TUMACACORI Historic Resource Study

In the spring of 1935, during the depths of the Great Depression, three men met for several nights in the village of Snaketown just north of the dry carcass of the Gila River. One was Julian Hayden, a young archaeologist working on the excavation of the Hohokam site at Snaketown under the direction of Emil Haury. Another was William Allison Smith, the foreman of the digging crew, an Akimel O'odham who spoke English and was a Presbyterian deacon in the Snaketown church. The third was Juan Smith, one of the last narrators of the full Pima version of the O'odham creation story (Bahr et al 1994).

There, in an old mesquite-roofed adobe storeroom, Smith recited the thirty-six stories that reveal O'odham history from the creation of the universe to the end of the Apache wars. Allison Smith translated Smith's Pima narration into English. Hayden scribbled furiously to record the word-for-word transcript. One evening, Hayden also snapped a picture of Smith by the light of a Coleman lantern. Smith is seated in front of a jumble of wagon wheels leaning against an adobe wall. His broad face, framed by a shock of black hair, is patient and intent. You can almost smell the dry, dark earth enclosing the three men.

The text, in English with a sprinkling of Pima words, is one of the most complete glimpses we have into the O'odham world before the Europeans arrived. The first part is a Pima Genesis. A pure spirit named Jeoss [1] heavens out of darkness. Then he makes Earth Doctor—Jewed Ma:kai—who fashions another being, Siuuhu. Earth Doctor and Siuuhu take clay, shape a man, and breathe life into him. They do the same to produce a woman. These two first humans become the original ancestors of the Pimas.

After them, Earth Doctor and Siuuhu create the dawn, the sun, the deer, the jackrabbit, the windstorm, the clouds, and the rain. Grasses spring up after the rain. "The woman went out and picked some of the grass and took it and cooked it and ate it," Smith recited. "When through eating, the man went out with bow and arrow and killed a deer, and they had it for supper that evening. At that time, the sun went down, and darkness fell over the earth, as it had been before. This was the end of the first day" (Bahr et al 1994:49).

Creation continues with the making of the moon, Coyote, the Milky Way, and Buzzard, who shapes the mountains. A great flood destroys this first creation. In other versions of the narrative, the flood is caused by the tears of a baby born from the penis of a promiscuous young man. Only Earth Doctor, Siuuhu, and Coyote survive. Earth Doctor returns to the heavens, but Coyote and Siuuhu take refuge in a flute and a house. Siuuhu emerges first and begins to be called S-e'ehe, or Elder Brother. Later in the narrative, he is also called I'itoi.

This is where the creation narrative leaves the realm of mythology and enters "prehistory." In Smith's account, the first people all die. In others, however, they take refuge in the underworld after Earth Doctor opens a passage into the earth with his cane. Siuuhu then makes more people, including the O'odham, whom he teaches how to cultivate a variety of crops, including corn, cotton, and tobacco. A series of stories follow with two interwoven themes: the domestication of plants and the domestication of sexual relations. The people also learn to make saguaro wine and to irrigate their fields from canals. Anthropologist Donald Bahr (Bahr et al 1994) calls these tales the "Hohokam chronicles" because they tell the story of the people who built the great platform mound communities like Casa Grande along the Salt and Gila rivers.

As these people's mastery of canal irrigation grows, their societies become more complex. The origins of turquoise and parrots—two highly prized items that linked the Hohokam in trade networks extending from the Four Corners to Mesoamerica—are explained. Warfare breaks out and Siuuhu pioneers the four-day ceremony to purify warriors after they have killed. The Hohokam grow so arrogant they kill Siuuhu himself.

It is here in the narrative that "history" begins. After four years of death, Siuuhu resurrects himself and contacts the people Earth Doctor either created or saved. They agree to join him in making war against his killers. The people who accompany Siuuhu are called Wooshkum or Vupshkam (Wu:skam), which means "Emerger" or "Emergent." They arise from the underworld far to the east. Medicine men among both the Hohokam and the Wu:skam have visions about the impending conflict. The Hohokam medicine men try to conceal their communities. The Wu:skam medicine men attempt to divine their locations.

Finally, however, battle is joined. The Wu:skam send an owl to frighten the Hohokam, a gopher to chew up their weapons, and a rattlesnake to bite and kill a powerful Sivañ (chief). Then they descend upon a series of communities with Great Houses (Vahki) beginning with Casa Grande, killing the medicine men chiefs and routing their followers. In addition to Casa Grande, Smith's narrative and the more fragmentary narratives recorded by others mention communities that can be identified as Sweetwater, Casa Blanca, Lower Santan, Upper Santan, Los Muertos, Los Hornos, and Pueblo Grande, ruled by the powerful priest-chief Yellow Buzzard (Teague 1993; Bahr et al 1994). According to archaeologist Lynn Teague, "The settlements identified in the oral traditions are an archaeologically accurate inventory of the late Classic period platform mound sites along a lengthy segment of the Gila River. Further, it can be shown that the ethnohistoric reports provide an account of these settlements too accurate and detailed to have been simply a tale created to explain the existence of numerous sites in the region" (Teague 1993:440).

