User manual | INFORMATION TO USERS

INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the boot
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.
A Bell & Howell Information Company
300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA
313/761-4700 800/521-0600
DATING CONSTRUCTION EVENTS AT GRASSHOPPER PUEBLO:
NEW TECHNIQUES FOR ARCHITECTURAL ANALYSIS
by
Charles Ross Riggs, Jr.
A Thesis Submitted to the Faculty of the
DEPARTMENT OF ANTHROPOLOGY
In Partial Fulfillment of the Requirements
For the Degree of
MASTER OF ARTS
In the Graduate College
THE UNIVERSITY OF ARIZONA
1994
UHI Number: 1361980
UMI Microform 1361980
Copyright 1995, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MX 48103
2
STATEMENT BY AUTHOR
This thesis has been submitted in partial fulfillment of requirements for an
advanced degree at the University of Arizona and is deposited in the University Library to
be made available to borrowers under rules of the Library.
Brief quotations from this thesis are allowable without special permission,
provided that accurate acknowledgment of source is made. Requests for permission for
extended quotation from or reproduction of this manuscript in whole or in part may be
granted by the head of the major department or the Dean of the Graduate College when in
his or her judgment the proposed use of the material is in the interests of scholarship. In
all other instances, however, permission must be obtained from the author.
SIGNED:
APPROVAL BY THESIS DIRECTOR
This thesis has been approved on the date shown below:
i4-&ep^4
J. Jefferson Reid
Professor of Anthropology
Date
3
ACKNOWLEDGMENTS
I owe a great deal of thanks to many people who had a hand in the completion of
this thesis. First of all, I would like to thank the members of my committee. J. Jefferson
Reid, Barbara J. Mills and John W. Olsen all provided useful comments on various drafts
of this work and exhibited a great deal of patience in what, in retrospect, seems like a long
drawn out process. I owe special thanks to Jeff Reid, not only for the construction phase
data, without which this thesis would not have been possible, but also for all of the advice
and support he has given me over the past several years.
I would also like to extend thanks to several people who provided additional
comments upon this work. T. J. Ferguson, Jeffrey S. Dean, and Barbara K. Montgomery
all provided useful criticisms at various stages in the writing process. I also wish to thank
M3 Engineering and Technology Corporation for both training me in the use of the
AutoCAD® program and for providing me with access to the computer equipment on
which much of this study was done. Additional thanks go to all of the staff, students, and
Apache crew members whose efforts at Grasshopper over the years generated the data
upon which this study was based.
Last, but by no means least, I want to offer special thanks to Maria Molina, whose
editorial skills were vital in the overall organization of the final work and whose personal
support and faith in me has gotten me over numerous obstacles, the least of which was this
thesis. And, to my daughter Salina who, although not yet born during much of the work
on this study, was my greatest inspiration and who will remain the guiding force behind all
my future endeavors.
TABLE OF CONTENTS
LIST OF ILLUSTRATIONS
6
LIST OF TABLES
7
ABSTRACT
8
1. INTRODUCTION
Historical Overview
1880 - 1930: Architecture as Artifact
1930 - 1960: Architecture as a Classificatory Device
1960 -1980: Architecture as an Indicator of Social Relationships
1980 - Present: Architecture as Artifact Rediscovered
Discussion
9
11
11
12
13
16
18
2. ARCHITECTURAL ANALYSIS AND COMPUTER TECHNOLOGY
Partitioning of Space
Qualitative Analysis
Quantitative Analysis
Stylistic Characteristics
Engineering and Structural Characteristics
Discussion
20
21
21
23
27
28
29
3. TIME IN ARCHITECTURAL ANALYSIS
Dating Architectural Units
Intrinsic - Relative Dating
Independent - Absolute Dating
Relative Growth and Actual Growth
31
32
32
33
33
4. RE-ASSESSING PUEBLO GROWTH AT GRASSHOPPER
The Construction Phase Model
Grasshopper Construction Phases
Construction Phase I
Construction Phase 2
Construction Phases 3 and 4
Construction Phase 5
Construction Phases 6 and 7
Construction Phases 8 through 14
Discussion
Absolute Dates for Growth
Grasshopper Tree-Ring Dates
38
38
43
44
45
46
48
49
51
57
59
62
5
TABLE OF CONTENTS - continued
Methodology
Problems and Assumptions
Methods
Definitions
Principles
Results
Room Block 1
Room Block 2 and the Great Kiva
Room Block 3 and the Southern Corridor
Discussion
Dating Construction
Relative and Actual Growth
Pueblo Growth
Summary and Conclusions
65
65
66
70
70
71
71
75
79
82
82
83
87
88
APPENDIX A
APPENDIX B
93
97
REFERENCES CITED
106
6
LIST OF ILLUSTRATIONS
FIGURE 1. Schematic Diagram: Relative vs. Actual Growth
FIGURE 2. Grasshopper Pueblo: Site Plan
36
39
FIGURE 3. Construction Phase 1
44
FIGURE 4. Construction Phase 2
45
FIGURE 5. Construction Phase 3
46
FIGURE 6. Construction Phase 4
47
FIGURE 7. Construction Phase 5
48
FIGURE 8. Construction Phase 6
49
FIGURE 9. Construction Phase 7
50
FIGURE 10. Construction Phase 8
51
FIGURE 11. Construction Phase 9
52
FIGURE 12. Construction Phase 10
53
FIGURE 13. Construction Phase 11
54
FIGURE 14. Construction Phase 12
55
FIGURE 15. Construction Phase 13
56
FIGURE 16. Construction Phase 14
57
FIGURE 17. Relative Growth Curves: Room Blocks 1, 2 and 3
58
FIGURE 18. Composite Relative Growth Curve
59
FIGURE 19. Construction Units at Grasshopper Pueblo: Main Ruin
67
FIGURE 20. Dated Rooms and Construction Units at Grasshopper Pueblo
69
FIGURE 21. Schematic Diagram: Illustration of Principles for Dating
72
FIGURE 22. Actual Growth at Grasshopper Pueblo: Main Ruin
77
FIGURE 23. Actual Growth Curves: Room Blocks 1, 2 and 3
84
FIGURE 24. Composite Actual Growth Curve
85
7
LIST OF TABLES
TABLE 1. Sedentary Period House Areas at Snaketown and La Ciudad
24
TABLE 2. Courtyard Groups at Snaketown
25
TABLE 3. Courtyard Groups at La Ciudad
26
TABLE 4. Construction Phase Data
43
TABLE 5. Sampling Chart for Construction Units
60
TABLE 6. Dates of Rooms with Construction Unit Designations
64
8
ABSTRACT
The analysis of architecture as a separate but important class of material culture
has seen a resurgence of interest in archaeology in recent years. However, a body of
analytical techniques equivalent to those used for the analysis of other types of material
culture is still lacking in architectural analysis. Computer aided drafting programs offer
one means of facilitating architectural analyses by providing both an analytical tool as well
as a means of organizing spatial information. Computer techniques are used to combine a
construction phase model with tree-ring dates at Grasshopper Pueblo. In the course of the
analysis, principles for assigning temporal information to undated construction units are
discussed and applied.
Finally, the results of the combination of these two sets of
information are discussed and a slightly revised site chronology is offered.
9
CHAPTER 1
INTRODUCTION
An architectural analysis represents but one important avenue of inquiry into past
human behavior. As in the manufacture of other classes of artifacts, a complex set of
behaviors is involved in the construction of a structure or dwelling. The physical
expression of an architectural form is best conceived of as the product of a number of
variables which can be broadly subdivided into the larger categories of environment,
tradition, social organization, functional considerations, performance characteristics, and
time. While the same can be said about ceramics or any other class of artifact,
architectural spaces contain much more information concerning social relationships and
cultural identity than any other class of material culture. Unlike portable material items,
only architectural space can both physically denote the patterns of daily interaction within
a group while shaping and reinforcing these same interactions (Hegmon 1989; Hillier and
Hanson 1984; Reid and Shimada 1982). Furthermore, while ceramics and other material
items are subject to movement through exchange relationships and can often travel long
distances, architectural forms themselves are permanently fixed upon the landscape. For
this reason they are far better indicators of cultural differences than portable materials.
