"In the Midst of a Loneliness":
The Architectural History of the Salinas Missions
NPS Logo


In the patios of the conventos of Abó and Quarai are two kivas. Historians have generally accepted that they were built before the conventos, and that the convento patios were centered on the kivas as a form of superposition: that is, a symbolic overbuilding to indicate the dominance of Christianity over the pagan religion of the Indians. There are, however, a number of problems with this interpretation. A few historians and anthropologists argue that a better explanation for the kivas is that the Indians built them after the conventos had been constructed, during one of the periods when there was no friar stationed at the mission. This position has problems, too. Some scholars have toyed with the idea that the kivas were built in the convento patios after the buildings were abandoned, but this never received much attention because all the Indians were supposed to have left the pueblos at the same time as the missionary.

All of these propositions suffer from basic problems. For example, the idea of superposition in the Province of New Mexico derives from a single example, that of Awatovi. There the church is built so that the main altar stands over a kiva that the Franciscans intentionally backfilled. John Otis Brew excavated beneath the altar because Ross Montgomery, the architect on the excavation crew, insisted that a kiva would be found there because of the "theory of Superposition." The care with which the structure was backfilled with clean sand instead of the usual midden earth, leaving the roof beams intact, was confirming evidence, in the mind of Montgomery, that the relationship between the kiva and the altar had symbolic significance. However, the kiva was somewhat offcentered, so that the wall between the apse and the sacristy passed over it. In fact, the central hatchway of the kiva was several feet north of the wall, under the sacristy. Only about the south third of the kiva was actually under any part of the sanctuary. Further, the location of a second kiva, backfilled like the first and also partly under the apse, but ignored by all the discussion in the Awatovi report other than a passing reference to its existence, suggests a second explanation.1

The second kiva makes it quite possible that there were several kivas in the area selected for the church and convento, and that all of them received the same careful backfilling. The most likely scenario would then go something like this: the Franciscans made arrangements with the pueblo authorities to use the area where the church was built. Several kivas already existed in this area. The pueblo authorities allowed the Franciscans to remove the roofing and fill the kivas in order to form a solid platform on which to build, on condition that the job be done carefully and with respect.2 Superposition, therefore, does not seem to have been the intent. If it had, it seems more likely that the main altar would have been centered precisely over one of the kivas.3

Considering the situation at a new pueblo, it is difficult to understand how the Franciscans could begin with such an arrogant move as to take over and destroy kivas without the approval of the local authorities. Fray Francisco Letrado, for example, was killed at Hawikuh for no more reason than that he called the people to Mass on one of their festival days. It would appear that when the soldiers left after the first few days, Franciscans had to be very careful how they conducted themselves.

At Abó and Quarai, the situation is even less indicative of superposition than at Awatovi. The kivas are carefully centered in the patios of the convento, so that no part of the structure of church or convento covers it. Rather than implying dominance, this plan almost suggests that the kivas were "centerpieces" of the conventos, as though a certain amount of respect for the structures had dictated the layout. More important, the kivas were built after the construction of the artificial platform on which the conventos were built, and therefore could not have been the object of an act of superposition.

At Las Humanas, Hayes remarked that Letrado probably picked the site he did for San Isidro in order to avoid kivas.4 Even more indicative of the situation, Letrado built his church next to kiva D, which continued in use concurrently with the church until perhaps 1661. Letrado had to pass the kiva almost every day while he was at Las Humanas, as did Acevedo for thirty years after him. In fact, none of the kivas of Las Humanas were destroyed until the anti-kachina campaigns by the Franciscans beginning in 1661. This demonstrates that a certain co-existence between the kivas and the churches was tolerated, at least at Las Humanas, by the Franciscans prior to 1660.5

The final objection to the "superposition" idea, however, is that the kivas at Abó and Quarai were built on the artificial platform that the Franciscans constructed for the missions, not the other way around. At Quarai, it is likely that the mission platform was constructed on the mound of a circular pueblo ruin; but quite unlikely that the ruin, which would have dated to about 1300 AD, would have had a square kiva high in its central rooms.6 At Abó, there is no evidence of ruins of any sort under the mission platform. The kiva had to have been built after the mission platform was constructed.

