Revays auto service harrisburg pa

Week 2 - Paleoindian/Echo-Hawk 2000 - Ancient History in the New World.pdf
Society for American Archaeology

Ancient History in the New World: Integrating Oral Traditions and the Archaeological Record in Deep Time Author(s): Roger C. Echo-Hawk Reviewed work(s): Source: American Antiquity, Vol. 65, No. 2 (Apr., 2000), pp. 267-290 Published by: Society for American Archaeology Stable URL: http://www.jstor.org/stable/2694059 . Accessed: 05/03/2013 20:05

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

.

Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to American Antiquity.

http://www.jstor.org

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=sam
http://www.jstor.org/stable/2694059?origin=JSTOR-pdf
http://www.jstor.org/page/info/about/policies/terms.jsp
http://www.jstor.org/page/info/about/policies/terms.jsp
ANCIENT HISTORY IN THE NEW WORLD: INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL

RECORD IN DEEP TIME

Roger C. Echo-Hawk

Oral traditions provide a viable source of information about historical settings dating back far in time-a fact that has gained increasing recognition in North America, although archaeologists and other scholars typically give minimal attention to this data. The Native American Graves Protection and Repatriation Act (NAGPRA) lists oral traditions as a source of evidence that must be considered by museum andfederal agency officials in making findings of cultural affiliation between ancient and mod- ern Native American communities. This paper sets forth the NAGPRA standards and presents an analytical framework under which scholars can proceed with evaluation of historicity in verbal records of the ancient past. The authorfocuses on anArikara narrative and argues that it presents a summary of human history in the New World from initial settlement up to the founding of the Arikara homeland in North Dakota. Oral records and the archaeological record describe a shared past and should be viewed as natural partners in post-NAGPRA America. In conceptual terms, scholarship on the past should revisit the biblio- centric assumptions of "prehistory," and pursue, instead, the study of "ancient American history "-an approach that treats oral documents as respectable siblings of written documents.

Las tradiciones orales proveen un manatial de informaci6n sobre escenas historicas muy antiguas-una realidad que ha aumen- tado en reconocimiento en Norte America, aunque arqueologos y otros academicos tipicamente le prestan atencidn minima a estos datos. La ley de repatriacion y protecci6n de tumbas indigenas de 1990 lista tradiciones orales como evidencia que debe consid- erar se en el establemiento de afiliaci6n cultural entre las comunidades indigenas del pasado y las de tiempo moderno. Museos, agenciasfederales, tribus indigenas y academicos en los Estados Unidos confrontan un reto especial en dirigir este aspecto de la ley porque exist poca direccion en el uso efectivo de tradiciones orales en el estudio de epocas antiguas. Este articulofija el estdn- dar de la ley de 1990, y tambken presenta una estructura analitico, donde se puede proceder con la evaluaci6n de la historicidad en el testimonio verbal del pasado. Enfocdndose en la le yenda de origen de los Indios Arikara y otras narraciones indigenas, el autor enseha como testimonies orales dan luz a la historia humana en una epoca muy antigua-en este caso, de la poblaci6n ini- cial del Nuevo Mundo un tiempo reciente en los grandes llanos. Este andlisis tiene implicaciones importantes para la construc- ci6n de modelos de la historia humana. Los testimonies oral y arqueol6gico se deben ver como complementos la ley de 1990. Esta perspectiva coneptualiza a la historia indigena norteamericana como dependiente no solo en documentos escritos una disciplina pero en tradiciones orales.

T hroughout the twentieth century, a complex dialogue on the cultural world of ancient North America has emerged from archaeol-

ogy and other disciplines, unfolding from a vast spec- trum of journals, books, technical reports, and popular media. Through the development of taxo- nomic systems and the analysis of artifact assem- blages and sites, an ever-growing community of scholars has sought to define cultural units in the American archaeological record, understand rela- tionships between them, and trace processes of change over time. With the passage of the 1990

Native American Graves Protection and Repatriation Act (NAGPRA), Congress entrusted federal agen- cies and museums with a mandate to evaluate rela- tionships between ancient and modern Indian societies. It is no wonder that archaeology has played a prominent role in this process. The NAGPRA con- cept of "cultural affiliation" requires that the formal affirmation of connections between groups be based upon a review of readily available evidence, and archaeology contributes a valuable source of infor- mation for this purpose. NAGPRA also lists oral tra- ditions as a source of evidence on cultural affiliations.

Roger C. Echo-Hawk * Denver Art Museum, 100 West 14th Avenue Parkway, Denver, CO 80204-2788

American Antiquity, 65(2), 2000, pp. 267-290 Copyright ( 2000 by the Society for American Archaeology

267

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
268 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

Verbal literatures raise a special challenge for museum and tribal officials who may have little com- mon ground in assessing such information for evi- dence on ancient history. Although the academic community and Native Americans have embraced a broad spectrum of attitudes toward historicity in oral literature, extreme perspectives have tended to dom- inate discourse between these groups. In this polar- ized world, tribal historians and religious leaders frequently rely on oral traditions as literal records of ancient history, while most academically trained schol- ars respond with skeptical rejection of verbal litera- ture as a vehicle for transmitting useful information over long time spans. Tribal leaders, museum admin- istrators, and federal agency officials who may have little or no background in working with oral traditions are still required by NAGPRA, as a practical matter, to render judgments and set policies on what consti- tutes "evidence." Even parties who may have every intention of finding common ground face serious prob- lems in making effective use of verbal records.

Throughout much of the twentieth century, the academic study of ancient America focused on devel- oping models based on archaeology and other fields of anthropology, with only sporadic attention on oral traditions. The final decades of the century, how- ever, saw a dramatic increase in interest among schol- ars in exploring oral literatures for information about ancient events, partly as a result of the passage of NAGPRA, but primarily because oral documents have potential for shedding light on historical set- tings dating back far in time, considerably enriching academic constructions of ancient human history.

The concept of "prehistory" presumes the absence of firsthand written records before a specific point in time, and the employment of this terminology as a primary taxonomic tool suggests that oral traditions either do not contain any information relevant to the time period or that they preserve something other than "history." The study of oral traditions has only recently begun to reveal the degree to which verbal messages can preserve firsthand observations over long spans of time, but it is clear that oral and writ- ten documents both deserve comparable status as records that can be analyzed for valid evidence about human history. The replacement of "prehistory" with "ancient history" as a conceptual framework would recognize oral traditions as records of history.

In this paper, I explore aspects of the ancient his- tory of Caddoan America. An archaeological litera-

ture of great vitality investigates the commingled roots of the modem Arikara, Pawnee, Wichita, and other groups, and the verbal literatures of these groups reflect a similarly complex history. Linked by a com- mon linguistic heritage, as well as by material cul- ture and lifeways, the shared history of diverse populations of the Central and Southern Plains extends into deep time, revealed in both oral records and the archaeological record. As a matter of careful scholarship, a range of useful analytical tools can be applied to oral traditions to illuminate their historical content, and under NAGPRA, the careful study of all information controls the assessment of ties among modem and ancient Native American communities.

Cultural Affiliation under NAGPRA

Three categories of claimants have standing to assert repatriation claims under NAGPRA: lineal descen- dants, federally recognized Indian tribes, and Native Hawaiian organizations. The rights specified for lin- eal descendants in claiming human remains and asso- ciated funerary objects reflect the principle accepted throughoutAmerican society that next-of-kin should have the authority to make suitable arrangements for the disposition of the remains of deceased kin. The status of Indian tribes and Native Hawaiians in NAG- PRA is not so much derived from racial classifica- tion as it is drawn from acknowledgment of the sovereign rights retained by these groups in their complex relationships with the United States. Thus, Native American next-of-kin and sovereign com- munities have the ability to assert authority under NAGPRA over human remains and associated funer- ary objects to which they can show a connection. This connection, "cultural affiliation," has a specific statu- tory definition in NAGPRA, with the meaning that "there is a relationship of shared group identity which can be reasonably traced historically or prehistori- cally between a present day Indian tribe...and an identifiable earlier group."

NAGPRA requires that, following consultation with affected communities, federal agencies and museums make findings of cultural affiliation for human remains and associated funerary objects held in their collections. Federally recognized Native American communities and lineal descendants not identified and affiliated through this process can assemble and present a preponderance of the evi- dence showing a cultural affiliation. Museum and federal agency officials necessarily sit in judgment

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 269

of any such submissions of evidence by claimants. A cultural affiliation under NAGPRA is deemed

to have been "reasonably traced" when it is sup- ported by a "preponderance of the evidence," con- sisting of more than 50 percent of the total realm of relevant evidence. This evidence can be drawn from "geographical, kinship, biological, archaeological, anthropological, linguistic, folkloric, oral traditional, historical, or other relevant information or expert opinion." In U.S. law, evidence is loosely defined as information that supports a conclusion, and in schol- arship, evidence is viewed as information that has survived critical scrutiny according to applicable aca- demic standards. Neither environment is particularly friendly toward unsupported opinion, religious belief, or speculation. Proper findings of cultural affiliation favor or disfavor a relationship on the basis of fair consideration of the full spectrum of readily available evidence. Since evidence cannot be scien- tifically quantified or weighed, the existence of a "preponderance" of it must be determined through informed interpretation.

Under regulations issued by the National Park Service (NPS) in December 1995, a cultural affilia- tion has been established when a preponderance of the above evidence "reasonably leads to such a con- clusion." Three criteria have been set forth which must be met to support a finding of cultural affilia- tion. First, a present-day Indian tribe must have stand- ing to make a claim; in other words, claimant tribes need to be federally recognized. Second, evidence must support the existence of an "identifiable earlier group." Support for the existence of such groups may include evidence that 1) establishes the group's "identity and cultural characteristics"; 2) shows "dis- tinct patterns of material culture manufacture and dis- tribution methods"; or 3) establishes the group "as a biologically distinct population." Other unspecified support for the existence of the earlier group also is permissible to include under these guidelines. Third, evidence must show that both groups have a shared group identity which can be reasonably traced and a preponderance of the evidence "must establish that a present-day Indian tribe ... .has been identified from prehistoric or historic times to the present as descend- ing from the earlier group." This finding "should be based upon an overall evaluation of the totality of the circumstances and evidence pertaining to the connection...and should not be precluded solely because of some gaps in the record." Nothing in

NAGPRA prevents the existence of multiple cultural affiliations of human remains and objects to more than one present-day Native American tribe.

