Formative-era exchange, interaction, and social organization in west-central Colorado

Material Information

Formative-era exchange, interaction, and social organization in west-central Colorado a ceramic study utilizing instrumental neuron activation analysis
Bedingfield, Kenneth Brian
Publication Date:
Physical Description:
233 leaves : ; 28 cm


Subjects / Keywords:
Nuclear activation analysis -- Colorado -- Western Slope ( lcsh )
Indians of North America -- Social conditions -- Colorado -- Western Slope ( lcsh )
Indian pottery -- Colorado -- Western Slope ( lcsh )
Antiquities ( fast )
Indian pottery ( fast )
Indians of North America -- Social conditions ( fast )
Nuclear activation analysis ( fast )
Antiquities -- Western Slope (Colo.) ( lcsh )
Colorado -- Western Slope ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 208-233).
General Note:
Department of Anthropology
Statement of Responsibility:
by Kenneth Brian Bedingfield.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
436223328 ( OCLC )
LD1193.L43 2009m B42 ( lcc )

Full Text
Kenneth Brian Bedingfield
B.A., Metropolitan State College of Denver, 1997
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts

This thesis for the Master of Arts
degree by
Kenneth Brian Bedingfield
has been approved
Tammy Stone

iz A^/'z-poS

Bedingfield, Kenneth, B. (MA, Anthropology)
Formative-era Exchange, Interaction, and Social Structure in West-
Central Colorado: A Ceramic Study Utilizing Instrumental Neutron
Activation Analysis
Thesis directed by Professor Tammy Stone
This thesis examines the use of ceramic materials among populations that
occupied areas of west-central Colorado near the Uncompahgre Plateau
during the Formative Era (ca. A.D. 1 A.D. 1300) in an effort to better
understand the social organization of these groups and the ways in which
they interacted with surrounding populations. Questions of how, why,
and from where these populations obtained and utilized ceramic goods
are particularly interesting given the geographically intermediary location
of the study area between relatively well-understood Formative-era
cultural units (Fremont and Anasazi). This work provides background
information regarding theoretical, cultural, and environmental
considerations pertaining to the area in question, followed by an
explanation of the methodological approach utilized in data analysis. As
a means of maximizing the explanatory capacity of the data, a multi-
dimensional approach to ceramic analysis has been employed, utilizing
conventional macroscopic and technologically-oriented microscopic
techniques of examination of ceramic artifacts in combination with
instrumental neutron activation analysis of pottery samples and raw
materials. Results indicating that ceramic material analyzed was
imported from locations south of the study area are inferentially linked to
aspects of social organization among the occupants of the sites from
which pottery was collected.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.

Tammy Stone

To Krista, my favorite archaeologist, and to Logan, one of the most
extraordinary individuals I have had the pleasure to know.

The completion of this thesis would not have been possible
without the support and encouragement of numerous colleagues, friends,
and family members. My thesis advisor, Tammy Stone, merits special
thanks for her undying patience in guiding an unconventional student
through the conventions of academia. Thanks also to fellow committee
members Chris Beekman and Jon Kent for sharing their expertise, which
allowed me to substantially improve upon the thesis in many ways.
Special thanks also go to Todd McMahon of the Colorado Historical
Society for initially piquing my interest in the problems treated in this
thesis and his consistent support throughout the project. The bulk of the
ceramic material considered here was made available by Mark Stiger of
Western State College. William Lucius graciously donated his time,
knowledge, and the research facilities of the Institute for Archaeological
Ceramic Research (LACR) to the project. I am also indebted to Sarah
Jennings and Krista Bedingfield who graciously volunteered to assist in
the field. Financial support for the project was provided through the
Ward Weakly Scholarship granted through the Colorado Council of
Professional Archaeologists (CCPA) and the Alice Hamilton Scholarship
granted through the Colorado Archaeological Society (CAS). Michael
Glascock and Leslie Cecil of the University of Missouri Research Reactor
Laboratory (MURR) were able to provide analysis at subsidized rates
made possible by a grant by the National Science Foundation (NSF).
Finally, I must thank the members of my immediate and extended family
(hi, mom!), particularly my father for consistently demonstrating the
importance of dedication and tenacity; their continued support through a
sometimes trying endeavor has been invaluable. Any errors of inclusion
or omission, inconsistencies, or misconceptions found herein are solely
the responsibility of the author.

Tables ..........................................................xi
1. INTRODUCTION...................................................1
The Study Area.............................................3
Methodological Approach....................................7
Organization of the Study..................................9
2. THEORY........................................................14
Historical Development of Exchange
and Interaction Theory....................................15
Contemporary Approaches to Exchange
and Interaction...........................................22
Ecological Models.......................................23
Sociopolitical Models...................................26
The Potential of Interaction Studies in the
Uncompahgre Plateau Area..................................31
Modeling the Study Area.................................31
Interregional Relationships.............................41

HISTORY: A.D. 1 TO 1300...................................48
The Northern San Juan Anasazi.........................50
Basketmaker II: 1,000 B.C. to A.D. 500...............52
Basketmaker III: A.D. 500 to750......................55
Pueblo I: A.D. 750 to 900............................57
Pueblo II: A.D. 900 to 1150..........................62
Pueblo III: A.D. 1150 to 1300........................67
The San Rafael Fremont................................75
The Gateway Tradition.................................85
4. ENVIRONMENTAL CONTEXT....................................96
Soils.............................................. 106
Flora and Fauna Associations.........................Ill
5. METHODOLOGY.............................................121
The Data Base........................................122

Pottery and Raw Material Analysis........................128
Typological Classification.............................130
Ceramic Characterization Studies.......................132
Technological Characterization.........................136
Compositional Characterization: Instrumental
Neutron Activation Analysis............................139
Data Interpretation......................................147
6. RESULTS....................................................150
Recorded Sites...........................................151
Technological Characterization...........................162
Instrumental Neutron Activation Analysis.................167
Summary of Results.......................................177

7. DISCUSSION AND CONCLUDING REMARKS......................182
The Weimer Ranch Collection........................184
Exchange and Spatial Distribution..................187
Change through Time.................................188
Exchange and Social Organization....................191
Limitations and Perceived Shortcomings..............195
Suggestions for Further Work........................196
AND TABULAR DATA......................................200
REFERENCES CITED...............................................208

1.1 Traditionally Ascribed Formative Era
Cultural Units in the Four Corners Area............................4
2.1 General Locations of Fremont Variants................................35
2.2 General Location of Gateway Tradition................................38
4.1 Study Area and Surrounding Physiography
of western Colorado at 1:200,000...................................98
4.2 Profile of Commonly Exposed Geologic Formations along the
Dolores River Canyon and Elsewhere in Study Area...................104
4.3 Biotic Communities and Representative Plant Associations in the
Project Area with Approximate Elevation of Transition Zones
Provided in Meters above Mean Sea Level............................113
5.1 Pottery Location by Vessel Number...................................123
5.2 Raw Material Specimen Locations.....................................124
5.3 Process of Neutron Capture by Target Nucleus and
Subsequent Emission of Gamma Rays..................................142
5.4 Geographic Distribution of Specimens
Submitted for INAA.................................................144
6.1 Ceramic Specimens of Northern San Juan Area Compositional Groups
with Affiliated Specimens from this Study Projected onto the First and
Third Principal Components....................................172
6.2 Ceramic Specimens of Northern San Juan Area Compositional Groups
with Affiliated Specimen from this Study for the Lowry Group
Projected onto the First and Third Principal Components............173
6.3 Plot of Rubidium and Hafnium Base-10 Logged Concentrations
Showing the Separation of Pottery Specimens Associated with
Previously Unidentified Compositional Groups..................175
6.4 Temporal Distribution of Vessels by Percentage......................181

3.1 Important Gateway Tradition sites......................................90
4.1 Climatic data from Gateway, Colorado for 1948 2005..................111
5.1 Ceramic Specimens Submitted for INAA..................................143
6.1 Individual Vessels and Associated Attributes as
Identified through Technological Analysis............................164
6.2 Northern San Juan Area Compositional
Group Affiliation of Ceramic Specimens...............................171
6.3 Affiliation of Ceramic Specimens with Previously
Unidentified Compositional Groups....................................174
A. 1 Element Concentration (PPM)
from Short Irradiation...............................................204
A.2 Element Concentration (PPM)
from Long Irradiation................................................205
A.3 Eigenvalues and Percentage of Variation
Explained by Principal Components....................................207

The presence of prehistoric exchange networks has long been recognized
as an indicator of social differentiation amongst the parties involved (Plog 1989).
This appreciation of the embedded quality of exchange activity within broader
forms of social relationships has permitted archaeologists to utilize studies of
exchange and interaction as a means of identifying forms of prehistoric social
organization and explaining social and economic change through time (Earle
1982). Existing models of prehistoric exchange often tend to stress the primacy of
either ecological or sociopolitical factors in affecting change, as influenced by the
researchers theoretical orientation. A widely accepted and distinct interaction
paradigm (Schortman and Urban 1992:3) that emphasizes social aspects of
exchange has yet to emerge within the discipline, yet some general points of
consensus have been established. Among these is the recognition that exchange is
driven by social as well as economic necessity, and furthermore, that the
mechanisms of exchange will be contingent on localized processes that encompass
environmental, social, and historical processes (Hodder 1982, 1991; Saitta 2000).
Existing models of interaction can be particularly problematic when
applied to non-stratified communities (Cobb 1993). In the absence of direct
linkages between forms of exchange and the social organization of trading

partners, it is essential that approaches to the study of prehistoric exchange,
particularly among pre-state, middle-range societies that may exhibit elements of
both egalitarianism and hierarchal structure, examine interaction within localized
contexts as processes which guide and influence interaction between regionally
situated populations. Such interaction can, in turn, be assessed in terms of its
effect on the historical trajectory of the communities in question. This requires an
attempt to recognize the degree to which exchange strategies are driven by
individual and/or group interests and how the interplay of these potentially
divergent or convergent interests may be manifested archaeologically.
This thesis examines the use and distribution of ceramic materials among
populations that occupied areas of west-central Colorado near the Uncompahgre
Plateau during the late Formative era (ca. A.D. 900 A.D. 1300) in an effort to
better understand the nature of sociocultural organization of these groups and the
ways in which they interacted with surrounding populations. As applied to the
study area, the Formative era is considered here as a time when horticultural
adaptations were developing throughout the southwest and no further assumptions
regarding the subsistence practices of the people living at this time are employed
(Reed and Metcalf 1999:5). This area is particularly well suited for an inquiry into
interaction due to its geographically intermediate location between sedentary
populations to the north and south (i.e. Fremont and Anasazi populations
respectively). As a potential zone of economic, social, and cultural interaction,

this area appears exceptionally conducive to inquiry into localized exchange
activity, which can in turn be considered in a regional perspective. In
consideration of the currently limited understanding of prehistoric social
organization and interaction in this area, questions of how, why, and from where
these populations obtained and utilized ceramic goods are particularly interesting.
The Study Area
The study area corresponds roughly with the spatial extent occupied by the
Gateway Tradition culture unit within Colorado (Figure 1.1) as postulated by Alan
D. Reed (1997). Reed proposed the Gateway construct to address the anomalous
nature of archaeological sites in the area, which generally do not exhibit material
manifestations that conform to established concepts of Anasazi or Fremont
affiliation. Though the study is not designed to directly test the Gateway
hypothesis per se, it considers evidence that will facilitate further examination,
and potentially explanation, of some of the anomalies observed by archaeologists
who have worked in the area.
Social dynamics among Formative-era inhabitants of the Uncompahgre
Plateau and surrounding areas of west-central Colorado are poorly understood at
the present time. The areas spatially intermediate position between relatively
well known Northern Anasazi groups to the south and Fremont groups to the north
and west, combined with a general lack of intensive archaeological inquiry has

frustrated researchers attempts to understand this area within a broader regional
context. Much of the discussion regarding the nature of the Formative-era
occupation here in the past decade (e.g. McMahon 2000; Reed 1997; Reed and
Metcalf 1999) has been productive yet largely inconclusive in terms of providing
concrete insight regarding the cultural affiliation of the areas inhabitants or the
forms of social organization that were in place.
Figure 1.1. Traditionally ascribed Formative Era cultural units in the Four Corners area
(modified from Cassells 1997:144).
A reassessment of existing collections is required to enhance our understanding of
the archaeological record in this region and to provide direction for future research
capable of illuminating a host of archaeological questions regarding problems

such as cultural chronology, technology, settlement patterns, extraregional
relationships, and social organization.
Due to a paucity of archaeological evidence indicative of ceramic
manufacturing here, it has largely been assumed that ceramic artifacts located in
the area are representative of exchange activity between inhabitants of the
Uncompahgre Plateau and Puebloan groups to the south and/or Fremont groups to
the north and west (Crane 1977; McMahon 2000; Reed and Metcalf 1999). While
the relative scarcity of ceramic artifacts on and around the Uncompahgre Plateau
appears to generally support this assumption, little has been offered in the form of
positive, empirical evidence to bolster this claim. The reliance on negative
evidence (i.e. the apparent absence of artifacts and/or features indicative of
ceramic manufacturing) has been necessarily cited, yet is particularly fragile
considering the lack of controlled, problem-oriented excavations that have been
conducted in the area (Reed and Metcalf 1999:110).
This thesis specifically addresses the question of where pottery utilized by
the Formative-era inhabitants of the Uncompahgre Plateau area originated.
Assessing the provenance of these artifacts will provide a basis for determining
whether they are imported or locally produced and a foundation for addressing
subsequent questions within a regional context. If imported, it is of interest
whether the artifacts represent exchange activity, migration or a combination of
the two. Compositional analysis will further permit an assessment of the nature of

exchange networks in place by indicating whether the ceramic assemblages
represent multiple trade partners or relatively restricted avenues of interaction.
Local production could also possibly represent migration, but an assessment of
technological and stylistic elements considered in context with other
archaeological manifestations may then address whether the artifacts represent a
unique regional ceramic tradition or a spatial extension of stylistic and
technological production methods previously recognized elsewhere in the region.
Through a multidimensional, comparative approach to ceramic analysis
considered within the archaeological context of the study area and surrounding
regions, the project described in this thesis utilizes material from the
archaeological record to examine the nature of social relationships among groups
in the Uncompahgre Plateau area and to establish a basis for inference on how
these groups may have interacted with neighboring regions in the context of
exchange-driven interaction. In addition to the goal of illuminating the nature of
Late Formative-era social interaction, it is believed that this study will enhance the
quality of baseline data available to archaeologists studying Late Formative-era
occupation of the Uncompahgre Plateau and facilitate the incorporation of
archaeological studies here into a broader regional framework.