In other words, O'odham living in the northern Sonoran Desert in the twentieth century possessed a detailed oral tradition that their ancestors conquered the Hohokam living in the same area 400 to 500 years earlier. According to the creation narratives, the Hohokam rulers and some of their followers died while others fled west to the Colorado River or north to Zuni country and the Hopi mesas. But others were incorporated into O'odham society, perhaps into the Buzzard moiety (Hayden 1970). As such, the creation narratives contain tantalizing slivers of evidence about the antiquity of the O'odham in Arizona and their relation to the Hohokam.

The Hohokam-Piman Continuum and the Tepiman Connection

The nature of that relationship remains a matter of debate. Most archaeologists believe in some variant of the Hohokam-Piman continuum, which contends that the Upper Pimas of Arizona are biological and cultural descendants of the Hohokam. After Hohokam civilization collapsed because of social conflict, environmental degradation, or perhaps even the devastation of Old World epidemic diseases, the Pimas were the remnants who survived (Haury 1945, 1950, 1976; Simpson 1946; Gumerman and Haury 1979; Doyel 1979, 1991; Reff 1991; Dobyns 1988; Ezell 1963). Charles DiPeso, on the other hand, argued that the Hohokam intruded upon the O'odham, who had been living in the northern Sonoran Desert since Archaic times (DiPeso 1956, 1979; see also McGuire 1991). A third scenario, advanced by ethnohistorian Bernard Fontana (1976) and ethnobiologist Amadeo Rea (1983, 1997), views the Upper Pimas as relatively recent newcomers who moved into the region as Hohokam civilization was disintegrating. Because they had no public architecture that rivaled the platform mounds of the Classic Hohokam, because they buried rather than burned their corpses, and because they relied upon small-scale rather than massive irrigation systems, Rea argues that the Upper Pimas were too different from the Hohokam to be "the degenerate remnants of a pre-existing Puebloan or puebloid desert culture" (Rea 1997:8).

The debate is complicated even further by the so-called "Tepiman connection." People speaking Piman languages within the far-flung Uto-Aztecan language family extend from central Arizona to northern Jalisco. They include the Upper Pimas (Akimel and Tohono O'odham), who occupied most of southern Arizona and northern Sonora when the Europeans colonized the Pimería Alta in the late 1600s. They encompass the Lower Pimas of central Sonora: the Névome, who lived along the middle Río Sonora and Río Yaqui, the Mountain Pimas, who occupied the western foothills of the Sierra Madre Occidental, and a number of poorly understood groups such as the Cocomacaques, who ranged across the desert north and west of modern Hermosillo. Tepiman speakers also comprise the Northern Tepehuan of Chihuahua, the Southern Tepehuan of Durango, and the Tepecano of northern Jalisco. This long s-shaped curve of Piman speakers, broken by Taracahitan speakers (Yaquis, Mayos, Warihios, and Tarahumaras) in southern Sonora and Chihuahua, snakes for a thousand miles across the deserts and mountains of North America.

The languages of these people, who call themselves O'odham, O'odaam, Ootoma, and Odami, are so closely related that linguists Kenneth Hale and David Harris state, "It is somewhat misleading to speak of time-depth within Piman; it is more probable that, until very recent times, Piman represented a more or less continuous chain of dialects belonging to a single language" (Hale and Harris 1979:176). Linguistic anthropologists David Shaul and Jane Hill (1998) observe that the various Tepiman languages share cognate densities of 74.4 percent to 86.4 percent, suggesting that they split from one another a millenium ago; a rough glottochronological rule of thumb is that cognancy of 80.5% represents a divergence of about 1,000 years. "Glottochronological dates should, of course, be used only with great caution," they go on to say, "but the fact that the lowest percentage of shared cognacy within Tepiman (between Southeastern Tepehuan and Tohono O'odham) is 73 percent is consistent with a hypothesis that the proto-Tepiman speech community existed sometime during the first millenium A.D., well within the Hohokam period" (Shaul and Hill 1998:379). Relationships between Hohokam and O'odham in Arizona, then, have to be viewed within a much larger framework, one that encompasses a world of Tepiman speakers that climbs from the creosote flats of the Phoenix Basin to the pine uplands of the Sierra Madre Occidental.

