, the pan-pipes, flutes, and drums are most carefully fashioned and preserved. In fact, these take precedence over all domestic implements, and even most ornaments.

The native singing voice is loud, high, and shrill. The male leader—as a rule it is a man who is appointed, and he may be any one who knows the old songs—sings the solo, to give the chorus their cue, in a high falsetto which is very penetrating, and marks both time and tune for the others to follow in canon. The song is started softly, and gradually increases both in volume and speed. According to the circumstances, the subject, and the occasion, the men sing alone, the women sing alone, or the men and women combine as in the tribal dances. Most of the singing is done in unison, with a regular drone accompaniment from those not actually articulating the words. The songs are sung in regular time, to the accompaniment of stamping, but not with hand-clapping. The melodies are simple, and in the definite tribal songs consist of little more than a single phrase that seems to admit of no variation, and is repeated ad libitum, as, for example, Mariana Keibeio, a Boro tribal song. The tune of this, notated from memory, and in part from a phonograph record, runs approximately, so far as it can be rendered in our notation:

What this implies no Indian now knows, for with all tribal songs the natives offer no explanation of their meaning or their origin. They are the songs that their fathers sang, and one can find no evidence of the amendation or emendation of the score on the part of their descendants. These tribal lays are so old that the words are obsolete and no longer understood by the singers; what is of importance is the rhythm, and to that, as is common with uncivilised peoples, the music is largely subordinated. It is but an accompaniment to the dancing. “The sense of time” in the Indian, as Stevenson noted among the South Sea Islanders, “is extremely perfect,” and one might complete the quotation and add, “I conceive in such a festival that almost every sound and movement fell in one.”[335] It is not an easy matter to discuss, because the English and the Indian standpoint are so diametrically opposite. So far as I could judge the tunes are usually in a minor key, both melody and harmony being of the simplest.

There are no love-songs among the Indians, for the poetic conception of love does not exist. Sacred songs and nursery songs are equally lacking. A mother never croons to her baby; she does not understand a lullaby. War-songs are merely the expression of the war-dance; they depend for their significance upon the words and for their ferocity upon the grim accentuation of the chorus.

At the time of the harvest of pine-apples, when the great dance is held, the men sing the challenge, and the women reply in their own defence. The songs are similar to that sung at the manioc-gathering dance, and I have previously tried to give some idea of such a song.

Apart from the traditional songs of the tribes, which are sacred and unchangeable, the Indians are very fond of a form of song which is really a game rather than a musical effusion. More correctly, perhaps, it should be called a ballad.[336] A leader of acknowledged fertility of imagination and fluency of expression is appointed, as for the Muenane riddle dance, and will collect the members of the tribe for what is actually an impromptu dance. He, or she, will chant to an improvised air with a simple rhythm, while the chorus repeat each line or its burden as a refrain. Such songs give opportunity for all the wit of the tribe. They are designed either to honour or to ridicule the subject of the ballad. In reality a composition of this description takes hours to sing. The first wit propounds the question, the chorus repeat it, and the second wit then suggests the answer, which is again repeated by all amid much laughter, and the repetition is continued not once but twenty times, until the first wit breaks in with a new query. This is a very favourite game among the women.

The following is an attempt to suggest the song-words of a dance performed by some Witoto for my benefit, though I do the Indians too much justice, give too great an idea of continuity, in this version. There is no cohesion in their productions, and reiteration is the salient feature of all. The sound and the rhythm suggested to me at the time the metre of Hiawatha, so I give this song in an attempt at Hiawathian measure. But the adaptation is really too varied for the Indian original. I was outside the maloka when the women started—no men took part—and they danced in front of me. After a time I went inside, and the performers promptly followed me, and continued to dance in the central space of the house. Naturally not one word would have been sung if these dancers had known it would be interpreted to me.

This went on ad nauseam. The true object in all such songs is to bring in and discuss sexual matters, and no song has advanced far before it has become essentially carnal in idea and thoroughly licentious in expression.

Although instruments are always employed at the dances they do not seem to be introduced with any idea of organised accompaniment, but only to help swell the body of sound. The natives, being ignorant of the use of metal, have been forced to make their instruments entirely of vegetable substances; the only other material used is bone, human bone, bien entendu, and judging from a specimen presented by Robuchon to the British Museum, the shell of a small land tortoise. Their instruments of percussion are drums, castanets, and rattles: their wind-instruments are flutes and pan-pipes. Very rarely a solitary Indian may be found playing the flute, apparently for his personal amusement and solace. As a rule, it is merely used in combination with its fellows to increase the volume of sound without heed to its proper place in harmony.

