John Ware is a Southwestern anthropologist and archaeologist whose teaching and research concerns focus on the Native American cultures of the northern Southwest, where he has worked for over 40 years. From 2001 until his retirement in 2014, Dr. Ware served as executive director of the Amerind Foundation in southern Arizona. Before that he directed the Laboratory of Anthropology and was founding director of the Museum of Indian Arts and Culture in Santa Fe. He is the author of A Pueblo Social History: Kinship, Sodality, and Community in the Northern Southwest (SAR Press, 2014).
Pre-Colonial Southwestern Migrations: A Critical Review
During the processual 1960s and 70s the word “migration” was rarely uttered in polite company (there was a great deal of forced conformity and very little polite company in those contentious days). During the post-processual 1980s and 90s it was possible to once again speak and write the ‘m’ word, as long as it was modified by words like “contingency” and “agency.” The discipline was too immersed in post-modernism, post-colonialism, and post-various other things to engage in a serious discourse on migration. In the last decade or so there have been several important treatments of the subject and migration has crept slowly back onto the stage of subjects permissible to address (e.g., Clark 2001; Clark and Lyons 2012; Cordell et al., 2007; Duff and Wilshusen 2000; Lyons 2003; Ortman 2012), although graying eye brows are still occasionally raised in protest.
In this paper I offer an overview of recent attempts to explain the migration of Kayenta Ancestral Pueblos to southern Arizona and Mesa Verdeans to the northern Rio Grande. Although the details are still being worked out, the Kayenta to southern Arizona case is now more or less settled. A strong consensus is also emerging for the Mesa Verde to Rio Grande case around the award-winning book Winds from the North by professor Scott Ortman of the University of Colorado. In this paper I unpack Ortman’s case and compare it to alternative narratives proposed by a handful of Rio Grande archaeologists led by Eric Blinman of the Museum of New Mexico. In a final section I summarize a discussion of the paper that took place at the School for Advanced Research in Santa Fe in December, 2015.
Western Pueblo Migrations
The case for Kayenta migrations to the Mogollon Rim and beyond to the Gila and San Pedro River valleys and the Tonto and Phoenix basins of southern Arizona has been building for years and is now, I think, almost universally accepted. First recognized by Emil Haury during his excavations of the Maverick Mountain component at Point of Pines Pueblo in the 1940s (Haury 1958), recent work by archaeologists primarily from the University of Arizona and Archaeology Southwest of Tucson have documented unambiguous site unit intrusions of Kayenta migrants stretching from the Gila Valley in western New Mexico to the Salt River Valley of central Arizona. The most thoroughly documented cases are two sites in the northern San Pedro Valley near Redington Pass, just east of Tucson.
Reeve Ruin and the Davis Ranch Site were excavated by the Amerind Foundation in the 1950s (Di Peso 1958; Lyons in press). The main component at the Davis Ranch Site consists of a semi-circular line of small “pocket” pit houses that form an arc to the north of a rectangular Western Pueblo-style kiva, and immediately west of the kiva a room block enclosing a small plaza. The pocket pit houses are similar to the earliest migrant architecture at Point of Pines and tell a similar story: a group of migrants arrived around 1300 CE and built expedient pit structures as they assembled materials for construction of a masonry pueblo. Before work began on the surface rooms, however, a kiva was constructed, complete with southern recess, perimeter bench, central slab-lined fire pit, sipapu, and two lines of floor anchors for upright looms (Figure 1). If the kiva at Davis Ranch were transported 200 miles north it would be indistinguishable from late prehistoric Hopi kivas.
In addition to the western pueblo kiva, the identity of the Davis Ranch migrants is preserved in domestic architectural details, in portable material culture, and perhaps even in food preferences (Patrick Lyons, personal communication). Some of the surface rooms at Davis Ranch have Kayenta-style “box entryways” found throughout the Kayenta heartland in northeastern Arizona. These entryways consist of a step from the ground surface down into a stone-lined box and over a lateral stone threshold, with the anterior slab of the box serving as a deflector for an interior slab-lined hearth (Figure 2). In addition, the use of unshaped sandstone blocks and slabs set in copious amounts of earth mortar echoes Kayenta masonry styles, and ground stone, ceramic containers, and other technologies have northern derivations as well.
The most diagnostic of northern ceramic forms are “perforated plates,” part of a pottery production technology that are present from the sixth through the thirteenth century in the Kayenta region and appear at sites in southern Arizona in the late thirteenth and throughout the fourteenth centuries (Figures 3). These plates appear to have served as base molds for coil and scrape pottery production (Lyons and Lindsay 2006). Scholars have debated the function of the perforations but there is little consensus, and they may simply be style elements or potter identity markers.
Locally manufactured Roosevelt Red Ware (Salado polychromes) dominates the decorated ceramic assemblage at Davis Ranch. Maverick Mountain series pottery is the second most frequent “ware,” especially Tucson polychrome, a locally produced Maverick Mountain type, which is a ceramic and iconographic “cognate” of Tusayan Polychrome—another obvious link to the north (Lyons in press). Polychrome ceramic technology can be traced from the four corners south to large late Pueblo III-IV villages on the Mogollon Rim and beyond to Point of Pines and from there to the Gila and San Pedro River Valleys in southern Arizona: an unbroken style horizon that would soon morph into Roosevelt Red Ware that would help to define a Western Pueblo diaspora in southern and central Arizona.
Reeve Ruin is located on a bluff overlooking the San Pedro River floodplain less than a quarter mile southwest of Davis Ranch (Di Peso 1958). A contemporaneous masonry pueblo with ceramic assemblage virtually identical to Davis Ranch, Reeve has all the earmarks of a fortified refuge for the Davis Ranch settlement. Reeve is separated from the San Pedro by a 60-foot vertical rock face, and its north and west approaches are via steep, easily defended trails. The only easy access to Reeve is from the southwest, along the top of the low bluff, but the narrowest point of this approach is protected by a masonry wall. The obvious defensive positioning of Reeve Ruin suggests that the intrusion of Kayenta peoples into the San Pedro Valley was locally contested. Although there are no contemporaneous Hohokam communities in the area around Redington Pass, just to the north near the confluence of the San Pedro and Aravaipa Creek are a large cluster of Hohokam settlements. Dispersed before the arrival of Kayenta immigrants, the Hohokam aggregated into about a dozen walled communities with small internal platform mounds after the migrants arrived, providing additional evidence of conflict during the early years of contact (Clark and Lyons 2012).
In less than a generation, evidence of accommodation between residents and newcomers is apparent. The Davis Ranch kiva is filled with trash sometime in the mid-1300s, suggesting a lapse of plateau-based ritual, and Roosevelt Red Ware replaces northern-derived Maverick Mountain series ceramics in the San Pedro and other western pueblo enclaves throughout southeastern Arizona. By 1350, Roosevelt Red Ware is the dominant decorated ceramic not only at Kayenta enclave sites but also at Hohokam platform mound villages, where the local manufacture of red-on-buff/brown ceramics was apparently curtailed in favor of importing Roosevelt Red Ware from former enclaves (Lyons 2012). By the end of the 1300s settlements in the San Pedro had shifted downstream toward its confluence with the Gila and included both Hohokam and Kayenta descendant groups, whose ethnic distinctions were becoming increasingly blurred. In the early 1400s, the entire region was depopulated. Where data is available, it appears that similar histories of initial conflict followed by close interaction and finally demographic decline and abandonment played out throughout the eastern Hohokam and western Mogollon regions of central and southeastern Arizona. From 1325 to the end of the pre-contact period, much of the Roosevelt Red Ware production appears to have been carried out by Kayenta potters and major stylistic shifts, such as the transition from Gila to Cliff Polychrome in the mid-to-late 1300s, occurred simultaneously across the entire Salado region, suggesting a closely linked Kayenta diasporic community (Clark et al. 2013; Lyons and Clark 2012).
Patty Crown’s seminal study of Roosevelt Red Ware (1994) describes its consistent use of design elements that symbolize water and fertility, exemplified by cloud, bird, and horned/plumed serpent motifs. Crown (1994) and others (e.g., Clark et al. 2013) have argued that these symbols were associated with an inclusive ideology that spread throughout the southern southwest in the late prehistoric period, revitalizing a Kayenta community in diaspora and perhaps challenging the more hierarchical ideology of local Hohokam platform mound communities (Borck and Clark forthcoming). By 1450, of course, the Hohokam world was in disarray and at least some western pueblo groups began a reverse migration to northeastern Arizona and northwestern New Mexico. In the Tusayan area of northern Arizona room counts double in the 1350 to 1450 era (Clark et al., in preparation), and migrants to Zuni bring with them Roosevelt Red Ware and human cremation practices, whose southern origins are unmistakable.
Eastern Pueblo Migrations
If many of the details of Western Pueblo migrations to southern Arizona have been worked out and the sequence is more or less accepted, contemporaneous migrations of ancestral Pueblo people to the Rio Grande Valley in the late thirteenth century have been debated for over a century and, until recently, consensus was as elusive as ever. The rich archaeology of the San Juan region of the northern Southwest has attracted explorers and archaeologists from the mid-nineteenth century to the present. One of the earliest themes to emerge from those investigations was the thirteenth century depopulation of the region, which every medium from the popular press to textbooks to Park Service interpretive panels describe as the “mysterious disappearance” of the Anasazi. Of course, archaeologists have known for years that the decline of population in the San Juan River basin immediately preceded the rather dramatic growth of pueblo population in the Rio Grande basin, and most assumed that migrations helped account for the demographic reversal (Figure 4). Anthropologist Alfonso Ortiz delighted in telling audiences that “the Anasazi didn’t disappear, they’re running bingo parlors in the Rio Grande” (today, the bingo parlors have morphed into casinos, but Ortiz’s point remains the same).