The idea that the Indians constructed the kivas when the priest was gone also has its problems. Extended absences of resident friars have long been accepted by historians as typical of the New Mexico missions. The argument has been that the number of friars was so limited that they constantly had to travel from one mission to the next in order to attend properly to all the Christianized Indians. However, this does not seem to be supported by the records. In reality, it appears that the core missions were continuously manned, and secondary or peripheral missions were placed in "visita" status when the number of available friars dropped too low. In the Salinas area, for example, Chililí was reduced to a visita of Tajique after about 1660, and Las Humanas moved in and out of visita status as the number of friars changed. Quarai and Abó, however, were core missions. It is unlikely that they would have been left unmanned for any extended period.7

The assumptions required for the "absence" idea include the "mission in the wilderness" concept. This idea is, for example, implicit in the title of this report, "In the Midst of a Loneliness."8 The assumption is that the friar was alone at his mission, with any other Spanish presence at some great distance. This does not appear to be valid, because the records indicate the probable presence of faithful Indian sacristans, a lay brother, or a second friar at those missions that needed one (those, for example, that had a visita and required that one friar be travelling frequently, such as at Abó in the 1630s). They also document several privately-owned estancias within only a few miles of Quarai. Such estancias were probably also within a short distance of Abó. In addition, architectural and archeological information demonstrates the probable presence of Spanish civil authorities in the pueblos; at Abó, for example, the Spaniards built a large compound against the north side of house block I just west of the church of San Gregorio I containing what appear to be stables and storage rooms. Documents support this: apparently all converted pueblos, or at least those given in encomienda, had Spanish alcaldes in permanent residence throughout most of the century.

Beyond the presence of all these persons, the existence of a completed mission building indicates the acceptance of the Franciscans by influential factions of the pueblo. An anti-Franciscan faction building a kiva in the convento would be acting against the interests of the powerful pro-Franciscan faction in full view of the entire pueblo. Since the pro-Franciscan factions of Abó and Quarai seem to have been stable and powerful, such an act would probably have been foolish on the part of the less powerful and less influential anti-Franciscan factions. It is highly likely, therefore, that the conventos of Abó and Quarai were always under the protection of someone associated or allied with Spanish authority.

The construction of the kivas after the abandonment of the conventos is more likely than previously thought. "Abandonment" meant only that the Franciscans withdrew their missionary from the pueblo, and, if possible, persuaded most of the Christianized Indians to go with them. This may also have signalled the loss of the tribute of the pueblo to its encomenderos. The anti-Spanish factions of the pueblos, undoubtedly involved in some sort of alliance with the local Apache, may not have left immediately. They, in fact, would be the people most likely to build such a kiva. However, two things argue against such a sequence of events: 1) the kivas are precisely centered in their patios, in a manner more suggestive of European planning methods than of the less compulsive Indian approach to building; and 2) the kiva at Abó was built centered on the first convento patio, which was changed to a different plan after about 1645--in fact, the evidence suggests that the kiva at Abó was filled in about that year.

At Abó, then, the kiva had to have been built between 1622 and 1645. The first church and convento were under construction from 1622 to about 1628 or 1629. From 1629 through about 1640 two friars were stationed at Abó in order to avoid periods of time when the convento was left empty. Acevedo became guardian of Abó in about 1640, and began work on the second church of San Gregorio soon afterwards. He appears to have been the only friar at the mission beginning in 1640. Therefore, the only time there is a reasonable chance that all friars were absent from the convento for longer than a few days is from about 1640 to about 1645. This was one of the most disrupted periods of New Mexico's history, as well as one of the least known.9 It is possible that during this period Acevedo was away from Abó long enough for anti-Franciscan factions to built a kiva in the center of the convento patio. However, such an interpretation does not explain the other oddities about the kivas at Abó and Quarai.

Consider the structural history of the kiva at Abó. It was laid out after 1622 with its center within one foot of the center of the first patio, and its roof apparently level with the artificial surface of the convento platform, giving an interior height of about seven feet, more like a Spanish roof height than that used by the Pueblo Indians.10 It was constructed with firepits, a deflector, a ventilator shaft, four large wooden pillars supporting the roof, and probably a central entranceway in the roof. It was used for a period of time, as demonstrated by the ash in the firepit. It was then unroofed and left an open hole for a period of time, as shown by the laminated layers of sand up to three feet deep against the walls in some areas of the floor. The unroofing may have occurred at the time of the rebuilding of the convento to its second plan in ca. 1645. Finally, earth from a midden near the convento was used to fill and level the interior at a depth of about 5 1/2 feet below the surface of the patio.11 The kiva was left open as a circular, stone-lined hole about 5 1/2 feet deep for the rest of the life of the convento. After abandonment, the side walls slowly collapsed into the pit along with blown sand, filling the hole until it was a barely detectable depression at the time Toulouse found it in 1940.

If the kiva had built by the Indians as a gesture of defiance against the Franciscans, then they would have excavated a hole in the patio without real regard for such niceties as carrying the backdirt away; therefore the backdirt would have been piled here and there around the patio at the edges of the kiva, or dumped in one of the convento rooms. The first act of the friar upon returning and discovering such an act of disrespect would be to rip off the roof of the kiva and shovel the dirt back into the hole. This would get the backdirt out of the way, and get rid of an embarassing episode in the history of the convento. Obviously this did not happen. What happened to the backdirt from the kiva?