In short, all information that qualifies as legitimate evidence must be considered, and support for a cul- tural affiliation must be based on an opinion that this evidence tends to favor - even slightly-a connec- tion. The lack of a complete chain of connection is not grounds for denying the existence of a relation- ship. Moreover, the preponderance of evidence stan- dard permits affirmation of a cultural affiliation even when much uncertainty exists. In assessing connec- tions over time between "cultural units," archaeolo- gists as a rule aim for a higher degree of certainty than that called for under NAGPRA. That is, in cases where evidence points to a connection between units, archae- ologists may be reluctant, as a matter of professional accountability, to assert a cultural affiliation between these units if the evidence falls short of a fairly high standard, such as "beyond a reasonable doubt."

Most scholars prefer not to draw conclusions, pub- lish findings, or present conference papers arguing for cultural connections on the basis of a mere pre- ponderance of the evidence. In addition, when evi- dence hovers anywhere near the 50% threshold, little agreement may be forthcoming as to which side of the line the evidence should fall. Therefore, it is rea- sonable to wish for more evidence in hand than NAG- PRA requires in affirming a cultural affiliation. NAGPRA is not intentionally designed to facilitate the settling of archaeological debates about taxonomy, although it will spark substantive refinement of tax- onomic assignments and relationships. The purpose of the law is to ensure that Indian tribes and lineal descendants have roles in determining the appropri- ate disposition of Native American human remains and funerary objects. This respects the status of these parties as sovereign governments and as next-of-kin.

For Indian tribes, an ideal assertion of cultural affiliation is one based upon evidence showing some form of transmission of sovereignty from the earlier group to the claimant tribe. To the extent that archae- ology contributes evidence on connections between groups, the profession should encourage the pro- duction of technical reports purposefully useful to Indian tribes, museums, and federal agencies in implementing NAGPRA. Applying the lower thresh- old of a preponderance of the evidence does not mean that a lower standard of scholarship should be employed. The forming of professional opinions

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
270 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

about cultural affiliation must be based on the best possible scholarship no matter which threshold of certainty is applied to the results of research. The pre- ponderance standard is primarily intended to help guide the comparison of conflicting evidence; infor- mation that fails to qualify as evidence has no weight under this standard.

Spiritual information holds great value to tribes since the treatment of the dead in every human soci- ety is a fundamentally religious activity. Such knowl- edge can usefully guide the development of research projects and tribal repatriation agendas, but personal visionary experiences of a religious nature are sub- ject to highly idiosyncratic interpretation and should not be submitted or accepted as evidence on cultural affiliation under NAGPRA. In matters of academic scholarship, spiritual insights and unsupported opin- ions-no matter how popular or strongly stated- typically elude critical analysis and cannot qualify as historical evidence.

Using Oral Traditions under NAGPRA

The study of oral literature as history features two major subdivisions: oral history and oral traditions. Oral history is best defined as the verbal memoirs of firsthand observers, while oral traditions are ver- bal memoirs that firsthand observers have passed along to others. Oral history is the focus of a bona fide and well-established segment of the academic community, receiving much attention from cultural anthropologists, ethnohistorians, and other scholars. The investigation of oral traditions that pertain to ancient settings lacks a similarly strong disciplinary infrastructure, but the most widely accepted approach for academic study of verbal literature per- taining to both shallow time and deep time is to seek independent corroboration of statements and set- tings discussed in oral records (Fentress and Wick- ham 1992:76-86; Ritchie 1995:95-101). Douglas Parks has succinctly characterized the range of views embraced by anthropologists:

Anthropologists are by no means in agreement on the historical validity of events and locations occurring in myths. Some, like Robert Lowie, completely rejected all myths as accurate sources of any historical fact, while others, like Paul Radin, believed that historical events and past cultural patterns can be reconstructed from myths. Perhaps the majority, though, subscribe to Edward Sapir's belief that authentic informa- tion can be found in myths when it is corrobo-

rated by other lines of evidence (e.g., archaeo- logical, linguistic, or ethnographic). For most cultural historians it seems fair to say that myths can indeed provide historical clues if used judiciously and in conjunction with inde- pendent forms of corroboration. (Parks 1985:57)

Only a small number of twentieth-century schol- ars have looked to oral traditions for insights into ancient historical events and cultural settings, but such studies appeared with increasing regularity dur- ing the final decades of the century (Bacon 1993; Bahr et al. 1994; Begay and Roberts 1996; Benn 1989; Ellis 1967,1979; Fewkes 1898; Hall 1983,1997; Henning 1993; Levi 1988; MacGregor 1943; Malotki and Lomatuway'ma 1987; Mann and Fields 1997; Moodie et al. 1992; Patterson-Rudolph 1997; Pen- dergast and Meighan 1959; Schlesier 1987; Sheppard 1998; Strong 1934; Teague 1993; Vehik 1993). These publications integrate knowledge derived from archaeology with knowledge from oral traditions, revealing, in some cases, vastly richer depictions of human history than can be uncovered through the archaeological record alone or oral traditions alone.

As a matter of convenience and necessity, admin- istrators at museums and federal agencies must rely heavily upon archaeological literature as a guide to NAGPRA cultural affiliations for ancient human remains and funerary objects, but the law calls for a new commitment to investigating and utilizing oral traditions. Oral traditions must be considered together with extant archaeological, biological, and other pertinent available evidence, and the full weigh- ing of all of this evidence must establish that it tends to favor or disfavor a cultural affiliation. Archaeol- ogists and historians form conclusions about human history through the analysis of evidence from many realms of scholarship, and oral traditions should properly be viewed as simply one more body of evi- dence. Opinions will vary in creating, comparing, and weighing categories of evidence. This process should therefore ideally include some effort at dialogue between interested parties to establish a common understanding of appropriate approaches to weigh- ing this evidence.

If necessary, it might be helpful to treat the "archae- ological" evidence as consisting of multiple separate units, such as osteological analysis, radiocarbon dates, mortuary treatment, and artifactual evidence-with oral traditions included as one more additional cate- gory, perhaps with its own subdivisions, depending

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 271

upon the variety of oral traditions that can be applied in a given situation. Clearly, the weighing of evidence under NAGPRA can never be a purely scientific exer- cise, with totally objective measurements of weight assigned to each applicable body of evidence. Instead, the evaluation of the evidence must be performed in a manner that can be justified as reasonable.

Assertions of cultural affiliation necessarily involve the assembling and interpreting of informa- tion, and it is reasonable to approach this task by scru- tinizing each element of evidence according to applicable academic standards. A standard rule of historiography is that source materials, whether con- sisting of written records or oral documents, should be critically evaluated rather than simply taken at face value. Reconciling data from different sources and assessing the reliability of eyewitnesses provides his- torians with the basic means of critiquing materials, but in situations where no other evidence exists for comparative purposes and eyewitnesses cannot be scrutinized for reliability, two analytical standards, which I term "compatibility" and "reasonability," can provide minimal assurance of historicity or potential historicity. NAGPRA requires the consid- eration of oral traditions that arguably contain actual evidence about the historical past.

First, the historical content of the oral or written information should be compatible with the general context of human history derived from other types of evidence. In cases where existing models are sub- stantially modified or overturned, then a rational jus- tification for accepting such changes needs to be presented. In other words, if acceptance of informa- tion from an oral tradition would generate conflict with an existing model of historical settings based on osteology, then such acceptance must include appropriate scholarship that leads to justifiable doubt as to the osteological evidence. The osteological evi- dence and the conclusions based on it cannot sim- ply be disregarded.

Second, the oral information must present a per- spective on historical events that would be accepted by a reasonable observer. Under this standard, all evi- dence used to construct past historical settings should pass a level of scrutiny that might be applied by a well-informed third party with no vested interest in the outcome. Defining the standards that this edu- cated bystander would follow may always prove a matter of art rather than science, but presumably such a person would value guidance from experts

who have studied oral traditions for information about ancient times.

As a general matter of academic scholarship, oral information is regarded as a weak source of evidence when it cannot be corroborated. One authority on oral history, Donald Ritchie, observes that "the more con- troversial the subject, the less an interview can stand alone"-a view aimed at verbal memoirs of firsthand observers, but which is especially true of oral tradi- tions handed down from firsthand observers (Ritchie 1995:94, 99-100). If an oral tradition about ancient times cannot be supported by other evidence, skep- ticism and even rejection is warranted. In formulat- ing ideas about the past, people are free to adopt an uncritical approach to oral traditions, but when such information is submitted as evidence, then scholars are obligated to treat it accordingly. The standards of compatibility and reasonability may have utility in salvaging some evidence from oral information that cannot be successfully identified through more reliable means, but critical assessment in scholarship means that some information will be accepted and some will be rejected as useful evidence.

The NAGPRA consultation process may elicit oral traditions from tribes (in addition to published oral traditions), but the study of this information can yield minimal or questionable results. For adminis- trators who wish to establish or enhance positive relationships with Indian tribes, it will be a challenge to solicit oral traditions, obtain such information, and then do a credible job in utilizing it according to applicable academic standards. The energy devoted to such efforts will vary and may yield problematic outcomes, but no archaeologist deliberately strives for minimal scholarship as a basis for drawing con- clusions. Since NAGPRA cultural affiliations rely on a preponderance of evidence based on information from relevant sources, the law should not be read as a carte blanche empowerment of the view that all oral traditions ought to be accepted as literal history. Appropriate analysis can identify the historical con- tent in oral records with at least minimal reliability.

Analytical Criteria for Oral Traditions

In an important work that appeared in its most recent form in 1985, Jan Vansina, a scholar of African his- tory, argued that a "three-tiered" hourglass pattern can be perceived in the preservation of historical information by oral means. In the top portion of this hourglass, a great deal of chronologically ordered

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
272 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

information typically exists that tapers off in quan- tity as the stories move farther back in time. The sec- ond "tier" is actually a gap of information that Vansina terms the "floating gap." The bottom tier contains information that has been fused into a dis- organized period of origins lacking any coherent sense of relative chronology. This body of knowl- edge flares out to encompass a mass of information about undated events beyond living memory, and as societies move forward in time, the floating gap also can move forward, with the lower tier in the hour- glass absorbing information that has lost chrono- logical specificity.