The problem with the bingo parlor theory has always been the absence of unambiguous physical evidence of migration at the Rio Grande destination. If Mesa Verde people moved from the northern San Juan to the northern Rio Grande, they didn’t bring much San Juan ritual or domestic architecture and portable material culture with them (Lipe 2010). A.V. Kidder expressed his concern about the lack of physical evidence in the conclusions of his classic Introduction to the Study of Southwest Archaeology: “I think the connection between these two sets of phenomena, abandonment of the outlying districts and sudden increase in population in the central areas, cannot be mistaken. The puzzling thing about it is that the incoming people brought with them so little of their local culture” (1924:342). The debates following Kidder’s statement are worth reviewing as background for the contemporary debate (a more comprehensive treatment can be found in Ortman 2012).
In 1935, H.P. Mera, who did pioneering work on Rio Grande ceramic series and was the first to document many of its large prehistoric sites, argued that Tanoan-speaking peoples were likely long-term residents of the Rio Grande and that Keresan-speaking migrants from the San Juan brought Mesa Verde tradition pottery to the Rio Grande, influencing a fusion of Mesa Verde and local Santa Fe white ware to produce Galisteo Black-on-white. Galisteo B/W dates to the late thirteenth century, so its timing coincides with purported migrations out of the central Mesa Verde region (Mera 1935). To this day, some scholars cling to Galisteo as a “cognate” of Mesa Verde Black-on-white, despite the fact that recent studies (Wilson 2008) suggest that Galisteo is simply a local variety of Santa Fe Black-on-white with some design and paste similarities to northern white wares.
Mera’s was the principal game in town until the late 1940s when archaeologist Eric Reed brought the study of contemporary language distributions to the debate (Reed 1949). Reed agreed with Mera that Tiwa speakers were likely the original Rio Grande inhabitants and that Keresan speakers expanded their range from the San Juan into the lower Rio Jemez and the Santo Domingo and Galisteo Basins in the late thirteenth and early fourteenth centuries. He departed from Mera, however, in arguing that the Keres derived mostly from the southern San Juan Basin and Cibola regions and that Tewa was the principal language of the northern San Juan. According to Reed, the Keres and Tewa migration pulses of the thirteenth century combined to drive a wedge between northern and southern Tiwa speech communities on the Rio Grande.
Eric Reed’s model of how migration contributed to historic language distributions would have a profound influence beyond archaeology. A year after Reed’s publication Fred Eggan published his classic Social Organization of the Western Pueblos (1950) in which Reed’s migration arguments were adopted as the latest word from the Southwest archaeological community. Building on Reed, Eggan argued that disruptive migrations from the northern San Juan combined with Spanish acculturation, especially the compulsory adoption of Catholic marriage practices, to transform the Tewa into the bilaterally organized communities described by twentieth century ethnographers, while Keresan migrants were able to retain elements of their ancestral matrilineal organizations. For archaeologists at least, Eggan’s “Keresan Bridge” hypothesis would become the dominant social history narrative of the Pueblos for the next fifty years (despite challenges from Robin Fox  and a few others).
In 1955 Reed teamed up with Fred Wendorf (Wendorf and Reed 1955) to present a new cultural chronology for the Rio Grande Valley. Departing from the iconic Pecos Classification, the central fact of this alternative chronology was that major investments in sedentary farming were not made in the Rio Grande until the beginning of the second millennium CE, more than a thousand years after corn became the principal source of food calories in the San Juan (Vierra and Ford 2006). In Wendorf and Reed’s chronology, a pit house dwelling, Basketmaker-like Developmental Pueblo (CE 600-1175) adaptation persisted into the twelfth century in the northern Rio Grande and was eventually superseded during the Rio Grande Coalition (CE 1175-1325) by dispersed farming villages that would coalesce to form the foundations of later Classic Period (CE 1325-1600) plaza pueblos. In characterizing their new phase system for the Rio Grande, Wendorf and Reed emphasized material culture continuities between Developmental and Coalition/Classic periods, acknowledging that there was little material evidence of migrations from the northern San Juan to the Rio Grande aside from some Mesa Verde-like black-on-white pottery in buried horizons at several Galisteo Basin Classic era villages (the aforementioned Galisteo Black-on-white). Wendorf and Reed’s migration story ended up looking more like Mera’s than Reed’s earlier model (the irony may have escaped Eggan, who didn’t revisit his social history model until the early 1970s [Eggan 1972]).
Reed’s earlier Tewa migration theory was resurrected in the 1960s, however, when Charles McNutt published on his excavations at the Tesuque Bypass Site north of Santa Fe (McNutt 1969). The site consisted of at least three temporal components ranging in age from late Developmental through early Coalition (circa 900 to the 1200s CE), and McNutt saw evidence of migration from the San Juan in nearly every component. He argued that the earliest Red Mesa and Kwahe’e Black-on-white components (900s-1100s) were tied ceramically and architecturally to the central San Juan Basin: “It is most probable that the earliest Puebloans in the northern Rio Grande Valley were immigrants from somewhere within the area south of Mesa Verde, west of the east Puerco, north of the Datil Mountains, and east of the Chuska-Lukachukai-Carizzo Mountains” (1969:100).
Seven miles north of Tesuque Bypass is the contemporaneous Pojoaque Grant Site (LA 835) consisting of a cluster of over a dozen 10-20 room unit pueblos and a great kiva that is also dominated by Red Mesa, Kwahe’e, and neck-banded utility ware. This was additional evidence, McNutt argued, for a migration from the central San Juan in the tenth through the twelfth century: "The apparently uninterrupted change from Red Mesa B/W to Kwahe'e B/W at LA 835, the specific design treatment of the latter pottery, and the appearance of the Great Kiva all suggest that the greater part of the immigrants were from the northern Chaco-San Juan area" (1969:105).
McNutt argued that migration pulses continued in the 1200s, citing several lines of evidence, including: Coalition period population increases and the growth of larger pueblos, a shift from mineral to carbon paint in Santa Fe Black-on-white, a shift from trough metates to the slab metate-mealing-bin complex, side notched to corner notched projectile points, and the appearance, for the first time, of full grooved axes, bird bone awls, and domestic turkeys in the Rio Grande Coalition. Of course, these changes were occurring over a large area of the plateau about the same time, so inter-regional trade and other forms of interaction may have brought these new technologies to the Rio Grande (Cordell 1979). The absence of pilasters in Rio Grande pit houses/kivas troubled McNutt, but he suggested that “small social units drifting into the area rather than a planned and organized population movement” would likely have had minimal effect on local religious architecture (1969:109). (We will return to this important point later in the paper).
In a seminal article from the early 1970s, Richard Ford, Albert Schroeder, and Steward Peckham (Ford et al. 1972), summarized the then current knowledge of ceramic distributions, historical linguistics, and inferred Plateau to Rio Grande migrations, and they famously split on the subject of Tewa origins. All agreed that the Tiwa were long-term residents of the Rio Grande but they disagreed on the reasons for the Tiwa split. Schroeder and Ford agreed with Reed’s 1949 study that Keres and Tewa migrations from the San Juan drove a wedge between the northern and southern Tiwa language communities. Peckham concurred with the Keres intrusion but argued that the Tewa developed independently in the Tewa Basin from 900 CE onward. Both arguments relied heavily on contradictory glottochronological reconstructions of the Kiowa-Tanoan language e family (Davis 1959; Trager 1967; Whorf and Trager 1937).
This is where the debate stood until recent years when continued work in the northern San Juan and Rio Grande Valley, as well as several regional demographic modeling studies, brought migration questions to the forefront once again. A collaboration among the Museum of Northern Arizona, the Center for Desert Archaeology (now Archaeology Southwest), and Geo-Map, Inc., resulted in the Coalescent Communities GIS Database (CCD) that gave researchers a tool for modeling large-scale demographic changes in the greater Southwest from about 1200 to 1700 CE (Wilcox, Doelle, and Hill 2003). With over 3500 sites and nearly 6000 temporal components in the database, major demographic trends were immediately apparent. After nearly 1000 years of sustained population growth a brief stasis occurred in the 1200s CE (Figure 5) followed by a striking decline in overall population across most of the region. By 1450 CE, the only large concentration of population in the northern Southwest was in the northern Rio Grande, with smaller population nodes in the Estancia Basin and along the southern edge of the Colorado Plateau in the vicinity of Acoma, along the Zuni River in western New Mexico, and on the Hopi Mesas (Figure 6). Significantly, by 1450 CE there appear to be few if any large (>12 rooms) communities in the Hohokam heartland of southern Arizona (which helps explain why, almost a century later, Coronado would describe the land over which his expedition marched from northern Sonora to Zuni as La tierra despoblado—a land without people).
At about the same time that the CCD database was being compiled, a major research project was launched in the central Mesa Verde region with the goal of modeling demographic changes leading up to the depopulation of the northern San Juan in the thirteenth century. A collaboration among Washington State University, Crow Canyon Research Center, and more recently the University of Notre Dame, the Village Ecodynamics Project (VEP) documented a punctuated occupation of slow but continuous population growth through the 800s CE, interrupted by a population shift and abandonment of some regions in the 900s, reoccupation and additional growth in the 1000s, collapse and reorganization in the 1100s, and concluding with a turbulent period in the 1200s that culminated in the final depopulation of Mesa Verde between 1260 and 1280 CE. In a second phase of the VEP, archaeologists compared the decline of population on Mesa Verde with the subsequent growth of population in the lower Chama Valley of the Northern Rio Grande (e.g., Kohler et al. 2007).