At Quarai, the square kiva is precisely centered on its patio. Diagonal lines drawn through the outside corners of the patio pass exactly through the center of the kiva.12 The kiva is surprisingly precise: its north and east sides are 15.9 feet long, its south side is 16.1 feet long, about 2 1/2 inches longer, and its west side is 15.2 feet long, or about seven inches shorter. In addition, the kiva at Quarai is the only known square kiva in the Salinas pueblos. Quarai's kiva was carefully filled during the life of the convento, so that excavators did not recognize it until the final cleaning of the surface of the patio in 1934.

Few details about the kiva at Quarai were preserved. Its roof, again, was at the level of the patio floor, for an interior roof height of about seven feet. The kiva contained a fire pit containing ash, a ventilator shaft, and what the excavators interpreted as a sipapu, but which could have been the ladder pit. The fill contained at least one item manufactured for Europeans, and the lowest foot contained a large quantity of lime or gypsum plaster, apparently fallen from the walls. The kiva seems to have been unroofed and filled until it was level with the patio--the excavators saw no trace of it until the final cleaning of the patio floor surface, indicating no depression remaining above it.

The usual theories for the presence of the kivas do not explain these oddities. The author suggests a new idea: the kivas were built by the Indians under Franciscan supervision to serve as transitional churches or classrooms while construction continued on the principle church at each mission.

If the kivas were built under the eyes of Spanish allies or authorities, probably during the first few years after the establishment of the missions, and carefully centered in the patios, then it is most likely that the Franciscans themselves approved of the buildings. The locations imply that they were associated with the religious life of the missions; they were built in the middle of things but out of the way of major construction efforts. In the 1620s, when the patio kivas were apparently built, these little structures appear to have been considered rather innocuous. They were originally called estufas, or sweat-houses. The missionaries were aware that they were the equivalent of churches, but the intense Franciscan opposition to kivas because they were the focus of Pueblo Indian religious activities (the "kiva wars") did not begin until the 1660s.13 If the Franciscans permitted or encouraged their construction, then the most likely use for these kiva-like structures was as a church. Their small size may indicate that they were for the conversion and training of only a few influential Indians. The first efforts of conversion at a new pueblo were usually directed at the "Caciques and captains of the pueblo."14 The kiva was a building familiar to the Indians in which the Franciscan could teach them about the new religion. The symbology of bringing the Indians up from the darkness of the kiva into the light of the new church building would have been attractive to the missionary.

The Franciscans had no problem with using a kiva-like building as a Christian church. The shape or structure of a church was not important.15 On the road to New Mexico they used a tent and a wooden table to celebrate Mass, and upon entering a new pueblo, they established a church in one of the pueblo rooms built by the Indians. Finally, there is one case on record where a kiva built by the Indians and used for Indian religious ceremonies was converted to a Christian church. In December, 1693, Diego de Vargas ordered that a kiva near the palace of the governors be refurbished and used as a church. After having the interior of the structure whitewashed and an altar built, de Vargas invited the custodio to inspect the place. The custodio objected that "one could not celebrate mass in the estufa because it had served as a place for their idolatry and diabolical meetings and dances." De Vargas replied that "the principal cathedrals of Spain had been previously mosques of the Moors." In his journal, he added that "this reasoning was so convincing that I had my proposition accepted and had the estufa rearranged and made ready."16

The author does not "believe" that the Franciscans built the patio kivas. The attribution of the patio kivas to Franciscans is the hypothesis that best fits all the available evidence, not an idea concocted by the author for its shock value. It is the simplest explanation of the available information. If an idea suggested by the available evidence does not fit with what is considered known, it is not automatically wrong, but only indicates that something may be wrong. The problem could just as easily be with the body of facts considered to be known. The author's position is that modern scholarship is a long way from being able to say flatly what Franciscans would or would not have done in the 1620s. The only way to discover the rules within which they worked is by a willingness to hypothesize and then to test the hypothesis.

1 Ross Gordon Montgomery, Watson Smith, and John Otis Brew, Franciscan Awatovi: The Excavation and Conjectural Reconstruction of a 17th-Century Spanish Mission Establishment at a Hopi Indian Town in Northeastern Arizona, Reports of the Awatovi Expedition, No. 3, Papers of the Peabody Museum of American Archaeology and Ethnology, vol. 36 (Cambridge: Peabody Museum of American Archaeology and Ethnology, 1949), pp. 64-67, 77, and figure 22d.

2 Brew's description makes it clear that, although the beams that had supported the roof were left in place, the roofing itself had been removed. This made the job of filling the kiva much easier.

3 This is not to argue that acts of superposition did not happen during the conquest of Mexico. Such activities are well-documented; see, for example, Montgomery's discussion in Awatovi, pp. 134-36, 265-72. The evidence simply does not seem to justify the assumption that superposition was the determining factor in the location of the church of Awatovi.