Archaeologist David Wilcox (1986) speculates that this corridor of Tepiman speakers was the principal route over which elements of Mesoamerican civilization diffused into the Southwest. Rea, in contrast, believes that the ancestors of the O'odham were "invaders" from further south in the Sonoran Desert. "Their northward expansion into former Hohokam country I see as relatively recent, perhaps stimulated by Old World diseases that swept through northwestern New Spain every few years (Reff 1991:16, 132-274; Roberts 1989)," Rea argues. "This view is not inconsisent with archaeological evidence to date (Ravelsoot and Whittlesey 1987). The protohistoric Piman period in southern Arizona thus may be temporally shallow" (Rea 1997:8).

Shaul and Hill (1998), in contrast, present evidence that the center of maximum linguistic diversity in Tepiman is in the north, suggesting that Proto-Tepiman may have originated there. They also examine both phonological and lexical borrowing between Yuman and Tepiman languages, expanding upon arguments made by Shaul and Andresen (1989).

The Proto-Tepiman word for water, *suu-dagi for example, is found in some form in all Tepiman languages, even though the Proto-Uto-Aztecan root for water is **pa-. The Tepiman languages retain that Proto-Uto-Aztecan root as *va- only in combining forms (e.g., wa:k "reedy place" in Tohono O'odham). The words for water itself probably derive from River Yuman words for "blue, green" including the sequence hava:sú, which the Tepimans may have interpreted as 'their-water-green/blue' and may have used in ritual contexts. Because some form of the element -su is found in all Tepiman languages, the borrowing from Proto-Yuman to Proto-Tepiman must have occurred where the two groups were in contact, i.e. in the northwestern Sonoran Desert near the Colorado River.

Shaul and Hill (1998), again following Shaul and Andreson (1989), point out that Upper Piman dialects exhibit phonemic borrowing from Yuman as well. Most salient is the hardening of a system of glides to produce a new series of stops, which sets Tepiman apart from most other Uto-Aztecan languages but is also found in River Yuman. "In Tepiman, the Proto-Uto-Aztecan (PUA) glides **y and **w, and the PUA labiovelar **k^w 'hardened,'" they note. "PUA **y became Proto-Tepiman (PT) *d. PUA **w became PT *g. PUA **k^w became PT *b. This resulted in a distinctive array of 'voiced' stops, *b, * d, *g, in all the Tepiman languages, contrasting with a 'voiceless' series, inherited unchanged from PUA as *p, *t, *k," (Shaul and Hill 1998:379). The PUA word for 'nose' (yaka), for example, became daka in PT, da:k in Tohono O'odham and daka in Northern Tepehuan but yahká in Guarijio and yeka in Yaqui. "Comparative lingiustic method argues that it is more parsimonious in such a case to assume that the change took place before the breakup of Proto-Tepiman into its present daughter languages rather than to assume that the change arose independently in every single daughter language," Shaul and Hill (1998:379) argue.

According to their reconstruction, Proto-Tepiman and River Yuman groups lived in close association with one another in southwestern Arizona for several centuries around 1,000 A.D. "Speakers of Proto-Tepiman and Proto-River Yuman were probably in a sufficiently intense contact with one another for a substantial population of bilinguals to exist, among whom convergences developed in the sound system, the morpho-syntactic system, and to a limited degree in the lexicon of the two languages," Shaul and Hill (1998:392) postulate. They also contend that Proto-Tepimans expanded from the north to the south rather than vice-versa. Following and elaborating upon the argument of Fowler (1983), Shaul and Hill believe that the Proto-Tepiman word *ha:sani refers specifically to the saguaro (Carnegiea gigantea), as it does in modern Akimel and Tohono O'odham, and not just to any tall cactus with edible fruit. If so, Proto-Tepiman would have originated within the biogeographic range of the saguaro, rather than further south. Proto-Tepimans were not Mesoamerican immigrants to the Sonoran Desert. On the contrary, Proto-Tepimans, Proto-River Yumans, and probably some Zunis [2] part of the Late Classic Hohokam system.

O'odham and Hohokam Regional Systems in the Late Classic Period

Shaul and Hill do not make the same claim for speakers of the modern Upper Piman languages, Tohono O'odham and Akimel O'odham. The most innovative modern dialect of Tohono O'odham—Aji/Totoguañ, which dominates most of the Papagueria, including areas where Hohokam archaeological remains are located—may instead reflect people who "spread into abandoned Hohokam regions in the late prehistoric period. This is also consistent with Aji-Totoguañ oral tradition that they represent a later emergence from underground, which resulted in their displacing autochthonous peoples—the Hohokam—in their region" (Shaul and Hill 1998:391). In other words, the ancestors of the modern O'odham may indeed be Rea's invaders from the south.

None of this evidence is conclusive. When all the strands are woven together, however, one possible scenario emerges—a scenario that incorporates the insights of Rea as well as Shaul and Hill. A thousand years ago, Proto-Tepimans lived in close proximity to River Yumans and may have participated in the Classic florescence of the Hohokam. Other Proto-Tepimans, in contrast, moved south through the Sonoran Desert and into the foothills and upland valleys of the Sierra Madre Occidental. Gradually, linguistic and cultural distinctions emerged. Sometime in the fifteenth or sixteenth century, however, the ancestors of the Akimel O'odham and the Tohono O'odham who speak the Aji-Totoguañ dialect turned their eyes northward once again and entered Hohokam territory from the southeast, perhaps along the San Pedro or upper Santa Cruz drainages. These O'odham helped overthrow the Hohokam elites controlling the platform-mound communities along the Gila River.

No mention is made of battles fought along the Salt River. A disastrous flood in 1358 A.D. may have washed out the irrigation systems of Hohokam communities there and prompted their abandonment a century or more before the invaders arrived (Doelle and Wallace 1990). Whether the conquerors were seizing new territory or reclaiming ancestral homelands remains to be determined. Nonetheless, the modern Akimel O'odham and Tohono O'odham do not appear to be the biological or cultural descendants of the Late Classic Hohokam elites themselves (Sheridan in press).

The Pimas of the Santa Cruz and San Pedro river valleys, on the other hand, may have been descendants of the people who occupied those watersheds during Late Classic times. Doelle and Wallace (1990) argue that three smaller regional systems—organized around communities dominated by platform mounds—existed south of the Phoenix Basin, the Salt-Gila heartland of the Hohokam. One was along the lower San Pedro. Another was in the Tucson Basin. The third clustered in the eastern foothills of the Coyote Mountains. These regional systems may have reflected tribal or ethnic differences in southern Arizona.

Doelle and Wallace (1990) also believe that the settlement patterns, ceramic distribution, and architectural features of all three systems exhibit evidence of increasing warfare during the Classic period. Villages become larger and more aggregated. Large, depopulated "buffer zones" separate the systems from one another and from the Phoenix Basin itself. Some of the platform mound communities, particularly in the lower San Pedro, are located on easily defensible alluvial mesas. O'odham populations in these regional systems may have adopted many of the ideological and sociopolitical trappings of the Phoenix Basin Hohokam but resisted their attempts to control them. The battles recorded in the Hohokam chronicles of the Gila Pimas may have been part of a larger pattern of conflict that wracked Arizona in late precolumbian times. The very presence of platform mounds themselves suggests hierarchy and conflict. In the words of Henri Lefebvre, "Monumentality...always embodies and imposes a clearly intelligible message. It says what it wishes to say—yet it hides a good deal more: being political, military, and ultimately fascist in character, monumental buildings mask the will to power and the arbitrariness of power beneath signs and surfaces which claim to express collective will and collective thought" (Lefebvre 1991:143).

There undoubtedly were other regional systems as well—systems outside the orbit of the Phoenix Basin Hohokam. The Upper Santa Cruz does not seem to belong to the Tucson Basin system. There are no platform mounds south of Martinez Hill and no large late Classic sites along the river between the Tucson Basin and Palo Parado south of Tumacácori. The frequency of Gila Polychrome plummets while the percentage of pottery types from the south such as Trincheras ??? rises (Henry Wallace, personal communication). Archaeologist David Wilcox, who reanalyzed Charles DiPeso's excavation of Palo Parado Ruin, concluded that Palo Parado "appears to lie near the northern end of a settlement system different from the one in the Tucson Basin" (Wilcox 1987:241). Wilcox goes on to say, "Other sites similar to the Classic component at Palo Parado exist in the Rio Rico area, and sites as far up the Santa Cruz as the San Rafael Valley exhibit analogously aggregated compound site structure (Sauer and Brand 1931; Damson 1946; personal observation)." Because most of the dwellings at Palo Parado were burned at the same time, Wilcox believes that the community may have been the victim of "regional warfare in the 14th century"—another indication of escalating tension and bloodshed in late Classic times (Wilcox 1987:239). [3] Perhaps the inhabitants of Palo Parado and other settlements in the upper Santa Cruz belonged to that division of the O'odham who went on to topple the platform-mound communities along the Gila a century or two later.

The Sobaipuris and Their Neighbors

Conflict continued to divide the people of southern Arizona when Europeans first settled there in the late 1600s, two centuries after Hohokam civilization disintegrated. When Jesuit missionary Eusebio Francisco Kino crisscrossed the Pimería Alta in the early 1690s, he mentioned two cases of warfare among the O'odham themselves. The first pitted the western Sobas of the Río de la Concepción drainage against the eastern Pimas of the Dolores Valley, who sought revenge because the Sobas had killed Podenco, headman of Dolores, a decade or so earlier. The second matched Sobaipuris under El Coro in the middle San Pedro Valley against ??? Sobaipuris under El Humari along the lower San Pedro. Meanwhile, nomadic groups identified as Sumas, Mansos, Janos, Jocomes, and Apaches raided O'odham living along both the San Pedro and the Gila. The region was no more stable in the late 17th century than it had been in the 14th and 15th centuries.