Partly as a result of these regional studies, the case that Eric Reed made for historical links between the central Mesa Verde and Rio Grande Tewa has now been revived (e.g., Cordell et al. 2007; Duff and Wilshusen 2000), and the revival has recently coalesced around Scott Ortman’s seminal volume (2012). Ortman acknowledges that the archaeological evidence for migrations of Mesa Verde people to the Rio Grande is ambiguous at best, so he looks beyond archaeology to other lines of evidence, including historical linguistics, human cranial morphology, and oral history. Ortman lays out his multidisciplinary arguments in over 400 pages that I will attempt to summarize in two (my apologies in advance to Scott).
The Ortman Narrative
Attempting to reconstruct the divergence and resulting distribution of the dominant Eastern Pueblo language family, Kiowa-Tanoan, Ortman eschews glottochronology in favor of a dating method called “words for things,” based on the following logic: If we know from archaeology that a certain technology was acquired by a proto speech community around a certain time, and the languages that diverged from the proto speech community have cognate words for that technology, then it is reasonable to infer that the technology was acquired sometime before linguistic divergence occurred. If, on the other hand, words for the technology are not cognate, then a case can be made that language divergence occurred sometime after the date of technology acquisition. Using a suite of reasonably well dated technological introductions and innovations (e.g., maize agriculture, stone axes, pottery, pit houses, gourd rattles, vigas, turquoise, cotton, etc.), Ortman argued that Kiowa-Tanoan existed as a proto-language during the early centuries CE; that Kiowa split off from proto-KT first, sometime before 1,550 years ago (450 CE); that Towa diverged from Tiwa-Tewa sometime before 1,275 years ago (725 CE); that Tiwa and Tewa diverged sometime before 1,020 years ago (980 CE), and that Northern and Southern Tiwa separated sometime before 800 years ago, or around 1200 CE (Ortman 2012:169). Significantly, Ortman’s words and things approach yields much more recent divergence dates for the major languages of Kiowa-Tanoan than standard glottochronological estimates (e.g. Davis 1959).
Ortman goes on to argue that the ancestral home of Kiowa-Tanoan was likely the northern San Juan region, where he believes most of the historical divergences took place. This is at odds with what historical linguists call the “maximum diversity principle,” which states that the ancestral home of a language family is often the place where the greatest diversity of languages and dialects of the family are spoken (Sapir 1916). For Kiowa-Tanoan, that place would be the Rio Grande Valley and its principal tributaries where Northern and Southern Tiwa, Northern and Southern Tewa, Towa, Piro, and Tompiro languages were historically concentrated. Ortman acknowledges this but presents several lines of linguistic evidence, including Kiowa-Tanoan words for specific place names, plant and animal distributions, and important Tewa cultural metaphors, to argue that the Tiwa-Tewa and other more ancient K-T splits occurred outside the Rio Grande Valley (Ortman agrees with most other scholars that the northern-southern Tiwa split likely occurred within the Rio Grande Valley).
Of course, people can learn new languages and acquire new technologies more readily than they can alter the genetic expressions of their phenotypes, at least in the short term, so Ortman uses cranial morphology as a proxy for genetic variation to test his theory of Mesa Verde origins. Ortman compiled a 1,200 individual data set from more than 120 archaeological sites from the greater San Juan and Rio Grande regions and used 12 standard cranio-facial dimensions to measure bio-distance in regional populations. A principal coordinates analysis of the data set of bio-distance measures produced a scatter plot of regional similarity values, with morphological data forming a cluster that links Mesa Verde, SE Utah, and McElmo with data sets from the Chama, Pajarito, Tano, Pecos, and the Salinas areas (see Ortman 2012:Figure 5.2). These data are not without their problems. Rather than collect the basic measurements himself, Ortman used metric data compiled by multiple analysts who measured different sets of attributes. Consequently, analyst variability must be considered. In addition, all individuals with at least four of the twelve measurements were used in the analysis, so a significant proportion of Ortman’s measurements were estimated using maximum likelihood methods. Nevertheless, the bio-distance methods Ortman used are standard in the field and provide at least tentative support for his hypothesis of Mesa Verde-Northern Rio Grande similarity.
A third line of evidence Ortman presents in support of his Mesa Verde origins hypothesis are Tewa oral history and metaphors about ideas and places that link the Rio Grande Tewa to the northern San Juan region. Investigators from Harrington to Ortiz have documented Tewa oral traditions that mention an ancestral Tewa homeland called Tewayó, located north and west of the Rio Grande Valley, presumably in southwestern Colorado (Ortman 2012:188-189). Ortman documents several Tewa place names for topographic features in the northern San Juan as well as the site of Yucca House in southwestern Colorado, which he says “…provides concrete evidence that the Tewa language was spoken in the Mesa Verde region during the final decades of ancestral Pueblo occupation in that region” (2012:199). Other traditions mention the place of Tewa emergence beneath a brackish lake translated roughly as “Sandy Place Lake” (Ortman 2012:196). Harrington (1916:567) believed that the most likely location of Sandy Place Lake was the San Luis Valley of southern Colorado, but Ortman counters that the place name was probably learned through contact with the northern Tiwa. Ortman acknowledges that oral history alone cannot prove that Mesa Verdeans were Tewa speakers, so he complements his oral history data with reconstructions of Tewa cultural metaphors that he argues are characteristic of Mesa Verde but not Tewa Basin culture (e.g., pottery vessels as textiles; buildings, communities, and the world as containers). All of the words describing these cultural metaphors consist of Tewa morphemes, which challenges the hypothesis that Tewa diversified from Tiwa within the Rio Grande Valley (Ortman 2012:247).
Needless to say, despite the various positive linguistic, morphological, oral history, and metaphorical connections between Mesa Verde and the Tewa Basin, Ortman must still explain why so much of Mesa Verde architecture and material culture was left behind by migrants bound for the Rio Grande. He concludes that such dramatic culture change within a single generation was likely motivated by a religious movement:
…the pace at which the final depopulation off the Mesa Verde region took place, the rate at which smeared-indented-corrugated pottery replaced a variety of earlier forms, the synchronized way in which plaza pueblos replaced earlier community patterns in the Tewa Basin, the complete absence of 1200s Mesa Verde architectural forms, and the almost total disappearance of many 1200s Mesa Verde vessel forms all suggest that these changes did not take place as a result of everyday processes of social reproduction among indigenous people, charter groups, and recent arrivals. Instead, it appears that the migration itself was connected to discourses that had the power to convince most of the migrants, and to coerce the rest, to leave their homeland, move to the Tewa Basin, and replace many current material practices with new practices modeled after indigenous Rio Grande practices and earlier Mesa Verde practices. The only discourses that could have wielded such power over so many people are religious discourses (Ortman 2012:350).
Ortman proposes a revitalization movement in the late thirteenth century in the central Mesa Verde region that encouraged a massive migration of people and a return to earlier cultural practices that prevailed before the turmoil of the 1200s. Citing the Pueblo Revolt and the cultural atavism strictly enforced by Popé and colleagues, Ortman suggests that similar processes may have played out during the abandonment of the Mesa Verde region five hundred years earlier. He even identifies the most likely thirteenth century incarnation of Popé: The Tewa culture hero P’oseyemu.
Ortman’s arguments seem to have persuaded many historical scholars, but then, few archaeologists have sufficient expertise to evaluate arguments from historical linguistics, cranial morphology, oral traditions, and cultural metaphor analysis, so caution is warranted. In fact, a number of challenges to Ortman’s model have come out of the Rio Grande archaeological community, and as one might expect, most of these challenges focus on the absence of unambiguous archaeological evidence for a large scale migration (Boyer et al. 2010).
If thousands of pueblo people left Mesa Verde for the northern Rio Grande in a massive and highly coordinated migration event, why don’t we see material evidence of migrant communities similar to those documented for Kayenta migrants to southern Arizona? One argument in Ortman’s favor is that Kayenta migrants were moving into a region dominated by a very different cultural tradition, so Kayenta migrants were bound to “stand out” from their Mogollon and Hohokam neighbors. In contrast, Mesa Verde migrants were moving into another puebloan region where social distance was minimal and assimilation processes may have helped mask material culture differences. As some Rio Grande archaeologists point out, however, Keresan migrations to the Rio Grande in the thirteenth and fourteenth centuries are hardly invisible and can be traced through multiple lines of material culture evidence. Moreover, there are many candidate cases of thirteenth century Mesa Verde site unit intrusions west of the Rio Grande around the margins of the San Juan Basin, along the Chuska front in the west, the Rio Puerco in the east, the Rio San Jose in the south (Roney 1996), and further south at Gallinas Springs north of Magdalena (Davis 1964) and the Pinnacle Ruin north of Truth or Consequences (Lekson, et al. 2002). Nearly all of these receiving areas were part of a generalized puebloan culture area, so why did Mesa Verde migrants stand out there and not in the Rio Grande? Admittedly, a great deal more work needs to be done at these candidate sites before they can qualify as true site unit intrusions, since most are known only from pottery and other surface indications. Nevertheless, the presence of Mesa Verde ceramics at multiple sites in the San Juan Basin and farther south, combined with the virtual absence of Mesa Verde ceramics in the northern Rio Grande Valley, should be enough to give us pause.