The primary focus of Vansina's Oral Tradition as History was on "verbal messages" concerning events of recent centuries, and the author made no effort to establish criteria for determining the possible longevity of orally transmitted information. The portion of the hourglass that encompasses discussions of "origins" thus has no definite floor against which the sands of time fall. While Vansina warned us that events men- tioned in "origin" stories may only date back several centuries or less, he also acknowledged that "heavily fossilized" information can persist in origin stories, and such traditions "may or may not remain stable over long periods of time" (Vansina 1985:21-22). This underscores the fundamental necessity for scholars to evaluate the historical information in a given oral tra- dition by measuring its content, where possible, against other relevant data about the past.

On a case-by-case basis, three main possibilities can be said to exist. First, a given narrative may have been simply manufactured at some point in the near or distant past as an entertaining fiction or for other nonhistorical purposes. Second, a given narrative may offer an unadorned account of ancient histori- cal events or settings, carefully preserved and handed down over unknown spans of time. Finally, a given narrative may contain some historical information that has become encrusted with fictional trappings. With this range of possibilities, how can we distin- guish individual verbal texts? I have identified three tests as appropriate to this process:

Test 1. The oral tradition or element of a tradition should tend to fit into Jan Vansina's classification of a "group account" and/or "traditions of origin and genesis" (Vansina 1985:19-24). In terms of his "hourglass" pattern, the verbal information selected for analysis should clearly fall into the bottom por- tion of the hourglass; that is, it should at best exhibit

only vague chronological indicators in its relation- ship to historical events mentioned in other oral tra- ditions of the society in question.

Test 2. The oral tradition should be presented in its native context as a story about events that are presumed to be historical. In some cases, a specific element in an oral tradition might be presumed to be historical, while the tradition itself is viewed as fictional. For example, we might agree that Gone with the Wind is a work of fiction, whereas a major event described in the story-the Civil War- actually occurred.

Test 3. The historical content of verbal literature must be supported or verified through evidence gath- ered from independent, non-verbal sources, such as through archaeological data, written records, or other accepted sources of evidence about the historical past. In other words, to the greatest degree possible, the "historical" messages in oral traditions must be generally consistent with constructions of the past that are based on non-verbal sources and are broadly viewed as reasonably acceptable (or at least theo- retically possible) models of the historical past.

If oral traditions (or specific elements within the texts) pass these various tests, then a presumption favoring historicity can exist-narratives failing the third test may point the way to needed future research in archaeology. Improbable results from the appli- cation of these tests should be subjected to the stan- dards of compatibility and reasonability for further evaluation. Finally, oral records which have survived long-term social developments may not survive unscathed. Verbal literature represents an inherently malleable medium of discourse, and changes occur in textual content from generation to generation, though the "rate" of such change may be quite vari- able. I conceptualize some of these changes in terms of a "principle of memorability."

My principle of memorability predicts that the transmission of historical oral traditions over long periods of time will inevitably introduce changes to texts involving one or more of the following factors: 1) elisions, omissions, or conflations will most likely serve to enhance the entertainment value or memo- rable quality of historical information; 2) the most memorable elements of a historical narrative may be emphasized at the expense of complex, detailed data; 3) data and stories that are viewed as important doc- uments may incorporate elements that begin as spec- ulative interpretation and end up as elements that enhance the entertainment value and color of the

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 273

data/story; 4) only those historical stories that are seen as inherently valuable texts and display ele- ments making the text more memorable will survive long transmission periods; and 5) information about the ancient past will more likely persist if it is encrusted with nonhistorical cultural meanings and narrative elements that are specific to transmitting societies. In short, for a verbal text on ancient his- torical events to endure for millennia, it must be col- orful as well as explanatory.

The principle of memorability helps to explain why oral traditions generally do not respect the ten- dency of historians to hold forth at length in dry detail on obscure events of the past. Over time, densely "footnoted" verbal narratives must inevitably give way before the interpreted versions of story- tellers who can enliven the dullest historical narra- tive by emphasizing its most memorable aspects. With the principle of memorability in mind, it is unnecessary to postulate the existence of successive generations of carefully trained oral historians with eidetic memories to explain the long-term preserva- tion and transmission of verbal information. Sys- tematic training of oral historians does occur in many societies, but this may not adequately explain verbal durability even in those situations. It would be a rare human society that has ever been wholly without a storyteller, and even a dull storyteller can transmit the specifics of memorable stories. In fact, we might presume that the very invention of written methods of preserving information resulted from social needs that required the preservation of information so dull and bland that even the best storytellers could not hope to retain such dry material or meet the chal- lenge of interesting bored audiences.

All analytical tools, such as the standards of com- patibility and reasonability and the principle of mem- orability, have utility only to the degree that they can be successfully employed to reconcile diverse realms of information and help build workable models of the past. Archaeologists necessarily have a primary focus on the material manifestations of past popula- tions, but if oral traditions can be successfully inte- grated with the archaeological record, this has important implications for the construction of ancient historical settings.

A Spoken Past

The durability of oral literatures is a matter of debate, but folklorists generally accept the idea that fictional

stories can persist for millennia (Lankford 1987:243; Thompson 1966:xxi-xxiii; Wiget 1985:6). More- over, a spectrum of oral traditions from around the world have potential for shedding light on Pleis- tocene settings (Echo-Hawk 1994). These include the endurance of memories of such phenomena as Arc- tic Circle patterns of solar movements, the observed transition from Arctic Circle to lower latitude diur- nal/nocturnal cycles, descriptions of permafrost thawing/freezing, Pleistocene weather patterns, the existence of European and American glacial ice sheets, sea-level changes associated with ice sheet expansion and melting, glacial lakes, the onset of Holocene seasonality, and human interaction with extinct megafauna. Verbal literature arguably pre- serves glimpses and echoes of the long-vanished Pleistocene world of our ancestors, so we should also search oral records for perspectives on more recent time periods.

Scholars generally do not see value in assessing oral traditions against an archaeological record extending back much further in time than a thousand years or so, because it is widely assumed that some form of barrier or boundary prevents information from being effectively conveyed into the present from distant time periods. Although scholarship has estab- lished the malleable nature of verbal literature, it is difficult to find viable arguments that set justifiable limits of transmission time. Most scholars would be dismayed to discover that little or no support exists in scholarship that sustains their favored presumption on the limits of verbal durability, whether the pre- sumed limit is set at 100 years or 10,000 years beyond the living memory of firsthand observers.

It is important to construct a reasoned basis for determining a possible chronological boundary for the maximum length of time that verbal information of any intricacy can be sustained. To date, where such boundaries have been drawn, their existence has relied largely upon the absence of demonstrable con- nections between oral traditions and other acceptable evidence about datable past events. A reasonable boundary for the long-term preservation of verbal lit- erature might be linked to the beginnings of com- plex social interaction requiring the regulation of knowledge, and the oldest settings and events dis- played in human origin stories are bounded by their artifactual nature as records generated in communal, multigenerational social settings (Echo-Hawk 1994:150-162).

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
274 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

On this basis, I speculate that the majority of oral traditions that contain historical information gener- ated by firsthand observers can go back no farther in time than about 40,000 years, though this boundary must vary greatly from region to region. Many schol- ars suggest that human history over the last 40,000 years is primarily a story of sustained social com- plexity, and this argues strongly for the concurrent preservation of oral information. If the level of human social interaction up to a given point in time does not require the preservation and regulation of informa- tion, then there can be little need for the generation of a literature that provides a sense of group history.

Scholars do not typically look for history in cre- ation and origin stories. In fact, the term "pseudo-his- tory" has been coined as a description that reflects the general attitude of scholars who study Native American stories about human origins (Wheeler- Voegelin and Moore 1957:72). Academic suspicion of origin stories gelled in the views of Robert Lowie (1917), who issued a scathing denunciation of oral traditions as a source of history. Lowie's extreme views were not uniformly endorsed among his col- leagues, but creation/origin stories subsequently became the exclusive domain of cultural rather than historical scholarship. Analysis of origin stories for cultural content represents legitimate intellectual inquiry, but the identification of cultural meanings in statements and scenarios does not automatically pre- clude historicity. The principle of memorability expects historical elements to better endure in verbal literature when embedded in meaningful cultural data.

Oral traditions, like other forms of information, create a bounded world in which some aspects of past historical processes are illuminated, while other aspects may be submerged and hidden from view or otherwise distorted. Thus, oral accounts of the past do not provide us with the equivalent of a carefully fossilized duplication of the structures of ancient his- torical events. In terms of the known intricacies of historical processes that account for specific social formations at any point in time, origin stories may feature echoes of such circumstances, but the prin- ciple of memorability explains how verbally trans- mitted documents tend to oversimplify past settings. One indication of this deception at work is that ori- gin stories often give the impression that particular social groups have moved through time and space as unchanged monolithic cultural units. Most typically, however, populations intermingle, fission, and

undergo various forms of major and minor changes in population composition. Oral traditions can erase these complexities, looking into the distant past to see an exact reflection of a group's contemporane- ous social state, but societies simply do not travel unchanged into the present from the distant past.

Migration stories constitute a widespread form of origin story in North America. Such stories com- monly refer to multiple locations as stopping points in the journey of an ancestral group - rest areas along a migration superhighway. These narratives can pre- serve firsthand observations of population move- ments and sojourns of specific tribal groups in various localities, but other historical processes might also be at work. What sounds like one population moving intact from place to place, for example, might actually reflect a history in which several groups from various locations came together in some fash- ion over time. The memories of population subdivi- sions residing at different contemporaneous sites across a given region might come to be portrayed among consolidated descendants as a migration, per- haps with the story of one subgroup eventually pre- served as the history of the whole society. Despite these problems of interpretation and theory, it seems unlikely that migration stories were fabricated from purely cultural materials, or as colorful expressions of creative imaginings. Since migration traditions can potentially explain, as well as obscure, a variety of past circumstances, ideal assessments of such sto- ries would utilize archaeological and other evidence to help assess traditions of population movements from one region to another.