The political dynamics of the 1500s and early 1600s are less clear. Alvar Núñez Cabeza de Vaca and his fellow náufragos (shipwrecked ones) may have nicked southeastern Arizona in the 1530s, but recent ethnobotanical evidence suggests that they traveled across northern Mexico, not the southwestern United States (Olson et al 1997). [4] The expedition of Francisco Vásquez de Coronado passed to the east, perhaps through the San Pedro Valley, but neither Coronado nor his chroniclers had much to say about that stretch of their journey. Not until the advent of Kino in 1687 do the O'odham of southern Arizona begin to take shape on European paper. The descriptions of Kino and his companions, particularly Juan Mateo Manje, provide our first indisputable window into the O'odham world. Unfortunately, that window opens more than a century after European influences, including European diseases, may have buffeted O'odham society and culture (Reff 1991).

Kino and Manje distinguished among different groups of O'odham, including the Sobas along the Río de la Concepción and the Papabotas [Papagos] between the Tucson Basin and the Colorado River (Fontana 1995; Sheridan 1988). The distinction between those they called Pimas and those they named Sobaipuris, a term with no known etymology, is less clear. According to historian Herbert Bolton, "In the southeastern portion of the area, nearest the Spanish settlements, were the people then regarded as the Pimas proper. They lived on all the slopes of the watershed which zigzags roughly westward from Huachuca Mountain to Nogales, having villages on the upper waters of the south-flowing Sonora, San Miguel, and Cocóspora [sic; Cocóspera] rivers, on the west-flowing San Ignacio and Altar, and on the north-flowing San Pedro and Santa Cruz" (Bolton 1984:247). He goes on to say, "North of the Pimas proper lived the Sobaípuris (Sobajípuris). A line drawn eastward from Tubac through Fairbank approximates the old boundary between the two peoples. Of the Sobaípuris there were three groups: one living all down the San Pedro River northward from the vicinity of Fairbank; another on the middle Santa Cruz between San Xavier del Bac and Picacho; a third on the Gila River from the Casa Grande westward nearly to the Bend of Gila River" (Bolton 1984:247).

If Bolton's reconstruction is correct, the distribution of Pimas in the Santa Cruz and San Pedro watersheds in the late 17th century parallels the cultural and political divisions of the late Classic. But what do Kino and Manje themselves say? Kino first encountered Sobaipuris in January 1691, when he accompanied Father Visitor Juan María Salvatierra on his inspection of the O'odham communities of the Río Magdalena-Altar-Concepción watershed. Kino and Salvatierra planned to return to Cocóspera but Sobaipuris from Bac and Tumagacori [Tumacácori] intercepted them with crosses and implored them to visit their settlements as well.

At Tucubavia, in the headwaters of the Altar, the two missionaries veered northeast to the "Valley of Guebavi" and reached Tumacácori. There, according to Kino, they met with "some of the Sobaipuris [sic] headmen, who had come twenty and twenty-five leagues from the north," presumably from Bac and other O'odham communities in the Tucson Basin (Bolton 1919 1:119). Kino noted that there were "more than forty houses close together" at Tumacácori. Kino and Salvatierra then headed south to "the ranchería of Guebaui [Guevavi] and the valley and ranchería of Santa Maria [Santa María Suamca, now the pueblo of Santa Cruz, Sonora], a journey of fifteen leagues, where we remained five days, catechizing and baptizing infants and adults" (Bolton 1919 I:120). Kino's short account seems to indicate that the Pimas from the Tumacácori-Guevavi area were in close contact with, and may even have been subordinate to, Sobaipuris from the north.

Kino visited the Sobaipuris a second time in late August and early September of 1692. His first destination was the "valley and river of Santa María" [the Santa Cruz], where he "found the natives very affable and friendly, and particularly so in the principal ranchería of San Xavier del Bac, which contains more than eight hundred souls" (Bolton 1919 I:122). From there he turned east and dropped into the San Pedro Valley. "Captain Coro and the rest of them received me with all kindness," he noted. Then he added, "It is true that I found them still somewhat less docile than the foregoing of the west [the Sobaipuris at Bac]" (Bolton 1919 1I:123). Despite his relentlessly optimistic promotion of the Pimería Alta, Kino's comment about the Sobaipuris of the San Pedro suggests that these O'odham were true frontiersmen—independent, a little unruly, willing to ally themselves with the charismatic Jesuit but not ready to accept the discipline missionization entailed.

Following the 1695 rebellion in the western Pimería Alta, many Spaniards accused the Sobaipuris, who had not participated in the revolt, of allying themselves with the Jocomes and stealing livestock. "It was falsely reported, also, that these Pimas were involved in the tumults and revolts of Taraumara, on the testimony of the Taraumares themselves," Kino angrily responded, "but the Taraumares could not have been speaking of the Pimas of this Pimería, who are more than one hundred and fifty leagues distant from the Taraumares, but only of the Pimas near them, who are those of Tapipa and near Yecora" (Bolton 1919 I:163).