Ortman’s linguistic arguments have also been challenged. There is, of course, the general criticism of all historical linguistic reconstructions based on their contradictory conclusions. The origin and spread of proto Uto-Aztecan is a case in point. Prominent linguists have used essentially the same linguistic evidence to argue for a southern, northern, and intermediate origin and spread of proto U-A (e.g., Hill 2001; Fowler 1983; Shaul 2014), so it would appear that, at least in the U-A case, historical linguistics is not yet able to yield evidence susceptible to refutation. Much more troubling are the archaeological implications of Ortman’s northern San Juan Kiowa-Tanoan origin hypothesis. If, as Ortman argues, the Tiwa-Tewa split occurred in the tenth century in the northern San Juan, then his model must account not just for a thirteenth century migration of Tewa speakers to the Rio Grande but also a tenth century migration of proto Tiwa speakers to the Rio Grande as well, and there is virtually no physical evidence for either migration event. Moreover, a tenth century migration from the San Juan to the Rio Grande should stand out like a sore thumb. By the turn of the tenth century the ancestral pueblos of the San Juan had developed a highly distinctive cultural package that included unit pueblos clustered into segmented villages, small as well as great kivas, and distinctive ceramic and other material cultural traditions that would have stood out from the residentially mobile, Basketmaker-like expressions of the contemporary Rio Grande Developmental period.
Ortman explains the lack of San Juan architecture and portable material culture in his Tewa origins model by proposing that a revitalization movement in the central Mesa Verde encouraged migrants to jettison most of their traditional material culture in favor of a new suite of atavistic practices in their new Rio Grande homeland (Ortman 2012:357-361). However, Ortman is virtually silent on the lack of empirical evidence for an earlier proto Tiwa migration to the Rio Grande. We must ask: How likely is it that lightening struck twice to produce an earlier case of revitalization-inspired atavism?
Another possible weak link in Ortman’s chain of inference involves Tewa oral history. Tewa informants make a compelling case for historical connections to the north, which Ortman interprets as Mesa Verde. Moreover, Tewa ritual encodes northern origins in its ceremonial circuit that recounts the emergence narrative. Ortman tells us, for example, that in both kiva ceremonies and plaza dances, the ceremonial procession always begins in the north and moves counter clockwise to the other directions (Ortman 2010:255). In fact, of course, all pueblos, with the possible exception of the Tiwa, either trace their origins to the north or view the north as the most important ritual direction (Parsons 1939:367-368). Keresan speaking pueblos cite historical linkages to Chaco and Mesa Verde and follow a counter clockwise ritual procession beginning in the north. The most important ritual direction at Zuni is to the north, the head rain priest’s kiva is associated with the north, and Zuni origin stories highlight migrations from the north, specifically Grand Canyon and the Pajarito Plateau. Several ritually important Hopi clans (Snake, Antelope, Blue and Gray Flute) and their associated ceremonies are said to have come from the north, from the vicinity of Navajo Mountain and the confluence of the San Juan and Colorado Rivers, and one of the most important ancestral villages of the Hopi is Kawestima, whose consensus location is Tsegi Canyon, far to the north of the Hopi Mesas. In short, oral traditions that Ortman claims link the Rio Grande Tewa to Mesa Verde appear to be virtually pan-pueblo in scope. Although Ortman cites evidence for Tewa knowledge of specific sites and place names in the central Mesa Verde region, such as Ute Mountain and Yucca House, he does not consider that some of these place names might have been created during the historic period when Tewa auxiliary troops—who comprised the vast majority of Spanish auxiliaries—accompanied numerous Spanish expeditions to the San Juan in search of Navajo and Ute raiders and slaves.
Alternative Migration Narratives
If Ortman can be challenged on multiple evidentiary grounds, what are the alternatives? The most obvious alternative narrative is that speakers of the closely related Tewa and Tiwa languages conform to the maximum diversity principle and are long-time residents of the Rio Grande, along with related Piro and Tompiro speech communities, whereas the dominant language of the core San Juan region, including Chaco and Mesa Verde, was Keresan, not Tewa (Ellis 1967). This scenario solves some problems with Ortman’s model but it creates some new problems of its own. Why are there ancient loan words linking proto Kiowa-Tanoan with proto Uto-Aztecan, which suggests ancient contact between K-T and U-A speakers on the plateau (Hill 2008)? If Keres was the dominant language in both Mesa Verde and Chaco, as well as the southern San Juan Basin, why does modern Keres have such a shallow time depth of 500 years (Miller and Davis 1963)? Why did the ancestral Tewa and Tiwa diverge on the Rio Grande and how did Northern Tiwa become separated from Southern Tiwa? Eric Blinman and his colleagues at the Office of Archaeological Research in Santa Fe have addressed these questions and more in developing an alternative to Ortman’s model. Following is a brief summary of the alternative culture history narrative advanced by Blinman and colleagues (henceforth, Blinman). Many of these arguments remain unpublished but all have been presented in public lectures and debates and some of the most important points were summarized in Boyer et al. (2010).
Blinman argues that proto Kiowa-Tanoan has ancient roots on the eastern Colorado Plateau stretching back to at least the late Archaic. R.G. Matson’s (1991) Eastern or Los Pinos Basketmaker II populations, best known from the Upper Animas Valley in southwestern Colorado (Morris and Burgh 1954), were likely proto K-T speakers who were in contact with Uto-Aztecan speaking western Basketmaker II populations in the early centuries of the Common Era (Hill 2008). This would explain the ancient U-A loan words in K-T, most having to do with corn horticulture, and the reverse borrowing of K-T words into U-A for plateau fauna and flora, such as “pinyon” and “antelope.” Blinman suggests that proto K-T speakers were likely responsible for some of the earliest brown ware ceramics (e.g., Sambrito Brown) that begin to show up in sites at the cusp of Basketmaker II–III, along with shallow round pit structures with east-oriented ramp entryways. Early brown ware sites have been documented over a wide area from Gallup in the south to the Dolores Valley in the north, but they were supplanted around 600 CE by classic Basketmaker III cultural expressions, complete with Lino Gray ceramics, the bow and arrow complex, full-groove stone axes, and pit structures with south-oriented entryways, sometimes via an attached antechamber. Blinman argues that Basketmaker III represents an intrusion of proto-Keres speaking peoples from the south who displace proto K-T speakers to the north and east. According to the argument, northern K-T speakers would move under pressure from the advancing Keresan front into the Fremont region of eastern Utah and eventually out onto the western plains where they would become the ancestors of the historic Kiowa (on this last detail, Blinman and Ortman appear to agree). Proto Tiwa-Tewa speakers would be displaced east to the Upper San Juan and Rio Grande where they would be the principal agents in Wendorf and Reed’s Rio Grande Developmental to Coalition phase sequence. Proto Towa speakers would remain in the upper San Juan and eastern San Juan Basin where they would be the principal agents of the Rosa-Gallina cultural tradition. In the thirteenth century, Towa speakers from the north and from the Rio Puerco to the west would expand into the southern Jemez where they would become the Jemez pueblos and interact intensively with Keresan speaking people who were moving into the Rio Grande along the lower Jemez River in the late pre-Hispanic period.
This scenario accounts for ancient U-A and K-T loan words and it acknowledges the earliest known ceramic sites on the plateau which do not figure in Ortman’s account. More important, however, Blinman’s model is consistent with the lack of an obvious cultural boundary between the northern and central San Juan for most of the 7th through 13th Century. Archaeological evidence suggests that the Core San Juan region was home to a relatively homogenous population that moved back and forth in response to regional environmental swings (e.g., Euler et al., 1979; Wilshusen and Van Dyke 2006). During times of above average rainfall and cooler temperatures, the higher elevation regions of the northern San Juan emptied out (perhaps due to growing season constraints) and the lower elevation regions of the central basin filled up. This was rather clearly the case in the late 800s through the early 1100s CE. During warmer drier intervals, from the early 700s to late 800s and again from the early to mid 1100s to the late 1200s, the central basin lost population to the more mesic uplands of the central Mesa Verde region. Differences in ceramics, architecture, technology, and other material culture expressions between the central and northern San Juan are mostly a function of differences in raw material sources. That is, the same people appear to have been doing essentially the same thing with different local materials. According to Blinman, they were also speaking the same proto-Keresan language.
If Blinman is right that the core San Juan post 600 CE was populated mostly by proto-Keresans rather than Kiowa-Tanoans, then the shallow time depth of modern Keresan is another hurdle to clear. As Miller and Davis (1963) pointed out, only about 500 years separate the western-most Keresan village of Acoma from the eastern-most village of Cochiti (there are dialectical differences between the two ends of the Keresan bridge but the dialects remain mutually intelligible). If Blinman is right that Keresan speakers spread throughout the plateau in the seventh century (he even includes the Kayenta region in his narrative), why don’t we see the kind of linguistic diversity in Keresan that we see among Uto-Aztecan, Yuman, and Kiowa-Tanoan, whose speech communities presumably spread and eventually diversified under the influence of agriculture? In fact, the shallow time depth of modern Keresan led linguist Jane Hill to propose that Keresan speakers may have been a group of relic hunter-gatherers in the southern San Juan Basin who were late comers to an agricultural lifeway (Hill 2007). Other evidence, however, suggests that Keresan speakers may have been at the very center of the most precocious pre-colonial ritual-political systems in the northern Southwest. Significantly, much of the ritual language of the Pueblos is Keresan (Shaul 2014). At Hopi, the songs of the Snake-Antelope and Blue-Gray Flute societies are in Keresan and three out of four Wuwtsim societies are apparently dominated by Keresan words as well (Parsons 1939:977). At Zuni, the majority of its twelve medicine societies are Keresan in origin (Ladd 1979). Jemez appears to have adopted a Keresan ritual system whole cloth (Robin Fox once described Jemez as a Keresan village in everything but language ), and on the Rio Grande, Keresan medicine and clown societies are nearly ubiquitous among neighboring Tanoan speaking pueblos. How could a group of hunter-gatherers and late comers to cultivation have had such a profound influence on pan-pueblo ritual systems?