4 Hayes, Mound 7, p. 36.

5 It is possible that, rather than being an example of tolerance, kiva D is another example of Franciscan construction like those suggested at Abó and Quarai, discussed in this appendix. Considering its location and the available information about it, the possibility exists that Letrado built kiva D with the intent for it to be centered in the patio of a convento planned to be built against the north side of the church of San Isidro. Kiva D was apparently built late in the life of Las Humanas, was filled in the late seventeenth century, and resembled the kiva in the convento of Abó; see Vivian, Excavations, pp. 44-45, 54, 57, and 110; also Hayes, Mound 7, p. 58, 61; and Hackett, Documents, p. 166. Because kivas are excavated into underlying fill by their builders, it is usually difficult to determine the date of their construction.

6 In fact, Quarai has a second kiva under the mission compound. This is a round kiva under the terraced section of the second courtyard, only a few feet east of the east wall of the friary. The upper retaining wall of the terraces passes over the center of the kiva. The limited archeological and photographic information indicates that the kiva was in ruins and had partly collapsed by the time the Franciscans built the second courtyard terraces across it.

7 During the 1660s, for example, several missions had more than one Franciscan. In fact, two missions had three friars, and another seven had two, out of a total of twenty-five missions, or about thirty-six percent. See Scholes, "Documents," pp. 52-56.

8 The title was used for this report because it so precisely captures the feeling of the ruins today. However, the original use of the phrase by Fray Hans Lentz in 1969 referred to the conditions in which the Franciscans lived. One of the conclusions of this report is that the "mission in the wilderness" idea is incorrect for most of the life of the Salinas missions.

9 See, for example, James Ivey, "The Scholes Manuscript: Another Look at the Dating of An Important Seventeenth Century Document," manuscript in the files of the Southwest Regional Office, National Park Service, Santa Fe.

10 At the time of its discovery, about six feet of wall height within the kiva survived. The stratigraphy implies that one or two feet of wall collapsed into the pit of the kiva after the abandonment of the mission, and that the roof had been about level with the patio courtyard. This indicates a distance from the floor to the underside of the ceiling of perhaps seven feet, significantly higher than the usual kiva height of about 5 1/2 feet. See Joseph H. Toulouse, Jr., The Mission of San Gregorio de Abó: A Report on the Excavation and Repair of a Seventeenth-Century New Mexico Mission, Monographs of the School of American Research, No. 13 (Albuquerque: University of New Mexico Press, 1949), p. 11 and figure 5.

11 Toulouse says that the kiva was used as a dump from the nearby kitchen, but also indicates that the filling was a single event with fragments of the same ceramic vessels appearing at all levels of the material. Therefore, the use of the kiva as a dump over any length of time is ruled out. The only way for a single-event filling to occur in this way would be for the fill to be removed from a pre-existing kitchen midden and hauled to the kiva.

12 The centering error is smaller than the limits of error in the plan. The difference between the two centers is therefore less than two inches. Available colonial Spanish surveyor's notes indicate that the usual method for the layout of a building or compound was to pick the center of the site and then plot diagonals from that point to locate the corners of the structure. If the Franciscans had laid out the kiva at Quarai before allowing the Indians to build it, they would have done so in a manner that resulted in the kiva being precisely centered in the patio.

13 There was a brief anti-kachina movement at Pecos in 1620; Kessell, Kiva, Cross, and Crown, p. 110-11; and a short reference to difficulties with the return of the Indians to their old ways in 1644; Scholes, "Church and State," p. 324; but the literature does not indicate any wholesale, province-wide suppression of kivas until the kachina wars of the 1660s. For example, both Alden Hayes and Gordon Vivian believe that anti-kiva efforts did not begin at Las Humanas until after 1660; Hayes, Mound 7, pp. 7, 58; Vivian, Excavations, p. 29; Hackett, Documents, p. 166.

14 Vivian, Excavations, p. 24; Bloom, "Perea's Relacion," pp. 228-34.

15 That is, so long as it was not a consecrated church. See the discussion of levels of blessing of a church and the restrictions accompanying each level in Montgomery, Awatovi, pp. 178-81, 190-91, 273-76; see also Appendix 2 in this report.

16 A. von Wuthenau, "The Spanish Military Chapels in Santa Fe and the Reredos of our Lady of Light," New Mexico Historical Review, 10 (July 1935): 178-79. See also Fray Angélico Chavez, "Santa Fe Church and Convent Sites in the Seventeenth and Eighteenth Centuries," New Mexico Historical Review, 24 (April 1949): 90, where Chavez indicates that the structure in question may not have been the original tower chapel in the palace of the governors. Whether or not the building had originally been built for Christian services does not matter here--the main point is that the participants in the debate between de Vargas and the custodio considered the building a kiva.

<<< Previous <<< Contents >>>

Last Updated: 28-Aug-2006