Kino devoted much of his energy countering these accusations and others, including one that the "Pimas of the interior [to the north] and their neighbors were such cannibals that they roasted and ate people" (Bolton 1919 I:163). Because he realized that Sobaipuri goodwill was essential if the Jesuits were ever going to expand their missions to the Gila River or reconvert the "apostate" Moquis (Hopis), Kino wanted to forge a lasting alliance between the Sobaipuris and the Spaniards (Bolton 1919 I:184).

Kino made a third sweep through Sobaipuri country during the winter of 1696-97. This time he traveled down the San Pedro first, baptizing the young son of El Coro. Then he veered west and visited the Sobaipuris of San Xavier del Bac, "where I was received with all love by the many inhabitants of the great ranchería, and by many other principal men, who had gathered from various parts adjacent."

Kino returned to San Pablo de Quiburi in March 1697, but his largest expedition to the Sobaipuris began in November of that year. He and his frequent traveling companion, Juan Mateo Manje, left Dolores and rendezvoused with Captain Cristobal Martin Bernal and twenty-five soldiers from the presidio of Fronteras at Quiburi. "We found the Pima natives of Quiburi very jovial and very friendly," Kino observed. "They were dancing over scalps and spoils of fifteen enemies, Hocomes and Janos, whom they had killed a few days before. This was so pleasing to us that the Señor Captain Christobal Martin Bernal, the Señor alférez, the sergeant, and many others, entered the circle and danced merrily with the natives" (Bolton 1919 I:168-69). The victory dance reassured Bernal that the Sobaipuris were not in league with the Jocome, Janos, Suma, Manso, and Apache raiders who were running off livestock from the communities of northern Sonora and Nueva Vizcaya.

The expedition continued down the San Pedro River, passing through a depopulated stretch of the floodplain that extended for sixty miles north of Quiburi. The Jesuit and his companions entered villages under the leadership of Humari, who had been at war with Coro until Kino made peace between the two O'odham leaders. Once it reached the junction of the San Pedro and Gila rivers, the expedition followed the Gila as it curves northwest along the Tortilla Mountains and then heads south across the desert flats where the cotton fields of Florence and Coolidge now spread under the sun. Their destination was the great ruin of Casa Grande, where they admired the high, thick walls and "marveled" at the "great aqueduct" of the Hohokam that once irrigated thousands of acres of cultivated land. Kino noted "six or seven rancherías of Pimas Sobaipuris" nearby, "all of whom in all places received us very kindly, with crosses and arches erected and with many of their eatables, and with great pleasure to themselves, gave us many little ones to baptize" (Bolton 1919 I:172).

Kino's brief entries draw no explicit distinction between the O'odham of the Upper Santa Cruz Valley and the Sobaipuris to the north. Describing his second expedition northward in late summer 1692, however, he wrote, "I went in, with fifty pack-animals, my servants, and some justices, to the Sobaipuris, both of the north and the northeast. The latter are in the valleys of the river of Quiburi, to the east, and the former are in the valley and river of Santa María, to the west" (Bolton 1919 I:122). That statement suggests that Kino considered both groups Sobaipuris.

In September 1698, on the other hand, Kino was ordered to follow the "Río Grande"—the Gila River—until it reached the "Sea of California." His "Relación Diaria" of the expedition, translated by Fay Jackson Smith, contains a passage open to another interpretation. Kino, Captain Diego Carrasco, and their O'odham companions pass through San Luis de Baconacos, Guevavi, and San Cayetano (Tumacácori). After leaving Tumacácori, Kino wrote, "At sunset we traveled on another three leagues to sleep in a place from where on the following day we could reach more easily the Sobaipuris and their first great ranchería, San Francisco Xavier del Bac" (Smith et al 1966:13). This passage suggests a possible distinction between the Pimas of the upper Santa Cruz Valley and the Sobapipuris of the Tucson Basin.

Manje, a much better journal keeper than Kino, left two accounts of their 1697 expedition to the "pimas sabahípuris" (Burrus 1971:358). One forms part of Manje's famous Luz de tierra incógnita. Jesuit historian Ernest Burrus published a second, more complete account in its original Spanish in 1971. Kino and Manje arrived in San Lázaro at the bend of the Santa Cruz on November 5, where Manje counted seventy people and "more than twenty houses" scattered a short distance from one another. It was a beautiful spot, "adorned with many cottonwoods, pretty meadows to support livestock, and fertile fields, all under irrigation" (my translation from Burrus 1971:358). The two men continued on to Santa María, where Manje counted 200 O'odham and commented favorably about the agricultural fertility of the valley and the abundant grasslands surrounding it. He also noted that the inhabitants of Santa María wore clothing of cotton and animal skins (gamuzas). In neither of his two accounts did Manje refer to the inhabitants of these two communities on the Upper Santa Cruz by any specific term.

