Browse the book with Google Preview »
When I began working with the Kiowa in 1989, I was interested in traditional dance and music, particularly in the origins of the modern powwow. This led me into the world of Plains Indian military societies. I was introduced to the Kiowa Black Legs Society, on which I eventually wrote my master's thesis (Meadows 1991). As I began to comprehend the vast historical and symbolic importance which this society holds for the Kiowa people, I began to look at other Kiowa and, later, Comanche, Apache, and Cheyenne societies. Though they are qualitatively different, I found great general and temporal similarity in the structure, functions, and continuing symbolic importance of veterans within these communities. Even more importantly, however, I began to comprehend the extent to which traditional sociocultural forms were being associated with traditional ways of honoring veterans through public dance, song, and ceremony to commemorate the past and present martial heritage, ideology, and ethos. While some prereservation cultural elements had been lost and new ones adapted, much had survived, and other forms had even been revived through the social arena of martial-based cultural events. As other social institutions in these cultures declined in their enculturative roles, Southern Plains Indian military society events continued to provide arenas for enculturation and even increased in complexity.
This study of Southern Plains military societies delineates comparatively and ethnohistorically the martial values embraced by the Kiowa, Comanche, and Apache (KCA) since circa 1800, describing how military society structure, functions, and ritual symbols connect past and present. In contrast to most ethnographic religious studies, I combine ethnohistorical documentation and oral traditions with symbolic analysis to elucidate the temporal evolution and role of these more secular sodalities and their symbols, focusing upon the continuity and change, meaning, and functions of martial symbols. While Plains Indian sodalities were not homogenous, they are central to understanding past and present Plains social organization, law, politics, warfare, and religion. Individual tribal histories cover four periods: prereservation (pre-1875), reservation (1875-1900), postreservation (1901-1945), and post-World War II or contemporary (1945 to the present).
Larger patterns of social change and development applicable to other external populations are also discussed. Despite change and cultural syncretism, traditional views and concepts regarding veterans remain largely intact. These cultures had strong warrior traditions and a strong martial ethos in which military societies were the largest form of prereservation sodalities. Because they never relinquished the ideology surrounding the role of the warrior, these groups were able to regain status in their own terms as warriors with widespread participation in World War II. The traditional roles, symbols, and warrior ethos of these military society systems continue via military societies. While these societies and their associated ideology were formerly adaptive in fostering social order, integration, defense, and competitiveness, they continue to serve as an important vehicle for traditional enculturation today, as demonstrated by their adaptive significance for the preservation and maintenance of ethnic identity in a larger socioeconomic and politically encapsulating society.This study demonstrates the extent and variety of relationships between Plains military societies and their tribal-level social organization, which have been inadequately documented and investigated. I hope that it will fill a gap in the ethnological picture, while demonstrating that a critical revaluation of these societies is necessary to understand their role in Plains Indian social organization. This study also develops a set of more clearly defined military society characteristics which can be used to describe Southern Plains societies and compare others cross-culturally. While some functions and material culture have been examined (Wissler 1916b), this study looks at other organizational and social features.
Furthermore, I want this work to be useful for the Native American communities which participated in its making, as well as for the anthropological community. Many ethnographies today present only data which are needed to support the larger theoretical points of the scholar and offer little for the native community. During my fieldwork, members of several tribes often stated that many recent works on their respective tribes were "of no value to us," "had no practical use," were "too technical and too scientific to understand," and "included hardly any of the material which we worked so hard and long to provide them." One of the ethnohistorian's strongest commitments is to publish primary data. There is a growing interest in cultural awareness among many native peoples who are actively trying to research, document, preserve, and revive elements of their culture. During my fieldwork, many military society leaders and elders expressed concerns that these data be published as completely as possible to help them document and preserve their military societies. I am seeking a middle ground between the ethnography of old and the academic theoretical concerns of today, by providing solid descriptive accounts for each group, while also addressing anthropological concerns. To have deleted much of this material would have eliminated crucial elements such as regalia, song texts, and dance choreography, which describe and convey many of the ongoing symbols and ideology of these groups. A sound ethnographic description can make the results accessible to the native communities involved, better support their claims, and allow the reader to examine and critique the material on which my arguments are based. Raising the issue of practicality in current ethnohistory, many native peoples ask, "What's in it for us?" If scholars wish to continue to work in various communities they need to make the published results more rewarding for the individuals upon whom such work depends-the native community--for without their hospitality research opportunities become limited and more archival in nature. Part of this involves making the consultants active participants in the research process.
This study is primarily aimed at defining military societies on a regional level and then comparatively analyzing their interactions concerning various levels of symbolic and social organization and integration. Throughout the immense changes Plains Indian communities have experienced over the last two centuries, a large part of their social structure and fabric has remained. Much of the ideology and ethos associated with traditional veterans in these communities has also remained. As we shall see, these veterans have indeed endured.
In the anthropological literature, Plains Indian military societies have often been labeled voluntary associations (social groups based on voluntary membership) that assumed preparatory functions for communal ceremonies such as the Sun Dance and policed communal hunts and the camp. In addition to these functions, the societies are largely described as men's social dance associations differing primarily in songs, dances, and regalia. Virtually all studies of Plains military societies focus on functional explanations, concentrating on ecological aspects, such as their role in policing communal buffalo hunts, while giving little attention to their organizational roles. Economic aspects perhaps have been functionally overemphasized by scholars such as Symmes Oliver (1962). Beyond brief ethnographic descriptions and limited use of ecological data (Wissler 1916b), there has been little analysis of ethnographic or historical data to assess these organizations and the extent of their relationship to the larger levels of social organization, interpersonal relations, tribal integration, and economics. Archival and field research among Southern Plains populations indicates that military societies (1) often were much more structured than previously noted, (2) shared many general similarities yet varied significantly in detail from one population to another, (3) functioned in a variety of social and political contexts, and (4) differed in their individual roles within their respective communities and that (5) current societies often exhibit a high degree of cultural continuity from their antecedents. These data can be further explored to gain insights concerning not only past and present military societies, but their impact upon other components of past and present Plains Indian cultures. Moreover, a comparative analysis permits the temporal exploration of larger patterns of social behavior for the North American Plains cultures.
In anthropology, "military societies" are categorized as sodalities, from the Latin word sodalis, referring to voluntary associations in contrast to kinship and residence groups (Lowie 1948:14). As Elman Service (1962:21) describes, a sodality is a "nonresidential association that has some corporate functions or purposes." Such a definition resembles the German term Gesellschaft or "special purpose group," yet does not imply voluntary membership or a mandatory single special purpose. Sodality is therefore a broader concept that is explicitly defined as nonresidential, whereas other forms such as clan, nonlocal clan, various residence groups, corporations, and societies are less clearly defined in terms of residence and purpose and are nonresidential only by implication or expectation. Service (1962: 116) distinguishes between kinship-based sodalities (clans, kindred, and segmentary lineages) and non-kinship-based sodalities (associations, age-grades, and warrior and ceremonial societies). Unlike residential groups, sodalities are usually not directly demographic and do not respond in the same way to population increases and greater densities of residential groups. Because they are nonlocal, sodalities acquire political functions by crosscutting, and thus integrating, various residential units. Sodalities integrate residential units and configure society and therefore are important in social structure (Service 1962:22).
All sodalities have latent and manifest functions and contain sociologically integrative (latent) functions simply by being nonresidential, because they crosscut some segment of the residential groups, however small. Because a sodality's manifest functions are related to its structure, sodalities can vary greatly. They normally have one or more manifest functions, which are usually related to their structure. They are in a sense, then, special-purpose groups, with their structures dictated by their functions and with integration as a by-product. Service (1962: 23) differentiates residential groups from sodalities because the structure of the family is more varied and complex in form and because nearly any form of family structure can still accomplish its manifest functions.
Sodalities vary worldwide, but are most common in the anthropological literature of Africa, Melanesia, Taiwan, and North and South America (Ritter 1980). Cross-culturally, sodalities occur in numerous forms, with specialization in various areas, such as police, warfare, status, religion, recreation, and medicine or curing (Lowie 1927; Service 1962: 118). Although the distinguishing features categorizing sodalities are more functional than organizational or ideological, all societies perform overlapping functions in numerous social spheres. Men's sodalities typically occur in greater frequency, are more highly organized, and are more secretive than female sodalities (Hunter and Whitten 1976: 362). Sodalities are generally secondary groups or associations in which membership is voluntary in most cultures. Some cultures require sodality membership as a prerequisite to all major life events, such as marriage (Lowie 1920:80, 276). Membership can be by ascription, inheritance, achievement, purchase, and/or performance. Sodalities may also serve as special-purpose political associations organized on the basis of sex, age, economic status, and personal interests and are found in both public and secret spheres of action (Hunter and Whitten 1976:362).
Service (1962: 113) defines pantribal sodalities as "the means of solidarity that are specifically tribal additions to the persisting band-like means," the addition of an encompassing integrative structure to an existing form of social organization. He suggests that all sodalities are formed from preexisting ingredients found in band society that were transformed by two primary adaptive circumstances, the natural (organic and inorganic) environment and the superorganic environment (presence of competing societies). Clans, segmentary lineages, and kindreds are modeled after vertically formed family units, while warrior societies and age-grades follow the horizontal forms of brothers, old men, and cooperative hunting and ceremonial groups (Service 1962: 113, 116). Thus, he attributes the development of pantribal societies to competition between societies in the neolithic phase of cultural development and suggests that complete tribal integration would not occur without foreign political problems (Service 1962: 113-114). By adapting the way in which groups maintain a nonlocal base, pantribal sodalities unite bands beyond the limitations of intermarriage. Service (1962: 115-116) asserts that "pan-tribal sodalities are the emergent features which made bands over into a new level of socio-cultural integration and thus a new cultural type." According t0 Service (1962: 115, 181), pantribal sodalities, as one of five basic types of human integration, were prerequisite to tribal formation.
While most theoretical contributions concerning sodalities have come from Service (1962), his assertions are still undocumented, unqualified, and largely unproven. More recently James Boon (1982: 97-111) has reanalyzed Robert Lowie's and Franz Boas's earlier rejection of Lewis Henry Morgan's theoretical view that nonliterate cultures were centered exclusively on kinship. As Boon notes, Lowie suggested that state organization arose when the androcentric "voluntarist component" came to dominate in the absence of other social structures. More importantly for this study, Boon (1982:102) points out Lowie's observation that sodalities "reveal a tribe's capacity to sample alternate social forms, without necessarily adopting them as thecentral components of the social machine." He also shows that a tribal a group commonly experiments with patterns of social organization "that are fundamental to its neighbors." These points are essential in comparing Southern Plains military societies.
Plains Military Societies
In the anthropological and historical literature, the term "military" or "warrior society" most commonly refers to the martial-oriented sodalities of the North American Plains and Africa. The exact origins of the term, however, are unknown. For purposes of clarity, in this study the terms "military society," "society," and "sodality" are used synonymously. Thus Kiowa society (culture) and Kiowa societies (sodalities) are distinguishable by context.
Perhaps the earliest mention of an organization resembling a North American military society is Louis Hennepin's (1698:280; Abel 1939:187) 1680 description of a group near present-day St. Paul, Minnesota, who served as police by seizing the goods of individuals who had broken communal hunting laws as punishment. In 1803 Pierre Antoine Tabeau noted the temporary authority of appointed Brulè Lakota police to kill horses and dogs, to destroy lodges, to break weapons, and to confiscate or destroy goods to enforce civil decisions as well as bison hunting laws. Resistance generally brought a beating, while penitence resulted in restoration of one's lodge and goods, often in an increased amount, reflecting patterns of military society policing in many Plains communities. Although Tabeau, like most writers, was most impressed with the coercive powers associated with the collective hunt, he also mentioned public functions of Brulè societies, including initiations, feasts, and naming ceremonies (Abel 1939: 116, 245). Indeed, Clark Wissler's (1912) and Lowie's (1916a:908-910) findings indicate that the policed bison hunt and the association of police duties with Plains societies originated among the Western Dakota (Lakota). On August 30, 1804, Meriwether Lewis and William Clark describe a Teton Sioux military society with obligatory battle behavior and fraternal characteristics of sitting, camping, and dancing together, reportedly in imitation of one of several Crow military societies (Thwaites 1904:1:130). Their references further demonstrate the widespread temporal presence of Northern Plains Indian military societies, from which later scholars infer their diffusion to other tribes (Lowie 1916a).
Wissler (1914:12, 18) asserts that such "typical" Plains traits as the Sun Dance and military societies were well distributed among Plains populations prior to the diffusion of the horse and its subsequent role in diffusing other traits. William Christie MacLeod (1937:199) states that Plains military societies derived from some form of Woodlands culture police prior to adapting to the Plains, which seems likely because several of the later Plains tribes with strong military society traditions such as the Cheyenne, Lakota, and Crow (formerly Hidatsa) were previously agriculturalists or, like the Omaha and Pawnee, continued semisedentary, agricultural pursuits. Francis Haines (1976:171-194) suggests that Plains Indian men's societies became extremely important with the increase in warfare, which is also likely in light of Bernard Mishkin's (1940) and Jane Richardson's (1940) Kiowa data. Thomas Biolsi (1984) has demonstrated that resource competition and large-scale warfare on the Great Plains, to which military societies were related, resulted from the simultaneous effects of ecological and cultural causes.
By the mid-1800s Plains dream, police, dance, and military societies were known to Euro-Americans, but remained undefined and undifferentiated. Military societies continued to be poorly understood because of widespread similarities in society obligations, names, regalia, functions, and associations with warfare. Despite individual tribal variations, these components served to strengthen societal solidarity. Many early writers automatically assumed the historical and psychological unity of all Plains organizations under discussion and grouped them under the name of "military" or "warrior" societies (Lowie 1916a: 883). Because of frequent associations with Plains warfare, the term "Dog Soldiers" was often used to refer to all Plains Indian military societies, from an imperfect rendering of one of the Plains' most prominent and widespread forms of military societies (Mooney 1898:229; 1912d:862).
Hugh Lennox Scott and W P. Clark (1982:354-365) both used the terms "soldier," "soldier bands," and "societies" when discussing several Plains military societies in their works on Plains Indian sign language. Several consultants interviewed during this work frequently used terms such as "group," "band," "clan," "society," and "lodge" in describing tribal military societies. The first to attempt to provide an ethnological definition for the term "military societies" appears to be James Mooney, who writes:
Throughout the plains from N. to S. there existed a military organization so similar among the various tribes as to suggest a common origin, although with patriotic pride each tribe claimed it as its own.... In each tribe the organization consisted of 4 to 12 societies of varying rank and prominence ranging from boys or untried warriors up to old men who had earned retirement by long years of service 0n the warpath and thenceforth confined themselves to the supervision of the tribal ceremonies. The name of each society had reference to some mystic animal protector or to some costume, duty, or peculiarity connected with the membership.... Each society had its own dance, songs, ceremonial costume, and insignia, besides special taboos and obligations. The ceremonial dance of one society in each tribe was usually characterized by some species of clown play, most frequently taking the form of speech and action the reverse of what the spectators were expecting.... At all tribal assemblies, ceremonial hunts, and on great war expeditions, the various societies took charge of the routine details and acted both as performers and police (Mooney 1912d:863).
More recently there has been some question whether the term "military society" accurately categorizes such Plains organizations. Because most Plains military societies were non-kin-based and pantribally organized, they are not believed to have fought as units during any given battle. Thus, it has been suggested that the term "military societies" is not accurate. E. Adamson Hoebel (1978:40) supports this point:
The societies are not organized companies on the order of colonial militia in the early days of American settlement. They are, in their way, somewhat more comparable to local American Legion or V. F.W. Posts--social and civic organizations mainly centered on the common experience of the members as warriors, with rituals glorifying and enhancing that experience, and with duties and services performed on behalf of the community at large.
Hoebel maintains that Plains military societies are similar to American veterans' groups: members perhaps did not fight together in the same units and battles, but, upon returning home, entered the same veteran fraternity of their local residence area. New Plains data indicate that Hoebel's characterization is valid only in later (ca. post-1850) historical periods. The presence of a common and collectively shared ideology, ethos, and associated symbols significantly counters this assertion.
Prereservation Plains Military Societies, Sodalities, and Voluntary Associations
In this study I wish to distinguish between a voluntary association or sodality and a Plains Indian military sodality. A voluntary association or sodality includes a wide range of social forms; thus, any voluntary organization can take practically any form (poetic, athletic, theatrical, hunting, medical) and can function for various purposes (education, entertainment, exercise, common interest, economic, curing). A major problem arises in how to classify sodalities, with emphasis generally being given to elements of social structure or functional attributes. Service (1962:23) asserts that a sodality is often organized so that it can perform specific "manifest function"(s). However, a sodality can accomplish a variety of personal, group, and communal goals. Boon (1982) discusses Lowie's assessment of voluntary associations as a springboard for a sense of solidarity exceeding the family level, which holds true, as nearly every sodality provides integration, socialization, and benefits to a segment of the larger society. In addition to being based upon common interests and providing varied communal services, however, several elements more narrowly distinguish Plains military societies from other North American sodality forms and from regional patterns within the Plains.
Numerous Plains societies of several different types existed among both sexes. Alfred R. Kroeber (1907:63) associates three spheres of action with men's societies: civil, military, and religious. He notes that a men's society may include elements of one or all three spheres of action and that the difficulty in classification lay in differentiating the primary function of a society from its secondary functions, which were subsequent or subsidiary developments. Pointing out the high degree of variation in Plains societies, Lowie (1963:113) observes that some organizations had virtually no religious aspects, while others, such as the Pawnee and Arapaho societies, were religiously structured. However, as N. D. Humphrey (1941:431) contends, at tempts to classify Plains societies must include not only factors of social organization, but their functional roles: "the organization typifying the nature or character of these societies cannot be determined by denoting mere structural elements. Rather the character of societies is a reflection of their functional role in Plains culture, and in order to characterize these societies it is necessary first to formulate and understand their functional roles."
Several criteria distinguish the prereservation military societies of mobile plains hunting and gathering groups from those of the Prairie and other re as and from other types of Plains Indian sodalities or voluntary associations One of the most visible and perhaps overemphasized roles is policing large communal hunts. On the Plains, such duties were generally performed one of four ways: (1) by a particular society or clan (the Mandan Black Mouths Society or the Iowa Elk Clan), (2) by a temporarily appointed portion of an individual society (Ponca), (3) by various societies in turn (Kiowa and Cheyenne), or (4) by a group of temporarily appointed distinguished men without associational affiliations (Pawnee, Kansa, Osage). Like many semisedentary Siouan and Caddoan populations, other Southern Plains tribes such as the Caddo, Lipan Apache, Osage, Tonkawa, and Wichita had more temporarily organized social organizations that similarly performed the functions provided by Plains military societies. However, these social units were often kinship and not sodality based. As Lowie (1943:70-71) states, "Where military societies are lacking, such activities quite as naturally devolve on other preexisting units, such as clans or a general honorary class of braves." Many Prairie tribes lacked a form of military sodality which functioned collectively beyond temporary policing duties.
Ecologically the Plains area differs from other geographic areas and thus provided opportunities for different social organization duties and roles than in other ecological zones. More importantly, Plains military societies are not (1) regulated by a priesthood as among some Puebloan groups, (2) organized in military moietylike war and peace divisions as among the Sac and Fox, (3) clan or town oriented as among the Creek and some sedentary Prairie-Plains groups, or (4) temporary appointments of members not of the same sodality as among some Prairie-Plains groups. Plains military societies (1) tend to be based upon voluntary membership with the exception of some contrary societies, (2) contain cross-band membership, (3) are either age-graded or coordinately organized, (4) contain a group name which is generally associated with some aspect of their dress, dance, duty, or associated group ideology, (5) conduct meetings in conjunction with tribal aggregations, (6) may (but not always) engage in warfare as a collective unit (prior to circa 1850), and (7) contain a central focus on warfare and individual deeds, including coup recitations and performances celebrating and enhancing martial activities, as opposed to doctoring and shield societies focusing upon specific forms of war power and healing.
While some military societies contained items associated with war power, they differ from shield societies. In contrast to the pantribal, non-kin-based membership of most military societies, shield societies were much smaller sodalities, generally composed of limited units of bilateral kin, with a focus on shield and warfare power, were generally more secretive, had fewer public social events than the larger military societies, and were largely residential units according to Service's criteria (1962). Thus, there are several basic differences between Plains and Prairie sodalities in terms of social structure, functions, interpersonal relations, and economic and ecological roles. In view of such variations, Plains and Prairie societies should not be classified as a single homogeneous topic, but perhaps as subsets of a larger social form.
While all of these criteria are significant and are addressed in this study, it is the last criterion which plays a dominant part in the history of these organizations. Plains men's military societies visibly promote the common central theme of a warrior tradition--a warrior ideology, which at specific times can be argued to be an ethos. Much of the work of Margaret Mead, Gregory Bateson, and Ruth Benedict (Harris 1968:393-448) involved attempts to define ethos and national character in studies of culture and personality. Roger Pearson (1985:78) defines ethos as "the distinctive quality of any individual culture or society." Charles Winick (1968:193) defines it as "the totality of the distinctive ways of living that separate one group from another, especially its values. Ethos also denotes the emotional quality possessed by socially patterned behavior." Benedict (Harris 1968:398-403; Winick 1968:193; Pearson 1985:78) suggests that the ethos of a society might be classified as "Apollonian" or "Dionysian"; however, most theorists believe that she oversimplifies the concept of ethos, which is rooted in the complex totality of a society's mores, customs, and cultural traits.
For this study, I define "ideology" as the opinions and doctrines of a particular group and the associated expressive means (symbols, words, songs, etc.) that reflect them and "ethos" as the deeper-seated, more pervasive, core cultural principles that characterize a particular group. Any culture can of course have more than one type of ethos, and transition to and from an ideology to an ethos also occurs. That is, an ethos is flexible: it can expand and contract or increase and decrease in intensity, depending upon appropriate circumstances that overlap, enhance, or supersede existing ideology. Thus, while a military and patriotic ideology could be argued to exist in American society, nationalism is more reflective of its ethos. However, during the recent Persian Gulf War, the American ethos could be argued to have become more martially focused, yet still nationalistic. Likewise, a martial ideology can be argued to exist throughout the annual cycle of many prereservation Plains Indian military societies, which was periodically enhanced to an ethos during times of intensified group integration such as society meetings, communal aggregations, and horse and revenge raids. Boon's (1982) discussion of voluntary associations as a source of solidarity exceeding the family level suggests the flexibility of Plains military sodalities in periodically providing such solidarity in a martial context for various larger socially integrative purposes.
A martial ethos is only one type of ethos common to many Plains communities. On a general level, I contend that there is a similar shared martial ideology (and at times ethos) for the groups discussed in this work, which has continued through time and has remained an important part of ethnic identity. These sodalities have performed a number of enculturative functions that, though varying among individual military societies and among tribes through time, are of a generally similar pattern. As an adaptive form of social organization, these societies have changed through time (membership requirements, frequency of contact), yet have retained a number of organizational and functional forms (ceremony, organization, community services, cultural preservation). As they have changed, these societies have assumed new and adaptive strategies (integration of U.S. military participation, incorporation of other tribal traditions, etc.) in becoming new integrative social forms. In doing so, these groups exhibit a generally shared pattern of martial-based ideology and ethos, conveyed through rituals and symbols. Like an actual arena used to hold Plains society dances and powwows, military societies serve as a cultural foundation through which various enculturative forms (language, belief, cultural heritage, ethnicity, values, ethos) are maintained and into which changes (adaptation, syncretism) are introduced-arenas of enculturation. Thus, traditional cultural forms are maintained while adapting to new sociopolitical and economic circumstances. Although individual societies differ on a number of levels, many of their larger core functions, the types, content, and uses of symbols, and adaptations follow a general temporal pattern.
In addition to these criteria, prereservation Plains military societies (1) were voluntary associations focusing upon shared interests, (2) provided various forms of social services (such as aid to the needy) to the larger community, and (3) collectively fostered military spirit, ideology, and ethos while recalling and enhancing the military accomplishments of members individually and as a sodality. In view of the widespread traditional Plains emphasis on veterans, it is this third criterion which most visibly characterizes and defines Southern (and possibly all) Plains Indian military societies as a distinct type of sodality. Thus, an understanding of the associated ideology and ethos, and not solely the functions, of a sodality is necessary to differentiate types of sodalities. Prereservation Southern Plains Indian men's military societies were comprised of members of varying veteran status, who periodically and collectively met for social or semisacred purposes, including the representation of individual and collective war-related performances, provided social and economic services to the larger community, and enhanced tribal integration. Military societies also performed other socioeconomic functions such as camp security and alignment, charity, preparation for communal ceremonies, and regulation of the communal hunt, which was enforced by confiscating illegally seized game, destroying property, and inflicting public shaming and corporal punishment (Wissler 1916b). Although Plains women's societies have largely been overlooked, their spheres of action included religious, military, and craft guilds (Lowie 1954: 105-114). Like Plains men's societies, women's organizations also frequently contained overlapping functions in more than one sphere of action. Similarly, Plains Indian women's military societies participated in and promoted a shared martial ideology and ethos. They resembled men's military societies by being (1) voluntary organizations whose members performed (2) social and or semisacred activities (sometimes including prayer, blessings, supplication, and the use of supernatural power) to support and ensure the success of the military activities of their male counterparts, while (3) providing various community services.
Among prereservation Plains groups, this martial ideology was often maintained and promoted through a series of collective social and semisacred gatherings at both divisional and pantribal levels, focusing upon rituals, dancing and singing, feasts, and public validation of individual accomplishmentss and status via coup recitations and the redistribution or giving-away of property, while social and public services to the larger community were performed. These ceremonies validated individual and collective (societal) accomplishments while reinforcing individual (member), collective (societal), and communal (divisional or tribal) recognition of warrior status.
As this work covers a lengthy period (ca. 1800-1999) it includes significant changes in many aspects (dress, songs, membership requirements, functions). This baseline definition of military societies works well for the prereservation era, but subsequent periods and adaptations necessitate changes in definition. While an emphasis on a martial ideology and ethos can be seen as an Anglo romanticization of Plains life (such as Mails 1973), which neglects the varied socioeconomic functions of military societies, or as a retrospective ideology that twentieth-century Plains communities have to establish their own sense of history and continuity (perhaps in response to external stereotypes and expectations), I demonstrate that such a martial ideology has continued to the present, through strategic adaptation, and still exists. Much of the core ideology (emphasis on veteran status, means of honoring veterans, community functions, ethnic identity, social integration, and incorporation of other cultural elements) associated with prereservation military societies remains in successor organizations today, though not in the romantic form portrayed by popular writers and in movies. Although the specific social functions of a martial ideology have varied over time, important elements of a tribal heritage stemming from earlier martial traditions remain a prominent aspect of contemporary Southern Plains Indian communities.
Forms of Society Grading and De Facto Age-Grading
Nineteenth-century sodalities, including Plains Indian military complexes, have traditionally been categorized in two major forms of organization: age-graded and nongraded or coordinate. An age grade is a category of persons who fall within a culturally distinguished age range. Members typically enter age-graded systems by age and purchase (Lowie 1916a:884). In age-grade societies young men of a relatively similar age or age-set (a set of persons of similar age and sex who move through some or all of life's stages together as a group) joined the lowest or first society at an age deemed sufficient by the respective culture. Each age-set progressed upward through a hierarchy of societies as a collective unit, by periodically purchasing the ceremonies, songs, regalia, ritual knowledge, and position of the next higher group (Lowie 1963:107). Thus, advancement was heavily linked to ritual knowledge and age. The process continued, with the oldest age-set normally retiring with every series of upward purchases and movements.
Age-grading was limited to five Plains groups: the Arapaho, Blackfeet, Gros Ventre or Atsina, Hidatsa, and Mandan. W E Whyte (1944:68) defines age-grading as "a system of differential distribution of rights and obligations, of prohibited, permissible, and required activities, according to different (socially recognized) periods of life and according to the social distinctions established between the sexes." In general aspects, Plains Indian age-sets are similar to those of Africa, Melanesia, and elsewhere, consisting of age-sets of ranked, relatively homogeneous age-groups. The sequential passage of age-sets through a number of grades required a formalized right of passage for each new level. Lowie (1916a) determined that Plains societies (1) originated among the Northern Plains village tribes (especially the Mandan and Hidatsa), (2) developed from a series of previously ungraded societies, (3) represent a special and late development, and (4) were transmitted by the Hidatsa to the Arapaho and later to the Blackfeet and that (5) in some instances even the same society varied in order, age affiliation, and content between tribes.
Contrasted with age-graded systems were the non-age-graded or coordinate systems, in which a man of appropriate age would join a society through invitation, choice, or some sort of qualification, and, depending upon the individual tribe, could either remain in that society for his entire life or shift membership among other societies, also based on personal choice or invitation (Lowie 1963: 110-111). Nongraded military societies were more common than age-graded societies on the Plains and were found among the Kiowa, Cheyenne, Lakota, Crow, and others. Except among the Arikara and Sarsi, no Plains nongraded military societies required age qualification or entrance fees (Lowie 1916a: 884). Other variations existed, however: the Comanche developed divisionally specific societies at a relatively late date (Meadows 1995), while the Plains Cree had one warrior society per tribal band, but occasionally purchased new societies and dances from other bands or tribes (Lowie 1963: 110). Comanche and Cree military societies were based largely upon factors of proximity and dependent upon the territorial range of residence bands that exploited a common area.
This work focuses on Southern Plains tribes exhibiting nongraded military sodalities that exclude membership based on clan, moiety, or phratry functions. As previously defined, these groups all contained what could be called "classic" or "typical" military societies when speaking generally of nineteenth-century mobile hunting and gathering Plains cultures. Although descendants' organizations characterized by annual gatherings and traditional dances are common among the Kiowa, Comanche, and Apache (KCA), I have excluded them because of their kinship-based descent structure (usually from an apical ancestor) and their qualitative difference as a lineage-based structure and not a sodality.
New data on Southern Plains military society systems and a review of Wissler's (1916b) Societies of the Plains Indians suggest that the traditional classification of age-graded and nongraded societies needs revision, as the age-grade factor is not the sole determining factor for both variations. Lowie (1916a: 883) states: "Although systems of societies graded by age are confined to five of the Plains tribes ... many other tribes of this area have societies so closely resembling the age-organizations in name, regalia, and functions that a hard-and-fast line cannot be drawn with the age factor as the basis for classification." Several "ungraded" Plains societies exhibited numerous patterns of differential membership, status, ritual knowledge, economics, and functions associated with age; thus, characterizing them as "non-age-graded" is not totally accurate. Therefore, I suggest that it is perhaps more accurate to view the two types as variations of a single form of social organization (age-graded and de facto graded) rather than as polar opposites (age-graded versus nongraded). Kiowa, Apache, and other Plains data point to the existence of a widespread form of de facto age-grading or at least a sodality-based clustering of largely (but not totally) similar aged cohorts in many Plains populations previously considered to have ungraded or coordinate societies. Widespread characteristics involving differentially defined membership and a series of specific and socially ranked or de facto graded societies combined with children's societies, varied forms of intra-tribal intersocietal rivalry, differential ritual knowledge and assignment of duties association of general age cohorts, individual war rank, and wealth and social status all point toward this conclusion.
Most literature on sodalities exists in tribal ethnographies and varies enormously according to the individual ethnographers' assessments of the primary functions of various sodalities. The recorded foci of Plains Indian sodalities range widely, including religion, politics, recreation, warfare, property owning, wife owning, cooperative aid, and so on, making generalizations difficult (Stewart 1977:2-3, 7). Most works dealing with sodalities emphasize age-graded forms, with theories largely derived from Africa and Taiwan, where the most highly developed, numerically largest, and most widespread forms of age-group sodalities occur. The enormous literature on age-groups depicts sodalities as the central institution in many tribal societies and has generated larger theoretical questions following comparison with sodalities elsewhere (Stewart 1977).
Aside from ethnographic descriptions, little discussion of North American military sodalities exists. Theoretical works largely discuss origins and development (Schurtz 1902; Lowie 1920, 1927; Whyte 1944; Eisenstadt 1954 Oliver 1962; Hanson 1988), although Frank Henderson Stewart (1977) attempts a more comparative survey of the fundamentals underlying age-group systems. Most later works are concerned with determining their development, importance, and effectiveness as a means of coercive power in social control (MacLeod 1937; Provinse 1937; Lowie 1943; Hoebel 1954; Bailey 1980) or their role as germinal instruments in the process of state formation (Hoebel 1936; Llewellyn and Hoebel 1941). The largest concentration of anthropological work involving Plains military societies is volume 11 of the American Museum of Natural History publications (Wissler 1916b). This survey involved a detailed ethnographic investigation of Plains military societies and the Sun Dance, concentrating on tribal groups which had not yet received ethnographic attention. These works focused primarily upon trait distribution and clusters to reconstruct the historical development and diffusion of these complexes within and among various Plains tribes.
However, as Fred Eggan (1966) and Oliver (1962:6) discuss, these publications do not address how societies functioned. While they described society regalia, little emphasis was given to the more functional or organizational relationships of military societies to tribal integration, religion, kinship, or social, political, and economic organization. Moreover, those tribes which had already been studied were not revisited and Southern Plains populations were almost totally ignored. Lowie (1915a, 1916b) published a total of seventeen pages for the Kiowa and Comanche (visiting both only briefly), the Plains Apache were not included, and further Cheyenne and Arapaho research was precluded because of the brief society descriptions in Mooney (1907), Dorsey (1905), and Kroeber (1983). This paucity of material resulted from a lack of research time in the respective areas and not from the absence of such organizations.
Because many scholars considered Plains Indian military societies (especially nongraded forms) to have been adequately explained, they were largely dismissed. It was not until the mid-1930s that further data on Southern Plains military societies were collected by general ethnological field schools, by the Santa Fe Laboratory of Anthropology among the Comanche in 193 3 and the Kiowa in 1935 and by scholars such as J. Gilbert McAllister for the Plains Apache in 1933-1934 and E. Adamson Hoebel for the Northern Cheyenne in 1935- Yet few of these data concerning military societies have been published (McAllister 1935, 1937 Hoebel 1936, 1940; Llewellyn and Hoebel 942). These data are invaluable primary sources and counter the acculturationist attempts to document what was believed to be the last of the pristine and rapidly disappearing prereservation Indian cultures.
Methodology, Theory, and Approach
This study of Southern Plains Indian military societies is based on five major areas of focus: (1) a controlled regional and temporal comparison of three distinct ethnic populations (Kiowa, Comanche, and Apache, frequently referred to within these communities and in this work as KCA); (2) a multifield ethnohistoric approach combining personal fieldwork, critical examination, and use of extensive archival and oral history sources; (3) an approach countering the weaknesses of acculturation studies and using the strength of New Indian History to emphasize indigenous meaning, to document the active role of native communities in their own processes of culture change and to show cultural innovation, change, and continuity, rather than cultural loss; (4) a concern for the temporal processes of change, syncretism, and cultural revival; and (5) a focus on the cultural role of martial symbols through time.
This study combines extensive ethnographic fieldwork, archival research, and analysis of symbols to (1) reconstruct the history of KCA military sodalities and demonstrate how they reflect past and present tribal values, Particularly through military society rituals; (2) determine the significance of these societies for their respective tribal forms of social organization; (3) show how these groups serve as enculturative and adaptive forms of soorganization for the maintenance of a distinct tribal ethos and ethnicity and, later, for identity in a larger, encapsulating society; and (4) delineate ethnohistorically and diachronically the military society symbols embraced the KCA and how these functioned in social integration and enculturation through time. Traditional military society social structures have an enduring g reality and in being periodically reinvigorated have been adapted to modern circumstances.
The Southern Plains was chosen as a research area for three reasons: (1) Southern Plains military societies have received little ethnographic attention in comparison with the Northern Plains (Wissler 1916b); (2) ironically, the largest percentage of surviving ancestral-based military societies are on the Southern Plains; and (3) the KCA have shared a generally similar history throughout much of the nineteenth and twentieth centuries. I was also able to work with the last elders with direct information concerning older society activities. An ethnohistorical approach offers a broader view of the process of repetitious and changing structures and order.
A comparative regional approach is necessary to elicit both individual tribal data and broader regional patterns. This approach facilitates the comprehension of the social and cultural dynamics of each group as a whole in the context of its own relations and interactions, as well as those with the neighboring groups with which it continues to share a generally similar historical experience. Like distinct ethnic communities, individual sodalities react to those within its own population as well as to those in neighboring communities, especially to sodalities possessing similar cultural forms. To discuss the societies of one tribe without comparing and contrasting the others distorts the cultural dynamics of their historical experiences, which include change, borrowing, diffusion, perception, readjustment, innovation, and creativity." In many respects the Kiowa, Comanche, and Apache define and reinforce their identity in reference to one another, much as Loretta Fowler (1987) describes for the Gros Ventre and Assiniboine. In addition, they all contained what could be called "classic" or "typical" military societies for nineteenth-century mobile hunting and gathering Plains cultures. Numerous KCA similarities allow an in-depth comparison of military societies in the social structure of several ethnic units existing within a generally similar cultural and ecological region-the boundaries for an excellent controlled regional analysis and comparison.
Sources and Research Methods
While ethnohistorians generally emphasize archival or documentary sources over oral sources, both are necessary, when available, to achieve the most thorough assessment of any phenomenon. They must be used critically and in different ways. I have emphasized written (archival) and contemporary "oral history" sources over many other forms of archival sources due to concerns about quantity and relevance and because many sources are eitherinferential (contain limited, but sufficient data to infer a connection) or substantive (contain a great quantity of data). The difference often lies in the purposes of the recorders and their familiarity with the culture and topic being described. For military societies, most early European sources are inferential regarding time and location and often lack further content. Archival sources are not always sufficient in quantity or quality for historical reconstruction. The Kiowa Agency records contain a plethora of data concerning attempts to prohibit the Sun Dance, Ghost Dance, and intertribal powwows, but virtually no references to men's societies. The combined use of oral and archival sources is necessary to supplement and confirm one another.
The alleged invalidism of oral history is no longer credible (as shown by the works of DeLaguna 1960:205; Vansina 1965, 1985; Miller 1980; Brown and Roberts 1980; Wedel and DeMallie 1980; Keesing 1986; Moore 1987:5; Kracht 1989; Krech 1991:360; and others). I have relied heavily on "archival" oral histories in the form of ethnographic fieldnotes for several reasons. In contrast with many sources of archival data, ethnographers are trained for ethnographic work, have linguistic and cultural insights, follow a formulated research agenda and a more holistic approach, and appreciate cultural relativity, which often makes their data more accurate than other external accounts. As primary sources, ethnographic and ethnological materials (1) were obtained with an organized and planned approach, (2) generally contain more detailed data than other sources, (3) include a more sensitive line of inquiry, and, most importantly, (4) were often obtained firsthand from the last generation of actual participants in these prereservation systems. My reliance on fieldwork, ethnographic fieldnotes, and oral historical accounts for early portions of this work is stressed, yet they are used cautiously, to complement, supplement, and cross-check the archival record.
This study would not have been possible without long-term, intensive fieldwork, which is crucial for gaining an understanding of another people's culture. Fieldwork is necessary in order to reveal a people's ideology and ethos-their characteristics and habits, motivations, styles, and their view' point and manner of solving problems (Lurie 1961: 87; Keesing 1986: 290). Without fieldwork one can never truly have an accurate feeling for a culture, and it is from the small and intricate details that the larger patterns evolve. Personal and family histories, knowledge of the language, songs, genealogies, oral history, anecdotes, and previously undocumented ceremonial accounts provided the insights that led to new questions and directions of inquiry in this work. These produce a more complete and dynamic depiction of a culture. As Richard Perry (1991:9) has shown for the San Carlos Apache, traces of a people's heritage remain despite numerous social and ecological changes and continue to serve as important clues to understanding a culture's past.
The importance of contemporary fieldwork cannot be overemphasized. To reiterate Moore (1987:317): "any historian or ethnohistorian who raises issues that are within the experience of modern informants simply cannot be excused from the necessity of doing fieldwork." No archival study or brief anthropological fieldwork stay would have yielded the data or results included here. Only extensive, long-term, and frequent interaction permitted the elicitation of integral information and connections between various segments of these three cultural systems. The comparative approach allowed a constant cross-checking and inquiry, which in turn produced better methods of inquiry and more thorough results. This work demonstrates the possibilities for those interested in combining published ethnography, modern practice, and early documents (Moore 1987), as one documentary source or an elder in one community often prompted questions and led to answers in another.
Observations of society activities and interviews with tribal members were conducted from October 1989 to the present. Residence in Oklahoma allowed me to conduct lengthy summer visits as well as year-round fieldwork for six years. Reputable and knowledgeable society leaders and male and female tribal members ranging in age from sixteen to ninety-nine were consulted. Younger tribal and nonsociety members provided valuable internal (tribal) yet "external" (nonsociety member) voices.
The great variety in responses to contact by native peoples in North America suggests numerous factors behind the maintenance of distinct cultural communities. Acculturation studies often examine differences between earlier and contemporary social forms, while ignoring the meaning of those forms to native peoples themselves. These works thereby frequently proclaim cultural loss where there is in fact continuity and adaptation and thus are based on what the ethnographer rather than the people themselves deem important. While promoting the stereotype of inevitable culture loss and further assimilation, these approaches distort the process through which native communities change and disregard the perception, interpretation, reaction, and adaptation of native peoples. Subsequently, acculturationist approaches have been largely revised (Redfield, Linton, and Herskovits 1936; Linton 194o; Bruner 1957; Herskovits 1958; Spicer 1961, 1971; Fowler 1987, Garbarino 1983:72-73 Trigger 1986:265). As Fowler (1987:7-8) discusses, cultural continuity results from both persistence of certain ideas or customs and resistance to change, as well as from the emergence of new concepts and values as groups adapt to a changing social world. As borrowed ideas, objects, and customs are given new meanings which make them acceptable as innovations, they become, in time, traditions.
This study seeks to counter the weaknesses of acculturation and ethnicity studies, first, by examining the adaptation and cultural growth of Southern plains military societies as an ongoing and adaptive social institution which maintains valuable and conscious enculturative forms for these groups and, second, by examining the temporal use of military society symbols for ethnic maintenance.
Old and New Indian History
Ethnohistorians refer to many historical studies (such as Berthrong 1976; Hoxie 1984; and Hagan 1990, 1993) focusing on Indian-white relations as "Old Indian History" because they emphasize an Anglo viewpoint and often rely solely on archival sources. Such works are largely theoretical and documentary and focus on the detrimental effects of national policies on local populations, attributing cultural change to Anglo exploitation, which is seen as the seminal cause of contemporary situations. While sympathetic to the problems of native peoples, these works generally ignore native peoples' ability to perceive, to interpret, to form strategies, and to set goals, portraying them merely as passive recipients of new sociocultural forms dictated by a dominant society. Many current ethnohistorians emphasize Indian-Indian relations in what is known as "New Indian History" (Fowler 1982, 1987; Moore 1987; Kehoe 1989; Foster 1991; White 1991; Dowd 1992). Using a multifield approach, these works place native peoples at the center and seek to understand the reasons for their actions. Native peoples are seen as conscious actors and participants in the larger historical picture and no longer as simply the passive recipients of external and eventually dominating policies. Scholars are now concerned with recognizing and demonstrating a number of features, including ethnogenesis of social forms, exploring relationships involving change and continuity, demonstrating process or how the present evolved from the past, and recognizing that all cultures undergo multiple and varying adaptations in their evolution.