Inca Land - BestLightNovel.com
You’re reading novel Inca Land Part 13 online at BestLightNovel.com. Please use the follow button to get notification about the latest chapter next time when you visit BestLightNovel.com. Use F11 button to read novel in full-screen(PC only). Drop by anytime you want to read free – fast – latest novel. It’s great if you could leave a comment, share your opinion about the new chapters, new novel with others on the internet. We’ll do our best to bring you the finest, latest novel everyday. Enjoy
It is difficult to describe our feelings as we accepted Saavedra's invitation to make ourselves at home, and sat down to an abundant meal of boiled chicken, rice, and sweet ca.s.sava (manioc). Saavedra gave us to understand that we were not only most welcome to anything he had, but that he would do everything to enable us to see the ruins, which were, it seemed, at Espiritu Pampa, some distance farther down the valley, to be reached only by a hard trail pa.s.sable for barefooted savages, but scarcely available for us unless we chose to go a good part of the distance on hands and knees. The next day, while our carriers were engaged in clearing this trail, Professor Foote collected a large number of insects, including eight new species of moths and b.u.t.terflies.
I inspected Saavedra's plantation. The soil having lain fallow for centuries, and being rich in humus, had produced more sugar cane than he could grind. In addition to this, he had bananas, coffee trees, sweet potatoes, tobacco, and peanuts. Instead of being "a very powerful chief having many Indians under his control"--a kind of "Pooh-Bah"--he was merely a pioneer. In the utter wilderness, far from any neighbors, surrounded by dense forests and a few savages, he had established his home. He was not an Indian potentate, but only a frontiersman, soft-spoken and energetic, an ingenious carpenter and mechanic, a modest Peruvian of the best type.
Owing to the scarcity of arable land he was obliged to cultivate such pampas as he could find--one an alluvial fan near his house, another a natural terrace near the river. Back of the house was a thatched shelter under which he had constructed a little sugar mill. It had a pair of hardwood rollers, each capable of being turned, with much creaking and cracking, by a large, rustic wheel made of roughly hewn timbers fastened together with wooden pins and lashed with thongs, worked by hand and foot power. Since Saavedra had been unable to coax any pack animals over the trail to Conservidayoc he was obliged to depend entirely on his own limited strength and that of his active son, aided by the uncertain and irregular services of such savages as wished to work for sugar, trinkets, or other trade articles. Sometimes the savages seemed to enjoy the fun of climbing on the great creaking treadwheel, as though it were a game. At other times they would disappear in the woods.
Near the mill were some interesting large pots which Saavedra was using in the process of boiling the juice and making crude sugar. He said he had found the pots in the jungle not far away. They had been made by the Incas. Four of them were of the familiar aryballus type. Another was of a closely related form, having a wide mouth, pointed base, single incised, conventionalized, animal-head nubbin attached to the shoulder, and band-shaped handles attached vertically below the median line. Although capable of holding more than ten gallons, this huge pot was intended to be carried on the back and shoulders by means of a rope pa.s.sing through the handles and around the nubbin. Saavedra said that he had found near his house several bottle-shaped cists lined with stones, with a flat stone on top--evidently ancient graves. The bones had entirely disappeared. The cover of one of the graves had been pierced; the hole covered with a thin sheet of beaten silver. He had also found a few stone implements and two or three small bronze Inca axes.
On the pampa, below his house, Saavedra had constructed with infinite labor another sugar mill. It seemed strange that he should have taken the trouble to make two mills; but when one remembered that he had no pack animals and was usually obliged to bring the cane to the mill on his own back and the back of his son, one realized that it was easier, while the cane was growing, to construct a new mill near the cane field than to have to carry the heavy bundles of ripe cane up the hill. He said his hardest task was to get money with which to send his children to school in Cuzco and to pay his taxes. The only way in which he could get any cash was by making chancaca, crude brown sugar, and carrying it on his back, fifty pounds at a time, three hard days'
journey on foot up the mountain to Pampaconas or Vilcabamba, six or seven thousand feet above his little plantation. He said he could usually sell such a load for five soles, equivalent to two dollars and a half! His was certainly a hard lot, but he did not complain, although he smilingly admitted that it was very difficult to keep the trail open, since the jungle grew so fast and the floods in the river continually washed away his little rustic bridges. His chief regret was that as the result of a recent revolution, with which he had had nothing to do, the government had decreed that all firearms should be turned in, and so he had lost the one thing he needed to enable him to get fresh meat in the forest.