Methodological Approach
In exploring the questions of how and with whom occupants of this area
obtained and exchanged ceramic resources, this study examines stylistic and
compositional attributes of ceramic artifacts and the chemical composition of
locally available raw materials as a means of identifying sources of production and
patterns of distribution that may be evident in the archaeological record. This
entails stylistic, technological, and compositional analysis of pottery from the
study area through optical inspection of ceramic artifacts, and technological and
chemical ceramic characterization techniques. Data derived are subsequently
compared with data from previously conducted studies as a means of assessing
contemporaneity and potential sources of origination for the ceramic material
examined. Though the project is multidimensional in construct, the emphasis of
the study is placed on compositional analysis of pottery and raw material samples
undertaken through instrumental neutron activation analysis (INAA) in an effort to
examine factors pertaining directly to pottery acquisition and distribution within
the study area. This method, combined with other techniques that establish
cultural context, offers the potential to speak to the possible distinction between
the social transmissions of ideas, as expressed through adoption of manufacturing
techniques and stylistic conventions, and the physical exchange of material goods.
As a method of compositional characterization of ceramic material, INAA
has emerged as the method of choice over recent decades. The technique permits

characterization of archaeological specimens through association with chemical
signatures obtained through measuring the emission of gamma radiation from
activated samples. While essentially destructive in practice, the process can be
conducted with samples as small as two grams with 3-4 cm2 of surface area
(Glascock 2003; Harbottle 1982), leaving the bulk of material under consideration
available for future research efforts. While cost constraints have been prohibitive
to wide-scale application of INAA in the past, funding of Archaeometry
Laboratory at the University of Missouri Research Reactor (MURR) by the
National Science Foundation (NSF) has made the process relatively affordable
through subsidized rates available to academic researchers working through U.S.
institutions (Glascock 2005).
In addition to providing a highly sensitive, precise, and accurate method of
chemical analysis, INAA permits comparative analysis through the use of a
rapidly expanding data base of archaeological and raw material sources. Notably,
researchers have recently analyzed existing collections, including ceramic
assemblages, from sites in the Uncompahgre Plateau area utilizing INAA analysis
of ceramic specimens (Greubel et al 2006). Complementary studies such as these
are considered in the discussion presented here, and the combined contribution to
the collective INAA database will certainly prove valuable to future research

To augment and further contextualize the analysis and discussion of the
pottery considered in the study, revisitation of the sites from which the pottery was
collected was recognized as an essential step. Since many of these sites were
recorded over thirty years ago and have not been the subject of ensuing research,
on-the-ground investigation was necessary to bring existing documentation up to
modem standards. This process allowed verification and/or correction of spatial
data, the assessment of present site condition, and opportunity to comment on the
sites potential for further research. Site visitation also provided an opportunity to
assess potential physical routes of exchange and the availability of local raw
material sources, permitting collection of raw material samples for inclusion in
chemical analysis procedures. Site visits further provided a chance to observe and
record any additional surface materials that had been exposed since the initial
recordation of the sites resulting in the collection of additional ceramics for
inclusion in the study and the observation of previously unrecorded diagnostic
Organization of the Study
Subsequent chapters contain treatments of theoretical developments in
interaction studies and how they inform this project, environmental and cultural
contexts for the project area, explanations of the methods and techniques utilized,

results obtained through analyses, and finally data interpretation and conclusions
derived from results of analytical procedures conducted.
Chapter 2 contains a summary of archaeological theory associated with
studies of social interaction. First, an historical overview of theoretical
developments pertaining to studies of exchange and interaction that have taken
place within archaeology as a discipline is presented along with an assessment of
how such developments have impacted previous work in and around the study
area. This is followed with an overview of contemporary theoretical approaches
to studies of exchange and interaction and a brief assessment of their applicability
to the proposed project. The chapter concludes with a discussion of the theoretical
standpoint adopted for this project and how this directs the study in the formation
of the questions addressed.
Chapter 3 contains a summary of what is currently known regarding the
prehistoric occupation of the study area and immediately surrounding regions in
an effort to provide a broader and deeper temporal and cultural context for
consideration of the study. Since the study specifically addresses questions of
exchange and interaction, the discussion of Formative-era culture history includes
treatments of surrounding regions, particularly those that were occupied by the
Northern San Juan Anasazi and the San Rafael Fremont.
The content of Chapter 4 addresses the environmental context of the study
area. This includes physiographic features associated with Uncompahgre Plateau

and surrounding areas as well as an overview of the areas geology as it relates to
the availability of potential clay and temper resources. The chapter also contains
descriptions of climatic variability, and addresses availability of floral and faunal
resources. The chapter concludes with a section describing the areas
Chapter 5 concerns the methodology employed in this study, and presents
summaries of the techniques utilized in both project-specific terms and in more
general terms that consider the broader role of analytical methods in archaeology
as a discipline. A description of the data base used for the study is followed by a
summary of methods employed in field investigations of the specific sites from
which the ceramic specimens analyzed in the study were obtained. The chapter
continues with a description of the method employed for the analytical portion of
the project. This includes descriptions of techniques used for the optical
assessment and typological classification of the ceramic collection followed by a
general discussion of the techniques used for initial ceramic characterization
efforts, including the criteria employed in the subsequent selection of samples
submitted for INAA. A discussion of the neutron activation analysis method, its
applications in archaeological research, and the specific details of the use of the
method in this project is then presented. Chapter 5 concludes with a discussion of
the techniques of statistical analysis used in the study for data reduction,
presentation, and interpretation.

The results of the study are presented in Chapter 6. This begins with a
description of the sites investigated during the field investigation phase of the
project, which entailed the reevaluation of eight previously recorded sites and the
collection of 23 previously undocumented ceramic specimens. Descriptions of the
results of field work are followed by a summary of results derived through both
technological and chemical characterization analyses of the ceramic data base.
Technological characterization indicated that nearly all of the ceramic specimens
originated in the Northern San Juan region. INAA partially confirms these results,
indicating that roughly one third of the specimens were produced in previously
identified manufacturing tracts, also located in the Northern San Juan area.
The chapter also contains a brief synthesis of the variant lines of evidence
pursued as described in the previous chapters, and the studys results in evaluation
of existing models of exchange are presented. Here data are reevaluated in light of
what may be discerned regarding the movement of pottery across space and
through time in light of the results of analysis as described in Chapter 5. In this
chapter, the studys results are related to the question of pottery acquisition and
distribution within the study area and a brief discussion of how perceived patterns
may speak to social organization in the study area during the late Formative era is
The conclusion of the study is presented in Chapter 7. This chapter
features an expanded discussion of the projects results as they relate to studies of

exchange in general and how the completed study may be built upon and
incorporated into larger studies of interaction in the both the study area and the
within the larger regional contexts. Inferences that may be drawn between the
results of the analysis and social organization in the study area are explored.
Perceived shortcomings of the project are critiqued in this chapter, offering
general suggestions for improving similar studies that may be conducted in the
future. The chapter will conclude with specific suggestions for future
archaeological research that may further illuminate the problems explored in the

In large part, efforts to model social interaction in the area considered
herein have been centered on a series of regionally based categories that have been
interpreted as potentially representative of groups who maintained distinct cultural
identities. To better understand the social dynamics behind the archaeological
expressions on which existing typologies are constructed, a framework that
considers the exchange of commodities, wealth, and information across culturally
created prehistoric boundaries within specific historical contexts is essential
(Baugh and Ericson 1993:9; Plog 1995:196-202; Saitta 2000; Schortman 1989:59-
60). The presence of ceramic artifacts in the Uncompahgre Plateau area that
possess similar stylistic attributes to those produced in surrounding regions
suggests that a study of interregional interaction may be key to developing a
greater understanding of the social relations that were in place here. This thesis
presents an effort to derive inference regarding social relations through an analysis
of data associated with exchange activity.
In this chapter, a background of theoretical developments that have
informed the archaeological study of exchange and interaction and resultant
hypotheses addressing social organization is presented. The historical
development of the archaeological theory is considered here, followed by a brief

review of contemporary approaches. The ways in which theoretical developments
have impacted previously conducted research in the study area is also presented
here, providing background information on how problems concerning social
organization have been framed in the past. The potential for gaining further
insight regarding the nature of social organization within the area through the
study of interregional interaction is then discussed. The chapter concludes with a
series of questions and associated implications that can be further through the
analysis conducted for this thesis.
Historical Development of Exchange
and Interaction Theory
The exploration of linkages between interaction on a regional scale and
social development and change has occupied anthropological inquiry since the
nineteenth century (Baugh and Ericson 1993:4; Schortman and Urban 1992:5;
Trigger 2006:211-223). Rooted in the evolutionary paradigm of the mid
nineteenth century, early anthropologists saw culture as progressively developing
through sequential stages that increasingly resembled that of the then-current state
of Western European culture (Schortman and Urban 1992:5; Trigger 2006:211).
This cultural growth was considered correlate with the physical development of
the human brain, and thus, the observations on the cultural stages of human
development as related to the developing physical capacity of those who occupy

them could be generalized and applied to the species as a whole (Schortman and
Urban 1992:5; Trigger 2006:211-216). The goal of developing cross-cultural
generalizations that could be linked to sequential, biologically developed traits
provided little incentive for the examination of intersocietal interaction since the
consequences of such action could not exceed the level of mental development
within a given population to adapt advanced cultural traits. While the diffusion of
ideas and behavioral patterns was recognized as existing between populations with
sufficient evolutionary capacity, the impacts of such exchange were considered to
be locally expressed and of little importance in the understanding of broad cultural
development (Schortman and Urban 1992:5-6; Trigger 2006:217-218).
Challenging these evolutionary precepts in the early twentieth century was
the emergent Diffusion School, whose thought would become increasingly
dominant within anthropology in the following decades (Baugh and Ericson
1993:4-9; Schortman and Urban 1992:6-8; Trigger 2006:217-223). Diffusionists
saw culture change emanating from the geographical distribution of cultural traits
and embraced diverse theoretical perspectives (Baugh and Ericson 1993:4-5;
Schortman and Urban 1992:6-7; Trigger 2006:217-223). Encompassed within this
broad diffusionist perspective, was the American culture-historical approach to
archaeological research with an emphasis on the temporal and geographical
ordering and classification of culture areas (Baugh and Ericson 1993:5).
Methodological developments in the discipline during this period including,

seriation, dendrochronolgy, and improved techniques for analyzing stratigraphy -
contributed to archaeologists ability to refine chronological sequences within sites
and regions, but the emphasis on classification based on material similarity within
the culture-historical framework was pursued at the expense of recognizing and
analyzing variation within the archaeological record (Baugh and Ericson 1993:5-
6). Despite theoretical and methodological advances well beyond the position of
their evolutionist predecessors and several efforts to link stylistic material
elements with functional correlates of human behavior (see Haury 1985;
Schortman and Urban 1992; Trigger 2006), the largely unexamined processes of
cultural diffusion were generally seen by many as a stochastic, almost cancerous,
spread of technological characteristics between groups, contributing to the
diversity of the archaeological record through localized historical trajectories. This
seemingly unpredictable process of cultural borrowing fostered a belief among
diffusionists that cross-cultural regularities either did not exist or were so
thoroughly buried by historical accretions that they were virtually unrecognizable
(Schortman and Urban 1992:7). The diffusionist paradigm had essentially
abandoned the concern with general evolutionary concepts that characterized early
approaches to archaeological theory building.
Dissatisfaction with this position would come to the fore during the 1940s
and 1950s, culminating in a widespread resurgence of interest in evolutionary
processes, as influenced and illustrated through the development of cultural

ecology by the social anthropologist Julian Steward (1955). This approach
emphasized the functional linkages between cultural process and physical
environment and tended to focus on core elements of spatially recognized
cultural groups as those which contained the potential for contributing to
adaptively advantageous patterns of behavior (Schortman and Urban 1992:9;
Steward 1955:89). Within this scheme, aspects of behavior, including those
involving trade and social interaction, were relegated a secondary role to those
driven by ecological determinants.
This functionalist approach would be adopted and refined by proponents of
the New Archaeology in the 1960s (Schortman and Urban 1992:8; Trigger
1984:280, 2006:396) and continues to be highly influential today. Focus had
again shifted toward efforts to establish broad, cross-cultural generalizations in an
attempt to recognize broadly applicable laws concerning processes of cultural
development pertaining to total cultural systems (Binford 1962:225) as they
strive to achieve or maintain homeostatic balance within the physical environment.
This systemic perspective, combined with an increase in regional settlement
pattern studies precipitated by Willeys (1953, 1974) work in the Viru Valley,
Peru, led many archaeologists to approach interaction as an internal process
operating within component subsystems of cultural wholes (Baugh and Ericson
1993:9; Trigger 1984:280). Within this framework, studies of interaction have
often employed quantitative analysis combined with ethnographic analogy in

developing interaction models to assess how elements of exotic resource
procurement, utilization, and exchange function within adaptive contexts (Baugh
and Ericson 1993:8-9). Despite the contributions of such studies to our
understanding of material exchange in prehistory, the processual emphasis on
human/environment interaction has heavily favored approaches focusing on
broadly applicable functional aspects of subsistence economy and demography
over sociocultural dynamics and interaction (Baugh and Ericson 1993:9) as
demonstrated by the prominence of theoretical concepts addressing adaptive
processes (e.g. niche, ecosystem, symbiosis, etc.) over those that might illuminate
developments occurring within the context of social interaction (Schortman and
Urban 1992:10-11). This shortcoming has led Schortman and Urban (1992:3) to
advocate the development of a distinct interaction paradigm that is focused on
the domain of sociopolitical change processes, and to conclude:
We are still working with views of interregional interaction that were found wanting and
in need of serious revision within the diffusion paradigm itself 50 or more years ago...If
interaction research is to provide insights into cross-cultural regularities in human
behavior, then a break must be made with these outmoded views. We have to recognize
that, whatever its virtues, a theoretical framework dominated by cultural ecology has no
room for interregional interaction. Interaction studies must...proceed to the development
of a paradigm appropriate to the study of interaction questions. (1992:11)
While a coherent paradigm of interaction remains elusive, theoretical insights
emerging from postprocessual inquiry over the past two decades have bolstered
the potential for such development.
Emphasizing the role of history and agency in cultural development,
postprocessual archaeologists have tended to view culture as unfolding through