An extensive body of Caddoan creation, origin, and migration stories have been published. Scholars have suggested, with varying degrees of caution, that historical settings may be found in these stories (Blaine 1979; W. Wedel 1979), but few studies have appeared, and archaeologists have generally given little or no consideration to oral traditions. It is grad- ually becoming clear, however, that evidence from both oral traditions and the archaeological record can work together to more richly reveal the ancestry and ancient history of Caddoan America. Looking westward to the Rocky Mountains and adjoining regions, specific details of this history loom out of the distant past.

Tales of Travelers

A number of Arikara origin stories of several differ-

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 275

ent types have been recorded and published, includ- ing versions of an emergence story (Dorsey and Murie 1904a: 12-17, 18-23, 23-25, 26-30, 31-32, 32-35; Gilmore 1930, 1987:31-32, 173-175, 184-188; Grinnell 1916:186-194;). In this story, the people emerged from the earth and followed Mother Corn upon a great migration, encountering various obstacles. In one version, the leader of the "migra- tion" was a woman known simply as "Mother" (Dorsey and Murie 1904a:37-38), but most of the accounts name Mother Corn as the central player in the historical formation of the Arikara community. In a version published by George Bird Grinnell (1916; also see Grinnell ca. 1890:File # 156, "Ree Cosmology"), the people received assistance for their journey through the gift of a bundle to a young boy, and corn is not mentioned in the narrative until after the arrival of the Arikara in the Central Plains (also see Dorsey and Murie 1904a:26-30 for compari- son). As a group, the extant versions of the Arikara emergence story convey a general impression of somewhat disordered geographic settings and his- torical events floating in time. Ancient human his- tory unfolds in a swirl of details.

In the course of the great journey, as the people encounter each obstacle, a portion of the population becomes scattered. The barriers generally include an impassable body of water, a great forest, and a deep canyon, appearing in variable order, with dif- fering descriptions. One story, for example, describes the water as "wide, thick ice and deep water" (Dorsey and Murie 1904a: 33). Another nar- rative states that when the people first emerged from the earth, they found themselves on "an island in big waters" (Dorsey and Murie 1904a:28), and after crossing over to a place of darkness, they eventu- ally developed a stone technology and emerged from the darkness.

The migration includes a period of sedentary life, when the people settled down "for some time" after passing the third obstacle while Mother Corn "returned to the heavens" (Dorsey and Murie 1904a: 15-16). At this place they learned to gamble, and when players from two different communities met to compete at "shinny ball and four sticks," the losers attacked the winners and a pitched battle ensued. An alternate version associated the cessation of fighting with the founding of a new social order: "When Mother-Corn returned from the heavens she brought with her a man who said that Nesaru was

displeased with their doings; that now he was to give them rules and laws to go by; and that the people were to select a man whose name should be Nesaru, chief' (Dorsey and Murie 1904a: 16).

Arikara traditional history unfolds as a set ofjour- neys from various places into the Central Plains and then up the Missouri River to their present home- land. These stories associate very specific locations with Arikara ancestors, including Nebraska, South Dakota, the Republican River in Kansas, the Black Hills of South Dakota, the Cross Timbers of Okla- homa and Texas, the Rocky Mountains, and possi- bly the Grand Canyon andArctic Circle. In searching for historicity among these Arikara migration narra- tives, we should not necessarily look for the move- ment of one monolithic Caddoan-speaking population from one implied or named location to the next. Interaction among people of diverse cul- tures must account for the Arikara people at every point in time, so it may be more useful to search the places mentioned in the Caddoan origin stories for evidence that people in those regions contributed to the formation of later Caddoan populations.

Like the Arikara stories, Skidi Pawnee origin sto- ries envision a great journey beset with obstacles. In the Pawnee stories, however, Morning Star makes thejourney and overcomes various obstacles in com- pany with Sun, and the two male deities encounter obstacles that differ greatly from those met by humans in the Arikara story. The significance of these differences in origin stories is not immediately clear from the standpoint of historical content, but both the Arikara and Skidi origin stories touch upon the theme of relationships between the sexes. The Skidi story can be said to emphasize the deeds of a male deity, while the Arikara story places a female deity at the center of unfolding events. These elements can be investigated for coexisting cultural and his- torical content. In terms of history, for example, sto- ries on the theme of gender relations often interweave the cultivation of corn and the empowerment of women, and such associations can be investigated in the archaeological record.

People of the Ancient Blue Mountains

Sometime around 1890, George Bird Grinnell col- lected several Arikara origin accounts and subse- quently published them as an emergence/migration story. This story begins with a series of movements by all of humanity from an underground world

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
276 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

beyond various obstacles until the people "came to some high hills called the Blue Mountains... " (Grin- nell 1916:186-194). This published version does not stipulate the location or further identity of the Blue Mountains, but the original manuscript of this nar- rative, held at the Southwest Museum in Los Ange- les, specifies that "Blue Mountains" is the Arikara name for the Rocky Mountains (Grinnell ca. 1890:File # 156, "Ree Cosmology").

It is not clear why Grinnell excluded the refer- ence to the Rocky Mountains from the published version, but it would have been reasonable for him to believe that he was dealing with a tale of largely non-historical character. He may have felt it appro- priate to spare his audience from the distraction of pondering the apparent historical specificity of the "Rocky Mountains" as a location for Arikara ances- tors in a narrative that he presumed to be made up of a mostly imagined events or pseudo-history.

The reference to the Rocky Mountains could be an error of some kind, particularly since no other available Arikara origin story mentions either the Blue Mountains or the Rocky Mountains. Chester Ellis (personal communication 1999) suggests that Grinnell's mention of the Blue Mountains is proba- bly a reference to Blue Butte near Mandaree, North Dakota, rather than the Rocky Mountains. Grinnell could have dropped the reference to the Rocky Mountains from the published version because he became aware in some fashion that it misrepresented the actual story. Thus, it should not be assumed that this reference is historical and that it refers to the Rocky Mountains. Due to the state of knowledge about the archaeological record for the Rocky Moun- tains and adjacent regions at circa 1900, Grinnell was not in a position to conduct any useful investigation on this point. Before proceeding with such an inves- tigation, it is possible to devise a general outline of narrative elements that may be amenable to histori- cal analysis.

In the manuscript version of the story recorded by Grinnell, humanity journeyed forth from a dark underground world and encountered three major obstacles: a large body of water, a thick forest, and a huge ravine. After passing beyond the ravine, the people entered the Blue Mountains, where two gam- blers argued and then fought. This brought about divisive conflict among all the people, and some left, eventually separating into eight wandering groups that spoke diverse languages. The ninth group ulti-

mately left the Blue Mountains and immigrated to the Missouri River, where they split into the Arikaras, Mandans, and Pawnees, with each group receiving corn of different sizes. Members of one of the other eight groups resided in the Black Hills of South Dakota at the time when horses first appeared, and these people subsequently moved to join the Arikara on the Missouri River. In the published version, Grin- nell excluded mention of the Black Hills group. It is difficult to match these events to a model in which "Blue Mountains" refers to Blue Butte in North Dakota, so for this reason, it is appropriate to inves- tigate Grinnell's original presumption that the Blue Mountains are the Rocky Mountains.

This origin story features minimal explicit chronological indicators, and it is clear that it was presented in its native setting as a historical narra- tive rather than as a fictional story. It is therefore appropriate to investigate whether the story can be reasonably reconciled to historical settings of the recent or ancient past. In other words, can other evi- dence identify probable or possible ancestors of the Arikaras, Pawnees, or Mandans somewhere in the Rocky Mountains?

It can be suggested, as a beginning point, that the story preserves at least a broad historical framework in its references to a dark origin point and the Mis- souri River. The place of origin, described as an underworld, can be interpreted as an ancient mem- ory of the Arctic Circle and the Beringian homeland of the ancestors of people who settled in NorthAmer- ica during the final millennia of the Pleistocene. A variety of origin stories and creation stories world- wide have potential for relating to the Pleistocene worldscape (Echo-Hawk 1994). Emergence origin stories are common throughout North America, with most referring to underground worlds, while others simply portray a dark region of earth from which humans emerge to populate the earth. One survey of 120 such stories found that over half were linked to migration stories (Wheeler-Voegelin and Moore 1957:66-67). Scholars generally see purely metaphorical content in the emergence theme (see, for example, Vecsey 1988:34-63), and this also is the preferred interpretation of emergence origin sto- ries adopted by some Native American religious lead- ers (Ferguson and Hart 1990:21).

I view the theme of emergence from an under- world/region of darkness, as a distorted remem- brance of Beringia and the Arctic Circle. My

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 277

principle of memorability explains why this infor- mation was not handed down into the present as an exact, unchanged description of Beringia. The his- torical specificity of a land of lingering darkness as a place of origin became preserved as an under- world-a concept that lends itself more easily to cul- turally meaningful metaphorical usage. In terms of community identity, the preservation of a strict his- torical account of Beringia would serve a less use- ful purpose than the more memorable story of an underworld that opens conveniently upon a particu- lar homeland. New World origin stories preserve memories of ancient Beringia because they are man- ufactured from preexisting historical narratives rather than from imaginative confabulation. Origin stories that associate a place of darkness with the "first" humans are exceedingly rare in the Old World-a situation that I explain in historical terms.

These stories need careful assessment against a variety of historical backdrops in order to associate them with specific time periods, but the theme of darkness occurs in association with a variety of other elements that are congruent with Pleistocene world- scapes, so such connections must be studied. In light of this arguable connection to the Pleistocene, the reference in the Arikara story to a great body of water as an obstacle could represent a retained memory of the giant proglacial lakes that formed along the edges of the retreating ice sheets after about 15,000 B.P. Pawnee stories associating a giant buffalo with cat- astrophic flood events could point to circumstances involving the formation of Glacial Lake Missoula and Glacial Lake Columbia, and several stories told among the Flathead and Shoshone also may concern Lake Missoula and its remnant, Flathead Lake in Montana (Echo-Hawk 1994:186-189).