------ FIGURE
Saavedra and his Inca Pottery ------
------ FIGURE
Inca Gable at Espiritu Pampa ------
In the clearing near the house we were interested to see a large turkey-like bird, the pava de la montana, glossy black, its most striking feature a high, coral red comb. Although completely at liberty, it seemed to be thoroughly domesticated. It would make an attractive bird for introduction into our Southern States.
Saavedra gave us some very black leaves of native tobacco, which he had cured. An inveterate smoker who tried it in his pipe said it was without exception the strongest stuff he ever had encountered!
So interested did I become in talking with Saavedra, seeing his plantation, and marveling that he should be worried about taxes and have to obey regulations in regard to firearms, I had almost forgotten about the wild Indians. Suddenly our carriers ran toward the house in a great flurry of excitement, shouting that there was a "savage"
in the bushes near by. The "wild man" was very timid, but curiosity finally got the better of fear and he summoned up sufficient courage to accept Saavedra's urgent invitation that he come out and meet us. He proved to be a miserable specimen, suffering from a very bad cold in his head. It has been my good fortune at one time or another to meet primitive folk in various parts of America and the Pacific, but this man was by far the dirtiest and most wretched savage that I have ever seen.
He was dressed in a long, filthy tunic which came nearly to his ankles. It was made of a large square of coa.r.s.ely woven cotton cloth, with a hole in the middle for his head. The sides were st.i.tched up, leaving holes for the arms. His hair was long, unkempt, and matted. He had small, deep-set eyes, cadaverous cheeks, thick lips, and a large mouth. His big toes were unusually long and prehensile. Slung over one shoulder he carried a small knapsack made of coa.r.s.e fiber net. Around his neck hung what at first sight seemed to be a necklace composed of a dozen stout cords securely knotted together. Although I did not see it in use, I was given to understand that when climbing trees, he used this stout loop to fasten his ankles together and thus secure a tighter grip for his feet.
By evening two other savages had come in; a young married man and his little sister. Both had bad colds. Saavedra told us that these Indians were Pichanguerras, a subdivision of the Campa tribe. Saavedra and his son spoke a little of their language, which sounded to our unaccustomed ears like a succession of low grunts, breathings, and gutturals. It was pieced out by signs. The long tunics worn by the men indicated that they had one or more wives. Before marrying they wear very scanty attire--nothing more than a few rags hanging over one shoulder and tied about the waist. The long tunic, a comfortable enough garment to wear during the cold nights, and their only covering, must impede their progress in the jungle; yet they live partly by hunting, using bows and arrows. We learned that these Pichanguerras had run away from the rubber country in the lower valleys; that they found it uncomfortably cold at this alt.i.tude, 4500 feet, but preferred freedom in the higher valleys to serfdom on a rubber estate.
Saavedra said that he had named his plantation Conservidayoc, because it was in truth "a spot where one may be preserved from harm." Such was the home of the potentate from whose abode "no one had been known to return alive."
CHAPTER XV
The Pampa of Ghosts
Two days later we left Conservidayoc for Espiritu Pampa by the trail which Saavedra's son and our Pampaconas Indians had been clearing. We emerged from the thickets near a promontory where there was a fine view down the valley and particularly of a heavily wooded alluvial fan just below us. In it were two or three small clearings and the little oval huts of the savages of Espiritu Pampa, the "Pampa of Ghosts."
On top of the promontory was the ruin of a small, rectangular building of rough stone, once probably an Inca watch-tower. From here to Espiritu Pampa our trail followed an ancient stone stairway, about four feet in width and nearly a third of a mile long. It was built of uncut stones. Possibly it was the work of those soldiers whose chief duty it was to watch from the top of the promontory and who used their spare time making roads. We arrived at the princ.i.p.al clearing just as a heavy thunder-shower began. The huts were empty. Obviously their occupants had seen us coming and had disappeared in the jungle. We hesitated to enter the home of a savage without an invitation, but the terrific downpour overcame our scruples, if not our nervousness. The hut had a steeply pitched roof. Its sides were made of small logs driven endwise into the ground and fastened together with vines. A small fire had been burning on the ground. Near the embers were two old black ollas of Inca origin.