The division of space created by architectural features represents a three dimensional
blueprint of the social interactions that occurred at a site or within a household. The
architectural print left by a particular culture's attempts to impose order onto daily social
interactions is one of the most distinctive cultural patterns left in the archaeological
record.
The importance of architecture as an independent analytical category has gone in
and out of fashion throughout the history of archaeological inquiry in the Southwest.
Overall, architectural analysis has closely followed the various paradigmatic shifts in
10
archaeology. Currently, the study of architecture has seen a resurgence of interest,
especially with a revival of the notion of architecture as an artifact (Gilman 1987). To
accompany this reassessment, new models and methods of analysis need to be developed
for analyzing architectural phenomena. Along these lines, computerized techniques of
analysis are useful for the study and display of spatial information. These procedures
especially lend themselves to both quantitative analyses of space, as well as to more
qualitative studies, which require that large sites be broken into smaller units of analysis.
In this paper, it is argued that by applying computer based techniques of analysis to
architectural spaces, the current gap in analytical techniques between architectural analysis
and the study of other types of material culture can be eliminated. Several issues related
to architecture are addressed in an attempt to provide new means of looking at the
problem of dating the construction of Grasshopper Pueblo. The immediate purpose of this
study is to apply Graves' (1991) revised cutting date estimates to Reid's (1973)
construction phases at Grasshopper Pueblo. The ultimate goal is to present new
techniques for looking at architectural data.
The first section of this thesis is devoted to presenting the framework for the
current analysis. Chapter 1 provides a brief overview of the history of architectural
inquiry in the American Southwest. I suggest that the current perspective in archaeology,
which views architecture as a separate artifactual class with definable properties of
function, manufacture, and use, is an old notion that has only recently come back into
fashion. The next section (Chapter 2) focuses on the use of computer aided drafting
programs as a means not only for displaying graphic information but also as an analytical
and organizational tool useful in manipulating architectural entities. Chapter 3 presents a
scheme for conceiving of architectural additions as a function of behavior and time. The
third section of this paper examines pueblo construction as a relative sequence of
11
architectural additions upon which absolute temporal information can be added to
reconstruct the actual growth of a community. Finally, these computer techniques and
concepts of pueblo growth are brought to bear upon the problem of testing Reid's
construction phase model at Grasshopper Pueblo through the application of Graves'
revised tree-ring dates.
Historical Overview
Architecture as an analytical category has been overshadowed by ceramic analysis
in past years. In the early years of Southwestern Archaeology, considerations of ceramics
and architecture were viewed as equal in most investigations. However, early ceramic
studies demonstrated the utility of pottery as an indicator of time and long range social
relationships (Colton and Hargrave 1937; Kidder and Shepard 1936), and ceramic analysis
began to take precedence over architectural analyses to such an extent that by 1936, Roys
was already calling for a renewed interest in architecture as an important analytical
category. While the usefulness of ceramic analyses is not questioned, it is noted that a
great deal has already been accomplished in the development of sophisticated analytical
techniques. Currently, pottery is being studied on the sub-molecular level through various
techniques that attempt to break the materials used in pot sherds down to their individual
elements (see Rice 1987). In direct contrast, studies of architecture have scarcely evolved
to a stage of sophistication analogous to the ware level of ceramic classification.
1880 - 1930: Architecture as Artifact
Early in the history of southwestern archaeology, architecture was seen as but one
component in an overall complex of cultural traits (Cushing 1890; Fewkes 1907, 1912;
Haury 1936; Kidder 1927; Mindeleff 1891). Early analyses focused on architecture as a
12
separate artifact class, to be treated on the same level with other types of material culture.
The classic early study of this type was MindelefFs work on the architecture of Tusayan
and Cibola (Mindeleff 1891). This early analysis was a detailed description of both
prehistoric and historic architecture in the Hopi and Zuni areas. Taking a broad based
approach, Mindeleff examined every aspect of architectural form, from the smallest
architectural features to the rites and traditions surrounding architectural construction and
use. The importance Mindeleff placed on architectural remains was not unique to the
period. It can be argued that architecture in the early days of investigations in the
Southwest had a much more prominent place in the archaeological literature. In Fewkes'
work at Casa Grande and at Mesa Verde, for example, the site reports are largely
architectural in their focus. In fact, Fewkes habitually devotes a great deal of attention to
architectural descriptions, while all the "minor antiquities" are discussed in very little detail
(Fewkes 1911, 1912). Clearly, much of the architectural emphasis of these early studies
was based on the fact that most of what was available for study at the time consisted of
architecture while the ceramic collections in this period were not yet well understood by
many researchers.
1930 - I960: Architecture as a Classificatory Device
Architecture has long been one of the defining characteristics of cultural affiliations
and periodization schemes in the Southwest. As a result of the first Pecos Conference in
1927, architecture was put forth as being a major indicator of cultural change from the
Basketmaker tradition to the Pueblo builders (Kidder 1927). The Pecos classification was
in large part based upon a sequence of change in dwelling types, and its emphasis on
architecture as a classificatory device helped to reduce architecture to a marker of cultural
affiliation and change. Several studies from this period testify to the place of architecture
13
as an indicator of cultural association and development (Haury 1928, 1936, 1985; Reed
1948, 1956; Roberts 1935; Wheat 1955).
By 1936, architecture had lost much of its prestige as an individual artifact class.
The level of neglect is indicated by Roys' (1936:115-142) discussion of the architecture of
the Lowry Ruin, in which he points to the gap between ceramic analysis and architectural
analysis and calls for a better way of classifying, describing, recording, and sorting
architectural data . Roys argues that architecture ought to be viewed as any other class of
material item and that like ceramics, an analysis of masonry types can indicate not only
cultural history but also shed light on "the mental traits of the builders" (Roys 1936:116).
Despite the call by Roys for southwestern archaeologists to be more aware of architectural
characteristics, the situation went largely unchanged until the rise of processual
archaeology in the early 1960's, when architecture came to play a new role in
southwestern archaeology.
1960 -1980: Architecture as an Indicator of Social Relationships
Foreshadowed by the earlier work of investigators such as Prudden (1903, 1918)
and Steward (1937) and, coinciding in part with the coming of Processual Archaeology,
several studies became available that set the stage for later investigation of architectural
forms by processualists. As a result of the work of Bullard at the Cerro Colorado site
(Bullard 1962) and Rohn's work at Mug House (Rohn 1965, 1971), architecture began to
be recognized as an important analytical category once again. It was thought that
individual households and discrete intrasocietal divisions could be identified on the basis of
an architectural analysis (Dean 1969; Rohn 1965, 1971; Vivian 1970; Wilcox 1975).
With the coming of processual archaeology, and its focus on ecological and social
theory (Binford 1962; Willey and Sabloff 1993), architecture became the object of a
resurgence of interest in archaeological investigations. Dean's (1969) analysis of Betatakin
14
and Kiet Siel remains the standard by which all subsequent architectural analyses have
been judged. Dean's reconstruction of the growth of Betatakin and Kiet Siel incorporates
the concepts of social theory that were important at the time while maintaining the
perspective that architecture was an artifact constructed and maintained for human
occupation. Considerations of construction techniques, use of construction wood, and
intrasite processes of growth were all important to the study. Dean's analysis not only
demonstrates what can be done with well preserved tree-ring material and good site
preservation, but also how social groups can be identified and discussed through the use of
detailed architectural analysis. The classification of architectural groups became an
important concept in the work of subsequent analyses in other areas. Dean's investigations
in Tsegi Canyon and, to a lesser extent, the work of Rohn at Mug House were influential
to the excavation strategy at Grasshopper Pueblo in the late 1960s and early 1970s.