In his first account, Manje described Coro as the "chief Indian of the Pima nation" (my translation from Burrus 1971:336). In his second account, he noted that the two "Hocome" boys captured by O'odham in the village of Jiaspi on the San Pedro had been taken by the "sobahípuris de San Xavier del Vac" (Burrus 1971:363). When the expedition reached Ojío, the ranchería of Humari, Manje described the O'odham leader as "the Indian with the greatest following among all this nation of the Sobahípuris" (my translation from Burrus 1971:366). Finally, in his epilogue to the second account, Manje referred to the Indians of Quiburi, Bac, and Casa Grande as "sobahípuris." The Indians of San Cayetano de Tumagácuri (Tumacácori) and San Gabriel de Guevavi, in contrast, were not given a specific ethnic designation.

Based upon this ambiguous documentary evidence, it is difficult to determine how strong an ethnic boundary existed between the Sobaipuris and Bolton's "Pimas proper" at the end of the seventeenth century. O'odham seemed to have been able to move freely between the Upper Santa Cruz Valley and the Tucson Basin. The Sobaipuris of Bac also traveled thirty leagues to the lower San Pedro to join Sobaipuris under Humari in their battle against the Jocomes. [5] The documentary record provides evidence of hostility between Sobaipuris on the San Pedro under Coro and Humari but not between the Sobaipuris of Bac and their neighbors to the south. The ethnic boundary between the O'odham of Tumacácori and Guevavi and the Sobaipuris to the north and northeast did not seem to be as strong as the Late Classic cultural boundary that separated the inhabitants of Palo Parado and other sites in the Upper Santa Cruz Valley from those living in the platform-mound communities in the Tucson Basin or the lower San Pedro Valley.

A mosaic begins to emerge as these bits and pieces of documentary and archaeological evidence are fitted together. Southeastern Arizona at the end of the seventeenth century was a dynamic frontier, just as it had been in the Late Classic. The Sobaipuris may have been descendants of the O'odham who helped overthrow elites governing platform-mound communities along the Gila. O'odham in the Upper Santa Cruz Valley may have moved northward in their wake. If Shaul and Hill's linguistic scenario is correct, Tepimans migrated southward as Classic Hohokam society flourished. Some of those Tepimans may have reversed direction as Late Classic Hohokam society decayed.

Two additional factors intensified the fluidity of O'odham society at European contact: 1) pressure from nomadic raiders to the north and east, and 2) the devastation of Old World epidemic diseases. When Kino and Manje reached the ranchería of Santa Cruz de Jauanipicta, or Gaybanipitea, on the Babocómari River on December 7, 1697, Sobaipuris were celebrating a victory over sixteen Jocomes and Janos who had tried to run off their livestock. The Sobaipuris had killed thirteen of the raiders and seriously wounded the other three. Two days later, Kino and Manje found Sobaipuris in Coro's village of Quiburi performing a circular dance around a tall pole dangling nine scalps, three bows and arrows, and other paraphernalia taken from the same raiders (Burrus 1971:360-61). The introduction of Old World livestock had already made Sobaipuris along the San Pedro targets of raids.

The ethnic identity of the raiders was fluid as well. In Manje's second account of the 1697 expedition, the dead enemies were Jocomes. His first account, on the other hand, states, "We also found here [Jiaspi] the scalps of six enemy Apaches that they had killed a little while ago, and two boy prisoners that Indians from the great ranchería of Bac had carried away. For that reason, the soldiers disabused themselves of their opposition [to the Sobaipuris] and [realized] they [the Sobaipuris] had no confederation with the enemy Apaches" (my translation of Burrus 1971:338). Manje's apparent confusion was understandable. He was providing a snapshot of the southeastern Arizona frontier at a time when the Apaches were relentlessly moving south and west, assimilating or displacing the Jocomes and other poorly understood groups like the Janos, Mansos, and Sumas (Sheridan in press).

There were no such discrepancies in his account when he and Kino traveled along the Gila. The area to the north and east was clearly identified as Apache territory. By the late 1600s, Apache groups dominated the mountains along the Gila, including the northern slopes of the Pinaleño Mountains (Sierra Florida). The Apaches shared the Chiricahua and other southeastern ranges with the Jocomes, Janos, Sumas, Chinarras, and Mansos. By the early 1700s, however, these groups disappeared from the Spanish documentary record, at least in southeastern Arizona. For the rest of the colonial period and beyond, the Apaches controlled the sky islands and grassland basins east of the San Pedro River (Sheridan in press).