The pan-pipes are the simplest of all instruments of Amazonian music to make, and are the most universally popular. They consist of a bundle of reeds—three, five, six, seven, ten, or even seventeen in number—bound together with palm-fibre, or, on the Napo, with finely split cane. Although the pipes are cut to lengths yielding the necessary musical intervals, the number seems to be purely arbitrary. They are used in concert with all other instruments, and mark so much of tune as the Indian orchestra strives to attain. The pan-pipes shown in the accompanying illustration are Witoto instruments contrasted with the neater finish of one made on the Napo. The latter has the greater number of pipes, and all relatively smaller. There is nothing complicated about the make of either set. The cane pipes are cut immediately below the natural joint, and the node is thus made to serve as a stop.[337]

The ubiquitous bamboo also furnishes the material for a larger flute, and flutes or fifes are made out of the arm-bones of prisoners taken in battle. After the victim is killed and eaten the humerus is cleaned, its extremities opened, and the soft matrix scooped out. Finger-holes are bored in the shaft of the bone, usually three in number, but occasionally five. When human bones are not forthcoming the tribesman uses the leg-bone of a jaguar. This is opened at the end and furnished with a wax stop that leaves a small canal open to a three-cornered air-hole. Occasionally one of these flutes is made with both ends open, in which case a square or semicircular hole is cut out from the upper rim. The flute is held against the lower lip, and commonly has three, or more rarely four, sound-holes. Flutes are also made of heron-bones, open at the lower end, with a square air-hole, and generally four sound-holes. These have mouthpieces made of leaves, and their tones are exceedingly shrill. But the most curious instruments of which I have note are flutes made from skulls of animals, by covering them with pitch, and only leaving open the holes of the nose and the occipital bone. One hole is blown through, the other is the sounding-hole. Many of the Indian instruments, especially the bone flutes, are gaily ornamented with elaborate incised patterns that are dyed black and red with vegetable extracts. The flutes are also adorned with tassels of cotton or palm-fibre.

The flute or fife is played from the extremity that is rudely fashioned into a mouthpiece. No native trumpets are provided with sliding tubes like the familiar trombone, and there is no plug in the mouth-hole. Nor are any of the Amazonian wind-instruments fitted with a vibrating reed. There are no bagpipes, and, in the regions I traversed, no stringed instruments. Certain tribes north of the Japura, notably the Desana, use whistles made of clay, which they employ both as alarm signals and as adjuncts to the dance.

Trumpets of bark and bamboo have an irregular distribution. Many tribes dispense with them on all ordinary occasions, and confine their use to Jurupari music. These sacred instruments constitute one of the most profound mysteries of the Amazon. They are lengthy affairs, made from the hollow stem of a palm, and fitted with a trumpet mouthpiece. The note is akin to that of the bassoon. These trumpets are tribal possessions, and are kept concealed at a distance from the maloka, in a hut which the women are never permitted to enter, and where the various secret paraphernalia connected with boy initiation—such as the whips of tapir hide—are stored. It is a capital crime for any woman even to set eyes upon them. The Jurupari trumpet is as tabu to Indian women as the bull-roarer of the Australian native is to his women-folk.[338] The Indian girls are brought up in the belief that the music of the trumpets is an essential element in the exorcism of the evil spirit from the body of the youthful initiate, and that any interference on their part must lead to the eternal residence of such spirit in the novice, to the consequent disaster of the tribe, and this belief holds good all their lives.[339] No sooner is Jurupari music heard approaching the maloka than all the women and uninitiated hurry to the bush, and remain in hiding until the ceremony is concluded and the trumpets have been returned to their tabernacle. What the ceremony may be is held a profound secret, and the punishment for infringement is death.[340] As a rule two of these sacred trumpets are used, and they are tuned to the same pitch, though differing in their tone according to their length. They are only used north of the Japura; south of that river the tribes have no Jurupari music and only know them as employed ceremonially by their neighbours in connection with initiation secrets to frighten their women.

The Tukana when dancing use a trumpet alternately with their rattles; and the Indians north of the Japura have regular castanets, made of blocks of hard wood, which are manipulated with one hand, much in the manner that the nigger-minstrel plays the “bones.” All the tribes make rattles of small gourds by the simple method of partly filling the calabash with dried seeds, or fruit stones, and inserting a wooden handle so that they can be shaken in time to the dance. Some of these are of the roughest, the stick of the handle quite untrimmed; others are more finely finished, and the polished black surface of the gourd may be ornamented with designs in colour, or incised patterns. But these are by no means the only rattles used at a dance. The Indians have them of many kinds and descriptions. The smaller are worn as armlets, wristlets, leglets, and anklets. These are made of nuts, strung with coloured beads on palm fibre, and very carefully fashioned. The leg rattles are frequently handsome ornaments, the rich brown of the glossy nutshell making a splendid contrast with the blue or red of the Brummagem beads. The finest are made from a nut not unlike the Brazil nut of commerce in shape, but less angular. That shown in Plate XLIII. has natural groovings and marks which give the polished sections the appearance of being engraved. A section of the shell is cut off, thoroughly cleaned and polished, then attached by a short string of beads to the main leg- or arm-band from which these nut sections hang bell-like. The arm rattles are made of smaller nuts, some are not unlike an oval hazelnut, flat on one side, cut in half and highly polished. The nut is, roughly speaking, some three-quarters of an inch across and long. These also are hung on threads of beads pendant a quarter of an inch apart from the connecting beaded string. Leg rattles are made of larger nuts, and one variety is made in the form of a bunch, not a band or chain. The beads used for these are blue and red in colour, and the bunch of nuts on their beaded strings is fastened with plaited palm-fibre beneath the knee. The whole effect is most distinctly ornamental. The jangle of two or three of these nutshell bells is not unpleasant: there is almost a tinkle in their clatter, but the volume of sound obtainable from a number of them is remarkable, and so is the precision with which they accentuate the rhythm of movement.