The cornering project and subsequent growth project instituted by the University
of Arizona Archaeological Field School at Grasshopper was a direct outgrowth of the
excavation of Broken K Pueblo and the later work of Dean at Tsegi Canyon. Beginning in
1967, the project was initially directed toward generating a site map and developing a
typology of room types as had been done at Broken K (Longacre and Reid 1974). To this
end, opposing room corners were excavated to facilitate mapping and estimation of room
size (Longacre and Reid 1974; Wilcox 1982). As the cornering project progressed, the
focus shifted toward identifying sets of room additions, perceived as representing social
groups similar to those documented by Dean and Rohn. In 1969, the cornering operations
came under the direction of David Wilcox who formalized procedures into the "Cornering
Project." The identification of "construction units" became the central goal of the
cornering project and, based on the idea that new architectural units which are added to
existing units are always abutted to them, the interpretation of bond-abut relationships
15
became important (Wilcox 1975, 1982). Later, due to some of the problems inherent in
the interpretation of corner relationships, and following the work of Rinaldo (Rinaldo
1964; see also Dean 1969) at Carter Ranch, wall face analysis (Reid 1973; Reid and
Shimada 1982) was also adopted as a means of identifying architectural additions (Reid
and Shimada 1982).
The data recovered by the cornering project spawned the "Growth Project," which
began in 1971 under Reid's direction to measure the rate and process of pueblo growth.
With the major construction units identified by the cornering project, rooms in the earliest
(core) construction units were excavated to obtain datable tree-ring samples to date the
founding of the pueblo (Longacre and Reid 1974:22). The construction units were
classified into sets of typological contemporaneity (see below) called "construction
phases" as part of the attempt to measure site growth (Reid 1973; Reid and Shimada
1982). Through 1972 and 1973 the cornering and growth projects continued, and moved
out of the room blocks of the main pueblo and into the outliers. The investigation of the
outliers by the cornering project had determined them to represent largely jacal structures
with masonry foundations, while investigations by the growth project found them to be
constructed slightly later than the main pueblo (Longacre and Graves 1982; Longacre and
Reid 1974; Reid 1973, 1989). The identification of social groups through the
establishment of construction units guided much of the excavation strategy at Grasshopper
in the late 1960s and early 1970s. The material and reconstructions generated by these
analyses form the primary data base for the present analysis as well for reconstructions of
the growth and abandonment of the community (Reid 1973; Reid and Shimada 1982).
The methods and models developed for identifying social groups through
architectural relationships in the puebloan Southwest were later influential to work in the
southern desert region. In the Hohokam area, investigators did not begin to formulate
16
methods for recognizing social groups until the late 1970s and early 1980s. In Wilcox's
reanalysis of Snaketown, he identified several clusters of pithouses that he referred to as
"courtyard groups" (Wilcox et al. 1981). These assemblages of pithouses arranged around
central courtyard areas were conceived of as representing a cooperative social unit
comparable to those documented at Betatakin, Kiet Siel (Dean 1969), and Mug House
(Rohn 1971). Since courtyard groups were identified at Snaketown, they have been found
at several Hohokam sites (Doelle 1985; Doelle et al. 1987; Henderson 1987; Sires 1983,
1987), and are thought to represent an important part of the Sedentary to Classic
transition (Sires 1987; Wilcox et al. 1981). As a result of this recent work, analysis of
Hohokam domestic architecture is now at approximately the same juncture that puebloan
architecture reached in the early seventies. In contrast, when considering Hohokam public
architecture, archaeologists have come considerably farther in looking at architecture as an
artifact.
1980 - Present: Architecture as Artifact Rediscovered
Hohokam archaeologists began to take a more artifactual view of architecture
beginning in the late 1970s. Wilcox and Shenk's (1977) reanalysis of the architecture of
Casa Grande is but one example of this refocused approach. The intent of the reanalysis
was to obtain previously uncollected information about various aspects of Casa Grande
construction and use. The analysis includes considerations of structural stability,
construction techniques, bond-abut patterns, as well as functional and social aspects of the
Casa Grande.
The most comprehensive study that considers the artifactual nature of Hohokam
public architecture is Gregory's (1987) analysis of Classic Period Hohokam platform
mound sites. This analysis views site structure as "the basic patterns of organization
17
imposed by a population on the form of their settlement" (Gregory 1987:184). In other
words, site structure is viewed, as with any other artifact, as the product of human
behavior. To Gregory, Hohokam platform mounds are viewed as an artifact class that can
be discussed in terms of the techniques of manufacture, use, and associated behaviors.
The study incorporates a developmental sequence of platform mound construction as a
means of determining the processes of construction and mound use. Finally, the
association of various architectural features such as platform mounds, compound walls,
and mound-top structures is used in combination with considerations of community layout
as an association of "artifacts" to define a typical Classic period Hohokam community.
Although there is a long standing tradition of architectural analysis in Chaco
Canyon, the 1980's saw an intensification of this trend. Architecture has always been the
central focus of research in Chaco Canyon due to both the architectural presence of the
ruins as well as the paucity of other artifact categories. Chacoan sites both in Chaco
Canyon itself and throughout the San Juan Basin have been traditionally defined by
architectural characteristics such as the large size of the ruins, core and veneer masonry,
and their symmetrical layouts (Judge 1991; Lekson 1991; Vivian 1970). Additionally,
Chacoan communities have also been traditionally defined by the presence of architectural
types. A "typical" Chacoan community consists of several unit houses clustered around a
central great house and great kiva (Lekson 1991:32). On the regional scale as well, it is
architecture that defines the Chaco system as a whole (Lekson 1992; Lekson et al. 1988;
Powers et al. 1983). The widespread similarities of architectural form throughout the
eastern Anasazi region has spawned a number of regional models based largely on
architectural similarities (see Vivian 1990 for an overview of the various models).
Probably the best known and most complete architectural analysis in recent years is
Lekson's discussion of Chacoan great house architecture (Lekson 1984). Analogous to
18
Mindeleffs study a century before, Lekson's analysis of Chacoan architecture takes a
more holistic view. Construction techniques are considered in addition to discussions of
form and the place of the architecture in the broader context of the Chacoan phenomenon.
Broadening the discussion of architecture as artifact to include Grasshopper
Pueblo and the current analysis, it is necessary to take the discussion one step further. For
this study, not only is Grasshopper Pueblo as a whole seen as an artifact with all of the
accompanying properties, but each construction unit is also viewed as an individual
artifact. While the models and methods of pueblo growth developed in the 1960's and
1970's (see above) are critical to the application of dates to Grasshopper construction
phases, the analysis could not have been conducted without viewing Grasshopper as an
assemblage of separate yet related artifacts. Individual construction units are as much a
product of human behavior as any other artifact and, as such, are subject to the same sorts
of theories and methods as any other artifact class. This conceptualization allows the
construction units to be classified by means of phases of construction (Chapter 4) as well
as in terms of individual artifacts with measurable properties of manufacture and use.
Discussion
The concept of architecture as an important analytical category has been back in
fashion for over ten years now (Gilman 1987; Hunter-Anderson 1977; McGuire and
Schiffer 1983). However, architecture as an artifact has yet to be subjected to the same
sorts of rigorous theories and tests as the other categories of archaeological data. While
comprehensive summaries of techniques and theory for both ceramic analysis (Rice 1987)
and lithic analysis (Jelinek 1976) are available, comparable material focusing on the
analysis of prehistoric architecture remains scattered and incomplete. With the reemergence of the idea of architecture as artifact, it is now necessary to develop more
19
sophisticated techniques to look at architecture as a separate analytical category.
Structural spaces are a part of the constructed environment in which human beings interact
on a daily basis. Because of this, the idea of an architectural unit with measurable
properties and definable techniques of manufacture ought to pervade all stages of field
work and data analysis, and be the guiding principle in any study of architectural form.