The Arikara story includes an obstacle described variously as a "deep chasm" or "deep crevice" or "deep ravine." Given the associations of underworld = Beringia, water obstacle = Great Basin proglacial lakes, and Blue Mountains=Rocky Mountains, then the impassable chasm must refer to the Grand Canyon. Framing the story in terms of a migration may obscure the real significance of "obstacles" as reflecting the geography of the world known to ances- tors of the Arikara-a world bounded by the mem- ory of proglacial lakes to the north, great forests to the northwest, the Grand Canyon to the south, and the Rocky Mountains to the east. Only in deep time do the obstacles assume a viable historical presence.

At the end of the Arikara story, the Missouri River reference must relate to more recent historical settings involving Caddoans of the Central Plains tradition and Coalescent tradition along the Missouri River.

The compatibility of this general chronology with known history may distract attention from the degree to which the episodes of the story actually represent discrete events that have become artificially associ- ated over time. It also remains possible that any or all elements of the story are better explained as social discourse that documents cultural behavior, fictional color, or speculative musings rather than history. Thus, the potential historical significance of the episodes as well as their relationships to each other offer legitimate topics of inquiry.

Although the final events of Grinnell's version of the Arikara origin story occur in recent centuries, it is unlikely that the episode involving residence in the Blue Mountains occurred in any postcolumbian time period. No Caddoan or Mandan occupation sites have been identified for any region of the Rocky Mountains, and no historical documents preserve a record of Arikara, Pawnee, or Mandan groups resid- ing in the Rocky Mountains. A non-Caddoan group residing in the mountains at ca. 1700 or earlier could have joined the Arikara on the Missouri River, but, aside from the oral tradition, no evidence of such an event has yet been identified. Whether these moun- tain people were Caddoan speakers or some other group, the implied antiquity of many of the described events suggests that it would not be reasonable to seek confirmation of the story by looking for recent Caddoan earthlodge cities in the mountains. Sug- gested references in the story to the Pleistocene world give notice that some narrative elements could reflect a potential antiquity dating far back in time.

Statements that can be interpreted as chronolog- ical markers in the story are sparse but significant, providing clues to the antiquity of the residency in the Blue Mountains. The occupation of the Blue Mountains began and ended during a time before horses first appeared, so this portion of the story must be set in some period prior to ca. 1650-1700. The adoption of ideology associated with corn cultiva- tion is also said to have occurred after these moun- tain immigrants settled in the Central Plains, with the suggestion that the people were aware of corn when they resided in their high altitude homeland in the Blue Mountains. In addition, the tradition asserts that the initial settlers of the Blue Mountains were

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
278 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

believed to have given rise to descendant groups speaking many mutually unintelligible languages- in fact, the narrative purports to describe the earliest human occupation of the region.

The following sequence of events exists in the oral tradition: 1) the origins of ancient ancestors of many Indian tribes are associated with settings that could relate to the terminal Pleistocene, including a memory of long Beringian arctic nights and a memory of proglacial inland seas; 2) at an unknown date, a population residing west of the Rocky Moun- tains expanded into or migrated into some region of the Rockies; 3) these settlers of the Rocky Moun- tains contributed to the composition of many dif- ferent later groups; 4) at some point in time, possibly prior to or during the general period of the formulation of Mother Corn ideology, this Rocky Mountain population helped to form Plains popu- lations, particularly the Arikara, Pawnees, and Man- dans; and 5) a group residing in the Black Hills joined the Arikara communities at the time when horses first appear. As mentioned earlier, it may be inadvisable to seek an exact description of an unfolding chronology of episodes in the oral tradi- tion, but this general outline seems reasonable as a starting point for research.

As required by the standards of analysis set forth earlier, confirmation for the scenario set forth in the Arikara story, or elements of it, must be sought in the archaeological record or other historical evidence. The suggested correspondence of certain narrative elements to Pleistocene settings in the Great Basin would gain substance if evidence were available showing that Paleo Indians in the region became ancestral to populations in the Rocky Mountains- populations who subsequently gave rise to the Arikara in some arguable manner. As detailed below, the basic model of the Arikara story is strikingly compatible with current explanations of the archae- ological record in Colorado.

The Mountain Tradition

In formulating a new taxonomic construct for the central and southern Rocky Mountains, Kevin Black (1991) argues that about 9500 B.P., late Paleo Indian (Western Pluvial Lakes tradition) populations began to immigrate from the Great Basin into the Rocky Mountains as a result of environmental conditions, taking up residence throughout a broad region from Montana to Colorado. The Mountain tradition is seen

by Black as enjoying an unbroken, continuous pres- ence in the Middle Rockies of Montana and Wyoming for about 5,000 years, but in the Southern Rockies, cultural continuity extends for a much longer period-up to about A.D. 1300, when the expansion of Numic speakers apparently replaces Mountain tra- dition cultures in the archaeological record. A survey of the distribution in time and space of Paleo Indian projectile points in southwestern Colorado led Bon- nie Pitblado (1998) to conclude that the extant pat- tern, although based on limited data, provides support for Black's contention that the Mountain tradition has strong cultural roots in the West.

Mountain tradition populations maintained a con- tinuing connection with the West, but they are not viewed by Black as culturally uniform throughout the millennia of their presence in the archaeological record. Instead, he endorses the view set forth by J. D. Keyser in a 1985 publication, that cultural diver- sity among "local groups" becomes especially appar- ent in the Southern Rockies after 5000 B.P. Black acknowledges (1991:4) that the material culture of Mountain tradition sites bears great similarity to sites associated with Numic speakers, but he nevertheless defers to the Numic Expansion model of replacement of resident Mountain tradition populations by Numic speaking groups ancestral to the Ute and Shoshone.

Black offers no model for the fate of the Moun- tain tradition peoples, nor does he speculate as to the linguistic identity of these populations. Who did they become? If they left their Rocky Mountain home- land, where did they go? What are the connections between the Mountain tradition, Western Pluvial Lakes tradition, DesertArchaic tradition in the Great Basin, Numic sites, Oshara tradition, Puebloan com- munities, and other neighboring groups throughout the Rockies and Great Plains? The Arikara oral tra- dition suggests that we look for archaeological evi- dence of Mountain tradition people entering the Central Plains to join the Caddoan-speaking ances- tors of the Arikara.

Black lists a number of complexes that fall under the Mountain tradition umbrella, and he lists other groups that might belong, but their membership is in some doubt due to evidence of having "relation- ships with low-land-based cultures"-in other words, archaeologists have caught them consorting with Plains flatlanders. One of these groups is "Woodland/Hogback" in Colorado, considered sus- pect because of ceramic technology adopted from

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 279

Plains Woodland tradition neighbors and from the succeeding Central Plains tradition. Along the Col- orado Front Range north of Pike's Peak, an evolv- ing mosaic of groups is represented by sites extending throughout eastern and central Colorado, identified under various taxonomic schemes as the Hogback phase or complex. The presence in Hog- back sites of pottery that draws upon ceramic tradi- tions of the eastern Plains for more than 600 years suggests a long-term relationship with neighbors in the Central Plains.

Definition of the Hogback phase has proven a challenge for archaeologists, as well as reconciling this phase to the morass of taxonomic schemes applied to the Colorado region. As summarized by Peter Gleichman, Carol Gleichman, and Sandra Karhu (Gleichman et al. 1995:122-123), the Hog- back phase spans a time from about A.D. 500 to 1200. This phase includes various sites along the Colorado Front Range north of Pike's Peak as well as related sites in the mountains. A key characteris- tic of Hogback sites is that they feature pottery influ- enced by technological developments in the east rather than the west or south. For this reason, Black leaves open the possibility that this may represent a Plains group rather than an indigenous Mountain tra- dition population. Although the Gleichmans and Karhu provide no assessment as to the origins of the Hogback phase, they argue that it represents an indigenous population that simply imitated the ceramic technology of the Plains Woodland and Cen- tral Plains tradition.

Grinnell's version of the Arikara story focuses on a young boy who at one point had been given a bun- dle by the Creator, and at different points, the boy opens the bundle and finds objects that help to resolve dilemmas of various kinds. After entering the Plains, the boy finds corn within the bundle and distributes it to the Arikaras, Pawnees, and Mandans. This could be read to suggest that some interaction with corn cul- tivators occurred while the people resided in the Blue Mountains, but corn only assumed prominence in the lifeways of these people after migration into the plains.

Hogback sites do not feature corn horticulture, but the archaeological record does indicate that Hog- back folk could have been familiar with corn from its presence among neighboring groups. During the 1940s Clarence Hurst excavated the Cottonwood Cave site in western Colorado and found a bundle containing corn in a pit capped by a stone slab, and

later radiocarbon tests showed this bundle to date back to about 270 B.C. (Stiger and Larson 1992). It is likely that Mountain tradition people knew of corn. It was first brought from Mesoamerica into thenAmer- ican Southwest during the Late Archaic after 1500 B.C., and by A.D. 200 it had spread into eastern North America (Cordell and Smith 1996:210-211, 247). It has a limited presence in the foothills of east- ern Colorado south of Colorado Springs prior to A.D. 200 (Zier and Kalasz 1991), and the earliest known presence of corn in the heartland of the Central Plains occurs about A.D. 250, but it is not until after about A.D. 1000 that systematic cultivation of corn and otherplants became widespread in the Central Plains (Adair 1988:114-115), creating fertile ground for Mother Corn ideology.

If, as the Arikara oral tradition implies, Mother Corn ideology and some form of migration were unfolding during the same era, then these dates sug- gest that an appropriate time to look for an emigra- tion from the mountains to the plains would be sometime after AD 900. It is therefore proper to search the archaeological record of the Rocky Moun- tain region for candidate circumstances that could arguably pertain to the Arikara story. In the Arikara oral tradition, the people looked out from the Blue Mountains and "saw a beautiful country" that seemed desirable to dwell in, but they decided not to leave their mountain homeland until a later time (Grinnell 1916; Grinnell ca. 1890:File #156, "Ree Cosmol- ogy"). The people of the Blue Mountains had an interest in the nearby plains, and this could reflect a memory of the Hogback lifestyle, which involved periods of residence in both the mountains and the High Plains.

The Arikara story urges us to look for long-term historical processes that contributed to the formation in the Blue Mountains of multiple groups speaking different languages. On the basis of geography alone, Puebloan, Caddoan, Numic, andAthapaskan groups all present themselves as candidates for having Col- orado Mountain tradition ancestry, and modem com- munities as diverse as the Tewa Pueblos, Navajos, Arikaras, Pawnees, Mandans, Hidatsas, Crows, Hopis, Utes, Shoshonis, Comanches, and others could well be descended from various Mountain tra- dition groups in Colorado. Pueblo origin stories gen- erally refer to ancestral movements from the north. One Tewa Pueblo oral tradition mentions the Col- orado Sand Dunes region as an origin point (Jean-

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
280 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

con 1931 manuscript: no page number), and Zuni ori- gin accounts associate one group of ancestors with the Rocky Mountains (Ferguson and Hart 1990:21-23).

Standing alone, the archaeological record is often hard-pressed to explain the fate of specific groups, but for the Mountain tradition some clues are avail- able. Based upon the similarity of material culture to Numic sites, as well as indications of ongoing ties of Mountain tradition people to neighbors in the Great Basin, it seems reasonable to presume that, for the most part, Mountain tradition groups were absorbed into the Utes, Shoshones, and other Numic speaking tribes. One Southern Ute oral tradition places Pike's Peak at the center of the world created for the Mouache Band (Arbogast et al. 1996, Pt 2:Tape 2, p. 4; Tape 3, p. 12). This tradition supports the idea that long-term populations of the region could have contributed in some manner to the Utes.

A Shoshone oral tradition also might have some bearing on the ancient history of the Colorado region. According to Robert Lowie (1909:233), although the Shoshones as a group do not have any story about ancestral migrations, the Wind River Shoshones "are said to have had a tradition that they originally came from the south." It is impossible to determine with any degree of certainty how this vague tradition may relate to Colorado's Front Range. It could relate to a very shallow time frame-perhaps involving the seventeenth- or eighteenth-century formation of the Comanches as a separate group from the Shoshones. Oral traditions relate that this separation occurred in Colorado (Wallace and Hoebel 1952:9). The Shoshone tradition of southern roots could, however, refer to a more ancient time frame related to Moun- tain tradition populations in the southern Rockies who could have moved northward into Wyoming. More definite and detailed oral traditional informa- tion is needed to clarify this possibility.

It would be improper to embrace a standard of analysis for the Arikara oral tradition that would look for an exact correspondence between story elements and the archaeological record. In other words, the story cannot be rejected because it neglects to pro- vide any explicit discussion on settlement patterns, split cobble technology, microtools, projectile point styles, architecture, and rock art in the Rocky Moun- tains. The principle of memorability explains why tribal historians who might have been tempted to include detailed consideration of these topics in their

accounts of the past would have ultimately failed to inspire their successors to retain such information.

The Apishapa Phase

Connections between the Arikara oral traditions and the Rocky Mountain archaeological record are made possible through theApishapa phase and possibly the Sopris phase. The Sopris phase (also known as Upper Purgatoire complex) flourished during AD 1000-1225, in the vicinity of Trinidad, Colorado (Baugh 1994:273-274; Crum 1996:70-71;). The people of the era of the Sopris phase were immedi- ately preceded in the archaeological record by a pop- ulation that built pithouses similar to structures found in eastern Colorado, supporting the idea that groups throughout the region gave rise to Sopris. Study of human remains indicates that the Sopris population could have also included Athapaskan speakers (Baugh 1994:275), but this interpretation is based upon dental characteristics that are not unique to Athapaskans. This complex continued to ca. A.D. 1225, relying upon corn cultivation as well as hunt- ing (Baugh 1994:273-274).

No clear evidence exists that shows Hogback peo- ple moving into the Caddoan Plains, but they prob- ably contributed to the Apishapa phase in the Plains south of Pike's Peak, which overlaps the end of Hog- back and shares similar ceramic technology. The use of dry-laid rock masonry as an architectural element occurs in both Hogback and Apishapa sites (Kalasz et al. 1995:337), as well as in Sopris sites dating between A.D. 1150 and 1225 (Baugh 1994:273).

The Apishapa presence in the archaeological record is recognized from about A.D. 1 100 to 1350, and the culture is viewed as having originated from the previous Plains Woodland residents of south- eastern Colorado (Gunnerson 1989:125-127; Lintz 1986:26-27). TheApishapa people focused on hunt- ing as their primary subsistence activity (Baugh 1994:278), but indications of corn cultivation on a limited scale have been found (Zier and Kalasz 1990). The most recent synthesis of the archaeolog- ical record in southeastern Colorado subsumes the Hogback phase and Plains Woodland tradition into a Developmental period geographically distributed from northeastern Colorado to northeastern New Mexico (Zier and Kalasz 1999). Developmental period populations of this region gave rise to the Apishapa phase and Sopris phase during the suc- ceeding Diversification period.

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 281

It is reasonable to presume that some Hog- back/Developmental groups, at least, participated in Apishapa. Hogback ceramic technology and archi- tecture are echoed in Apishapa sites, and the termi- nation of Hogback by about A.D. 1200 coincides with the cultural changes that gave rise to Apishapa at about A.D. 1100. Hogback families and groups who favored a hunting lifestyle could have chosen to move among the Apishapa, while those who saw benefit in farming would have had the choice of tak- ing up residence among the Sopris people. As already noted, others could have ultimately led to groups that became the Ute, Shoshone, Comanche, or some other people. Both Hogback and Apishapa are gen- erally viewed as long-term Colorado residents who were influenced by groups farther out in the Plains.

Christopher Lintz (1986:3) includes theApishapa phase and neighboring Antelope Creek phase under a single taxonomic umbrella, and Timothy Baugh (1994:274) adds the Buried City complex as a third group. The Antelope Creek phase was located to the east of Apishapa, distributed throughout the Texas Panhandle, while the Buried City complex lay even further east, cutting across the Texas and Oklahoma Panhandles into southwestern Kansas. Beyond the Buried City complex lay the homelands of other related peoples across Oklahoma into southern Mis- souri, Arkansas, and Louisiana (Perttula 1992:6-9; Vehik 1994).

Waldo Wedel traced the Wichita in the Oklahoma archaeological record back to "Washita-focus mate- rials from...about A.D. 1070-1612"-a time frame that makes the Washita people, in part, contempo- raries of Apishapa, Antelope Creek, and Buried City (Wedel 1983:232-233). He also noted similarities between Washita and earlier Plains Woodland mate- rial culture, implying that Washita is derived from Plains Woodland people in the Southern Plains (Wedel 1983:226-227). Later refinement of South- ern Plains archaeological taxonomy has continued to retain the basic idea that Plains Woodland groups gave rise to populations that ultimately came to form the Wichita (Vehik 1994; Cordell and Smith 1996:250; Drass 1999). The proposal that the Wichita and related groups represent long-term residents of the Southern Plains has broad acceptance among archaeologists.

Caddoan Roots in the Southern Plains

After A.D. 1000 the Southern Plains and Central

Plains were dominated by groups that contributed directly to the formation of Caddoan-speaking tribes. In Colorado, the Apishapa phase is viewed by some archaeologists as the westernmost expression of Cad- doan culture. James Gunnerson (1989:122-124), for example, noted close cultural similarities between Apishapa and neighboring Antelope Creek, Custer, Washita River, and Upper Republican populations, and he concludes that "one could think of the Clas- sic Apishapa phase as though it were a less seden- tary, less horticultural version of Upper Republican." In assessing Gunnerson, as well as earlier research by Christopher Lintz, Timothy Baugh (1994:278) infers that the Apishapa phase may be closely related to the Caddoan Upper Republican phase of the Cen- tral Plains tradition, ancestral to the Pawnees, while the Antelope Creek people and their neighbors in Oklahoma served as sources for the Wichitas, via the Wheeler phase (Drass and Baugh 1997). Steve Cas- sells (1997:222) agrees that it is reasonable to pre- sume that residents of the Southern Plains, Central Plains, and southeastern Colorado "shared a com- mon background and language, or were involved in social interaction."

Complex interactions among diverse people throughout this region ultimately gave rise to the later Wichitas, Pawnees, andArikaras. Oral traditions from these three Caddoan groups reflect these cir- cumstances and document ancient connections to the Southern Plains. One Skidi Pawnee tradition reported by Rush Roberts associated the Kitkahahki Pawnee with the general region of the Antelope Creek and Buried City sites at a period dating "before the Westward migration of the Indian comprising the Sioux an [sic] linguistic family," when the Kitka- hahki dwelt in several communities on both sides of the Red River "in what is now Oklahoma & Texas" (Roberts ca. 1950s: #20). This mention of a Siouan "Westward migration" probably refers to the move- ments of such groups as the Oto, Missouria, Omaha, Ponca, Kaw, and other related tribes into the eastern periphery of the Plains. This oral tradition also places the ancestors of the Chaui to the east on the Ohio River, with the ancestors of the Pitahawirata Pawnees in "eastern Oklahoma, Southern Missouri, and West- ern Arkansas."

In 1890, George Bird Grinnell recorded an oral tradition from an Arikara named Fighting Bear that referred to a location in Oklahoma "down by the Cross Timbers about 2 days south of the Big Tim-

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
282 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

ber" as the place where the Arikaras separated from the Pawnees (Grinnell ca. 1890:Journals # 315). Another Arikara tradition asserted that northern Kansas was a destination for the firstArikara settlers in the Central Plains: "Mother-Corn led them away on through the country to what is known as the Republican River, in Kansas, where there is only one mountain"; and at this location, the Arikara settled and "had their ceremonies" (Dorsey and Murie 1904a: 16). TheArikaras are reported as early as 1866 as having an association with Pawnees in Oklahoma. A document prepared by the U. S. Indian agent for the Pawnees in 1866 also mentions the Red River:

The Pawnees also claim that the "Uricarees" of the Northern Missouri River are another branch of the original Pawnee Nation. They speak the same language and have the same manners, cus- toms, habits, and legends as to their origin and claim themselves as a part of the old Pawnee Nation, which at one time was master of the plains from the Red River on the North to the Gulf on the South (Wheeler 1866).