It is far more likely, I think, that Keresan was the dominant language of the core San Juan region where complex ritual expressions reached their apogee in Chaco Canyon in the eleventh century and persisted in the valley of the San Juan River well into the thirteenth century. The shallow time depth of Keresan may be the result, instead, of a demographic bottleneck that stripped Keresan of some of its linguistic variability, and participation in the Chaco ritual sphere may have also dampened linguistic diversity. The bottleneck hypothesis finds some empirical support in the late thirteenth century depopulation of the central Mesa Verde region. Based on evidence from Sand Canyon Pueblo and several other Pueblo III sites in the Mesa Verde region, the depopulation of the northern San Juan was accompanied by a high level of structural violence, including large scale massacres and perhaps isolated cases of cannibalism (LeBlanc 1999; Kuckleman 2010). Scholars working on the Village Ecodynamics Project have estimated that thousands of people could have survived the structural violence of the thirteenth century to migrate out of the northern San Juan, but momentary population estimates are always difficult to substantiate in archaeology, so we may never know how many people made it out of the northern San Juan in the thirteenth century. If endemic warfare persisted for two or three generations, steep fertility declines would likely have had trouble keeping up with heightened mortality rates, and comparative studies show that large populations can crash very quickly. From scattered Mesa Verde occupations around the edges of the San Juan Basin and points farther south we can be confident that some migration occurred, but the scale of the migration and the number of Mesa Verde migrants may always be in doubt.
Perhaps the biggest challenge for Blinman’s model is the Tiwa-Tewa divergence on the Rio Grande and the Northern Tiwa-Southern Tiwa split. Blinman addresses these questions as follows: Proto Tiwa-Tewa farmers in the Rio Grande would expand up White Rock Canyon, the Santa Fe River Valley, and into the Tewa Basin around 900 CE, displacing “Latest Archaic” hunter-gatherers as they went (Vierra and Ford 2006). This movement, according to Blinman, was the beginning of the divergence of Tiwa and Tewa on the Rio Grande, and it coincides roughly with the date Ortman assigns to the split using his “words for things” methodology. Three hundred years later Keresan speakers from the San Juan would move up the Jemez River Valley and into the Santo Domingo Basin, western Galisteo Basin, and onto the Pajarito Plateau, displacing local Tiwa speakers who leap-frogged over the Tewa Basin and moved north into the Taos Valley where they would contest for dominance with Tewa speakers on the northern edge of the Tewa frontier.
Archaeological evidence for the tenth century expansion up the Santa Fe River Valley and into the Tewa Basin is uncontroversial, and the later expansion of Keres speaking people into the Santo Domingo Basin, southern Pajarito Plateau, and western Galisteo Basin is equally well established. The diversification of Tewa and bifurcation of northern and southern Tiwa populations is probably the most controversial aspects of Blinman and colleagues proposal. Would the La Bajada escarpment and Caja del Rio uplands pose a sufficient barrier for Tiwa-Tewa language divergence or did the absorption of local forager populations by proto-Tewa speaking farmers in the Tewa Basin contribute to the split, and how would we know? Moreover, there is linguistic diversity among the Rio Grande Tewa that is not explained by either Ortman or Blinman’s model, and oral traditions that link the northern Tewa speech community (Ohkay Owingeh, Santa Clara, and San Ildefonso) with the Upper San Juan and Piedra regions of southwestern Colorado complicate both migration scenarios (R. Ford, personal communication, December 2015). Ortman rejects a Rio Grande diversification because the Tiwa and Tewa share only a fraction of cognates for Tewa Basin place names, but in Blinman and colleague’s alternative model we would not expect there to be much cognate overlap in the Tewa Basin. In Blinman’s model, the greatest degree of cognate overlap should be south of La Bajada and White Rock Canyon in the Santo Domingo Basin, before the tenth century divergence of proto Tiwa and Tewa, and perhaps in the Taos area where Tewa and Tiwa people eventually merged on a highly contested landscape. Needless to say, there is much more work to do on these important questions. However, it seems to me that the task of explaining a Rio Grande divergence of Tiwa-Tewa and the subsequent Tiwa split poses less of a challenge than accounting for two late prehistoric proto Tanoan migrations from the San Juan to the northern Rio Grande that left few if any archaeological traces.
A final hurdle for Blinman to clear is the growth of population in the northern Rio Grande without major influxes of migrants from the periphery. Growth rates for the Tewa Basin are hard to estimate given the almost total lack of large-scale aerial survey data, with the notable exception of the Pajarito Plateau where cultural inventories have been completed for virtually all of Bandelier National Monument and for most of the Los Alamos National Laboratory lands. We know, however, that there were no substantial Developmental period (600-1175 CE) populations on the Pajarito or in the lower Chama and Ojo Caliente Valleys, or the Galisteo Basin, and yet all of these areas literally filled up with farming populations in the 1200s and early 1300s during the Rio Grande Coalition (1175-1325 CE). Blinman has suggested that most of this growth and dispersal of farmers was internally generated and a reaction to regional improvements in moisture and dry farming potential (Boyer et al., 2010). In my judgment, however, sufficient internal growth to fill up nearly every adjacent region of the northern Rio Grande is hard to justify without some input from migration, and the results of the Coalescent Communities Database study rather clearly indicate a late pre-colonial concentration of population in the northern Rio Grande that precisely correlates with depopulation of most outlying regions. Fortunately, although Blinman and colleagues reject Ortman’s massive coordinated migration of Mesa Verdeans to the northern Rio Grande in the thirteenth century, they do not rule out the possibility of significant numbers of people from the San Juan and other peripheral regions making the journey in small family or descent-based groups, which coincides with McNutt’s (1969) and other’s conclusions about the organizational scale of Rio Grande migrations (see Dick Ford’s comments below). The conundrum remains, however. Why is there so little unambiguous archaeological evidence? Where is the smoking gun?
A Compromise Explanation
Perhaps part of the answer lies in the pueblos’ historic ability to assimilate migrants of all stripes and complexions? Small groups on the Colorado plateau faced constant threats to their survival from droughts, crop failures, conflicts, and other challenges, so there was always the need to form regional relationships in order to solve problems requiring extra-local action. In non-state societies, regional alliances are most commonly formed through intermarriage, but most modern pueblos are matrilocal-matrilineal and, therefore, endogamous, which would have likely ruled out inter-pueblo marriage alliances (Levy 1994). The matrilineal Iroquois of the northeast and the matrilineal Creeks of the southeast used political confederacies to build regional alliances. The matrilineal Pueblos of the southwest used pan-tribal ritual sodalities for the same purpose (Ortiz 1994; Ware 2014). Ritual affiliations would have been a convenient and highly flexible way to incorporate migrants into existing communities. If immediate kin ties could not be established, perhaps moiety affiliation, or membership in a ritual society, or knowledge of a certain ceremony would have expedited the assimilation process. The benefits of assimilation are, of course, bilateral. The host community gains an expanded mating pool and a larger labor and defense force, and presumably, assimilation would have occurred on the host community’s terms: You can move in with us but you have to adhere to our authority structures, follow our ritual cycle—and share any rituals you might have that complement our ceremonial suite—and eventually learn our language and become full-fledged community members (Kroskrity 1998). There is, of course, ample archaeological and ethnographic evidence of Pueblo migrant assimilation along precisely such lines.
The multiethnic character of late pre-colonial communities above and below the Mogollon Rim in eastern Arizona has long been appreciated, and pueblos that survived into the historic period exhibit similar ethnic diversity. Hopi communities are an assortment of ethnicities who came together in the late pre-contact period to form the historic Hopi villages, including, apparently, large numbers of Kayenta reverse migrants from southern Arizona and a large contingent of Keresan speakers on Antelope Mesa. The Hopis also accepted Tewa, Jemez, Sandia, and other Eastern Pueblo immigrants following the Spanish Reconquest. Both ethnohistorical (Cushing 1896) and archaeological data (Smith et al. 1966) support the hypothesis of Zuni multi-ethnicity. As mentioned above, a significant number of human remains from Hawikku Pueblo consisted of cremations in association with Roosevelt Red Ware vessels, and Zuni origin stories identify the Grand Canyon as well as the Pajarito Plateau as sources of founding populations, suggesting that ancestral Hopis and Keresans were once assimilated into Zuni. Nearly half of M. C. Stephenson’s informants from Taos in the early 1900s traced their origins to the Jicarilla Apache. In addition, many Utes were also assimilated into Taos, and apparently there was a large complement of Santo Domingo Keresans who had assimilated into Taos as well (Fowles in progress). A substantial number of Tanos from Galisteo Pueblo were assimilated into Santo Domingo in 1793 following a smallpox epidemic in the Galisteo Basin (Simmons 1979:187). Isleta assimilated Laguna migrants in the late nineteenth century, in part to obtain katsina ceremonialism that had lapsed at Isleta (Parsons 1928). Jemez Pueblo assimilated Navajo families during the eighteenth century, as well as migrants from Pecos in the early nineteenth century (who may or may not have been native Towa speakers—see discussion below). And this is hardly an exhaustive list of an understudied phenomenon—graduate students take note.