The artifactual nature of architecture is a result of the same set of behaviors that produce
other types of material culture. Thus, the generation of a site or room map represents not
the end point of architectural analysis but rather the data collection phase analogous to a
surface collection or a test excavation; it is the primary source of important architectural
information. To conduct a complete investigation of an architectural entity, considerations
of function, technology, aesthetics, and tradition are only some of the criteria that ought to
be evaluated. This study is conducted from a perspective that it is possible to measure
architectural spaces and to classify architectural forms in the same ways as portable
artifacts have been measured and classified. To facilitate this study, it is necessary to turn
to computer techniques as a means of analyzing and storing architectural information.
20
CHAPTER 2
ARCHITECTURAL ANALYSIS AND COMPUTER TECHNOLOGY
Having discussed the need for more sophisticated techniques of architectural
analysis in Chapter 1, this section describes one method for looking at architecture as an
artifact. The availability of methods for classifying and analyzing the data contained in a
given assemblage is of initial concern in any analysis of prehistoric artifacts. Due to their
very nature, architectural entities are difficult to manipulate in the same ways as other
more portable artifacts. The great advantage of ceramic and lithic analyses is that the
artifacts can be taken back to the laboratory and studied systematically. Architecture, on
the other hand, must be well documented in the field so that analyses can be conducted
after the site is backfilled. One way of facilitating the analysis of architecture is by
inputting field data into a computer mapping program such as AutoCAD®. Entering site
or house plans into the computer allows them to be manipulated in ways that were difficult
or impossible using traditional mapping and graphic representation techniques.
In effect, the use of computerized mapping programs allows the analyst to take a
full scale, three-dimensional representation of an architectural space in hand and look at it
from all sides just as if it were a portable artifact. Furthermore, sites can be put on the
same sheet and regularized for scale and orientation, and then compared visually just as
any set of artifacts can be. The use of computer drafting programs also greatly facilitates
the measurement of irregularly shaped spaces such as plazas and room blocks. Estimation
of area, for example, can be made at a high level of precision, down to the millimeter level
(well within the margin of error of the site plan). Computer mapping also allows sites to
be pulled apart into smaller units such as households or construction units enhancing our
ability to evaluate construction sequences and intrasite architectural relationships. Finally,
21
the use of computer aided drafting software, in conjunction with various engineering
packages, allows for the study and manipulation of architecture not only as dwellings and
culturally produced artifacts but also as buildings with structural and spatial requirements.
Architecture, more than any other class of artifactual remains, by virtue of its
physical characteristics, lends itself to graphic representation. As indicated above, our
inability to take an architectural space back to the lab puts even greater emphasis on the
need for more accurate display of architectural information. The graphic nature of
architecture allows analyses to be divided into three broad categories: 1) considerations
of the ways in which space is partitioned; 2) analyses based on stylistic characteristics;
and 3) studies of the engineering and structural requirements of an architectural entity.
None of these three classes is totally independent of the other two nor, as mentioned
above, are any of them the result of a single behavioral variable. Each of these areas is
interdependent; each is a result of the myriad of behavioral responses to the need to
provide and maintain shelter and to denote territorial boundaries.
Partitioning of Space
The way in which space is partitioned can be broken down into two areas of study.
Each of these benefits enormously from the use of computer aided drafting packages.
Space can be conceived in both a qualitative and a quantitative fashion, and computer
drafting programs are specifically designed to facilitate the manipulation of plan drawings
in both ways.
Qualitative Analysis
Qualitative analyses of spatial relationships consist of such factors as linkages
between rooms and houses, access to different areas of the site, and the relationships of
various construction events to one another. Recently, several studies have taken
22
advantage of the graphic nature of architecture to apply methods of graph analysis in order
to measure the relationship between spatial integration and social integration (Ferguson
1992, 1993; Hillier and Hanson 1984). The graphic nature of architecture allows for the
manipulation of discrete site units in order to reconstruct the various qualitative spatial
relationships of a prehistoric community. The single most useful tool in AutoCAD® for
the organization and manipulation of graphic information is the ability to draw in layers.
Layers are a series of overlays, each of which can contain a different set of information.
For example, the site plan presented in Chapter 4 (Figure 2) contains 52 different layers,
including one for topography, fourteen for construction phases, two for the water courses,
four for burials (these layers are turned off in Figure 2) and several other individual sets of
information. The ability to draw in multiple layers not only allows information to be
stored and presented by means of different categories of information but it also enables the
map to be divided into temporal categories such as those displayed in Chapter 4 (Figures
3-16 and Figure 22). In the past, either several maps had to be drawn or numerous
hatching patterns had to be utilized in order to display qualitative spatial relationships
synchronically or diachronically. Computer drafting increases the speed and efficiency of
these analyses by allowing several maps to be generated from one map simply by
manipulating the different layers designated in the drawing. Not only can computer
generated maps be reproduced numerous times at any scale (including full size) but,
because they are stored in an electronic medium, they can be modified and revised without
destroying the original drawing, and without retracing the original plan.
The current analysis largely focuses on the qualitative characteristics of the
partitioning of space and thus relies heavily on AutoCAD® as an organizational tool. The
construction phases and growth periods to be presented in the following pages were all
generated from a single map using multiple layers organized by construction phase and by
23
growth period. The division of Grasshopper Pueblo into these discrete units of analysis
would have been, at best, difficult without the use of a computer drafting package. This
analysis illustrates the usefulness of AutoCAD® not only as a device for displaying
graphic information but also as a critical tool for organizing and storing visual data.
Quantitative Analysis
The most useful aspect of computer drafting programs in architectural analyses in
archaeology is in the area of quantitative analysis of space. In the past, measuring the area
of an irregular space from a map required the use of a basic length - width measurement.
For example, attempts to accurately estimate the area of Hohokam courtyard groups have
been founded on the idea that a simple length - width measurement approximated the area
of houses and courtyards and that measuring all houses in the same manner consistently
biased the estimation in the same direction in all cases. However, after limited work with
two of the best documented sets of courtyard groups in the Hohokam area, it is apparent
that measurements based on simple length - width relationships have been found to
consistently misrepresent the actual area of houses and outdoor spaces.
Recently, I have attempted to measure more accurately the areas of courtyard
groups at Snaketown and La Ciudad using the AutoCAD® program. Although I applied
a different technique for delimiting courtyard groups (see Sires 1987), a comparison of the
individual house areas presented here to those of Wilcox and others (1981: Table 9) and
Henderson (1987: Table 3), illustrates the problems of using simple length - width
measurements (Table 1). In most cases, the area of Sedentary period houses presented in
Table 1 is on average two square meters lower than the original estimations. This is due to
the fact that using a length - width measurement to determine area assumes that a 90
degree quadrangle is being measured. The pit structures at Snaketown and La Ciudad are
24
Table 1. House Areas from Snaketown and La Ciudad
Snaketown - Sedentary Period Houses
La Ciudad - Sedentary Period Houses
House
No.
Floor Area Wilcox
et al. 1981
Floor Area
this Analysis
9G:3
10G:9
10G:18
10G:16
10G:3
10F:(9)
10F:3
10F:1
10G:6
10F:4
10G:2
10G:4
10F:10
10G:8
10F:19
10G:10
10F:15
10F:24
10F:4
10F:17
10F:23
10F: 16
10F:21
10F:9
10F: 11
10F:5
54.9
59.1
51.8
37.5
25.5
52.25
36.2
40.3
29.4
25.3
15.9
36.1
45.3
31.8
30.6
24.6
17.4
18.2
25.3
22.6
27.5
18.9
27.2
21.4
27.1
18.1
54.43
62.64
48.57
36.96
24.88
50.18
35.58
41.92
28.01
23.54
16.54
35.03
42.24
30.39
32.45
22.28
20.32
17.5
20.4
20.08
26.92
21.53
23.06
21.89
25.94
17.23
House
No.