In the little chacra, ca.s.sava, coca, and sweet potatoes were growing in haphazard fas.h.i.+on among charred and fallen tree trunks; a typical milpa farm. In the clearing were the ruins of eighteen or twenty circular houses arranged in an irregular group. We wondered if this could be the "Inca city" which Lopez Torres had reported. Among the ruins we picked up several fragments of Inca pottery. There was nothing Incaic about the buildings. One was rectangular and one was spade-shaped, but all the rest were round. The buildings varied in diameter from fifteen to twenty feet. Each had but a single opening. The walls had tumbled down, but gave no evidence of careful construction. Not far away, in woods which had not yet been cleared by the savages, we found other circular walls. They were still standing to a height of about four feet. If the savages have extended their milpa clearings since our visit, the falling trees have probably spoiled these walls by now. The ancient village probably belonged to a tribe which acknowledged allegiance to the Incas, but the architecture of the buildings gave no indication of their having been constructed by the Incas themselves. We began to wonder whether the "Pampa of Ghosts" really had anything important in store for us. Undoubtedly this alluvial fan had been highly prized in this country of terribly steep hills. It must have been inhabited, off and on, for many centuries. Yet this was not an "Inca city."
While we were wondering whether the Incas themselves ever lived here, there suddenly appeared the naked figure of a st.u.r.dy young savage, armed with a stout bow and long arrows, and wearing a fillet of bamboo. He had been hunting and showed us a bird he had shot. Soon afterwards there came the two adult savages we had met at Saavedra's, accompanied by a cross-eyed friend, all wearing long tunics. They offered to guide us to other ruins. It was very difficult for us to follow their rapid pace. Half an hour's scramble through the jungle brought us to a pampa or natural terrace on the banks of a little tributary of the Pampaconas. They called it Eromboni. Here we found several old artificial terraces and the rough foundations of a long, rectangular building 192 feet by 24 feet. It might have had twenty-four doors, twelve in front and twelve in back, each three and a half feet wide. No lintels were in evidence. The walls were only a foot high. There was very little building material in sight. Apparently the structure had never been completed. Near by was a typical Inca fountain with three stone spouts, or conduits. Two hundred yards beyond the water-carrier's rendezvous, hidden behind a curtain of hanging vines and thickets so dense we could not see more than a few feet in any direction, the savages showed us the ruins of a group of stone houses whose walls were still standing in fine condition.
------ FIGURE
Ruins in the Jungles of Espiritu Pampa ------
One of the buildings was rounded at one end. Another, standing by itself at the south end of a little pampa, had neither doors nor windows. It was rectangular. Its four or five niches were arranged with unique irregularity. Furthermore, they were two feet deep, an unusual dimension. Probably this was a storehouse. On the east side of the pampa was a structure, 120 feet long by 21 feet wide, divided into five rooms of unequal size. The walls were of rough stones laid in adobe. Like some of the Inca buildings at Ollantaytambo, the lintels of the doors were made of three or four narrow uncut ashlars. Some rooms had niches. On the north side of the pampa was another rectangular building. On the west side was the edge of a stone-faced terrace. Below it was a partly enclosed fountain or bathhouse, with a stone spout and a stone-lined basin. The shapes of the houses, their general arrangement, the niches, stone roof-pegs and lintels, all point to Inca builders. In the buildings we picked up several fragments of Inca pottery.
Equally interesting and very puzzling were half a dozen crude Spanish roofing tiles, baked red. All the pieces and fragments we could find would not have covered four square feet. They were of widely different sizes, as though some one had been experimenting. Perhaps an Inca who had seen the new red tiled roofs of Cuzco had tried to reproduce them here in the jungle, but without success.
At dusk we all returned to Espiritu Pampa. Our faces, hands, and clothes had been torn by the jungle; our feet were weary and sore. Nevertheless the day's work had been very satisfactory and we prepared to enjoy a good night's rest. Alas, we were doomed to disappointment. During the day some one had brought to the hut eight tame but noisy macaws. Furthermore, our savage helpers determined to make the night hideous with cries, tom-toms, and drums, either to discourage the visits of hostile Indians or jaguars, or for the purpose of exorcising the demons brought by the white men, or else to cheer up their families, who were undoubtedly hiding in the jungle near by.