The Indians have no cymbals, gongs, or bells; but the drum is an important factor not only in native music, but in native life. The drum is the telegraph of the Amazons. In fact, the most remarkable of all the native instruments is the manguare or signal drum. Although the primary use of this drum is to signal, it is utilised on great occasions as an addition to the aboriginal orchestra. To make this important adjunct of the maloka two blocks of hard wood are chosen, some six feet in length, and about twenty-four inches in diameter. These blocks are very carefully hollowed out by means of heated stones that are introduced through a narrow longitudinal slit, and char the interior. Instead of endeavouring, however, as would be the case with an ordinary drum, to contrive as nearly perfect a cylinder as possible, the object of the signal-drum maker is to obtain a husk of varying thicknesses, so as to secure differences in note. Accordingly, with his rude implements, hot stones, capybara-tooth borer, and stone axe, he fashions the interior of the drum in such a manner that the outer shell, the sounding-board, varies in thickness from half an inch to four inches. Two blocks are used; the smaller is called the male, and the larger the female. The ends are simply the wood of the tree which is not removed, all the hollowing being accomplished by means of the grooved slit. When finished these are suspended by withes at an oblique angle, one end much higher than the other—say six feet and three feet respectively from the ground. They hang from the rafters of the maloka, or from an upright frame, and present the appearance of two barrels surmounted by a narrow slit.[341]

The musician takes his stand between these drums and, with a wooden mallet headed with a knob of rubber, beats out his message or his tune. Altogether he has a range of four notes—two low ones on the female manguare, and two high ones on the male. On these he rings the changes with great rapidity, and produces a sound which, though not startlingly loud, has such penetrating qualities that it can be heard twenty miles away. He beats very quickly in short and long strokes, not unlike the Morse Code. By means of the manguare a skilled signaller can carry on a conversation as accurately as a telegraph operator at St. Martin’s-le-Grand, or a soldier with a heliograph—but how he does it is another secret of the Amazonian bush. When used for its proper purpose as signal drum, the Boro and the Okaina can carry on conversations upon almost any subject within their ken. Other tribes are only able to distinguish between a warning of danger and an invitation to a dance. Brown could use the drum for small matters—he could hurry the bearers out of the bush for example. He said there was no code, but that the signaller tried to represent the sound of words with the drum, and Indians invariably told me that they made the words with the drum. However, with a language dependent on inflection, as theirs unquestionably is, there must be a code of some description.

India-rubber, which has added a new and awful terror to the life of the forest Indian, is only employed by these tribes to make the drum mallet, used with the manguare, and the latex for depilatory purposes. The Witoto call the mallet ouaki, the drum is hugwe.

These great signal drums have designs worked upon them in which the organs associated with the presumed sex of the instrument are prominent; and, after the manner of the natives, both instruments are invariably distinguished internally with the proper sexual characters, the female drum having two breasts pendant inside.

Even in the construction of a small playing drum much time and ingenuity are expended. First an aeta palm[342] is selected, cut down, and a section of the trunk laboriously hacked off. This section in turn is carefully hollowed, until only a thin shell remains. Some tribes use a section of bamboo in place of the hollowed palm, but these never secure so fine an instrument or so fine a note as the palm trunk makes. Over the two ends of the cylinder dried monkey skin is tightly stretched—preferably that of the howler monkey, as it is popularly supposed to produce a louder and more rolling sound. Some tribes then fasten across one end of this drum a very tight cord, into the centre of which has been tied a fine sliver of wood. By this means two notes are obtained—the open note where nothing interferes with the vibrations of the drumhead, and the closed note where the vibrations of the splinter intersect those of the skin. A very inferior instrument is made with agouti skin over a bamboo cylinder. The drums made on the Napo River look very much like an English child’s toy drum, rather high and narrow, and, of course, made entirely without metal. The sides bulge slightly, and have crossed threads of fibre string. The vellum of the drumhead is kept in its place tautly by a close-fitting ring. These drums are usually decorated, and are objects of barter among many of the tribes. They are played with the fingers only, not with drumsticks or mallet.