804
802
1328
1052
800
1060
1725
710
1706
715
160
688
132
807
780
157
1349
696
1056
808
129
Floor Area
Henderson 1987
15.58
13.00
13.75
12.00
15.00
11.25
21.00
16.63
12.96
11.86
22.68
11.16
13.39
24.50
11.75
16.10
12.74
10.00
12.70
21.00
22.01
Floor Area
this Analysis
14.59
11.35
9.35
8.94
16.00
11.36
21.08
14.82
12.28
14.91
20.17
10.03
15.31
24.41
8.79
14.78
11.44
11.57
10.74
14.3
26.87
not quadrangular, nor are they regular in shape. The same holds true with the
measurement of rectilinear spaces characteristic of the puebloan region. All of these
spaces are at best only rectangle-like and their area cannot be measured precisely using a
length - width measure.
One other related issue that must be addressed is the fact that the numbers
presented in Table 1 are not lower than the original estimates in all cases, as would be
expected given the previous discussion. Apparently, due to the extreme irregularity of the
pit structures as well as the presence of the entryway, the estimates presented here are
25
Table 2: Courtyard Groups at Snaketown
Average
Courtyard
Total
House
House
Area
(m2)
Area
Area(m2)
54.43
54.43
810.94
Total
Area
(m2)
865.37
Public Space
Public
per House
Space
(m2)
(% of total)
810.94
93.71
1
House
No.
9G:3
Floor
Area
54.43
2
10G:9
62.64
62.64
62.64
802.73
865.37
92.76
802.73
3
10G: 18
10G:16
10G:3
48.57
36.96
24.88
36.80
110.41
635.84
746.25
85.20
211.95
4
10F:(9)
32.63
195.77
510.83
706.6
72.29
85.14
10F: 1
10G:6
10F:4
10G:2
50.18
35.58
41.92
28.01
23.54
16.54
5
10G:4
35.03
35.03
35.03
186.58
221.61
84.19
186.58
6
10F:10
42.24
30.39
32.45
22.28
20.32
17.5
27.53
165.18
551.44
716.62
76.95
91.91
10G:8
10F: 19
10G:10
10F:15
10F:24
7
10F:4
10F: 17
10F:23
10F:16
20.4
20.08
26.92
21.53
22.23
88.93
338.54
427.47
79.20
84.635
8
10F:21
23.06
23.06
23.06
191.61
214.67
89.26%
191.61
9
10F:9
10F: 11
21.89
25.94
17.23
21.69
65.06
256.33
321.39
79.76%
85.44
Group
10F:3
10F:5
26
Table 3: Courtyard Groups at La Ciudad
House
Group No.
804
1
802
1328
Floor
Area
14.59
11.35
9.35
Average
House
Area
11.76
Total
Courtyard Total
Public
Public Space
Space
per House
House
Area
Area
Area(m2)
On2)
(m2)
(m2) (% of total)
89.96
105.40
35.29
316.21
351.5
2
1052
800
2060
8.94
16.00
11.36
12.10
36.30
213.52
249.82
85.47
71.17
3
1725
710
1706
715
21.08
14.82
12.28
14.91
15.77
63.09
248.71
311.8
79.77
62.18
4
160
20.17
20.17
20.17
N/A
N/A
5
688
132
807
10.03
15.31
24.41
16.58
49.75
247.29
297.04
83.25
82.43
6
807
780
157
1349
24.41
8.79
14.78
11.44
14.86
59.42
307.79
367.21
83.82
76.95
7
696
1056
11.57
10.74
11.16
22.31
219.46
241.77
90.77
109.73
8
1056
808
129
10.74
14.3
26.87
17.30
51.91
185.52
237.43
78.14
61.84
sometimes higher than the original estimates. This indicates that the assumption that the
use of a length - width measure consistently biases the results in the same direction, either
up or down is not true in all cases. The investigation of Hohokam courtyard groups
indicates that while there is a tendency for the computer generated numbers to be lower
than the estimates previously available, it is not always the case, and the idea that all of the
samples are biased in the same direction is inaccurate.
27
Once architectural spaces have been accurately measured, comparisons can be
made between various spatial relationships. Tables 2 and 3 illustrate the types of
comparisons that are possible once accurate estimations of area are made. These tables
point out the fact that while the Sacaton phase courtyard areas and house areas at
Snaketown and La Ciudad are markedly different from each other in terms of total area,
the amount of available public space per domestic space is very similar. The analysis of
Sacaton phase courtyard groups illustrates how computer aided drafting can be used to
measure various spatial relationships and the accuracy and ease with which such analyses
can be done.
Stylistic Characteristics
The ability to display several architectural entities on a single drawing regularized
for scale and orientation allows for numerous sites to be compared at the same time. This
is analogous to ceramic sorting, in which stylistic elements are used to subdivide sherds
into various types. This type of analysis becomes possible when an entire site is conceived
of as a classifiable artifact. Just as sherds can be sorted and classified by decorative style,
temper type, or any other physical attribute, site plans can be sorted along several lines
such as common orientation, plaza presence or absence, or using any feature that is
expressed architecturally. Manipulating sites in this way can allow us to begin to get
beyond the "ware level" in architectural analysis.
In terms of an intrasite investigation, the same can be done with masonry types,
room layouts, feature distribution, or any other graphically significant aspect of
architecture. For example, the study and classification of masonry types is greatly
enhanced through the use of computer drafting in that photographs of walls can be
digitized at a 1:1 scale and displayed on one drawing just as ceramic design elements can
be. In conjunction with data imported through data exchange files (DXF) from programs
28
such as Surfer, the distribution of features and artifacts can be plotted from existing data
bases without the time consuming process of plotting individual points on a scaled
drawing. Then by conceptualizing different combinations of features and artifacts as
representing styles of rooms or habitation areas, room and feature types can be displayed
and sorted as a ceramicist would do with an assemblage of sherds. Though these types of
analyses have been possible in the past (Ciolek-Torrello 1978, 1985), it has not been easy
to display and evaluate the spatial distribution of various rooms, artifacts and/or features
without retracing the site plan and plotting their individual distributions. Furthermore, the
availability of multiple layers allows distributions of architectural entities to be displayed in
numerous ways that previously involved the use of multiple hand-drawn maps and an array
of confusing symbols.
Engineering and Structural Characteristics
The usefulness of the AutoCAD® computer drafting program as a tool analyzing
the structural and engineering requirements of buildings is not immediately apparent
simply because, with the exception of Chacoan great houses, these avenues of inquiry have
been largely ignored by archaeologists. The use of AutoCAD® in conjunction with
engineering software allows the analysis of architecture on a more experimental level,
analogous to the experimental work done on ceramic samples (see Rice 1987). Thus, not
only does AutoCAD® facilitate the display, measurement, and storage of architectural
data, it also provides a means by which architectural spaces can be viewed in terms of both
compositional studies and structural requirements.
These types of analyses would serve to help clarify several current issues involving
architectural information in archaeology. The most prevalent of these is the pithouse to
pueblo transition, which in the past has been the subject of a variety of theories concerning
29
the factors influencing the change in dwelling types. Cost and labor requirements
(McGuire and Schiffer 1983), dwelling shape and spatial divisibility (Hunter-Anderson
1977), and seasonality (Gilman 1987) have all been seen as important variables in the
pithouse to pueblo transition. Viewing the transition from an engineering perspective
provides yet another avenue of research into the question of the pithouse to pueblo
transition. While many of these analyses are beyond the area of expertise of most
archaeologists, they represent an important line of evidence that must be evaluated in
order to have a complete architectural analysis of a structure.
Discussion
In the following analysis, community growth at Grasshopper is viewed as the
product of the relationship between time and social organization. As such, it is necessary
to classify individual units of construction into temporal categories so that construction
data can be easily displayed and manipulated. Various software packages are applied to
these data not only to provide a means of data display, but also to manage the abundance
of architectural information, to store and synthesize data, and to provide a means of
ordering temporally distinct units in time. Figures 3-16 (Chapter 4) and Figure 22
(Chapter 4) illustrate how multiple layers were used to sort Grasshopper into temporally
distinct units, first by construction phase and subsequently by growth period. In addition,
the data base presented in Appendix B was linked to the base map so that all of the
architectural information contained in the database can be retrieved for any single room
space simply by selecting the room number on the map. In essence, by applying these
techniques to construction data from Grasshopper, I have created a representation of the
site that contains as much temporal and construction information as was collected from
the ruin. Ultimately, through continued work, it will be possible to create databases for all
30
available architectural information and to link them to the site map, thereby creating a
computer version of Grasshopper that for analytical purposes will act as the site itself In
terms of the present analysis, however, the use of AutoCAD® is primarily helpful in
classifying construction units and for making the temporal distinctions that are necessary
to date undated construction units.