fluid and dynamic processes that are contingent on localized contexts (Hodder
1982a: 11-13, 1982b:207). Postprocessualists have successfully incorporated this
dynamic and diachronic perspective into the inquiry of archaeological exchange
systems. Some models generated through symbolic/structural approaches have,
however, been criticized (Cobb 1993:60-65; Saitta 1994:204-206, 2000:153-155)
as glossing over historical and material variation within prehistoric exchange
systems by overemphasizing the role of symbolic value and/or the paramount
position afforded to elite actors; two such models, including prestige goods
economy models and peer polity interaction models are discussed further below.
Importantly though, such approaches have clearly demonstrated that, in terms of
social interaction, as phrased by Cobb (1993:65), external relations do not occur
in a disembodied vacuum, but rather shape, and are shaped by, internal relations.
To address the intricacies of these nested relationships and the tensions
existing between and within them, approaches grounded in political economy have
attempted to integrate aspects of structure accentuated by processualists with those
of agency emphasized in symbolic perspectives. Rather than finding the locus of
cultural change in external perturbation of bounded systems or internally among
the interaction of individuals, political economy seeks to demonstrate, that a
study of agency cannot be separated from a study of structure: that agency is a
manipulation of an existing structure, a structure that is external to the
individual...and appears to that agent as a synchronic construct, as something to

be drawn upon (Johnson 1989:206). From this perspective, exploration of
negotiated relationships between diverse social agents in concrete historical
circumstances (Saitta 2000:156) becomes crucial to informing theoretical models
of interaction and exchange.
Broad trends in the development of anthropological thought have
obviously had their corresponding influences on how archaeologists approach
studies of exchange and interaction. Ecologically oriented studies have revealed
much about how material goods are transported across landscapes, and they have
contributed to an extensive database applicable to exchange studies in many parts
of the world. Processual approaches have largely failed, however, to address the
underlying mechanisms of exchange and interaction that is, people and how they
socially construct the systems in which they function. Since human populations
are rarely self-sufficient in even the most basic requirements (Spielman 1991a:3)
and are necessarily interactive at all levels of complexity, understanding of the
relationships between agents within and between the boundaries normally
recognized by archaeologists, at multiple levels of geographic and temporal scale,
must be pursued to provide a holistic understanding of cultural development and
change. This entails the application of interaction studies that incorporate multiple
variables, recognizing potential limitations imposed by ecology yet also realizing
the inherent potential for plasticity of human behavior and corresponding ability to
manipulate or influence socioeconomic conditions. In cases, this may require

extending the analysis beyond observables (Bender 1985:50), to illuminate the
underlying mechanisms and processes of the archaeological record.
Contemporary approaches to exchange and interaction have been proposed by
researchers embracing diverse theoretical perspectives in attempts to elucidate the
nature of interaction in terms of the empirically observable and beyond. Before
turning to the particular problems of settlement in the Uncompahgre Plateau area,
it will be useful to briefly review some of the specific theoretical positions behind
contemporary efforts to model exchange in the archaeological past.
Contemporary Approaches to
Exchange and Interaction
Despite the various perspectives of archaeologists examining exchange and
interaction in recent years, a broad consensus has been established regarding at
least three basic factors influencing such inquiry in the United States. As
summarized by Saitta (2000:151-152), these basic points include the following:
(1) Exchange is driven as much by social as by economic necessity...That is, exchange
cements political alliances between interacting groups and provides goods that can be
used on a local level to create and signal important status distinctions, as well as meet the
requirements of group ritual activities. (2) Exchange is variable across the continent in
terms of the kinds of goods exchanged, the social context of exchange, and in the intensity
of exchange. Exchange systems also wax and wane in scale and complexity depending on
the specific circumstances of time and place. (3) Because of the social nature of exchange
and its responsiveness to historical contingencies, particular forms (e.g. reciprocity,
redistribution) or scales (e.g. bounded, extended) of exchange do not neatly correlate with
other social characteristics or particular levels of social complexity.. .any society can
display a number of different forms or scales of exchange activity depending on
environmental, social, and historical circumstances.

Recent approaches to interaction and exchange have demonstrated varying degrees
of emphasis on these factors of social or economic necessity. Few have
successfully addressed the resultant, complex potential for variability within and
between exchange strategies. Below, some of the attempts to model exchange and
interaction are briefly reviewed. While there is sometimes considerable
conceptual overlap between various contemporary models and most researchers
would not advocate purely ecologically deterministic arguments, for heuristic
purposes, recent approaches can be roughly divided by the relative primacy
afforded to ecological versus sociopolitical factors.
Ecological Models
Spielman (1986; 1991a) identifies two complementary and potentially
interactive ecological models of exchange and interaction: (1) Buffering models in
which economic security is achieved through exchange for participating parties,
and (2) mutualism models where exchange is predicated upon the differential
distribution of resources (1991a:4). Different environmental circumstances -
environmental stochasticity or differential distribution of natural resources are
seen as influencing which approach will be selected for in the development of
intersocietal exchange practices.
Buffering exchange is seen as a risk aversion strategy in which participants
attempt to negate the effects of environmental stochasticity by establishing

reciprocal obligations among neighboring groups through the exchange of craft
and ritual items (Spielman 1986:281-282; 1991a:4). During times of
environmental stress, members of the affected population are able to supplement
subsistence goods through their reciprocal partners occupying adjacent niches, or
in cases, through the direct exchange of durable goods for foodstuffs (Dean et al.
1985:539-540; Spielman 1986:282; 199la:4). Dean and colleagues (1985) have
utilized a variation of this model to examine exchange on the southern Colorado
Plateau. Employing the concept of trade alliances extrapolated from the stylistic
variation in ceramic distribution, they consider mobility and settlement patterns,
subsistence mix, ceremonialism, agricultural intensification, and territoriality as
archaeological indicators of exchange strategies in their attempt to identify
relationships between environmental variability and population on the one hand
and cultural and behavioral variability on the other (Dean et al. 1985:549). Lintz
(1991:106) sees interactive environmental buffering strategies employed by
protohistoric Plains and Pueblo groups shifting to a system of mutualism as the
requirement of economic change necessitated major readjustments in the structure
of the Plains societies.
The concept of mutualism sees the exchange of goods or services as
benefiting all active participants by establishing a higher carrying capacity than
would be attainable in the absence of such interaction (Spielman 1986:285-286,
1991a:5-6) This strategy is proposed to be employed when the majority of

individuals within a population are able to participate in interactive exchange and
when the cost of exchange goods is relatively low in comparison to the value of
benefits received (Spielman 1986:287, 1991a:5-6). This approach focuses on the
interaction between groups occupying spatially segregated
environmental/economic niches involved in interdependent, mutually beneficial
exchange practices. The success of mutualistic interaction would be greatest when
resources were abundant and their availability reliable and predictable, conditions
conducive to increased specialization, which in turn, can bolster the strength and
intensity of trade to the benefit of all partners (Spielman 1991a:6). Such
arrangements are hypothesized to have been in place between northern Anasazi
groups and southern Plains populations, involving the exchange of carbohydrates
and proteins in form of com and bison meat respectively (Snow 1991; Speth 1991;
Spielman 1991b).
While models emphasizing buffering or mutualistic strategies acknowledge
the prospect of socially motivated agency to influence interaction, they largely
posit ecological factors as the ultimately determinant impetus for the initiation of
social relations and the forms they assume. Furthermore, the emphasis placed on
the exchange of economic goods largely ignores the effects of social or political
power in the realm of interaction, focusing on the form of exchange practices over
the potential significance of the interaction involved (Schortman and Urban
1987:61). The implicit assumption of such models is that the social mechanisms

of interaction arise in response to external perturbation, suggesting that
increasingly complex exchange strategies arise in response to problems
encountered by growing populations in circumscribed environments. This
perspective is therefore limited in its ability to analyze the social factors involved
in the manipulation of power existing within exchange and interaction
relationships. To address the social factors that are crucial to understanding the
nature of exchange and interaction adequately, a conceptual component focusing
on political interaction is required.
Sociopolitical Models
Researchers utilizing sociopolitical models typically credit political agency
- at group and/or individual levels as providing the necessary impetus for the
stimulation of intensified exchange and interaction. Here the control of labor and
distribution of goods are typically seen as having greater explanatory potential
than any intrinsic or culturally ascribed value of the goods themselves. While
various schemes regarding the distribution of power and the relative importance of
ideological factors have been proposed, the manipulation of power at some level is
generally regarded as the functional impetus for intensified interaction between
A major influence for this type of model is found in Wallersteins (1974)
world system theory. Developed to model the emergence of modem capitalism,

this scheme proposes that economic cores dominate economic systems by
extracting raw materials from underdeveloped peripheral areas whose populations
are, in turn, increasingly reliant on the finished goods produced in the core area.
While this approach provides an adequate model for colonial relationships, its
application to pre-modern exchange systems has been problematical. Arguments
against the application of world systems models in prehistoric archaeological
contexts (Kohl 1989:225; Spielman 1991a:6-7; Stein 1999:159, 2002:904) often
contend that the nature of pre-capitalist exchange systems was inherently different
in form and that the relative autonomy that could be maintained in peripheral
locations would work against the longstanding, stark inequity in power relations
proposed in Wallersteins scheme. While attempts to modify the core-periphery
concept of world system perspective for archaeological application have been
productive in illuminating some of the specificities of pre-modern exchange (see
Kohl 1989; Ratnagar 2001), the model tends to lose its explanatory power when
applied outside of its originally intended temporal context.
Remotely related to world systems theory (Cobb 1993:63), prestige goods
models present an alternative hypothetical structure for the development of
increasingly complex exchange practices. This type of model sees elite
individuals as manipulating the exchange of exotic, and often symbolically
charged, prestige items as a strategy of solidifying external relationships while at
the same time increasing their personal status within their particular communities

(Blanton et al. 1996:5; Brumfiel and Earl 1987:3; Cobb 1993:63; Saitta 2000:152).
In this scenario, political elites consciously and strategically employ
specialization and exchange to create and maintain social inequality, strengthen
political coalitions, and fund new institutions of control, often in the face of
substantial opposition from those whose well-being is reduced by such actions
(Brumfiel and Earle 1987:3). While this model has been applied extensively in
archaeological studies, its focus on wealth items and coercive domination has been
criticized as overlooking the variability within the organization of power among
prehistoric groups (Cobb 1993:64; Saitta 2000:153). The model has further been
criticized as potentially conflating types (i.e. prestige v. utilitarian or material v.
labor) of exchange processes, and of glossing over historical variation within
prestige goods economies (Cobb 1993:64; Saitta 2000:153-154). To address some
of the potential for variation in the organization of exchange, Brumfiel and Earle
(1987) among others (e.g. Blanton et al. 1996) make the distinction between the
types of strategies involved in exchange relationships and how they may be
differentially expressed in the archaeological record. Briefly, these distinctions
include those between subsistence goods and wealth, independent and attached
specialists, and staple and wealth finance (Brumfiel and Earle 1987:4).
A potential alternative to prestige goods models is found in the peer polity
interaction model. This approach attempts to consider the articulation of internal
and external interaction, focusing on the:

...full range of interchanges taking place (including imitation and emulation, competition,
warfare, and the exchange of material goods and of information) between autonomous (i.e.
self-governing and in that sense politically independent) socio-political units which are
situated beside or close to each other within a single geographical region, or in some cases
more widely. (Renfrew 1985:1).
The central criticism of this approach is that it tends to gloss over irregularities of
development within the broad spatial boundaries in which it is applied (Cobb
1993:65). Furthermore, this model was designed in terms of the emergence of
early states (Renfrew 1985:2), and becomes more problematic in analyzing small-
scale, nonhierarchical societies, particularly in assessing the relative autonomy of
adjacent groups.
Political approaches to exchange and interaction among middle-range
societies need to encompass the interaction within localized communities as
processes which guide and influence interaction between regionally situated
populations, which in turn affect the historical trajectory of the community in
question. Saitta (2000) believes this may best be accomplished by exploring the
conceivably fertile ground where selflessness and selfishness meet that is, where
elites may in fact function as first among equals but in some interesting and
unpredictable ways (155-156). This entails recognizing the distinction between
individual oriented network strategies, in which the power of elites is enhanced
through monopolization of long-distance exchange (as in prestige goods models),
and group oriented corporate strategies where power is achieved through effective
control over local goods and labor (Blanton et al. 1996:3-7; Saitta 2000:157).

Rather than viewing these as alternative, exclusionary models, however, a more
holistic picture emerges when they are viewed as complementary and interactive
strategies. Blanton and others (1996:3-4) propose a dual-processual theory
focusing on the contradictions and interactions between network and corporate
strategies as they pertain to the exercise of power among political agents and how
such power is expressed through elements of either strategy as they coexist or as
one emerges as predominant to the other. Saitta (2000:158) argues that it is more
likely that these strategies continuously coexist within given groups and that
exploring the tension between them provides a means of allowing for multiple
ideologies surrounding the social meaning of nonlocal places and goods, as well as
for a variety of managerial positions and strategies in the same social context.
This approach has been applied to the American Southwest in cases where
prestige goods models do not empirically fit, and views Puebloan social structure
as containing elements of both egalitarianism and stratification within complex
communal societies (McGuire and Saitta 1996). While this approach complicates
the study of interaction by increasing the number of recognized variables that must
be considered in deriving explanations, it provides an enhanced ability to consider
the specific historic frameworks within which social organization takes place,
thereby providing a more effective means of assessing ambiguities within the
archaeological record (McGuire and Saitta 1996:199; Saitta 1994:202, 210-211,

The Potential of Interaction Studies
in the Uncompahgre Plateau Area
Ambiguity can be considered one of the defining characteristics of the
Uncompahgre Plateau area sites of west-central Colorado as they are presently
understood. And while the development of sophisticated models of exchange and
interaction has provided potential tools for addressing this ambiguity, rigorous
efforts to model interaction in the area have seldom been pursued. Previous
archaeological approaches to this region have, in fact, presented an obstacle to
understanding how interaction within these communities and with their regional
neighbors has contributed to the unique archaeological manifestations observed
there. A review of previous studies and associated efforts to clarify the cultural
affiliation of this region is presented below in an effort to illuminate the nature of
these theoretical obstacles and suggest potential points of departure in the form of
interaction studies that can contribute to our understanding of Formative Era social
organization. A more in-depth presentation of the proposed cultural history of
groups in and around the study area is presented in the following chapter.
Modeling the Study Area
Interpretation of the archaeological record from the Formative Era (ca. 400
B.C. A.D. 1300) on and around the Uncompahgre Plateau in west-central
Colorado and east-central Utah has been problematic since the initial investigation

of sites there in the early twentieth century. Attempts to demonstrate cultural
affiliation for Formative groups in the area have resulted in classificatory schemes
attributing these sites variously to Anasazi, Fremont or more recently, Gateway
tradition populations. Geographically situated north of traditionally accepted
Anasazi territory and east and southeast of areas normally attributed to Fremont
groups, settlements of the area have defied ready cultural assignation. The
problem is compounded by seemingly anomalous material assemblages -
including diagnostic elements of Anasazi and Fremont and the generally poor
quality of existing data from past work in the area (Reed and Gebauer 2004:70-71;
Reed and Metcalf 1999:110). Early documentation of structural remains in the
Paradox Valley of western Montrose County suggested that these features
represent Anasazi field houses that were utilized seasonally by groups traveling
northward up the Dolores River from the Anasazi core area to the south (Jeancon
1924). The model depicting archaeological sites in this area as peripheral Anasazi
settlements would remain dominant for the next three decades. In the early 1950s
investigators noted that these same features were similar to those found at sites in
Utahs La Sal Mountains (Hunt 1953), which had been attributed to the Fremont
cultural tradition as defined by Morss (1931). The Fremont association was
further reinforced two years later in Wormingtons (1955) reappraisal of Fremont
culture, which incorporated the Paradox Valley sites among others located on the
Uncompahgre Plateau.

More recently, some archaeologists (Cassells 1997:182-186; Reed 1997;
Reed and Metcalf 1999:132-140) have rejected the application of the Fremont unit
for the area in favor of the term Gateway Tradition (Reed 1997) to signify an
indigenously derived cultural occupation of the area during the Formative Era.
While some view the Gateway concept as heuristically useful and the designation
has made its way into regional overviews (e.g., Cassells 1997; Reed and Metcalf
1999), it has been subjected to limited criticism (e.g., McMahon 2000). The
utility of the Gateway designation is somewhat questionable and the concept
warrants further scrutiny for multiple reasons, which are discussed in greater detail
The Trouble with Fremont. The concept of Fremont, as used to describe
Formative Era populations of the Great Basin and northern Colorado Plateau, has
been generally fraught with similar kinds of arguments and counterarguments
regarding typology and utility which are particular to the archaeology of west-
central Colorado. A.V. Kidders (1924) designation of portions of what is now
generally considered the Fremont area as the Northern Peripheral District of the
Greater Southwest based on observed similarities between Fremont and Anasazi
material culture, architecture, and subsistence strategies would be challenged by
Wormington (1955) and others (Jennings and Norbeck 1955), who viewed the
Fremont as representing an in situ development influenced by the preceding

indigenous Archaic groups of the Desert Culture with limited diffusion of select
traits from the Anasazi to the south. Aikens (1966) has alternately proposed that
Fremont origins stem from Athabaskan groups on the Northwestern Plains who
adopted some Southwestern characteristics. More recently, Madsen (1979) has
rejected the term as applied to groups of the eastern Great Basin in favor of an
entity he refers to as the Sevier Culture who relied largely on a marsh resource
gathering subsistence strategy supplemented by com agriculture. Madsen does,
however, retain the use of the term Fremont Culture for groups occupying the
Colorado Plateau from about 450 1250 A.D. who appear to be considerably
more reliant on com agriculture than the eastern Sevier.
Despite serious questions regarding issues of behavioral continuity and
ethnic association among groups that have fallen under the label of Fremont, the
term has endured as being useful in distinguishing Formative groups of the
intermountain west from the hunter-gatherer groups that preceded them and later
Protohistoric populations. And though the term itself has come to be characterized
by variability, Fremont retains utility in distinguishing these groups, who are
generally more similar to one another than they are to surrounding groups and
appear to share ritual concepts, from horticulturalists in the Anasazi region to the
south (Stone 1999:118). Refinement of the concept has led to a general, if
somewhat contentious, designation of five Fremont variants based on local
variability (Figure 2.1): the Parowan (A.D. 450 1250) centered in southwest

Utah, the Sevier (A.D. 780 1260) of west-central Utah and east-central Nevada,
the Great Salt Lake (A.D. 400 1350) of northern Utah, the Uinta (A.D. 650 -
950) in northeast Utah and northwest Colorado, and the San Rafael (A.D. 700-
1250) of east-central Utah and west-central Colorado (Jennings 1978; Marwitt
1970; Stone 1999:118-119).
Figure 2.1. General locations of Fremont variants (modified from Stone 1999:118).
Recognized shared attributes of these groups include, coiled pottery, one-rod-and-
bundle basketry, unique one-piece moccasins, clay figurines, and distinctive rock
art featuring elaborately ornamented trapezoidal anthropomorphs (Madsen 1989).

The area considered in this study is spatially situated within areas traditionally
ascribed to the eastern San Rafael Fremont of the above classificatory scheme.
Structural remains and ceramic assemblages observed at San Rafael
Fremont sites may be indicative of a greater reliance on agriculture and more
extensive networks of interaction (Barlow 2002; McDonald 1994; Talbot
2000:282-284, 2004) in comparison to some of the other recognized Fremont
variants. Circular and square habitation structures featuring masonry architecture
that resemble Puebloan structures (McMahon 2000:12-18; Talbot 2000:282-283)
are found in the area, as are slab-lined hearths and pit houses and dome-shaped
adobe granaries (Jennings 1978: 184-187; Marwitt 1986:170; Stone 1999:126).
Ceramic assemblages of San Rafael Fremont sites usually contain Kayenta and
Northern San Juan Anasazi types (Stone 1999:126; Talbot 2000:283) in addition
to the locally produced Emery gray (Madsen 1977:31). Though the nature of
interaction between the San Rafael Fremont and the Northern San Juan Anasazi is
poorly understood, it is not unexpected given the spatial proximity of these two
groups (Stone 1999:126).
The Gateway Alternative. The problem of interpreting Formative Era behavior as
illustrated by the general difficulties in defining coherent social units for the area
as a whole are uniquely acute in considering the archaeology of the Uncompahgre
Plateau and surrounding areas. Reed (1997; Reed and Metcalf 1999:131-140) has

rejected the hypothesis that Formative groups of west-central Colorado and east-
central Utah represent peripheral Anasazi settlements based on the absence of key
elements of the Anasazi tradition such as ceramic production, architectural
conventions including the use of kivas and water control structures and
evidence of complex interregional relationships with Chacoan logistical networks.
Reed (1997) also believes it is unlikely that these groups are affiliated with the
Fremont based on the absence of the Fremont traits cited above as defined by
Madsen (1989). Reed (1997) does, however, note that architectural elements,
namely circular habitation structures, observed at Fremont sites to the west and
north are evident at some of the Uncompahgre Plateau sites. More recently Reed
(Reed and Metcalf 1999:113,121-122) has acknowledged that sites in the Glade
Park area on the northern Uncompahgre Plateau west of Grand Junction are likely
of Fremont origin due to the presence of identified Fremont rock art motifs of the
Sieber Canyon Style that has been associated with the Uinta Fremont (Cole
To facilitate discussion of the archaeology in this area, Reed (1997) has
proposed the use of the term Gateway Tradition. Geographically, the Gateway
tradition is roughly bounded by Interstate 70 to the north, the Uncompahgre River
to the east, the Green River to the west, and extends as far southward as
Monticello, Utah (Figure 2.2). The Gateway Tradition has been tentatively dated

from 500 B.C. to A.D. 1250, a period coinciding with evidence for com
horticulture in the area (Reed 1997).
Figure 2.2. General location of Gateway Tradition (Reed 1997:20).
Traits attributed to the Gateway Tradition include limited reliance on corn
horticulture, manufacture of small arrow points, lack of ceramic production,
construction of storage units in rock shelters, habitation of noncontiguous circular
masonry structures and possibly pithouses, and evidence of the exchange of
technology and ideas with Anasazi and Fremont groups as evidenced by Anasazi,
and to a lesser degree, Fremont ceramics and rock art motifs exhibiting the

influence of both groups (Reed 1997; Reed and Metcalf 1999:131). Reeds (1997)
impetus for designation of the Gateway Tradition is his belief that traditional
application of the Fremont unit has impeded research in the area by masking
significant variation that can contribute to our understanding of past human
The unique nature of the archaeological record on the Uncompahgre
Plateau is undisputed, and thus far there have been few published critical
assessments of Reeds Gateway construct. McMahon (2000) suggests that
application of the concept, despite its intended purpose of clarification, is
premature, and that until positive evidence demonstrating in situ cultural
development from indigenous Archaic populations can be demonstrated, the terms
Fremont or San Rafael Fremont are more conducive to furthering our
understanding of the relationships of these sites with surrounding regions. In
support of retention of the Fremont nomenclature, McMahon (2000) cites
architectural similarities with known Fremont localities and, to a lesser extent,
similarities in artifact assemblages and rock art motifs among the sites of western
Montrose County.
Additional theoretical and empirical problems attest to the premature
nature of the Gateway concept. Firstly and not insignificantly, the name itself,
though reputedly taken from the small Colorado town west of the Uncompahgre
Plateau, is rife with unfounded and unwarranted implications. While evidence

may emerge which places Gateway sites in an interactively intermediate position
between Fremont and Anasazi groups, existing evidence does not provide support
for the implicit suggestion, and the form of organizational relationships that
apparently did exist remains unclear. Until these relationships can be
meaningfully modeled in ways that illuminate the exchange of goods and
information and attendant power relations within and between sites in this area,
terms that imply the functional significance of the area within a broader context
should probably be withheld. It is also notable that the Gateway Tradition is
defined almost entirely by the absence of select material manifestations and/or the
presence of presumably exotic material. This sort of negative definition provides
very little in the way of heuristic utility, and despite the intent to represent
variation in the archaeological record, this approach actually holds potential to
mask variation by applying a typological construct that could be interpreted as
being representative of a bounded cultural entity.
Further limitations inherent in the Gateway construct have recently been
acknowledged by Reed (2004; 2005) in light of data obtained through the
TransColorado Natural Gas pipeline mitigation project (Reed et al 2001). Based
on accumulating evidence including that which suggests a prolonged hiatus in
horticultural subsistence strategies in the area coupled with intensification of
procurement and processing of wild plant food resources between A.D. 400 and
A.D. 900 Reed (2004; 2005) believes it appears increasingly unlikely that

Formative era groups in the area represent indigenous groups who simply
borrowed elements of the Anasazi and Fremont (Reed 2005:17) as suggested in
the initial formulation of the Gateway concept. Though he still believes that the
Gateway designation may retain utility in reference to some sites west of the
Uncompahgre Plateau that date between A.D. 900 and A.D. 1100 (spatial and
temporal spans which roughly coincide with those considered in this thesis), Reed
(2005:30) suggests that the application of this unit seems untenable in reference to
groups predating A.D. 900.
Interregional Relationships
So while retention of the Fremont designation is far from ideal, the
Gateway construct fails to provide a satisfactory alternative. It is argued here that
attempts to apply such labels, purportedly indicative of cultural affiliation, have
hindered efforts to understand Formative era developments among occupants of
the Uncompahgre Plateau and surrounding areas. To gain insight into the complex
and fluid social structure of these groups, a different approach is required. To this
end, this study focuses on how Formative era occupants of the area interacted with
groups from surrounding areas and how these relationships may have changed
through time as an approach to providing a more clear and nuanced understanding
of the nature of social structure in the area within the context of surrounding

Key elements that shaped the nature of such interaction, such as
directionality, scale of exchange, and network strategies should be observable to
varying degrees through examining variation within and between assemblages. A
quantitative, compositional comparison of locally produced wares to those
imported from surrounding areas will provide an initial indicator of the scale and
relative directionality of exchange that was occurring between populations in
west-central Colorado and those in surrounding regions. Complementary stylistic
analysis can provide a basis for interpreting the extent to which the transmission of
ideas as manifested by commonalities in production technique and decorative
elements of surrounding areas impacted groups living within the study area.
Subsequent analysis of between-site variation within the Uncompahgre Plateau
area can then provide a basis of assessing the forms of exchange which were
practiced. These data, considered within environmental and historical contexts,
should contribute to a more in-depth understanding of the underlying mechanisms
of exchange in the area, and in turn will permit more informed assertions
regarding the now poorly-understood nature of internal social structure in the area
and how such structure relates to surrounding regions. Furthermore, an
examination of how exchange relationships change through time can provide
insight into the fluid of the social constructs in which they occur.
With emphasis on social organization, this thesis evaluates the
applicability of elements of both environmental and socio-political explanations of

interaction in exploring the underlying nature of regional and interregional
interaction. Existing ecologically-based models are often premised on the
presence of undeniable and insurmountable environmental constraints which
cannot be simply dismissed. While these models consider the conditions which
contribute to the settings conducive to various forms of interaction, they often fall
short in their consideration of the human-driven and often archaeologically
ambiguous components of exchange systems that permit such systems to function.
Here, an effort is made to transcend normative constructs which inform the bulk of
the collective knowledge of the area in question and to consider the interplay of
environmental constraints with historical and social conditions as a means of
furthering such knowledge.
What emerges from the archaeology of the Uncompahgre Plateau and
surrounding areas is a complex picture of sustained and possibly intense
interaction between people employing variant social identities as reflected in the
diverse assemblage of materials they left behind. The spatial and temporal
position of Formative sites on and around the Uncompahgre Plateau demands an
approach that seeks to understand how such interaction contributed to the
formation of the peculiarities evident in the archaeological record of this area.
Normative approaches to classification can only obstruct studies that foreground

interaction by placing a theoretical emphasis on boundedness and separation rather
than elements of interaction and exchange. While evidence may, in fact, point to a
unique and indigenously derived cultural unit in this area, existing data seemingly
indicate somewhat fluid interaction through the exchange of material and ideas
over prolonged periods. It seems unlikely that a cohesive unit such as that implied
by Gateway will clearly emerge, and attempts to identify such units may prove
counterproductive to developing an understanding of the dynamics of exchange
and interaction that are currently evident. The development of theoretical models
addressing interaction since the inception of archaeology as a discipline, and
particularly over the last two decades, has provided possible lines of inquiry that
have largely remained unexplored in relation to the Uncompahgre Plateau area.
This thesis represents an attempt to employ a socio-politically oriented approach
that incorporates environmental variables within historic frameworks as a
potentially more promising approach to examining social interaction in the region
than previous typological divisions of cultural affiliation.
The requisite initial step in modeling interaction for the area is the
determination of the provenance of potential trade items. It has largely been
assumed based on typological similarities and the absence of direct evidence for
ceramic production that ceramic artifacts recovered in the study area were
imported from neighboring groups. Until recently (i.e. Greubel, et al 2006),
however, this assumption has generally not been tested through alternate lines of