The next day the savages and our carriers continued to clear away as much as possible of the tangled growth near the best ruins. In this process, to the intense surprise not only of ourselves, but also of the savages, they discovered, just below the "bathhouse" where we had stood the day before, the well-preserved ruins of two buildings of superior construction, well fitted with stone-pegs and numerous niches, very symmetrically arranged. These houses stood by themselves on a little artificial terrace. Fragments of characteristic Inca pottery were found on the floor, including pieces of a large aryballus.
Nothing gives a better idea of the density of the jungle than the fact that the savages themselves had often been within five feet of these fine walls without being aware of their existence.
Encouraged by this important discovery of the most characteristic Inca ruins found in the valley, we continued the search, but all that any one was able to find was a carefully built stone bridge over a brook. Saavedra's son questioned the savages carefully. They said they knew of no other antiquities. Who built the stone buildings of Espiritu Pampa and Eromboni Pampa? Was this the "Vilcabamba Viejo"
of Father Calancha, that "University of Idolatry where lived the teachers who were wizards and masters of abomination," the place to which Friar Marcos and Friar Diego went with so much suffering? Was there formerly on this trail a place called Ungacacha where the monks had to wade, and amused t.i.tu Cusi by the way they handled their monastic robes in the water? They called it a "three days' journey over rough country." Another reference in Father Calancha speaks of Puquiura as being "two long days' journey from Vilcabamba." It took us five days to go from Espiritu Pampa to Pucyura, although Indians, unenc.u.mbered by burdens, and spurred on by necessity, might do it in three. It is possible to fit some other details of the story into this locality, although there is no place on the road called Ungacacha. Nevertheless it does not seem to me reasonable to suppose that the priests and Virgins of the Sun (the personnel of the "University of Idolatry") who fled from cold Cuzco with Manco and were established by him somewhere in the fastnesses of Uilcapampa would have cared to live in the hot valley of Espiritu Pampa. The difference in climate is as great as that between Scotland and Egypt, or New York and Havana. They would not have found in Espiritu Pampa the food which they liked. Furthermore, they could have found the seclusion and safety which they craved just as well in several other parts of the province, particularly at Machu Picchu, together with a cool, bracing climate and food-stuffs more nearly resembling those to which they were accustomed. Finally Calancha says "Vilcabamba the Old"
was "the largest city" in the province, a term far more applicable to Machu Picchu or even to Choqquequirau than to Espiritu Pampa.
On the other hand there seems to be no doubt that Espiritu Pampa in the montana does meet the requirements of the place called Vilcabamba by the companions of Captain Garcia. They speak of it as the town and valley to which Tupac Amaru, the last Inca, escaped after his forces lost the "young fortress" of Uiticos. Ocampo, doubtless wis.h.i.+ng to emphasize the difference between it and his own metropolis, the Spanish town of Vilcabamba, calls the refuge of Tupac "Vilcabamba the old." Ocampo's new "Vilcabamba" was not in existence when Friar Marcos and Friar Diego lived in this province. If Calancha wrote his chronicles from their notes, the term "old" would not apply to Espiritu Pampa, but to an older Vilcabamba than either of the places known to Ocampo.
The ruins are of late Inca pattern, not of a kind which would have required a long period to build. The unfinished building may have been under construction during the latter part of the reign of t.i.tu Cusi. It was t.i.tu Cusi's desire that Rodriguez de Figueroa should meet him at Pampaconas. The Inca evidently came from a Vilcabamba down in the montana, and, as has been said, brought Rodriguez a present of a macaw and two hampers of peanuts, articles of trade still common at Conservidayoc. There appears to me every reason to believe that the ruins of Espiritu Pampa are those of one of the favorite residences of this Inca--the very Vilcabamba, in fact, where he spent his boyhood and from which he journeyed to meet Rodriguez in 1565. [13]
In 1572, when Captain Garcia took up the pursuit of Tupac Amaru after the victory of Vilcabamba, the Inca fled "inland toward the valley of Sima-ponte ... to the country of the Manaries Indians, a warlike tribe and his friends, where balsas and canoes were posted to save him and enable him to escape." There is now no valley in this vicinity called Simaponte, so far as we have been able to discover. The Manaries Indians are said to have lived on the banks of the lower Urubamba. In order to reach their country Tupac Amaru probably went down the Pampaconas from Espiritu Pampa. From the "Pampa of Ghosts"
to canoe navigation would have been but a short journey. Evidently his friends who helped him to escape were canoe-men. Captain Garcia gives an account of the pursuit of Tupac Amaru in which he says that, not deterred by the dangers of the jungle or the river, he constructed five rafts on which he put some of his soldiers and, accompanying them himself, went down the rapids, escaping death many times by swimming, until he arrived at a place called Momori, only to find that the Inca, learning of his approach, had gone farther into the woods. Nothing daunted, Garcia followed him, although he and his men now had to go on foot and barefooted, with hardly anything to eat, most of their provisions having been lost in the river, until they finally caught Tupac and his friends; a tragic ending to a terrible chase, hard on the white man and fatal for the Incas.