Grinnell also set forth a variety of convoluted movements for the Arikara across the Central Plains and Southern Plains, and he portrayed these events as sequential, but they are actually conflated from diverse historical settings. A sojourn that is probably a memory of the residency of the Skidi near the Wichita Mountains during the 1770s, for example, is surgically joined to later events of the 1830s, when a large Arikara group dwelt in Pawnee country for three years (Grinnell 1961:231-232 [1889]). Confusion also derives from the probability that not only did Pawnee and Arikara ancestors emerge from South- ern Plains Caddoans in ancient times, but Pawnee and Arikara groups also resided in Oklahoma during a later period, with both historical settings becoming mingled in subsequent Caddoan oral traditions.

South Band Pawnee connections to Oklahoma are indicated in traditions gathered during 1870s by John B. Dunbar(1880:25 1): 'The traditions of three of the bands, the Xau-i, Kit-ke-hak-i and Pit-a-hau-e-rat, coincide in stating that the Pawnees migrated to the Platte River region from the south, and secured pos- session of it by conquest." This migration occurred at a time "so remote that they have failed to retain any of its details, except in a very confused form." The Pawnees were accompanied in this northward move- ment by the Wichitas, who ultimately "left them long ago and wandered away to the south...."

Luther North became acquainted with Pawnee oral traditions during the 1860s and 1870s. He informed George Bird Grinnell that the South Band Pawnees and the Wichitas "were one and the same tribe, they separated many years ago, the Pawnees coming north to hunt for Buffalo" (Grinnell ca. 1890:File #19, North to Grinnell 6/26/1884 corre- spondence). While living in the south, apparently, the Pawnees hunted buffalo and raised corn, but "had very little of both." North's knowledge of Pawnee traditions may have been minimal, and he conflated a variety of events together, associating, for exam- ple, this first entry of Pawnee ancestors into the region with the acquisition of horses.

Grinnell also reported another tradition stating that the Pawnees moved northward from Missouri or Arkansas into the Central Plains, hunting buffalo, and the Wichitas "accompanied them part way on theirjourney, but turned aside when they had reached southern Kansas, and went south again" (Grinnell 1961:225-226 [1889]). According to another Pawnee tradition, the Pawnees and Wichitas separated dur- ing an eastward movement of Pawnee ancestors to the Mississippi River (Peters ca. 1960s; also see Grin- nell 1961:224-225 [1889]). Rush Roberts (ca. 1950s: #20) indicates associations of the Kitkahahki and Pitahawirata with Texas, Oklahoma, southern Mis- souri, and western Arkansas, suggesting that popu- lations scattered throughout this region gave rise to various South Band Pawnee groups.

Dunbar (1880:251) also mentioned a Wichita tra- dition locating the ancestral home of the Wichita- Pawnee people upon the Red River below the mouth of the Washita. The Wichitas were "dissatisfied with the migration, or its results," and "they attempted to return to their old home." This information could represent a conflation of different events involving the termination of the Great Bend aspect, and the later presence of the Skidi on the Red River during the 1770s.

These traditions may concern a variety of events and periods-but some probably describe the found- ing of Quivira (Great Bend aspect) in Kansas. Susan Vehik (1994:261) suggests that the Antelope Creek phase terminated by "joining the Washita River phase and/or moving northeastward to join members of the Great Bend aspect." Pawnee and Wichita oral traditions support this model, and if we presume close ties between Apishapa and Antelope Creek, then Vehik's suggestion also provides one route for a group or groups

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 283

ancestral to the Arikara to enter the Central Plains. The principle of memorability cautions us against a too lit- eral treatment of Caddoan oral traditions, but both the archaeological record and oral traditions point to the Oklahoma region for insights into ancient Plains Cad- doan history, suggesting that complicated movements of groups and other interactions occurred among res- idents of the Central and Southern Plains. Peering back into the period ofA.D. 1000-1400, we should not look too hard for "Pawnees," "Arikaras," or "Wichitas"; instead, we should seek to understand the elaborate dynamics of regional population interactions to grasp the formation of later tribes.

The ancestors of the modem Wichita were located in south-central Kansas when the Coronado expedi- tion encountered them in 1541, living in a number of grasslodge towns scattered along the streams of the region and cultivating corn (M. Wedel 1979:183; W. Wedel 1979:274). One Wichita tradition (Mead 1904:173) associated the Arkansas River in Kansas with ancestors who settled there, "cultivating gardens and hunting for subsistence, using implements of stone or bone...." Coronado wrote to the King of Spain that the people of "Quivira" resided in "not more than twenty-five towns, with straw houses" (Hammond and Rey 1977:188), and based on the existence of numerous town sites in the region, Waldo Wedel wrote that he believed "Coronado could eas- ily have...counted up to 25 villages south of Smoky Hill River within 100 miles of the present town of Great Bend" (Wedel 1942:12-13).

While visiting Quivira, Coronado met a large del- egation of Indians from a "province" in the north known as "Harahey"; Martha Blaine (1982:113-115) suggested that these were either Pawnees orArikaras, and reported several Pawnee traditions describing encounters with the Spanish. Based on her survey of Pawnee oral traditions, she suggests that grasslodges were widely used in the Caddoan Central Plains, particularly during the sixteenth and early seven- teenth centuries, and must relate to the Quivira and Harahey mentioned in the annals of the Coronado expedition and later Onate expedition (Blaine 1979). This coincides with a Wichita tradition published by Elizabeth A. H. John (1983), which describes the homeland of the Wichita as "on the river Platt" an indication that Caddoan populations with similar lifeways extended throughout the Central Plains and gave rise to the later South Band Pawnees, Wichi- tas, and the Skidi Pawnees.

During the seventeenth century, the Wichitas and other associated Caddoan groups moved back south- ward from Quivira into Oklahoma, as one tradition recalled: "Many hundreds of years ago, when our people came from the north on their way to this point where we are living..." (Dorsey 1904:3 10). Mildred Wedel (1982:124) suggested that conflict with the Skidi Pawnee, Missouria, and Osage during the late 1600s contributed to the decision of the Wichitas to leave Quivira: "It was the Skiri who in the 1670s had first captured La Salle's Pana slave...." In a Wichita tradition, Elizabeth A. H. John (1983) also published a tradition recorded during the early nineteenth cen- tury describing how the Wichita left their home in the Central Plains due to warfare with the Osages. By 1719, the Wichita had returned to their ancient home- land in northeastern Oklahoma on the Arkansas River.

Southwestern Ancestors

Accounts of ancient Pawnee history generally refer to a time when ancestors of the tribe dwelt in the south- west. These stories may be explained as relating to the Plains Woodland and laterApishapa groups, and pos- sibly the Sopris phase. A Skidi tradition reported that the ancestors of the Pawnees resided somewhere in the southwest located "away beyond the Rio Grande" (Grinnell 1961:225 [1889]); aChaui version described this land as "far off in the southwest...beyond two ranges of mountains" (Grinnell 1961:224 [1889]). An 1866 report of this tradition made by the agent for the Pawnees, mentions New Mexico as a region associ- ated with Pawnee ancestors:

The old men of the tribe inform me that the Pawnees formerly lived in the Southern portion of what is now a part of the United States. That is in a portion of New Mexico. They have no distinct idea of their numbers at that time, only that the Pawnees were like the Buffalo, Elk, & Deer on the Plains, almost innumerable. They claim that at that time they owned and con- trolled all the land between the Rio Grand and Platte Rivers (Wheeler 1866).

The Pawnees associate their ancient ancestors in the southwest with houses that incorporated the use of stone in some fashion. George Bird Grinnell (1961:225 [1889]) wrote that the ancient ancestors of the Pawnees dwelt in "stone houses," but he pro- vided no description of the structures. A later version of this tradition was set down in more complete form by a Pawnee named Bert Peters sometime before the mid-1970s. This account states that the ancestors of

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
284 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

the Pawnees originated from an unknown location in the Southwest, where they "lived in houses of mud and stone" (Peters ca. 1960s). Archaeologists have speculated that these oral traditions relate to archi- tecture of theAntelope Creek phase (Lintz 1979:162, 178), to the Apishapa phase (Gunnerson 1989:128), and possibly to the Great Bend phase (W. Wedel 1979:277). The Sopris phase also offers itself as a plausible setting for these stories.

A Chaui Pawnee tradition on the history of tribal sacred bundles attributed the time of their origin to when the ancestors of the Pawnees lived in the south- west: "All the sacred bundles are from the far-off coun- try in the southwest, from which we came long ago. They were handed down to the people before they started on their journey" (Grinnell 1961:352-353 [1889]; also see Fletcher and Murie 1996:156 [1904]). A European visitor in 1823 received "a gift of very valu- able wampum" from a Skidi priest that seemed to reflect the southern ancestry of the Pawnee: "Put together of the seeds of a variety of palm and the seeds of a leguminous plant (Glycine?) from tropical regions, it was purported by the priest that it had been bequeathed from father to son. This was evidently of southern origin and was valuable to me as proof of the migration of these people" (Wilhelm 1973:394). It seems implausible that a Skidi priest would give away a tribal heirloom of great antiquity-and thus, of great cultural meaning-to a casual European visitor. This object was probably of more recent vintage, dating back to the eighteenth-century residence of the Skidi on the Red River, but it could have nevertheless served as a reminder of oral traditions about Caddoan movements and ancestral ties extending back many centuries.

Caddoan oral traditions and the archaeological record reflect a complex past. In the Colorado region, Caddoan roots reach back to a generalized Plains Woodland population ranging from northeastern New Mexico up the Front Range of the Rockies into northeastern Colorado. This population served as the primary source of the Apishapa phase and Sopris phase. The Upper Republican phase in the Central Plains displays cultural resemblance to the Apishapa phase probably because both have roots in the Plains Woodland residents of Colorado and New Mexico, and some interchange of population may have occurred. Archaeologists broadly accept the view that Plains Woodland groups of the Colorado region gave rise to Apishapa, but the fate of Apishapa-as well as the Sopris phase-is less clear.