31
CHAPTER 3
TIME IN ARCHITECTURAL ANALYSIS
Once architectural units have been properly classified and organized, applying the
appropriate temporal information becomes the next step in developing a reconstruction of
site growth. Two features of prehistoric pueblo building allow for time to be estimated:
the intrinsic behaviors related to room additions, which allow construction units to be
ordered from earliest to latest in a horizontal sequence, and the use of datable wood
species, which provides an independent means of assigning calendrical dates to
construction events. The current analysis combines these two types of temporal methods
in assigning absolute dates to the construction phases at Grasshopper Pueblo. Of primary
concern to the present analysis is the distinction between intrinsic dating techniques and
independent dating techniques (Dean 1978) as they relate to the examination of
architectural additions.
Intrinsic techniques, as traditionally defined, are those dating methods that derive
their temporal properties directly from the activities and behavior of humans. Intrinsic
techniques include both relative dating methods such as seriation and horizontal
stratification and absolute dating methods like ceramic cross-dating. In contrast,
independent techniques derive temporal information from phenomena not directly related
to human behavior. Independent dating techniques are also sub-divided into relative
techniques including natural stratification and absolute techniques such as radiocarbon
dating and dendrochronology. (Dean 1978; Willey and Sabloff 1993).
32
Dating Architectural Units
Intrinsic - Relative Dating
This analysis takes advantage of the fact that units of construction added over time
serve to provide a stratigraphic sequence of site development in the horizontal dimension.
As an intrinsic dating technique, the identification of construction episodes can place
architectural units into relative sequences based upon the characteristics and requirements
of wall construction. As with any type of stratigraphic technique, the construction phase
model (Chapter 4) attempts to order units into a relative sequence based on their position
with respect to one another. Instead of ordering materials into a vertical sequence based
on the principles of natural deposition, however, construction phases are ordered in a
horizontal sequence such that the earliest construction phases are the ones to which all
subsequent phases are added. Of primary importance to the relative ordering of
construction units is the notion of typological (classificatory) contemporaneity (Dean
1969).
Based on Reid's (1973) "Construction Phase Model" (see below), typological
contemporaneity is assumed for all of the additions of the same construction phase at
Grasshopper. Typological contemporaneity is assigned to each construction phase based
on its relationship to both prior and subsequent construction phases. All units of a single
construction phase are abutted to units of the immediately preceding construction phase
and are themselves enclosed by units of the immediately following construction phase.
Having seriated units of typological contemporaneity by means of construction phases
provides a horizontal stratigraphic sequence to which absolute temporal information can
be applied.
33
Independent -Absolute Dating
One goal of any analysis of community growth is to provide absolute dates for the
units of architectural addition. Having determined the relative growth sequence of a site
from initial to final construction is not entirely adequate to describe the trajectory of
community growth. As the following analysis will indicate, simply ordering units of
community growth does not adequately address issues of rate of growth nor does it
illuminate differential growth patterns between non-contiguous architectural units.
Fortunately, the nature of prehistoric architecture in the Southwest allows for highly
accurate estimations of construction dates.
The advantage of architecture in the puebloan Southwest is that datable wood
species were used for almost all construction purposes. With proper control over the
recovery process (Dean 1978) and an understanding of the cultural and natural formation
processes that affect wood preservation (Schiffer 1987), room construction can be dated
through the use of dendrochronology. Although an in-depth discussion of tree-ring dating
is not appropriate here (see Dean 1986), it should be noted that it is the only means by
which absolute construction dates can be applied to room additions at pueblo sites.
Following Dean (1969) once again, the notion of absolute contemporaneity becomes
crucial to discussions of community growth. By providing absolute temporal ranges to
construction units, our ability to perceive the overall pattern is greatly enhanced. While
seriating pueblo additions provides a rough measure of site development, at best it is only
a framework for the application of absolute temporal information.
Relative Growth and Actual Growth
The distinction between relative and absolute time allows growth to be envisioned
in two ways: relative growth and actual growth. The behaviors related to room
34
construction provide one type of temporal signal, i.e., a sequence of architectural additions
through time. The growth pattern derived from such an intrinsic-relative analysis is
referred to as relative growth. Relative growth, in this case, lacks absolute temporal
information and simply represents the aggregate of all construction from earliest to latest.
The construction phase model (Reid 1973) is the means used in this analysis for
reconstructing relative growth. Actual growth, on the other hand, requires the application
of some sort of absolute temporal information. Actual growth is here defined as a relative
growth reconstruction that benefits from the availability of absolute dates. It is the use of
wood as a construction material in pueblo roofs that links these two ways of estimating
time.
Provided that roofs are constructed as part of all room additions, and taking into
account factors such as reuse or stockpiling of timbers and remodeling, a newly
constructed roof maintains the same relationship to a previously existing structure as the
newly constructed walls that support it. Thus, the association of wooden roofs with their
supporting walls provides the connection between the intrinsic behaviors that allow
relative time to be measured and the independent characteristics of trees that allow
absolute dates to be assigned to them. It is important that the distinction be made as to
which type of growth is being discussed in an architectural analysis due to the fact that a
relative growth reconstruction does not provide information as to contemporaneity of
construction units or the rate of growth at a site.
In applying a relative/actual growth reconstruction it is best to conceive of the two
types of growth as representing opposite endpoints on a continuum. At one end of the
continuum is true relative growth which consists of a sequence of architectural additions
that contains no absolute temporal information. True actual growth, on the other hand,
refers to a construction sequence in which every architectural unit has an absolute date.
35
The best example of an architectural reconstruction that comes close to true actual growth
is Dean's analysis of Betatakin and Kiet Siel (Dean 1969), while the 14 construction
phases at Grasshopper presented in Chapter 4 represent a true relative growth
reconstruction. Ultimately, the goal is a reconstruction of actual growth that comes as
close to true actual growth as possible. The primary factor in determining the extent to
which a relative growth reconstruction can be developed into an actual growth
reconstruction is the resolution of temporal information available for the given
architectural sequence. As the level of temporal resolution increases, the necessity of
relying on the assumptions of a relative growth sequence such as typological
contemporaneity and equal time intervals for phases decreases. As a result, relative phases
are broken down into individual dated units which can then be regrouped into actual
growth phases. (Figure 1)
Figure 1 shows a hypothetical growth sequence as it might be reconstructed using
a relative growth model and using an actual growth model. As this figure illustrates, the
basic relationships between constructed units remain the same in terms of their physical
association to one another in both examples. However, the pattern of growth is expressed
as being markedly different in each of the two cases. Relative site growth simply
represents an ordering of construction units into groups of typologically contemporaneous
phases. In Figure 1, the relative growth sequence causes units to be ordered in such a way
that the construction phases cross-cut the individual room blocks to give the impression
that the three room blocks grow at the same rate . However, when absolute temporal
information is applied to the same units, growth cannot be represented in the same simple
fashion. The actual growth pattern illustrated in Figure 1 indicates that the relative growth
sequence, while accurate in terms of the physical pattern of room additions, may not
adequately illustrate site growth. A comparison of the two examples in Figure 1 points to
RELATIVE GROWTH
ROOMBLOCK 1
r~
CONSTRUCTION PHASE 1
i
i
I 1-3* I
CONSTRUCTION PHASE 2
1
1
CU 1-1*
CU I-2BJ
CU 2-2*
CU 2-3*1
CONSTRUCTION PHASE 3
ROOMBLOCK 2
1
1
|
I
CU 2-1*
CU 7-2BJ
NOTE: A CONSTRUCTION PHASE REFERS TO ALL
UNITS OF CI ASSIFICATORY CONTEMPORANEITY
ACTUAL GROWTH
ROOMBLOCK
[" A D. 1301
I
"J
CU 1-2*1
A.D. 1298
CU 1-1*
GROWTH STAGE 1
A.D. 1310
CU 1-3*1
I
A.D.1310
.cui-raj
GROWTH STAGE 2
GROWTH STAGE 3
GROWTH STAGE 4
ROOMBLOCK 2
r
A.D. 1306
A.D. 1310
•
CU 2-3*1
A.D. 1298
CU 2-K
A.D. 1306
i
NOTE: SAME UNITS AS ABOVE GROUPED INTO PHASES OF
ABSOLUTE CONTEMPORANEITY WITH TREE RING DATES.
Figure 1. Schematic Diagram:
Relative vs. Actual Growth
37
three important considerations in community growth analyses: 1) Growth in one room
block may not reflect the growth trajectory of another; 2) The hiatus between
construction events is not measurable without the use of absolute dating techniques; and
3) While construction units are valid as discrete entities, construction phases, as defined
below, are best viewed as an analytical construct and may not be representative of actual
community growth. However, it is important to note that as the construction phases begin
to break down into a series of individually dated construction units, the construction phase
itself is never invalidated in terms of the sequence of physical architectural relationships
that allow it to be created, it merely ceases to be useful as a classificatory device.
The following pages apply the methods discussed thus far to the construction data
from Grasshopper Pueblo. The construction units are organized and classified using the
computer techniques presented in Chapter 2 and are then discussed in terms of relative and
actual growth. Reid's (1973) construction phase model is used as a means of
reconstructing the relative growth of Grasshopper pueblo. Following this, temporal
information taken from Graves' re-analysis of Grasshopper tree-ring dates is applied to the
relative growth sequence in order to test the construction phase model and develop an
actual growth reconstruction.
38
CHAPTER 4
RE-ASSESSING PUEBLO GROWTH AT GRASSHOPPER
The Construction Phase Model
Using the data obtained by the cornering project and necessitated by the poor
quality of the tree-ring dates at Grasshopper Pueblo (Figure 2), a construction phase
model was developed and applied as a means of measuring community growth (Reid
1973; Reid and Shimada 1982). Although the community growth model developed by
Reid (1973) did take into account the tree-ring dates from Grasshopper, the following re­
assessment of the construction phase model as well as the current community growth
reconstruction is undertaken in order to verify Reid's community growth phases and to
evaluate the construction phase model in light of Graves' revised tree-ring dates.
The construction phase model relies on the premise that architectural form delimits
behavioral space (Dean 1969; Reid 1973; Reid et al. 1975; Reid and Shimada 1982; Rohn
1971; Wilcox 1975). At the nucleus of the construction phase model is Wilcox's
construction unit (Wilcox 1975, 1982). A construction unit (CU) is a set of continually
bonded walls and their associated spaces (Wilcox 1975, 1982). Core construction units
(CCU) are the earliest discernible units of construction in a room block upon which all
succeeding additions are made. Core construction units are constructed independently of
any other architecture, while non-core construction units are added to pre-existing ones.
The largest architectural entity is the room block. A room block (RB) is defined as all
additions forming a single contiguous set of room spaces (Reid 1973; Wilcox 1975).
The construction phase model assumes that all new construction units determined
to be added onto existing construction units, through bond-abut or wall face analysis,
were built either at the same time or later than the pre-existing rooms. Construction
phases (CP), as defined by Reid (1973), are sets of rooms added on to existing ones. The
39
Figure 2. Grasshopper Pueblo: Site Plan
40
previously existing sets are considered to represent a single phase, while those added to
them are part of the immediately succeeding phase. All typologically contemporaneous
construction units (see above) are classified as a single construction phase, regardless of
spatial connectivity (Figure 1).
The construction phase model is an analytical construct to seriate individual
construction units. The addition of construction units is assumed to represent behaviorally
significant clusters of habitation spaces. Basic to this notion is the supposition that
individual rooms or groups of rooms, representing some fundamental level of
organization, are being added in response to increasing requirements for domestic space
(Wilcox 1975, 1982). Whether these additions represent steady growth through the
natural budding off of resident household groups or whether they represent the influx of
people into the site, while important to considerations of rate of growth, is irrelevant to
the basic assumptions of the model.
The primary method of determining a construction sequence throughout the
Southwest has been through bond-abut analysis ( Rinaldo 1964; Rohn 1971; Wilcox 1975,
1982). Any bond-abut analysis is based on the simple fact that a bonded corner represents
a single construction event while an abutted corner represents a coeval or later
construction episode (Reid 1973; Wilcox 1975, 1982). While this seems to be a fairly
straightforward concept, there are problems with using bond-abut data for ordering
construction events. As with any artifactual material, architecture is subject to a wide
range of cultural and natural formation processes (Schiffer 1987). The practice of
remodeling can have dramatic effects upon the original corner relationships at a site
(Wilcox 1982:22). Walls are often remodeled, rebuilt, and replastered, and in some cases,
this can be quite extensive (Ferguson et al. 1990). In addition to the complexities arising
from the human behaviors associated with walls, natural processes of decay can further
41
obscure corner relationships. Pueblo walls collapse in a number of ways, creating a jumble
of rubble where corners may have once existed. Not only can walls fall inward or
outward, but a single wall can fall one way at the top and the other at the bottom, while
collapsing in an entirely different manner at the other end and in the middle. With the four
walls of a structure collapsing in this unpredictable manner, it is easy to see how corner
relationships can become obscured.
Partially in response to the difficulties inherent in the use of bond-abut data,
investigators turned to the use of wall face analysis as a tool for further clarifying
construction sequences. Wall face analysis is based on the assumption that walls are
constructed from the outside rather than the inside. As a result of this behavior, the
exterior face of the wall is generally smoother due to its use as the plumb line for the
construction of the wall (Reid 1973; Rinaldo 1964:49). For example, determining the
range of variability in wall types at Grasshopper (see Scarborough and Shimada 1974) has
allowed for more precise evaluation of the types of wall faces found. This in turn has
augmented bond-abut data to make the identification of construction units possible.
Because rooms are added to the outside of existing structures rather than to the inside, the
identification of a smooth faced wall indicates that the wall in question was once an
exterior wall and the space enclosing it was added on later. Again, remodeling may affect
wall-face relationships to some extent but unless the wall collapses completely, the natural
processes that can blur corner relationships do not act as dramatically on wall faces. In
both cases however, good control and competent data recovery can alleviate many of the
aforementioned problems. The ultimate limitation of both techniques is, of course, that
neither method can do more than provide a relative sequence of construction events.
Although the construction phase model is very basic in its conception and is
founded on a fairly firm set of assumptions, there are several possible sources of variability
42
that must be accounted for. As Wilcox points out (Wilcox 1982), there is a hiatus
between the building of one unit and the subsequent addition of another unit. The length
of this gap can only be determined precisely through the use of an independent dating
technique. Without applying independent dating to the construction phases at
Grasshopper, there is no way to assign absolute contemporaneity to architectural units.
Non-contiguous room blocks also pose a problem in working out a relative growth
reconstruction. One section of the site can grow at a different rate than surrounding units.
For example, based on an analysis of bond-abut and room connectivity, Rohn
reconstructed the growth of Mug House and found that in the early period of growth,
there were two distinct clusters of habitation, the northern of which grew at a faster rate
than the southern cluster (Rohn 1971). Of course, it was only through the availability of
adequate dendrochronological data that the temporal distinction between the two units
was possible. Turning to the Grasshopper example, Figure 3 illustrates a similar situation,
except there are eight core construction units (as defined by Reid, personal
communication) which eventually coalesced to form the three large core room blocks at
the site (Figure 2). Although the display of these units makes them look is if they are
absolutely contemporaneous, it must be kept in mind that they are only contemporaneous
because of their inclusion in the analytical constructs of construction phases. Figures 3-16
show the 14 construction phases as cross-cutting the three room blocks. Although these
units abut one another in the same sequence in all three of the room blocks, there is no
way of physically relating them to one another across the core room blocks due to the
spatial discontinuity between them. Thus, the construction phase model makes the
necessary assumption that all units of typological contemporaneity (of the same
construction phase) were added at the same time. When, in reality, of course,
construction units, even within a given phase, may be added differentially at different
43
times. Again, with adequate dating, the problems of comparing room additions between
non-contiguous blocks of rooms can be minimized.
Grasshopper Construction Phases
Both bond-abut and wall face data have been used to reconstruct the relative
growth of Grasshopper (Reid 1973; Reid and Shimada 1982) where a total of fourteen
individual construction phases have been identified (Figure 3-6, Table 4). Relative growth
of the site, based entirely on the ordering of these construction phases, reflects the pattern
put forth by Reid (Reid 1973; Reid and Shimada 1982). The earliest units added (Phases
1-5) are all fairly large, indicating a process of demographic change in line with what
would be expected for the influx of new social groups (Longacre 1975, 1976; Reid 1973;
Reid and Shimada 1982). By contrast, the addition of units after Phase 6 seems to
represent a marked decrease in the rate of site growth.
Table 4: Construction Phase Data
Construction
Phase
Additions to
Room Block 1
# CUs
URms
Additions to
Room Block 2
U CUs
URms
1
2
3
4
5
6
7
8
9
10
11
12
13
14
3
5
5
8
8
5
4
2
2
2
2
1
1
0
17
13
14
15
11
10
4
2
2
2
2
1
1
0
2
5
5
4
4
2
1
0
0
0
0
0
0
0
32
13
16
10
16
2
3
0
0
0
0
0
0
0
TOTALS
48
94
23
92
Additions to
Room Block 3
U CUs
URms
3
3
2
4
8
7
5
5
6
4
2
3
3
1
56
14
7
5
9
19
11
9
7
7
4
2
3
3
1
101
Total
Additions
U CUs
URms
8
13
12
16
20
14
10
7
8
6
4
4
4
1
127
63
33
35
34
46
23
16
9
9
6
4
4
4
1
287
44
PLAZA
PUZA I
PUZA
METERS
Figure 3.
Construction Phase 1
Construction Phase 1
Construction Phase 1 consists of eight spatially separate core construction units
containing a total of 63 ground floor rooms. Three units are located on the east side of
Salt River Draw while the remaining five are to the west of the draw. The clusters of
rooms on the west side of the draw form two larger groups each arrayed around a central
plaza area (Figure 3).
45
ezza
PUZA
PLAZA - I
PLAZA
LEGEND
NEW ARCHITECTURAL ADDITIONS
METERS
EXISTING ARCHITECTURE
Figure 4.
Construction Phase 2
Construction Phase 2
In phase two, a total of 13 construction units, consisting of 33 ground floor
rooms, are added to the largest blocks of Construction Phase 1. Additions on the eastern
side of Salt River Draw are mostly to the west of the largest core unit. To the west of the
draw rooms are typically added to the north and south of the existing room blocks (Figure
4).
46
PLAZA
PLAZA I
PLAZA
LEGEND
NEW ARCHITECTURAL ADDITIONS
EXISTING ARCHITECTURE
Figure 5.
Construction Phase 3
Construction Phases 3 and 4
Construction Phases 3 and 4 see the addition of 69 rooms as part of 28 individual
construction units. Additions follow the same general pattern as that noted in
Construction Phase 2. The majority of spaces added are to the section of the site that later
becomes Room Block 2 with the six-room cluster to the north (Construction Unit 2-3a)
defining the eastern corridor into Plaza II. Additionally, the plaza spaces as a whole are
being defined in all three locations (Figures 5 and 6; Table 4).
47
PLAZA
PLAZA I
PUZA
LEGEND
NEW ARCHITECTURAL ADDITIONS
METERS
EXISTING ARCHITECTURE
Figure 6.
Construction Phase 4
48
PLAZA
PLAZA I
GREAT
KIVA
LEGEND
NEW ARCHITECTURAL ADDITIONS
METERS
EXISTING ARCHITECTURE
Figure 7.
Construction Phase 5
Construction Phase 5
Construction Phase 5 consists of 20 construction units and represents the largest
construction phase in terms of the total number of rooms added. Forty-five ground floor
rooms are added to the three core room blocks. Additions to Room Blocks 2 and 3
consist of several large construction units while the growth of Room Block 1 seems to
already be experiencing a slow down in its rate of growth. The rooms necessary for the
roofing of the southern corridor are added to the southern end of Room Block 3 and
Plaza 3 is formalized into the shape it was to attain as the Great Kiva. Finally, the process
of infilling is beginning in Plazas 1 and 2. (Figure 7; Table 4).
49
PLAZA
PLAZA I
GREAT
KIVA
LEGEND
J
NEW ARCHITECTURAL ADDITIONS
I
EXISTING ARCHITECTURE
Figure 8.
Construction Phase 6
Construction Phases 6 and 7
Construction Phases 6 and 7 together consist of only of 24 construction units with
a total of 39 rooms. These construction phases represent the first significant decrease in
the number of added spaces (Table 4), with only 24 rooms being built in Construction
Phase 6 and 15 rooms added in Construction Phase 7. By this period in the relative
sequence, Room Block 2 reaches its final configuration and from here on site growth
slows dramatically in Room Blocks 1 and 3 (Figures 8 and 9).
PLAZA
PLAZA I
GREAT
KIVA
LEGEND
NEW ARCHITECTURAL ADDITIONS
EXISTING ARCHITECTURE
Figure 9.
Construction Phase 7
51
/
PLAZA
\
RB-3
\
PLAZA I
RB-1
GRCAT
KIVA
LEGEND
20
METERS
I/ I
EXISUNG ARCHITECTURE
Figure 10.
Construction Phase 8
Construction Phases 8 through 14
These periods represent a dramatic decline in the rate of construction at the site.
Some infilling still occurs but the majority of rooms are being added to the western side of
Room Block 3 and to the southern end of Room Block 1 (Figures 10-16). The sum of all
room additions for these seven phases amounts to only 39 additional room spaces in 34
construction units. This compared to a total of 249 room spaces in 93 construction units
for the first seven phases indicates a dramatic reduction in the rate of architectural
additions. This significant decline in the pace of construction argues for a major
demographic shift sometime between Phases 5 and 7.
LEGEND
20
_l
3
NEW ARCHITECTURAL ADDITIONS
EXISTING ARCHITECTURE
Figure 11.
Construction Phase 9
LEGEND
NEW ARCHITECTURAL ADDITIONS
EXISTING ARCHITECTURE
Figure 12.
Construction Phase 10
54
PLAZA
R0-3
PLAZA I
RB-1
GREAT
KIVA
RB-2
0
I
I
20
NEW ARCHIIECTURAL ADDITIONS
I
METERS
EXISTING ARCHITECTURE
Figure 13.
Construction Phase 11
PLAZA
PLAZA I
GREAT
KIVA
20
J
9
NEW ARCHITECTURAL ADDITIONS
EXISTING ARCHITECTURE
Figure 14.
Construction Phase 12
PLAZA
PLAZA I
GREAT
KIVA
LEGEND
20
_|
NEW ARCHITECTURAL ADDITIONS
EXISTING ARCHITECTURE
Figure 15.
Construction Phase 13
57
/
PLAZA
\
RB-3
PUZA I
RB-1
GREAT
LEGEND
RB-2
20
NEW ARCHITECTURAL ADDITIONS
METERS
EXISTING ARCHITECTURE
Figure 16.
Construction Phase 14
Discussion
The analysis of the sequence of construction at Grasshopper indicates that there is
a distinct pattern to the relative growth of the site. Figure 17 shows the cumulative
growth curves for the three individual room blocks of the Main Pueblo based on the
assumption of equal temporal intervals for each of the construction phases. