What are the gender roles in indigenous communities?

Historically, a number of social norms in Eastern Woodland communities demonstrate a balance of power held between women and men . Men and women have traditionally both had the final say over who they would end up marrying, though parents usually have a great deal of influence as well.

What were the two traditional roles often viewed about females?

Marriage and motherhood were still considered the most important job for women according to societal standards, but unlike just a few decades earlier, a woman's standing in her community was not solely hinged on her starting a family.

What are some examples of gender roles in the community?

Traditionally, women are seen as caregivers, nurturers, homemakers, and helpers . Men's gender roles revolve around characteristics such as dominance, assertiveness, and strength. Traditionally, men's gender roles include breadwinner, leader, and protector.

How were gender roles determined in indigenous contexts?

Aboriginal Women's Traditional Roles & Power Women were respected for their spiritual and mental strength and men were respected for their spiritual and physical strength . Women were given the responsibility in bearing children and were given the strength and power to carry that responsibility through.

What are the three major types of gender roles?

Gender role ideology falls into three types: traditional, transitional, and egalitarian .

What are the traditional gender roles in relationships?

Generally, most communities agree that the traditional role of a woman is more nurturing, and men are more dominant. Traditional gender roles in relationships exist to divide the work that needs to be done to support families and society at large . The idea is that we leverage our natural skills.

How have gender roles changed throughout history?

New ideas like socialism, nationalism and women's rights helped transform traditional attitudes and expectations . As a result, gender roles began to shift and change. The labor-intensive Industrial Revolution brought many women out of the home to work in factories. Colonized people began to resist European control.

What were the gender roles in Native American culture?

Gender Roles & Family Men and women played mutually supportive roles. Men were responsible for hunting, warfare, and ceremony, while women were responsible for childcare and the preparation of food and clothing .

What are the roles in indigenous culture?

There are three basic groups of roles; agricultural, pastoral, and hunter-gatherer . Depending on the land and its seasons, many tribes and cultures had a combination of these three basic roles.

What is the gender role in traditional society?

Traditionally, men and women had completely opposing roles, men were seen as the provider for the family and women were seen as the caretakers of both the home and the family . However, in today's society the division of roles is starting to blur.

What are the third genders indigenous people?

Some contemporary Zapotec peoples in Mexico embody the traditional third gender role known as muxe . They consider themselves to be "muxe in men's bodies", who do the work that their culture usually associates with women.

Many Faces of Gender: Roles and Relationships Through Time in Indigenous Northern (Boreal) Communities

ROLES AND RELATIONSHIPS THROUGH TIME IN INDIGENOUS NORTHERN COMMUNITIES

LISA FRINK, RITA S. SHEPARD, AND GREGORY A. REINHARDT, EDITORS

MANY FACES OF GENDER

MANY FACES OF GENDER ROLES AND RELATIONSHIPS THROUGH TIME IN INDIGENOUS NORTHERN COMMUNITIES

Edited by LISA FRINK, RITA S. SHEPARD, and GREGORY A. REINHARDT

UNIVERSITY PRESS OF COLORADO UNIVERSITY OF CALGARY PRESS

University of Calgary Press 2500 University Drive, N.W. Calgary, Alberta, Canada T2N 1N4 www.uofcpress.com

All rights reserved Printed in the United States of America The University Press of Colorado is a proud member of the Association of American University Presses. The University Press of Colorado is a cooperative publishing enterprise supported, in part, by Adams State College, Colorado State University, Fort Lewis College, Mesa State College, Metropolitan State College of Denver, University of Colorado, University of Northern Colorado, University of Southern Colorado, and Western State College of Colorado. The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials. ANSI Z39.48-1992 Library of Congress Cataloging-in-Publication Data Many faces of gender : roles and relationships through time in indigenous northern communities / edited By Lisa Frink, Rita S. Shepard, and Gregory A. Reinhardt. p. cm. — (Northern lights, ISSN 1701-0004 ; v. 2) Includes bibliographical references and index. ISBN 0-87081-677-2 (hardcover : alk. paper) — ISBN 0-87081-687-X (pbk.) (University Press of Colorado) ISBN 1-55238-093-9 (pbk. : alk. paper) (University of Calgary Press) 1. Indians of North America—Psychology. 2. Indians of North America—Sexual behavior. 3. Indian women—North America—Social conditions. 4. Inuit—Psychology. 5. Inuit—Sexual behavior. 6. Inuit women—Social conditions. 7. Gender identity—North America. 8. Sex role—North America. 9. Sex differences—North America. 10. Ethnopsychology—North America. I. Frink, Lisa. II. Shepard, Rita S. III. Reinhardt, Gregory A. IV. Northern lights (Calgary, Alta.) ; v. 2. E98.P95 M35 2002 305.3'089'971—dc21 2002007131 Design by Daniel Pratt 11

CONTENTS

List of Illustrations Acknowledgments 1 Many Faces: An Introduction to Gender Research in Indigenous Northern North America Lisa Frink, Rita S. Shepard, and Gregory A. Reinhardt

vii ix 1

I CONTEMPORARY RESEARCH 2 Kipijuituq in Netsilik Society: Changing Patterns of Gender and Patterns of Changing Gender Henry Stewart 3 Gender Equality in a Contemporary Indian Community Lillian A. Ackerman 4 Celebration of a Life: Remembering Linda Womkon Badten, Yupik Educator Carol Zane Jolles

13 27

II HISTORICAL AND ETHNOARCHAEOLOGICAL APPROACHES 5 Changing Residence Patterns and Intradomestic Role Changes: Causes and Effects in Nineteenth-Century Western Alaska Rita S. Shepard

vi 6 Re-peopling the House: Household Organization Within Deg Hit’an Villages, Southwest Alaska Jennifer Ann Tobey 7 Fish Tales: Women and Decision Making in Western Alaska Lisa Frink

CONTENTS 81 93

III MATERIAL AND SPATIAL ANALYSIS 8 Child and Infant Burials in the Arctic 111 Barbara A. Crass 9 Puzzling Out Gender-Specific “Sides” to a Prehistoric House in Barrow, Alaska 121 Gregory A. Reinhardt 10 Broken Eyes and Simple Grooves: Understanding Eastern Aleut Needle Technology Through Experimental Manufacture and Use of Bone Needles 151 Brian W. Hoffman 11 Gender, Households, and the Material Construction of Social Difference: Metal Consumption at a Classic Thule Whaling Village 165 Peter Whitridge IV SYNTHESIS AND PROJECTIONS FOR INDIGENOUS NORTHERN GENDER RESEARCH 12 Gender Dynamics in Native Northwestern North America: Perspectives and Prospects Hetty Jo Brumbach and Robert Jarvenpa

195

Notes References Contributors Index

211 217 247 249

vii

ILLUSTRATIONS

FIGURES 1.1 4.1 4.2 5.1 5.2 5.3 7.1 7.2 7.3 7.4 7.5 7.6 9.1 9.2 9.3 9.4

Map of volume localities Sketch map of St. Lawrence Island Linda in her late teens or early twenties Excavated features at the Tagilgayak Site on the Unalakleet River, Alaska Structural remains found in Features P3 and P9 at the Tagilgayak Site Stone net sinker and carved bone fish arrow recovered at Tagilgayak Yukon-Kuskokwim delta landscape in July Chance Hill setting whitefish net Ulrich Ulroan brings whitefish to his grandmother, Mrs. Angelina Ulroan Mrs. Angelina Ulroan processing herring at her camp Filleted salmon with fish mark ownership Dried and smoked salmon at the foot of the smokehouse The “male” artifact packets Gender-uncertain objects Gender-female objects Gender-male objects

2 45 53 70 71 72 95 99 101 102 105 106 138 139 139 140

viii

ILLUSTRATIONS

9.5 Leather (and other) scraps and debitage 10.1 Grooved needle and red fox left tibia with needle blank removed via groove and snap technique 10.2 False Pass High School class replicating needles 10.3 False Pass needle manufacturing experiments 11.1 Map of Qariaraqyuk, southeast Somerset Island, Nunavut 11.2 Model of Thule social relations 11.3 Map of Eastern Arctic showing metal sources and probable routes by which trade goods reached Qariaraqyuk during Classic Thule times 11.4 Distribution of blade thicknesses for 42 Thule blades from Qariaraqyuk, Deblicquy, Brooman Point, and Skraeling Island 11.5 Distribution of slot widths for all knives from Qariaraqyuk 11.6 Distribution of slot widths for all projectiles from Qariaraqyuk 11.7 Distribution of slot widths for all heavy-duty manufacturing tools from Qariaraqyuk 11.8 Distribution of slot widths for men’s end-slotted, composite, and side-slotted knives from Qariaraqyuk 11.9 Distribution of slot widths for ulus and men’s knives from Qariaraqyuk

144 152 158 160 167 171

176 178 179 180 180 181 188

TABLES 8.1 8.2 8.3 8.4 8.5 8.6 9.1 9.2

9.3 10.1 10.2 11.1 11.2 11.3

Sites with infants, children, or both 112 Individuals 113 “Other” burial types 113 Individuals in single and multiple burials 114 Multiple burial associations 114 Significant goods 117 Reproduction of table 1 from Newell’s “Artifact inventories of three functional partitions of the [Mound 44 House]” 126 Additions and corrections to table 1 from Newell’s “Artifact inventories of three functional partitions of the [Mound 44 House]” 128–129 Gender-ascribed and gender-sided inventory of floor-related Mound 44 artifacts 132–136 Needle production steps, materials, and debris 155 Comparison of eyed and grooved sewing needle diameters 162 Distribution of inorganic finds at Qariaraqyuk, by house 183 Inferred blade/bit material for Qariaraqyuk hafts based on slot width analysis, by house and artifact class 184 Distribution of metal finds at Qariaraqyuk, by artifact class and gender category 186

ACKNOWLEDGMENTS

WE ARE GRATEFUL TO SEVERAL PEOPLE who helped make this book happen. We wish to thank John W. Schoenfelder for creating the place map and extend our appreciation to three anonymous readers who gave us exceptional suggestions for improving the manuscript (and, of course, all omissions and commissions are ours alone). We recognize and thank the participants in the original “Approaches to Gender in the North” symposium at the 25th Annual Meeting of the Alaska Anthropological Association held in Anchorage, Alaska, March 1998 (organized and moderated by Lisa Frink): Barbara Crass, Lisa Frink, Gregory A. Reinhardt, Petra Rethmann, Rita S. Shepard, Henry Stewart, Steven R. Street, and Barbara Bodenhorn (discussant). And thanks to all of our contributors for their innovation and hard work. We also acknowledge Allen McCartney, who inspired the publication of this book. Furthermore, our gratitude goes out to our editor, Darrin Pratt (director, University Press of Colorado), for his thoughtful support, and to Cheryl Claassen and Gary Dunham who helped us navigate the publishing waters. This book is dedicated to all of the women who have blazed the trail and continue to keep the path open.

MANY FACES OF GENDER

1 MANY FACES AN INTRODUCTION TO GENDER RESEARCH IN INDIGENOUS NORTHERN NORTH AMERICA LISA FRINK, RITA S. SHEPARD, AND GREGORY A. REINHARDT

A demure ten-year-old, wearing typical girl’s attire, sits quietly playing with a doll as he dreams of killing his first game animal. A family sleeps surrounded by the objects of their daily lives, unaware they will soon be crushed by intruding sea ice. A nursing mother dies, and her family kills the infant and buries mother and child together rather than allow the child to starve to death. On a Bering Sea island, a woman born in a dome-shaped house of driftwood and walrus hide dreams of seeing a real lawn of green grass. A man brings the fish he just caught to his wife on the shore, where she cleans, cuts, and marks them as her own. During a time of major social transformation, seamstresses abandon drilled-eye needles and adopt grooved needles in their stead.

THESE ARE A FEW OF THE MOMENTS whose significance this volume explores. By its nature, our focus on gender takes us within houses and families,

LISA FRINK, RITA S. SHEPARD, AND GREGORY A. REINHARDT

Figure 1.1 Selected localities discussed in this volume: (1) Ekwen; (2) Point Hope; (3) Barrow (North Slope Borough); (4) Gambell; (5) Chevak; (6) Cape Newenham and Goodnews Bay; (7) Bethel; (8) Anvik and Holy Cross; (9) Unalakleet and Tagilgayak; (10) Allakaket (Koyukuk R.); (11) Fairbanks; (12) Amchitka Island; (13) False Pass; (14) Sitka; (15) Colville Reservation; (16) Knee Lake (Chipewyan area); (17) Qariaraqyuk; (18) Pelly Bay; (19) Narssarssuaq; (20) Ammassalik.

putting “the people” back into our studies of broad social, political, and economic trends. This volume has two basic aims. First, we want to begin to fill a gap concerning descriptions and analyses of women’s and men’s prehistoric, historical, and contemporary roles in northern Native communities. Second, we wish to present anthropologically comprehensive discussion and analysis of past and present gender roles and relationships. Our book casts a wide net of current research on gender issues, with a vast geographical and temporal breadth. The research sites range from Greenland to the Canadian north, Alaska, and into the Pacific Northwest. Chapters include investigations of pre-Christian child and infant mortuary behavior, nineteenth-century Deg Hit’an household organization, archaeological material patterning and emergent social difference in a prehistoric Thule whaling village, and a contemporary Alaskan Yupik woman’s navigation of her life amid a rapidly changing cultural and economic landscape.

MANY FACES: AN INTRODUCTION

GENDER AND ANTHROPOLOGY Since the last third of the twentieth century, anthropologists have produced a staggering array of research focused on gender. But relatively little of this outpouring is attributable to scholars’ analytically addressing gender dynamics among the indigenous people of northern North America.1 Furthermore, the comparatively meager number of published northern works does not fully reflect the possibilities for synchronic and diachronic anthropological contributions from such a broad and diverse region.2 At the outset, we must discuss what we mean by gender. Clearly, gender analysis has entered the mainstream of anthropological thought.3 As others before us have stressed, we also want to get readers past the idea that female equals gender, because they are not the same (Moore and Scott 1997). Gender studies are ultimately about people and the social significance, performance, and marking of gender. The cultural meaning of gender can change whether a person is elderly, disabled, a child or an infant, a woman or man, or a member of any other chosen social category. Moreover, the past no doubt contains the same degree of diversity the present holds (and maybe more). Gender is not just about females and males, who peek out from the data in selective studies; it is not just about sex roles but about relationships. Ideally, it is about complex interpersonal interactions rather than two-dimensional dichotomous stick-figure people (who behave like we do because, let’s face it, they are often tacitly modeled on our society’s standards of how the sexes act). We believe gender studies should consider human values and emotions, loyalty and rivalry, triumph and tragedy, love and fear, success and selfexpression, desire and competition. That is, gender should examine the limitless ways humans in different cultures choose to create, re-create, and change their personal identities and their interactions within (and between) their societies and environments in meaningful ways. Imagine the possibilities if we invested more into putting faces (Tringham 1991) to the lives that we, as anthropologists, want to understand. Having said this, a certain danger is inherent in gender research (especially in analogy-founded archaeological interpretations) in foisting today’s patterns on the past, a warning initially sounded by Margaret Conkey and Janet Spector (1984). This stasis can create what Susan Kent (1998:18) aptly characterized as a gender that is “timeless,” denying recognition of differently constituted gendered categories and changing roles and power relations over time (Lepowsky 1993). In fact, Brumbach and Jarvenpa (Chapter 12, this volume) caution us that “we

LISA FRINK, RITA S. SHEPARD, AND GREGORY A. REINHARDT

may be on Arctic ‘thin ice’ when we employ Euro-American frameworks of gender and gender roles for interpreting other people’s lives.” For instance, although Arctic ethnographers have analyzed gendered assumptions (Ackerman 1990a; Bodenhorn 1990; Guemple 1995), few researchers have tested the somewhat “petrified” model of northern Native women’s and men’s complementary but equal roles. This book, however haltingly, endeavors to address the complex diversity that makes up synchronic and diachronic social, economic, political, and material gender relationships in the circumpolar north. In addition, this volume contributes in large measure to the basic understanding of the position of women in Native northern societies, a relatively neglected topic in northern research. Indeed, Laura Klein and Lillian Ackerman (1995:3) claim that “silence surrounds the lives of Native North American women.” Our authors give fresh voice to some of these women past and present, revealing their vital individual, familial, and communal roles; their life stories and modes of generational transmission of knowledge and tradition; and their perspectives and impacts on social, cultural, economic, and technological change. Each author has individual aims in revealing aspects of gender in anthropology. Still, the editors hope to influence readers’ thinking on two salient issues concerning anthropological research in the north. “ANTHROPOLOGY” AND GENDER RESEARCH The first issue is that anthropology as a four-field approach is coming under increasing scrutiny. In this time of division within anthropology, gender provides a rich conceptual focal point from which cultural, physical, archaeological, and linguistic anthropologists can inform one another’s research.4 Our intent is not to deliberate on this debate but to demonstrate the convergent position gender research continues to offer anthropology’s four subfields. This effective interdisciplinary approach has a long tradition in gender studies (di Leonardo 1991; Hager 1997; Reiter 1975; Rosaldo and Lamphere 1974). The data and analyses resulting from sociocultural, biological, and archaeological gender studies will enhance, enlighten, and caution each other, as well as benefit from discipline-wide critical review. This book’s materially focused chapters uncover aspects of diversity and complexity that make up the past and consider dimensions of present behaviors that can be integrated into archaeological interpretations. All of the authors recognize the difficulty of finding gender in the archaeological record at a fine-grained level. For instance, Reinhardt cautions us that, although discovering “female” and “male” house-floor

MANY FACES: AN INTRODUCTION

sides may appear methodologically straightforward (based on artifact distribution), it is in fact rife with difficulties in “puzzling” spatial relationships among material remains and the roles that may be ascribed to them. Hoffman and Whitridge illuminate change, agency, and the utility of gender-focused research in archaeological theory building. They challenge us to understand the social issues embodied in seemingly simple objects, showing that material items can signal countless (if not disparate) relationships and active social and economic shifts through time. As Whitridge emphasizes, gender relations may be the seed from which emergent inequality grows. Shepard and Tobey steer their examinations to include households as well as families, reiterating that both can be diverse, fluid, and clouded. They point out that lines dividing men’s and women’s roles, their association with a distinct space, or their daily activities within a household can be very blurry. What is more, children and elders need to be integrated into the study of household organization and production. Adding to the complexity that goes beyond just gendered women and men, Frink introduces readers to the profound resource management and economic contribution of Native elder women. Going further, Crass encourages us to incorporate infants and children into our analyses. These types of multifaceted and diachronic considerations of gender relations and change should be seen as a valuable and useful comparative data set for all anthropologists. To appreciate the complexity in archaeological communities, we must continue to enrich our models of the past and become more aware of the broad variability of gender patterns among contemporary gendered individuals, families, and communities. The cultural anthropologists in this volume—Ackerman, Jolles, and Stewart—invite us to value the intricate nature and emergent patterns of people’s lives today. Stewart questions both the dichotomous representations of gender and the incorporation of a third gender category, for the Netsilik kipijuituq does not comfortably fit into this gender paradigm. Jolles, relying on one indigenous woman’s life history, accentuates the tremendous changes that have accompanied the relatively recent contact history in Alaska, emphasizing the ability of individuals to engage and respond actively to sometimes dramatic social and economic shifts. In addition, Ackerman extends gender and equality investigations to the community level. Her work reinforces the idea that indigenous gender relations (and ideology) can endure (as well as mitigate and mediate) tremendous economic and social change. These researchers

LISA FRINK, RITA S. SHEPARD, AND GREGORY A. REINHARDT

impel us to accept the profound complexity of experiences in the present day and so provide an additional guide in understanding and questioning the archaeological record. Finally, this interdisciplinary approach allows us to continue to create models of past and present gendered social behavior and relations and to answer some thorny questions in cultural, biological, and archaeological anthropology. Who or what determines someone’s gender, and how do we identify possible markers? Was European contact a detriment or a benefit to indigenous women? Do changes in living space imply changes in social structure? Does ideology prescribe or describe living space? Does women’s production within the household indicate their subjugation or their power? Can archaeological activity areas and tools accurately identify women’s and men’s patterned actions? How can we understand human-environment interactions within the framework of variable and shifting roles in modes of production and social and economic relationships? Is gender at the heart of emergent inequality? These are the threads of questions and musings, relevant to many anthropologists, that weave through this volume. NATIVE PEOPLE AND ANTHROPOLOGICAL RESEARCH The second issue we underscore relates to continuity. Three things combine to make Native northern North America an optimal location for anthropologists to examine modern, historical, and prehistoric representations of gender in action: (1) the late contact (from mid–eighteenth to late nineteenth century) between indigenous people of the far north and Europeans, (2) the fairly well-preserved archaeological record (as a result of prevalent permafrost and minimal development), and (3) the history of cooperation between Native peoples and anthropologists (Cruikshank 1990; Mather 1995; Morrow and Schneider 1995). Thus the modern anthropological data and historical records generated in this vast region can provide details not readily available in other environs. Moreover, many of these Native communities have adaptively retained some of their economic and social institutions, such as subsistence pursuits and community celebrations. As Brumbach and Jarvenpa (Chapter 12, this volume) observe, the activities and pursuits of “both women and men in contemporary northwestern communities are highly visible.” This visibility allows archaeologists to expect reasonable success in analogical model building. Besides that, it offers an especially useful tool for analyzing cultural resources (such as oral history) and testing archaeological models of social development. We believe studies of contemporary behavior, combined with

MANY FACES: AN INTRODUCTION

the relatively recent connection to the past, provide a fairly sturdy translational bridge from which we can try to “see” past societies and in particular to understand and test modes and directions of change. In addition, archaeology in the far north can offer cultural anthropologists (as well as those in other disciplines) an invaluable diachronic model of gendered behavior, a resource these anthropologists seldom appropriately recognize and employ. For instance, first contact with Europeans led to few social or cultural changes among many Native peoples for several decades, although new material goods appeared in settlements almost immediately. It seems some Native North Americans revised their living spaces by integrating new tools and technologies into their old lifeways while retaining both their physical and social structures (houses and communities). The collection of new objects in traditional places indicates enduring cultural and social conservatism. Such a model of social action can be used and tested for comparative analytical and theoretical purposes (see Chapter 5, this volume). AVENUES TO UNDERSTANDING GENDER Although our strategies vary, we do not seek to reinvent the methodological or theoretical wheel. One of our aims is to present data and analyses that can and must be integrated into mainstream anthropological currents. Doing so will throw light on the complicated nature of social, economic, and material relations in past and present northern Native communities. This volume begins in the ethnographic present, considering contemporary issues of gender identity and change. In his captivating portrayal of Canada’s Netsilik kipijuituq, Stewart informs readers about the social production, performance, and perception of self and gender. His research serves as a caveat throughout this book, a warning that there are certain pitfalls to relying on Western dichotomies, given that the kipijuituq does not quite fit typical anthropological notions of a third category such as the Native Californian two-spirit (Hollimon 1997). Furthermore, in her case study of the Northwestern Plateau’s Colville Indians, Ackerman investigates an anthropological paradigm first posited by Eleanor Leaco*ck (1978) that sexual inequality necessarily follows participation in an industrial economy. According to Ackerman, this Native Plateau community is steadfast in its ideology that women are valued for their productive contributions and remain the “backbone” of the economy. Finally, Jolles takes us on a life history journey with an Alaskan Yupik woman, including the familial (and community) shift from use of

LISA FRINK, RITA S. SHEPARD, AND GREGORY A. REINHARDT

semisubterranean structures to aboveground houses. She uses oral history and personal testimony to chart this Native woman’s extraordinary accomplishments and ordinary daily activities in order to help us understand the challenges facing an individual in the midst of dramatic economic, social, and material changes. Many social issues broached by the ethnographers (e.g., impacts of missionization on Native domestic life, gender marking and labor, materiality, and culture change) are further addressed in the work of the archaeological anthropologists. Both Shepard and Tobey use ethnohistoric data to assess household and community organization and culture change resulting from missionization. Shepard suggests that the position of women and men in western Alaskan Koyukon Athapaskan households underwent a fundamental transformation when belief systems changed in the mid- to late nineteenth century. The changing organization of household space helps to identify these shifts in social relations. Tobey also investigates the likely effects of missionization on household and community organization among the Deg Hit’an Athapaskans of southwestern Alaska. She suggests that the anthropological conception of a “household” must incorporate the full range of active members during different daily routines. She further believes Native spatial, social, and economic patterns were significantly disrupted (and likely challenged) during the American missionary period. An issue that winds through several of the chapters concerns women and the meaning of their productive activities. Frink further develops previous anthropological inquiries concerning Native women and modes of labor and production. She integrates interviews of western Alaskan women with participant-observation studies at a subsistence fish camp. Frink demonstrates that women are not just passive processors but are essential both to production and control and as critical resource managers—patterns that ought to impact archaeological model building and interpretation in this region. Crass, Hoffman, Reinhardt, and Whitridge analyze material culture as a route to understanding gender identity, the production of place, and inequalities between women and men. As Ackerman (Chapter 3, this volume) points out, a community’s interest in the well-being of women and children can inform gender equality. Thus the “visibility” of children should also be a basic concern of archaeologists interested in past social systems. Crass reviews and analyzes 305 pre-Christian Inuit and Eskimo infant and child interments, discussing the differential postmortem treatment of infants, children, and adults. She then considers the possibilities for interpreting these burial discrepancies.

MANY FACES: AN INTRODUCTION

By using experimental replication with Unangan (Aleut) schoolchildren and incorporating data from the Agayadan village site located on Unimak Island in the eastern Aleutians, Hoffman investigates the enigmatic shift from eyed to grooved bone needles. He explains how late prehistoric shifts in needle manufacture may signal gendered relationships, particularly women’s active involvement in social, economic, and material change. According to Hoffman, Native women likely intentionally altered the design of needles to facilitate producing finely made parkas, considerable status symbols and trade items in the changing Unangan political economy. Extending the household analysis approached by Shepard and Tobey, Reinhardt reanalyzes one interpretation of a “frozen family,” a prehistoric North Alaskan Thule household in which the occupants succumbed to an encapsulating ice flow. The excavated objects were subsequently used to posit female and male “sides” within this prehistoric house. In his thorough reconsideration of the data, however, Reinhardt alerts us to the dilemmas that arise when trying to assess the relevance of spatial data and gender attribution in archaeological interpretations. Finally, by focusing on household activity areas and material remains, Whitridge challenges readers to think more deeply about “egalitarian” and “simple” household and community organization and social change. His interpretation of archaeological data at the prehistoric Classic Thule site of Qariaraqyuk in the Central Canadian Arctic connects the ways materials are used and activities take place within a particular space and thereby identifies economic and social jockeying between the sexes. Whitridge explores the emergent inequalities in this small-scale huntergatherer community and calls for archaeologists to consider an analysis that may “logically” precede more complex, less egalitarian systems. In the final chapter, Brumbach and Jarvenpa contextualize each of the contributions in this volume. They elucidate the critical nature of gender research in the American region of the circumpolar north and emphasize that this research will encourage fresh approaches and new lines of inquiry. Beyond that, they further develop the cautionary tale woven through this volume, that gender attribution and the interpretation of the “female-male nexus” are filled with biases and traps. They warn that unevaluated gendered associations may confound rather than clarify the many ways individuals and communities live and impact their social, economic, and material lives. The editors and authors of Many Faces of Gender hope our book contributes to fuller understanding of the endless variations of social and cultural life found among indigenous peoples of the far north. We

LISA FRINK, RITA S. SHEPARD, AND GREGORY A. REINHARDT

encourage other anthropologists to look more broadly at the diverse roles and relationships in the communities they are studying and implore archaeologists to remember that people—people with faces—once lived at the sites they excavate and once used the artifacts they now scrutinize. We want these chapters to be enlightening and constructive, and we hope they will motivate our readers to delve further into the ideas we have raised.

MANY FACES: AN INTRODUCTION

I CONTEMPORARY RESEARCH

2 KIPIJUITUQ IN NETSILIK SOCIETY CHANGING PATTERNS OF GENDER AND PATTERNS OF CHANGING GENDER HENRY STEWART

GENDER STUDIES, dating back to the late–nineteenth-century suffragist movement, underwent great change in and after the 1960s. Recent gender studies—variously referred to as the second feminist movement, second wave gender studies, or the radical feminist movement—fostered many cultural anthropological studies of gender, biological sex, and sexuality. At first these studies tended to center on Europe, the United States, and other regions of Judeo-Christian influence. Recent reevaluations of the berdache in North America since 1980, however, and other examples of transvestism, hom*osexuality, gender change, and gender role change throughout the world, have raised questions concerning the universality of European (Judeo-Christian) concepts of gender, biological sex, and sexuality (Bodenhorn 1990; Callender and Kochems 1983; Davis and Whitten 1987; Fulton and Anderson 1992; Gilmore 1993; Guemple 1986; Jacobs, Thomas, and Lang 1997; Lang 1998; Morris 1995; Murray 1994; Robert-Lamblin 1981; Saladin d’Anglure 1990, 1993, 1994a, 1994b). In these studies there is general agreement that the dichotomy of man/woman, either as gender or as sex, may not be a universally applicable model. For example, the existence of a third gender among the Inuit (Eskimo)1 and other non-European societies (Saladin d’Anglure 1994b; see also Murray 1994) is one aspect of an ongoing reinterpretation of a Western folk model of gender in Northern studies.

HENRY STEWART

Here I shall discuss the kipijuituq in Netsilik society, a newborn male child raised as a female, and how the kipijuituq may reflect on EuroAmerican concepts of gender and biological sex. Data used in this chapter are primarily those I gathered in the hamlet of Pelly Bay, Nunavut, Canada, from 1989 through 1996, in addition to supplementary data from the literature. Based on my observations, it appears that the kipijuituq is unique and differs from other examples of gender change known for the North American Inuit/Yupik. For example, gender change in the case of the kipijuituq apparently did not serve a particular communal social function, as did other examples of gender change in the Arctic. Following a review of Inuit sex and gender in the literature, I propose that in the Netsilik and other Inuit societies ideally there may have been one “sex” and three or possibly more “genders.” THE KIPIJUITUQ In the Netsilik (Arvilingjuarmiut) society of Pelly Bay in Nunavut Territory of Canada, certain biologically male (in the scientific sense of the term) children are brought up as females until they take a prescribed animal, usually at age fourteen to sixteen.2 Such children are subject to various taboos (a state referred to as tiringnaqtaq), foremost of which is that the child’s hair must not be cut before catching a prescribed species of game.3 BECOMING KIPIJUITUQ The decision of whether a newborn male child should become kipijuituq is in many cases made by the grandparents, mainly in the following ways. First, if the namesake of the infant is (was) kipijuituq, the child automatically becomes kipijuituq. Second, the grandmother or grandfather talks to (hanaurajuk) the infant, and by observing the child’s reaction to those words, the grandparent may know what course of life, or destiny, is in store for the infant. An infant, named either in utero or when it is born, takes on the personality and disposition carried in the name (Rasmussen 1931:219) and thus is believed to be born with a fully developed personality. Knud Rasmussen (1931:259, 505), discussing Netsilik naming practices, wrote that the child becomes the name(s) spoken when it leaves the mother’s womb. This latent personality becomes manifest as the infant grows. It is said that by observing the reaction of the infant thus endowed with a full-fledged personality, it is possible to determine whether the infant is destined to become kipijuituq. Male infants judged to be kipijuituq become socially female: they are dressed in female clothing, expected to act as girls, and are referred to by female kinship terminology. They do

KIPIJUITUQ IN NETSILIK SOCIETY

not, however, take on all elements of culturally defined feminine gender roles. That is, a kipijuituq may play with dolls and otherwise behave in a feminine manner but is not taught sewing, cooking, and other traditionally feminine activities. In the event that an infant is judged to be kipijuituq, that judgment must be obeyed by the child, its parents, and kinspeople. It is (was) believed that if that judgment were ignored, the infant and its kinspeople would be visited by misfortune, such as a hunting accident or a poor catch. Many restrictions apply to the activities of the kipijuituq and to how members of the community treat the kipijuituq. The most stringent restriction is an admonition against cutting or trimming the hair. As mentioned earlier, the hair must be allowed to grow naturally until the first catch of the game animal prescribed by the grandparents. If this admonishment were disregarded, misfortune would be visited upon the kipijuituq and, in some cases, also upon its kinspeople. Such misfortune might include poor hunting, health problems, or an accident during subsistence activities. Other taboos include prohibitions against eating certain game or portions thereof. Several years ago one kipijuituq prematurely cut its hair and was severely chastised by its parents and grandparents. B of Pelly Bay was a kipijuituq until his first catch.4 I first noticed tenyear-old B in 1989, but it was several seasons later before I realized B was a male. For several seasons I thought B, always demure and unobtrusive, with hair done up in two long braids, was a female. Because persons in a kipijuituq condition have become rare in recent years, B said he was at first embarrassed by his situation.5 By the time he reached age eight, he became used to being kipijuituq and ceased being embarrassed. Having to use the girl’s toilet at school was the admonition that troubled him most. B took a seal when he was ten years old, but because his grandparents felt a seal was insufficient in B’s case, it was not until he was fifteen and took a polar bear in 1994 that he was released from the kipijuituq condition. TRIMMING THE HAIR Informants were unable to explain why so much importance is attached to the admonition that a kipijuituq must not cut or trim the hair, but an excerpt from an ethnography of another Inuit society suggests a possible explanation. Gustav Holm (1914) noted that among the Angmagsalik (Ammassilik) of eastern Greenland the hair is worn long and has never been cut, it being regarded dangerous to lose any of one’s hair (Holm

HENRY STEWART

1914:32). When Holm asked for a lock of hair for research purposes, the Ammassilik usually said their father had told them they were never to cut their hair or they would die. Holm (1914:86–87) surmised that they were afraid that an ilisitsok (exorcist or witch) might use the hair in a tupilek, an animal effigy of bad omen with power to cause death (Kleivan 1984:619). William Thalbitzer stated that Ammassilik men almost always had long, uncut hair (1912:600). According to informants, the fact that some Netsilik shamans did not cut their hair to retain vitality may indicate a similar belief in the Netsilik society. RELEASE FROM THE STATE OF BEING KIPIJUITUQ The most common way a person is released from the kipijuituq condition is by taking the prescribed game animal. If the person(s) judging that an infant is to be kipijuituq states that a caribou or a seal is the suitable game, taking an animal of that species can release the person from the state of being kipijuituq. In some cases, if so specified by the grandparents or namesake, a kipijuituq may be obliged to take several kinds of game or more than one of the same species. For reasons I discuss later, the polar bear seems the most potent game for release. There is another situation by which the kipijuituq state is terminated in the Pelly Bay society. That occurs when on their deathbed the kipijuituq’s grandparents, parents, or namesake leave testament that the kipijuituq need no longer be bound by the taboo against cutting his hair. At that time the child becomes male by gender and is no longer considered a kipijuituq. Traditionally, when the kipijuituq took the specified game animal or was freed from his tiringnaqtaq (tabooed) state by a deathbed injunction, all the people in the camp gathered for a celebration.6 During the celebration all adult females gathered in one house (tent or snow house), and the men formed a single line in front of the entrance. Women came from the house one at a time and kissed each man in the Inuit fashion (kunigatuk), holding the man’s face in her hands. A woman did not kiss men of her own kin group (ilagiimariktut), however. As each woman came out she grabbed the former kipijuituq from the preceding woman and kissed its face. After being kissed by all of the women, the boy was completely released from being kipijuituq. Following this ceremony a drum dance was held for the former kipijuituq. In recent times the kissing and drum dance ceremonies have not usually been performed, and they were not performed in the case of B. After taking a polar bear, B cut his hair in the presence of his grandparents and was released from taboo restrictions. He seemed happy to

KIPIJUITUQ IN NETSILIK SOCIETY

be a normal boy and, at the same time, seemed proud of having undergone the experience. While a kipijuituq, B did not seem to receive special treatment from his cohorts. It is difficult for an anthropologist such as myself who is over fifty to commune with young teenagers, but I was able to observe interaction between B and his best friend Keith (pseudonym), the youngest boy of the family with whom I stayed in Pelly Bay. B visited often, and the two boys joked and played together as other teenagers. I was struck, however, by the fact that B never joined in contests of strength and wrestling or other Inuit male games that took place often in the house and at camp. It appears his female gender restricted participation in such male activities, although he hunted like other young males. The kipijuituq is becoming a rare, possibly discontinued social phenomenon in Pelly Bay. Informants told me this was because of the influence of Christianity. My impression is that television, projecting forceful images of gender roles corresponding to biological sex (in the Western sense), is also a possible factor. I have seen many young people in Pelly Bay model their appearance and actions on images of gender portrayed on television and in videos. Many studies have been conducted on the effect of television on Inuit/Yupik society since concerted broadcasting commenced in 1980 (e.g., Alia 1999; Koebberling 1999; Stenbaek 1999; Watson 1980), especially its influence on language (Graburn 1982). Although Richard Condon discussed in detail the influence of television on the behavior and values of Inuit teenagers in Holman (Condon 1987:125–128), his discussion of gender and sexuality (Condon 1987:132– 143) does not refer to the influence of television on Inuit/Yupik views of gender, sex, and sexuality. Laila Sorenson’s (2000:174) apt observation on the role of television in conveying, formulating, and transforming beliefs, values, and myths guiding modern perceptions, however, is without doubt applicable to arguments concerning the influence of television on Inuit/Yupik society. GENDER CHANGE IN THE ARCTIC Many facets of the kipijuituq phenomenon have yet to be recorded and interpreted. For example, there are conflicting statements as to whether a female infant could be kipijuituq. Based on my observations at Pelly Bay, it appears that females do not become kipijuituq, although for different reasons they may be classed as tiringnaqtaq (under a special state of taboo). As I discuss later, many researchers have recorded instances of Inuit/Yupik female children being male gendered until coming of age or, in some cases, throughout their lifetimes (Lang 1998:165–166, 186,

HENRY STEWART

302). Some of these instances may parallel the kipijuituq, but no data exist to either support or repudiate such a supposition. The kipijuituq differs from the 148 examples of the berdache7 recorded for North American indigenous peoples (Callender and Kochems 1983; Fulton and Anderson 1992; Lang 1998:4–5) and also from the catamite reported from the Aleutians. In the Aleutians the catamite (shupan) acts as a sexual partner for renowned hunters or powerful leaders (Musk-Ox 1994) but remains male in gender. As far as I was able to ascertain, the kipijuituq did not participate in sexual acts with adult males but was only treated socially as a girl before menarche. A review of the literature reveals several instances of gender change among the Inuit and Yupik.8 Sabine Lang (1998:4–5), in her definitive study of gender change in North America, has gleaned from the literature instances of gender change in Inuit/Yupik society from southwest Alaska to Greenland: southwest Alaska (Chugach, Koniag [Alutiiq]), Kotzebue Sound (Malimiut), St. Lawrence Island, Quebec, Cumberland Sound, and southeast Greenland. Lang (1998:281) stated that male gender change (woman-man) has been reported only for western Alaska, although her own data (Lang 1998:78, tables 2, 4, 5, 6, 7, 8) seem to contradict this statement. It is not certain whether the notion of gender change applies to the Chugach. According to Kaj Birket-Smith, in Chugach society no known cases exist of boys raised as girls or of women who lived or behaved like men, but a half man/half woman (aranu’tiq) appears in historical traditions. The aranu’tiq are said to be male on one side and female on the other; they performed the work of both sexes and were considered more skilled than ordinary persons (Birket-Smith 1953:94). Although BirketSmith refers to the aranu’tiq as a transvestite, in the Chugach intellect the aranu’tiq more aptly refers to a third gender, not an either-or category. Women-men (anasik) on St. Lawrence Island were considered to be especially powerful shamans, and in fact all anasik were shamans, although all shamans were not necessarily anasiks. In the case of the anasik it appears that gender change was a means of becoming invisible to malevolent spirits, possibly an influence from the Chukchi (Lang 1998:165–166). Lang (1998:261) also mentions men-women in the St. Lawrence society, as well as among the Mahlemut (Malimiut) of the Kotzebue Sound coast. Because of geographical proximity and a common historical background rooted in the Thule tradition (McGhee 1984b:376, 1996:211), examples of gender change in Greenland and the eastern Canadian Arctic

KIPIJUITUQ IN NETSILIK SOCIETY

may likely have more relevancy to the kipijuituq, so I shall discuss those examples in greater detail. In her study of exchange in the Ammassilik society of southeast Greenland, Joelle Robert-Lamblin (1981) notes that not only were girls raised as boys (tikkaalia), that is, made into boys, but also that boys were brought up as girls (nuliakkaalia), that is, made into girls. Several reasons are cited for reversing gender roles, one being that when a newborn was not of the sex the parents desired, newborn children were raised in the gender opposite their biological sex. Thus a boy might be raised to fill a woman’s position or a girl to be a hunter (see also Holm 1914:67). Or if a daughter (son) died prematurely, the next-born daughter (son) might be raised in the gender role of the dead child. Concerning this transformation, Robert-Lamblin (1981) does not cite the metaphysical reasons known for the Netsilik kipijuituq but theorizes that this practice was intended to remedy sex-ratio disequilibrium within a family. This social device may have functioned as a substitute for infanticide as practiced among the Netsilik (Remie 1985). To the best of my knowledge, however, the kipijuituq did not function as a sex-ratio equalizer. Another instance of gender change in east Greenland is the piaarqusiaq (piaaqqusitat), a young boy or girl dressed in a ridiculous manner and wearing a specially made cap (Ostermann 1938:191; Thalbitzer 1912:588, 1941:602). The piaarqusiaq is a child whose brother or sister died in infancy and for this reason requires special treatment to reach adulthood safely (Ostermann 1938:191; Thalbitzer 1912:588). The special clothing is thought to make the child invisible to hostile spirits (Ostermann 1938:191; Thalbitzer 1941:604). In this way, the piaarqusiaq is a practice to ward off evil spirits and as such differs from the nuliakkaalia and tikkaalia, a practice said to equalize gender balance and assure division of labor according to gender. An example of gender change to avoid a hostile spirit is cited by Bernard Saladin d’Anglure (1994b:98) for the Quebec Inuit. In this case a person ordered by the group to execute a dangerous camp member became afraid the soul of the executed person would seek revenge and became a transvestite to evade that hostile soul. An example of sex crossing is found in traditions recorded by William Thalbitzer (1923:511–516) from east Greenland and Franz Boas (1901:248, 323) from Cumberland Sound and Hudson Bay. In these traditions, an old woman took to living like a man and married her daughter-in-law. In the Greenland version the old woman changed into a man and procreated as a man.

HENRY STEWART

Saladin d’Anglure has also recorded instances from the Quebec Inuit society of girls dressed and socialized as boys (1994b:91), a boy dressed alternately as a girl and boy every other day (Saladin d’Anglure 1994b:94), and a boy taught both sewing and hunting and dressed in the coat of a boy and the pants of a girl (Saladin d’Anglure 1994b:92). The reason given for these gender changes was that the sex of the child’s namesake differed from the sex of the child; therefore the child was gendered according to its name. In all of these examples, the transvestite girl or boy eventually had to make the transition to his/her biological sex. In the case of a boy, the transition to his biological sex was accomplished when he killed his first game animal (Saladin d’Anglure 1994b:97), at which time he abandoned his feminine clothing. It is not certain whether this phenomenon corresponds to the Netsilik kipijuituq, as there is no mention of taboos associated with the Quebec transvestite. The Quebec example, however, is similar to the Netsilik kipijuituq in that a boy’s transition to his biological sex was effected by his first catch. DISCUSSION As I pointed out earlier, the kipijuituq most commonly makes the transition to the male gender by taking a polar bear. The role of the polar bear in gender conversion may be rooted in the Inuit worldview. In Inuit creation myths, according to Saladin d’Anglure, at the beginning of the world humans and other animals were undifferentiated and thus were able to mutually metamorphose. In the mythical world, of all the animals the polar bear was most closely associated with humans. This close association is evidenced in Inuit mythology as follows: (1) at the beginning of the world, a sterile woman came to possess a polar bear cub and brought it up as a human (Saladin d’Anglure 1990:179); (2) polar bears also stand up on two legs; (3) polar bears eat the same things as humans (marine mammals and fish); (4) they hunt in the same manner as humans; (5) like humans, polar bears can travel either on the sea ice or on land; and (6) they live in a winter shelter similar to an iglu (snow house) (Saladin d’Anglure 1990:183). Another aspect of Inuit creation myths relating to the importance of the polar bear is the allusion to antagonism, or opposition, between human females and polar bears. This allusion suggests that the polar bear symbolizes males as well as sexual desire (Saladin d’Anglure 1990:180, 184) and thus may be the most suitable first game for the kipijuituq transition to male status. Also, in mythology the polar bear is portrayed as the original male and as such is considered to occupy a status equal

KIPIJUITUQ IN NETSILIK SOCIETY

to humans, males in particular, and is seen as a supermale with the power to fertilize sterile women (Saladin d’Anglure:1990, 1994a:208). In some cases, the polar bear is depicted as a kinsperson and a rival of high respect. Thus by taking a polar bear, which symbolizes reproduction and vitality, the kipijuituq would be imbued with vigorous vitality and hunting ability and would be blessed with numerous descendants (Saladin d’Anglure 1990:186, 189). This worldview is one plausible explanation of why, by taking a polar bear, the kipijuituq may become a male and how in many cases a young boy becomes recognized as an adult. As mentioned before, B had taken a bearded seal when he was ten, but because his grandparents had directed that the state of being kipijuituq would not terminate until he took a polar bear, only then was he able to make the transition to the male gender. The concept of sex and gender in Inuit societies also needs to be discussed. Pelly Bay informants state that the fetus is male and that the conduct and condition of the mother-to-be at the time of childbirth determine whether the fetus will be born a female or a male. Saladin d’Anglure (1994b:84), based on his research in Quebec, also stated that from the time of birth, neither sex nor gender is fixed. Saladin d’Anglure wrote that the embryo always begins as a male, but during pregnancy or under certain conditions during childbirth it may be born as a female. If the mother breaks a taboo during pregnancy or experiences a difficult delivery, the penis of the male fetus will audibly rupture and change to a vulva. Such a fetus will be born as a female and is called a sipiniq9 (Saladin d’Anglure 1994b:84, 94). Sex change of the fetus is often accompanied by signs such as a cracking sound when the vulvar fissure is formed and the penis retracts, or a cl*toral hypertrophy (Saladin d’Anglure 1994b:85). According to one Pelly Bay informant, the male fetus holds on to its penis when coming down the birth canal. In the event of a difficult birth or other extenuating circ*mstances, however, if the male fetus lets go of its penis, the newborn will be a female. Such beliefs may be rooted in Inuit legend. In his volume on the intellectual culture of the Iglulik Inuit, Rasmussen (1929:252–253) recorded these stories: Women appeared on earth. They came from hummocks of earth: they were born so. They were already fully grown when they emerged from the ground. They lived together as man and wife, and soon one of them was with child. Then the one who had been husband sang a magic song:

HENRY STEWART A human being here A penis here. May its opening be wide And roomy. Opening, opening, opening! When these words were sung, the man’s penis split with a loud noise and he became a woman, and gave birth to a child. . . . Also, women grew up from a hummock of earth. They were born and fully grown all at once. And they wished to have children. A magic song changed one of them into a woman, and they had children.

For the Netsilik, Rasmussen (1931:208–209) recorded this story. That was the time people lived in darkness, in the very first beginning, when there were only men and no Woman was made by man. All the animals died, and there were only two men left. They lived together. They married, as there was nobody else, and at last one of them became with child. They were great shamans, and when the one was going to bear a child they made his penis over again so that he became a woman, and she had a child.

The Ammassilik have a similar belief. Pierre Robbe (1981:74, quoted in Saladin d’Anglure 1994b:86) noted that the Ammassilik believe all fetuses are masculine and that the fetus holds his penis until birth. At that time if he wants to be a girl he does not hold his penis as he moves down the birth canal, and the penis is transformed into a vulva, resulting in the birth of a girl. Many legends also tell of change of sex by individual volition (cf. Rasmussen 1929:158–159, 302; Thalbitzer 1941:511). CONCLUSION In this chapter I have discussed the kipijuituq, which to my knowledge has not been heretofore reported in the literature. Other examples of gender change in the Arctic literature are reported to relate to gender/ sex equilibrium (Greenland), a tactic to avoid malevolent spirits (St. Lawrence Island, Greenland, Quebec), dual sexuality by birth (southeast Alaska), spiritual powers (St. Lawrence Island, Kotzebue Sound), or a replacement of a child who died prematurely (Greenland). None of these reasons applies to the kipijuituq, except that when the namesake of a newborn male is or was kipijuituq, the newborn male necessarily becomes kipijuituq. The kipijuituq differs from gender change in Quebec as a result of one namesake, however, in that (1) in Quebec, gender change

KIPIJUITUQ IN NETSILIK SOCIETY

arising from one namesake applies to both male and female infants (Saladin d’Anglure 1994b), but only male infants become kipijuituq; (2) the kipijuituq always changes back to the male gender; and (3) the kipijuituq is not recognized as having special powers. Although being tiringnaqtaq, or subject to special taboos, leaves the kipijuituq vulnerable to misfortune or visitation by evil forces, this condition may apply to any person and is not unique to the kipijuituq. What, then, was the function of the kipijuituq? In spite of repeated questioning and observation, I was unable to establish a communal function for the kipijuituq. Negative evidence leads me to surmise that birth conditions may deprive a male infant of his male completeness or full male potential. That is, what should have been born a girl came out as a boy, and raising that child as a girl until the age at which it should attain manhood allows the child to be transformed back into a male, imbued with sufficient masculinity to cancel sex/gender ambiguity. A particularly deficient case could be remedied by taking a polar bear, which would imbue the child with supermale power. Although I repeatedly asked why only certain male infants were destined to become kipijuituq, I always received the rejoinder “I don’t know.” When I suggested the reasoning posited earlier, the response was neither positive nor negative, probably because the Inuit refuse to hypothesize or make presumptions concerning suppositional situations or matters of which they lack intimate knowledge (e.g., Briggs 1969:44, 53). A newborn male child is determined to be kipijuituq through the process of hanaurajuk (talking to). Here also, repeated questioning did not reveal what criteria or circ*mstances obliged a male infant to become kipijuituq. The answer to my queries was always that one can know by hanaurajuk. Again, I can only surmise that during hanaurajuk it becomes apparent that the infant is predestined to be kipijuituq. Except when the namesake was a kipijuituq, in which case the newborn child automatically becomes kipijuituq, I was unable to ascertain what criteria or circ*mstances played a part in that determination. It appears that through hanaurajuk the grandparents perceive signs that the child should become kipijuituq. Data from the literature cited earlier, and information provided by informants at Pelly Bay, suggest that in the cosmologies of many eastern Canadian and Greenland Inuit there was only one sex, the male. In the beginning, a man’s penis ruptured and formed into a vulva, or a magic song changed a man into a woman—a transformation making procreation possible. Such a transformation may be the basis for the explanation that the fetus is fundamentally male but may be born a

HENRY STEWART

female if he does not hold on to his penis during parturition. From this I conclude that the Netsilik and other eastern Arctic groups posited a single original sex: the female sex deriving from the original male. There is, however, a problem in this interpretation—namely, that the word inuk (and its plural forms) appearing in myths may refer to both male and human being. Where Rasmussen includes the original text with a translation, inuk is translated as human being (Rasmussen 1929:252). He does not record the original text for his other stories (Rasmussen 1929:253, 1931:208–209), but if consistency in translation can be assumed, inuk in creation myths concerning humankind should mean male (cf. Saladin d’Anglure 1994b:89). In conclusion, I shall address three questions. Is the kipijuituq a phenomenon restricted to the Pelly Bay Netsilik (Arvilingjuarmiut)? The term kipijuituq does not appear in the literature and, as discussed earlier, differs from other examples of gender change in the Arctic. For these reasons I propose that the kipijuituq is one example of gender change widespread throughout the Arctic but is distinctive in its rationale and manifestation. As such, the kipijuituq may be postulated as a unique Netsilik phenomenon. Did becoming a kipijuituq and later acquiring a male gender endow certain males with the symbolic reproductive attributes often associated with females (cf. Bodenhorn 1990)? This question can be answered only with further data. Finally, in the Inuit intellect are (were) there categories or concepts that correspond to gender, biological sex, and sexuality? There are terms for human males (anut) and females (arnaq), but I have not found terms in Inuktitut corresponding to gender, biological sex, or sexuality. In this chapter I have used gender not because it correlates to an Inuit concept but because it is essential to describe an Inuit social phenomenon in English. It is important to remember that gender and associated terms are vehicles to translate Inuit social phenomena into English and are not translations of indigenous concepts. As Henrietta Moore (1994b) has pointed out, social scientific research contests the premise that sex difference is natural and that outside of biomedical discourse, sex is not a universal concept rooted in presocial, physiological parameters. Examples presented in this chapter aptly emphasize the fact that gender is a process rather than a category (Moore 1994b:820). Research is needed to establish whether the concept of gender and sex exists (existed) in non-Western societies. Stated otherwise, we need to reexamine our projection of oppositional dualism as a paradigmatic structure (Kehoe 1997:266; see also Chapter 12, this volume).

KIPIJUITUQ IN NETSILIK SOCIETY

Research and hypotheses concerning these problems are prerequisites to discussions of the third sex/gender or conjectures concerning the ascendancy of the biological male alluded to earlier in this chapter. All of these a priori suppositions implicitly presuppose the pertinence and universality of a two-sex dichotomy in anthropological research. ACKNOWLEDGMENTS I thank Simon Inuksaq (deceased May 1997), Josie Angutinguiniq, Martha Kutjuutiqu (Tunnuq, deceased October 1998), and Guy Kakkiarniun of Pelly Bay for their hours of consultation and patience with my interminable questions. I also express appreciation to Petra Rethman (McMaster University) and Gregory A. Reinhardt (Indiana University), Lisa Frink (University of Wisconsin, Madison), Rita S. Shepard (UCLA), and Masao Kashinaga (Tokyo University) for pertinent comments and advice.

3 GENDER EQUALITY IN A CONTEMPORARY INDIAN COMMUNITY LILLIAN A. ACKERMAN

GENDER EQUALITY HAS BEEN DESCRIBED in several hunting-gathering societies (Draper 1975; Schlegel 1977), but no one so far as I know has described gender equality operating in a society that is part of an industrialized nation. Indeed, Eleanor Leaco*ck (1978) wrote that such equality was impossible in a complex society. She believed gender equality and industrialization were incompatible. On the Colville Indian Reservation in north-central Washington, however, gender equality and industrialization do appear together in the same society. The people of the reservation are descended from Indians who had a fishing-gathering-hunting culture in which the practice of gender equality was not only present but necessary. Women provided at least 50 percent of the society’s subsistence, and their other functions were extremely important in keeping the economy and other cultural systems viable. The modern period of Colville culture is rooted in the Euro-American conquest of the Plateau Culture Area, which is a fairly recent event. Plateau Indians were forced to sign treaties in 1855, consenting to be placed on reservations. When the Colville Reservation was formed in 1872, eleven tribes/bands were eventually located on it, but they shared the same culture; that is, they were all Plateau Indians. Traditional Plateau cultures became very similar as a result of the extensive intermarriage

LILLIAN A. ACKERMAN

required by their incest regulations, which forbade marriage with any known kin (Ackerman 1994:289; Hunn 1990:217). Consequently, tribes and villages far apart geographically often intermarried, evening out cultural differences. This practice is followed to the present day. After the reservation was formed, the people farmed as the government expected them to, but they continued their foraging lifestyle as well as they could with the decreased land area available. This situation continued for some years. A crisis point in cultural change took place in the late 1930s. In 1938 the separate bands or tribes were required to surrender their independent political structure of chiefs and assemblies. These institutions were dissolved—not without a furor—and a reservation-wide tribal council was formed that administered the reservation as a whole. Then in 1939 the construction of Grand Coulee Dam began, and by 1941 it flooded much of the reservation land, destroying areas where wild root crops flourished and also destroying the salmon runs that could not get past the dam to spawn. As salmon provided one-third of the food supply and roots and other plants provided about half, the people were forced to give up extensive foraging and turn almost completely to the Euro-American economy to survive. I will not go into detail here as to how the people adjusted to these changes. Suffice it to say that after great effort and astute management by the tribal council beginning in 1970, the Colville Reservation today is one of the most prosperous in the nation. This may not be saying much, since unemployment is still over 50 percent, but the reservation has been successful in many economic projects and continually plans for more. Today the reservation has organized businesses including lumber mills, logging, grocery stores, gambling casinos, and houseboating for tourists. The tribal bureaucracy provides employment mostly to tribal members to run these establishments and to perform the administrative work on the reservation. In 1986 the reservation injected about $5 million into the economy of northcentral Washington as a result of purchasing materials, paying salaries, and paying taxes. The economy of the reservation, then, is similar to the Euro-American economy in that the reservation staffs offices, manages forests and businesses, maintains roads, plans economic development, and administers health, education, and welfare offices. Nevertheless, the economy remains a somewhat separate system from the rest of the country in that the tribal administration does its hiring by its own rules and runs its businesses based on its indigenous traditions and values.

GENDER EQUALITY IN A CONTEMPORARY INDIAN COMMUNITY

EQUALITY IN THE ECONOMIC SPHERE Because the Colville people have a legacy of gender equality inherited from their traditional foraging culture (Ackerman 1995:78–98), they have achieved without pain that elusive goal in Euro-American society: equal pay for equal work. Men and women not only earn equal salaries for the same work but have equal access to all jobs including management, logging, ranching, and, at one time, mining. I once asked a council member if she was aware that equal pay for women was not the norm in Euro-American society. She replied that everyone on the reservation was aware of it, but the council took as a given that women had to support their families, and the philosophy was that the welfare of their society was dependent on the welfare of children. Thus ideology is a factor in the preservation of gender equality. As noted earlier, women in the past were extremely influential, as they provided at least half of the food supply by gathering roots, berries, and other plants and sometimes capturing small animals. Their work was regarded as the backbone of the economy, recognized by both genders. Today women are still considered the backbone of the economy, still recognized by both genders. Women are expected to support their families today, as they did in the past with their gathering activities. In 1979, when the economy was good, almost 90 percent of women were employed in a formal job. The remaining 10 percent, who had no employable skills or no child care resources, took care of other people’s children along with their own to earn money. Often Colville women are the sole supporters of their children because of the high divorce rate in Plateau societies, past and present. As a consequence, girls as well as boys are raised as potential breadwinners and are equally encouraged to obtain higher education or vocational training. Since the work of both genders is recognized by the society at large as of equal value and importance, a ceremony in the form of a feast persists that recognizes that value. Some wild resources are still available, so when a boy captures his first fish or animal or a girl brings back food from her first independent gathering effort, a feast is held in which the child’s food is completely consumed. Elders particularly are invited to attend the feast because their presence bestows good fortune on the young person. New permutations of this custom have arisen and are observed as well. A girl’s first loaf of bread is given to an elder as a sign of respect. A child’s first powwow prize money is shared with elders. Individuals of both genders commented that women are generally more efficient and reliable workers than men. One male Euro-American

LILLIAN A. ACKERMAN

consultant admired the speed and skill displayed by a group of Colville women who built a road on the reservation. He noted that women generally are more industrious and more thorough about completing a task, and they keep an eight-hour day more conscientiously than men. This modern opinion is reminiscent of the phrase “men don’t work,” which encapsulates the emic or Native perception of the male economic role in the traditional culture. Of course men “worked” in the past by providing fish and animal flesh to the diet, and their defense of the community was risky “work” that was highly valued. Apparently, however, these activities were not regarded as “work.” Male economic activities in the past occupied long hours during a few weeks each season and occurred in strenuous spurts, with leisurely periods in between, whereas women more often were continually occupied with tasks on a daily basis. In my opinion, the traditional economic role of Colville women better prepared them for the eight-hour day and five- to six-day work week required of workers in an industrial society. The traditional pattern, on the other hand, did not prepare men for this schedule. Accentuating this difference is the cultural lack of a “moral imperative” for Colville Reservation men to earn money for its own sake, which equals power and prestige in Euro-American society. The traditional Indian male pattern of intermittent work may account for the observation that women in many North American tribes acculturate more readily than men to modern society (Maynard 1979; McElroy 1979). It is argued that this differential acculturation among the Oglala Sioux occurs because women are able to continue their homemaker roles in contemporary times and thus experience less cultural disruption than men. In contrast, men are completely deprived of their former roles as warriors and hunters (Maynard 1979:12–13). This explanation cannot apply to the Plateau tribes of the Colville Reservation because little continuity has occurred in women’s roles either. Although childbearing and rearing continue, all else is changed. Office employment is as different from gathering and preserving wild foods as lumbering is from hunting. What remains from the past is the ethic that women do what they must to support the family and even provide the major share of support if needed, as they did in the traditional culture. Women’s better adjustment in contemporary times may result from their being accustomed to sustained rather than strenuous intermittent work. It would be interesting to see if other North American tribes have conceptions on the nature of male and female work similar to those in the Plateau and to correlate those conceptions with the differential adjustment of the sexes to the Euro-American economy.

GENDER EQUALITY IN A CONTEMPORARY INDIAN COMMUNITY

EQUALITY IN THE DOMESTIC SPHERE Consultants say that in the past, men were seldom home, since they were away getting food. Indeed, since the annual round often separated a married couple involved in disparate tasks in different areas, they must have been together consistently only in the winter, when the subsistence round was completed, and perhaps in the fall, when parties consisting of both sexes participated in the hunt. Women butchered the game the men brought in and dried the meat. Whether men were home or not, it was considered the domain, or at least the realm, of women. They were considered the owners of all the food in the household; they processed it, distributed it, and used the surplus for trade. Within the home today, often an extended family, Colville women still control the economic resources of the family, as they did formerly. The oldest woman runs the household and makes most of the decisions. Women are considered more economically astute than men and still handle their husband’s income as well as their own. The idea that women control all resources within the household still lingers, even though men are present year-round (see Chapter 7, this volume). The separation of men and women continues to some extent. I observed that men socialized in groups, whereas women socialized in their groups at large social gatherings. When I mentioned this pattern to consultants, they agreed that this is common. Women today usually keep young children after a divorce, although some inequality is beginning to creep in. In the past, no one expected a man to provide for his children after divorce because it was impossible for him to do so. One spouse or the other usually left the community. This caused no economic hardship for the woman and her children, for her work was extremely productive, and it was easy to acquire a surplus for trade. Women could trade their vegetal foods for meat or fish, and they always received meat and fish in communal distributions. Today I have seen elders deride the few women who defy custom and seek support after a divorce. Unfortunately, although most women work, divorced or not, it is harder to support children on only one income in modern conditions. Consequently, single mothers are finding it harder to sustain their families than in the past. A married couple continues to own property separately. A husband and wife will each have their own car. Consequently, if a divorce occurs, the woman, who sought a loan for her car and paid for it out of her income, will automatically own it. Neither a husband nor a wife, even in a long-term stable marriage, will borrow the other’s vehicle without permission.

LILLIAN A. ACKERMAN

Each marital partner pays for different things to run the household. He might pay the taxes and make the mortgage payments, whereas she will pay for the children’s dental work and food. Each partner keeps a separate savings account. The division of financial responsibility is different for each household, but an attempt is made to make it equal. EQUALITY IN THE POLITICAL SPHERE In the political arena, men and women both speak publicly with equal frequency, and they vote independent of each other. I was told that a husband or wife never learns how the other votes in national or reservation elections. In a spirit of experimentation I once informed a young Colville man that married couples in Euro-American society discuss candidates before an election and influence each other on which candidates to vote for. He was incredulous, even scandalized. His reaction indicates how seriously married people view their political independence from their spouses. A problem of women’s access is developing in the political sector. When the tribal council started in 1938, fairly equal numbers of men and women were elected to the council. Around 1970, after the tribal council successfully fought termination of the reservation, the workload for the council increased tremendously, and all members were required to travel extensively. Young women felt the travel particularly was incompatible with their parental duties to their young children. Older women who have older or grown children, however, do seek and win office without trouble. This leads to a situation in which young men sit on the council along with mostly older men, but young women are not present alongside older women. Despite the age imbalance, the gender balance is about equal. The chairperson is currently female (2002). Many consultants say they generally prefer to have women represent them on the council, as women are more aware of problems and will aggressively fight Euro-American demands more readily than men. Women, they say, “have a stronger spirit” and will fight for what they believe, whereas men are trained to be reticent and dignified. DISCUSSION In all aspects of contemporary Colville Reservation culture, gender equality has survived industrialization and modernization, with only the two exceptions noted earlier: the currently threatened welfare of children after divorce, and young women’s limited access to leadership in the political sphere. Overall, however, I would judge that gender equality continues to be a necessary element in the contemporary culture. A

GENDER EQUALITY IN A CONTEMPORARY INDIAN COMMUNITY

significant difference occurs in the nature of gender equality from the past, however. During the foraging period, men and women exercised mostly balanced, complementary rights. In the present they exercise mostly identical rights in every aspect of culture, suggesting that access to identical rights is the only way to have equal rights in an industrial society. Why has female status been reduced in other parts of the country and the world by colonialism and capitalism and yet escaped that fate on the Colville Reservation and, in fact, in the entire Plateau Culture Area? Missionaries were successful in destroying several female leadership positions in the culture, but they were not successful in ranking men and women by, for instance, giving superior value to men’s work. Perhaps women’s work was too important. This may not sufficiently explain the persistence of gender equality, but a few other speculations may provide some clues. Gender equality may have persisted on the Colville Reservation because Euro-American settlers did not appear in the Plateau in large numbers until the 1890s, reducing the time for negative impact on the culture. Further, the eleven tribes on the Colville Reservation were not confined there until 1872. The missionaries may have failed in the area because they had too little time to solidify the culture changes they engineered before the influx of settlers reduced their influence. Moreover, the Plateau people have a strong oral tradition and trace their genealogies five to seven generations back, along with the life histories of most of their ancestors (Ackerman 1994:302). Consequently, they also have a strong cultural tradition, which has not been easily suppressed. Finally, gender equality may have survived because it is not something that a casual observer can see. It is largely invisible to outsiders. With part of Euro-American society trying to win equal rights for women, the existence of gender equality on the Colville Reservation might even be applauded rather than condemned. I have stressed the importance of the economy in this chapter, since so much theory points to it as the source of gender equality (e.g., Sanday 1974). I believe that ideology (Schlegel 1977) also supports the gender equality found among the Colville people; thus equal pay for equal work is a principle among them, as explained earlier. Their ideology extends further. Despite the theoretical universal inferiority ascribed to women by some researchers (Chodorow 1974; Ortner 1974) because of their childbearing, child-rearing, and lactating functions, Colville women are honored by men for those functions. Consultants of both genders say men honor not only their own mothers and grandmothers—important figures

LILLIAN A. ACKERMAN

in a boy’s development—but all women. Men readily confide in women and expect their mothers, grandmothers, and wives to be strong. They expect their daughters to be strong and train them appropriately. The Colville women’s enculturating functions earn them a position of authority within the extended family in their old age, and they are able to exert influence within the community as well. At any age, women are not considered inferior within Plateau culture because of their natural functions but are honored and receive economic and domestic rewards for exercising those functions. The Colville Reservation data reveal that gender equality exists not only among foragers but also among participants in an industrial society. The facts suggest that equality may flourish at any level of socioeconomic integration and may be present in other unlikely or unexpected places, since modern economic conditions do not appear to be incompatible with gender equality after all. Where the Colville Reservation people have the authority and power to direct their society, they have adapted modern economic conditions to fit their ideology of gender equality. IDENTICAL AND CONTEMPORARY RIGHTS In contrast to contemporary Colville culture, where gender roles tend to be identical, gender roles in traditional Colville culture were different from each other. Men hunted and fished and women gathered, for instance. Both genders, however, had equal access to power, authority, and autonomy in the various spheres of the culture. This access was not identical as in modern Colville Reservation culture but was balanced or complementary. One example is that only men were chiefs in most groups, but women chose them. Interdependence of husband and wife occurred in the Plateau as it does among the Inupiaq and Yup’ik (Eskimo) cultures. The descriptions of this interdependence leads many ethnographers, myself included, to view the status of Yup’ik and Inuit men and women as at least approximately equal, although it is an equality of the complementary kind. A fine example of complementary access to the social spheres of a culture is described by Barbara Bodenhorn (1990) for the traditional Inupiaq of the Alaska North Slope. She points out that Euro-American culture often uses the Eskimo in general as an example of inequality between men and women. Because men hunt, the reasoning goes, they dominate. Further, men control the public sphere because they “work”— that is, they hunt—while the women stay home. In contrast, the Inuit view of reality is that hunting is not simply a matter of men capturing animals. The wife of the hunter attracts the animals to the husband so

GENDER EQUALITY IN A CONTEMPORARY INDIAN COMMUNITY

he can catch them. Since hunting is a sacred act, the wife who is generous and skillful at womanly tasks leads animals to give themselves to the hunter. As Bodenhorn (1990:61) says, “It is the woman to whom the animal comes.” Ideology enters the picture here, as it does in the Colville gender system. A man of the North Slope whose wife has died can realistically seek help from female relatives to raise his children and butcher his catch, but he does not have a wife to whom an animal can give itself. Thus his view is that his hunting is handicapped. Consequently, women’s activities—such as sewing, butchering, and sharing with others—are viewed as hunting skills by the Inupiaq (Bodenhorn 1990:62). This situation is comparable to the concept that the whale comes to the whale captain’s wife at Point Barrow. There the wife has an important role during the launch of the whaleboat, and her behavior and generosity at home while the men are hunting affect the outcome of the hunt (Bodenhorn 1990:61). I argued in a recent article that Yup’ik women must have had even greater autonomy than many women in hunting-gathering societies, since their men lived in so-called men’s houses (see also Chapters 5, 6, 7, and 11, this volume). Men’s houses had the effect of isolating men as individuals (Ackerman 1990a:214). They were excluded from the company of their male kin, since they were scattered over a wide territory into patrilineal dispersed clans, whereas residence was and is matrilocal, at least in Goodnews Bay, Alaska, and some other Yup’ik villages (Ackerman 1990b). Thus women lived with their mothers, sisters, children, aunts, and cousins, which formed a support network, whereas men were isolated from their kin of both genders. The men of King Island in their men’s houses were reported to be often hostile, competitive, and suspicious of each other (Bogojavlensky and Fuller 1973:73, 76). Thus the existence of men’s houses within a community is not an index of male dominance within the society (for a different view see Chapter 11, this volume). It seems unlikely that the separation of men and women in different dwellings has anything to do with gender status, and in fact such an arrangement may be socially disadvantageous for men because of their social isolation, referred to earlier (see also Chapter 9). Bodenhorn (1990:67) pointed out that one translation of the word kashim, generally translated as “men’s house,” refers to a ceremonial, communal, dancing, or festival house. This definition seems to fit the situation more accurately than “men’s house.” I attempted to evaluate gender status among the Yup’ik (Ackerman 1990a) using my research at Goodnews Bay, Alaska, plus the literature,

LILLIAN A. ACKERMAN

and concluded that the genders were approximately equal. I leaned heavily on Margaret Lantis’s study of the Nunivak Islanders, using such data that men secured half the food and 90 percent of the raw materials necessary to sustain life (Lantis 1946:245), whereas women provided the rest of the food. This share of food production compares favorably with the gathering prowess of traditional Colville Indian women, who collected 50 to 70 percent of the food supply in a milder climate. Yup’ik women also produced clothing, boots, and pottery and prepared and stored all the food. Finally, as Lisa Frink (Chapter 7, this volume) reports for women in Chevak, the women distributed the food and were regarded as its owners, like their counterparts on the Plateau. Lantis (1946:246) judged that the status of Yup’ik men and women on Nelson Island was little different. Bodenhorn (1990:61) wrote that Alaska Native women today are moving into power positions, thus extending gender equality from the traditional past into contemporary Inupiaq and Yup’ik cultures. Men and women in all three cultures—the Yup’ik, Inupiaq, and Plateau—were equally valued for the economic and other contributions each could make, leading to an evaluation of equal status in all three societies. ACKNOWLEDGMENTS The early part of this research was supported by fellowships from the Woodrow Wilson National Fellowship Foundation and the American Association of University Women Educational Foundation and by grants from the Phillips Fund of the American Philosophical Society and Sigma Xi. I thank them for their support.

4 CELEBRATION OF A LIFE REMEMBERING LINDA WOMKON BADTEN, YUPIK EDUCATOR CAROL ZANE JOLLES

LIFE HISTORY NARRATIVES HAVE BEEN RECORDED in conjunction with North American ethnographic research since the 1920s. Until the 1970s, however, these locally generated texts were often regarded as complementary, colorful data that could buttress and enliven formal research. It is not my purpose to detail the emergence of the life history narrative as an ethnographic endeavor in its own right. Rather, in this chapter I describe one among many of the remarkably skilled and knowledgeable women who have contributed in major ways to northern Native research and the enduring friendships that so often develop from collaboration with them. Thus my intent is less theoretical than descriptive. As an aside, it is worthwhile to note that the movement of life history narratives from the wings to the stage of ethnographic research can be used to illustrate the relatively greater importance assigned to men’s roles and men’s lives in the ethnographic record. By the late 1970s, narrative life histories coauthored by indigenous women and female ethnographers became more commonplace, the result of a more general attention given to women in anthropological research. Such attention, spurred in part by the dynamics of the feminist movement and the development of feminist anthropology, paralleled a revitalized interest in matters of gender. Marjorie Shostak was especially critical in popularizing indigenous women’s narratives.

CAROL ZANE JOLLES

Shostak, who died in 1997, recorded the story of “Nisa” (Nisa: The Life and Words of a !Kung Woman, 1981), a !Kung-San woman from the Kalahari Desert region of northeastern South Africa. Nisa’s story, compiled after dozens of interviews with a cross section of !Kung women, included a carefully edited and restructured life history narrative plus ethnographic background and interpretation. For researchers focused on Native American societies, indigenous biographies and autobiographies are familiar resources. Many will recognize the coauthored narratives of Albert Yava (Big Falling Snow: A Tewa Indian’s Life and Times and the History and Traditions of His People, 1978), Don Talayesva (Sun Chief: The Autobiography of a Hopi Indian, 1942), Left Handed (Son of Old Man Hat, 1938), the stories of Lame Deer and Black Elk, and numerous other narratives featuring Native American men. At the time they were written, these too were regarded as complementary to “real” research, although they often captured the public imagination. Coauthored life history narratives of Native American women from the “States” were few. Stories revealing the lives and contributions of northern Native American women (Alaskan and Canadian women of the Arctic and Subarctic regions) were practically nonexistent. THEORY Postmodernism, in conjunction with feminist thought of the 1970s and 1980s (see, for example, Behar and Gordon 1995; Clifford and Marcus 1986; di Leonardo 1991; Finn 1995; Marcus and Fischer 1986; Mukopudhyay and Higgins 1988; Ong 1995; Pratt 1986), changed our understanding of the gendered landscape even as other efforts were made more generally to open up our view of history, to give voice to the silent, and to balance the cultural and historical record (Wolf 1982). One corollary was greater attention to first-person narratives collected over the last century. Regardless of our new willingness to acknowledge the lives and contributions of women, however, Native American women are still less likely than others to be featured in life histories. Many Native women’s narratives remain in archives along with other basic data. I would speculate that this lack of attention comes in part because Native women’s lives even now are often perceived as extensions of men’s lives. In the North, for example, because of a persistent subsistence-systems bias in local research, more attention is paid to male hunters than to women food harvesters, processors, and seamstresses; and neither men nor women are given much consideration as members of the modern global community. For many of us who began research in the late 1980s and early 1990s, the exceptional work of two women anthropologists, Margaret Blackman

CELEBRATION OF A LIFE

and Julie Cruikshank, critically influenced our own investigations. The portrayals of Florence Edenshaw Davidson, a Haida woman from the village of Masset in the Queen Charlotte Islands, Canada (Blackman 1982); Sadie Brower Neakok, an Inupiaq woman from Barrow, Alaska (Blackman 1989); and three Yukon women elders of Athapaskan and Tlingit descent from interior British Columbia, Canada (Cruikshank 1990), grew out of longitudinal studies in the women’s home communities (see, for example, Blackman 1991). Each anthropologist worked closely with her indigenous collaborator-consultants, spending several years tape-recording them. In dialectic fashion, each author also attempted to shape and edit the stories of these northern Native women in ways that could be appreciated by the women themselves, by their own families and communities, and by a readership that included the broader Native and non-Native public. Unlike Shostak’s text, the works of Cruikshank and Blackman amply illustrate the contributions these women made to their societies, stressing their talents and accomplishments. The women were not chosen as “Everywoman.” Rather, they were selected for their individual and unique qualities. Nevertheless, Blackman’s two volumes, During My Time (1982) and Sadie Brower Neakok: An Inupiaq Woman (1989), and Cruikshank’s Life Lived Like a Story (1990) are exemplary not simply because they highlight the very interesting lives of their consultants. Both authors present their consultants in ways that adhere to local traditions of tact and propriety without either unnecessary sensationalism or abrogation of traditional values that preclude drawing attention to the individual at the expense of family or clan. Blackman’s descriptions of Florence Davidson, for example, emphasize Davidson’s accomplishments within the framework of her marriage and the demands of Haida tradition. Blackman’s collaboration with Sadie Neakok is executed in a similar fashion. Cruikshank’s groundbreaking work in Life Lived Like a Story not only records the lives of its three Yukon elder protagonists, Mrs. Smith, Mrs. Sidney, and Mrs. Ned, but also takes into account postmodernist concerns with voice and perspective. Cruikshank relinquishes Western tradition with its reflection, attention to explication, and reliance on a linear chronology of events in favor of the categories and perspectives preferred by her collaborators, creating as she does so cultural relevance deriving from the protagonists’ cultures rather than her own. PURPOSE It is with the work of these two anthropologists, especially with the women whose stories are the subjects of their cooperative enterprises, in mind that I turn to this chapter. My task is both to pay tribute in some

CAROL ZANE JOLLES

small way to my friend, adviser, and consultant, Linda Womkon Badten, a St. Lawrence Island Yupik1 woman of extraordinary character and talents who passed away in the spring of 1997, and to draw attention to the critical and various roles northern indigenous women play in documenting their cultures. DISCUSSION A life is many things. At the very least it is constituted of a multifaceted self. Obviously, that self is composed of the singular characteristics of personality and personhood that distinguish self from other. These include a congeries of qualities that result from cultural background, on the one hand, and from interactions with others both within and without one’s own culture on the other. Self is shaped by tradition and historical moment, by belief—the monumental elements of spiritual congress that form the self—and by discourse, the language with its cognitive and substantive qualities through which that self, that life, is expressed and made familiar. These forces are at work as self emerges, “becomes aware,” gains in confidence, and achieves adulthood (for a discussion of Yup’ik personhood, see Fienup-Riordan 1986a). Postmodernist theory suggests that anthropology too is bound up with such processes. If this is so, then any rendition of a life story should be marked by its acceptance of the multiple voices and embodiments of experience of which it is constituted. Minimally, it must acknowledge interactions between consultant and interviewer (Clifford and Marcus 1986; Marcus and Fischer 1986; Pratt 1986). In the present context I attempt to acknowledge the relationship between myself as friend-anthropologist-interviewer and Linda as friendconsultant-adviser. Eventually, it must do much more. Cruikshank noted that a life history narrative “simultaneously reflects continuity with the past and passes on experiences, stories, and guiding principles in the present” (Cruikshank 1990:x). Here I combine Linda’s collaboratorprotagonist’s story and her (protagonist’s) past and present, my own story and my past and present, and the common experience of the two of us. TRIBUTE: A COLLECTION OF THEMES In tribute to Linda I present a brief outline of her extraordinary life along with recollections of the moments where our lives came together, and I try to place her life within the frame of her own (Yupik) and our own (her Yupik and my laluramka, or “white person”) cultures. I begin with an account of our friendship and our work together; I follow with

CELEBRATION OF A LIFE

background information about Linda’s culture, that of the St. Lawrence Island Yupik, and a much abbreviated account of Linda’s life. I conclude by comparing aspects of my friend’s life to those heroines of Yupik traditional stories whose adventurous mythic journeys encompass so much of Yupik values. By drawing on traditional stories, I hope to place Linda’s life within the larger context of Yupik experience while adding a second implicit Yupik voice to the discussion. I look to these stories for possible explanations of her achievements, her cheerful multicultural perspective, and her successful management of a complex multicultural existence. AN EXTRAORDINARY WOMAN Linda Badten was a close friend. The circ*mstances of our friendship and my personal sense of loss at her passing therefore color what I say here and my understanding of her life. That said, let me begin. I met Linda in 1989. I had completed ten months’ residence and research in her birthplace, the village of Gambell, St. Lawrence Island, Alaska. We had not met previously because she had lived for many years in Fairbanks, Alaska, several hundred air miles away. Although we had not met, I had been told about her by younger women in her large extended family, the members of her ramket, or clan, who still live in the village of her birth. For these women Linda was a role model, a living example of what faith, intelligence, perseverance, and hard work could accomplish. I knew little about her personally except for her publicly acknowledged achievements. When her family members learned I would be traveling to Fairbanks to work in the Polar Archives at the University of Alaska, they suggested I call her. I did so, and Linda immediately invited me to stay with her, not just that one time but whenever I was in Fairbanks. Thus began our friendship, based originally on her hospitality, our shared interest in her culture, and her knowledge that I had spent almost a year living among her relatives and caring for her father’s brother’s elderly widow. I stayed with her thereafter whenever I traveled to Fairbanks. As Linda and I came to know each other, I found she had the gift of humor. Each time I visited her we stayed awake until two, three, and four in the morning. We talked, we laughed (sometimes without much of an excuse beyond the pleasure of laughing together), and we sang songs—Linda was an ardent and enthusiastic piano player with a large repertory of Broadway show tunes, old-time favorites, and an endless supply of hymns. She played with such gusto and forte that we used to joke about her having to have a house that wasn’t too close to anyone else. And we played Scrabble. Night after night Linda beat me at Scrabble.

CAROL ZANE JOLLES

I can’t remember winning a single game. At her memorial service I learned I was not the only person Linda had beaten at Scrabble, leading the minister to ask for a show of hands among those in attendance who had once “played a board game with Linda.” Many hands in the church were raised. Sometimes Linda cooked for the two of us, sometimes we cooked together, and sometimes we went out to eat. Often we visited her daughter and her family. Only after several years of friendship did we begin to work together formally. Almost from the beginning, though, Linda served as my critical reader and adviser, going over an article I was writing or had just finished. The two of us discussed the points I was trying to make and mulled over my supporting evidence and examples from my personal experience within the community. Often the writing became a starting point for a philosophical discussion on the nature of marine mammal hunting communities and culture change to which we would return several times in the course of a visit. She expressed her approval of the work I was carrying out in the community, saying she felt both the past and the present history of the community needed to be documented. She was particularly interested in the combined political, economic, and cultural changes taking place in Gambell and saw those processes as history in the making. And she was curious about my experiences as a non-Native person living in her home community. As we talked about the community and her perceptions of it, I became aware of the great psychological distance Linda had traveled from her homeland. Often she seemed surprised at the conservatism of the Gambell community. Once, after she had returned to Fairbanks from a visit to Gambell, she said with some amazement that she had encountered what she called “clan politics.” She said she had not realized anyone in Gambell still attached much importance to the ramka, or clan system. She wondered after her visit whether she had actually experienced culture shock in her own birthplace. And we pondered together over the differences between rural tribal village life and life lived in the city, regardless of identity and origins. In 1991 we worked together formally for the first time on a project to document the ethnographic content in the paintings and drawings of a well-known St. Lawrence Island artist. We recorded a number of audiotapes that described that content, employing an interviewer-consultant format. In 1992–1993 Linda and I continued our work on St. Lawrence artists and received, along with several others, a planning grant to explore possibilities of an exhibit at the Otto Geist Museum. Toward the end of 1994 she and I began to discuss recording her own story. With

CELEBRATION OF A LIFE

what seems in retrospect a sense of premonition, Linda was preoccupied with two concerns. The first was to leave a meaningful narrative for her children and grandchildren. The second was her desire to find a publisher who would allow the greatest percentage of profit to return to her family. She wanted to leave behind not only her story but also some small measure of financial support for her family. She thought a book about her life might do both. In this, Linda’s reasoning was similar to other women from St. Lawrence Island with whom I have worked. The conventions of the life history format, combined with the cultural constraints at work in northern communities, make drawing attention to individual accomplishment a complex task. Certainly, Linda’s modesty about her experiences and achievements was a product of the Yupik values she cherished, which suggest that one should not broadcast one’s achievements. Nevertheless, she wished to go ahead with a book about her life. I have no doubt that she would have agreed with another of my St. Lawrence Island consultant-collaborators who told me, “I’m not doing this for myself; I’m doing it for the grandchildren.” In 1995 I obtained a small grant from the University of Washington (Seattle) that would allow us to begin our project together. Thus in the fall of 1995 Linda and I recorded what would become the only tapes of her life story narrative. Because Linda was troubled by family concerns and the first signs of physical discomfort eventually diagnosed as cancer, we spent far more time talking, going for walks, and eating than we did recording. Nevertheless, those tapes substantiate Linda’s capacity for objective cultural analysis, her drive, her enduring and scholarly interest in and curiosity about her own culture, and her energy. During that visit we began to discuss traveling to the Smithsonian National Anthropological Archives together. We hoped to persuade one of her close childhood friends from Gambell to go with us. We envisioned this trip as an exciting opportunity to explore life histories together. We planned to stay in my mother’s house where I had grown up, a house filled with my familiar ancestral memorabilia, and we anticipated working each day at the Smithsonian surrounded by the ancestral memorabilia of Linda and her friend. Our plan was distinctly postmodernist in concept, involving three women plus the unpredictable but interesting possible contributions of my mother, but, sadly, it never took place. By the fall of 1996 Linda was too unwell to undertake such a journey. When I last spoke to Linda, at the end of April 1997, she had just learned that the cancer which had plagued her for more than two years was terminal. She was clearly shaken by this news and was trying to

CAROL ZANE JOLLES

sort out the collapse of her physical strength from the enduring religious strength that sustained her. Neither she nor anyone else realized that she had only a few weeks to live, not the six to eight months she imagined. When she died she left a heartfelt void: in her family, among her friends, and in the small community of northern scholars who focus on the history and culture of Alaskan and Siberian Yupik peoples. The memorial service for Linda brought together many people: her children, grandchildren, and in-laws; her Yupik relatives, among them highly respected bowhead whale hunters; college professors, researchers, and former students from the University of Alaska; students and friends from her Bible study group and Sunday school classes; a wellknown Alaskan author; the nurse who attended her in the hospital; parishioners from her church; former colleagues and students from the elementary school where she had taught so many children to read and to love language as she did; and a collection of people from all walks of life who had met her and wished to pay their respects on her passing. GROWING UP YUPIK (LINDA WOMKON BADTEN, 1923–1997) In many cultures the lives of natives span periods of critical and rapid culture change. (Blackman 1982:5) You learn everyday. It seems like, even as old as I am today, I learn new things. You don’t cease to be a student (Sadie Brower Neakok, in Blackman 1989:221).

Linda Badten was a gentle, determined, intelligent woman. On her death, many people recalled her personal qualities and described events that illustrated her generosity, her devotion to friends and family, her willingness to share her cultural knowledge, and the deep religious faith that supported her in all things. Beyond this Linda was a person whose achievements were celebrated in two cultures. Linda’s birthplace was the Yupik (Eskimo) community of Gambell, St. Lawrence Island, Alaska, in the north Bering Sea. In the 1920s, the time of her youth, Gambell was effectively isolated from contact with mainstream American life for nine months a year. Radio contact was limited in those days, and the first plane landed on the frozen lake that borders the south side of the present community in 1916. Gambell had no regular, dependable scheduled air travel until the late 1970s, no television until 1980, and no running water until 1994. During the brief summer months, ships frequented the shores with deliveries of staple goods and the few luxury items that came to the community: oranges

CELEBRATION OF A LIFE

Figure 4.1 Sketch map of St. Lawrence Island

and apples, dried fruit, hard candies, pilot bread and cookies in large wooden containers, flour, tea, and bolts of colorfully patterned cloth. When the supply ship arrived, the whole village turned out to help unload. These were some of the outward signs of the commercial world, which might have impressed a young girl growing up in the 1920s. Boat travel between the island and the Siberian shore only thirty-eight miles to the west was still common, although that would slow down and then come to an abrupt halt in 1948 when the border between Russia and the United States closed. For Linda and others of her generation, life centered on family and kin. In winter there was school and the local Presbyterian church. Some families spent the winter in fox trap–line camps and returned to Gambell each spring to participate in bowhead whale hunting and the walrus hunting that followed. Others stayed in the village during the school year while able-bodied men from the family left to hunt and trap. In summertime there was camping, and most families traveled to their traditional family lands. Linda’s family camped on the south side of the island.

CAROL ZANE JOLLES

Girls of Linda’s generation could expect to follow a predictable life pattern. Girls were carefully guarded and supervised in childhood by their parents, grandparents, and other close adult relatives whose ties were reckoned especially through the father’s side of the family. (St. Lawrence Islanders are distinguished from other Eskimo peoples by their patriclan kinship system.) Formal education in the local two-room schoolhouse operated by the Bureau of Indian Affairs generally ended in the fifth or sixth grade. The teachers were recruited from the “States.” Alaska was still a territory—remote both politically and culturally from the U.S. “mainland.” If parents needed a young girl’s help at home, her schooling ended before she finished the elementary years. Many girls of Linda’s generation did not complete the educational sequence available in the village, since formal education for girls was often thought unnecessary. High school was unavailable in the village. Those who desired further education traveled to Sitka or even to boarding schools in the States. If a girl did desire to attend high school, her parents were unlikely to give permission for her to travel to the mainland to do so. By the time a girl finished sixth grade, she could expect to be kept busy at home helping her mother, aunts, and grandmother with women’s chores. These included minding her younger brothers and sisters; fetching and carrying water or other items; assisting in routine gathering of greens, berries, mosses, and grasses (used for a variety of household chores) in summer and ice to be melted into water in winter; assisting in food processing tasks; learning to sew; and generally familiarizing herself with the different meats and other foods her female relatives prepared (for more on women and productive responsibilities see Chapter 7, this volume). Her homely chores anticipated a time when she would be responsible for such duties in her husband’s family. She stayed closer to home than her brothers and had more household responsibilities at an earlier age than they. Young men, after all, were destined to become hunters, and travel away from home was a routine aspect of the hunting experience. In her teens a girl undoubtedly became aware that the elders of her family had long since arranged for her marriage partner. The ideal was for her to marry someone in a cross-cousin relationship to her. Eventually, she went through lengthy traditional marriage ceremonies that included a series of gift presentations and a formal bride service (known in the community as “groom’s work”), and finally she would move to her new husband’s family home to begin a family. All of this constituted the expected and predictable pattern for a woman who would have

CELEBRATION OF A LIFE

reached her seventeenth year in 1940, as Linda had (see, for example, Jolles 2002; Jolles and Kaningok 1991. But it was not Linda’s life. Linda, looking back, said she was “always curious,” that she “always wanted to know about the world” beyond the borders of her own community. She felt she was always preparing to leave home. Although her friends seemed to satisfy their curiosity about the world beyond them with pictures from magazines and occasional forays into the pages of books, Linda imagined traveling to that world to see for herself. She exclaimed once that she used to dream of seeing a “real lawn of green grass.” And she speculated that this was the reason she so loved to take care of the lawn around her small blue clapboard home in Fairbanks. She didn’t consider herself a gardener, but her lawn was something special. It symbolized for her the curiosity that was a signpost of her intellect and the drive that had led her away from her birthplace while others remained at home. Linda was born in 1923. Her full name was Adelinde Aghanaghaughpik Womkon. Her English given name came from Otto Geist, a German self-educated archaeologist and associate of Charles Bunnell, president of the newly founded Alaska College of Mines that later became the University of Alaska, Fairbanks. The university’s museum is named after him, and its original collections donated by Geist came from his stay on St. Lawrence Island. Geist had befriended Linda’s father and apparently had asked that Linda be named for his sister, hence the German spelling of her name. She was not particularly fond of the name, although she felt comfortable with the Americanized short version, Linda. Her Yupik given name came from her family and drew on their historical attachment to St. Lawrence Island and the Chukotka Peninsula. Her heritage was a mix of St. Lawrence Island Yupik and Siberian Yupik. Her father was Wamkun. Either when the early missionaries were present in the community (1894–1911) or during the first remembered census on St. Lawrence Island (1937), his name became the formal surname of his family of marriage; he acquired the English given of Patrick. Thus Womkon, as it was spelled by those who gave it surname status, became Linda’s last name until her marriage. Linda, like many others of her generation, was born in a domeshaped two-room house, a type that came into general use in the community following the abandonment of semiunderground dwellings, or nenglus, in the late 1800s (see also Chapters 5, 6, 7, this volume). Its building materials consisted of driftwood and walrus hide. Linda’s earliest memory was of watching her mother scrub down the interior walls of this house. By the time she began to “be aware,” to use the Yupik

CAROL ZANE JOLLES

phrase, the walrus-hide walls had been replaced by lumber walls. Still, the house in which she spent her youth had one main room and an outer room (unheated) used for storing family belongings. There was also a frame summerhouse, difficult to keep warm in those days because it lacked insulation, and its shape was less amenable to heating with a seal oil lamp or the kerosene lamps that replaced them than the more traditional home with its domed roof and focused living quarters. The family also had a traditional camp on the south side of the island, in an area the family considers its traditional homeland and the source of its ramket name. Still later, after the trade in Arctic fox skins proved profitable for her father’s family, the family built a very large two-story wood frame dwelling that still stands, a monument to that productive and profitable time. Linda grew up with her parents, her grandparents, and her father’s brothers and their families. She regarded her cousins as brothers and sisters. All lived together in the two-room house. It was a life that brought a smile of fond remembrance to her face when she spoke of it. From that period she drew a strong sense of self, of embeddedness in her culture, and a rootedness in the family values that supported her throughout her life. My conversations with Linda generally focused on select realms of experience. An underlying theme was her religious faith. It was personal, it was deep, and it supported her through a range of encounters it is hard for me to imagine. Let me backtrack. I first met Linda in the written notes and comments recorded in manuscripts now housed in the Presbyterian Historical Society archives in Philadelphia. She came to my attention originally because so many who had met her as a young girl in the 1940s considered her highly intelligent, remarkable, and strong. There was only one photograph that I recall, though, and it hardly did justice to her beauty. Much later, her daughter showed me a picture of Linda from the time she married. It showed an elegant young woman with great personal poise and presence. It suggests that whatever else one might know of Linda’s personality and character, she was at one time a woman who must have attracted attention because of her loveliness. To return to the narrative, Linda’s early life was lived within the safety and security and love of a strong family. She spoke easily and happily of her admiration for the elders who were the centerpiece of her young life. She especially admired her father. And she spoke with nostalgia of the times when, as a young child, she was invited to listen to the stories of her grandparents, to move from her own nuclear family to

CELEBRATION OF A LIFE

her grandparents’ section. Each segment of the family group had its assigned place in the room. Her oldest sister, who was approaching adulthood, also had a special place. No one, according to Linda, entered another person’s space without implicit permission. This gave each space its own sanctity while reinforcing respect among all those who shared the home. She also spoke of mealtimes. Women of each family shared the resources brought in by their menfolk among the rest, and a general feeling of goodwill prevailed. If it sounds romantic, with the hard edges of life in a one-room space smoothed away, it probably is, but that is the nature of memory. It is the image of early life Linda carried with her. In a separate compartment, away from the delight Linda retained of her home and camp life, she carried her memories of the pain of that life before Western medicine stopped the ravages of tuberculosis, influenza, polio, and other once incurable disorders. Linda lost many in her family, including all of her siblings to tuberculosis (see also Chapter 7, this volume). In her own words (excerpted from the tape recordings of her descriptions of her life): Oh it was horrible! It got so that I used to look around the big family to see—I wonder who, you know, who’s going to be next? How terrible it was! That death. And they all looked dear, oh, which one of those dear people is going to be next? I had a torturous life because of that.

And in yet another compartment Linda kept both her own curiosity and her impatience and downright boredom with a life in which very little went on and she could envision her future by watching her female relatives. It was offset by her curiosity toward the world around her. One of the great excitements was school. Very early she was caught by the power of language, of words, of description. They transported her to other worlds, and unlike her girlhood friends, she was driven not simply to hear about them but to see them and experience them firsthand. She brought what she learned in school home with her. She says she was so anxious to do well in school that homework was always the first thing she did, not the last. Of her early life she said: Oh, I’m telling you! I didn’t prepare for anything over there [for a life on the island]. Yes, I did finish a couple of pairs of boots. But it was an ordeal! It took so much time away from doing homework for school. So because the school was becoming very important . . . fitting into the school was becoming important and doing your homework was stronger . . . you did a little bit of this for your mom, and then you’d get back to your own schoolwork. So I never

CAROL ZANE JOLLES did get into the full swing of becoming a member [of her society] that way, as the girl child did. . . . I don’t know what [the other girls] did about their school homework, but I was worried about mine. . . . I tried to keep ahead of everybody.

Her friends in school were older, but she held her own—in schoolwork but not always in the social environment of her friends. She remembered that she was often teased as a girl. Both her older brother and some of her classmates taunted her about being “plain.” She recalled that her brother was the favorite at home and at school. She didn’t like being teased, and she didn’t like the favoritism shown him, but he was her brother and she loved him. Before he was seventeen he was gone, swept away by tuberculosis. Her siblings disappeared, one by one, as the TB epidemic took its toll. Through it all, Linda studied hard. At home she did chores. She remembered that even when she wanted to be outside playing, the chores came first. She couldn’t believe how much water her family needed, for example. Her task was to go with her brother and bring home the sled with a block of ice on it, which would then be melted for drinking and cooking water. She remarked that they always seemed to need more water for something. One night when we were talking about the daily chores, Linda yawned uncontrollably. I joked that I was putting her to sleep. She replied that no, it was her early life that was putting her to sleep! She could not find any liveliness in the monotony of her childhood chores. Nevertheless, from the boredom of everyday life, she and her brothers and sisters were inspired to create games to while away the time spent indoors. She related how she and her cousins had invented games inside when it was impossible to go out. When I was bigger, they had converted the skin walls [of the house] into wooden walls, and that gave us an opportunity to tear the most colorful pages out of magazines to plaster the walls! To make them pleasant! And not only that, not only was that pleasant, [but] when we children did not have anything to do we learned to spell, [to do] our letters. [With] magazine pages. And we would continue to learn the hardest words we could find. I remember “hexol resorcinol” solution. There was a pharmacy page that had “hexol resorcinol” solution, and to compete with my cousins, I learned to spell that. There were times when we didn’t have anything to do; I guess we were bored, and we learned our ABCs backward and learned how to count as far as we could. In that way we began to learn the

CELEBRATION OF A LIFE

patterning of number systems. . . . I think that was a benefit of being in this type of a home where we were restricted from roaming around, because we had to stay, stay put much of the time in our own areas. So being there, not wanting to disturb the older people, we had to invent something . . . like learning to spell, competing with one another.

Linda liked math and proposed that we find someone who could help us write about the Yupik counting system. It seemed to her to have much in common with the “new” math concept she had taught to her students as an elementary school teacher. At her funeral, a friend related that Linda had once tried to explain the idea of atoms and molecules to her family but found that the Yupik language lacked the right words to convey her thoughts clearly. She also loved to tell stories. She once described how she had decided to tell some stories by Edgar Rice Burroughs and Rudyard Kipling to the younger children when she was home from school. Linda said, with a twinkle in her eye, that she had silently challenged herself to capture the attention of the adults in the room with her tale without seeming to do so. Carefully, she built up her story with dramatic voice and gesture. She said she was very aware of her uncles sitting in the room talking among themselves about “adult” matters—men she thought of as her second fathers. Studiously, she kept her eyes on the children but raised her voice and her dramatic tone, glancing out of the corner of her eye from time to time to see if the men were listening. Gradually, she said, the men stopped talking to each other and listened, spellbound, to her story. Later she heard one of her uncles repeat the story to some friends. Linda chuckled as she told me, “I had them in the palm of my hand.” Linda was drawn into the world through her fascination with language and ideas. She was a good student. When Linda approached the end of elementary school in the village, her teachers encouraged her to continue her education. It was what she wanted to do. I never knew which of the teachers was most important to her later schooling. Very early she had been taught by Paul Silook, a man from the village. She also remembered a Mr. Thompson whose nose was quite large and whose false teeth, which he pulled out of his mouth to surprise the children, absolutely terrified them. They wondered among themselves whether Mr. Thompson might not be a shaman. As Linda recalled those years, she said, “I was so busy with schoolwork that I never really learned the tasks that a girl was supposed to learn. I never finished those things.”

CAROL ZANE JOLLES

Linda’s desire to go to school away from the village was unusual. Imagine her delight when her parents reluctantly gave permission for her to attend Sheldon Jackson High School in Sitka, Alaska, many hundreds of miles away. In Sitka Linda was among strangers, but all of the students had in common the newness of their experience. There were few opportunities to travel back and forth. Letters were the only way to communicate, and that was hardly reliable in Alaska in the 1940s and 1950s. It is not exactly clear how old Linda was when she attended Sheldon Jackson, but it seems likely she entered around 1948–1950. She apparently did well in her classes and again impressed her teachers with her work and her faith. She was offered support so she could continue her studies at the College of Emporia in Emporia, Kansas. In the early 1950s Linda A. Womkon became the first person from her tribe, from the St. Lawrence Island Yupik community, to earn a bachelor’s degree. During her years in college, she stayed with fellow students over vacations and holidays. She went on trips, traveling as far as New York City, and met students from all over the world. Each visit to someone’s home or to a new city had its own story, its own appeal for her. With degree in hand, Linda returned to Alaska to teach elementary school. Her first teaching assignment was in the Athapaskan village of Fort Yukon. She went on to teach in Kotzebue, where she met and married her husband. The couple moved to Fairbanks, where in 1958, waiting for the birth of her second child, she lost her husband, a pilot, in a tragic plane crash. She never remarried. She raised her two children and became known as a highly respected teacher. Linda was at home in the non-Yupik world in which she had chosen to raise her children. At the same time, she retained an abiding interest in and devotion to her own heritage. She expressed that interest by sharing what she knew with others and by making a substantial contribution to the study of St. Lawrence Island culture. In 1960 Linda was approached by linguist Michael Krauss, a young professor at the University of Alaska, Fairbanks. He was interested in Native languages and was searching for someone who could help him with a Yupik text he had recently acquired by the Russian scholar Rubtsova, written in the Cyrillic alphabet. Linda listened to Krauss’s description of how the Cyrillic alphabet worked and set out to learn to use it to read some of the very first materials she had ever encountered written in her native Yupik. At that time Yupik was not a written language in the United States, and Linda became instrumental in transforming Yupik from a spoken-only language to a written language.

CELEBRATION OF A LIFE

Figure 4.2 Linda in her late teens or early twenties

CAROL ZANE JOLLES

Krauss and Linda became close professional and personal friends, a relationship that ceased only with her death. From their professional association came the first St. Lawrence Island/Siberian Yupik dictionary (Badten, Kaneshiro, and Oovi 1983; Jacobson 1987), numerous transcribed texts, and cooperative endeavors of all sorts. Linda became as at-home in the offices of the Alaska Native Language Center on the University of Alaska campus as she was in her own classroom at Nordale Elementary School. Scholars with international reputations called on her to speak to their students about her language and her culture. Linda inspired others from her community to engage in scholarly pursuits as well, and the dictionary project was a product of her work with two of her cousins, women she considered sisters. In all of this, what is striking is Linda’s uncompromising optimism, her willingness to go forward. Unlike her peers, life away from home was not desolate for her, although at times it must have been lonely. She said later that her ability to survive must have developed from a very young age, from her very early desire to see the world: It’s a mysterious thing, when I think back, that I would have thought of these things as small as I was. Little dreams. I guess magazines did have a lot of influence to arouse my curiosity about other places, because I loved looking at pictures of outside, outside of St. Lawrence Island. Unbelievable lawns and things like that . . . trees, lawns, what it would feel like to be among trees.

When I reconsider Linda’s life, with its myriad accomplishments, it seems appropriate to compare her life with those of the Yupik heroines from the tales I have heard over the last decade. One in particular comes to mind, which I will outline here, drawing attention to the qualities associated with Yupik heroines. Perhaps heroine is not the right term; admirable and virtuous womanhood is probably more appropriate. This is the story of two young girls. The protagonist of the story must leave home on a journey. She travels a great distance in unknown lands. Everywhere she goes she is presented with challenges. Some come in the form of huge tasks that must be finished. For example, she is confronted with a great pile of fish, not yet cleaned, that sits in the middle of her path. Although no one is in sight, the girl feels obliged to clean all of the fish and leave everything in good order before continuing on her way. On another occasion she finds a stack of partially sewn boots, which she finishes. She also cleans meat that has been left on her trail. Her honesty is tested. At one point she is asked by an old woman she has met to sort through a tray of beautiful beads but to take none for

CELEBRATION OF A LIFE

herself. This she does. She obeys the instructions from the elderly woman without question. As she travels along her road, doing exactly as she has been told, it becomes evident that the journey that has led her from home is also the journey that will return her safely to her home. Even in moments of greatest stress, she withstands the temptation to disobey. She is told on one occasion that if she looks behind her she will die. She does as she has been told, even though she can hear the deep growls of polar bears directly behind her on the path. The reward at the end, once she has faithfully met each challenge, is a strong, handsome young hunter who weds her and cares for her and her family. The second part of the story tells of a lazy girl who also wants a handsome husband. She copies her virtuous sister and sets out on a journey of her own, but she fails to complete any of the tasks set before her or to respond obediently when instructed by the old woman. She is rewarded in the end with a weak, miserable old man for a husband. She accepts her fate, and because of her willing resignation to circ*mstance, albeit of her own making, her less attractive husband becomes materially successful, and she too lives, as they say, a long and prosperous life. A number of Yupik tales exist in which the protagonist, often the youngest daughter, must undertake a difficult journey. The heart of the tale is not how she outwits the enemy or escapes danger so much as it is a test of her virtues: her ability to demonstrate through action her adherence to strong Yupik values. Linda believed these stories were designed to affirm the social order and said with a sigh that she was glad she didn’t have to live in those olden times, that she would never have survived. I think there are parallels between Linda’s life and the lives of traditional Yupik female protagonists, parallels she was too modest to notice. First, there is the journey. It seems obvious that Linda envisioned herself traveling toward the unknown from a very early age, caught by the excitement of what lay beyond her own small island. Unlike her peers, she was not frightened by the world outside. When she finally did travel to Anchorage, to Sitka, to Emporia, Kansas, and beyond, she met each challenge in two ways: she worked hard, and she set high standards for herself. She also found pleasure in the challenges of the work, in learning something new, and especially in meeting new people. Unlike her friends who became desperately homesick when they left the island and returned home as quickly as possible, Linda continued the journey. When she did return home periodically, she brought her new knowledge and experience with her and shared it as much as possible with her family.

CAROL ZANE JOLLES

Most stories about Eskimo women are told quietly, within families. Many of the stories come as legends or as instructive tales that have mythic or metaphoric qualities. Only occasionally are stories told of women one knows or might have known. Only a few are about the recent past or the present. Of these few very real and quite remarkable women, most have engendered the respect of others within their own homes, their own homelands. Their admirable qualities were appreciated within the context of a local community, often the community into which they had been born. Linda Badten was not one of these women who gained prominence because of her homegrown achievements in her birthplace. She was an exception. In an era in which few Native women succeeded away from their traditional homes, she stands out as an uncommon woman, one who managed successfully not only to exist in two cultures but also to contribute in amazing ways to each. As her memorial service so eloquently affirmed, her positive example touched dozens and dozens of people. In spite of her accomplishments, Linda remained modest and unassuming. Her personal goals had to do with her children, her faith, and her sustained interest in her heritage. Because of this and because she was generous in fact and in spirit, she never became estranged from her family, although she lived most of her life away from her homeland. She became for many a model of the possible, and in this her life, like that of a traditional Yupik heroine, has come to partake of the ideal. As anthropologist Aihwa Ong noted, “Life stories have been defined as testimonials . . . that both reinterpret and remake the world” (Ong 1995:355). I have no doubt that Linda Womkon Badten’s life will be held up as an example for young girls to follow and will become instrumental in reinterpreting and remaking the future for new generations of Yupik young women. ACKNOWLEDGMENTS This chapter is the result of my friendship with Linda. It is only natural, then, that I offer my thanks to Linda’s daughter, Jayne Badten Harvey. I am truly indebted to her for her friendship over the years, for her support, and for her help. I am also grateful to Edna Apatiki Anungazuk, Linda’s niece, who followed in Linda’s footsteps and became a teacher. It was Edna who impressed upon me the important role Linda had played in her life and in the lives of other women of her generation as a role model. And it was Edna who first encouraged me to seek out Linda when I traveled to Fairbanks. I am indebted, too, to the institutions that made work in Gambell and later with Linda possible. That work was

CELEBRATION OF A LIFE

supported by the National Science Foundation (Grants NSF-8721726 and DPP-922032), by the Royalty Research Fund of the University of Washington, and by a grant from the Alaska Humanities Forum. Although each of these grants provided opportunities to work with Linda, only the Royalty Research Fund grant funded the recording of Linda’s life history. All, however, made it possible for me to spend time with Linda, to work with her, and to deepen our friendship. For that I am most grateful.

II HISTORICAL AND ETHNOARCHAEOLOGICAL APPROACHES

5 CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES CAUSES AND EFFECTS IN NINETEENTH-CENTURY WESTERN ALASKA RITA S. SHEPARD

SIGNIFICANT MARKERS OF SOCIAL TRANSFORMATION may be hard to find in the archaeological record, but historical archaeology allows greater insight into such processes because researchers are able to employ many archaeological and nonarchaeological tools in sketching a group’s past lifeway. We have the good fortune to be able to use written records, oral history, photographs, and personal interviews. Yet even these tangible and personally interactive strategies do not make it easy to distinguish a point or cause of social change. Symbols that identify socioideological structure do exist, however. When they are replaced and accepted by the community, the new symbols “crystallize” social change. I have argued previously that although Native Alaskans readily adopted the technological advances introduced by Europeans in the eighteenth century, they did not change the features of their social organization and their intertwined ideologies until the economic and social values propagated by the nineteenth-century Protestant and Roman Catholic missionaries had been widely accepted (Shepard 1997). In this chapter I suggest further that the roles of women and men within a household also underwent a radical transformation when belief systems changed, leading Natives to move into Western-style houses and out of their traditional domestic systems (for more on ideological systems and change see Chapter 3, this volume). The question for the archaeologist is, can

RITA S. SHEPARD

these role changes be identified in the archaeological remains we recover and analyze? First, it is necessary to understand both a “domestic system” and a “household” as I define them. The domestic system includes dwellings and their associated storage facilities, ceremonial structures (i.e., qasgis) or other special buildings, and the surrounding plazas, courtyards, or outside use areas (Shepard 1997:12). In the mid- to late nineteenth century, Native winter settlements might contain all or only some of these attributes. Indeed, in their journals explorers often noted nothing more than a single semisubterranean house as a stopover locale. Jennifer Tobey (Chapter 6, this volume) also calls for expanding the analysis of a household. She maintains that archaeologists must consider the surrounding locale of a house and the associated artifacts to understand the dynamics of the people inhabiting a house. She and I agree that the people who occupy a particular house, although making use of the entire domestic system and drawing on the surrounding locale, must be considered a “household.” QUESTIONS The move into a Western-style dwelling, in which a nineteenth-century Western-concept nuclear family lived, occurred for the most part from the 1890s through the early twentieth century, after the U.S. missionaries were firmly established. I have suggested previously that much of the sociocultural change seen during this period was based on Native acceptance of Christianity with its new moral standards (Shepard 1997). Consider what a residence change for the household might mean: 1. Did a change of dwelling style discernibly affect household social organization and division of labor? 2. Did a change in living space create observable changes in personal communication and allocation of individual space? 3. If so, how can archaeologists uncover these social changes in the archaeological record? 4. Most important, can archaeologists discover something useful to add to our understanding of household relationships?

Since people of both sexes comprise a domestic unit, we can look for evidence of change in social relations at a microscale, at the level of the household. Obviously, analyzing changes in the cultural remains to understand changes in interfamilial relationships could be a very slippery exercise. Nevertheless, I think it is worth a try.

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

The remainder of this chapter will address these questions, beginning with a consideration of the problem from a broader perspective. Descriptions of western Native Alaskan households during the Russian and early American periods (i.e., mid- to late nineteenth century), followed by brief discussions of the Moravian mission at Bethel in the Kuskokwim River Valley and the Swedish Covenant missionaries in Unalakleet on Norton Sound, will provide a backdrop for understanding late-nineteenth-century changes in familial responsibilities and gender interactions. Next, strategies and methods of three previous archaeological studies of gender underscore the difficulties of the pursuit; attempts to identify gender at my own research site are included in this section. The chapter concludes with discussion and recommendations for future work. CONSIDERATION OF THE PROBLEM Archaeologists often disagree when it comes to identification and interpretation of ideology in the archaeological record. Fewer debates over how to identify ideology might occur if better accord existed over its meaning. It is important to remember that ideology is not just about religion, although it does include it. In the present discussion I define ideology as an explicit body of concepts about human life, accepted by and characteristic of a specific individual or group. As archaeologists, we should be able to find and identify concrete physical manifestations of ideology, or a materialization of ideology (DeMarrais, Castillo, and Earle 1996). If, moreover, we consider ideology to be both a source of social power and a part of the framework that shapes social organization and economic activity, then we can contend that people within groups will be wont to create tangible forms in which to ground and focus their power (DeMarrais, Castillo, and Earle 1996:16). Ideology can be manifested in various ways. Shared rituals and ceremonies, storytelling, symbolic objects or icons, architecture, and written texts are tangible markers of belief; they reflect political and economic patterns as well as ideological systems (DeMarrais, Castillo, and Earle 1996:16). In highly structured societies like chiefdoms and states, manipulation of one or more of these tangible symbols allows leaders to exploit beliefs and gain and maintain social power. Materialization plays a different but equally significant role in smaller hunter-gatherer-fisher societies. Elizabeth DeMarrais and her colleagues (1996) examined monuments and ordered landscapes within complex societies (i.e., Inca, Moche, and the Danish Neolithic) to understand how

RITA S. SHEPARD

people “materialized” their ideological system. I suggest that domestic structures and interior household design organize social relationships and establish boundaries in less complex social groups. Domestic houses and ceremonial buildings, symbolic images, and artistic motifs are the materialized representations of ideology in nonstratified societies. When a new lifeway supplants an earlier one, the material culture changes; so also do the tangible components of an original ideology change when a new belief system replaces an old one. It stands to reason that male and female interactions and activities are altered as well. The challenge for archaeologists is to recover and describe these modified characteristics. Specifically, I suggest that archaeologists investigate historical transition-era sites known to be occupied through the period of missionization (see Chapter 6, this volume). HOUSEHOLDS WITH FACES Ruth Tringham (1991) has proposed that we think about “the faces” in ancient households. In other words, she wants archaeologists to visualize living people in the archaeological assemblages they discover. Many archaeologists, however, believe such descriptions of archaeological “faces” are fancies of the imagination. Most often they have not been taken seriously by those who approach the field “scientifically” and prefer testable models. And other problems exist. Archaeologists must guard against interpretations that are limited to analysis of sherds, flakes, bones, and botanicals. It is too easy to simplistically stereotype gender activities or activity loci. Households are diverse and complicated. Lines dividing men’s and women’s roles or space or activities within a household can be very blurry (see Chapter 9, this volume); moreover, the faces of children and elders need to be incorporated into the study of household organization and production. As Henry Stewart has discovered (Chapter 2, this volume), even personal gender identification can be called into question. If we want to see the full variety of roles present in archaeological communities, we must enrich our models and become more aware of the broad variability in gender relations and household groups (Tringham 1991:118). What does all this mean in the study of late-nineteenth-century Native Alaskans? The ethnographic record has rendered intensely clear pictures of traditional life in Native settlements (Dall 1870; FienupRiordan 1988; Nelson 1983 [1899]; Oswalt 1963a, 1963b; VanStone 1955, 1959, 1968; Whymper 1868). Today’s researchers have the rare opportunity to compare household organization before and after contact and

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

the abandonment of qasgis (men’s residential houses, frequently also used as community ceremonial buildings). This fundamental change in familial structure must have necessitated change in individual activities, whether men’s, women’s, or children’s. Thus when the social order was changed, domestic groups began to relate to each other very differently. It will not be easy to “see” shifts in male and female domestic roles archaeologically; however, thinking about the process at the household level will lead eventually to a clearer understanding of more general culture change. Our own daily personal interactions and behavior are varied and complex, dependent on those with whom we are associating and in what context. The interactions of people who lived “long ago” were not necessarily simpler. Therefore we must consider the probability of myriad interconnections when we are conducting archaeological investigations. NATIVE ALASKAN SETTLEMENTS During the early historic period the predominant characteristic of domestic life for the Kuskokwim Eskimos was that female and male activity spheres were centered on clearly defined separate dwellings (Oswalt 1963a:136).1 The men’s house, or qasgi2, was the center of the community (both physically and figuratively) in Eskimo villages. This structure was a man’s world that women visited on errands and for some ceremonial functions but in which they never lingered long (Oswalt 1963a:51). Several journal entries provide vivid word pictures of some historicera residential arrangements in western Alaska. In 1841 Lavrentiy Zagoskin described an Eskimo qasqi in the area of Norton Sound: It is in the kazhim that the men do all their work. Here they tan hides, weave fishnets, rig sled. Here all councils are held and decisions made on communal matters. The kazhim serves as a guesthouse for receiving visitors, a dining room for entertaining them, a hall for communal plays, a general dormitory for all the male population except infants, and finally as a bathhouse, for baths are one of the chief delights of all the Kang-yulit peoples. (Michael 1967:115)3

Although his account of the activities in and around small winter houses is not nearly as descriptive as those of the qasgi, he clearly specifies separate winter dwellings for the women and children (Michael 1967: 114, 121–122, 152, 193).

RITA S. SHEPARD

Nearly twenty-five years later, George Adams, a member of the Western Union Telegraph Expedition, gave us a peek into the inside of a winter house at Ulukuk, an Indian settlement on the banks of the Unalakleet River: “We entered these underground houses thru a small covered log passage, dropped down by ladder a hole for six feet and then crawled on hands and knees along an underground passage for ten feet into the main house of one room, which was twenty feet square with six feet walls on the four sides” (Adams 1982:46). He goes on to describe a fire that burned on the ground in the center of the room, with straw on which to sit or lie covering the earthen floor along the walls. When Adams entered the smoke-filled room, approximately twenty people already occupied half of it, and the odors were “complicated and awful. Besides the smoke there was an aroma of damp rotting decayed wood and rancid oil” (Adams 1982:46). In 1877 Edward W. Nelson manned a weather station in St. Michael, Alaska, for the U.S. Army Signal Corp; while he was there the Smithsonian Institution also engaged him to collect ethnographic material from the Native Alaskans (Nelson 1983:11). Thus he traveled widely in southwestern Alaska, collecting artifacts, myths, and stories, as well as documenting subsistence techniques, ceremonial practices, and building construction and use details. He described “family” dwellings as women’s houses that one to three families shared, each family independent of the other families living in the house (Nelson 1983:288–289). Each woman who is the head of a family has an oil lamp beside her sleeping bench where she sews or carries on her household work. Her own cooking utensils and wooden dishes for food, together with the stock of seal oil, dried salmon, and other articles of domestic economy, are kept at one side of the platform. . . . When the time approaches for the preparation of a meal, a fire is built in the middle of the room and the food made ready, after which each woman places a quantity in one or more wooden dishes, takes it to the kashim, and sets it beside her husband, father, or whoever she has provided for. (Nelson 1983:288–289)

By the turn of the twentieth century people who made their homes in Unalakleet and along the banks of the Unalakleet River had begun to live in aboveground log and plank cabins. The oldest resident with whom I visited in 1993 was born in Unalakleet in 1908. Until she was nearly ten years old, her family, along with about ten other Eskimo families, lived in log cabins approximately 25 km upriver from the village of Unalakleet

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

near the place still called Sauyaq (personal communication). She had never seen people living in semisubterranean houses. NINETEENTH-CENTURY MISSIONIZATION IN SOUTHWEST ALASKA In searching for information about how missionaries affected day-today household life in Alaska, I came across three excellent studies. Wendell Oswalt’s (1963a) ethnohistoric study of the missionization process in the Kuskokwim delta stands out and proved an invaluable resource. Thomas Correll’s (1972) linguistic research and Joseph Jorgensen’s (1990) examination of present-day Native people in three “bush” villages also provided useful data. THE MORAVIANS In an early historic-era Eskimo village, the domestic house was not the home of a Western-style familial household; men’s and women’s activities were dichotomized (Oswalt 1963a:147–148). In winter settlements, during a typical day men could be found sitting naked on the floor of the qasgi, carving or flaking wooden, bone, ivory, or stone into tools and ceremonial articles. The debris from their work would be scattered on the floor around them. After the day’s work the men took daily sweat baths in the qasgi, during which the older men often recounted old myths or tales of travel to distant and faraway places (Oswalt 1963a:55– 56). At mealtimes the women and young girls would bring food to their men in the qasgi (see also Dall 1870:406). Children accepted the fact that they were surrounded by women most of the time, with the men visiting only occasionally and for short times. In tundra camps (for summer fishing or fall hunting), where there was no qasgi, the men did live with their spouses, children, and other blood relatives; but even then frequent and extended hunting and trapping trips left little time for household interaction (see also Chapters 3, 7, this volume). Yet in a relatively short time after the arrival of the Moravians, the mission church had supplanted ancestral ceremonial uses of the qasgi in the winter village. Soon it became little more than an inn for travelers and a meeting place for the older men. Even the traditional Eskimo hot bath was replaced by the Russian equivalent conducted in an aboveground structure that eliminated smoke and in which women also frequently participated (Oswalt 1963a:148–149). The qasgi lost its fundamental meaning within the settlement. The men and boys began living in houses with the women and children, following the example set by the Moravian missionaries and their wives.

RITA S. SHEPARD

By the late nineteenth and early twentieth centuries, aboveground sod houses and log cabins had succeeded the semisubterranean dwellings. At first, the people living in one of these houses may have been an extended kindred group living as a single residential unit, but the nuclear family residence was the eventual result (Oswalt 1963a:149). In addition, with a firmly established Christian ideology, marriage became more formalized and immutable. Men, discouraged from living in the qasgi, began living with women under a much more formal contract of marriage, one that considered premarital sex and adultery sinful acts. It soon was commonplace for the bride to move into her in-law’s house while she and her husband built a separate dwelling. New social relationships had to be forged to accommodate the new interactions between husbands and wives, couples and in-laws, parents and children, and siblings of opposite sex (Oswalt 1963a:149–150). Children attending mission schools had taken on the new role of translator between their parents and the new American settlers (Oswalt 1963a:158). Change from traditional lifeways was relatively rapid. The first Moravian missionaries founded their station in Bethel in 1885, and by 1894 they had suppressed Eskimo traditional life to the degree that the major traditional annual ceremony for the dead was not held at any of the largest villages on the lower Kuskokwim River (Oswalt 1963a:157). Nevertheless, it was sixty years from the time the Moravians arrived until the last qasgi was abandoned along the central Kuskokwim River (Oswalt 1963a: 148–149). “When the last kashgee [men’s house] in any community was neglected, then abandoned and finally torn apart for firewood, the old way of life lingered only in the memories of the people” (Oswalt 1963a:51). SWEDISH COVENANT MISSION CHURCH In 1887 Axel Karlsen arrived in Unalakleet and established the Swedish Covenant Mission Church; by 1889 the school had been started as well. He provided medical services and encouraged residents to cultivate gardens, teaching them to plant potatoes, onions, and other coolseason vegetables. In his capacity as the healer of the soul and deliverer of the Christian message, Reverend Karlsen also suppressed and finally eliminated traditional plays, feasts, dancing, and songs (Jorgensen 1990:63). Karlsen and his church brought the people a new value system (Correll 1972:82). The qasgi in Unalakleet was also the hub of the village; it was the point where all information chains intersected. In early-twentieth-century Unalakleet, even though Karlsen had declared himself opposed to

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

both the concept and the structures, two qasgis were still used for nontraditional social activities, such as playing cards, as well as for the more traditional purposes of tool making and information exchange (Correll 1972:199–200). Eventually, however, the Unalakleet qasgis were abandoned, and the community allowed them to deteriorate and crumble away. No sign of either of them remains in the village today, although some elders can point out where they once were located (for more discussion of qasgis, see Chapters 3, 6, 7, 9, 11, this volume). The residents of present-day Unalakleet live in single-family houses, the children attend a large public school, medical and dental clinics provide health care, two general stores vie for business, and the Covenant Church remains as one of the largest buildings and most influential organizations in the village. Neighbors now meet each other at church, in the schools, or at the stores. ARCHAEOLOGICAL CONSIDERATION OF GENDER IN ALASKA The ideological shifts that resulted from the acceptance of Christianity seem to have led to community reorganization and the transformation of the physical structure of houses. These changes must have also brought about changes in roles and relationships within the household. This section begins with my own research that uncovered gender-specific loci in the archaeological context of a traditional domestic system. Then I examine two other archaeological investigations that also considered changes in roles and relationships as cultural mores changed (Ackerman 1970; Clark 1996). UNALAKLEET RIVER ARCHAEOLOGICAL PROJECT My research examined traditional Native settlements situated on the banks of the Unalakleet River. My field project included extensive excavations at one site (Tagilgayak) out of the four I surveyed (see Figure 5.1). I have interpreted the site as having one qasgi (P3) and two winter houses (P4 and P10); we also excavated storage structures (P1 and P5) associated with the winter house and a small enigmatic houselike structure—possibly a temporary residence (P9). Historical trade items greatly outnumbered the locally made artifacts, and ceramic chronology of English transferware placed occupation in the mid- to late 1800s. I identified a qasgi based on the size of the structure and tools normally attributed to men (see Figure 5.2). The building was significantly larger than the other two houses at Tagilgayak, and it contained the only fishing-related equipment found at the site: two bone fish arrows and a net sinker (see Figure 5.3). Additionally, although berry remains were

RITA S. SHEPARD

Figure 5.1 Excavated features at the Tagilgayak Site (UKT-022) on the Unalakleet River, Alaska; hatch lines indicate unexcavated features

found in the smaller structures, the higher density and wider variety of berry remains recovered from the building suggested that ceremonies took place there (see also Dall 1870:153; Osgood 1958:73–146). We must remember, however, that archaeology is not an exact science; artifact context and provenance can be equivocal. Like artifacts from the houses in A. McFadyen Clark’s study (1996), those found in the buildings at Tagilgayak can be interpreted in several ways. An antler handle for an ulu blade (i.e., a woman’s knife) was recovered from the (probable) qasgi. If it was used there by a woman, it could mean the structure was not a qasgi but rather just a larger winter house. On the other hand, a man could have made or repaired the knife in the qasgi, supporting the argument that it was a men’s house. Moreover, as Tobey (Chapter 6, this volume) points out, the house and the qasgi (or kashim in the dialect of

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

Figure 5.2 Structural remains found in Features P3 and P9 at the Tagilgayak Site; gray outlines in P3 show features uncovered in the lower Level 2 excavations. Drawing by Corinna M. Schoenfelder.

the Deg Hit’an) are not discrete entities but part of a dynamic sphere of activity and interaction across the settlement. To understand the qasgi data better, archaeologists need more comparative material from ceremonial structures, but very few qasgis have been excavated. The main living area of the house structures at Tagilgayak ranged in size from 32.5 m2 to 7.0 m2 (Shepard 1997:37) (see Figure 5.2). The smallest building (P9) presented the biggest challenge to interpretation, and gender played a role in the analysis (see Shepard 1997, appendixes B, C, F, G for data). There was no indication of a delineated sleeping area or bench, but an ephemeral hearth was detected in the center of the floor. Botanical analysis found that this small house had a very high density of grass seed, over ten times that recovered from the large structure thought to be a qasgi (Shepard 1997:262). Botanists also identified several types of berry seeds from its float sample; faunal analysis revealed caribou and bird bone in this small dwelling. Glass trade beads dominated the artifact collection from this unit, complemented by fragments of European ceramics and a metal knife. This combination of cultural material suggests that the building may have been used by women as a seclusion space, either for birthing or at

RITA S. SHEPARD

Figure 5.3 Upper: Stone net sinker, UA-92-129-28, recovered from just outside the walls of the NW corner of P3 at Tagilgayak; lower: Carved bone fish arrow, UA-92129-43, with metal point (lower right) found under a plank on north wall of P3 at Tagilgayak. Drawing by Corinna M. Schoenfelder.

the onset of menstruation (Clark 1981:591; Giddings 1961:153; Nelson 1983:289; Reinhardt 1986:152–156). This scenario entails the accumulation of food remains and a high density of grass to be used either as bedding or for making grass mats or baskets, since young women at their menarche began to learn domestic skills (Clark 1981:591; Giddings 1961:154). Decorative glass beads fit well with this hypothesis and would be expected, too. Other interpretations can be constructed, but this one

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

best unites the archaeological data and the ethnographic record (Clark 1996; Dall 1870; Fienup-Riordan 1988; Nelson 1983; Osgood 1958; Michael 1967). CAPE NEWENHAM In 1966 Robert Ackerman questioned the revision and retention of physical characteristics of Native residences in the face of cultural change. He considered the households and household goods found during three periods of occupation: (1) archaeological context at sites on Chagvan Beach, (2) the late–nineteenth-century historic village of Tzahavagamute on Cape Newenham (from data collected in government, missionary, and ethnographic reports), and (3) living people in the modern Native community of Goodnews Bay. He redirected his research from object recovery to feature analysis within these specific cultural frameworks (Ackerman 1970:15–17). Ackerman hoped his methods of comparison would help archaeologists to establish activity spheres and study the stability of cultural patterns from one time period to the next (1970:42). Over the years, the village residential pattern in Goodnews Bay has shifted gradually toward the store, American school, post office, and church (Ackerman 1970:36). The professed reason many households moved closer to the new “center” was “to be nearer the water” to make washing clothes and personal bathing easier. Ackerman does not question this motive. It seems more likely to me, however, that these institutions, taken as a whole, had replaced the village qasgi. Together they had become the new community center. And as along the Kuskokwim and Unalakleet Rivers, people met to exchange information, make group decisions, and worship within the new ceremonial structure. Next, Ackerman wondered whether distribution of household items could identify gender-linked use of space within a modern house (an aboveground single-room building) that was home to a family of five. He planned to compare archaeological distributions to modern use and work patterns. To do this he compiled a plan view of one present-day home indicating the storage area of tools specific to women’s and men’s activities, finding fairly direct correlations between gender-specific tools (that suggest particular behavior) and associated activity or storage areas (Ackerman 1970:37–39). Gender mixing, however, appeared in work zones and in tool use (Ackerman 1970:40). Frequent forays by either parent outside of the house meant both men and women often had to be completely self-sufficient and able to use tools or take over tasks more commonly performed by the opposite sex (e.g., cooking, fishing, sewing). Therefore if people living

RITA S. SHEPARD

in the past engaged in similar behavior, an archaeologist might have a hard time trying to prove female or male activity at prehistoric sites (see Chapter 9, this volume). In spite of these nonconformities, Ackerman believed the study of modern behavior patterns enabled his group to better discern gender-specific activity areas in a sixteenth-century house pit at the nearby prehistoric site on Chagvan Bay (Ackerman 1970:40–41). THE KOYUKON: ALASKAN ATHAPASKANS In her recent book Who Lived in This House? Clark (1996) documented three nineteenth-century traditional houses in three different winter occupation sites along the Koyukuk River. In addition to providing a comprehensive analysis of each building’s architectural elements and construction, she considered the spatial distribution of the artifacts found in each house. Originally, Clark planned to compare the distribution of objects and faunal remains to information she had collected from informants about use of space and household organization. Clark’s idealized representation of a semisubterranean house, based on her excavations and personal interviews, had a central hearth and hosted two domestic groups. Each group occupied half of the dwelling; the female children were situated closest to the entry tunnel, followed by the wives and babies, and the husbands resided in the center on either side of the hearth. Elders and unmarried adults lived in the back of the house (Clark 1996:145). She tried to compare her artifact collections to the theoretical spatial arrangement within the house. Unlike Gregory Reinhardt (Chapter 9, this volume), however, her artifact assemblage was small, making comparisons inconclusive—both among the three houses and between houses and oral traditions. Clark created a composite floor plan from two of the excavated houses. She plotted the distribution of two selected groups of artifacts that comprised the majority of items recovered, hoping they might reflect gender-specific use (Clark 1996:142–144). She included scraping stones, metal scrapers, baskets, and a snowshoe needle as probably female associated; ammunition, powder cans, a strike-a-lite, and other iron objects were attributed to men (Clark 1996:153–154). The resulting distribution pattern fit reasonably well with the use of space described by the local elders. Patterns could not be seen, however, in the faunal material, which seemed randomly dispersed around the living area (Clark 1996:158–159). In fact, particularly telling was the fact that excavators found bear bones randomly distributed around the house floors, including areas traditionally allocated to women and children. Additionally, a single lynx scapula

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

was recovered from the woman’s area in one of the dwellings. Clark makes special note of this find because neither bear nor lynx bones would be expected in a traditional Athapaskan household with women in residence. Bear meat, bones, and hide are strictly forbidden to premenopausal Koyukon women, as is lynx, and would never be taken into a house where women were present (Clark 1996:5–6, 159, 203). The appearance of animal bones normally prohibited in the presence of women suggested several social scenarios—all dependent on knowledge of present-day gender roles within the group. The simplest explanation is that the people living in these houses were not Athapaskans; they may have been Inupiat who did not hold gender prohibitions against the bear and lynx. Koyukon oral tradition indicates, however, that both male and female shamans were exempt from taboos and restrictions; thus the houses may have been occupied by shamans. Alternatively, elders (including women past menopause) were exempt from food taboos, implying that perhaps only old people lived in the houses. Then again, severe food stress would allow personal survival to take precedence over cultural rules and mores; people would eat anything available rather than starve. Finally, the houses may have been inhabited after long-term exposure to Russians or other Europeans, and the inhabitants may have already begun to accept new beliefs and ideologies, abandoning tradition (Clark 1996:204–206). ANOTHER NORTHERN EXAMPLE OF THE EFFECT OF CULTURE CHANGE ON GENDER ROLES Nearly two decades later and worlds away, in an insightful investigation of household change (or what she calls changes in “the internal domestic arena”), Ellen Pader (1988) looked at how objects, organization of objects, and a man’s or woman’s relationship with objects within a household reflect social action and cultural mores. She studied two very different cultural groups: the nomadic, pastoral Mongolians, and British “Traveler-Gypsies,” arguing that the spatial appropriation within a specific locale brought together society and its structures (Pader 1988:251). The Mongolian research specifically addressed issues of gender that can illustrate identifiable changes in material culture in the wake of socioideological change. Travelers’ journals dating back to the thirteenth century detailed the placement of furnishings within a Mongolian domestic tent, or ger, and described the ideological values assigned to objects and spaces inside the living area. The western part of the tent and its contents were associated with men and held higher status than objects located in the eastern

RITA S. SHEPARD

and female portion of the dwelling. A north-south (or upper-lower) division also indicated status; thus the upper western quadrant held the highest honor (Pader 1988:255). After the Russian Revolution, the socialist government mandated a change in the lifestyles of the nomadic peoples of the high desert. Studies found that, although the people had obtained many new goods and tools (e.g., sewing machines), they placed them traditionally within the ger, retaining the preestablished symbolic order. Status assigned to women and children underwent observable transition, however (Pader 1988:256– 257). The children who used to sleep on the floor at the head of the parents’ bed moved into their own bed on the male side of the tent, apparently indicating the high importance the Soviet state placed on future generations. Moreover, seating arrangements of male and female children were no longer distinguishable from each other, reflecting a change not only in social relationships but in future power. Perhaps most significant, books became a sign of social change. Prior to the revolution they were the purview of senior men and religious leaders, and women were forbidden to read them. When the Soviets made literacy a priority, women not only began to read, but books began to be kept on a shelf at the head of the bed on the woman’s side (Pader 1988:259). Looking “inside” domestic spatial relations leads to a clearer understanding of the larger process of learning, using, interpreting, and reinterpreting the values and codes of a society (Pader 1988:265–266). Pader argued that members of a domestic group are also part of a spatial interaction encoded by the social group, and the organization of space within a home is part of the social structure. Therefore when a society adopts new socioideological practices, its social roles and structure change and are reproduced by and through the individuals within it. The resulting “recursive relationship between ideology, action, and spatial relations is intensified” as each reflects the others (Pader 1988:266). The Mongolian study should spur archaeologists to search for material evidence that will substantively demonstrate these kinds of sociospatial interactions (see Chapters 9, 10, 11, this volume). ARGUMENT Originally I asked whether archaeologists perceive macroscale social change at the microscale level of the household. Eric Wolf (1982:19) has suggested that cultural forms are the “determinate orderings of things, behavior, and ideas” and as such play a major part in guiding human association and interaction. Thus cultural forms influence, and maybe even mold, social relationships.

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

Therefore I suggest the following sequence of events: (1) New things appeared in the community (“community” includes several settlements, linked by their interactions, within a specific geographic area). They were either completely new introductions or traditional items produced by “new” technologies or made of new materials. (2) New people arrived with the new goods (and controlled the access to them), and eventually their different ideas and lifeways permeated the aboriginal domestic systems, affecting the people living in their traditional modes. (3) Native people modified their dwellings to facilitate their changed lifestyle and newly adopted materials. After ideas changed, behavior changed. In due time Alaskans expected their community to include a store, a school, health care, and a church. The ethnographic literature has shown that as the Native population began to accept Christian doctrine and mores and live by U.S. law, they also began to abandon their semisubterranean domestic system and gradually moved into Western-style log or plank cabins (Correll 1972; Jorgensen 1990; Lenz and Barker 1985; Oswalt 1963a, 1963b, 1990; Porter 1893). Thus in western Alaska, as new people (Russians, Americans, Scandinavians) were integrated into the community, their ideas about education and religion also began to be accepted. Fundamental concepts of what made up a community or a settlement or even a proper house and household changed, too. Eventually, wood stoves that radiated better heat replaced open hearths, reducing the physical need for the insulation of semisubterranean houses. Ready-made tools precluded the need to spend time making them in the qasgi; wool and cotton fabrics for clothes released women from much of the time spent preparing skins. Moreover, to have easy access to the new goods, some groups moved their entire settlement closer to the trading posts. Studies to determine gender activity areas are becoming more common. Traditional Native Alaskan wintertime activities took place within the entire domestic system—inside the qasgi and the small winter house and outside, in work areas adjacent to buildings and between structures. Therefore changes in the amount and place of work should be visible in the archaeological record. Archaeologists need to consider social changes that might have accompanied such movement. Archaeologists might learn a great deal about transitional social organization and household interactions if we would examine transitional dwellings built in settlements occupied around the turn of the twentieth century (see Chapter 7, this volume). Alaskan archaeologists need to relate material recovered from semisubterranean structures during the

RITA S. SHEPARD

late-nineteenth- and early-twentieth-century period to artifact and feature loci in the earliest aboveground cabins. These data would help to identify changes in work places and habits. Thus more homes need to be examined and compared with the ethnographic literature (see also Chapter 6, this volume). RECOMMENDATION FOR FUTURE STUDIES All of this raises the question, now what? As always, we must start with standard archaeological method. Without data we cannot interpret anything. But we must try to remember that actual people lived at our sites; they ate and slept, laughed and cried, argued and agreed, made and broke tools, talked about ways to get food, and made love. They participated in all of the day-to-day trivia that we do. We must consider these activities and relationships when formulating our research models. To do this, archaeologists must think about how people might be likely to behave in certain situations, then use all the research tools available to them—especially ethnographic material and historical records—to find pattern changes that may indicate social change (Tringham 1991:121–124). Clark may have raised more questions than she provided answers to regarding who lived in the Koyukuk River houses, but she did propose “faces” for those houses. She may not have known exactly which faces belonged in her houses, but she could paint pictures of real people—working, celebrating, eating, sleeping, and perhaps changing within the parameters of a household. Could it be that Native Alaskan households moved into aboveground cabins not so much because the Americans brought with them the cast iron, pot-bellied stove that provided much more heat much more efficiently than a small open fire, but rather because, in the Christian faith, married couples were expected to live together in one house with their immediate family? And what were the spatial implications when women then had to share their work space with men? Did less space cause activity areas (e.g., for flaking and grinding stone tools, preparing and cooking meat, sewing skins) to become smaller and more clearly defined, more diffuse, or overlapping? Did tools once kept in separate buildings now commingle in one, or were they stored in very separate places in the new house? Was work once done in the qasgi or house now done outside? What happens to ordered space as seasons change? Did outside work areas take the place of the qasgi floor? Were more outbuildings constructed for storage or to act as special work spaces? The list can go on and on, but what is important to remember is that, with any of these changes, personal communication and spousal interactions change, too.