The Arikara tradition of the Blue Mountain resi- dency and Pawnee stories of a homeland located vaguely in the Southwest consistently refer to some period before horses were common in CaddoanAmer- ica, and the postcolumbian archaeological record offers little help in explaining the stories. For this rea- son, many Caddoan oral traditions must be handed down from earlier periods, including from Apishapa groups, possibly Sopris, and the preceding Plains Woodland tradition in Colorado/New Mexico. Some Apishapa and Sopris groups could have moved directly into the Central Plains, while others entered the Southern Plains and took a different route to ulti- mately contribute to the later Pawnees, Arikaras, Wichitas, and possibly other groups. Quivira in Kansas may have included descendants of the Apishapa, Sopris, Antelope Creek, Buried City, and other groups in Oklahoma and Arkansas. Central Plains tradition groups like the Smoky Hill phase and Upper Repub- lican phase also may have contributed to Quivira (W. Wedel 1979:274-275), but they flowed more directly to the neighboring "provinces" of Harahey and Gua in northern Kansas and Nebraska. Waldo Wedel (1979:277) suggested that Upper Republican may have moved south to help form Antelope Creek, and then moving northward to establish Quivira, and Cad- doan oral traditions support the existence of complex interactions among these and other groups. Neigh- boring populations distributed throughout the South- ern Plains and Central Plains served as ancestors of the Caddo, Wichita, Kitsai, Pawnee, and Arikara.

In addition to ancient roots to the west, southwest, and south of the Central Plains, long-term residents of the Nebraska-Kansas region also served as ances- tors of the Pawnees. Evidence not discussed in this paper supports the view that Siouan and Caddoan tribes such as the Mandans and Skidi Pawnees-and probably the South Band Pawnees, Arikaras, Hidat- sas, and Crows-have links to the Plains Woodland residents of the Central Plains. Also not considered here are ties to groups on the Mississippi River, and movements of populations into South Dakota and then back into the Central Plains.

From ancient times into the present, a complicated social history accounts for the cultural character of the North American Great Plains world. As many scholars have observed, it is unrealistic to look for discrete social units moving unchanged through time from one location to the next, or from one archaeo- logical taxonomic unit to another. More elaborate

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
Echo-Hawk] INTEGRATING ORAL TRADITIONS AND THE ARCHAEOLOGICAL RECORD 285

dynamics are at work, and this is reflected in the com- plexity of both oral records and the archaeological record. Careful study of Caddoan oral traditions has great potential for yielding many insights into the same past that accounts for the archaeological record of the ancient Colorado plains, New Mexico, Rocky Mountains, Central Plains, and other places in time. Upon these vast regions, configurations of Caddoan America move through time, and we can trace the journey as it unfolds into the present.

A Spoken Future

Archaeologists frequently say that the sites they exca- vate and artifacts that they recover can "speak" to us across the centuries, and physical anthropologists often think of collections of human skeletal remains as "libraries." In oral traditions, we can hear echoes of the actual voices of the people who made those artifacts and who were the original owners of the skeletons. As researchers explore the contribution to history of oral traditions in Africa, Australia, and the Americas, it has become increasingly difficult to ignore arguments that historical information has been preserved through verbal means for great lengths of time.

As a concept, "prehistory" interferes with recog- nition of the validity of the study of oral traditions because it presumes an absence of applicable records. Its ubiquitous presence as a term in academic and popular discourse also reflects the degree to which twentieth-century American archaeology has dis- placed Native American oral traditions as the source of valid knowledge about ancient human circum- stances. It may be technically correct to apply the term to periods in time for which no writings exist, but its usage as a taxonomic device emphasizes writ- ten words, while presuming that spoken words have comparatively little value.

A bibliocentric research agenda imposes needless limits upon legitimate scholarship about the ancient past. My proposed shift in terminology from "pre- history" to "ancient history" would place oral tradi- tions generated by firsthand observers on an analytical par with written records generated by first- hand observers, and both categories of records would receive acknowledgment as legitimate documents for scholarly study. Written words and spoken words need not compete for authority in academia, nor should the archaeological record be viewed as the antithesis of oral records. Peaceful coexistence and

mutual interdependence offer more useful paradigms for these "ways of knowing."

The Quest for Connections

In formulating findings of cultural affiliation under NAGPRA, the tracing of relationships among popu- lations into the ancient past is a matter of complex scholarship. Oral records and the archaeological record interact in intricate ways to both reveal and obscure connections between ancient and modem communi- ties. For this reason, NAGPRA cultural affiliations are most convincing when sustained by careful analysis that best explains all relevant bodies of evidence. To ignore any applicable realm of evidence would not serve either the law or academic scholarship.

As with all scholarship, NAGPRA cultural affil- iations that emerge from the integration of oral tra- ditions and archaeology will be subject to various forms of peer review and constructive criticism. The refinement of models of human history is a perpet- ual condition of academic study, and analytical delib- erations typically move at a highly situational pace, but the law calls for timely action and timely deci- sions. The NAGPRA requirement for setting forth findings of cultural affiliation really aims at expe- diting the involvement of Indian tribes and lineal descendants as decisionmakers.

The law has already generated much new research and dialogue among scholars, and it has considerably enhanced the presence of Native American communi- ties in this new discourse, but dialogue needs to be viewed as an ongoing process, and findings of cultural affiliation need to respond to new information and new insights. NAGPRA cultural affiliations should thus be viewed as flexible constructions designed to identify proper parties of interest for consultation and repatri- ation purposes rather than as inflexible representations of the past. This situation, in the short term, can lead to much inconsistency in institutional findings of cul- tural affiliation, but tribes can help by assembling thor- ough documentation that sets forth reliable evidence for museums and federal agencies that may otherwise be left to blaze their own idiosyncratic trails.

Research aimed at integrating oral traditions and archaeology is becoming common in the wake of NAGPRA. Since the spirit of NAGPRA expects aca- demic institutions and Indian tribes to engage in dia- logue about the cultural identities of persons who lived and died long ago, it would be ideal for tribes and museums to work together to investigate how

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions

http://www.jstor.org/page/info/about/policies/terms.jsp
286 AMERICAN ANTIQUITY [Vol. 65, No. 2, 2000

oral traditions and archaeological evidence might be reconciled. Ultimately, however, oral traditions deserve serious attention from scholars because aca- demic constructions of ancient human history can benefit substantially from the study of verbal records created and handed down from firsthand observers.

For researchers interested in oral traditions, anthro- pologists have recorded and published a vast body of materials that can be productively examined. In addi- tion, useful guidelines exist for archaeologists inter- ested in working with tribal experts on oral traditions (Anyon et al. 1997). Not every archaeologist will be inclined to track down, study, and apply information from oral sources, but those who choose to pursue such inquiry deserve support and encouragement, particularly because this research has interesting implications for scholarship on ancient America.

Reorganizing Prehistory into Ancient History

Revision of archaeologically based taxonomic sys- tems will be a natural consequence of the study of oral memoirs. To date, scholars of Caddoan oral lit- eratures have focused primarily on sorting materials into various literary genres, with significant atten- tion to formulating groupings according to histori- cal criteria (Dorsey 1904; Dorsey and Murie 1904b, 1906; Parks 1991). These efforts also have yielded important insights into Caddoan traditional views of the organization of human history, but this has had no measurable effect upon the development of archaeological taxa for Caddoan America.

George Dorsey (1904:20-22) divided the tradi- tional history of the Wichita andAffiliated Tribes into four eras and organized his collection of traditions according to these divisions. The first era includes "the origin of the world," human creation, and the establishment of social and spiritual relationships. During the second era, "the people scatter out over the earth" and enter a "period of change and unrest and of transformation," ending in a flood. The third era opened with the lives of the survivors of the del- uge, and brings human history up to the present. The fourth era concerns a future time (as foretold during the first era), "which the Wichita suppose to be rapidly approaching...."

In this view of history, the envisioned past goes back to when the first people dwelt in regions of dark- ness. The woman was given the knowledge that corn would sustain future generations, and the first man was given the idea to journey into the east to a place

of equal days and nights. This journey was followed by a time of increasing population, and the first man and first woman helped the people to develop life- ways and technology. The second era of Wichita tra- ditional history opened with population dispersions and proliferating lifestyles, and human communities acquired social identity, but these people were destroyed in a flood. The survivors entered the third era of history to repopulate the earth. During this period ("present time"), the people acquired grass- lodges made with cedar frames, and they at last received corn to cultivate.

Wichita, Arikara, and Pawnee origin stories all include the idea that some form of great journey serves as a key event in the unfolding story of human existence, and all of these journeys occur in the geo- graphic context of a place of darkness. Arikara and Caddo stories refer to an underground world; a Skidi story associates the journey with stars in the night sky; a Wichita account locates the first people in a region of darkness and the journey results in the cre- ation of day and night. This collection of ideas can suggest some definitive aspects of the earliest period of the human past recalled in Caddoan oral tradi- tions-aspects that can be connected to the archae- ological model of Beringia and the peopling of the New World. These stories also concern a more recent period in Caddoan history. References to Mother Corn, together with a focus on relationships between the sexes, must reflect cultural settings dating around AD 1000, when corn cultivation became prominent in Caddoan America, accompanied by social upheavals related to the economic empowerment of women. Caddoan origin stories represent a collec- tion of discrete memoirs preserving knowledge of human events in deep time conflated with events of more recent centuries, and Caddoan historians deserve recognition for having preserved a wealth of details about the ancient past.

Other scholars of New World oral literatures have given attention to the taxonomic implications of treat- ing oral traditions as sources of historical knowl- edge. Summarizing archaeological evidence for the presence of Southwestern Zea mays in the Northeast, Barbara Mann and Jerry Fields (1997:119) observe that corn begins to appear after A.D. 800 and is well- established by A.D. 1100. Iroquois oral traditions attribute the origin of corn to a female deity known as Otsitsa and her daughter. Mann and Fields men- tion that, according to a Cayuga oral tradition

This content downloaded on Tue, 5 Mar 2013 20:05:49 PM All use subject to JSTOR Terms and Conditions