We should not be surprised, therefore, if Keresan-speaking migrants (and perhaps other Puebloan groups) escaping climate disruptions and extreme violence in the thirteenth-century core San Juan region and elsewhere were able to integrate into emergent Tewa communities in the northern Rio Grande, where they would have assimilated into Tewa culture, learned the language, and contributed their genes along with their own rituals and cultural metaphors to the syncretic amalgamation of Tewa culture. This compromise solution allows us to accept the bulk of Ortman’s interdisciplinary analyses without the excess baggage of an implausible Tiwa-Tewa homeland in the greater northern San Juan and a mythical culture hero given agency in a massive coordinated migration for which there is little or no physical evidence. Ortman establishes a plausible link between the Rio Grande Tewa and Mesa Verde through the timing of linguistic divergence, skeletal morphology, oral history, and cultural metaphor analysis, but in the end he is still confronted with the lack of Mesa Verde architecture and portable material culture in the northern Rio Grande, which has challenged migration explanations for over 100 years. Ortman solves this problem by conjuring up a revitalization movement that encouraged atavistic practices that obscured the identity of Mesa Verde migrants in the Rio Grande, but not in the San Juan Basin and other destinations to the south. I point out these discrepancies with some reluctance because I think that the kind of multidisciplinary analyses that Ortman conducts are precisely the kind of studies that promise to lead us out of the thicket of a century-old archaeological debate. We need more, not fewer ambitious multidisciplinary studies of this scale. Unfortunately, in his conclusions Ortman seems more interested in advancing a Tewa version of Northern Rio Grande precolonial history than in doing critical historical anthropology, which would require incorporating other Rio Grande traditions in his study. Fortunately, there are explanations of Rio Grande migrations that aren’t weighed down by the confirmation biases that suffuse Ortman’s otherwise excellent work.
Do any of the migration scenarios sketched above help to explain the growth and consolidation of population at Arroyo Hondo Pueblo during the fourteenth and early fifteenth centuries? This was the subject of an informal seminar at the School of Advanced Research in December, 2015. The seminar was organized and chaired by Doug Schwartz and participants included fellow Rio Grande archaeologists Eric Blinman, Steve Post, Curtis Schaafsma, Polly Schaafsma, Jay Shapiro, and anthropologist and Tewa specialist Richard Ford. The discussions ranged from Arroyo Hondo in particular to well beyond the range of the present paper in general. Following is a summary of the main discussion points, beginning with the particular.
There was general agreement that throughout both of its late pre-colonial occupations, the architecture and portable material culture of Arroyo Hondo is consistent with local origins for most if not all Arroyo Hondo occupants. The adobe construction styles at the site are identical to those used at contemporary sites on the nearby Santa Fe River and Cañada de los Alamos, and pit structure styles and floor feature types and orientations exhibit continuity with the late Developmental through the Rio Grande Classic pit structure styles (Shapiro). The overwhelming majority of Rio Grande pithouse-kivas, including all but one excavated at Arroyo Hondo, have east-oriented entryways in contrast to the south orientation typical of San Juan kivas, so Arroyo Hondo’s ritual architecture conforms to Rio Grande architectural styles (C. Schaafsma). Moreover, participants saw strong continuity between the first and second occupations at Arroyo Hondo, suggesting that the village was probably reoccupied in the second half of the fourteenth century by the descendants of the people who left the site in the early 1300s (Schwartz).
Most participants saw sufficient population in small sites within a 10-20 mile radius of Arroyo Hondo to provide most if not all of the population that aggregated at Arroyo Hondo in the early 1300s (Schwartz). A similar aggregation of mostly local population happened in the Galisteo Basin a hundred years later when dozens of small hamlets and villages coalesced into the Classic Period pueblos of Galisteo, San Cristobal, San Lazaro, and San Marcos (Blinman). In all likelihood, one of the first Arroyo Hondo founding groups came from Upper Arroyo Hondo, located two miles upstream and with occupation dates that preceded and partly overlapped with the initial occupation of Arroyo Hondo. The early 1300s was a period of dynamic population movement and growth throughout the region and this may have motivated the residents of Upper Arroyo Hondo to establish a residential foothold and thus a claim on the Arroyo Hondo spring (Post). Participants discussed whether the spring at Upper Arroyo Hondo was perhaps less reliable than the Arroyo Hondo Spring, but Doug Schwartz noted that both springs were fed by a sixty square mile watershed that would likely have provided water in even the driest of years (as it continues to do today).
Regarding the larger picture of migrations in, to, and through the northern Rio Grande, there was general agreement that despite rapid population growth and settlement expansion in the fourteenth and fifteenth centuries, there is no obvious archaeological evidence of a large-scale coordinated migration from the San Juan or anywhere else. Archaeologists who have argued that Galisteo B/W was a Rio Grande cognate of Mesa Verde White Ware frequently cite an American Antiquity article by Bertha Dutton (1964). In fact, however, Dutton herself denied the connection after a careful comparison of the two ceramic types (Blinman). Curtis Schaafsma, who excavated at the site of Las Madres in the Galisteo Basin in the 1960s under Dutton’s general direction, recalled that Bertha saw the strongest connections between Galisteo B/W and the ceramics from Walnut Canyon, Arizona! Unfortunately, the mythology of Galisteo B/W and Mesa Verde B/W continues to this day (e.g., Cordell, et al. 2007).
No one questions that Tewa oral traditions claim an origin in the north, but participants agreed that other pueblo language groups make similar claims. Polly Schaafsma pointed out that some pueblo origin stories now reflect detailed knowledge of the archaeological literature. She mentioned by way of example that some Hopis now talk about migrations from Páquime in northern Chihuahua, and some contemporary stories even claim Hopi visits to Teotihuacan. These facts raise serious methodological concerns about the scholarly evaluation of traditional oral histories. What influence does the present have on the construction of a particular past, and how do historical scholars filter out the potential effects of such bias, especially in the absence of documentary evidence?
According to Dick Ford, Ortman’s migration account simplifies the complexity of Tewa cultural and dialectical differences. Ford pointed out that there are three main dialects of Tewa: 1) a southern dialect in the Galisteo Basin that would likely have included Arroyo Hondo and Cañada de los Alamos; 2) a middle dialect spoken at Tesuque and Nambe, one or two other pueblos along Tesuque Creek that are no longer occupied, such as Cuyamungue, and perhaps the Santa Fe River Valley; and 3) a northern dialect spoken today at Ohkay Owingeh, San Ildefonso, and Santa Clara (Santa Clara may include a large complement from the southern Tewa speech community). In consultation with Tesuque members during excavations at the Santa Fe Convention Center, which Tesuque claims is an ancestral community, there was no mention of Mesa Verde origins, and the southern Tewa whose descendants ended up at the Western Tewa village of Hano trace their origins no earlier than the Santa Cruz Valley north of Santa Fe where southern Tewas relocated from the Galisteo Basin sometime after the Pueblo Revolt of 1680 (Blinman). According to Ford, Ortman’s book paints with perhaps too broad a brush.
Ford favors a trickle as opposed to a wave of Tewa migrants, initially from the Piedra region of the Upper San Juan and the Western San Luis Valley down into the Chama and eventually south into the Tewa Basin, Santa Fe River Valley, and the Galisteo Basin, where linguistic diversification occurred following population aggregation. Ford mentioned a Tewa shrine at Chimney Rock and at Arboles, at the confluence of the Piedra and San Juan Rivers, that were visited by Tewa elders from Ohkay Owingeh throughout the 1920s and 30s. Apparently, all of the lakes around Dulce, New Mexico, had Tewa names that were simply translated from Tewa into Jicarilla, and Tewa priests would routinely visit these sites along with multiple shrines along the lower Chama. Ford suggests that the principal unit of Tewa migration was very likely the Maatu'in, an ambilocal extended family that forms the principal residential unit among the Northern Tewa during the historic period (Ford, in progress). The movement of such small family groups would have left little more than campsite-size “footprints” on the landscape.
Citing evidence from rock art, Polly Schaafsma pointed out that the late pre-contact ritual imagery of the northern Rio Grande comes mostly from the south and east rather than the north. For example, shield figures are the most distinctive elements of ancestral Tewa rock art and these have rather clear origins on the Plains, whereas many other elements of the Rio Grande rock art style have southern origins. Whether or not southern groups came into the northern Rio Grande or whether their ideas diffused from the Mimbres and Jornada Mogollon is a question that needs to be answered, but there is no question that most Tewa ritual imagery comes from the south and east and washed over everything in the Rio Grande, regardless of linguistic affiliation. There is other material evidence of southern influence in the form of Socorro B/W ceramics and Mogollon brown wares in the southern Tewa region (Ware), as well as migrations of ideas, if not people, from the west in the form of true indented corrugated ceramics showing up in the eastern Galisteo Basin in the last decade of the 1200s (Blinman).
There was general agreement that most pueblos are multiethnic polyglot communities, consistent with the compromise explanation outlined above. The multiethnic character of the Hopi and Zuni pueblos is well known and most participants agreed that the Rio Grande Pueblos have equally diverse origins. Ford reiterated that large numbers of Navajos were assimilated into Jemez after the Spanish reconquest (and Navajos reciprocated by assimilating many Jemez refugees following the reconquest), and he mentioned the movement of Pecos Pueblo’s remnant population to Jemez in the 1830s. Ford noted several lines of evidence suggesting that Pecos was predominately a southern Tewa Pueblo rather than a Towa speaking pueblo, as some authorities have suggested (e.g. Schroeder 1979). He mentioned showing baptismal records from Pecos to Joe Sando of Jemez who didn’t recognize any of the names as Towa, whereas Herman Agoyo of Ohkay Owingeh recognized Tewa names on the list. According to Ford, Jemez and Pecos had long been active trading partners, which may explain why Pecos migrants ended up at Jemez.