CHANGING RESIDENCE PATTERNS AND INTRADOMESTIC ROLE CHANGES

Imagine the woman saying to the man, “Take that ivory outside to work— the chips and dust are getting in the food.” Traditionalists (and cynics) may say we are playing at storytelling and that, as archaeologists in the new century, we can never know how the people felt who were living and working at our sites 100 or 1,000 years ago. Indeed I have said so in the past. My recent work at sites where people have modified social organization, switched trade partners, added Western foods to their diet, and adopted new ideology, however, has led me to believe that archaeologists can discover and know a more intimate level of change—if only they will look. ACKNOWLEDGMENTS The United States Bureau of Land Management in Anchorage provided excavation permits and partial financial support for my research. Further financial support was contributed by the National Science Foundation, Arctic Social Sciences Program: Division of Polar Programs (DPP 9216347). The Cotsen Institute of Archaeology at UCLA provided me with the time and space to write and edit this chapter. Special thanks to Jeanne Arnold for her many insightful comments. Thanks also to the anonymous reviewers for valuable suggestions on earlier drafts of this chapter. Much appreciation goes to Corinna Schoenfelder, who produced the artifact and house excavation drawings. I am especially grateful to Lisa Frink for the invitation to be a part of the original symposium that inspired this book and for encouraging me to look at my research with an eye to gender. I thank Greg Reinhardt for years of friendship and guidance, as well as for his valuable comments on this work. Finally, my sincerest thanks to John Schoenfelder, my colleague, friend, and partner, for his computer mapping skills, his thoughtful critiques, and his unconditional support.

6 RE-PEOPLING THE HOUSE HOUSEHOLD ORGANIZATION WITHIN DEG HIT’AN VILLAGES, SOUTHWEST ALASKA JENNIFER ANN TOBEY

THE HOUSEHOLD IS A FUNDAMENTAL SOCIAL FORM within human society; it embodies many dimensions of a society’s social activities. Examinations of the household can provide insights into economic practices, symbolic meanings, class relations, status differentiation, kin relations, and gender relations. Using a household approach to highlight gender organization in Deg Hit’an society can prove useful to understanding the process of missionization in Alaska and in creating an interpretation that depicts people as dynamic and interactive. The process of missionization among the Deg Hit’an of southwest Alaska has not been explored in detail at the household level. This chapter offers an approach that focuses on the spatial and temporal organization of the household. Using this approach, I illuminate the importance of research at the household level and provide a brief analysis of the premissionary household. I then discuss important aspects of the missionary period that would have influenced life at the household level and address the need to explore these different aspects in order to understand the routine life of the Deg Hit’an during this period. A considerable lack of accessible information exists concerning Deg Hit’an social dynamics at the household level during the American missionary period.1 In general, studies have focused primarily on broadscale approaches (see Mitchell 1997; Simeone 1982; VanStone 1974, 1979a,

JENNIFER ANN TOBEY

1979b). Cornelius Osgood’s work in the Anvik-Shageluk area (Osgood 1940, 1958, 1959) is an example of a study conducted at the village level, although his research focused on premissionary rather than missionary life.2 All of these studies are informative, providing useful comparative and background data to incorporate into household studies. When we apply a household approach, however, the formerly obscure mundane and integral aspects of Deg Hit’an culture are illustrated. This chapter emphasizes the need to examine these integral aspects and broaden our understanding of the American missionary period in Alaska. I suggest that the way to do so is to use a household approach that permits us to examine the changes and consistencies of daily practices, kin relations, and gender relations and roles. Furthermore, the house is a social space represented as a fluid entity whose composition, meaning, and function are transformed during the day. To understand the effects of the missionary period, one must first examine the premissionaryperiod household and use this as a baseline of information. In what follows I describe many facets of the household and discuss the importance of a household study. Using Osgood’s data, I offer a brief household analysis, exploring the spatial and temporal organization of the household and the house using a sociospatial lens. By describing a household that was dotted across the village landscape,3 its members active throughout the day and the year, I show that the house itself transformed, taking different meanings and different roles at various times of the day. In addition, I show that three household-type groups were found in Deg Hit’an society. These groups overlapped, as individuals were members of more than one group. With a dynamic view of premissionary village life as seen at the household level, I set the stage for future examinations of the missionary period, highlight some of the forces that may have affected the household and gender roles, and suggest possible avenues of research. HOUSES AND HOUSEHOLDS The house is an important structural element in human societies. Archaeological remains of houses often appear as synchronic signatures; however, through the course of its use life, the house is a diachronic entity. One way to access this diachronic nature is to examine the social group associated with the house and house locale—the household. To envision the household of the past as a dynamic entity, we must first recognize the complexities of that entity and work with them accordingly. Archaeologists utilize a multitude of evidence to reconstruct the household of the past. As Carol Kramer (1982:673) described it, archae-

RE-PEOPLING THE HOUSE

ologists use “architecturally bounded spaces, patterns of circulation within and among them, and structural and perhaps artifactual redundancies within them to construct the number and possible relations among the inhabitants of those spaces, which they may term households.” Although “bounded spaces” and “structural and artifactual redundancies” may be synchronic, “patterns of circulation” implies movement of household members—a dynamic characteristic. The numerous forms of evidence we study indeed suggest that the household is a complex entity. All too often, however, the household is reduced to a simple unit of analysis, such as a production unit or domestic unit (for example, see Sahlins 1972; Kramer 1982, respectively). Focusing on a particular aspect or function of the household leads to a manageable analysis, but it often excludes important elements. Although researchers are aware of the multiple aspects, representations often continue to reflect households as simple units, and this can lead to a static view of the household. The difficulty of presenting the household as complex and dynamic is compounded for archaeologists, who must rely on material remains as a data source. Examining households in terms of activity patterns has been useful to archaeologists who cannot see the people they study (e.g., Flannery and Winter 1976; see also Chapters 9, 10, 11, 12, this volume). When we study the house, associated artifacts, and surrounding locale, we are indeed looking at an assemblage of static remains. It is our job through interpretation to put the dynamic life back into these remains. The first step is to recognize that the house and the household have multiple meanings and functions (Bailey 1990). Because the house assemblage is intimately related to the household, we must acknowledge that the assemblage that represents a dynamic social unit is itself dynamic. For example, when we look at a house and its associated artifacts and features, we must look at the entire assemblage as a single categorical unit. We must acknowledge that just as the house/house locale serves many functions, the household served many functions and filled many roles. Furthermore, examining the spatial and temporal organization of household activities can reveal the complexities of the house and household—for instance, that activities overlap or vary through time, both in a cyclical or seasonal sense and in a progressive sense. GENDER At the most basic level of interpretation, gender is a social category that is based on biological sex, age, and sexuality and is specific to a particular culture. It is not analogous to biological sex but rather is distinct

JENNIFER ANN TOBEY

from it. Sex is a fixed quality unrelated to culture, whereas gender is “a thoroughly socio-cultural creation” (Pollock 1999:24; see also Nelson 1997:15). Gender roles and classifications may indeed be accepted as natural, predetermined, and unchanging within the society to which they belong. Thus examining gender highlights how culture is organized. Although it may relate to sexuality, roles, activities, behaviors, and responsibilities to others, gender is more than a mere classificatory device; it is “a process that is constructed as a relationship or set of relationships, necessarily embedded within other cultural and historical social institutions and ideologies such as status, class, ethnicity, and race” (Conkey and Gero 1991:9). Moreover, gender varies from culture to culture, both by specific gender designations and by the number of different genders present in society (Nelson 1997). Gender categories may—and often do—consist of more than men and women. The roles and behaviors of unmarried young women or girls may be significantly different from those of married women, and thus unmarried young women may be considered a different gender from married women in certain cultures. Gender, like the household, is an important organizational principle that transcends multiple aspects of society (Conkey and Gero 1991; Tringham 1991; Yanagisko 1979). Both gender and the household are embedded in, as well as imbued with, multiple characteristics of society. Therefore an approach that explores both gender relations and the household will reveal many aspects of society, and will produce a dynamic integrated view of culture at the village level. I suggest focusing on the organization of household members by exploring the overlap as well as the separation of activity patterns. It is necessary to keep in mind that these activities are not segmenting but rather are transforming the village landscape. For instance, the overlapping activity signatures may give us insights into how activities changed through time—daily and seasonally—and thus into how the house space transformed through time. Furthermore, since we are interested in the social unit, we must follow the members as they came to and went from the house. This means studying activity patterns throughout the village, as well as those in and around the house. With the use of ethnographic and ethnohistoric records during missionization, this avenue of study is feasible (see also Chapter 5, this volume). A sociospatial approach integrates the physical environment with spatialized experience within the term space. Thus space becomes both medium and outcome of social practices (Giddens 1984; Saunders 1990). We must view the house as an entity/artifact in itself. An individual’s experience of space occurs through the dialectical relationships among

RE-PEOPLING THE HOUSE

the physical locale, social practices and meanings, and the phenomenological (Tilley 1994). Moreover, different individuals experience space differently based on differences in social practices and relations that correspond to age, status, and gender (Soja 1989; Tilley 1994). We must consider how the house incorporates all of these different functions and meanings. THE DEG HIT’AN The Deg Hit’an are an Athapaskan-speaking people from the southwest interior of Alaska. Around the time of contact, Deg Hit’an villages were established along the Kuskokwim and Yukon Rivers and their tributaries (de Laguna 1947; Osgood 1940, 1958, 1959; VanStone 1979b). These rivers were an important resource, supplying a variety of fish and waterfowl and serving as transportation and communication links between villages (Osgood 1940, 1958, 1959; Snow 1981). Because the resources of the Kuskokwim were less rich than those along the Yukon, the Deg Hit’an along the Kuskokwim depended more on hunting and less on the rivers than the Yukon Deg Hit’an did (Snow 1981:602). Native peoples in Alaska had a well-established trade network with people in Siberia, and indirect contact with the Russians began with the introduction of Russian material items around 1789 (Griffin 1996; Michael 1967:100; Simeone 1982; Staley 1992:16). Trade and contact increased when Russian fur traders reached the Aleutian Islands and later extended into the Alaskan mainland. Members of the Russian Orthodox Church also traveled through Alaska. The Deg Hit’an may have been less preoccupied with trade than other Alaska Natives (see Michael 1967; Snow 1981), but they were not untouched. Even the smallest element of contact would have had an influence. After the United States purchased Alaska in 1867, the lives and practices of the Deg Hit’an, and of Alaska Natives in general, were affected by U.S. fur traders and missionaries. Unlike the Russian American Company and the Russian Orthodox Church, the American Commercial Company and U.S. churches established trading posts and missions within Deg Hit’an territory. As a result, Americans had a bigger impact than the Russians on Deg Hit’an daily life. The Deg Hit’an began to depend less on traditional means of subsistence and began to rely on U.S. commodities (VanStone 1974, 1979b). PREMISSIONARY LIFE The Deg Hit’an occupied three types of villages during the year: the winter village, the canoe village, and the summer village. These different

JENNIFER ANN TOBEY

villages were located relatively near each other, with the winter village being the base camp (Snow 1981). Canoe villages consisted of ephemeral shelters, and summer villages were closer to the winter village and were smaller than the canoe villages (Snow 1981:604). Summer villages were characterized by aboveground plank or bark houses, smokehouses, and caches (Osgood 1940). Deg Hit’an winter villages were characterized by semisubterranean houses, caches, and a kashim—commonly defined as a men’s house (de Laguna 1947; Osgood 1940; VanStone 1979a, 1979b; see also Chapters 3, 5, 7, 11, this volume). I will focus on the winter village. Women told stories in the house, and men told stories in the kashim. Numerous games and ceremonies were held, including the Animal Ceremony and the Mask Dance (Osgood 1958). Men, women, and children of the village, and sometimes those of other villages, gathered for ceremonies; during such time the kashim was the center of activity for the entire village. Osgood claimed the house belonged to married women (Osgood 1940, 1958); however, it is not clear if this ownership was shared among all married women or was limited to a specific woman within the household. Even so, it served many members of the community. Two to three families were associated with one house, which stored material items for the many household members. At the start of the day it housed a large group of people, as families gathered to eat their morning meal, after which men and boys left for the kashim (or for hunting, trapping, and similar activities). Women and girls remained—sewing, cooking, making pottery, talking, and telling stories. At night, women, their husbands, daughters, and young sons slept in the house; and occasionally the house functioned as a council place for women of the village (Osgood 1940, 1958). While women worked and socialized in the house during the day, men worked in the kashim. This space was the focus of men’s lives in the winter (Osgood 1958:33), where they socialized during the day and evening. Here they manufactured and repaired items such as traps, wood bowls, and arrow shafts (Osgood 1958). In the late afternoon they built a large fire and took a sweat bath. Later, women brought the evening meal to their husbands and unmarried sons. After the meal, men and women listened to men tell stories. Like women in the house, old men occasionally held council in the kashim (Osgood 1958). ORGANIZATION OF SPACE From ethnographic research and archaeological discussions, we can recognize a division of labor and social relations based on gender (see Osgood

RE-PEOPLING THE HOUSE

1940, 1958; Snow 1981). Within married couples, men provided the majority of food, whereas women cared for children, prepared food, and obtained food from areas near the village (Snow 1981:611). According to Osgood, women and men manufactured different items. Women manufactured twisted lines, tanned skins, nets, grass baskets and mats, birch bark baskets, skin containers, pottery, dry fish, and clothing. Men made awls, chisels, gouges, thimbles, scrapers, knives, axes, adzes, wedges, mauls, grinding stones, weapons, hunting implements, fishing implements, wood dishes and bowls, ceremonial items, and skin lines (Osgood 1940:440–441). These different social roles and activities were then transcribed onto the landscape. For example, as mentioned, women spent their days in the house, and the men were in the kashim. It would be easy to create a dichotomy and categorize the house as women’s space and the kashim primarily as men’s space. Following this spatial distinction, archaeologists would expect to find women’s artifacts in the women’s space and men’s artifacts in men’s space. The organization of space within Deg Hit’an communities, however, was more complex than this simple distinction (for further discussion of gender and spatial analysis, see Chapters 9, 11, 12, this volume). Physical places took on different meanings because they housed many different activities during the day. Examination of the house further supports this view of complexity. As mentioned previously, the house was considered to be owned by women; yet both men and women built the house (Osgood 1940:432–448). It housed men, women, and children, its composition changing throughout the day. Also, men and women kept their belongings in it. The roles and functions of the house changed as the household members moved throughout the village landscape. For purposes of this chapter I define the household as the group of people regularly associated with a particular house-type structure. With this definition, three different household-type groups can be distinguished. The first is the multigender household with married men, married women, unmarried men, unmarried women, and children. This consisted of all members of the two to three families associated with a particular house (see also Chapter 7, this volume). Although the married women owned the house, men took part in deciding who would make up the social unit that lived together. Osgood noted that men who were hunting and/or fishing partners often shared the same house (Osgood 1958). This household group met for the morning meal but did not remain spatially bound together throughout the day. These individuals had particular relationships, roles, and responsibilities with respect to

JENNIFER ANN TOBEY

each other; and the house was a social and physical setting in which these activities and interactions took place. After the morning meal, the house was transformed from a space for a multigendered group to a place occupied by married women, unmarried women, and sometimes children. This second household group consisted of the women and girls of the original two to three families. Although part of a larger household group, they formed another unique group that was based on their association with the house. As they conducted their daily activities within this structure, they formed relationships specific to their social group. The third social group of interest here is that associated with the kashim. The kashim-hold included the married and unmarried men of the village (members of this group were also members of household groups) who spent their days manufacturing and repairing tools. Again we see a group in spatial flux. In the morning its members were dispersed at their wives’ and mothers’ houses, and most men reassembled in the kashim during the day. Then, in the evening many left for the night. As we see, group membership overlapped with respect to the landscape. The three social groups were not only dynamic across the landscape and through time but were also closely linked. Examining the landscape with a social lens, we see that places—such as the house and the kashim— are not discrete, isolated entities but are tightly woven into society. REORGANIZATION IN THE MISSIONARY PERIOD American missionaries introduced new practices and deliberately tried to persuade the Deg Hit’an to change some of their traditional practices, such as discontinuing their feasts (Simeone 1982). A shift in village residence took place as people converged on the larger communities of Anvik, Holy Cross, and Shageluk4 (VanStone 1979a). Boarding schools were established at all three villages and enrolled children from neighboring villages (Mitchell 1997; Simeone 1982; VanStone 1979b). Because they removed members from the household groups for significant lengths of time, the schools impacted the groups’ labor and activity forces. Church missions were established at Anvik and Holy Cross, and the missionaries worked to convert people and educate them in the ways of American life (Simeone 1982). The missionaries played a significant role in attracting people to these villages. As a result, the household group may have reorganized tasks, which in turn impacted the spatial as well as social organization of gender-differentiated activities. Missions at Anvik and Holy Cross affected spatial organization in the Deg Hit’an landscape by creating physical distinctions between mission

RE-PEOPLING THE HOUSE

and nonmission land. In Anvik, Natives who became Episcopalian moved across the river from the village (Chapman 1896:523, cited in VanStone 1979a:77). The mission charged residents living on mission land a dollar a year and emphasized that undesirable people were to be kept off that land (Simeone 1982). This created the idea that the mission land was a privileged place, and by 1900 all Anvik residents lived on that land (Simeone 1982). At Holy Cross tension existed between the Roman Catholic Church and a number of Deg Hit’an residents. The latter, led by an influential shaman, moved across the river from the mission; the others moved onto mission land with Father Lucchessi (VanStone 1979b; Simeone 1982). House form also changed as a result of the American mission period. In Anvik, for example, all inhabitants lived in traditional semisubterranean houses as late as 1892 (VanStone 1979a:77; see also Chapter 4, this volume). Five years later two-thirds of the people lived in aboveground log houses on the mission land (VanStone 1979a:77). Although this change may have taken place only in the structure itself, the household composition may also have changed (see Chapter 5, this volume). The missionaries influenced many aspects of Deg Hit’an life, and it is likely that the role of the house was affected, too. What other changes occurred with respect to spatial organization and role of the house have yet to be examined; however, by appraising how the missionary period is reflected in the house, we can gain an understanding of how society was affected at the household level. With the increased presence of the church, the importance of the kashim declined (Osgood 1940, 1958). The disappearance of the kashim may have had significant impacts on ceremonial life, social organization, and gender-differentiated activities. James VanStone summarized the demise of the kashim: The decline of the Ingalik ceremonial cycle obviously affected the importance and function of the kashims in the various settlements. As workshops and lounging and sleeping places for men, they would continue to be important for some time to come, but once the ceremonial aspects of kashim activities were diminished, the structure as an institution in Ingalik life would never be the same again. The kashims at Old Shageluk and Holikachuk continued as meeting houses until the villages moved in 1966 and 1963, respectively. The structure at Anvik was destroyed by fire in the 1940s and not rebuilt. (VanStone 1979a:76)

The continued use of the kashim by men suggests that a division of labor based on gender continued. That division would have changed,

JENNIFER ANN TOBEY

however, and, at the least, spatial organization, which was based on gender, continued. Moreover, the activities in the kashim may have changed, but, unfortunately, the daily role of the kashim during the American missionary period has yet to be fully examined. MOVING FORWARD To understand the effects American missions had on mundane life, a household approach as outlined in this chapter should be applied. Following the provided interpretation of the premissionary household, research ought to focus on the organization of household members through time. In particular, an activity pattern analysis should be conducted that investigates artifact distributions and the distribution of household members with respect to the landscape and to the time of day and year. An appropriate research approach is ethnohistoric in nature, such as described by Oswalt in Bashful No Longer (1990). This approach combines multiple lines of evidence such as archaeological data, oral traditions, and historical documents—missionary diaries, journal articles, other articles, and texts. Incorporating multiple lines of evidence is neither a new approach to historical archaeology in Alaska (for example, see Crowell 1997; Kan 1996; Oswalt 1980; Oswalt and VanStone 1967) nor foreign to studies of the Deg Hit’an (see VanStone 1979a, 1979b). Ethnohistory is more than a “fitting” of different lines of evidence. The different forms of evidence tend to have different biases and agendas. As a result, variabilities and discrepancies in historical portrayal may arise, which in turn can contribute to a complex view of household life. CONCLUSION The examination of premissionary households presented here investigated several aspects of Deg Hit’an life, such as composition of the household groups and the spatial and temporal organization of daily activities. Unlike activity-area studies that have focused on activities directly associated with the house structure and the house locale, this brief examination focused on activities of the members of the social group associated with the house. By doing so, a dynamic view of both the household and the house was created. People moved across the village landscape, and the house’s meaning and functions were transformed throughout the day. Besides providing a dynamic view of the past, this interpretation highlights different variables that may have been influenced by the American missionary period. Because the household and activities were integrated within society, we are made aware that one change or effect can influence many other aspects. Thus, for example, the removal of

RE-PEOPLING THE HOUSE

children from the household unit to attend boarding schools likely affected the household as a production unit. With fewer individuals within the household, fewer people were present to obtain and procure food (yet there were also fewer to feed). Furthermore, young children in school were unlikely to learn traditional practices of subsistence, ceremony, and daily activities. Research examining missionary pressures at the household level would increase our understanding of Deg Hit’an life during this time period. Once data from these multiple lines of evidence are gathered, they should be analyzed within a framework that explores the spatial and temporal organization of the household. This type of approach, one that is multifaceted, will illuminate various aspects of Deg Hit’an society—such as issues of gender relations, family relations, and identity— creating a dynamic view of the missionary village. Further, when this interpretation is compared with one of premissionary life, we can then explore the impacts of the missionaries; and issues such as cultural change, ethnic continuation, and acceptance and resistance can be examined. This type of research will not only increase our understanding of Deg Hit’an past, it will also provide information beneficial to understanding the American missionary period in Alaska. ACKNOWLEDGMENTS First, I would like to thank Lisa Frink, Rita Shepard, and Greg Reinhardt for inviting me to participate in this volume. Their comments have been both inspiring and informative. Donna MacAlpine of the Anvik Historical Society has not only been a good friend but also educated me about the Innoko Lowlands and the people who have lived there. Thank you. Because this chapter was based on my master’s research, I wish to thank the faculty of Binghamton University for their support and guidance. I especially thank Drs. Albert Dekin, Charles Cobb, William Isbell, and Susan Pollock.

7 FISH TALES WOMEN AND DECISION MAKING IN WESTERN ALASKA LISA FRINK

ALTHOUGH ANTHROPOLOGISTS HAVE LONG RECOGNIZED the role of women as subsistence processors, insufficient attention has been paid to the entire spectrum of processing activities, including both the details of the skills operating through the full array of the processing stages and the sophistication of postprocessing management and distribution responsibilities. This chapter seeks to begin to fill the gap by examining more closely the mechanics of the production process at a Native Alaskan fish camp, highlighting the scope of women’s productive activities. Clearly, women’s productive roles have proven more complex and influential in understanding present and past human behavior (e.g., Brumbach and Jarvenpa 1997a, 1997b; Gifford-Gonzalez 1993; Moss 1993). For several years Arctic anthropologists have increasingly paid attention to the significant role of women and subsistence fish processing. Archaeologists, however, have yet to take full advantage of this bank of knowledge and the new questions these data stimulate (for exceptions, see Chang 1988, 1991; Frink, Hoffman, and Shaw n.d.; Knudson et al. n.d.; Romanoff 1992; Schalk 1977; see also Chapter 12, this volume). Food processing and production is an anthropologically valuable research focus for at least three reasons. First, it leaves the “day-to-day” material remains that are relevant to archaeological investigation (Lightfoot, Martinez, and Schiff 1998:199); second, processing is critical

LISA FRINK

to understanding the full spectrum of women’s labor and the social and economic dynamics of prehistoric, historic, and modern people; third, viewing women’s processing as “productive” informs our models of women’s economic, social, ideological, and material contributions and impacts in hunter-fisher-gatherer communities (see also Ellanna and Sherrod 1995:18). The data set used for this chapter is the result of ongoing archaeological, ethnoarchaeological, ethnohistoric, and oral historic research with the Cup’ik-speaking Eskimo (a subdialect of Central Yup’ik; Woodbury 1992:12) community of Chevak in the Yukon-Kuskokwim delta of western Alaska. This chapter explores the role of women in fish production, and the research results demonstrate that processing fish is a highly skilled occupation (a concept not often applied to women’s activities), that women’s roles are highly managerial (e.g., Ellanna and Sherrod 1995; Jolles and Kaningok 1990), and that fish production is complex and warrants continued research. ENVIRONMENT AND CULTURAL BACKGROUND The western Alaskan Yukon-Kuskokwim delta appears to be an endless, windy, wet, and cold space; however, for at least part of the year this swampy flat is brimming with life (Selkregg 1976). The approximately 75,000 km2 triangular-shaped delta is edged on the north by the Yukon River, the fifth largest in the United States and the most extensive in Alaska, and to the south by the Kuskokwim River, Alaska’s second largest watercourse. The outwash from these rivers makes up the delta—a flat, treeless, highly active, wet tundra environment. The most prominent features are the Nulato Hills in the north and the Kilbuck Mountains to the south. The low and poorly drained soils of the lower Yukon region are home to the largest seasonal waterbird breeding grounds in Alaska and contribute to providing residents with an amazing quantity and variety of foods including fish, sea mammals, and floral edibles (Ager 1982; Selkregg 1976). The climate in the delta is variable, with extremes from 44 to 80 degrees F (Selkregg 1976:15–16). The region has 20” of mean annual precipitation, including the water equivalent from snow (Selkregg 1976:12– 13), and discontinuous permafrost remains underneath the low ground vegetation of grasses and sedges rooted in mosses and lichens (Selkregg 1976). Around May, ice on the many rivers, lakes, and sloughs starts its rapid retreat to the Bering Sea, and the slowly melting snow heralds the “season of plenty” for the delta residents, in which highly seasonal resources abound (Testart 1982:524, 1988b, 1998c; Ager 1982; Barker 1993;

FISH TALES

Figure 7.1 Yukon-Kuskokwim delta landscape in July

Fienup-Riordan 1986b; Fitzhugh and Kaplan 1982; Oswalt 1963b, 1990). (Of course, this abundance can vary according to year, season, and individual circ*mstances.) For instance, particularly during the early spring and late fall, men hunt several species of seal including the bearded (Erignathus barbatus), ringed (Phoca hispida), and harbor (Phoca vitulina) seals. The Yukon-Kuskokwim delta is famous for its diversity and quantity of migrating waterfowl. Plentiful streams, small thaw lakes, and marshy sedges create an extremely productive habitat for birds (Selkregg 1976:197). April is Tengmirvik, or “where geese arrive” or “geese come” (Andrews 1989:259). Millions of seasonally migrating fowl are a copious resource1; they include several species of geese like the Canada (Branta canadensis) and black brandt (Branta nigricans). Also available throughout the summer are various greens collected by women, such as sourdock (Rumex articus), and during August several species of berries are eagerly gathered, such as salmonberry (Rubus chamaemorus) and blueberry (Vaccinium uliginosum).

LISA FRINK

The most abundant and paramount delta food, however, is fish. Today men spend many hours using nets to procure the fish, and women work to process this intensively harvested resource (Menager 1962:157– 162; Nelson 1983 [1899]; Pete 1991; Wolfe 1989b). The first fish to offer itself to delta residents is the herring (Culpe harengus), which begins to run in mid- to late June. Not long after that the most important fish, salmon, continues to run through August. Five species of these anadromous fish—the king (Oncorhynchus tshawytscha), the red (O. nerka) the silver (O. kisutch), the pink (O. gorbuscha), and the chum (O. keta)—are available in the surrounding rivers and sloughs. Also included in the catch are flounder (Platchthys stellatus) and halibut (Hippoglossus stenolepis), as well as freshwater fish like the sheefish (Stenoduc levichthys) and blackfish (Dallia pectoralis). The “season of scarcity” begins in the fall (Testart 1982:524, 1988b, 1988c; Barker 1993; Fienup-Riordan 1983, 1986b; Oswalt 1990). Migratory birds begin to head south for the winter, and the rivers begin to freeze in November. Although some hunting, trapping, and fishing occur, most people rely on properly stored and managed provisions to make it through the sometimes long and severe winter months. For these storage-based economies, planning, production, and storage of the seasonally abundant harvests continue to be imperative.2 Procurement of fish can be an enormous investment of time and energy, since winter prosperity depends on this short-term summer activity (Pete 1991; Wolfe 1989a, 1989b). For instance, Victor Shnirelman (1994:175) estimated for the Cherchez le Chien of the Kamchatka Peninsula that eight fisherman using nets can catch thousands of fish per day.3 He also documented that the production process was highly intensive, keeping most women busy for twelve hours of continuous labor (Shnirelman 1994:179). During this heightened labor load, Beth O’Leary (1992:vii) observed that Southern Tutchone women cut, on average, 100 fish per woman per day. CULTURE HISTORY Very few excavations have been conducted in the delta, making it in many ways archaeologically little known (VanStone 1984a). Evidence suggests that people inhabited the western coast of this region 3,000 years ago (Okada et al. 1982) and lived in the inland territory since A.D. 800 (Shaw 1998). The first recognized culture in the delta is late Norton (Giddings 1964; Oswalt 1952; Shaw 1983, 1998), followed by western Thule culture, the ancestors of modern Eskimo groups (Dumond 1977; Shaw 1998).

FISH TALES

As recounted by Linda Womkon Badten (in Chapter 4, this volume), many Native Alaskan people lived in rectangular semisubterranean houses made of wood, hides, and sod into the early/mid–twentieth century (Barker 1979; Shaw 1982; see also Chapter 6, this volume). Typically, particularly in winter villages, the cultural sexual division of space was structurally reinforced by the use of the qaygiq, or men’s house (for a discussion of the use of the men’s house, see also Larson 1991; Lutz 1973; Chapters 3, 5, 6, 11, this volume), and the ena, or woman’s house (Fienup-Riordan 1983, 1986b; Menager 1962; Nelson 1882, 1983 [1899]; Oswalt 1963a, 1963b, 1967, 1979, 1990; Oswalt and VanStone 1967; Ray 1966; VanStone 1967). Residence in the men’s house depended on age (Fienup-Riordan 1983:43; Oswalt 1963a:123); young boys (5 to 10 years of age) and single men most often stayed in the qaygiq, whereas married men spent time in both village spaces. The qaygiq served a multitude of functions including the men’s fire bath, workshop, community ceremonial space, and a place for male visitors to stay (for more on gendered use of space and what it means, see Chapters 3, 5, 6, 9, 11, 12, this volume). The ena was considered a female space, where women and their children resided. Grandmothers, mothers, and daughters worked, cooked, ate, and slept in these smaller “complex households” (Blanton 1993:5; Chapters 5, 6, this volume) that could house three families and twenty individuals (Menager 1962; Nelson 1983 [1899]:288; Ziff, Pratt, and Drozda 1982). EUROPEAN CONTACT Attesting to the ecological productivity of western Alaska, populations were among the highest in Alaska at contact (VanStone 1984b). European influence in the lower Yukon region was kept to a minimum for several reasons (Fienup-Riordan 1994; Oswalt 1990); because of the low Bering Sea shelf this “damn’d unhappy part of the world” was difficult to travel (Oswalt 1990:11). Moreover, unlike other regions in Alaska that produced timber, furs, marine fish, and the like, the Russians found no compelling economic draw in this isolated region and hence had relatively little influence on the Native groups in this part of western Alaska (Fienup-Riordan 1996; VanStone 1984a; cf. Black 1984). Native communities in the north-central Yukon delta had less contact with Russians or even early Americans than did those in the Kuskokwim area or the far northern Yukon delta (Ray 1966). In fact, although Andrey Glazunov conceivably explored further down the Yukon River in 1832 (VanStone 1984a; Ziff, Pratt, and Drozda 1982), Edward Nelson, a U.S. naturalist, traveled through the northern extent of the delta for the first time during

LISA FRINK

the winter of 1878 (Nelson 1882, 1983 [1899]), collecting ethnographic data and Native material items for the Smithsonian Institution (Collins 1982; Fitzhugh and Kaplan 1982; Nelson 1882, 1983 [1899]; see also Chapter 5, this volume). MISSIONIZATION AND DISEASE Most major social and demographic changes in the delta region resulted from missionization and disease (Oswalt 1990). Russian Orthodox priests were present in the delta in the mid-1800s, and Catholic and Moravian missionaries began seeking converts in the late nineteenth and early twentieth centuries (Fienup-Riordan 1991; Flanders 1984; Llorente 1988; Menager 1962; O’Connor 1947; Ray 1966; Schwalbe 1951). Epidemics of influenza and smallpox continued into the early 1900s, causing tragic numbers of deaths and substantial movements of people (Fienup-Riordan 1991, 1994; Chapter 4, this volume; for an overview of early historical disease in Alaska, see Fortuine 1992). Death rates at times were staggering; for instance, in 1942 in the village of Nunaraluq (an ancestral village of Chevak), according to Father Jules Convert, seventeen children died of whooping cough in a nineteen-day period (Barker 1979:29). The disruption of disease caused major demographic shifts, especially the aggregation of village members and the incorporation of numbers of orphaned Native children into the newly formed mission boarding schools (FienupRiordan 1991, 1994; Llorente 1988; Oswalt 1990; Ray 1975; Woodbury 1992; see also Chapter 6, this volume). RESEARCH PROJECT Since 1996 I have been conducting research in the village of Chevak and its attached archaeological village sites.4 Chevak, located 27 km east of the Bering Sea, rests on a bluff above the Ninglikfak River and is home to over 800 Cupiit Eskimos5 (Morrow and Schneider 1995:9; Woodbury 1992:12). This region is suitable for ethnoarchaeological research, since many villagers lead a very modern lifestyle, yet traditional ways continue to be essential. For instance, although Western goods are available in this mixed-subsistence economy, they are viewed as secondary to subsistence foods; people continue to hunt, gather, and—most significantly—fish (see also Barker 1993; Fienup-Riordan 1983, 1986b; Wolfe 1984, 1989a, 1989b). Moreover, the people of Chevak have an archaeologically unique connection to their past; Chevak is the youngest in a series of four sequentially inhabited village sites. The oldest prehistoric site, Qavinaq, is connected to the people of Chevak through oral history; however, many

FISH TALES

Figure 7.2 Chance Hill setting whitefish net at the mouth of a slough

village elders remember living in the subsequently occupied site, Nunaraluq, a pre-proto- and historic mound site. In the 1940s people moved from Nunaraluq to Old Chevak and eventually in the 1950s to present-day New Chevak. This community is also appropriate for gender research, since women and men practice a sexual division of labor—particularly in the collection and processing of subsistence foods. Even though this system has complexities, the general custom among this modern generation of Eskimos is that the subsistence division of labor runs along sexual lines with little task “overlap”6 (Lepowsky 1993:113–116; Ackerman 1990a; Giffen 1930; Ray 1966). Men most often hunt birds, as well as sea and terrestrial mammals; fish with nets; and gather some foods such as berries and clams. Women gather grass, collect vegetation, dig clams, fish using lines, and—most salient to this research—are the primary processors of all food and are essential as decision makers and managers of the critical subsistence harvests.