inquiry. Assessing points of origin for ceramics recovered in the Uncompahgre
Plateau area permits the formation of new hypotheses regarding pottery
distribution in the region and provide a basis for a more meaningful discussion of
the ways in which groups that occupied the area interacted with their neighbors.
As previously stated, the question of where ceramics recovered in this part
of west-central Colorado is, therefore, the primary question addressed in this thesis.
Archaeologists currently conducting research in this area must contend with a
limited database resulting from a lack of controlled excavation. Through the study
of materials which are commonly available for research here, namely surface
context cultural material, this study utilizes complementary lines of evidence with
an emphasis on compositional analysis, in an effort to determine where ceramic
assemblages originate and thereby provide a point of reference from which a host
of secondary questions regarding social organization, subsistence, settlement and
space-time systematics can be more satisfactorily addressed.
In this study, 221 ceramic artifacts collected from 15 sites within the
Gateway tradition area are analyzed. With the exception of an additional 224
ceramic artifacts recovered from three sites (Greubel et al 2006:37) in the area,
this represents the bulk of known ceramics to have been collected from Gateway
area sites within the OAHP database that were available for analysis at the time
this study was initiated. Moving forward from the initial question of where these
artifacts were manufactured (are sherds largely imported, or do they represent

local ceramic production?), a context for consideration of subsequent questions is
provided. Such questions and associated implications include:
If most or all of the sherds are locally produced, are manufacturing loci
dispersed throughout the study area or concentrated among relatively few
sites? A focused area of production within the study area could be
indicative a regional center of specialization and potentially the presence
of disproportionate economic influence within the study area, whereas a
more dispersed pattern of production could signify a more egalitarian and
relatively autonomous arrangement existing between various sites. In
either case, the preponderance of locally produced pottery would
necessitate the revision of the Gateway construct as representative of an
aceramic cultural tradition.
If locally produced, how do the sherds compare stylistically with those
from surrounding areas? Strong stylistic similarity in this case could
indicate the transmission of ideas such as might occur through migration or
other social mechanisms in the presence of relative economic
independence in the region.
If most or all of the sherds are imported, were they produced in multiple
locations, or relatively homogeneous, restricted manufacturing locales?
Limited trading partners would imply a particularly strong relationship
between Uncompahgre Plateau area sites and an exchange system

involving balanced reciprocity, while multiple, diverse trading partners
could potentially signify the presence of a more complex system
possessing elements of redistributive exchange mechanisms.
How do the intensity and/or directionality of pottery production or
exchange fluctuate through time? A temporal trajectory closely paralleling
those known for adjacent regions may reflect the association of
developments taking place across regions, whereas divergent
developments through time restricted to the study area could imply a
relatively independent, restricted development of social arrangements.
Considered in the environmental and cultural contexts of the study area and
surrounding regions, all of these questions hold the potential to provide
information regarding interregional relationships and social organization among
inhabitants of the Uncompahgre Plateau area. While the thesis cannot provide
comprehensive answers to all of the questions considered, it does provide a means
of eliminating some possibilities and supporting alternate explanations.

HISTORY: A.D. 1 to 1300
A chronological synopsis of what is currently known regarding the
Formative-era occupation of the study area and immediately surrounding regions
is presented in this chapter. Archaeological investigations that have been
previously conducted in the area and informed the existing culture history are
considered. Data regarding culture history are presented in an effort to provide a
larger context for the current study. Such a review is particularly important since
it provides a framework for the diachronic perspective underlying the study, and
because the study seeks to provide data capable of contributing to the refinement
of the existing culture history as it applies to Formative-era occupation. As
discussed in chapter 2, the current understanding of Formative-era cultural
affiliation in the study area is uncertain and the subject of recent archaeological
debate. A summary of the generally accepted sequence of prehistoric occupation
of the area and descriptions of some of the cultural content on which
archaeologists base the current discussion of temporal and spatial divisions is
presented in this chapter.
Much of the information here is gathered from Reed and Metcalfs (1999)
Context for the Northern Colorado River Basin, which is generally applicable to
the study area. Discussion of neighboring Anasazi groups to the south of the study

area is derived in large part from Lipe et als (1999) Context for the Southern
Colorado River Basin, which provides an in-depth treatment of the archaeology of
southwestern Colorado. The focus of the discussion presented in this chapter and
the temporal period in question for this study is the Formative era (A.D. 1 to 1300).
The preceding Paleoindian (11,500 to 6,400 B.C.) and Archaic (6,400 B.C. to A.D.
1) eras are not treated extensively here though numerous in-depth treatments can
be found in the above-mentioned sources and elsewhere (e.g. Frison 1992; Madsen
1982; Pitblado 1994).
As applied to the area from which sites in this study are located, the use of
the archaeological construct Formative era should be, interpreted as a time
when the horticultural adaptations were fluorescing in many parts of the western
United States; no inference about the actual subsistence practices of the peoples
subsumed by the Formative era should be made (Reed and Metcalf 1999:5).
While the focus of the study is interaction, especially as inferred through the
movement of ceramic artifacts across space, associated regional Formative-era
developments in areas such as technology, architecture, and subsistence strategies
area also considered here as factors that contributed to the environment in which
interaction occurred.
An understanding of cultural change through time in surrounding areas is
essential to the discussion of exchange in the study area. While it is beyond the
scope of this work to provide a comprehensive, in-depth treatment of the culture

histories of all geographically adjacent areas, this section contains a general
overview of Formative cultural developments in immediately surrounding regions
as well as those in the study area. While some broad and far-reaching
developments which took place among the cultures of the Formative era in the
Southwest and the Great Basin namely the Anasazi and the Fremont are
touched upon in this chapter, the emphasis is on those branches of these groups
which area immediately adjacent to the study area, the Northern San Juan Anasazi
and the San Rafael Fremont. Finally, this chapter contains a summary of the
Gateway Tradition, which has been proposed as an alternate cultural unit within
the study area, applicable where the evidence of Anasazi and/or Fremont
occupation is ambiguous.
The Northern San Juan Anasazi
Due in large part to the work of early twentieth-century archaeologists
employing methods stressing stratigraphic-chronological (Lipe 1999a:61) goals,
the diachronic cultural development of ancestral Puebloan, or Anasazi, groups to
the south of the study area are better understood than those in surrounding regions.
The subsequent development of an intact and continuous dendrochronological
sequence further contributed to archaeologists ability to apply relatively fine-
grained temporal assignation in this area. More recently the accumulative body of
knowledge regarding the prehistory of areas to the south of that considered in this

thesis was bolstered substantially through the Dolores Archaeological Program
(DAP). The DAP was a large-scale mitigation project conducted on behalf of the
Bureau of Reclamation by multiple individuals representing several academic and
government institutions between 1978 and 1983 prior to the construction of
McPhee Reservoir in Dolores and Montezuma Counties, Colorado (Robinson et al.
1986). The DAP provided a wealth of data that is still frequently drawn upon by
the areas archaeologists, and thus, much of the discussion below is derived either
directly from DAP efforts or subsequent interpretations.
Despite shifts in nomenclature and periodic modification over time, the
chronological sequence proposed by Kidder (1927) at the original Pecos
conference, though not appropriate to the Southwest as whole as originally
suggested, is still widely accepted by southwestern archaeologists as largely
applicable to the San Juan Anasazi groups that neighbored Formative era groups
of west-central Colorado. And while the concept of a progressive cultural
continuum posited at Pecos (Kidder 1927) has proved untenable, the chronological
framework in which it was presented retains utility. The first recognized periods
in this area that correspond to the earliest dates of west-central Colorados more
general Formative era is represented by Basketmaker culture.

Basketmaker II: 1,000 B.C. to A.D. 500
Though the designation of Basketmaker II is now generally used to refer to
the period in which evidence of widespread agriculture first appears in the Four
Comers area between A.D. 1 and A.D. 500 (Lipe 1999b: 133), in its broadest
application the term encompasses the postulated Basketmaker I period, which
Kidder (1927:490) tentatively characterized as ...preagricultural, yet
adumbrating later developments. This postulated, preceding period never fully
materialized, and Southwestern archaeologists have yet to arrive at a consensus on
a chronology for the area that distinguishes the long period of largely
preagricultural occupation from the subsequent Basketmaker II period (Lipe
1999b: 132-134). This discussion follows Lipe (1999b: 133-134) in considering
Basketmaker II as extending from the appearance of maize in the region to the
first widespread occurrence of pottery at about A.D. 500. In general terms, this
period can be considered representative of a transition from earlier Archaic
practices to the established agricultural practices that characterize the subsequent
Basketmaker III period, distinguished by both cultural continuity and the
formation of persistent traits which characterize later periods of the Anasazi
tradition (Cassells 1997:146-149; Janetski 1993; Lipe 1993, 1999b; Stone
Basketmaker II occupation of the Four Corners area is, in fact, mainly
defined by the presence of com. The period is characterized as one during which

hunting and gathering continued to play a crucial role in subsistence strategy
alongside the emergence of increasing reliance on maize agriculture (Lipe
1999b: 159-161). Squash is also noted among cultigens utilized during
Basketmaker II (Lipe 1993:2), though the completion of the Southwestern triad
through the addition of beans occurs during following periods. The period is
further defined, as the name suggests, by widespread utilization of basketry and
the relative absence of pottery. Though Basketmaker II groups utilized clay in
lining baskets and storage features, they are generally considered to have been
aceramic with the advent of pottery taking place during the subsequent
Basketmaker III period (Cassells 1997:148; Lipe 1993; Stone 1999:91-92). As
during the preceding Archaic period, hunting technology among Basketmaker II
groups relied largely on atlatl and dart as indicated by large, side and corner-
notched projectile points (Lipe 1993:9) sometimes found in association with dart
foreshafts (Janetski 1993:224); cordage for nets and snares has also been
recovered in association with Basketmaker II occupations (Cassells 1997:148).
Other textiles, including sandals, twined bags, winnowing trays, and blankets from
rabbit fur intertwined with cordage (Lipe 1993; Janetski 1993) are also attributed
to Basketmaker II occupations. A growing reliance on agriculture is suggested by
the presence of increasingly specialized groundstone tools within Basketmaker II
assemblages which have been interpreted to be demonstrative of a progressively

greater emphasis on com processing as the period progresses (Lipe 1993:10; Stone
Increased seasonal sedentism is evinced during the period, especially in
late Basketmaker II, by the presence of typically shallow pithouses and associated
storage features. Habitation sites are also often located in rock shelters, and open
campsites were presumably important as well though more difficult to attribute to
Basketmaker II with certainty. The degree of relative mobility among these
groups is poorly understood and appears to vary geographically and temporally
(Lipe 1996b; Janetski 1993). It merits note here that Basketmaker II assemblages
often contain a higher relative frequency of exotic materials such as shell than
those attributed to later Puebloan groups (Lipe 1993:8; Lipe 1999b: 165). The
source of these materials and the mechanisms through which they were
incorporated into Basketmaker II assemblages is still open to interpretation, but
the suggestion of a more broad scope of external interaction is certainly of interest
in this context due to potential implications it could have regarding interaction
with Uncompahgre Plateau area groups during the period.
Little is known about the social structure of Basketmaker II society.
Architectural and settlement pattern data generally do not support the presence of
social differentiation (Lipe 1993:8), and whether these groups were migrant
populations or local, indigenous groups adopting new subsistence strategies
through diffusion remains open to question (Lipe 1999b: 164). Despite the

absence of clarity in existing interpretations of the archaeological record during
this time, it is clear that Basketmaker II groups established a series of regionally
accepted practices, such as those related to maize agriculture, that persisted into
subsequent periods.
Basketmaker III: A.D. 500 to 750
As with the transition from Archaic to Basketmaker II, the understanding
of how Basketmaker III groups relate to their immediate predecessors and
subsequent Pueblo I groups is limited. An increasing dietary emphasis on corn is
the chief characteristic of continuity from the preceding period. Coprolite analysis
(Minnis 1989) indicates that corn had, by this time, assumed a level of importance
in the regional diet that would persist for the next six centuries, through the Pueblo
III period. Beans had by now also been introduced into the variety of cultigens
propagated by Basketmaker farmers (Lipe 1993:2-3; Minnis 1989). A potentially
related technological innovation of note that takes place during Basketmaker III is
the widespread adoption of the bow and arrow (Lipe 1993:2), which may represent
a more tactically efficacious tool for logistical hunting forays taking place away
from areas of permanent habitation.
The technological hallmark of Basketmaker III culture, however, is
represented by the widespread appearance of pottery in the archaeological record.
Some occurrences of brown wares are attributed to the early years of the period,

but it is the proliferation of plain gray ware jars and decorated vessels that
provides the distinguishing archaeological indicators of Basketmaker III
components (Lipe 1993:2-3). Among the most prominent of early gray wares in
areas adjacent to the study area considered in this thesis is Chapin Gray, a pottery
type made through a coil and scrape method, tempered with igneous rock or
sandstone, and fired in a partially oxidizing atmosphere (Bretemitz et al. 1974:1-2;
Oppelt 1991:49). Chapin grey appears in the record during early Basketmaker III
(ca. A.D. 575) and its presence persists until the end of the subsequent Pueblo I
period (ca. A.D. 900) (Blinman 1986:72; Bretemitz et al. 1974:2; Oppelt 1991:50).
Chapin Black-on-white is the dominant decorated ware in the area during
Basketmaker III (Blinman 1986:71), a type similar to Chapin Gray in production
method and composition but distinguished by applied monochrome designs,
usually of mineral paints of reddish-brown to black and generally confined to
bowl interiors (Bretemitz et al. 1974:25-27; Oppelt 1991:58).
Architectural developments occurring during Basketmaker III are
perceptible in the appearance of pit structures that usually feature antechambers
and are larger, more substantial, and relatively deep compared to those attributed
to Basketmaker II (Lipe 1993:2-3; Stone 1999:93; Wilshusen 1999a: 174).
Pithouse depth continues to increase throughout the period (though other structural
attributes remain quite consistent throughout) and features such as interior hearths,
and interior and exterior storage bins are commonly associated with these

domestic habitation units (Wilshusen 1999a: 174). Exterior space also becomes
more formalized during the period, with some sites exhibiting outdoor work areas
and ramadas (Stone 1999:93). Small hamlets consisting of one or two households
represent the most commonly observed intrasite settlement pattern during this
period; these sites are often situated within larger, dispersed communities of five
to nine households, and settlements generally appear to have been occupied for
longer periods than during the preceding Basketmaker II period (Wilshusen
1999a: 175). Though little is known regarding the intricacies of Basketmaker III
social structure, evidence suggests that large hamlets consisting of tow or more
contemporary pitstructures may have served as focal points of communal
organization (Wilshusen 1999a: 188) which could be interpreted as a precursor of
the increasingly centralized organization that emerges during the ensuing Pueblo
Pueblo I: A.D. 750 to 900
As implied by the term pithouse to pueblo transition, this period
witnesses the rise of above ground architecture among the Anasazi of the Four
Comers area. This is a period of rapid and substantial change in demographics
and social organization signaled by evidence for rising populations and the
establishment and growth of aggregated villages (Stone 1999:95; Wilshusen
1999b). It is estimated that villages of over fifty contiguous or near-contiguous

rooms served as habitation for over half of the population of the northern San Juan
area by the end of the period (Wilshusen 1999b:210). Pit structures remain in use
and become increasingly deep (Wilshusen 1999b:201-203) throughout Pueblo I,
but are now more commonly associated with surface room blocks containing
multiple, contiguous rooms constructed of jacal over sandstone slab foundations
early in the period, increasingly incorporating more stone masonry as the period
progresses (Wilshusen 1999b:201; Stone 1999:97). Though there is considerable
variation in village layout, as few as two and up to 18 room blocks each
containing domestic household suites and storage areas for agricultural produce
and housing between three and 19 associated households situated in clusters
with associated pit structures and great kivas is typical of Pueblo I village
configuration by A.D. 840 (Wilshusen 1999b:213-219; Stone 1999:97). The
increased occurrence of above ground dwellings including interior storage space
compliments other lines of evidence at site and regional levels suggesting that
corn and bean agriculture provided the mainstay of Pueblo I subsistence.
The changes in Pueblo I lifeways related to increased sedentism and
intensification of agriculture also are also associated with a concomitant rise in
social complexity as suggested by the presence of great kivas at some villages. Pit
structures are generally not used as primary habitation structures during Pueblo I,
and they appear to assume an increasingly ceremonial role in Pueblo society
(Cassells 1997:151). While the role of some smaller subterranean structures, or

the clan kiva is open to debate (see Stone 1999), the development of the great
kiva essentially a pit structure of greater than 10 meters in diameter can be
seen as the first instance of monumental public architecture occurring in the area
with the function of establishing social structure at community levels and
reinforcing concepts of social identity (Wilshusen 1999b:219-221). Further
efforts to reinforce concepts of integrated identity that necessarily underlie the
formation and maintenance of aggregated communities are also reflected in
Pueblo I rock art motifs (see Cole 1990:130-163), the presence of landscape
features such as shrines (Wilshusen 1999b), and increased conformity to
established ceramic traditions (Blinman and Wilson 1993:76).
Changes in pottery production that took place among the northern Anasazi
during the Pueblo I period include: 1) the development of regionally specialized
red wares at the beginning of the period, 2) improved production efficiency
occurring around A.D. 825, and 3) a decreased level of specialization in gray ware
vessels (Blinman and Wilson 1993:77). Unlike previously manufactured gray and
white wares, red wares required a firing technique that provided an oxidizing
atmosphere to impart the red color to the finished pottery and were generally
manufactured from clays not locally available to occupants of the northern San
Juan area (Blinman and Wilson 1993; Oppelt 1991:75). San Juan Red Wares,
nonetheless, comprise as much as eight percent of ceramic assemblages in the
Dolores area over 50 kilometers from the source of production in southeast Utah

- from this period (Blinman 1986:94; Blinman and Wilson 1993:75), reflecting
the importance of long-distance exchange during the early Pueblo I period.
Improvements in pottery forming and firing techniques are indicated by an
increased occurrence of pottery scrapers and scraping marks and increased vessel
strength resulting from higher temperatures and/or longer duration in the firing
process in assemblages postdating A.D. 800. These technological innovations are
accompanied by a subsequent decrease in gray ware production specialization as
indicated by the presence of fewer non-producing households occurring after A.D.
825 (Blinman and Wilson 1993:76).
By the mid 800s red ware exchange declines and large villages begin to
disperse. While it is unclear whether the decline in red ware exchange reflects
decreased production in the source area or disruption of exchange networks
resulting from the dissolution of villages, the trend toward aggregation that
characterizes Pueblo I settlement patterns in the northern San Juan area had clearly
reversed by the end of the period. The rise in population that facilitated village
aggregation during early Pueblo I also reverses dramatically during the periods
terminal decades as the areas occupants apparently emigrated from southwestern
Colorado to areas along the San Juan River in New Mexico (Wilshusen
1999b:234-238). Accompanying these shifts in demographics and settlement
systems is an apparent constriction of trade networks in the area. As red wares
become less prominent in the archaeological record there is a notably high rate of

intersite variation among non-local grey and white wares, suggestive of the
continuation of multiple, independent avenues of exchange rather than the
emergence of any kind of homogeneous, unidirectional flow of trade goods across
the landscape (Blinman and Wilson 1993:78).
The causes of Pueblo I village aggregation and subsequent dissolution are
manifold and variant. Environmental factors affecting resource availability,
including periodic fluctuation in precipitation rates resulting in episodes of
drought (Petersen 1988; Stone 1999:21-22; Van West and Dean 2000) certainly
played a role (see Schlanger 1983), and Wilshusen (1991) has convincingly
argued that the selective advantage of cooperation at the suprafamily level,
required to maintain aggregated communities, diminished as environmental
changes contributed to a decrease in the spatial variability of agriculturally
suitable land. Wilshusen (1999b:237-238) further suggests that emigration from
the area may be related to greater risks involved in the dryland farming dependent
on bimodal (i.e. winter and summer) precipitation pattern of southern Colorado
compared to the relatively lower risk practices encountered in northern New
Mexico which were reliant on runoff from primarily summer-dominant patterns of
precipitation. Even in consideration of the strength of these environmentally
based accounts, Wilshusen (1999b:238) stresses that, any explanation must also
make sense of the cultural histories of the Basketmaker III and early Pueblo I

communities that come before and the early Pueblo II communities that follow
closely on the heels of the migration from this region.
Pueblo II: A.D. 900 to 1150
The onset of the Pueblo II period is marked by a low point in population
density among the northern Anasazi that would persist throughout the tenth
century (Lipe and Varien 1999a:253; Schlanger 1983). The aggregated villages of
the Dolores area were totally abandoned by A.D. 920 (Kane 1986:368-369), and
sites of this period are generally concentrated further south along the drainages of
the McElmo and Mancos Rivers (Lipe and Varien 1999a:244). Dispersed clusters
of small habitation sites including surface rooms of jacal or masonry, small pit
structures, kivas, and midden areas are typical of early Pueblo I (A.D. 900 to 1050)
settlement systems. The presence of great kivas as central features persists
throughout the period (Lipe and Varien 1999a:244-256), though these features and
others indicative of community centers are relatively rare when compared with the
preceding Pueblo I period (Lipe and Varien 1999a:256). There are few datable
sites in the area from this period when compared to earlier and later periods, with
the exception of the Chimney Rock area in the Upper San Juan-Piedra drainages
where small aggregates exist (see Eddy 1993), and evidence suggests that this
dispersed settlement pattern was more or less the norm for the early Pueblo II
period throughout the tenth century. During the A.D. 1000s populations in the

area appear to rebound somewhat, and new settlements are established in
previously unoccupied areas throughout the northern southwest (Lipe and Varien
1999a:256). Widely dispersed habitation patterns continue to dominate in the area
into eleventh century, but a notable change occurs around A.D. 1075 with the
appearance of Chacoan great houses on the landscape.
Built largely within existing dispersed communities, sites containing
Chacoan great houses are distinguished from typical Pueblo II habitation sites in
the area by an increased level of architectural formality and higher proportion of
ritual to domestic space (Lipe and Varien 1999a:258; Stone 1999:100). These
sites are characterized by the presence of massive core and veneer masonry
structures, often two stories high, and the near ubiquity of associated formal,
masonry-lined great kivas that are often fully enclosed within quadrangles
bounded by surrounding structures; and while these sites were not necessarily built
to a single, standardized regional plan and they were subject to incremental
modification through time, they do in contrast to earlier Pueblo I villages -
exhibit evidence of architectural planning, especially in the large room blocks that
represent initial great house construction episodes (Cassells 1997:155; Lipe and
Varien 1999a:258; Stone 1999:100). The great houses themselves were not
necessarily pueblos as might be thought of in the context of the preceding Pueblo I
period; rather than representing contiguous habitation room blocks of an
aggregated village, great houses were highly visible public spaces which were

surrounded by loosely distributed clusters of habitation sites (Lekson and
Cameron 1995:185-187; Lipe and Varien 1999a:259). In addition to the
aforementioned Chimney Rock, other prominent great house sites in southwestern
Colorado include Lowry Ruin, Escalante Ruin, Mud Spring Ruin, and Yucca
House among others.
Interpreted as Chacoan outliers, these sites clearly exhibit symbolic and
economic ties to the influential power center of New Mexicos Chaco Canyon
where great house construction peaked from around A.D. 1080 to A.D. 1100
culminating in the establishment of several large towns such as Penasco Blanco,
Pueblo Bonito, and Una Vida (Lipe and Varien 1999a:258; Stone 1999:101). The
Chacoan system dominated vast areas of the southwest from A.D. 1050 to 1150
(Lekson and Cameron 1995; Stone 1999:100), though the Uncompahgre Plateau
area is bereft of material clearly indicative of Chacoan influence (Reed and
Metcalf 1999:98-107). This being the case, it is surely beyond the scope of this
work to thoroughly explore the ill-defined way (Lekson and Cameron 1995:185)
in which Chacoan dominance of the region was achieved and maintained. It is
interesting to note, however, that the period of great house construction in areas
immediately south of the study area takes place during a time in which the locus of
Chacoan influence appears to shift northward from Chaco Canyon itself to the
Totah area, centered near Aztec, New Mexico (Lekson and Cameron 1995; Lipe
and Varien 1999a:258-260). Whether this apparent spatial shift represents a

decline in the primacy of Chaco Canyon or an increase in its dominance of
regional exchange marked by northward expansion (see Lekson and Cameron
1995), the fluorescence of the relatively nearby Totah district clearly had a
substantial impact on populations in southwest Colorado.
While great houses are the most obvious indicators of northern Anasazi
participation in the Chacoan network, ceramic assemblages from the period also
provide evidence of regional involvement in this encompassing system of
interaction, and the period is characterized by a general increase in both local and
regional ceramic exchange. Gray ware production occurs at most households
throughout the period, and while white wares from the period are highly variable
throughout the region a fact that complicates the assignation of precise
production localities they are generally more abundant in Pueblo II assemblages
than in those attributed to earlier periods. Innovations in temper use during this
period, namely the use of sherd temper, further complicate provenance
identification since inclusions in sherd temper often cannot be identified with
binocular microscopes (Blinman and Wilson 1993:79). Another innovation of
note is the introduction of corrugated wares during this period, occurring around
A.D. 910 to 930 (Lipe and Varien 1999a:261). The function of corrugation is not
fully understood, but it is suspected that it may be related to ease of handling the
cooking and storage vessels on which this surface treatment is typically found
(Oppelt 1991:41-42; Rice 1987:238). Corrugated types primarily Mancos

Corrugated and Dolores Corrugated in sequence tend to dominate gray ware
collections following their introduction until the end of the period (Lipe and
Varien 1999a:261).
Based on sherds that can be sourced, the occurrence of non-local gray
wares indicates a trend toward increased exchange through time during the Pueblo
II period in the northern San Juan area, rising from two percent of sherds at the
beginning of the period to seven percent by A.D. 980 (Blinman and Wilson
1993:79). Non-local white wares generally occur in higher frequencies throughout
the region and further indicate increased exchange activity through time, rising
from seven percent of sherds at the beginning of the period to 20 percent by A.D.
1100 (Blinman and Wilson 1993:80). Red ware frequency continues to decline
early in the period, occurring at only about half the rate of abundance noted for the
late Pueblo I period; San Juan Red Ware production, and by extension, acquisition
of this type by northern Anasazi groups, appears to cease entirely by A.D. 1100
(Blinman and Wilson 1993:79-80). Notably, great house sites in the area show
evidence of higher rates of long-distance exchange than other Pueblo II period
sites (Lipe and Varien 1999a:258-259). Sites such as Chimney Rock, Lowry Ruin,
Escalante Ruin, and the Parking Lot site contain occurrences of Tsegi Orange
Ware and Tusayan White Ware manufactured in the Kayenta area of northeastern
Arizona and lesser amounts of White Mountain Red Ware from the Cibola region
of west-central New Mexico (Eddy 1977; Hallasi 1979; Martin 1936).

The nature of northern Anasazi connections to the Chacoan network as
expressed in ceramic assemblages is somewhat ambiguous. Great house sites do,
however, exhibit a relatively high frequency of exchange goods when compared to
contemporary non-outlier sites (Lipe and Varien 1999a:259). This suggests the
presence of a form of social differentiation, signified by the occurrence of great
houses with presumably restricted access that was not present in previous periods.
The mechanics of this apparent social stratification and its impact on the lives of
Pueblo II period occupants of southwest Colorado is still poorly understood, and
the significance of regional exchange is as of yet unclear. The period exhibits
abundant material evidence for extraregional interaction, but the questions of how
such material arrived and how it functioned in community social organization
remain open to further research (Lipe and Varien 1999a:288-289). Chacoan
influence in the region largely dissipates by the end of the Pueblo II period. This
development curtails participation on the part of northern San Juan area groups in
widespread regional exchange and sets the stage for reorganization of social
structure associated with the growth and fluorescence of the Mesa Verde culture
area that characterize the ensuing Puebloan period.
Pueblo III: A.D. 1150 to 1300
Due in part to the highly visible and often spectacular nature of the cliff
dwellings constructed during Pueblo III, this period has been the focus of intensive

investigation since the earliest days of Southwestern archaeology and has played a
large role in forming the popular conception of the prehistoric southwest globally.
Along with distinctive and prevalent above-ground architecture, a robust body of
previously conducted research and a highly refined tree ring sequence have
permitted archaeologists to pursue an array of anthropological problems
encompassing environment, demography, and social organization. As in previous
sections, only a cursory overview of the myriad complexities of the period is
presented here. The paramount, overarching themes that have occupied the
attention of archaeologists and which characterize the period as a whole include
population growth, aggregation, and the eventual abandonment of the region
(Stone 1999:106).
Habitation sites from early Pueblo III (A.D. 1150-1225) in southwest
Colorado are generally small and widely dispersed between community centers
situated mainly on agriculturally viable mesa tops along tributaries of Montezuma
Creek or the Mancos River, yet settlement patterns indicate a trend toward
residential aggregation through time (Lipe and Varian 1999b:292-300). As with
the Pueblo II period, there is a scarcity of absolute dates for the earliest Pueblo III
sites, but an increase in available tree ring dates begins around A.D. 1170 and rises
substantially at about A.D. 1200 (Lipe and Varian 1999b:292). While the
apparent hiccough in the dendrochronological sequence could possibly be
attributed to site formation variables such as reuse and preservation issues, the

number of Pueblo III settlements in southwest Colorado clearly rises through the
A.D. 1180s and 1190s, apparently due to rapid increases in population (Lipe and
Varian 1999b:292-300). Chaco-style great houses that were built during the
previous period are centrally located at some of the early Pueblo III communities,
but the question of whether these features were continually occupied, reoccupied,
or used at all during this time in the northern San Juan area remains largely open,
and the degree to which the great houses were utilized, if at all, appears to vary by
site (Lekson and Cameron 1995; Lipe and Varian 1999b:300). Generally speaking,
the habitation sites that comprised the dispersed early Pueblo III communities are
similar in form to those from previous Puebloan periods, consisting of small room
blocks, associated small kivas and middens (Lipe and Varian 1999b:300; Stone
Some important changes in settlement patterns take place during the late
Pueblo III (A.D. 1225-1300) period. Firstly, the tendency for residential site
location on mesa tops gives way to a marked preference for canyon environments
(Varien 1997:177). In southwest Colorado these canyon-oriented centers often
take the form of cliff dwellings such as Mesa Verdes Cliff Palace and Mug House,
though in areas adjacent to those considered in this thesis, settlements tend to be
situated at canyon heads with structures on both canyon rims and below the rim in
rock shelters or on talus slopes such as those at Sand Canyon Pueblo and many of
the sites within Hovenweep National Monument (Lipe and Varian 1999b:303-312).

In either case, large complexes of the period are consistently located near reliable
springs and incorporate water control features such as agricultural terraces, check
dam systems and reservoirs (Cassells 1997:157; Lipe and Varian 1999b:347;
Stone 1999:106).
The growth of large tightly aggregated villages, which are occupied by the
majority of the areas population by around A.D. 1250, occurs concomitantly with
the shift from mesa top to canyon settings (Lipe and Varian 1999b: 303; Varien
1997:177). Proposed explanations for Pueblo III aggregation are generally similar
to those proffered for the Pueblo I period and place varying degrees of emphasis
on factors such as competition and cooperation as related to resource availability
(see Schlanger 1983, Wilshusen 1991). Since any satisfactory explanation of
aggregation must consider multiple, interacting social, economic, and
environmental variables, they are necessarily complex and fall beyond the scope
of this of this work. It is clear, however, as indicated by the rapid and substantial
changes in settlement patterns during Pueblo III that the factors contributing to
aggregation were in flux. The growth that started in the 1180s culminated with
populations approaching 30,000 individuals in southwestern Colorado by the mid-
thirteenth century (Rohn 1989; Wilshusen 1996).
Population levels begin to decline by about A.D. 1270, and the area was
nearly or entirely depopulated by about A.D. 1290 (Lipe and Varian 1999b:312).
The possible reasons for such rapid fall off in the areas population have long

formed one of the central questions of Southwestern archaeology. The
abandonment of the region clearly corresponds to a period of drought from A.D.
1276 to 1299, as inferred through tree ring records (Douglas 1929), but as
archaeologists have further explored the question of regional abandonment, the
great drought itself has lost much of its explanatory power. It appears unlikely
that drought alone was sufficient to account for the total abandonment of the
region. More recent studies (e.g. Van West 1994) indicate that southwestern
Colorado as a whole could have continued to support large populations practicing
maize agriculture during drought years despite resultant crop failure that would
have been experienced in some particular communities. While environmental
stress certainly played a role in thirteenth century emigration, satisfactory models
of regional abandonment must also consider additional internal factors, such as
violent conflict (see LeBlanc 1999, Turner and Turner 1999), as well as external
pull factors with the potential to make other areas attractive alternatives.
Though logistical details are still poorly understood, the destination of most
immigrants from the northern San Juan area was presumably south, to the northern
Rio Grande (Lipe and Varian 1999b:339; Stone 1999:107). When compared to
the increasingly marginal environment and potentially unstable social atmosphere
of southwestern Colorado, the northern Rio Grande possessed numerous attractive
qualities, such as more reliable climatic patterns conducive to agriculture as well
as developing new regional systems with the potential to offer greater economic

opportunity and social stability (Lekson and Cameron 1995; Lipe and Varian
1999b:341-343; Stone 1999: 108-109). By the close of the thirteenth century the
Mesa Verde region was unoccupied by agriculturalists and would remain largely
devoid of permanent settlements for hundreds of years.
The social organization underlying such dynamic demographic shifts and
associated settlement systems appears to have been relatively dispersed when
compared to the preceding period. The collapse of the Chacoan regional system
seems to have been followed by to a general lack of centralized, hierarchal
organization at both regional and local levels among Pueblo III period occupants
of southwestern Colorado. Rank-size analysis based on data from on 81
settlements of the Mesa Verde culture area indicates a loosely integrated system
rather than one centered on a particular locus of dominance (Lipe 1994 cited in
Lipe and Varian 1999b:331). The absence of a centralized regional polity persists
throughout the period, and power appears to become further decentralized as
aggregation intensifies during the second half of the thirteenth century (Lipe and
Varian 1999b:334).
While some data suggest social differentiation within communities,
evidence of hierarchal status at the community level during the late Pueblo III
period is somewhat ambiguous. Much of the available information regarding
Pueblo III social complexity in southwest Colorado has been informed by research
conducted by various researchers through the Crow Canyon Archaeological

Center in the Sand Canyon area. This includes the aforementioned Sand Canyon
Pueblo, a site containing over 500 structures, Castle Rock Pueblo, a nearby
contemporaneous site with about 75 structures, and numerous smaller, associated
sites. The most compelling evidence for centralized control at the community
level is in the form of probable storage facilities associated with public
architecture at Sand Canyon Pueblo. A total of over 40 rooms may have been
dedicated to food storage equivalent to the storage capacity of 5 to 10 habitation
units (Lipe and Varian 1999b:335-336). The distribution of faunal remains, which
exhibits high frequencies of deer at Sand Canyon Pueblo and much lower
frequencies at nearby smaller sites, has also been interpreted as potentially
indicative of hierarchal social structure (Driver 1993 cited in Lipe and Varian
1999b:337). Despite these suggestive data, when assessing the research conducted
by Crow Canyon during the 1990s, Lipe and Varian (1999b:335) find that,
overall, the evidence from the Sand Canyon locality is equivocal with regard to
whether there were individuals or groups who enjoyed high status and had control
of economic, political, or religious sources of power. This assessment is based,
in part, on a dearth of exotic materials and the scarcity of other artifacts indicative
of wealth or status. Additional indicators of differential status such as
discrepancies in labor investment as exhibited in residential architecture or
elaborate funerary offerings are also absent (Lipe and Varian 1999b:336-338).

A relatively egalitarian form of Pueblo III social organization in southwest
Colorado is also reflected in the periods pottery. Wilson and Blinman (1991 cited
in Lipe and Varian 1999b:315-316) recognize the following three temporal-
stylistic complexes for the northern San Juan area during the Pueblo III period:
A.D. 1140-1180Dolores Corrugated is the dominant corrugated type, with some Mesa
Verde Corrugated and only traces of Mancos Corrugated. McElmo Black-on-white is the
dominant white ware type to the near exclusion of Mancos Black-on-white. Red wares
are rare, are dominated by Tsegi Orange Ware polychromes, and may include small
amounts of White Mountain Red ware sherds.
A.D. 1180-1225Both Dolores Corrugated and Mesa Verde Corrugated are abundant,
and white ware sherds include an equal mixture of McElmo Black-on-white and Mesa
Verde Black-on-white sherds. Red ware sherds are scarce to absent, and when present
are limited to White Mountain Red wares.
A.D. 1225-1300Mesa Verde Corrugated sherds are more abundant than Dolores
Corrugated. Mesa Verde Black-on-white is more abundant than McElmo Black-on-
white, and the former type increases in abundance toward the end of the period. Red
wares are usually absent, but if present they consist of White Mountain Red wares.
These trends indicate that, as might be expected in an area undergoing
depopulation, extraregional trade among the Anasazi of the northern San Juan was
in persistent decline during the Pueblo III period. Moreover, remaining exchange
networks were becoming increasingly unidirectional as the period progressed
toward its terminal years.
Overall, the picture that emerges of northern Anasazi exchange and
interaction during all Puebloan periods considered here is remarkably complex
and encompasses a number of interrelated local and regional patterns.
Interpretations of how the distributions of ceramic artifacts speak to Anasazi social
structure and spatial boundaries vary. Based on analysis of artifacts from sites on

Black Mesa in Kayenta area of northeastern Arizona, Stephen Plog (1989) sees
potential support for centralized control of widespread exchange networks
associated with differential control of ritual centers; Plog further suggests that the
reason such patterns may have been previously overlooked by archaeologists is
that these forms of interaction simply dont conform to existing conceptions of
prehistoric Southwestern exchange as reciprocal and occurring largely at the
household level. Blinman and Wilson (1993), on the other hand, interpret the high
intersite variability and spatially widespread stylistic similarities among the
northern Anasazi as indicative of multiple independent and overlapping exchange
networks characterized by generally diffuse communication and interaction rather
than differentiation rooted in political or social factors.
The San Rafael Fremont
North of the areas discussed in the preceding sections of this chapter,
cultural affiliation, development, and trajectory are understood with far less clarity
than that provided by the collective opus of Anasazi research. This is in part a
result, no doubt, of the relatively limited amount of time and effort devoted by
archaeologists to understanding such problems north of the Anasazi homeland.
The relative dearth of studies pertaining to groups of the far northern Colorado
Plateau and adjacent portions of the Great Basin can be partially attributed to the
comparative lack of cultural material; architecture is relatively scarce here, and

sites are generally smaller and less densely distributed. The nomenclature applied
by early researchers has probably contributed to a persistent bias in archaeological
research as well; as Richard K. Talbot (2000) among others has noted, Kidders
(1924) designation of the area as the Northern Peripheral District contains the
implication that the archaeology here may be only peripherally interesting.
Nonetheless, extensive research has been conducted in these areas over the years,
and as applied to what is referred to herein as the Formative era, the bulk of such
research has fallen under the category of Fremont studies.
The term Fremont was first applied by Morss (1931) to describe a series of
observed traits in sites along Utahs Fremont River which were considered to be
distinct yet similar to traits observed within Anasazi sites to the south. Subsequent
researchers applied this trait list approach, with modification, over an expanding
geographic area, consistently defining the Fremont presence in opposition to the
Anasazi (Madsen 1977; Madsen 1979; Talbot 2000). The emphasis on
distinctions from Anasazi groups rather than shared similarities increased during
the 1950s when some researchers redefined Fremont as a local development
stemming from the indigenous, archaic period Desert Culture (Jennings and
Norbeck 1955; Wormington 1955). Throughout the ensuing decades these two
primary and contrasting models of Fremont origins have persisted, with some
researchers (e.g. Gunnerson 1969; Berry 1975) emphasizing Fremont-Puebloan

connections and others (e.g. Aikens 1970, 1972; Marwitt 1970) arguing for the
validity of the Desert Archaic model.
Material culture observed at Fremont sites has done little to foster
agreement on Fremont origins one way or the other. In an attempt to organize a
comprehensive trait list for the Fremont as a whole utilizing 115 designated traits,
Ambler (1970:7) finds:
There are...several drawbacks to both this general approach and the specific applications
of it to the Fremont. Many traits listed are so generalized (e.g. maize agriculture) as to be
useless in distinguishing Fremont from Anasazi, although they are traits useful in
separating Fremont from contemporaneous groups to the west, north and east. Other
traits, such as bone gaming pieces, tubular pipes, and trough metates, although
considered to be major traits of the Fremont, are also found among other prehistoric
groups, although perhaps not at the same time or in the same quantity as among the
Fremont. One of the principal difficulties in the use of trait lists to define the Fremont is
the problem that there are actually rather few distinctive and typical traits that are
found over the entire area usually considered to be Fremont.
The persistent inconsistency in cultural material and derived inferences regarding
behavior has led Madsen (1979:713) to assert that Fremont, not a
shared artifact tradition. Over time, the assorted groups initially referred to as
Fremont have come to be collectively defined by variability as much as
anything else, leading some (Madsen 1979) to reject the use of the term as applied
to groups who inhabited the eastern Great Basin in preference of Sevier Culture
to distinguish these eastern groups, who relied largely on marshland foraging
supplemented by com agriculture, from those of the western Great Basin and
Colorado Plateau who apparently relied more heavily on agriculture for
subsistence. Other researchers (i.e. Marwitt 1970) have proposed models defining

various Fremont variants rather than attempt to approach Fremont broadly across
geographic space as archaeologists have become increasingly aware that problems
inherent to the Fremont concept are often associated with the broad expanse over
which the term has been applied.
The most enduring classificatory schemes are often premised on Marwitts
(1970) designation of five variants of the Fremont culture, which has provided a
general regional and temporal framework for Fremont studies that is still largely
accepted by many researchers. Based on the presence, absence, and relative
frequency of material culture, namely ceramics, these variants consist of: the
Parowan (A.D. 450 1250) centered in southwest Utah, the Sevier (A.D. 780 -
1260) of west-central Utah and east-central Nevada, the Great Salt Lake (A.D. 400
- 1350) of northern Utah, the Uinta (A.D. 650 950) in northeast Utah and
northwest Colorado, and the San Rafael (A.D. 700-1250) of east-central Utah and
west-central Colorado. Despite the utility and endurance of Marwitts
classification, the appropriate scale of analysis in Fremont studies continues to be
a major point of contention among various researchers. For the regional approach
applied in this thesis, Marwitts designations are suitable; and as previously stated,
the San Rafael variant is considered most germane. And while the San Rafael
variant serves as the focus of the this somewhat abbreviated discussion of Fremont
culture, the possible influence of other groups, particularly the Uinta Fremont (due

to the relatively close proximity of their geographic locale north of the study area),
cannot be entirely dismissed.
Academic contention aside, numerous material traits have been identified
and generally accepted as distinctive characteristics of Fremont material culture as
a whole. Bone and stone gaming pieces, stone balls, deer-hock moccasins (Aikens
1970:103-109), one-rod-and-bundle basketry (Madsen 1989), and unfired
trapezoidal figurines (Jennings 1978) are viewed as uniquely Fremont, as are
certain rock art styles featuring trapezoidal anthropomorphic motifs found
throughout Utah (Schaafsma 1971) and northwestern Colorado (Cole 1990:172-
222 ). The bow and arrow appear to have been used throughout areas of Fremont
occupation (Coltrain and Leavitt 2002:454), but their adoption too appears to vary
spatially and temporally throughout the Great Basin and Colorado Plateau
(Madsen and Simms 1998:260).
Architectural style and technique is also variable through space and time
amongst Fremont sites, but common architectural features may include granaries,
pithouses, and multi-room surface structures exhibiting mud and pole, adobe, or
stone masonry construction techniques (Barlow 2002; Jennings 1978; Marwitt
1986). Early Fremont architecture is typically represented by circular, basin-
shaped pithouses featuring conical roofs and associated underground storage units,
stylistically similar to Basketmaker II pithouses. Between A.D. 900 and A.D.
1100, there is a perceptible shift in architectural style from these circular structures

to rectangular and subrectangular pit structures. Though pithouses are presumed
to have continued serving as primary residences, at some point after A.D. 1000
there appears to be what has been described as, the beginning of a pithouse-to-
pueblo transition of sorts (Talbot 2000:282), including the appearance of above-
ground room blocks characterized by extensive use of stone masonry, plastered
interior walls and slab-lined hearths (Jennings 1978: 184-187; Marwitt 1986:170)
in the San Rafael Fremont area that are similar in style to those characteristic of
Pueblo I and Pueblo II periods (Jennings 1978; Marwitt 1986; McMahon 2000;
Talbot 2000).
Fremont ceramic assemblages are dominated by undecorated gray wares
produced through coil and scrape techniques. Madsen (1977) has grouped
Fremont pottery into a two-tiered classification system based on method of
manufacture, surface treatment, and temper material that includes three ceramic
traditions Desert Gray, Promontory Gray, and Ivie Creek Black-on-white and
numerous subsequent types (Madsen 1977). Though there have been recent
efforts (focusing primarily on the Parowan Fremont) to reclassify Fremont
ceramics into a type-series-ware hierarchy as is conventional in Southwestern
ceramic analysis (Watkins 2006), Madsens (1977) system of classification
remains predominant in Fremont ceramic studies. Only Desert Gray ware,
including Emery Gray (smoothed and to a lesser extent, corrugated varieties) and
Uinta Gray, is reported to occur with any significant frequency within the study

area (Crane 1977; McMahon 2000). Temper material is variable, but is primarily
crushed igneous rock in Emery Gray wares while calcite is used almost
exclusively in Uinta Gray wares (Gunnerson 1960; Madsen 1977; Watkins 2006).
Jars are the most common vessel form in San Rafael Fremont assemblages though
bowls and pitchers are occasionally found; while paint is absent, vessels are
sometimes decorated with applique, punched, or incised surface treatments
(Madsen 1977:31-33). Reported occurrences of Desert Gray wares within the
study area are quite rare, and recent research (Greubel et al. 2006; Reed and
Metcalf 1999: 111) suggests, as discussed below, that these thirty-year-old
designations may in fact be erroneous.
The degree to which the Fremont relied on agriculture also remains
questionable. It is generally accepted that the Fremont pursued a mixed economy
with both cultigens and locally available wild resources playing large roles in
dietary composition. Debate has largely centered on the relative importance of
domesticates in the Fremont diet (Talbot 2000:279). As mentioned above,
Madsen (1979) has argued that western Great Basin groups were largely
dependent on locally available marsh resources while groups occupying the
eastern edge of the great basin and the Colorado Plateau were considerably more
reliant on agricultural production. Berry (1974) suggests a high level of
dependence on maize among the Parowan supplemented with wild resource
procurement according to the relative success of crops annually. Simms (1986)

model of Fremont adaptive diversity proposes three coexistent strategies,
including: 1) localized foraging to supplement horticultural production, 2) a
variable strategy of periodic shifts from farming to foraging strategies, and 3)
distinct groups of farmers and foragers occupying the same territory. More
recently, Barlow (2002:69) has argued that:
...the archaeological record of the Fremont simply represents a continuum of variation
in the economic importance of maize and the degree to which people inhabited central
places or employed residential mobility to exploit maize fields and wild food
procurement areas.
She concludes that the eventual abandonment of maize agriculture, occurring ca.
A.D. 1300 1350, was contingent on the increased availability of higher-ranked
wild foods, (Barlow 2002:70) rather than the relative success of maize crops as
suggested in previous models.
Regardless of which, if any, of these economic models emerges as
preeminent, two basic points regarding Fremont agriculture are revealed through
the existing literature: Fremont groups did cultivate com to varying degrees for a
prolonged period of time, and maize agriculture among the Fremont never
assumed the central economic position that it did among neighboring Anasazi
populations. Among the eastern Fremont in general, dated com specimens and/or
associated features (Greubel 1996; Talbot and Richens 1996; Wilde and Newman
1989) combined with stable carbon isotope (13C) analysis of human remains
(Coltrain 1996) indicate that maize production was firmly in place by the first
century A.D. And it has been suggested that due to their association with large

quantities of corn, the presence of anthropomorphic figurines a hallmark of the
San Rafael Fremont may be associated with ceremonial activity related to maize
agriculture (Marwitt 1978:187-206), though the hypothesis has not been subjected
to rigorous testing. So while emerging data are suggestive and continue to
contribute to a more complete understanding of the role of agriculture in Fremont
culture, the implications of agriculture on Fremont social structure remain unclear.
As with other elements that define archaeologists current understanding
of the Fremont, ideas regarding social organization have undergone revision
through time as additional data are revealed. Conventionally held views that
hypothesize Fremont groups as typically occupying small, politically isolated
communities of fewer than 25 30 residents (e.g. Jennings 1978; Sammons-Lohse
1981), may prove to be oversimplified in light of recent research. Sites in Clear
Creek Canyon (Janetski and Talbot 1995) and the Nawthis (Metcalfe 1984) site of
central Utah, for instance, each contain localities with over 35 habitation structures
- many of which were contemporaneously occupied situated in spatial
arrangements that suggest integrative sociopolitical structure. As with Fremont
material culture however, evidence for complex social structure beyond the
household level is not consistent across the Fremont region.
Permanent, multi-room habitation sites potentially indicative of decision
making beyond the household level within the San Rafael Fremont area include
Turner Look (Wormington 1955), Innocents Ridge, (Schroedl and Hogan 1975),

Crescent Ridge, Power Pole Knoll, Windy Ridge Village, the Bull Creek Sites
(Jennings and Sammons-Lohse 1981), and the Nine Mile Canyon locality (Gillin
1938). These sites are typically located on low ridges in close proximity to water
sources and arable land (Marwitt 1986:170). Within the study area, the Paradox I
site in western Montrose County, Colorado, appears to possess a Fremont
component represented by multiple pit structures and storage units associated with
Fremont ceramics (Kasper 1977; McMahon 2000; Reed and Metcalf 1999:111-
112). It is noteworthy that thus far Paradox I represents the only site in western
Montrose County where the San Rafael Fremont designation is accepted with
anything approaching consensus among current researchers. While McMahon
(2000) has argued in favor of retention of the Fremont nomenclature for other
architectural sites in the area, in the absence of modem, controlled excavations, he
does so with reservations.
Reed and Metcalf (1999:113,121-122) acknowledge Fremont occupation
in the Glade Park area on the northern Uncompahgre Plateau west of Grand
Junction based on the presence of identified Fremont rock art motifs of the Sieber
Canyon Style that have been associated with the Uinta Fremont (Cole 1990:173).
Despite the presence of some characteristically Fremont architectural elements,
namely circular habitation structures, Reed and Metcalf conclude that, evidence
of a Fremont occupation is western Montrose County is scant and ambiguous,
dismissing the presence of one-rod-and-bundle basketry at the Tabeguache and

Dolores cave sites (Hurst 1942, 1947) as likely representing exchange episodes
rather than evidence of Fremont occupation (1999:111). These authors (Reed and
Metcalf 1999:109) assert that:
In regions where Fremont affiliation is uncontested, such as in northwestern Colorado,
group assignments area more freely made, because the implications of classification
error are less substantial. In areas such as western Montrose County, however, where
cultural affiliation of Formative-era sites is debated, higher standards are
implemented for unit assignments. Open sites in western Montrose County [to be
considered Fremont] must either yield Fremont pottery types or consist of rock art
motifs widely accepted as Fremont.
So while some evidence, seemingly consistent with the ambiguous and variable
definition of Fremont itself, hints at some level of Fremont occupation within the
study area and at least some evidence of exchange between the Formative-era
occupants of this area and Fremont groups has been demonstrated, currently
available data permits only a tenuous designation of Fremont cultural affiliation
for the area at best.
The Gateway Tradition
As Reed and Metcalf (1999:109-112) largely reject a substantial or
prolonged Fremont occupation in west-central Colorado, they also contend (Reed
and Metcalf 1999:107) that the areas archaeology does not support, the existence
of a bona fide Anasazi occupation, noting that:
The degree of cultural continuity characterizing Anasazi culture over broad areas of the
American Southwest makes it untenable to maintain that a diluted form of Anasazi
culture existed such a short distance from the Anasazi homeland, especially considering
similarities in environments.

To occupy the Formative-era void that is seemingly encountered here, Reed (1997)
has proposed the Gateway tradition cultural unit. While this construct made its
initial appearance in the literature over ten years ago and has been utilized in some
regional overviews (Cassells 1997:182-186; Reed and Metcalf 1999:131-140), due
to a general lack of large-scale excavation projects in the area, the concept remains
somewhat nascent. Recent research (Greubel et al. 2006; Reed 2001) has however
generated progress toward assessing the viability of the Gateway tradition and
permitted refinement of its spatial and temporal application.
Reeds 1997 articulation of the Gateway concept stemmed from his belief
that traditional application of the Fremont unit has impeded research in the area by
masking significant variation that can contribute to our understanding of past
human behavior. The absence of key elements of the Anasazi tradition such as
ceramic production, architectural conventions including the use of kivas and
water control structures and evidence of complex interregional relationships
with Chacoan logistical networks is perceived as inconsistent with an established
Anasazi occupation. Reed (1997:23) further asserts that it is unlikely that these
groups are affiliated with the Fremont based on the scarcity of the Fremont
material traits. Noting that previous researchers (Crane 1977; Fetterman and
Honeycutt 1990; Schroeder 1964) had long noted these inconsistencies and
suspected the presence of a culture distinct from the adjacent Anasazi or Fremont,
Reed put forth the Gateway nomenclature as a means of formally recognizing

these groups and avoiding potential masking of variation within the archaeological
record that might occur as a result of inaccurate designations of cultural affiliation.
As originally proposed, Gateways spatial expanse is roughly bounded by
Interstate 70 to the north, the Uncompahgre River to the west, the Green River to
the East, and extending as far southward as Monticello, Utah (Reed 1997:20), an
area (especially those portions within the state of Colorado) which roughly
corresponds to that considered for this thesis. Initially dated from 500 B.C. to A.D.
1250, a period coinciding with evidence for com horticulture in the area (Reed
1997:24), Reed (2005:30) has more recently reconsidered the applicable time span,
suggesting that the term may most meaningfully be restricted to the period roughly
spanning A.D. 900 to 1100. This chronometric refinement is spurred by a
perceived 500-year hiatus in com production for the area commencing around A.D.
400 (Reed 2001; Reed 2005:30). Since one of the predicating attributes of the
Gateway tradition is reliance (albeit limited) on com horticulture, Reed (2005) errs
on the side of caution here with the recently proposed temporal restriction, but he
does not entirely dismiss the possibility that Gateway may be applicable to earlier
dates as new data emerge.
In addition to the above-noted reliance on com horticulture, Gateway is
further defined by the following criteria (Reed 1997:24; Reed and Metcalf
Manufacture of small corner-notched projectile points, such as the Rosegate series.

Procurement through trade small quantities of Anasazi and, much less frequently,
Fremont ceramics. Such trade with the Anasazi may have occurred primarily during the
period between A.D. 900 and 1050.
Apparent lack of ceramic production.
Habitation of circular and rectangular masonry surface structures. In a few cases, rooms
may be contiguous.
Possible habitation of pit structures.
Relatively short-term use of habitation structures, as indicated by shallow middens.
Construction of granaries and storage cists in rockshelters.
Rock art with both Anasazi and Fremont influences.
Though few sites simultaneously exhibit the entire suite of Gateways defining
characteristics, Reed and Metcalf (1999:132-135) consider no less than thirteen
architectural sites and/or localities in Montrose and San Miguel counties to be
representative of the Gateway tradition (Table 3.1). Problems associated with
previous research efforts at all of these sites have been well documented (see
McMahon 2000; Reed and Metcalf 1999:132), and data collection and
interpretation have suffered from an array of problems ranging from now-outdated
methodology, to breeches of archaeological ethics, to loss of data through fire. So
while eleven of the thirteen localities identified by Reed and Metcalf as Gateway
have been excavated, the overall quality of the database is generally poor. If, in
fact, the Gateway tradition proves to retain viability as a descriptive unit, these
localities will presumably signify a representative sample of a more far-reaching

cultural tradition, which will conceivably emerge as additional systematic surveys
and controlled excavations occur within the area.
Architectural developments occurring during the proposed Gateway
occupation of west-central Colorado include the familiar Formative-era pattern of
a shift from pit structures to above ground, round or rectangular masonry surface
structures. Round structures, such as those recorded by Crane (1977:16) at the
Weimar Ranch sites (including 5MN517, 5MN518, 5MN368, 5MN652, 5MN653,
5MN654, 5MN7720, 5MN7721), tend to be single rooms measuring five to six
meters in diameter, possess unlined, interior hearths, and lack remaining evidence
of doorways. Rectangular structures are more often situated in contiguous room
blocks (Reed and Metcalf 1999:138). It has been suggested (Crane 1977:45;
McMahon 2000:14-18) that a progressive shift from round to rectangular
structures associated with increased reliance on agriculture and population growth,
as has been demonstrated elsewhere in the Southwest, occurred in this area. As
Reed and Metcalf (1999:138) note however, precise dating of Gateway occupation
remains problematic, and relative dating based on diagnostic ceramics indicates
that both round and rectangular structures may have been utilized
contemporaneously. Recently obtained radiocarbon dates (Greubel et al. 2006:50,
56) seem to further support the idea of contemporaneous occupation.