As pueblo intermarriages have increased in recent years, assimilation processes have become even more apparent. One of the most accomplished potters at San Ildefonso learned her pottery craft at her birth community of Zia. After marrying into San Ildefonso, however, she abandoned the Zia style and began making pottery in the Tewa tradition (Blinman). Ford mentioned a similar case of an Ohkay Owingeh potter who married into Picuris and became an outstanding micaceous pottery maker, in the Northern Tiwa tradition. These examples suggest that pueblo cultural boundaries, though porous to migrants and now, intermarriage, are nevertheless highly durable. When you move into another pueblo, you are expected to become a member of that pueblo and forsake your previous identity (Blinman). This evidence is clearly at odds with Ortman’s hypothesis that local Rio Grande populations adopted the language, if not the architecture and material culture, of migrants from the northern San Juan.
Finally, all participants agreed that the late pre-colonial demographic history of the Rio Grande is much more complicated than any published model of migration and ethnogenesis has thus far described—certainly more complicated by far than contemporary migrations of Western Pueblo people to southern Arizona. The multidisciplinary analyses accomplished by Scott Ortman are exemplary and may eventually help to resolve these complications, but future studies must incorporate a much broader sample of ethnographic cases, oral traditions, and archaeological data sources such as rock art. In addition, future studies must acknowledge the fundamental multiethnic and polyglot character of most if not all pueblo communities and focus much needed research on the mechanics of assimilation and ethnogenesis. These, we believe, are the most important frontiers of future migration studies.
Acknowledgments: Thanks first of all to Doug Schwartz and the participants in the Arroyo Hondo Seminar for their lively discussion and many contributions to the thoughts and themes of this paper. Jeff Clark, Peter Pillis, Eric Blinman, Scott Ortman, and Polly Schaafsma offered additional comments on the manuscript, and Jeff Clark, Eric Kaldahl, Patrick Lyons, Scott Ortman, and Peter Pillis kindly granted permission to use previously published illustrations. Finally, special thanks to Scott Ortman for courageously leading the way and stirring the pot.
Boyer, Jeffrey L., James L. Moore, Steven A. Lakatos, Nancy J. Akins, C. Dean Wilson, and Eric Blinman
2010 Remodeling immigration: A northern Rio Grande perspective on depopulation, migration, and donation-side models. In Leaving Mesa Verde: peril and change in the Thirteenth-century Southwest, edited by T. A. Kohler, M. D. Varien, and A. M. Wright, pp. 286-323. Amerind Studies in Archaeology 5. University of Arizona Press, Tucson.
Clark, Jeffery J. ,
2001 Tracking prehistoric migrations: Pueblo settlers among the Tonto Basin Hohokam. Anthropological Papers of the University of Arizona No. 65. University of Arizona Press, Tucson.
Clark, Jeffery J., and Patrick D. Lyons (editors)
2012 Migrants and mounds: Classic period archaeology along the Lower San Pedro River. Anthropological Papers of the University of Arizona No. 45. Center for Desert Archaeology, Tucson.
Clark, Jeffery, J., J. Brett Hill, Patrick D. Lyons, and Stacey N. Lengyel
2012 Of Migrants and Mounds. In Migrants and Mounds: Classic Period Archaeology of the Lower San Pedro Valley, edited by J. J. Clark and P. D. Lyons, pp. 345-405. Anthropological Papers No. 45. Archaeology Southwest, Tucson.
Clark, Jeffery J., Deborah L. Huntley, J. Brett Hill, and Patrick D. Lyons
2013 The Kayenta Diaspora and Salado Meta-Identity in the Late Pre-Contact Southwest, In Hybrid Material Culture: the Archaeology of Syncretism and Ethnogenesis, edited by J. Card, pp. 399-424. Center for Archaeological Investigations Occasional Paper No. 39. Southern Illinois University, Carbondale.
Cordell, Linda S.,
1979 The Prehistory of Santa Fe Country. New Mexico Geological Society Special Publication No. 8. New Mexico Archaeological Council, Albuquerque.
Cordell, Linda S., Carla R. Van West, Jeffrey S. Dean, and Deborah A. Muenchrath
2007 Mesa Verde settlement history and relocation: Climate change, social networks, and ancestral Pueblo migration. Kiva 72:379-405.
Crown, Patricia L.
1994 Ceramics and ideology: Salado polychrome pottery. University of New Mexico Press, Albuquerque.
Cushing, Frank H.
1896 Outlines of Zuni Creation Myths. 13th Annual Report of the Bureau of American Ethnology for the Years 1891-1892, pp. 321-447. Washington.
Davis, Emma Lou
1964 Anazazi mobility and Mesa Verde migrations. Unpublished PhD dissertation, Department of Anthropology, University of California, Los Angeles.
1959 Linguistic clues to northern Rio Grande Prehistory. El Palacio 66(3):73-84.
Di Peso, Charles C.
1958 The Reeve Ruin of Southeastern Arizona. Amerind Foundation Technical Report No. 8, Dragoon, Arizona.
2002 Western Pueblo identities: Regional interaction, migration, and transformation. University of Arizona Press, Tucson.
Duff, Andrew, I. and Richard H. Wilshusen
2000 Prehistoric population dynamics in the northern San Juan region, A.D. 950-1300. Kiva 66:167-190.
Dutton, Bertha P.
1963 Las Madres in the Light of Anasazi Migrations. American Antiquity 29:449-454.
1950 Social organization of the western pueblos. University of Chicago Press, Chicago.
1972 Summary. In New perspectives on the Pueblos, edited by A. Ortiz, pp. 287-305. University of New Mexico Press, Albuquerque.
1967 Where did the Pueblo People Come From. El Palacio 74(3): 35-43.
Euler, Robert C., George J. Gumerman, Thor N. V. Karlstrom, Jeffery S. Dean, and Richard H. Hevley
1979 The Colorado Plateaus: Cultural Dynamics and Paleoenvironment. Science 205:1089-1101.
Ford, Richard I., Albert H. Schroeder, and Stewart L. Peckham
1972 Three perspectives on Pueblo prehistory. In New perspectives on the Pueblos, edited by A. Ortiz, pp. 22-40. University of New Mexico Press, Albuquerque.
Fowler, Catherine S.
1983 Some lexical clues to Uto-Aztecan Prehistory. International Journal of American Linguistics 49(3):224-257.
In Preparation Taos Social History: A Rhizomatic Account. Draft manuscript prepared for SAR Advanced Seminar, October 2015, Santa Fe.
1967 The Keresan Bridge: a problem in Pueblo ethnology. Athlone Press, London.
Harrington, John P.
1916 The Ethnogeography of the Tewa Indians. 29th Annual Report of the Bureau of American Ethnology, pp. 29-618. Washington.
1958 Evidence at Point of Pines for a prehistoric migration from northern Arizona. In “Migrations in New World culture history,” ed. R. H. Thompson. Univ. Ariz. Bull. 29(2), Soc. Sci. Bull. 27:1-8
Hill, J. Brett, David Wilcox, William Doelle, and William Robinson
2012 Coalescent Communities Database Version 2.0. Archaeology Southwest, Tucson.
Hill, Jane H.
2001 Proto-Uto-Aztecan: A community of cultivators in central Mexico? American Anthropologist 103(4):913–934.
2007 The Zuni language in southwestern areal context. In Zuni origins: Toward a new synthesis of Southwestern Archaeology, edited by D. A. Gregory and D. R. Wilcox, pp. 22-38. University of Arizona Press, Tucson.
2008 Northern Uto-Aztecan and Kiowa-Tanoan: Evidence of Contact Between the Proto-Languages? International Journal of American Linguistics 74(2):155-188.
Kidder, Alfred V.
1962 (1924) An introduction to the study of southwestern archaeology with a preliminary account of the excavations at Pecos. Papers of the Southwest Expedition 1. Reprinted in 1962 by Yale University Press, New Haven, Connecticut.
Kohler, T. A., C. D. Johnson, M. D. Varien, S. G. Ortman, R. Reynolds, Z. Kobti, J. Cowan, K. Kolm, S. Smith, and L. Yap
2007 Settlement and Ecodynamics in the Prehispanic Central Mesa Verde Region. In The Model-Based Archaeology of Socionatural Systems, ed. T. A. Kohler and S. van der Leeuw, pp. 61-104. School of Advanced Research Press, Santa Fe.
1998 Arizona Tewa Kiva Speech as a Manifestation of a Dominant Language Ideology. In Language Ideologies: Practice and Theory, ed. B. Schieffelin, K. Woolard, and P. Kroskrity, pp. 103-122. Oxford Studies in Anthropological Linguistics No. 16. Oxford University Press, Oxford.
Kuckelman, K. A.
2010 The depopulation of Sand Canyon Pueblo, a large Ancestral Pueblo village in southwestern Colorado. American Antiquity 75(3):496-526.
Ladd, Edmund J.
1979 Zuni Social and Political Organization. Handbook of North American Indians, Vol. 9: Southwest, ed. by A. Ortiz, pp. 482-491. Smithsonian Institution Press, Washington, D.C.
Lakatos, Steven A.
2007 Cultural continuity and the development of integrative architecture in the northern Rio Grande Valley of New Mexico, AD 600-1200. Kiva 73(1):33-68.