100

LISA FRINK

METHODOLOGY Data for this chapter were generated from taped interviews with ten elder women conducted during the 1998 field season in the village of Chevak. These women ranged from sixty-one to eighty-three years of age, and most lived at the previously occupied villages of Nunaraluq and Old Chevak. My interviews focused primarily on women’s roles in subsistence fish processing and management and how practices may or may not have changed in these women’s lifetimes. With assistance from a Native interpreter, the format was one of discussion with open-ended questions. Additional data were collected at a herring fish camp (located on the Ninglikfak River within walking distance of the village) where I visited an elder woman (in her early eighties) while she managed her camp and cut her fish. At the early summer seasonal camp I observed, audio recorded, photographed, and participated in the processing, storage, and disposal of herring (Culpe harengus) and whitefish (Coregonus spp.). The results of this ongoing seasonal camp research suggest several critical components to understanding the role of women in Eskimo society; two are pertinent to this research.7 First, fish processing is a learned complex skill requiring accumulated knowledge for proper processing and handling of the harvest; and second, fish handling requires advanced management skills, such as short- and long-range decision making. DISCUSSION Fortunately for archaeologists, a fair amount of data have been collected concerning fish processing.8 Cutting fish, unlike what Paul Drucker and Robert Heizer (1967) recognized as fairly simple, is a learned talent requiring technical prowess and skillful decision making. Or as Robert Schalk said, processing is “not as simple as cleaning and hanging fish to dry” (Schalk 1977:232). Given the climate of western Alaska, the temporal window for successful harvesting of fish is quite small. Many fish are processed within a short period of time; for instance, the main run of herring during the early summer lasts several weeks. The entire intensive season of fishing begins (approximately) in mid/late June and continues into August. Women must cut, dry, smoke, and store the catch in a very demanding and intensive manner. Therefore they must differentiate (according to which species of fish they are handling) and prioritize their processing activities to maximize the yield of fish harvested so that fish are not lost through spoilage. This requires knowledge about the properties of different species of fish, the condition of the fish at the time of capture, the

FISH TALES

101

Figure 7.3 Ulrich Ulroan brings whitefish to his grandmother, Mrs. Angelina Ulroan.

effect of weather (humidity, temperature, rainfall) and insects on processing, and the productive capacity of the laborers available (see also O’Leary 1992; Pete 1991; Romanoff 1992; Schalk 1977)—all crucial management factors and decisions negotiated and made by women. Processing is an intergenerational activity in which this knowledge is taught and learned through years of experience. As occurs in other Eskimo communities, grandmothers, mothers, and daughters usually cut fish together (Ellanna and Sherrod 1984; Wolfe 1989a). Similar to what Linda Ellanna and George Sherrod (1995) found with a Native Inupiaq group, elder women are active managers of the fish camp operation; for example, many elder women from Chevak establish their own camp each season, often with or near extended family. The fish camp I visited was repeatedly referred to as the elder woman’s, and all the fish her grandson caught were brought specifically for his grandmother.9 The equipment at the fish camp—including the tent, drying rack, pits, and smokehouse—is considered the elder woman’s property. The ownership of fish is an interesting issue and one I will

102

LISA FRINK

Figure 7.4 Mrs. Angelina Ulroan processing herring at her camp (note salmon-skin boots).

return to. During my visits to this camp I learned about several examples of decision making in the processing of fish, similar to the types of decisions James Barker (1993) observed on Nelson Island. Not all fish are processed the same way; each species is treated differently, and within individual species other factors differentiate processing techniques. For instance, herring are cut and processed differently based on size; the “larger” fish lose their heads, and the “small” fish keep theirs. Large fish are then cut on one side for gutting, whereas the small herring are completely opened up. After drying for approximately four days (weather dependent), the herring are hung using a herring braid (made of rye grass and cotton); heads are threaded on the larger fish, and tails are used for the smaller herring (Barker 1993; Pete 1991). The skill of braiding herring is deceptive; several middle-aged Chevak women were still learning to master this processing activity. Deciding which fish should be first in the processing queue is also a complex art. Often the run of fish is so intense that decisions must be made about which species to prioritize; some demand immediate atten-

FISH TALES

103

tion, whereas others can wait. For example, herring, according to my elder host, can last in a pit for up to two weeks, whereas whitefish are known to deteriorate and rot much faster and need to be cut and hung to dry in a more timely manner. The whitefish could also be eaten at camp, since they require only limited drying to be edible. Several elders also discussed the necessity, during an intensive salmon run, of deciding which individual salmon to cut first. Some of the salmon will be set aside and stored in large pits lined with grass and covered with sod blocks. The women will then decide when to return and cut these fish; in addition, further processing decisions continue after the fish have been initially processed. Each species requires specific methods and timing of processing. For example, herring are dried for several weeks and must be kept dry during the consistent rains of the delta, whereas whitefish require much less drying and can also be smoked. Storage for hunter-fisher-gatherers in the Arctic is critical to survival. It was and is women’s responsibility to store the foods properly for use by the family throughout the entire year. Traditional methods included the use of handwoven straw baskets and underground pits in which to store fish. (Today the luxury of freezers makes the job of food storage less difficult and risky.) Clearly, too little has been discussed in the literature exploring the processes by which perishable resources are stored (Schalk 1977). One notable example suggests that storage was a learned art, requiring the right positioning of items in pits, the appropriate layering of grass for circulation, and a watchful eye for the remaining stores (Lantis 1946; see also Fienup-Riordan 1983). My preliminary data suggest that women store different species in different ways at different times of the year. For instance, herring is most often stored aboveground in baskets, but needlefish—an important winter food— likely was stored in small circular floor pits within the semisubterranean houses. Again, more data need to be generated to understand the full spectrum of choices women had and have regarding proper storage. ISSUES OF OWNERSHIP Ernestine Friedl (1975) described Arctic women’s labor as directly related to men and their economic and social success, intimating that women merely process and may not own the fruits of their labor. More recent research has demonstrated, however, that this is a simplistic understanding of women’s labor and their control over the product of their work (Ellanna and Sherrod 1995; Jolles and Kaningok 1990). These data suggest the women of Chevak, like other Native northern North American women (Ellanna and Sherrod 1995; Chapter 3, this volume),

104

LISA FRINK

not only process the bulk of subsistence products but also manage them, and this oversight extends to control over distribution and consumption as well (the full economic spectrum). Chevak women control the daily allocation of foodstuffs. As Ann Fienup-Riordan (1986b:182, 184) found, the food caches are “owned” and managed by an extended family’s elder women, who “decide what is to be eaten and when” (for more on food and ownership, see Chapter 3, this volume). This is less true today in the village; several elder women suggested that unlike today (when kids can go to the store and get snacks), when they were young their mothers had jurisdiction over all the food. They talked of their mothers’ portioning at mealtimes and recalled that they would get only a certain amount of food. (One elder remembered that her brothers always got more than she.) In making decisions concerning daily allocation, women must project long-range usage plans. A woman must decide what resources will be used for her nuclear and her extended family over the coming year (O’Leary 1985; Romanoff 1992). In addition, her planning must include nonhuman and social needs such as feasting (Fienup-Riordan 1983, 1986b; Koranda 1968; Nelson 1882, 1983 [1889]; Oswalt 1963a, 1963b, 1967, 1979, 1990; Ray 1966; Michael 1967; for the importance of feasting, see Dietler and Hayden 2001). Feasting still occurs in the winter months, especially December, and requires that large quantities of food be shared. One elder woman spoke longingly of her youth and the foods that were so available during feasting periods, unlike the controls placed on her consumption during the rest of the year. Moreover, today most families have only one dog to feed, but prior to available mechanical transportation (mid-twentieth century) women had to plan to feed the family dog team, which could range from five to eleven dogs—each requiring at least half a salmon ration per day while traveling10 (Ray 1966:82). While at fish camp I observed that most of the heads and backbones of fish were tossed in the river; however, the elder women I spoke with remarked that this was never done in their day, when everything was used (see also Wolfe 1989a). Women’s authority over fish production is also demonstrated by the use of ownership marks. Traditionally, when a man is catching fish, they are his; but when he brings them to his sister, wife, or mother, they become hers (see also Fienup-Riordan 1986b:184). As Fienup-Riordan (1986b:184) has found for the Native Nelson Islanders, the stores “become the property of the older women of the extended-family household for processing and for distribution” within and beyond the household. As the wife does not often hunt seal, “her husband rarely goes into the

FISH TALES

105

Figure 7.5 Filleted salmon (note fish mark at top); courtesy, Brian Frink, University of Minnesota, Mankato.

106

LISA FRINK

Figure 7.6 Dried and smoked salmon at the foot of the smokehouse and fish rack in the background

FISH TALES

107

cache to bring in fish for the evening meal” (Fienup-Riordan 1986b:184). As previously discussed, however, this all-encompassing authority over food distribution is likely diminishing over time. A physical declaration of ownership is the Chevak women’s ownership mark, which several elder women described during the village interviews. These marks are simple cuts placed on both sides of the fish flesh near the tail (see Figure 7.5). I was told most women do not do this anymore (however, one elder continues to make these marks when she remembers) and do not know when the practice stopped, but one elder woman suggested it may have ceased in the 1950s when the village moved to its current location. The fish marks are suggestive concerning ownership. According to my elder consultants, the marks are considered feminine designations of ownership (see also Romanoff 1992) and are unlike the impressions men place on their catch or material items (see Nelson 1983 [1899]); instead, these marks were passed on in a woman’s line from mother to daughter. It is interesting that these female designations of ownership were a complete surprise to several middle-aged Chevak women, making it likely that the transmission of this practice has ceased. More data need to be collected concerning the use, complexities, and relevance (especially for the archaeological record) of this provocative practice of marking fish. ARCHAEOLOGICAL IMPLICATIONS Food production is a critical research focus for archaeologists. Fish processing in present-day Native western Alaska is no exception, and new questions born from these data abound for archaeologists. For instance: How has technological change affected efficiency and so harvest capabilities? Can archaeologists detect past fish camp activities using soil chemistry analysis? How will envisioning women as resource managers and active producers inform our theoretical models of hunter-fishergatherer communities and social change? With more complete documentation and appreciation of the complexity of women’s productive activities, we can continue to more completely understand modern, early historic, and prehistoric production capabilities; more accurately represent women’s roles; and increasingly apply these data to the interpretation of archaeological assemblages. CONCLUSION This research adds new and vital observations concerning Native western Alaskan women and their processing and management of fish and

108

LISA FRINK

has critical implications for archaeological investigations, especially concerning women and their roles in a hunter-fisher-gatherer society. These data illuminate the fact that women’s direction of production activities is a highly skilled craft; it requires a knowledge of processing, storage techniques, and management skills that have been critical to survival in the harsh Arctic environment and must be learned over an extensive period of time. New insights and documented data concerning the complexities of economic roles, including women as not only processors in a subsistence economy but as resource managers and producers, will increasingly broaden our recognition of the economic and social contributions of women in the present and past. ACKNOWLEDGMENTS I wish to thank the Elders of Chevak and the Elder Advisory Council for sharing their knowledge with me. I also am grateful to the Chevak Traditional Council and the community of Chevak for their interest in and support for this work. This research could not have been accomplished without the financial and logistical support of the United States Fish and Wildlife Service and Debbie Corbett (Anchorage Office), Mike Rearden, and Paul Liedberg (Bethel Office). Also, many thanks to Mary Berthold, Hetty Jo Brumbach, Brian Hoffman, Bob Jarvenpa, Phyllis Morrow, Mary C. Nanuwak, Greg Reinhardt, and Rita Shepard for intellectual engagement and remarks on the manuscript and to three anonymous reviewers (of course, this chapter is singly my responsibility). My heartfelt thanks also go to Brian Frink for his artwork. Finally, much gratitude goes to Kate for her continuing observations, unwavering support, and enduring partnership.

III MATERIAL AND SPATIAL ANALYSIS

8 CHILD AND INFANT BURIALS IN THE ARCTIC BARBARA A. CRASS

THE ARCHAEOLOGY OF GENDER INCLUDES ALL “INVISIBLE” PEOPLE (e.g., Moore and Scott 1997), not just women or gender transformers. Gender is not based solely on cultural definitions of male and female but is age associated as well. Children are currently one of the foci of gender research, especially in Europe (e.g., Baker 1997; Meskell 1994, 1996; Siemoneit 1997; Sofaer Derevenski 1994, 1997). The Arctic also provides a rich basis for looking at children and infants in the past. Children and infants have been described as one group of invisible people in the archaeological record (e.g., Moore and Scott 1997). Although children undoubtedly left their mark in the past, as they certainly do in the present, these marks are rarely obvious archaeologically. One area where these marks may be more apparent is in mortuary contexts. In an extensive survey of burials in the Arctic (Crass 1998), 305 infants and children were identified in pre-Christian Inuit burials. These infants and children will be compared with 1,459 adults. Since the reports were written and the archived material collected over the past century by various individuals, the data are not always complete. The numbers are large enough, however, that general trends should be apparent. Data were gathered from fifty sites located across the Inuit range from northeastern Siberia to Greenland.1 The sites vary from large, permanent settlements with associated cemeteries providing hundreds of burials,

112

BARBARA A. CRASS

Table 8.1—Sites With Infants, Children, or Both Site

References

SIBERIA Ekwen Uelen

Arutiunov & Sergeyev 1975 Arutiunov & Sergeyev 1969

ALASKA Cape Krusenstern Cook Inlet Dovelavik Bay, St. Lawrence Island Gambell, St. Lawrence Island Kitnepaluk, St. Lawrence Island Naknek Point Hope Prince William Sound Tulaagiaq Uyak, Kodiak Island

Giddings & Anderson 1986 DeLaguna 1934 Hofman-Wyss 1987 Bandi 1984 Hofman-Wyss 1987 Dumond 1981 Larsen & Rainey 1948; Rainey n.d. DeLaguna 1956 Anderson 1978a, b Heizer 1956

CANADA Kamarvik, Chesterfield Inlet Kulaituijavik, Chesterfield Inlet Naujan Niutang, Baffin Island Saglek Bay, Labrador Silumiut, Chesterfield Inlet Southampton and Walrus Islands Tasioya, Baffin Island GREENLAND EAST GREENLAND Dødemandsbugten Imertiit, Illuluarsuk Region Kangerlussuaq Qoornoq, S. Skjoldungen Sound Skærgårdshalvø, Knud Rasmussen Land Suess Island, Clavering Island Suukerti Timmiarmiut, Frederick VI’s Coast WEST GREENLAND Ammaasaq, Upernavik Illorsiut, Disko Bay Illutalik, Disko Bay Illutalik, Julianehåb District Kangaamiut Area Narsarsuaq, Julianehåb District Saattoq Island, Upernavik Tuttutuup Isua, Julianehåb District Upernaviarsuk, Upernavik District

McCartney 1971; Merbs 1967, 1968a, 1968b McCartney 1971; Merbs 1967, 1968a, 1968b Mathiassen 1927; Fischer-Møller 1937 Salter 1984 Way 1978 McCartney 1971; Merbs 1967, 1968a, 1968b Collins 1955, n.d.; Collins & Emerson 1954; Emerson 1954 Salter 1984

Larsen 1934 Felbo et al. 1993 Mathiassen 1936b; Degerbol 1936 Felbo et al. 1992 Larsen 1938 Glob 1935 Mathiassen 1933 Mathiassen 1936c Hjarnø, Jorgensen, & Vesely 1974 Mathiassen 1934 Mathiassen 1934 Mathiassen 1936a Mathiassen 1931 Mathiassen 1936a Hjarnø, Jorgensen, & Vasely 1974 Mathiassen 1936a Hjarnø, Jorgensen, & Vasely 1974

CHILD AND INFANT BURIALS IN THE ARCTIC Table 8.2—Individuals Infant

113 Table 8.3—“Other” Burial Types

Child

Adult

Graves Cairns Other

16 25 7

170 74 13

759 629 71

Total

257

1,459

Infant

Child

House Fissure Midden Surface Cave Pit

1 2 4 0 0 0

5 3 2 1 1 1

Adult 10 20 9 10 14 8

Total

such as Ekwen in Siberia and Point Hope in Alaska, to a handful of houses and a few associated cairns providing less than ten individuals, such as Narssarssuaq or Sûkersit in Greenland. Infants and/or children were found in thirty-five of these sites, as shown in Table 8.1. For this chapter, infants are defined as 0–3 years of age and children as 3–12 years. The burials are divided into three types: graves, cairns, and other. Graves are in-ground burials found primarily in Siberia and Alaska. Cairns are aboveground stone chambers and are common in the Canadian Arctic and Greenland. The majority of the burials are either graves or cairns. The “other” category consists of less common burial methods including burials in inhabited houses or pits; placement in rock fissures, middens, or caves; and surface burials. The distribution of the 48 infants, 257 children, and 1,459 adults in these three burial types is provided in Table 8.2. A further breakdown of the distribution of infants, children, and adults in the “other” category is given in Table 8.3. From these two tables, it is clear that child burials are infrequent compared with adult, and infant burials are rarely found. Of the 1,764 individuals, 82.7% are adults, 14.6% are children, and only 2.7% are infants. Many of the these individuals are interred together in multiple burials. Individuals are discussed instead of burials because about a third of the individuals are found in multiple burials: 30.7% (290 of 945) of the individuals in graves, 36.4% (265 of 728) of the individuals in cairns, and 28.6% (26 of 91) of the individuals in other burial types (Table 8.4). Whereas the adults are often buried with other adults, infants and children are usually buried alone or with one or more adults (Table 8.5). Although the sex of all the adults is not known, a trend is apparent. In graves, children are buried with adult males more often than with adult females (13 versus 10). The reverse occurs with children in cairns, with 11

114

BARBARA A. CRASS

Table 8.4—Individuals in Single and Multiple Burials

Table 8.5—Multiple Burial Associations

Infant

Child

Adult

GRAVES Single Multiple

14 12

113 57

538 221

CAIRNS Single Multiple

20 5

53 21

390 239

OTHER Single Multiple

6 1

10 3

49 22

M/F

GRAVES Infant Child

0 10

0 13

1 3

5 24

6 7

CAIRNS Infant Child

3 8

1 3

0 2

0 7

1 5

Burials contain either both an infant and child or more than one infant or child.

being buried with females and only 4 with males. Some of the multiple burials, as seen in the last column, contain either both infants and children or more than one infant or child, with or without adults. The ethnographic record offers some explanation for the number of burials of infants and children with adults of both sexes. Most early ethnographers agree that when the mother died, infants and any children who were not weaned were killed—usually by strangling—and buried with their mother (cf. Boas 1907:117; Bogoras n.d.:26; Crantz 1767:238; Gosling 1910:213; Hawkes 1916:139; Holm 1914:62; Kroeber 1900:31; Lyon 1825:276, 380; Oswalt 1963a:144, 1979:133; Peary 1898:506; Rasmussen 1929:159). Older children not yet able to fend for themselves were also sometimes killed (Cook 1894). Exceptions were made if a woman could be found who could nurse the child or if the child was particularly strong and struggled against death (Day 1973; Steensby 1910:377). A child who fought was seen as an individual who would likely survive. Strangling small children was seen as kinder than letting them slowly starve to death because of lack of proper food. Similarly, sometimes small children were strangled and buried with their father, since there would be no one to provide for the family (Cook 1894; Day 1973; Oswalt 1979:133). The mother’s chances of remarrying would be better if she was not accompanied by small children. Again, the logic was that a quick death was preferable to starvation. Another possibility may be that small children were buried with adults or older children so they would have someone to help them get to the afterlife realm. In Greenland, small children are said to have been buried with the head of a dog so the dog could lead them to the afterlife,

CHILD AND INFANT BURIALS IN THE ARCTIC

115

since small souls could not find their way alone (cf. Birket-Smith 1924:66; Crantz 1767:237; Egede 1818:153: Schultz-Lorentzen 1928:245). Unfortunately, most child and infant burials from Greenland in the study were excavated in the first half of the twentieth century when animal bones were ignored. Six single infant or child burials were excavated in the 1990s, however (Felbo et al. 1992, 1993), and one—a child—does contain the mandible of a dog. Although graves and cairns are the most common burial type, infants and children are frequently found in the other burial methods. Some of these types are difficult to find archaeologically, such as surface burials. Others, such as burials in occupied houses, are not usually found when excavating burial sites. The majority of the house burials are from sites in Greenland, where random houses were excavated and the associated visible cairns were investigated (cf. Degerbol 1936; Larsen 1934, 1938; Mathiassen 1936a, 1936b). If we compare the number of infants with the number of adults found in burials other than graves and cairns, we find significantly more infants (chi2 = 8.83, p>0.005). We even find that infants are found more often than children in these alternative burials (chi2 = 6.2, p>0.025). Since the middens and fissures are the most common types for infants, this may indicate a preference for a quick or easy disposal. The midden burials had no indication of interment in the area surrounding the infants, whereas the infants found in fissure burials were accompanied by a few burial goods. Curiously, no significant difference exists between children and adults in these other burial types (chi2 = 0,005). Fewer children than expected had tools in cairns as well (chi2 = 5.8, p>0.025), although to a less significant degree. A trend for children to have more miniature items than adults was also found (chi2 = 5.05, p>0.025), although the numbers are small (2 of 18 for children and 4 of 206 for adults). The numbers are also small for pendants (3 of 18 for children and 6 of 206 for adults), but a much more significant number of children had pendants than adults (chi2 = 8.35, p>0.005). The amount of goods can also indicate variation in burial treatment. Approximately half of the child and infant single cairn burials did not have goods. When compared with adult single cairn burials, a signifi-

CHILD AND INFANT BURIALS IN THE ARCTIC

117

Table 8.6—Significant Goods

Grave Cairn

Tool category Tool category Miniature/toy Pendant

Chi2

9.57 5.80 5.05 8.35

>0.005 >0.025 >0.025 >0.005

cant difference was found (chi2 = 8.46, p>0.005 ). Graves, on the other hand, did not have a significant difference between burials with goods and those without (chi2 = 2.24). A particular problem exists when looking at goods in cairn burials, as cairns are often conspicuously visible and can be opened and closed. This made them an easy target for collectors, and many cairns were noted by the investigators as obviously disturbed (Crass 2001). The problem is compounded when one takes into consideration the Inuits’ practice of borrowing or exchanging items from cairns as needed (cf. Hawkes 1916:136; Jenness 1922:176; Kumlien 1879:28; Rasmussen 1908:114). For example, it would be perfectly acceptable to remove a knife from a cairn and replace it with some food or a clothing item that would likely not be discernable archaeologically. Graves, however, are not so easily accessible, so items found in graves were likely placed there at interment. Variation in the number of goods associated with individuals was tested by the Wilcox two-sample test with two-tailed probability. For cairns, p=0.17384, indicating little probability of a difference, whereas graves, with a p value of 0.01778, indicate a trend for adults to have a greater number of goods than children and infants. One interesting pattern found in graves was the distribution of skeletal elements. Infant and child remains appear to include skulls more often than adult remains (chi2 = 6.5, p>0.025). If remains consisting only of skulls are compared, the difference is highly significant (chi2 = 14.24, p>0.001). This could be the result of taphonomic processes, but it is doubtful that is the sole explanation. In some areas the preservation is so good that several close-to-term fetuses have been recovered, fully articulated and in excellent condition! Again, the ethnographic record may provide some clues. Several ethnographers claimed a child’s or infant’s corpse was handled more freely and kept around longer than an adult’s (Lantis 1946:229; Lyon 1825:369). If the child died when the ground was too frozen to dig a grave, perhaps it was stored somewhere and later buried in the grave. If the graves are in fact a secondary burial,

118

BARBARA A. CRASS

this could explain the higher incidence of skulls. Although uncommon, a few ethnographic accounts exist describing parents whose care and love for their deceased children went beyond what we would expect. For example, Lucien Turner (2001 [1894]:29), describing the Inuit around Hudson Bay, claimed parents, when moving to another site, would sometimes take their child’s corpse with them if it was not too badly decomposed. Any tendency to curate the remains of children, no matter how casual, could help explain the presence of child and infant skulls in burials. Finally, a pattern can be seen in the skeletal orientation in cairn burials. Previously I have shown that a distinct north to east and south to west distribution is associated with beliefs in cosmology and the afterlife (Crass 1998, 2000). The Inuit traditionally have two afterlife realms, one in the sky and one under the sea or land. Unlike our heaven and hell, both of these realms are viewed in a positive manner. The destination of a soul to any one sphere is associated with how one died. People who die a violent death, such as victims of murder, suicide, and accidents, go to the Land of the Day in the sky, or east. In contrast, those who die a nonviolent death go to the Land of the Sea, or west. Of the infants and children with a known orientation, 20 of 38, or 52.6 percent, were placed with their heads oriented from north to east. Although 44.7 percent of the adults were also oriented in this direction, 58.2 percent of the identified male adults were aligned with this orientation. The percentage of infants and children is very similar to that of adult males, the group most likely to die a violent death from causes such as accident, murder, or warfare. Interestingly, death in childbirth is also seen as a violent death, and 37.3 percent of the known females were found oriented from north to east. Conversely, 10 of 38 infants and children (26.3%) were placed with their heads oriented from south to west, supposedly indicating a nonviolent death. Although 28.6% of the adults were oriented in this direction, for the adults whose sex is known, 36.4% were female and only 16.5% were male. CONCLUSION I have shown that, traditionally, child and infant burials vary in several aspects from adult burials among the Inuit. Infants and children were not always buried in the same manner or in the same places as adults. Infants and children were usually buried either alone or with adults of either sex, rather than with other infants and children. This supports the claim of ethnographers that when one parent died, small children and infants were often killed and buried with the parent rather than slowly starving to death.

CHILD AND INFANT BURIALS IN THE ARCTIC

119

The souls of these strangled individuals, who definitely died a violent death, would then go to the Land of the Day in the east. The analysis of head orientation in cairns supports a fairly high incidence of violent death, with 52.6 percent of infants and children oriented between north and east. A significant number of infants, but not children, were found in burial types other than graves and cairns. The most common types were fissures and middens, which may represent quick and easy disposal methods. When burial goods were examined, little significant difference was found between adult and infant and child burials. The main exception was the tool category, with adults having tools more often than infants and children in either graves or cairns, although the significance is much greater in graves. Although the numbers were small, pendants and miniature items were found more often with infants and children than with adults. Although no difference was found among individuals in graves with goods and those without, adults in cairns had a much higher probability of having goods than children or infants. From the ethnographic record, however, we know items in cairns were often removed, exchanged, and replaced, which could easily mask what was originally deposited in the cairns. Still, one would not expect infants’ and children’s cairns to be preyed upon more often than those of adults. The overall number of goods found with adults and infants and children appears fairly evenly distributed, with cairn burials having individuals with many goods. A trend for proportionally more adults to have a greater number of goods than infants and children was indicated in graves. One of the oddest finds is that infant and child remains consisting only of their skulls were found significantly more often than adult remains of only skulls in graves. Although taphonomy may be partly responsible, some form of casual curation may also be the cause. More research needs to be done in this area. I hope I have shown that Inuit infants and children are not invisible people. Although they resemble adults in many aspects of burial, enough differences exist that their presence should be noted. Our next problem is understanding what these difference may tell us. ACKNOWLEDGMENTS The research on which this chapter is based was made possible through a Dissertation Improvement Grant from the National Science Foundation,

120

BARBARA A. CRASS

Department of Polar Programs; a Full Fulbright Grant to Denmark; and the generous financial and editorial support of my husband, Paul R. Holzman. An earlier version of this chapter was presented at the 6th Gender and Archaeology Conference, Northern Arizona University, October 6–8, 2000.

9 PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA GREGORY A. REINHARDT

MOUND 44’S DISCOVERY AND SIGNIFICANCE IN 1982 A FEW BARROW TOWNSPEOPLE inadvertently dug into a set of frozen bodies, which had been crushed to death 400–500 years earlier when their semisubterranean house caved in on them. That distinctive discovery caused a brief stir in local and national news, led to a symposium at the 1983 Alaska Anthropological Association annual meeting, and resulted in an entire issue of Arctic Anthropology (1984) dedicated to scientific studies of the corpses and their archaeological setting. The next notable publication about Mound 44 was an article in National Geographic (Dekin 1987). Three years later, having been the 1982 archaeological crew chief at Mound 44, I described its architectural elements and detailed all artifactual contents unearthed there that season (Reinhardt and Dekin 1990). Other authors presented data from Mound 44’s 1983 excavation season in the same set of reports (Hall 1990). What made Mound 44 such an appealing find was its unusual preservation history. Normally, in ethnographic times (and presumably prehistoric as well) the Inupiat people of Utqiagvik abandoned their houses in summer, in large part because the subterranean tunnel connecting the house to the outside world would fill with water thawed from winter’s snow and frozen soil. After the tunnel water froze, by late September, the tunnel ice could be dug out with picks and mattocks (Murdoch 1988

122

GREGORY A. REINHARDT

[1892]:76). The occupants took most things with them when they left for the summer, especially if they chose not to return in the fall. Still, most prehistoric and contact-era Inupiat homes that archaeologists excavate contain artifacts, but not many. Mound 44 differs, however, because this is not a case of “normal” abandonment (Cameron and Tomka 1993). Rather, the residents died inside their home, leaving scattered about them virtually all the artifacts they kept in the house plus the occupational debris they generated. Exemplifying archaeology in its most popular state, journalistic allusions to a “snapshot in time” larded the first tantalizing news reports about Mound 44, and most archaeologists would agree this was a veritable freeze-frame of daily life in a prehistoric Eskimo house. Besides illustrating the architectural remains, we mapped and point provenienced over 300 artifacts and dozens of discarded bones directly associated with the house interior in 1982. We also charted hundreds of other artifacts and waste items found buried in soil matrix beneath the house floor (indicating earlier occupation of the same house pit). In 1983 the Utqiagvik Archaeology Project extended its work to areas outside the house proper: the tunnel, the kitchen, a storage pit, and soils surrounding these house locations. This study considers only the 1982 data from the house interior, focus of the gender-space premise. RELATING MOUND 44 TO OTHER STUDIES OF HOUSEHOLD AND GENDER The importance of considering other household and gender studies, as applied to this one, should not be minimized. At the same time, because this work is principally descriptive and a reassessment of an earlier behavioral assertion, the thrust of discussion here is to examine select cases that have some bearing on Mound 44. I divide them into two kinds of relevance. The first kind of study relates theoretically to processes affecting the archeological record of households. In their discourse on the formation of house-floor assemblages, Vincent LaMotta and Michael Schiffer (1999) posited three processual stages (Habitation, Abandonment, and Postabandonment) and divided each stage into two processes: accretion or depletion. Within the habitation stage, primary deposition (discarded or lost in situ) and provisional refuse deposition (removing refuse from one spot to a separate location) are accretion processes, whereas secondary refuse deposition (storage or caching instead of discarding) is a depletion process (LaMotta and Schiffer 1999:20–22). Habitation pertains the most to archaeological circ*mstances surrounding the Mound 44 house. Raymond Newell (1984) dealt with both

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

123

primary and secondary deposition. Abandonment does not really apply to Mound 44, but the postabandonment depletion process of scavenging probably occurred. Gregory Reinhardt and Albert Dekin (1990) described the secondary deposition and scavenging processes at Mound 44 in some detail, whereas Georgeanne Reynolds (n.d.) looked beyond toolkits at the spatial distribution of artifacts from Mound 44’s house floor. The second kind of study is specific to recent examples attempting to “discover” gender in archeological settings. In her excavations at a mid-nineteenth-century gold-mining town in Australia, Susan Lawrence (1999) sought to identify women through the various structures and artifacts found there. She was able to discriminate households exclusive to men from households occupied by both sexes (Lawrence 1999:138). In the same book Suzanne Spencer-Wood (1999:185–186) took to task longstanding “androcentric constructions of the unitary gender ideology and norms of a culture.” Her wide-ranging documentary research reveals how changes in architecture and other reflections of material culture reflect communal shifts in the social structure of, and gender interaction within, mid-Victorian domestic life (Spencer-Wood 1999:185–186). More challenged by a lack of historical documentation, but simultaneously influenced by discussions with local elders, Annette McFadyen Clark (1996) struggled to disclose female-male differences in behavior based on the household distributions of artifacts and ecofacts. She creatively analyzed a “composite plot” made up of material remains from three houses, projected onto one floor space, in hopes of determining whether the occupants were Koyukuk Athapaskan or Inupiat Eskimo. In the process, she concluded that the cultural remains found inside houses fit well with her interview results: gender-based activity areas within the idealized two-monogamous-family house conform to her composite plot of material remains (McFadyen Clark 1996:159, 199, 203, figs. 5-4, 5-11, 5-12). RECONSIDERING THE GENDER-BASED “SIDES” PREMISE THE PREMISE AS A “PROBLEM” The longest article in Arctic Anthropology’s Mound 44 volume focused on the house’s archaeological, human physical, and relative settings (Newell 1984). Unfortunately, the representation of artifact contents in the house (Newell 1984:fig. 14) and the corresponding breakdown as to “functional partitions” of those artifacts (Newell 1984:table 1) are not complete, internally consistent, or entirely correct. The distribution map (Newell 1984:fig. 14) shows a somewhat small assortment of artifacts (and no

124

GREGORY A. REINHARDT

refuse) cast about the floor around a single corpse. Moreover, the accompanying data (Table 9.1), which list those objects in three columns (Men’s Side, Women’s Side, and Front of Iglu), were cited to further a hypothesis that “the gender-specific artifacts appeared to be clustered toward the western and eastern halves of the [house]” (Newell 1984:22). This excerpt from a single sentence—coupled with the map and tabulated list (Newell 1984:fig. 14, table 1)—constitutes the original genderbased “sides” premise. Nothing further accounts for either the entire floor-level artifact inventory or for tabulated discrepancies between the gender-based sides and the artifact types; and, importantly, none of those artifacts was actually linked to either sex (Newell 1984:fig. 14, table 1). FLOOR DIMENSIONS AS ANOTHER “PROBLEM” Perhaps the oddest thing to contemplate first in trying to compartmentalize the Mound 44 house into male and female areas is the space in question. From 1981 to 1983 the Utqiagvik Archaeology Project excavated four complete house floors in Barrow (all of which, like the Mound 44 floor, were rectangular). Their average dimensions were just over 3.1 x 2.2 m (about 10 x 7 ft) and an average floor area of around 6.9 m2, or 70 ft2 (Reinhardt and Dekin 1990:table 4-1). The Mound 44 floor was incomplete because floorboards at the sleeping platform end (farthest from the tunnel entrance, or katak, the ingress-egress floor hole) were decomposed, making its precise length unknown. Its minimum dimensions were 2.67 x 1.64 m (roughly 9 x 5 ft), however. Of greater interest than Mound 44’s floor shape is its length relative to width. The length:width ratio of the four other Utqiagvik houses averages about 1:0.7, whereas the Mound 44 house floor, with its incomplete long dimension, must have been narrower, its ratio closer to 1:0.6. In the early 1880s one house in Utqiagvik had thirteen regular residents (Murdoch 1892:75), but six corpses were crushed in the Mound 44 disaster (Reinhardt 1990, n.d.a). So before we even begin to contend with gender-based sides in this house, we must understand that we are considering the possibility for a very small floor area occupied by six people—close to the presumed prehistoric average household size of 6.4 persons around Barrow (Lee and Reinhardt n.d.:appendix 1). Perhaps the original gender-based “sides” premise envisioned bisecting the floor lengthwise when it posited discrete Women’s and Men’s Sides, or “functional partitions,” to the Mound 44 house. Perhaps also the term partition connotes too discrete a division. In any case, the gender-based floor-halves idea must imagine the adults as having sexually segregated themselves

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

125

into areas merely 0.82 m (2.7 ft) wide. To put this in perspective, if adults sat leaning against one long-axis wall with their legs extended (a common Eskimo pose) in this house, their feet would project many centimeters into the opposite gender’s “partition.” Still, that should not be an issue until we can confirm or refute the gender assertion by looking at the distribution of all artifacts, not just Newell’s handpicked ones, from the house floor. METHODS CORRECTING FOR SELECTIVITY IN THE ORIGINAL DATA PRESENTATION Our challenge in assessing the previously published data is that those objects (Newell 1984:fig. 14, table 1) were chosen nonrandomly and comprise only about half of the formal artifacts and about 20 percent of the full cultural assemblage related to the frozen bodies (Reinhardt and Dekin 1990). In hindsight, then, the original gender-based data presentation seems anecdotal. Toward reevaluating the gender hypothesis, Table 9.1 repeats the artifact inventory according to the previous sequence (Newell 1984:table 1). Table 9.2 then tries to clarify the earlier data by indicating which of those artifacts are (1) likely misidentified—for example, “whale bone ice pick,” a harpoon part, should say “pick head,” a digging tool; (2) not shown (Newell 1984:fig. 14) and possibly not present but listed (Table 9.1)—for example, “carving knife”; (3) shown (Newell 1984:fig. 14) but not listed (Table 9.1)—for example, “medial labret”; (4) probably not from the house at all—for example, “lamp supports,” an identification lacking ethnographic precedent in this region but known elsewhere in Alaska (cf. Nelson 1983 [1899]:252, fig. 79); (5) incorrectly presented as having come from within the house—for example, “ivory ulu handle” came from soil outside the house—or incorrectly listed as having been found on the Women’s Side—for example, “skin bag w/sewing equip.”; and (6) from the Men’s Side and the Women’s Side—that is, those from the “Front of Iglu” column (Newell 1984:table 1). More important, Table 9.2 ascribes a gender role to those artifacts for which doing so seems ethnographically supportable. REANALYZING THE ORIGINAL DATA Using Table 9.2, we can statistically assess the strength of the original gender hypothesis. The issue at hand emerges from the “Gender” columns plus the “Side (Female/Male)” column relating to the undefined “Front of Iglu” (Newell 1984:table 1). Table 9.2 eliminates from analysis nearly half of the artifacts originally listed because I regard their typological identities as unclear, their gender ascriptions as uncertain, or

126

GREGORY A. REINHARDT

Table 9.1—Reproduction of table 1 from Newell (1984) “Artifact Inventories of Three Functional Partitions of the Kataligaaq Iglu [catastrophically abandoned house] near Mound 44” Men’s Side

Women’s Side

Front of Iglu

Chert arrow point Slate end blade Bird blunt

Bird blunt Antler arrow point Whale bone adze head with slate blade Boots

Chert flake Baleen comb Ivory comb Wooden buckets

Boot liners Boot soles Skin bags

Wooden platter or dish Ceramic pot Baleen bucket

Skin bag with ivory handle Skin bag with sewing equip. Needle case and holder Thimble holder/belt hook Bird-skin bag Gut bag Baleen cup with human hair Fish box

Wooden plank Walrus hide mat Sewn gut, from skylight

Bag of 6 bone & 1 chert arrow points Whale bone ice pick Adz handles Bag of bolas weights, tooth ornaments, worked bone, & ivory Snow goggles Block plane scraper Chert endscraper Wooden shaft Carving knife Chert drill bit Chert scraper Chert flakes Men’s trousers w/attached boots Hide fragments

Ivory ulu handle Child’s parka Lamp supports Slate ulu blade Slate blade Grooved stone abrader Hide scraper Sewn hide fragments Chert flake

both. For the remainder, totals at the bottom of Table 9.2, when combined according to house side, break down as seen in Graph 9.1. Fisher’s test (Langley 1971:292–297) is akin to the chi-squared test except that it is restricted to two classes, each with two qualities, and it deals with a total number of observations (N) between 8 and 50. Categorized artifacts appear in one of four possible cells: Women’s-Side Female Objects, Women’s-Side Male Objects, Men’s-Side Female Objects, Men’sSide Male Objects. The original gender premise should show high quantities clustering in the upper left and lower right cells. Instead, when applied to the data here, Fisher’s test yields a probability much larger than 5 percent (Graph 9.1), meaning the overall difference in numbers

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA Graph 9.1 Object’s Gender Ascription Female Male

127

Graph 9.2 House Side Women’s Men’s 8 9

4 9

Object’s Gender Ascription Female Male

House Side Women’s Men’s 8 9

4 40

among the four cells is not statistically meaningful. Therefore the hypothesis (that there are demonstrable Men’s and Women’s Sides, based on my added gender ascriptions for artifacts found on those sides) is not affirmed by the original data (from Newell 1984). It may be that the insignificant results are simply the outcome of “sample size rather than lack of a trend” (Peter Whitridge, personal communication 2000), but that does not belie the fact that no statistical test accompanied the original gender premise. The preceding test masks a quantification problem because it counts two packets on the men’s side of the floor as one item each. One packet was a “[bundle] of . . . arrowpoints,” which held seven items (Table 9.2), and the other was a “bag of bola weights [and similar items],” which contained twenty-four pieces (Reinhardt and Dekin 1990:fig. 4-20). These packets (Fig. 9.1) may have inspired the gender-based house-sides notion (Newell 1984:fig. 14). Certainly, their appearance (lying together next to the men’s-side wall), as well as their occurrence twice in his text (Newell 1984:fig. 14, 4 and 5, fig. 22), makes this guess tempting. For reasons detailed later, I prefer the minimalist approach—first because the packets do not represent male activity so much as storage of male objects, and second because including them produces an entirely different effect, as seen in Graph 9.2. Note the explosion in “male” artifacts on the men’s side and the contrasting paucity of “female” artifacts (Graph 9.2), which strongly skews the previous artifact distribution (Graph 9.1). Following this approach, one could posit an incidental female presence in the house—despite its having contained at least two adult female corpses (Zimmerman and Aufderheide 1984:53). An appropriate statistic to use here is Yates’s chisquared test (Langley 1971:285–287), suitable for 2 x 2 contingency tables (four cells in all) when N is more than 50 and there are matched observations on sets of things that have two qualities. In this case the test yields a very high probability: 0.2 percent. Statistically, the “tentative assumption of no significant difference or association [between genderlinked artifact and house side] cannot be denied on the evidence presented”

Boot liners Boot soles [& tops] Skin bags

Skin bag w/ivory handle Skin bag w/sewing equip. **MS Needle case & thimble holder Thimble holder

M ? M

? ?

? M F

[Bundle] of 6 [antler] & 1 chert arrow points Whale bone [pick head] Adz handles Bag of bolas weights, tooth ornaments, worked bone, & ivory Snow goggles Block plane scraper

Chert endscraper

Wooden shaft

Carving knife Chert drill bit Chert scraper

Chert flakes

Bird-skin bag Gut bag Baleen cup w/human hair Fish[-shaped] box **MS

Bird blunt Antler arrow point Whale bone adz head w/[jade] blade Boots

Women’s Side

M M M

Artifact Gender

Chert arrow point Slate end blade Bird blunt

Men’s Side

? ? ?

M F

? ? F

M M M

Artifact Gender

*Wood endscraper handle

Sewn gut, from skylight

Wooden [slat?] Walrus hide mat

6 wooden [tubs & 1 dish] Wooden platter or dish Ceramic pot Baleen bucket

Chert flake Baleen comb Ivory comb

Front of Iglu (side not specified)

M4; W2 & 1 ? W ?

M M M

Men’s/Women’s Side

? F ?

? ? ?

Artifact Gender

Table 9.2—Additions and corrections to table 1 from Newell (1984) “Artifact Inventories of Three Functional Partitions of the [Mound 44 House]”

Grooved stone abrader Hide scraper Sewn hide fragments Chert flake *Float nozzle **O *Medial labret *Polar bear hide mitten *Ivory seal effigy *Split polar bear tooth **O *Walrus tooth bolas weight **O

M M ? ? M M ? ? ? M?

? ? F ?

Women’ side 7F,9M

Child’s parka Lamp supports Slate ulu blade **O Slate blade

? F M ?

Men’s side 3F,9M

Ivory ulu handle **O

Men’s Side 1F,0M Women’s Side 1F,0M

[Brackets] = corrections to artifacts misidentified in Newell 1984:fig. 14 and table 1 Italics = items listed in Newell 1984:table 1 but not shown in Newell 1984:fig. 14, or probably misidentified, or apparently never in the house * = items shown in Newell 1984:fig. 14 but not listed in Newell 1984:table 1 ** = items shown in Newell 1984:fig. 14 but with misattributed locations in Newell 1984:table 1 [**MS = technically, shown and found on men’s side based on mathematical bisection of floor length; **O = actually found outside house, in matrix unrelated to the house floor]

TOTALS

Men’s trousers w/ attached boots Hide fragments Hide scraper handle *Flaker [handle] *2 polar bear hide mittens *Small ivory point

130

GREGORY A. REINHARDT

(Langley 1971:287). In lay terms the test confirms a heavily male-artifact– dominated men’s side. Whether we consider the small original sample size (Table 9.1) or my revision of those data (Table 9.2), either result still begs the question, can we really discern a men’s side and a women’s side to this house, based on all the artifacts known? If the answer is affirmative, using a larger sample of artifacts, it forces us to wonder something else. What, if anything, does this signify about gender-based proxemics—or use and perception of space—at Mound 44? REEXAMINING SEX-BASED “SIDES” AT MOUND 44 “SEXING” THE ARTIFACTS Testing the notion of discrete men’s and women’s sides at Mound 44 is simple enough as a four-step operation. One, divide the house floor lengthwise. Two, designate all artifacts, according to their recorded locations, as having been found on either the so-called women’s or men’s side of the house floor. Three, ascribe gender to the artifacts. Four, examine the results statistically, in this case using Yates’s chi-squared test. The first two steps are easy because I created the grid system used at Mound 44 in 1982, mapped most artifacts from the house floor (as well as above and below it), and kept copies of those maps. The difficulties with this study, of course, lie in step three. To begin with, specifying one’s theoretical assumptions is paramount. The original gender premise essentially forces me to suppose (as Newell [1984] undoubtedly did) “that the activity for which a tool was designed was carried out on the very spot where the tool was found . . . [and] that prehistoric people [from Mound 44] left everything just where they used it” (Flannery 1986:322). Thus my primary supposition is that, all things being equal, most artifacts’ recorded locations reflect essentially where they were when the house collapsed and therefore essentially their position of final antemortem use, storage, and disposal. This is essentially a presumption of “Pompeii effect” (see Binford 1981; Schiffer 1983) or “the Pompeii premise” (Flannery 1986:322), but Mound 44 surely satisfies “Pompeii” conditions. Furthermore, “For archaeologists interested in reconstructing social, economic, or demographic characteristics of households, an assemblage comprised mostly of objects both used in a house and deposited in that same house potentially provides the strongest line of evidence [for household activities]” (LaMotta and Schiffer 1999:21). Another assumption is that, although many objects came from the general house area, only convincingly floor-related ones pertain to this

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

131

study (Reinhardt and Dekin 1990:50). Archaeological remains from the tunnel, kitchen, and storage spaces—all outside the house proper—are beyond the scope of this study, as are ones buried in house-subfloor soil. Yet another operating idea, resulting in part from the difficulty in going back and forth between field specimen numbers and final catalog numbers (explained later), is that same-type objects (e.g., a set of bola weights or fishhooks) from a single location (e.g., from bundles, bags, a box, and the small caches) should count as only one item because they all came from a precisely defined spot. Similarly, if I excavated my own house, I would not count each button from a jar of them or every piece from a silverware set to use in a spatial distribution analysis, but I might count the different kinds of items in a kitchen drawer of mixed utensils. The worst part of step three is correlating genders with particular artifacts. Not wanting to turn this study into a pan-Arctic ethnological discussion (e.g., Giffen 1930) about gender-dominated or gender-exclusive use for each thing found at Mound 44, I lean toward conservative assignations. In refraining from forcing gender assumptions onto artifact types, however, I have to relegate fully two-thirds of the Mound 44 objects to a gender-unassigned category. Pieces in Table 9.3 that I consider gender-uncertain include clothing and footwear, which might be “sexed” for the adult items (but I lack that information); wood- and baleen-walled vessels; cut leather, fur, and gut swatches; thongs and cords of varying thicknesses and lengths; worked and unworked animal bones and teeth; nondescript flakes and blade fragments, both chert and slate; and rodshaped whetstones. I systematically left out all unworked bones unless they were in cache-type locations, whereas animal teeth tended to have uses as charms, amulets, and fishing lures, making them more worth considering but still left as gender uncertain. Table 9.3 artifacts that I present as more related to female-dominated tasks include drying-rack parts; small bags (of gut and bird—and squirrel—skin) and a possible bag handle; ulu parts (women’s slate knife blades and handles); jig-fishing items (lures and hooks, sinker parts, fish-shaped box, baleen line, leister barbs or blubber hooks); sewing items (needle case, bag); composite scraper parts (endblades, handles); scrapers of bone (modified scapulae and metapodials); a possible “crane” for hanging things (Newell’s “lamp support” [1984:fig. 14, 56]); and a ceramic cooking pot. Artifacts in Table 9.3 that I associate more with male-dominated tasks include a different set of objects: harpoon and dart parts (possible dart point, endblades, harpoon head preform, dart shaft, ice pick, float nozzles, finger rest, foreshaft, socket piece); sandstone whetstones; adzes (heads, handle preform);

Table 9.3—Gender-ascribed and gender-sided inventory of floor-related Mound 44 artifacts

FS # Object(s) — 24 25 30 32 100 101 102 103 104 105 106 107 108 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 223 224 225 226

House Side

end blade, slate animal-head cane (see FS107) worked wood drying rack slat polar bear maxilla frag. hide strip, walrus? drying rack slat, wood dart shaft, wood drying rack slat, wood vessel, baleen rim vessel, wood rim (see FS#202) leather frag. animal-head cane (see FS#24) drying rack slat, wood vessel, wood vessel, wood vessel, wood (see FS#105) large hide frag., walrus flake, red chert mukluk mukluk whetstone, sandstone drying rack slat, wood vessel, baleen adz head w/jade blade pendant, tooth (toy bola wt?) bag, gut bag, bird skin mitten, fur hair mukluk liner ulu blade, slate bird blunt seal skin ulu blade bag/bucket handle?, ivory fish lures, antler fish-shape box lid, wood mukluk mukluk liner & sole

Artifact Gender

M W

M ?

W W M W W M W W W

? F ? ? F M F ? ?

W —

? —

W W W — W W W W W W W W W

F ? ? — ? ? ? ? M? F ? M ?

W W W W W W ? W ? W

F? F? ? ? ? F M ? F F?

M M W W

F F ? ?

FS # Object(s) 228 229 230 231 232 233 235 236 237 238 239 240 241 242 243 245 247 248 249 250 251 252 253 254 255 256 257 263 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279

House Side

“pyrite lump” W ulu blade, slate W human effigy, ivory W blade frag., slate W sandstone frags. W fishing sinker top piece, W ivory “shaped & drilled wood” W “shaped wood” W sandstone W chert flake M chert biface, black M “notched wood” M “shaped wood” M bag, leather (see FS#590) M fish-shaped box, wood M worked & drilled wood W “mukluk, etc.” W “gut, mukluk, etc.” W 2 caribou teeth M (fishing lures) marcasite sinker, worked W arrowhead, antler W vessel, wood M “heavy wood shaft” M 2 ground slate frags. W “flake scraper” W flake, green W chert flake, large gray W bundle of arrow points M needle case, bone W labret, medial, stone W charm (bone in leather) W charm (feathers in leather)W scraper end blade, chert W scraper handle, wood M scraper end blade, chert W sewn gut W boot sole creaser, ivory W mukluk sole W fox canine M bag of bola weights, etc. M leather frag. M blade frag., purple slate W 2 sandstone frags. W

Artifact Gender ? F ? ? ? F ? ? ? ? M ? ? F F ? ? ? F F M ? ? ? F ? ? M F M ? ? F F F ? F ? ? M ? ? ?

continued on next page

Table 9.3—continued

FS # Object(s)

House Side

280 bola weight, walrus tooth W 281 cut hide frag., walrus W 282 ulu blade, black slate & W tar wad 283 bird blunt, ivory W 284 grooved white stone W 285 face effigy from bola W wt., ivory 286 bola weight, ivory W 287 bola weight, ivory W 290 bola preform, whale bone W 291 worked bear canine W 292 fishing sinker top piece W 293 vessel, wood W 295 chert flake W 297 wood (worked?) W 298 worked walrus tooth W 299 hide frag. W 300 sandstone W 301 thongs W 302 flake (chert?) M 303 amber M 304 drilled ivory cylinder M 305 cut leather M 306 gut frag. M 307 gut frag. W 309 hair, human M 310 flake (chert?) W 311 sandstone W 312 sinker, brown whale bone W 313 fishing sinker, ivory, W marcasite 314 fishing sinker, ivory, W marcasite 315 2 flakes, chert W 316 end blade made into knife, W slate 317 fish line leader, baleen W 319 tar-sand frags. W 320 leather thong & 5 cut W frags. 322 wood rod M 323 wood rod M 324 worked & pegged wood M 325 braided sinew M

Artifact Gender M ? F M ? ? M M M ? F ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? F F F ? M F ? ? ? ? ? ?

FS # Object(s) 327 328 329 330 331 335 336 337 338 339 340 341 342 343 345 346 347 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373

House Side

ulu frag., black slate slate frag., purple thong w/loop vessel, wood pointed stick worked & charred wood frag. mitten, polar bear fur steatite frag. sandstone frag. steatite frag. steatite frag. sandstone frags., 4 pick head, broken, bone arrow point, gray whetstone, orange chert rod worked bird bone esophagus, bird tooth, polar bear bound bone & wood cut seal leather bola weight vessel, baleen vessel, wood sinker, brown concretion thong, knotted leather, frags., cut, 2 slate frag., purple leather strips, 2 hair mat seal skin, cut bird skin, cut seal skin, cut caribou skin thong, single loop thong seal skin, cut sandstone frags., 8 caribou skin frag. fur mass caribou fur bird blunt wood frag., worked & pegged

Artifact Gender

W M M M M M

F ? ? ? ? ?

M M M M M M M M M

? ? ? ? ? ? M M ?

M M M M M W W W M M M M M M M M M M M M M M M M M M M

? ? ? ? ? M ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? M ?

continued on next page

Table 9.3—continued

FS # Object(s) 375 376 377 384 387 388 389 391 392 393 394 396 397 398 399 400 401 402 403 404 465 466 467 468 469 470 471 472 474 475 476 477 478 479 480 494 495 496 497 498 499 500 501

House Side

hair comb, baleen hair comb, ivory scapula scraper, caribou gut frag. teeth, 1 dog & 1(?) bone frag. fat scraper charm, drilled ivory sinew fur strip inserted bone flake, gray chert bola weight, ivory tooth, seal canine scraper, metapodial, caribou float nozzle, ivory whetstone, sandstone gauged drill, bone bola weight, walrus tooth harpoon ice pick, antler “worked bone/antler” worked whale bone flaker handle, ivory pyrite concretion frag. flake, gray chert harpoon finger rest, ivory wood rod sled shoe, whale bone gut mukluk frag. chert, drilled? harpoon ice pick, antler wood, cut square foreshaft, antler harpoon socket piece harpoon ice pick, antler bola weight whale epiphysis ulu handle, antler baleen strip leather frag. blade frag., slate wood shaft, cut frag. leather, cut square

Artifact Gender

M M M M M M M

? ? F? ? ? ? ?

M M M W W W W

? ? ? ? M ? F

W W W W M M M M W W W M M W M W M M M W M M M M M M M M M

M M M M M ? ? M M ? M ? ? ? ? ? M ? M M M M ? F ? ? ? ? ?

FS # Object(s)

House Side

502 biface, chert M 503 bola weight, seal M calcaneus 504 leather, cut rectangle M 505 leather, cut triangle M 506 tooth, human, extracted M 507 bag, squirrel skin W 509 baleen strips M 510 gut strip M 511 bird skin frag. M 512 bola weight, walrus tooth M 513 flaker, antler M 514 ulu blade, slate M 515 bag, seal flipper W 516 “crane”/blubber dripper, W wood 517 bead, red stone M 518 boot upper, seal skin M 519 scraper end blade, chert M 520 blade, purple slate M 521 boot lower, seal skin M 523 leather, cut M 524 scraper, caribou scapula M 525 hide, cut M 526 fur, bear W 527 leather frag. W 528 knife blade, purple slate M 529 hide, cut M 530 thongs, 2 M 531 fur, caribou M 532 thong M 533 whetstone frag., sandstoneM 534 leather, cut triangle M 535 hide, cut M 536 leather, rectangle M 537 bone, caribou thoracic M vertebra 538 drilled & incised frag., M ivory 539 baleen strip M 540 fur, bear M 541 mukluk & legging M 542 bird quills wrapped in W wings 543 seal skin frag. W

Artifact Gender M M ? ? ? F ? ? ? M M F ? F? ? ? F M ? ? F ? ? ? M ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?

continued on next page

Table 9.3—continued

FS # Object(s) 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 574 575 576 577 578 579 580 581 582 584 585 589A 589B 590 593 594

House Side

net, leather W baleen strip, broken W ladle charm, drilled ivory W netting needle, ivory W thongs & cut leather M ice pick, antler M gut frag. W caribou skin frag. M seal skin frag. M buckle?, incised ivory M blade, gray slate M bag, seal flipper M mukluk sole M seal skin frag. W leather frag., sewn W leather frag., cut, w/holes W arrowhead preform, antler W fur, bear W seal skin frag. W slate frag. M bola, 3-strand (toy?) M net float eyelet, ivory M slate, drilled frag W blade frag., purple slate W slate frag., gray W chert frags., orange, 2 W thong W seal skin strips, knotted W seal skin, cut triangle W flake, gray chert W arrow point, gray chert W leather frag., sewn W flake, gray chert W wood frags., fish-shaped boxW flake, chert W bag, bird skin? W bag, leather W seal effigy, ivory W fishhook shank? M feather, fletching W parka W parka, child’s W bag, leather (see FS#242) — tooth, fox canine W? needle case, bone W

Artifact Gender ? ? ? M ? M ? ? ? ? M ? ? ? ? ? M ? ? ? M M ? ? ? ? ? ? ? ? M ? ? ? ? F F? ? F ? ? ? — ? F

FS # Object(s) 595 777 779 780 781 782 783 784 785 786 787 788 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 811 812 813 814 817 818 819 825 826 827 832

House Side

slate frag. M drill bit, chert W tooth, walrus M bola weight, ivory M harpoon head preform M scraper, caribou scapula M leather frag., cut M arrowhead, barbless antler M flake, used chert M arrowhead, antler M slate frag., ground M leather frag., cut M arrowhead, antler or bone M labret preform M flake, used gray chert M slate frag. M fur, bear M adz head, whale bone M arrowhead, antler M bird skin frag. W adz handle preform frag. W teeth, 2 walrus, 1 bear W incisor flakes, gray chert, 2 W bone frag., caribou W vertebra worked rectangle, bone/ W antler leather strip W drying rack slat, wood W fishhook shanks, ivory M flake, used gray chert M ulu frag., purple slate M blade frag., purple slate M bola weight, whale bone M fur, bear M peg, wood M vessel, wood M flake, gray chert M bag, bird foot M pick/mattock handle, M wood sandstone slab, large M pyrite concretion M pot, ceramic W

Artifact Gender ? M ? M M F ? M ? M ? ? M M ? ? ? M M ? M ? ? ? ? ? F F ? F ? M ? ? ? ? F M M? M F

continued on next page

136

GREGORY A. REINHARDT

Table 9.3—continued

FS # Object(s) 834 835 836 839 907 908 909 910 911 912 913 969 983 984 985 986 988 990 991 992

House Side

wood, burned M gut frag. M pegged wood M cobble M bola preform M bola weights, whale M bone, 5 leister barb/blubber M hook, ivory arrow shaft, wood, toy? M pendant, green stone M rod frag., wood M whetstone frag., sandstoneM arrow point, chert—not W on map harpoon head, ivory, toy? M shaft frag., wood M peg, wood M dart point? M frag., ivory M stone, white iridescent M bola weight, walrus M bird blunt blank?, M whale bone

Artifact Gender ? ? ? ? M M F? ? ? ? ? M ? ? ? M ? ? ? ?

FS # Object(s) 993 mattock head, whale bone 1003 shaft frag., sharpened 1006 drill frag., gray chert 1007 flakes, chert, 2 1011 leister barb/blubber hook, ivory 1013 float nozzle 1014 shaft frag., wood 1015 leister barb/blubber hook, ivory 1016 whetstone frag., sandstone 1017 flake, used gray chert 1018 flake, used gray chert 1019 pegs, wood, 2 1020 arrow point, gray chert 1021 flake, gray chert 1038 worked whale bone 1043 flake, gray chert 1090 thong, knotted 1092 ½ jaw, squirrel, sinew wrapped

House Side

Artifact Gender

M W W W

? M ? F?

W W W

M ? F?

W W W W W M W M W

? ? ? M ? ? ? ? ?

FS#: field specimen number (distinct from Utqiagvik Archaeology Project catalog number) House Side: M = Men’s Side, W = Women’s Side, — = listed elsewhere, ? = uncertain excavation location Artifact Gender: M = presumed male-dominated task, F = presumed female-dominated task, ? = gender-uncertain task

arrow blunts (bird and small-game hunting); the hunting bundle (birdskin-wrapped antler arrowheads and chert arrowpoints); other arrow heads and arrow points; the bag of “men’s” items (bola weights, preforms, scrap ivory and antler, and similar objects); a medial labret and preform; bola weights and toy bola (bird hunting); men’s slate knives (not ulus); drilling items (gauged drill, chert drill bit); pyrite concretions (fire making); a flaker handle (“flint knapping”); netting items (needle, float eyelet); and digging items (pick handle, pick head, mattock head).

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

137

FLOOR SIDING THE ARTIFACTS The process of siding and assigning gender to Mound 44’s artifacts was easier said than done because of the way the Utqiagvik reports were published. Although I wrote an annotated catalog of all artifacts (Reinhardt and Dekin 1990), the editors deleted all catalog numbers from my text, arguing that the numbers would be captioned in the Utqiagvik reports’ photo-plates. As a result, problems abounded. First, not all items got photographed, so their numbers cannot be culled from the text. Second, there was no system for correlating lab-assigned catalog numbers with the field specimen numbers previously assigned to the items upon excavation. Third, Mound 44 contained so many artifacts and architectural timbers, many of which were large enough to sketch, that I produced eleven different field maps to indicate all their locations (i.e., eleven maps on which to find specific field-numbered artifacts). Fourth, some objects were never mapped or even reported as to general location. To produce artifact distribution maps for this study, I began by finding each field specimen number on one of the field maps and then reading the original field log’s description of that artifact to discover its identity. (Map sketches made some identifications self-evident, fortunately.) If the artifact type was still not clear, I looked it up in the Utqiagvik report (Reinhardt and Dekin 1990), hoping to learn more. If I found its description, I usually knew what the object was; if not, I had to look up its catalog number by going through the report’s photo-plate captions appendix. Then, to confirm that the numbers matched, I had to crossreference that catalog number against several computer printouts (generated in 1983) that list Mound 44 artifacts sequentially by field specimen number. If they did not match I had to read other sections of my text, looking again for the right description and, I hoped, for a matching catalog number. Once I had positively identified an item (by field specimen number), I could then map it. If I felt I could ascribe gender-prevalent use to the artifact, I marked it on either the Gender-Female or GenderMale map; otherwise I placed it on the Gender-Uncertain map (Figs. 9.2– 9.4). USING ADDITIONAL DATA TO TEST THE GENDER-BASED “SIDES” PREMISE The data set employed here consists of 365 mapped artifacts and other cultural debris (or sets of these things) associated with the terminal occupation of Mound 44 (Table 9.3). I ascribe 119 of these, in roughly equal numbers, to tasks related to one sex or the other. Three distribution maps present the results. Figure 9.2 (Gender-Uncertain artifacts)

138

GREGORY A. REINHARDT

Figure 9.1 The “male” artifact packets: several arrow points wrapped in birdskin, center, and bola weights and other items in a bag, right, partly covered with debris; a floorboard is visible under the arrow points, the “Men’s Side” wall sill is at the top (left to right), and an upright timber between floorboards is at the left (for orientation, see Fig. 9.4).

shows 246 items I could not ascribe to female- or male-oriented gender roles; it indicates a rather broad distribution of all sorts of objects spread across both sides of the house floor. Figure 9.3 (Gender-Female artifacts) bears 51 items that probably relate to primarily women’s tasks; it illustrates relatively few pieces, which tend to occur about the center of the floor and concentrate on the women’s side. Figure 9.4 (Gender-Male artifacts) displays 68 items I associate with men’s tasks; it presents slightly more “male” objects, which assume something of a U-shaped pattern. The “male” majority also occurs fairly differentially toward the back of the house (away from the katak) and somewhat more toward the men’s side (Fig. 9.4), compared with the greater centralization of “female” objects. Graph 9.3 shows the distribution of gender-ascribed artifacts, according to their appearance on the men’s or women’s side of the house floor.

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

139

Figure 9.2 Gender-uncertain objects; shaded areas=caches (large subfloor cache also shown in dashed lines)

Figure 9.3 Gender-female objects; shaded areas=caches (large subfloor cache also shown in dashed lines)

140

GREGORY A. REINHARDT

Figure 9.4 Gender-male objects; shaded areas=caches (large subfloor cache also shown in dashed lines)

Graph 9.3 Object’s Gender Ascription Female Male

House Side Women’s Men’s 34 28

17 40

Ideally, the upper left and lower right cells should contain higher quantities than those shown, but Yates’s chi-squared test of the distribution data from Figures 9.3 and 9.4 still expresses significant difference: P = 5%. Still, because the arrow-point bundle and the bag of bola weights (and similar items) add 31 pieces to the total, it is important to account for this larger number distribution (raising the total to 150 artifacts), as seen in Graph 9.4. In this case the preponderance of male artifacts on the men’s side far outstrips the predominance of female artifacts on the women’s side,

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

141

Graph 9.4 Object’s Gender Ascription Female Male

House Side Women’s Men’s 34 28

17 71

and Yates’s chi-squared result is a highly significant probability of 0.2%. I prefer the first and more conservative distribution, however, as explained later. DISCUSSION CONTEMPLATING THE RESULTS These statistics suggest that a measurable difference indeed exists in artifact dispersal within the house, divided lengthwise into opposing sides. Does this mean there is validity to the original notion of a women’s and a men’s side, if only vaguely expressed (Newell 1984:22)? Probably not, although the desire to concur seems compelling at first. True, the gender-female artifacts are twice as prevalent on the women’s side (Fig. 9.3). As for gender-male artifacts, they either slightly prevail on the men’s side (if we minimize the bundle and bag) or overwhelmingly occupy it (if we itemize the bundle’s and bag’s contents). I nevertheless hesitate to confirm the idea of men’s and women’s sides, as originally proclaimed, because we need to address the rather Ushaped dispersal to the gender-male artifacts (Fig. 9.4). That is, half of the gender-male artifacts appear either in (1) the large subfloor cache (under floorboards below the sleeping platform at the back of the house, opposite the katak) or in (2) other peripheral storage caches (on the sill timbers [between the bases of wall planks] or below or between floorboards) (Fig. 9.4). Peter Whitridge’s dissertation research at the Qariaraqyuk site (see Chapter 11, this volume) observed “a relatively widespread distribution for men’s cached and discarded items, and a relatively delimited distribution for women’s, with all that that might imply for the control of social space” (Whitridge, personal communication 2000). Eliminating the cached items at Mound 44—and focusing instead on the floor-surface artifacts—tends to “neutralize” the notion of gender-specific sides to the floor. The challenging issue that emerges from this study is what male and female house sides, if they existed in emic Inupiat reality, signify about proxemics in the first place. Do Yates’s chi-squared results given previously

142

GREGORY A. REINHARDT

imply that men worked on one side of the house while women worked on the other? Why are there more “male” than “female” objects, particularly in the large cache? More significant, why are there not more “male” artifacts centrally located on the men’s side? Salient to these questions is the issue of sea mammal oil lamps, source of heat and light in Inupiat houses. Even though we never found it in the Mound 44 house, the main lamp (if not the sole one) had been on the women’s side, which is consistent with the original gender-based-sides premise. I am certain of the lamp’s general position because during excavations I noted much more lamp oil residue near the wall on that side of the house, both across the floorboards and in the soil below. To either side of the katak frame below the floor, and just north from it [i.e., toward the back of the house], are cut sods . . . those on the [women’s side] being particularly seal-oil-soaked (where the seal oil seems to be most thickly accumulated under the floor). . . . Some seal oil extends to the [men’s-side corner near the katak] of the subfloor and some is also seen in the [women’s-side corner near the katak], but [there is] far less along the [men’s side] of the house. What did appear [on the men’s side] was most abundant along the [katak boards’ wall edges]. (Reinhardt n.d.b)

The lamp complication is that, if men were in the house and working on things there, it would make more practical sense for their items to occur near the lamp (the women’s side), where they could see better. Instead, objects such as those from the arrow-point bundle and the bola weight bag (Fig. 9.1; Table 9.1, left column) appeared about as far from the lamp as possible (Fig. 9.4)—near the wall on the presumably darker men’s side of the room. PROPOSING AN ALTERNATIVE INTERPRETATION These results are not as definitive as they might be. I therefore propose a kind of “independent test” that another marked difference exists, in artifact distributions, according to how objects cluster on the men’s and women’s sides of the Mound 44 house. Let us consider two sets of gender-uncertain objects not typically regarded as formal artifact classes: sewing scraps and chert flakes (Table 9.3; Fig. 9.2). Let us assume that certain scraps (cut-up leather, gut, hide, and fur—all either stitched or not) result from tailoring and hide preparation; that is, they reflect female tasks (Giffen 1930:33–34, 87). (Sinew, thongs, and knotted leather or fur strips are potentially unrelated to sewing and might as easily relate to male tasks, thus their exclusion from analysis.) Let us also as-

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

143

Graph 9.5 Gender-linked Waste Materials? Leather, etc. Chert

House Side Women’s Men’s 22 16

37 3

sume that chert flakes result from steps in lithic manufacture; that is, they reflect male tasks (Giffen 1930:33, 87). Granting that these premises might be wrong, the test still presents itself as an intriguing counterpoint to the preceding ones that deal with formal artifacts. Paralleling previous test constructions, the outcome is seen in Graph 9.5. Yates’s chi-squared test of these figures indicates a very significant difference amid the four cells: P = 0.2%. The materials just examined, leather (and other) scraps and chert debitage, are presumably by-products of basic manufacture (Figure 9.5). Together they constitute about 32% of all gender-uncertain objects from Mound 44 that relate to the house’s final occupancy and about 21% of all field-numbered objects in Table 9.3. (Although some field specimen numbers—FS#s in Table 9.3—represent more than one object, I assume that the cases of multiple items per number are about evenly split between men’s and women’s sides of the house, making the exact quantity unimportant in this and in the preceding Yates’s chi-squared tests.) Among the formal artifact classes, “female” items tended to be on the women’s side and “male” items on the men’s side (Figs. 9.3, 9.4). Among the analyzed “waste” objects, though, the relationship between presumed gender role and house side apparently reverses. That is, more “female” pieces (“sewing scraps”) occupy the men’s side, whereas “male” ones (“waste flakes”) strongly favor the women’s side (Fig. 9.5). These puzzles probably relate to another simple ethnographic reality. Except in midwinter, Inupiat men and older boys in this part of Alaska spent most of their waking hours in a qargi, or men’s house—a kind of clubhouse and workplace built elsewhere in the village—and returned to their domicile mainly to sleep (Murdoch 1892:80). This fact is rather consistent with the original gender thesis in that the Mound 44 disaster may have occurred “between early October and the end of November or between the end of April and May/June” (Newell 1984:26), which is basically outside the hard winter period. Therefore, better than insisting on male-dominated and/or femaleintensive sides to the house interior, it might be easier to suppose that

144

GREGORY A. REINHARDT

Figure 9.5 Leather (and other) scraps and debitage; squares=leather and similar items, triangles=chert debitage, open circles=used flakes, shaded areas=caches (large subfloor cache also shown in dashed lines)

women used most of the floor and usually shared it with the younger children. In this interpretation, women probably sewed by the lamp and either tossed scraps away from it or stored them in the large subfloor cache (Fig. 9.5). They also spent a good deal of time trimming the lamp wick, which kept them close to the lamp—in effect the women’s workplace, a spot we could expect to be freer of debris and formal artifacts (Fig. 9.3). The presence of chert flakes on the women’s side seems harder to explain at first blush, insofar as having small, sharp objects about the floor on which adults worked and children played could pose a slight health risk. Although speculative, one might argue that women threw their work-related refuse to the other side of the room, and men did the same thing (Fig. 9.5). It could be that women did their own flaking (although this idea appeals less), or it might even have happened that children (boys?) were trying their hand at this skill. Perhaps the floor had recently been swept clean of flakes, or maybe chert was not even pro-

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

145

cessed in the house but brought in for women’s use (by either women or men). If so, maybe the nineteen locations for chert debitage do not particularly mirror males’ stone knapping, which should generate hundreds of flakes around the house. Instead, it might reflect females’ keeping a few flakes nearby as relatively “disposable” cutting tools to supplement their general use of ulus, which (being made of slate) are consequently duller than siliceous cherts. The appearance of visibly used flakes on both sides of the floor probably does not affect this discussion either way; in any event, all five were cached (Fig. 9.5). We know men and older boys returned to the house to sleep, but they spent much of their waking time and did most of their tool work in a separate structure. Accordingly, they would not need their artifacts near the lamp in the house, which would clutter the women’s lamptending area and the house floor in general. The alternatives for males were either to place their artifacts and work materials on the opposite side of the floor—by the wall, where we found the arrow-point bundle next to the bag of bola weights—or to store them below floor level in the large out-of-the-way cache or in other lateral caches. Viewed in this light, the gendered proxemics at Mound 44 clearly defy the original premise of “functional partitions,” or seemingly assigned work spaces for women and men. To illustrate my point, I redefine the 119 gender-ascribed formal artifacts into cache locations (those peripheral to the house floor) versus noncache locations (those more toward the central floor area) (see Graph 9.6). Yates’s chi-squared test indicates that 5% >P >1% for the distribution in Graph 9.6. Statistically significant, the pattern is numerically almost identical to that for house sides. By a 2:1 ratio, female artifacts occur on the women’s side and toward the house center, whereas male artifacts are slightly more common on the men’s side and on the periphery. Adding in the 31 items from the arrow-point bundle and the bola weight bag affects the Central Location Male-Object cell, as seen in Graph 9.7. In this case Yates’s chi-squared test produces a probability greater than 10 percent, meaning the numbers in the four cells are not statistically significant. They clearly reverse the interpretation immediately previous, but they do so by falsely exaggerating the importance of two sets of parceled artifacts. Both sets, being against the far wall from where I think the lamp was located (Newell 1984:figs. 14, 22), were in a real physical sense removed from the house center. Therefore I believe my second center-versus-periphery analysis is far less valid. It confuses storage with

146

GREGORY A. REINHARDT

Graph 9.6 Object’s Gender Ascription Female Male

House Location Central Peripheral 34 31

17 37

Graph 9.7 Object’s Gender Ascription Female Male

House Location Central Peripheral 34 62

17 37

Graph 9.8 Object’s Gender Ascription Female Male

House Location Central Peripheral 34 31

17 68

presumed active use of artifacts in the house. If we reverse the location of the bundle and bag contents (i.e., move them to the Peripheral-Location Male-Object cell) to distinguish curation from work, the result changes dramatically and accords better with my thesis (see Graph 9.8). Now Yates’s chi-squared probability is 0.2 percent, and both genders’ pieces reflect a 2:1 ratio. Twice as many “female” items are in the center, and twice as many “male” ones are in the periphery. Furthermore, “female” formal artifacts occur around the lamp area (Fig. 9.3) more than “male” artifacts do (Fig. 9.4). Scrap materials, by contrast, tend to be either stored in the large subfloor cache or scattered far from the lamp area and toward the opposite wall (Fig. 9.5). “Male” formal artifacts (Fig. 9.4), alternatively, array themselves in a way very similar to the dispersal of scrap materials (Fig. 9.5), which I treat as added evidence for gender-based differences in the use of house space. Peter Whitridge arrived at very similar results when he examined distributions of artifacts found in five houses and a qargi at Qariaraqyuk (Whitridge 1999a). In his study the parallels had to do with hide-working refuse (“ ‘cast-offs’ of various kinds, sinew, clothing fragments”) at one scatterplot extreme compared with hard-material–working refuse (i.e., “ivory, antler, and whale bone debitage”) at the other extreme. In other

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

147

words, they tended not to co-occur, and more “male” debris appeared in the qargi, whereas more “female” debris was found in dwellings. In addition, “Within the dwellings female refuse tended to occur on the floor and in the kitchen, and male refuse on the sleeping platform and in the tunnel” (Whitridge, personal communication 2000). CONCLUSION No scientific study should ever be considered either complete or irrefutably conclusive. If it were, there would be no more questions—nothing left to say about that topic or, ultimately, about the science under which rubric the study appeared. In examining gender at Mound 44, more could have been done. For instance, I might have tried to create isopleths of house-floor artifact density to expose more visual representations of female and male household activity (Reynolds n.d.). I might have dismissed the idea of two gender-based “sides” and looked instead for “natural” artifact and debris distributions. I might also have produced two more distribution maps (one each for gender-female and gender-male artifacts), but with the cached objects eliminated; still, these should be discernible in Figures 9.3 and 9.4. (My thanks to Peter Whitridge for suggesting these lines of further inquiry.) A major thrust of this study concerns unexpected outcomes. I had long thought Newell’s gender suppositions (1984) were unsubstantiated— at least insofar as he surmised—and that seems borne out here. It surprises me, though, to discover that there is a statistically significant difference between the so-called men’s and women’s sides in terms of artifact ascriptions as either male or female. It intrigues me even more to realize that yet another behavior pattern emerges for two categories of waste materials and that the center-versus-periphery test seems to clarify that difference. In other words, Newell (1984) was right about male and female “sides” at Mound 44 but for the wrong reasons, both methodologically and ethnographically. In the final analysis, though, I am skeptical of my own certainty, with which I began this research, as to what is “male” and “female” and what we can comfortably and testably assign, based on gender, to the archaeological record. As Sara Nelson phrased it, “The dichotomies produced by our language and our categories [see Chapter 2, this volume] ntend to obfuscate what could be learned about the prehistoric past, thus preventing us from encountering the variety left by the actual users of the tools” (1997:92). Statistical probabilities might lend the data-testing here an aura of assurance about artifactual femaleness or maleness. That, however, should not supplant our intellectual responsibility to know

148

GREGORY A. REINHARDT

two things first: (1) what we mean by sex-based ascriptions, and (2) whether those ascriptions have any emic validity. Still, how can we fully apprehend or accurately visualize an alternative past for an alternative culture? The glaring gender problem surrounding Mound 44 is proving with total, or even reasonable, confidence that any adult males were actually members of this household when the roof crushed the people inside. We see “male” artifacts, but although two female corpses came archaeologically from the Mound 44 house, no identified male remains were ascertained, even though Newell (1984:35) posited a missing “Senior male.” One might argue that males were in fact altogether absent, owing to the dearth of male paraphernalia on the house floor. Seen differently, maybe men were absent but one or more boys or adolescent males present. Then there is the matter of whose artifacts belonged to whom (which sex or which person), what ownership itself means, and how we might ever archaeologically establish such property rules (Barbara Bodenhorn, personal communication 1998). For instance, Barbara Crass (personal communication 1998) found no statistical difference between the sexes in the number of harpoons (an indisputably “male” artifact) in her study of pan-Eskimo burial practices. A related consideration is the extent to which males and females might have shared their material things. The use by one sex of the other’s equipment is not uncommon and constitutes, particularly for the archaeologist, a difficult problem in social unit identification. What is the proof of a man’s or woman’s presence? Men frequently camp out by themselves for a day or two in the fall. They carry with them women’s gear (cooking utensils, food). In this situation, men are using women’s equipment and it is difficult if not impossible at times to determine from the camp debris or the items left at a permanent camp whether a woman was present. (Ackerman 1970:40)

So even if we can distinguish male from female artifacts and debris, and even if “sex-linked spatial control [can be] marked by the storage of items specific to male or female activities” (Ackerman 1970:38), this does not necessarily mean men were present or that women did not also use men’s things at Mound 44. Another caution is that the house’s collapse captured for archaeology and posterity only one precise instant in the lives of its residents, and this instant was only one phase in their annual round, and this phase affected only one household (no doubt with its own idiosyncra-

PUZZLING OUT GENDER-SPECIFIC “SIDES” TO A PREHISTORIC HOUSE IN BARROW, ALASKA

149

sies) out of an entire community and society. Archaeologists generally (if often tacitly) assume that, all things being equal, a part reflects the whole. We cannot resolutely conclude, however, that the analytical picture described here represents all lives in that period or all seasons of the Mound 44 householders’ yearly routine or all interactions between the sexes then. Pressing the data beyond the scope of this chapter to learn more about gender is a logical further step. In fact, “It seems that a responsible archaeologist would make the recovering of gender relationships an explicit goal, and would use all available powers to delineate specific social structures” (Nelson 1997:129). Still, caution is warranted: “While a consideration of gender in household studies will unquestionably increase the archaeological visibility of women, paradoxically it will only contribute to a further reification of the link between women and home unless there is an accompanying awareness of women’s activities outside the domestic environment and of men’s activities within it” (Lawrence 1999:122). Mound 44, with its unique preservation, has much to offer in these regards. If nothing else, it steers us away from simplistic interpretations of household roles for men and women among the prehistoric Inupiat of Arctic Alaska. In place of such thinking, Mound 44 evidently instructs us about two cultural behaviors, house-specific though they might be: a marginalized place for males in daily domestic activities, and a living arrangement no doubt struck by mutual consensus of both sexes. Putting a face to those who died in Mound 44, I have to imagine a lifestyle radically removed from my own. Consistent with ethnography, the women likely lived most days apart from men and managed their houses and children alone. If so, women and men agreed to a sort of daily divorce, and children learned a principally female worldview before recognizing the perspectives of males. Yet somehow, despite the seemingly compartmentalized inflexibility of adult sex roles (Giffen 1930), these Eskimo women and men nevertheless cooperated in composing a life together. They cooked meals, shared food, banged drums, survived disasters, gamboled naked, dug roots, fought frostbite, laughed heartily, built fires, picked berries, constructed houses, nurtured children, drove dogsleds, respected elders, wore jewelry, framed kayaks, hunted seals, held feasts, buried friends, sewed clothes, feared nature, argued sharply, hooked fish, assembled tools, endured famine, sang melodies, erected tents, killed whales, made love, tended lamps, rowed umiaks, danced fervently, mediated disputes, sacrificed dreams, nursed illnesses, battled invaders, grew old, and died.

150

GREGORY A. REINHARDT ACKNOWLEDGMENTS

I thank Lisa Frink, Tammy R. Greene, Kelly E. Guthridge, Stephen P. Nawrocki, Jeffrey A. Oaks, Christopher W. Schmidt, Rita S. Shepard, Peter Whitridge, and Matthew Williamson for insights that contributed to the thoughts in this chapter and Marijane Hetrick for assistance with remapping artifacts from Mound 44.

10 BROKEN EYES AND SIMPLE GROOVES UNDERSTANDING EASTERN ALEUT NEEDLE TECHNOLOGY THROUGH EXPERIMENTAL MANUFACTURE AND USE OF BONE NEEDLES BRIAN W. HOFFMAN

SOMETIME AROUND A.D. 1000 THE UNANGAN, or people of the eastern Aleutian Islands, switched from using bone sewing needles with delicately drilled eyes to needles with simple grooves carved on their ends for attaching the thread (Figure 10.1a). This seemingly inconsequential morphological change in a tool shape came at a time of great social and economic transformations. The Unangan of A.D. 1000 also switched from occupying relatively small single-family or extended-family homes to living in large multifamily longhouses (Hoffman 1999a; McCartney 1984). They began importing slate knives in large numbers, as well as acquiring metal tools (Holland 1988; McCartney 1988). Warfare probably increased (Maschner and Reedy-Maschner 1998), as did social differentiation and inequality (Hoffman 2002; Maschner and Hoffman n.d.; McCartney 1984). The exact timing of these changes and whether they occurred independently or as part of a package remain undetermined. What is known is that by the time Russian explorers first arrived, the Unangan were culturally complex maritime foragers with ranked lineages who engaged in long-distance trading and warfare while living in large permanent villages (Black and Liapunova 1988; Lantis 1984). For most archaeologists the change in needle form is simply diagnostic of “Late Aleutian Tradition” assemblages with little additional behavioral significance (Desautels et al. 1970; Holland 1982, 1988; McCartney

152

BRIAN W. HOFFMAN

Figure 10.1 (a) Grooved needle and (b) red fox left tibia with needle blank removed via the groove and snap technique; both artifacts recovered from Agayadan Village (from Hoffman 2002)

1967). Recent excavations at Agayadan Village on Unimak Island, however, reveal sewing needles as an important material component of highstatus households (Hoffman 2002). The change in needle form, far from inconsequential, helps explain the interplay among gender roles, prestige-goods production, and status differentiation in Unangan society. This chapter explores why the Unangan changed the form of their needles by examining the manufacturing costs and functional attributes of each needle type. Experimental production and use of both eyed and grooved needles confirm that reducing needle diameter was the critical advantage gained by switching to grooved needles. Thinner needles enabled Unangan seamstresses to sew finely stitched and elaborately decorated garments. The Unangan skills with the thin grooved needles made their sewn products highly desired status symbols in their increasingly complex social world. NEEDLES AND GENDER Investigating gendered behavior from a society’s material remains can be a difficult undertaking for archaeologists. Assigning a specific object to the activities of a single segment of a social group raises serious practical and theoretical questions, as discussed by several authors in this volume (Reinhardt, Shepard, Whitridge) and others elsewhere (Costin 1996; Hayden 1992; Owen 1999). Whether the Unangan needles were largely manufactured and used by women or by men makes little difference when answering this chapter’s principal question, which is why the needles changed in form. The answer to that question, however, acquires added interest when the gendered implications of this technological change are considered (cf. Brumfiel 1991; Costin 1996). Further-

BROKEN EYES AND SIMPLE GROOVES

153

more, assigning needles and sewing activities to Unangan women is an interpretation amply supported by cross-cultural studies (Byrne 1999; Hoffman 2000) and historical records. Virtually every eyewitness account of Unangan labor practices from the eighteenth and nineteenth centuries clearly states that Unangan women made their own needles and sewed all clothing (Klichka 1988:266; Krenitsyn and Levashev 1988; Merck 1980:77, 173, 203; Sarychev 1969:8; Veniaminov 1984:285, 287). Aleut men may have carried needles to make emergency repairs on their clothing or kayak covers when traveling, but the vast majority of needles recovered archaeologically from house interiors are the residue of sewing activities by Unangan women (Hoffman 2000). EYED AND GROOVED NEEDLES IN ALEUTIAN PREHISTORY Bone sewing needles are a regular component of Arctic archaeological assemblages (De Laguna 1975; Holland 1982; Issenman 1997b; McCartney 1967). The earliest needles have drilled or incised eyes. Eyed needles continue to be used in most regions up until contact. Needles with a slight groove for thread attachment appear in the Aleutians, the lower Alaska Peninsula, and the Kodiak Archipelago around A.D. 700–1000. This style needle completely replaces eyed needles in the eastern Aleutian region. Western Aleutian and Kodiak Archipelago sites apparently contain a mix of needles, demonstrating contemporaneous use of both styles by the people of these regions (Desautels et al. 1970:236; Heizer 1956; Deborah Corbett, personal communication 2001). Archaeologists have been unable to explain adequately the change in needle form. Frederica De Laguna, in her survey of Arctic artifact distributions, described the grooved needle as a “degenerate type” (1975:198). This description implies that Unangan seamstresses somehow became less skilled or lacked the motivation necessary to make eyed needles. Although no Aleutian historian today would agree that Unangan needles “degenerated” in form, many express amazement at the skill required in making the delicate eyed needles, particularly drilling the sometimes incredibly fine eye (Black 1982; Hoffman 2000). This perception led Kathryn Holland (1982:125) to hypothesize that Unangan seamstresses switched to the simpler grooved needle form because eyed needles were difficult to make. A. P. McCartney (1967:359) offers no explanation for the switch to grooved needles, but he does suggest that eyed needles were sturdier, whereas grooved needles could easily be regrooved after breaking. No one has suggested what I consider the most obvious advantage, which is that grooved needles can be made thinner because they are not limited by the size of the eye (Hoffman 2000).

154

BRIAN W. HOFFMAN

Assuming one or more of these hypotheses is correct, Unangan seamstresses redesigned their needles for practical reasons. They were evidently concerned with lowering their production costs, improving needle durability, reducing needle diameter, or a combination of these factors. Experiments in the manufacture and use of replica eyed and grooved needles offer us a means for evaluating these “practical” hypotheses. Working with the False Pass, Alaska, high school class over the course of two days, we made and used both needle styles. Our experiments were designed to address three simple questions: (1) Are grooved needles quicker to make than eyed needles? (2) Which needle style lasts longer? (3) And finally, can grooved needles be made thinner than eyed needles? Needle replica experiments made by Stephanie Butler (1997) and Jeffrey Flenniken (1978) provide additional insight for this chapter. I also draw on archaeological evidence of production recovered from Amchitka Island in the western Aleutians and Agayadan Village (49 UNI [Unimak] 067), a late precontact/early contact Unangan settlement on Unimak Island (Hoffman 1999a, 2002). MAKING NEEDLES— ARCHAEOLOGICAL AND DOCUMENTARY EVIDENCE The first issue with our project, as with any experimental program, concerned decisions regarding the materials we employed and the manufacturing steps we followed. We used both historical accounts of Unangan needle manufacture and archaeological data to guide our decisions. Several eyewitness accounts from the late eighteenth century describe the manufacture and use of bone sewing needles by Unangan seamstresses. One of the most detailed observations comes from Dr. Carl Merck, a naturalist and physician with the Russian-sponsored Billings-Sarychev Expedition that traveled in the Aleutians between 1790 and 1792.1 Describing Unalaska Aleut, Merck wrote that “[Unangan] women also cut seagull bones with [an iron knife] carefully. They sharpen these bits of bone on a spongy kind of volcanic rock, and so make sewing needles” (1980:77). The commander of the expedition, Captain Joseph Billings, added that “their needles can be considered a product of accomplished craft; they are cut from the femoral bone of birds, of a large type of gull. The very smallest needles are not more than a third of a vershok long [vershok = 4.4 cm] and as thin as hair. However, they have no eye, and on the blunt end they have a very narrow cut groove, in which they fasten the thread so artfully that it follows the needle in sewing unimpeded” (in Merck 1980:203).

BROKEN EYES AND SIMPLE GROOVES

155

Table 10.1—Needle Production Steps, Materials, and Debris (*recovered at Agayadan Village, † documented in historical accounts) Step

Materials Used

Debris Generated

1: blank production

metal knife† stone flake/knife*

a. bone cores w/blanks removed* b. “cut” ends (long bone articular ends removed from core)* c. blanks and blank fragments*

2: blank reduction

stone abrader*†

semiround needle fragments*

3: modify blunt end

acutely pointed stone

partially drilled eyes (?) flake*

Finally, Lieutenant Gavriil Sarychev (1969:8), second in command of the expedition, noted that even when given iron needles, Unangan women “immediately break off the eye, and rub it on the edge of a stone [until] they have made a notch, where they can tie the thread in their usual way.” From these observations and other eyewitness accounts (Klichka 1988:266; Krenitsyn and Levashev 1988; Veniaminov 1984:280), we can outline the basics of Unangan needle production. The preferred raw material was bird bone, with both gull and albatross mentioned specifically. These birds are strong fliers and as a result have relatively hard, dense bone, particularly in their wings. Fox and other medium-sized terrestrial mammals, along with walrus ivory, also provided suitable raw materials. The Unangan of Unimak Island at least occasionally used these alternative materials, as demonstrated by the recovery of a modified red fox (Vulpes vulpes) tibia found at Agayadan Village (Figure 10.1b). Manufacturing needles involved three basic steps (Table 10.1). First, a needle blank was obtained. Next, the blank was either carved or ground to size and shape. Finally, the blunt end was drilled or grooved to allow for attaching the thread. Eyed needles were probably drilled prior to final thinning and polishing while the blunt end was still relatively thick and sturdy. STEP 1—OBTAINING THE NEEDLE BLANK Unangan women of the late eighteenth century, as noted by Merck (1980:77), used iron knives to “cut” the needle blanks. Iron tools, obtained either through long-distance trade networks or from shipwrecked Asian vessels, occurred in the Aleutians by or before A.D. 1000 (McCartney 1988). Sharp stone flakes or ground stone knives were probably used by Unangan women who lacked access to metal tools.

156

BRIAN W. HOFFMAN

Production debris recovered from archaeological sites demonstrate that the needle blanks were not literally “cut” from the bone but removed by a “groove and snap” technique (Figure 1b; see also Jochelson 1925). This technique involves incising long parallel grooves into a bone “core” and then prying or snapping out the linear blank. The groove and snap technique is a method found around the world for working with bone and antler (Clark and Thompson 1953; Knecht 1997). This technique exploits the linear strength of long bone elements and produces thin, long blanks ideal for making needles, points, and other tool types (Scheinsohn and Ferretti 1995). Other methods for obtaining a tool blank, not documented in historical accounts, include simply smashing or twisting a long bone to produce irregular bone splinters. Flenniken (1978) employed this method in his experimental replication of Paleo-Indian needles. Archaeological evidence from the eastern Aleutians indicates the Unangan did regularly smash bird bone. We have no evidence, however, that these splinters served as needle blanks. Instead, most modified bird-bone splinters were simply sharpened at the tip and used as awls (Czederpiltz 1997; Hoffman 2002). Archaeologists working on Amchitka Island in the central Aleutians documented a third method for obtaining needle blanks. They recovered a bird radius “ground on two sides of the shaft leaving two, thin needle-like blanks” (Desautels et al. 1970:264). This “Amchitka” method does not appear to have been widely employed, as no other archaeologists have reported this distinctive production debris. Whether the Unangan obtained blanks from fresh, green bone or dried bone is not recorded in published eyewitness accounts. Butler (1997), in her needle replication experiments, found dried bone difficult to groove and soaked her bone in water to keep it soft. I suspect Unangan seamstresses may also have preferred fresh bone for their needle production. STEP 2—SHAPING THE NEEDLE Thinning and sharpening the blanks into a needle shape is the second step in the manufacturing process. Blanks could be shaved thin by carving with a knife or by grinding on an abrasive stone. Production debris recovered from Aleutian sites indicates most needle blanks began as slim 2–3 mm wide slivers (Hoffman 2002). I suspect Unangan seamstresses did not attempt to carve these already thin needle blanks but instead ground them to shape as described earlier by Merck (1980:77). A variety of coarse to fine abraders and whetstones are typically recovered in Aleut house interiors in general association with needle produc-

BROKEN EYES AND SIMPLE GROOVES

157

tion debris (Hoffman 2002). On Unimak Island these abraders included a sandstone-like tuff suitable for rough shaping and a tuffacious siltstone suitable for fine polishing and final shaping (Hoffman 2002). None of these abraders has needle-size grooves, however, like those created and described by Flenniken (1978:67) from his experiments. STEP 3—MODIFYING THE BLUNT END (EYES AND GROOVES) The final step in the production process, once the needles are polished to shape, involves modifying the blunt end by either drilling an eye or inscribing a narrow groove. Sarychev (1969:8) suggested the grooves were made simply by rubbing the needle against the sharp edge of a stone. The grooves could also be cut with an iron knife or stone flake. Unangan seamstresses of the eastern Aleutians did not make eyed needles at contact, so no regionally specific accounts of this manufacturing step exist. Butler (1997) and Flenniken (1978) both used acutely pointed stone flakes to drill needle eyes. Iron drill bits could also have been used, although we have no documentary evidence of metal drills. FALSE PASS NEEDLE PROJECT False Pass, Alaska, is presently the only village occupied on Unimak Island. It is a small community, with a 1990 census population of sixtyeight residents. Commercial fishing dominates the local cash economy. A 1987–1988 study by the Alaska Department of Fish and Game found that subsistence harvests continue to provide substantial economic benefit (Fall et al. 1996). On average, each False Pass household used, harvested, or received as gifts nearly 1,300 pounds (usable weight) of wild foods, mostly salmon and caribou (Fall et al. 1996:79). Students in the False Pass school are regularly involved in a wide range of Alaska Native cultural heritage studies. Many participate in traditional beadworking, weaving, and skin-sewing projects.2 The high school class (a multigrade class of six students and one teacher during the 1999–2000 school year) and I collaborated on the needle replication project because of our mutual interest in understanding the role of needle technology and sewing activities in ancient Aleut communities. The students and their teacher made replica sewing needles following the procedures outlined earlier (Figure 10.2). For raw material we used the femur, humerus, ulna, and tibiotarsus bones from ptarmigan (Lagopus sp.). We used ptarmigan because these birds were the only species hunted at the time of our study and the only fresh bone available. The ptarmigan bone is less robust than gull and albatross and was

158

BRIAN W. HOFFMAN

Figure 10.2 False Pass High School class replicating needles; left to right, Alan Hoggard, Carlin Hoblet, Teshla Freeman, Karita Freeman, Virginia Berlin (teacher), Nikki Hoblet, and Christopher Freeman

probably rarely, if ever, used for needle manufacture in traditional situations. Chert and basalt flakes, similar to the lithic materials available on Unimak Island, provided our cutting implements. Mostly, we used unmodified flakes, although a few students resharpened their flakes with an antler pressure flaker. Abrading the needle blanks was done using a variety of locally available coarse volcanic rocks. Included in our experiments were fragments of a sandstone-like tuff abrader recovered from disturbed contexts at Agayadan Village. The needles the students made were short and thick compared with archaeological specimens. The differences between the students’ replicas and actual needles reflect the poor quality of ptarmigan bone as a raw material, the limited time available for the experiments, and to some degree the students’ impatience with the needle fabrication process. One conclusion their efforts amply illustrated is that needle making is a timeconsuming activity. I believe the results of our experiments would remain unchanged, however, had the students produced full-scale needles.

BROKEN EYES AND SIMPLE GROOVES

159

The only significant difference is that the production time would have been at least double the time spent on our attempts. We treated each attempted needle/blank as a separate experiment. We recorded the time spent on each stage, whether the production was successful or not, and any reason for failure. We also measured and traced each blank/needle at each production stage. The students sewed with each successfully manufactured needle for 20 minutes or until the needle broke, recording the results of this experimental use. These data allow us to evaluate the manufacturing costs and durability of each needle type. NEEDLE PRODUCTION RESULTS We made our needle blanks using the groove and snap technique. We successfully removed needle blanks from seven (63.6%) of eleven experiments. Failures all occurred when attempting to snap the blank out of the bone core. Grooving the ptarmigan bone was a very time-consuming process. Each blank took an average of 27 minutes (range 10 to 40 minutes) to make. Grooving and removing longer blanks would have been even more costly. Butler (1997:8) worked for 30 to 45 minutes to make each groove in her experiments. Not included with our experiments was one student’s attempt to obtain a blank using a splinter technique. This attempt produced many irregular slivers, but none was judged adequate for our needs. Of the seven needle blanks, the students were able to grind and polish five (71.4%) into a needle shape. The two failures occurred when the needles snapped because too much pressure was applied as the tip was being sharpened. We found grinding the needles, like making the blanks, a very slow process. The students spent an average of 19.4 minutes on this stage (range 7 to 25 minutes). We could easily have ground our needles for an additional 20 minutes or more, since our finished needles were relatively thick compared with archaeological specimens. For the final production step we had six experiments, three eyed needles and three grooved needles (including one needle regrooved after the original groove broke during use). Drilling the needle eye took between 10 and 13 minutes, whereas inscribing a groove required only a minute or two. We had no failures during this stage. Our experiments in needle fabrication allow us to assess the hypothesis that Unangan seamstresses adopted grooved needles to reduce their manufacturing costs. Although we found drilling the eye took a little more time than inscribing a groove, this difference was negligible when considering the total time spent making the needles (Figure 10.3). Our

160

BRIAN W. HOFFMAN

Figure 10.3 False Pass needle manufacturing experiments; total time per experiment

production costs for the three eyed needles averaged 53 minutes (range 39 to 63 minutes), whereas production costs for the two grooved needles were essentially the same, averaging 52 minutes (range 43 to 61 minutes). The vast majority of the production costs (85.2%) involved obtaining the blank and grinding it into a needle shape. These production costs, which would have been even higher had we made full-sized needles, remain the same whether making eyed or grooved needles. Switching to grooved needles would not have significantly reduced the overall time involved in needle manufacture. Risk of failure also does not appear to differ significantly between needle styles. Most production accidents occurred when attempting to remove a needle blank from the bone core. Archaeologically recovered production debris suggests that Unangan seamstresses likewise experienced a relatively high rate of failure at this step (Hoffman 2002). We experienced no failures during the final production step, although our experiments were limited in number. Eyed needles may look impressive, but replicating them is not a particularly difficult task. The students were able to successfully drill delicate needle eyes despite the fact that none had ever worked with bone or stone tools before. Overall, our

BROKEN EYES AND SIMPLE GROOVES

161

results are inconsistent with the hypothesis that grooved needles were adopted to reduce manufacturing costs or because eyed needles were more difficult to make. NEEDLE DURABILITY RESULTS The second phase of our experiments was to use our needles and evaluate whether either style was more durable. We had six experiments counting the regrooved needle (three eyed and three grooved). Each experiment involved stitching heavy cotton fabric with polyester thread for twenty minutes or until the needle broke, whichever came first. One eyed and one grooved needle broke before the twenty minutes were up, for a success rate of 66.7 percent for each needle style. Butler (1997) obtained similar results in her experiments sewing animal skins (deer and rabbit) using sinew thread. These experiments, although limited in number, suggest eyed and grooved needles do not differ significantly in durability. We did find it possible to regroove a broken needle, consistent with McCartney’s hypothesis, but our experiments did not fully explore the potential differences in needle repair or refurbishing. NEEDLE DIAMETER The final hypothesis evaluated in this chapter is that Unangan seamstresses switched to grooved needles because that style could be made thinner than eyed needles. The needles made by the False Pass students do not attempt to evaluate this hypothesis, since we intentionally made relatively thick needles of both styles. Butler’s (1997) replica experiments provide a better test of this hypothesis. Her eyed needles ranged between 0.75 and 4.00 mm, whereas her grooved needles ranged between 0.50 and 1.50 mm (Butler 1997:13). Butler’s (1997) results closely match the diameters obtained from archaeologically recovered needles (Hoffman 2000). For this analysis I compared a sample of twenty-six eyed needles recovered from a large midden site (49 RAT 31) dated between 600 B.C. and A.D. 1000 on Amchitka Island in the western Aleutians (Desautels et al. 1970) and seventy-six grooved needles and needle fragments recovered from house interior contexts at Agayadan Village (Hoffman 2002). The Amchitka eyed needles were substantially thicker, averaging 1.67 mm in diameter compared with an average diameter of 0.98 mm for Agayadans’ grooved needles (Table 10.2). Other samples of eyed needles closely match the Amchitka needle measurements (Hoffman 2000). Both the Amchitka Island and Agayadan Village needle assemblages included relatively robust needles (>1.5 mm) suitable for heavy-duty

162

BRIAN W. HOFFMAN

Table 10.2—Comparison of eyed and grooved sewing needle diameters (in millimeters)

EYED NEEDLES 49 RAT 31 (n=26) Butler 1997 (n=5) GROOVED NEEDLES Agayadan (n=76) Butler 1997 (n=3)

Minimum

Maximum

Mean

1.00 0.75

2.28 4.00

1.67 —

Archaeological context Experimental replicas

0.52 0.50

2.22 1.50

0.98 —

Archaeological context Experimental replicas

sewing tasks. The difference between the needle assemblages is that the Agayadan Village occupants also used a large number of very fine needles (

Many Faces of Gender: Roles and Relationships Through Time in Indigenous Northern (Boreal) Communities - PDF Free Download (2024)

FAQs

What are the gender roles in indigenous communities? ›

What are the third genders indigenous people? ›

References