copyright, Kellscraft Studio
1999-2003                                             
(Return to Web Text-ures)                                             
Click Here to return to
Inca Lands
Content Page

Click Here to return to
previous chapter

 (HOME)   

CHAPTER XII

THE FORTRESS OF UITICOS AND THE HOUSE OF THE SUN

WHEN the viceroy, Toledo, determined to conquer that last stronghold of the Incas where for thirty-five years they had defied the supreme power of Spain, he offered a thousand dollars a year as a pension to the soldier who would capture Tupac Amaru. Captain Garcia earned the pension, but failed to receive it; the “maņana habit” was already strong in the days of Philip II. So the doughty captain filed a collection of testimonials with Philip’s Royal Council of the Indies. Among these is his own statement of what happened on the campaign against Tupac Amaru. In this he says: “and having arrived at the principal fortress, Guaynapucara [“the young fortress”], which the Incas had fortified, we found it defended by the Prince Philipe Quispetutio, a son of the Inca Titu Cusi, with his captains and soldiers. It is on a high eminence surrounded with rugged crags and jungles, very dangerous to ascend and almost impregnable. Nevertheless, with my aforesaid company of soldiers I went up and gained the fortress, but only with the greatest possible labor and danger. Thus we gained the province of Uilcapampa.” The viceroy himself says this important victory was due to Captain Garcia’s skill and courage in storming the heights of Guaynapucara, “on Saint John the Baptist’s day, in 1572.”

The “Hill of Roses” is indeed “a high eminence surrounded with rugged crags.” The side of easiest approach is protected by a splendid, long wall, built so carefully as not to leave a single toe-hold for active besiegers. The barracks at Uncapampa could have furnished a contingent to make an attack on that side very dangerous. The hill is steep on all sides, and it would have been extremely easy for a small force to have defended it. It was undoubtedly “almost impregnable.” This was the feature Captain Garcia was most likely to remember.

On the very summit of the hill are the ruins of a partly enclosed compound consisting of thirteen or fourteen houses arranged so as to form a rough square, with one large and several small courtyards. The outside dimensions of the compound are about 160 feet by 145 feet. The builders showed the familiar Inca sense of symmetry in arranging the houses. Due to the wanton destruction of many buildings by the natives in their efforts at treasure-hunting, the walls have been so pulled down that it is impossible to get the exact dimensions of the buildings. In only one of them could we be sure that there had been any niches.

PRINCIPAL DOORWAY OF THE
LONG PALACE  AT ROSASPATA 
ANOTHER DOORWAY IN THE
RUINS OF ROSASPATA

Most interesting of all is the structure which caught the attention of Ocampo and remained fixed in his memory. Enough remains of this building to give a good idea of its former grandeur. It was indeed a fit residence for a royal Inca, an exile from Cuzco. It is 245 feet by 43 feet. There were no windows, but it was lighted by thirty doorways, fifteen in front and the same in back. It contained ten large rooms, besides three hallways running from front to rear. The walls were built rather hastily and are not noteworthy, but the principal entrances, namely, those leading to each hall, are particularly well made; not, to be sure, of “marble” as Ocampo said — there is no marble in the province — but of finely cut ashlars of white granite. The lintels of the principal doorways, as well as of the ordinary ones, are also of solid blocks of white granite, the largest being as much as eight feet in length. The doorways are better than any other ruins in Uilcapampa except those of Machu Picchu, thus justifying the mention of them made by Ocampo, who lived near here and had time to become thoroughly familiar with their appearance. Unfortunately, a very small portion of the edifice was still standing. Most of the rear doors had been filled up with ashlars, in order to make a continuous fence. Other walls had been built from the ruins, to keep cattle out of the cultivated pampa. Rosaspata is at an elevation which places it on the borderland between the cold grazing country, with its root crops and sublimated pigweeds, and the temperate zone where maize flourishes.

On the south side of the hilltop, opposite the long palace, is the ruin of a single structure, 78 feet long and 35 feet wide, containing doors on both sides, no niches and no evidence of careful workmanship. It was probably a barracks for a company of soldiers.

The intervening “pampa” might have been the scene of those games of bowls and quoits, which were played by the Spanish refugees who fled from the wrath of Gonzalo Pizarro and found refuge with the Inca Manco. Here may have occurred that fatal game when one of the players lost his temper and killed his royal host.

Our excavations in 1915 yielded a mass of rough potsherds, a few Inca whirl-bobs and bronze shawl pins, and also a number of iron articles of European origin, heavily rusted — horseshoe nails, a buckle, a pair of scissors, several bridle or saddle ornaments, and three Jew’s-harps. My first thought was that modern Peruvians must have lived here at one time, although the necessity of carrying all water supplies up the hill would make this unlikely. Furthermore, the presence here of artifacts of European origin does not of itself point to such a conclusion. In the first place, we know that Manco was accustomed to make raids on Spanish travelers between Cuzco and Lima. He might very easily have brought back with him a Spanish bridle. In the second place the musical instruments may have belonged to the refugees, who might have enjoyed whiling away their exile with melancholy twanging. In the third place the retainers of the Inca probably visited the Spanish market in Cuzco, where there would have been displayed at times a considerable assortment of goods of European manufacture. Finally Rodriguez de Figueroa speaks expressly of two pairs of scissors he brought as a present to Titu Cusi. That no such array of European artifacts has been turned up in the excavations of other important sites in the province of Uilcapampa would seem to indicate that they were abandoned before the Spanish Conquest or else were occupied by natives who had no means of accumulating such treasures.

Thanks to Ocampo’s description of the fortress which Tupac Amaru was occupying in 1572 there is no doubt that this was the palace of the last Inca. Was it also the capital of his brothers, Titu Cusi and Sayri Tupac, and his father, Manco? It is astonishing how few details we have by which the Uiticos of Manco may be identified. His contemporaries are strangely silent. When he left Cuzco and sought refuge “in the remote fastnesses of the Andes,” there was a Spanish soldier, Cieza de Leon, in the armies of Pizarro who had a genius for seeing and hearing interesting things and writing them down, and who tried to interview as many members of the royal family as he could; — Manco had thirteen brothers. Cieza de Leon says he was much disappointed not to be able to talk with Manco himself and his sons, but they had “retired into the provinces of Uiticos, which are in the most retired part of those regions, beyond the great Cordillera of the Andes.”1 The Spanish refugees who died as the result of the murder of Manco may not have known how to write. Anyhow, so far as we can learn they left no accounts from which any one could identify his residence.

Titu Cusi gives no definite clue, but the activities of Friar Marcos and Friar Diego, who came to be his spiritual advisers, are fully described by Calancha. It will be remembered that Calancha remarks that “close to Uiticos in a village called Chuquipalpa, is a House of the Sun and in it a white stone over a spring of water.” Our guide had told us there was such a place close to the hill of Rosaspata.


NORTHEAST FACE OF YURAK RUMI

On the day after making the first studies of the “Hill of Roses,” I followed the impatient Mogrovejo — whose object was not to study ruins but to earn dollars for finding them — and went over the. hill on its northeast side to the Valley of Los Andenes (the Terraces”). Here, sure enough, was a large, white granite boulder, flattened on top, which had a carved seat or platform on its northern side. Its west side covered a cave in which were several niches. This cave had been walled in on one side. When Mogrovejo and the Indian guide said there was a manantial de agua (“spring of water”) near by, I became greatly interested. On investigation, however, the “spring” turned out to be nothing but part of a small irrigating ditch. (Manantial means “spring”; it also means “running water”). But the rock was not “over the water.” Although this was undoubtedly one of those huacas, or sacred boulders, selected by the Incas as the visible representations of the founders of a tribe and thus was an important accessory to ancestor worship, it was not the Yurak Rumi for which we were looking. Leaving the boulder and the ruins of what possibly had been the house of its attendant priest, we followed the little water course past a large number of very handsomely built agricultural terraces, the first we had seen since leaving Machu Picchu and the most important ones in the valley. So scarce are andenes in this region and so noteworthy were these in particular that this vale has been named after them. They were probably built under the direction of Manco. Near them are a number of carved boulders, huacas. One had an intihuatana, or sundial nubbin, on it; another was carved in the shape of a saddle. Continuing, we followed a trickling stream through thick woods until we suddenly arrived at an open place called Nusta Isppana. Here before us was a great white rock over a spring. Our guides had not misled us. Beneath the trees were the ruins of an Inca temple, flanking and partly enclosing the gigantic granite boulder, one end of which overhung a small pool of running water. When we learned that the present name of this immediate vicinity is Chuquipalta our happiness was complete.

It was late on the afternoon of August 9, 1911, when I first saw this remarkable shrine. Densely wooded hills rose on every side. There was not a hut to be seen; scarcely a sound to be heard. It was an ideal place for practicing the mystic ceremonies of an ancient cult. The remarkable aspect of this great boulder and the dark pool beneath its shadow had caused this to become a place of worship. Here, without doubt, was “the principal mochadero of those forested mountains.” It is still venerated by the Indians of the vicinity. At last we had found the place where, in the days of Titu Cusi, the Inca priests faced the east, greeted the rising sun, “extended their hands toward it,” and “threw kisses to it,” “a ceremony of the most profound resignation and reverence.” We may imagine the sun priests, clad in their resplendent robes of office, standing on the top of the rock at the edge of its steepest side, their faces lit up with the rosy light of the early morning, awaiting the moment when the Great Divinity should appear above the eastern hills and receive their adoration. As it rose they saluted it and cried: “O Sun! Thou who art in peace and safety, shine upon us, keep us from sickness, and keep us in health and safety. O Sun! Thou who hast said let there be Cuzco and Tampu, grant that these children may conquer all other people. We beseech thee that thy children the Incas may be always conquerors, since it is for this that thou hast created them.”

It was during Titu Cusi’s reign that Friars Marcos and Diego marched over here with their converts from Puquiura, each carrying a stick of firewood. Calancha says the Indians worshiped the water as a divine thing, that the Devil had at times shown himself in the water. Since the surface of the little pool, as one gazes at it, does not reflect the sky, but only the overhanging, dark, mossy rock, the water looks black and forbidding, even to unsuperstitious Yankees. It is easy to believe that simple-minded Indian worshipers in this secluded spot could readily believe that they actually saw the Devil appearing “as a visible manifestation” in the water. Indians came from the most sequestered villages of the dense forests to worship here and to offer gifts and sacrifices. Nevertheless, the Augustinian monks here raised the standard of the cross, recited their orisons, and piled firewood all about the rock and temple. Exorcising the Devil and calling him by all the vile names they could think of, the friars commanded him never to return. Setting fire to the pile, they burned up the temple, scorched the rock, making a powerful impression on the Indians and causing the poor Devil to flee, “roaring in a fury.” “The cruel Devil never more returned to the rock nor to this district.” Whether the roaring which they heard was that of the Devil or of the flames we can only conjecture. Whether the conflagration temporarily dried up the swamp or interfered with the arrangements of the water supply so that the pool disappeared for the time being and gave the Devil no chance to appear in the water, where he had formerly been accustomed to show himself, is also a matter for speculation.

The buildings of the House of the Sun are in a very ruinous state, but the rock itself, with its curious carvings, is well preserved notwithstanding the great conflagration of 1570. Its length is fifty-two feet, its width thirty feet, and its height above the present level of the water, twenty-five feet. On the west side of the rock are seats and large steps or platforms. It was customary to kill llamas at these holy huacas. On top of the rock is a flattened place which may have been used for such sacrifices. From it runs a little crack in the boulder, which has been artificially enlarged and may have been intended to carry off the blood of the victim killed on top of the rock. It is still used for occult ceremonies of obscure origin which are quietly practiced here by the more superstitious Indian women of the valley, possibly in memory of the Nusta or Inca princess for whom the shrine is named.

On the south side of the monolith are several large platforms and four or five small seats which have been cut in the rock. Great care was exercised in cutting out the platforms. The edges are very nearly square, level, and straight. The east side of the rock projects over the spring. Two seats have been carved immediately above the water. On the north side there are no seats. Near the water, steps have been carved. There is one flight of three and another of seven steps. Above them the rock has been flattened artificially and carved into a very bold relief. There are ten projecting square stones, like those usually called intihuatana or “places to which the sun is tied.” In one line are seven; one is slightly apart from the six others. The other three are arranged in a triangular position above the seven. It is significant that these stones are on the northeast face of the rock, where they are exposed to the rising sun and cause striking shadows at sunrise.


CARVED SEATS AND PLATFORMS OF ŅUSTA ISPPANA


TWO OF THE SEVEN SEATS NEAR THE SPRING UNDER THE GREAT WHITE ROCK

Our excavations yielded no artifacts whatever and only a handful of very rough old potsherds of uncertain origin. The running water under the rock was clear and appeared to be a spring, but when we drained the swamp which adjoins the great rock on its northeastern side, we found that the spring was a little higher up the hill and that the water ran through the dark pool. We also found that what looked like a stone culvert on the borders of the little pool proved to be the top of the back of a row of seven or eight very fine stone seats. The platform on which the seats rested and the seats themselves are parts of three or four large rocks nicely fitted together. Some of the seats are under the black shadows of the overhanging rock. Since the pool was an object of fear and mystery the seats were probably used only by priests or sorcerers. It would have been a splendid place to practice divination. No doubt the devils “roared.”

All our expeditions in the ancient province of Uilcapampa have failed to disclose the presence of any other “white rock over a spring of water” surrounded by the ruins of a possible “House of the Sun.” Consequently it seems reasonable to adopt the following conclusions: First, Nusta Isppana is the Yurak Rumi of Father Calancha. The Chuquipalta of to-day is the place to which he refers as Chuquipalpa. Second, Uiticos, “close to” this shrine, was once the name of the present valley of Vilcabamba between Tincochaca and Lucma. This is the “Viticos” of Cieza de Leon, a contemporary of Manco, who says that it was to the province of Viticos that Manco determined to retire when he rebelled against Pizarro, and that “having reached Viticos with a great quantity of treasure collected from various parts, together with his women and retinue, the king, Manco Inca, established himself in the strongest place he could find, whence he sallied forth many times and in many directions and disturbed those parts which were quiet, to do what harm he could to the Spaniards, whom he considered as cruel enemies.” Third, the “strongest place” of Cieza, the Guaynapucara of Garcia, was Rosaspata, referred to by Ocampo as “the fortress of Pitcos,” where, he says, “there was a level space with majestic buildings,” the most noteworthy feature of which was that they had two kinds of doors and both kinds had white stone lintels. Fourth, the modern village of Pucyura in the valley of the river Vilcabamba is the Puquiura of Father Calancha, the site of the first mission church in this region, as assumed by Raimondi, although he was disappointed in the insignificance of the “wretched little village.” The remains of the old quartz-crushing plant in Tincochaca, which has already been noted, the distance from the “House of the Sun,” not too great for the religious procession, and the location of Pucyura near the fortress, all point to the correctness of this conclusion.

Finally, Calancha says that Friar Ortiz, after he had secured permission from Titu Cusi to establish the second missionary station in Uilcapampa, selected “the town of Huarancalla, which was populous and well located in the midst of a number of other little towns and villages. There was a distance of two or three days’ journey from one convent to the other. Leaving Friar Marcos in Puquiura, Friar Diego went to his new establishment, and in a short dine built a church.” There is no “Huarancalla” to-day, nor any tradition of any, but in Mapillo, a pleasant valley at an elevation of about 10,000 feet, in the temperate zone where the crops with which the Incas were familiar might have been raised, near pastures where llamas and alpacas could have flourished, is a place called Huarancalque The valley is populous and contains a number of little towns and villages. Furthermore, Huarancalque is two or three days’ journey from Pucyura and is on the road which the Indians of this region now use in going to Ayacucho. This was undoubtedly the route used by Manco in his raids on Spanish caravans. The Mapillo flows into the Apurimac near the mouth of the river Pampas. Not far up the Pampas is the important bridge between Bombon and Ocros, which Mr. Hay and I crossed in 1909 on our way from Cuzco to Lima. The city of Ayacucho was founded by Pizarro, a day’s journey from this bridge. The necessity for the Spanish caravans to cross the river Pampas at this point made it easy for Manco’s foraging expeditions to reach them by sudden marches from Uiticos down the Mapillo River by way of Huarancalque, which is probably the “Huarancalla” of Calancha’s “Chronicles.” He must have had rafts or canoes on which to cross the Apurimac, which is here very wide and deep. In the valleys between Huarancalque and Lucma, Manco was cut off from central Peru by the Apurimac and its magnificent canyon, which in many places has a depth of over two miles. He was cut off from Cuzco by the inhospitable snow fields and glaciers of Salcantay, Soray, and the adjacent ridges, even though they are only fifty miles from Cuzco. Frequently all the passes are completely snow-blocked. Fatalities have been known even in recent years. In this mountainous province Manco could be sure of finding not only security from his Spanish enemies, but any climate that he desired and an abundance of food for his followers. There seems to be no reason to doubt that the retired region around the modern town of Pucyura in the upper Vilcabamba Valley was once called Uiticos.

__________________

1 In those days the term “Andes” appears to have been very limited in scope, and was applied only to the high range north of Cuzco where lived the tribe called Antis. Their name was given to the range. Its culminating point was Mt. Salcantay.

Click here to continue to the next chapter of Inca Lands