1999 Prehistoric Warfare in the American Southwest. University of Utah Press, Salt Lake City.
Stephen H. Lekson, Curtis P. Nepstad-Thornberry, Brian E. Yunker, Toni S. Laumbach, David P. Cain and Karl W. Laumbach
2002 Migrations in the Southwest: Pinnacle Ruin, Southwestern New Mexico. Kiva 68(2):73-101.
1994a Ethnographic analogs: strategies for reconstructing archaeological cultures. In Understanding complexity in the prehistoric Southwest, edited by G. Gumerman and M. Gell-Mann, pp. 233-244. Santa Fe Institute Studies in the Sciences of Complexity Proceedings Volume 16. Addison Wesley, Reading, Massachusetts.
Lindsay, Alexander J. Jr., Richard Ambler, Mary Anne Stein, and Philip M. Hobler
1968 Survey and Excavations North and East of Navajo Mountain, Utah, 1959-1962. Bulletin No. 45, Glen Canyon Series No. 8. Northern Arizona Society of Science and Art, Flagstaff, AZ.
Lipe, William D.
2010 Lost in Transit: The Central Mesa Verde Archaeological Complex. In Leaving Mesa Verde: Peril and Change in the Thirteenth-Century Southwest, ed. T. A. Kohler, M.D. Varien, and A. Wright, pp. 262-284. University of Arizona Press, Tucson.
Lyons, Patrick D.
2003 Ancestral Hopi migrations. Anthropological Papers of the University of Arizona No. 68. University of Arizona Press, Tucson.
2012 Ceramic Typology, Chronology, Production, and Circulation. In Mounds and Migrants: Late Prehistoric Archaeology of the Lower San Pedro River Valley, Arizona, edited by J. J. Clark and P. D. Lyons, pp. 211-308 Anthropological Papers No. 45. Archaeology Southwest, Tucson.
In Press Excavations at the Davis Ranch Site, San Pedro Valley, Arizona. University of Arizona Press, Tucson.
Lyons, Patrick D., and Jeffery J. Clark
2012 A Community of Practice in Diaspora: The Rise and Demise of Roosevelt Red Ware. In Potters and Communities of Practice: Glaze Paint and Polychrome Pottery in the American Southwest A.D. 1200–1700, edited by L. S. Cordell and J. A. Habicht-Mauche, pp.19-33. Anthropological Papers of the University of Arizona No. 75. University of Arizona Press, Tucson.
Lyons, Patrick D., and Alexander J. Lindsay, Jr.
2006 Perforated Plates and the Salado Phenomenon. Kiva 72(1):5–54.
Matson, R. G.
1991 The origins of Southwestern agriculture. University of Arizona Press, Tucson.
McNutt, Charles H.
1969 Early Puebloan Occupations at Tesuque Bypass and the in the Upper Rio Grande Valley. Anthropological Papers, Museum of Anthropology, University of Michigan, No. 40. Ann Arbor, Michigan.
Mera, H. P.
1935 Ceramic clues to the prehistory of North Central New Mexico. Technical Series Bulletin No. 8. Laboratory of Anthropology, Santa Fe, NM.
Miller, Wick R., and Irvine Davis
1963 Proto-Keresan Phonology. International Journal of American Linguistics 20(4):310-330.
Morris, Earl H., and Robert F. Burgh
1954 Basket Maker sites near Durango, Colorado. Carnegie Institution of Washington Publication 604.
1949 Social structure. The Macmillan Company, New York.
1994 The Dynamics of Pueblo Cultural Survival. In North American Indian Anthropology: Essays on Society and Culture, edited by R. J. DeMallie and A. Ortiz, pp. 296-306. University of Oklahoma Press, Norman.
Ortman, Scott G.
2010 Evidence of a Mesa Verde homeland for the Tewa Pueblos. In Leaving Mesa Verde: Peril and change in the thirteenth-century Southwest, edited by T. A. Kohler, M. D. Varien, and A. M. Wright, pp. 222-261. Amerind Studies in Archaeology 5. University of Arizona Press, Tucson.
2012 Winds from the North: Tewa Origins and Historical Anthropology. University of Utah Press, Salt Lake City
Parsons, Elsie Clews
1928 The Laguna Migration to Isleta. American Anthropologist 26(3):333-339.
1939 Pueblo Indian religion, 2 volumes. University of Chicago Press, Chicago, Illinois.
Reed, Eric K.
1949 Sources of Rio Grande Culture and Population. El Palacio 56:163-184.
Roney, John R.
1996 The Pueblo III period in the Eastern San Juan and Acoma-Laguna areas. In The Prehistoric Pueblo World, A.D. 1150-1350, edited by M. Adler, pp. 145-169. University of Arizona Press, Tucson.
1916 Time Perspective in Aboriginal American Culture: A Study in Method. Anthropological Series 13, Memoirs of the Canadian Geological Survey 90. Ottawa.
Schroeder, Albert, H.
1979 Pecos Pueblo. In Handbook of North American Indians, Volume 9. Edited by A. Ortiz, pp. 430-437. Smithsonian Institution, Washington.
Shaul, David L.
2014 A Prehistory of Western North America: The Impact of Uto-Aztecan Languages. University of New Mexico Press, Albuquerque.
1979 History of Pueblo-Spanish Relations to 1821. Handbook of North American Indians, Vol 9: Southwest, ed. Alfonso Ortiz, pp. 178-193.
Smith, Watson, Richard B. Woodbury, and Nathalie F. S. Woodbury
1966 The Excavation of Hawikuh by Frederick Webb Hodge, Report of the Hendricks-Hodge Expedition, 1917–1923. Contributions from the Museum of the American Indian, Heye Foundation, 20. Museum of the American Indian, New York.
Trager, G. L.
1967 The Tanoan settlement of the Rio Grande area: A possible chronology. In Studies in Southwestern Ethnolinguistics, ed. D. H. Hymes and W. E. Bittle, pp. 335-350.
Vierra, Bradley J., and Richard I. Ford
2006 Early maize agriculture in the northern Rio Grande Valley, New Mexico. In Histories of Maize, edited by J. E. Staller, R. H. Tykot, and B. F. Benz, pp. 497-510. Elsevier, Amsterdam.
Ware, John A.
2014 A Pueblo Social History: Kinship, Sodality, and Community in the Northern Southwest. School for Advanced Research Press, Santa Fe.
Wendorf, Fred, and Eric Reed
1955 An alternative reconstruction of Northern Rio Grande prehistory. El Palacio 62(5-6):131-173.
Wilcox, David R., William H. Doelle, and J. Brett Hill
2003 Coalescent Communities GIS Database: Museum of Northern Arizona, Center for Desert Archaeology, GeoMap Inc. Manuscript on File at The Center for Desert Archaeology, Arizona.
Wilshusen, Richard H., and Ruth Van Dyke
2006 Chaco’s Beginnings. In The Archaeology of Chaco Canyon: An Eleventh-Century Pueblo Regional Center, edited by S. H. Lekson, pp. 211-259. SAR Press, Santa Fe, New Mexico.
Wilson, C. Dean
2008 Examination of trends for Galisteo Black-on-white. In Chasing Chaco in the Southwest: Papers in honor of Frances Joan Mathien, edited by R. Wiseman, T. O’Laughlin, C. Snow, and C. Travis, pp. 201-215. Archaeological Society of New Mexico No. 34. Albuquerque.
Whorf, B. L., and G. L. Trager
1937 The relationship of Uto-Aztecan and Tanoan. American Anthropologist 39:609-624
Figure1. Hopi style kiva at the Davis Ranch Site, San Pedro Valley, Arizona. Courtesy of the Amerind Foundation, Inc., Dragoon, Arizona.
Figure 2. Kayenta style entry box, plan and profile. Davis Ranch Site, San Pedro Valley, Arizona. Reproduced from Clark at al. 2012:Figure 6.17 (adapted from Lindsay et al., 1968:Figure 204).
Figure 3. Example of a Kayenta style perforated plate. Photograph by Peter J. Pillis.
Figure 4. Graph of population growth and decline in the Central Mesa Verde and the northern Rio Grande regions (after Ortman 2012, figure 4.8. Reproduced with permission of the author).
Figure 5. Coalescent Communities GIS Database: Population density estimates circa 1250 CE. Archaeology Southwest, derived from Hill et al. 2012.
Figure 6. Coalescent Communities GIS Database: Population density estimates circa 1450 CE. Archaeology Southwest, derived from Hill et al. 2012.
 This paper is dedicated to the memory of Douglas W. Schwartz (1929-2016), who organized and chaired the Arroyo Hondo Seminar where the paper was presented. Doug’s passion for understanding deep history inspired some of the best work in our field. He will be greatly missed.
Stanley Stubbs conducted a partial excavation of LA 835 in 1951 and concluded that despite the evidence of pottery from the San Juan Basin, the site was likely built and occupied by people indigenous to the Rio Grande.
 The author attended a seminar at the Amerind Foundation in September, 2015, in which Kayenta and Mesa Verde migrations were compared. Organized by archaeologists from Arizona who may have been unfamiliar with the details and nuances of his arguments, Ortman’s model was generally accepted as state-of-the-art for eastern Pueblo migrations.
 Polly Schaafsma (personal communication, December 2015) questions this connection considering the very different rock art styles of the Four Corners and Fremont regions.
 See Murdock (1949) for a discussion of the relationship between matriliny and endogamy. Even the bilateral Tewa and Tiwa are endogamous, which, along with surviving kinship terminologies, suggests the possibility of a prior matrilineal state (Whiteley forthcoming).
 In Hopi oral tradition, newly arriving clans were allowed to join existing communities if they could contribute new and effective rituals to the annual ceremonial cycle.