It was with great regret that I was now unable to follow the Pampaconas River to its junction with the Urubamba. It seemed possible that the Pampaconas might be known as the Sirialo, or the Cori-beni, both of which were believed by Dr. Bowman's canoe-men to rise in the mountains of Vilcabamba. It was not, however, until the summer of 1915 that we were able definitely to learn that the Pampaconas was really a branch of the Cosireni. It seems likely that the Cosireni was once called the "Sima-ponte." Whether the Comberciato is the "Momori" is hard to say.
To be the next to follow in the footsteps of Tupac Amaru and Captain Garcia was the privilege of Messrs. h.e.l.ler, Ford, and Maynard. They found that the unpleasant features had not been exaggerated. They were tormented by insects and great quant.i.ties of ants--a small red ant found on tree trunks, and a large black one, about an inch in length, frequently seen among the leaves on the ground. The bite of the red ant caused a stinging and burning for about fifteen minutes. One of their carriers who was bitten in the foot by a black ant suffered intense pain for a number of hours. Not only his foot, but also his leg and hip were affected. The savages were both fishermen and hunters; the fish being taken with nets, the game killed with bows and arrows. Peccaries were shot from a blind made of palm leaves a few feet from a runway. Fis.h.i.+ng brought rather meager results. Three Indians fished all night and caught only one fish, a perch weighing about four pounds.
The temperature was so high that candles could easily be tied in knots. Excessive humidity caused all leather articles to become blue with mould. Clouds of flies and mosquitoes increased the likelihood of spreading communicable jungle fevers.
The river Comberciato was reached by Mr. h.e.l.ler at a point not more than a league from its junction with the Urubamba. The lower course of the Comberciato is not considered dangerous to canoe navigation, but the valley is much narrower than the Cosireni. The width of the river is about 150 feet and its volume is twice that of the Cosireni. The climate is very trying. The nights are hot. Insect pests are numerous. Mr. h.e.l.ler found that "the forest was filled with annoying, though sting-less, bees which persisted in attempting to roost on the countenance of any human being available." On the banks of the Comberciato he found several families of savages. All the men were keen hunters and fishermen. Their weapons consisted of powerful bows made from the wood of a small palm and long arrows made of reeds and finished with feathers arranged in a spiral.
Monkeys were abundant. Specimens of six distinct genera were found, including the large red howler, inert and easily located by its deep, roaring bellow which can be heard for a distance of several miles; the giant black spider monkey, very alert, and, when frightened, fairly flying through the branches at astonis.h.i.+ng speed; and a woolly monkey, black in color, and very intelligent in expression, frequently tamed by the savages, who "enjoy having them as pets but are not averse to eating them when food is scarce." "The flesh of monkeys is greatly appreciated by these Indians, who preserved what they did not require for immediate needs by drying it over the smoke of a wood fire."
On the Cosireni Mr. Maynard noticed that one of his Indian guides carried a package, wrapped in leaves, which on being opened proved to contain forty or fifty large hairless grubs or caterpillars. The man finally bit their heads off and threw the bodies into a small bag, saying that the grubs were considered a great delicacy by the savages.
The Indians we met at Espiritu Pampa closely resembled those seen in the lower valley. All our savages were bareheaded and barefooted. They live so much in the shelter of the jungle that hats are not necessary. Sandals or shoes would only make it harder to use the slippery little trails. They had seen no strangers penetrate this valley for about ten years, and at first kept their wives and children well secluded. Later, when Messrs. Hendriksen and Tucker were sent here to determine the astronomical position of Espiritu Pampa, the savages permitted Mr. Tucker to take photographs of their families. Perhaps it is doubtful whether they knew just what he was doing. At all events they did not run away and hide.
------ FIGURE
Campa Men at Espiritu Pampa ------
------ FIGURE