Not long after Kino and Manje's expedition, hostilities between the Sobaipuris and their Janos, Jocome, Suma, Manso, and Apache enemies intensified. On February 25, 1698, those enemies sacked the mission of Cocóspera to the south. A month later, on March 30, a combined force of Jocomes, Sumas, Mansos, and Apaches, estimated to number about 600 by Kino, assaulted Gaybanipitea. When Coro heard of the attack, he hurried to the rescue. The enemy leader, El Capotcari, proposed a showdown between ten of Coro's bravest fighters and ten of his own, six of whom were Apaches. The Sobaipuris routed their ten adversaries and pursued the rest of the attackers, killing many of them.

After the attack, however, Coro and his people retreated to Sonoita Creek near Patagonia. Such movement may have been common among the O'odham of the San Pedro watershed. Archaeologist Deni Seymour (1989) contends that Sobaipuris along the San Pedro frequently abandoned sites and shifted the location of their settlements. They also came together into fewer but larger communities. Seymour recognizes that Sobaipuri mobility may have been in response to attacks by Janos, Jocomes, Sumas, and Apaches, but she also suggests "there is some indication that the pattern of population movement was a characteristic indigenous trait." According to Seymour (1989:220), "Archaeological evidence in favor of this position includes the extremely low density of artifacts, the lack of deep stratification, and the unsubstantial nature of the architecture on all known Sobaipuri and Upper Piman sites throughout southern Arizona."

At present, there is somewhat of a disjunction between the documentary and archaeological record. Manje enumerated 4,450 O'odham living in thirty rancherías along the San Pedro, Gila, and Santa Cruz rivers. He also described extensive Sobaipuri fields of corn, beans, squash, and cotton irrigated by earthen acequias (canals) along both the San Pedro and the Santa Cruz. As archaeologist Bruce Masse (1981:46) points out, however, "Ground stone is weakly represented at Sobaipuri village sites, as are domesticated plant remains. The archaeological record suggests that agriculture played only a minor role in Sobaipuri subsistence."

This disjunction may be more apparent than real. "A question that has plagued archaeologists for years is: If so many Classic period sites are known, where are the sites that date to the 16th and 17th centuries?" Doelle and Wallace (1990:256) observe. "Our response is that the sites are there and they have gone unrecognized for the most part." More archaeological survey and excavation focusing on this so-called "protohistoric" period may bridge the Late Classic and early mission eras.

That bridge may turn out to be tumultuous and pox-ridden. The Late Classic stratified regional system of the Hohokam in the Phoenix Basin was disintegrating about the same time the trading center of Paquimé (Casas Grandes) was collapsing in northwestern Chihuahua. The resulting power vacuum sucked less sedentary peoples into the region from all directions—O'odham from the south and west, Janos, Jocomes, Mansos, Sumas, and Chinarras from the east, Apaches from the north—before the Europeans arrived.

Silent invaders undoubtedly penetrated southern Arizona as well—smallpox scabs riding in raw cotton passed along Indian trade routes, microbes of measles lurking in the mucus or saliva of infected individuals fleeing Spanish northward expansion. The populations of North and South America never experienced the ravages of Old World contagious diseases—smallpox, measles, influenza, typhus, malaria—until 1518, when a pox-ridden soldier from the Narváez expedition landed on the Gulf coast of Mexico in pursuit of Cortés. His arrival triggered a terrible pandemic that devastated the densely populated empires of Mesoamerica. Anthropologists Henry Dobyns (1983) and Steadman Upham 1986) believe this pandemic may have reached the Southwest as well.

Anthropologist Daniel Reff (1991) takes issue with Dobyns and Upham, pointing out that the lowlands of Nayarit and Sinaloa were still densely populated when Nuño de Guzmán raped and pillaged his way north along the Pacific coast of Mexico in 1530 and 1531. Reff believes that epidemics of Old World diseases did not arrive in Sonora or Arizona until sometime after 1539, when Fray Marcos de Niza described large populations in the region. Nonetheless, Reff agrees with Dobyns that Old World diseases may have contributed to the collapse of the Hohokam sometime in the mid-16th century.

Most archaeologists disagree, contending that dates for the abandonment of platform-mound communities in the Phoenix and Tonto basins cluster in the 1300s and 1400s, not the 1500s. But it is hard to believe that the O'odham did not suffer from other pandemics eating away at the flesh and fabric of other Southwestern societies in the late sixteenth and seventeenth centuries. The impermanence of O'odham settlements may have been an adaptation to frontier conditions. But when Europeans like Kino and Manje first entered southern Arizona, the Pimería Alta must have been a disease frontier as well as a political frontier. The Sobaipuris and other O'odham may have moved frequently to flee the "wandering sicknesses" from the south, burning their simple brush dwellings (kis) to destroy the contagions that had ravaged their neighbors and kin (Bahr et al 1974; Dobyns 1959. [6]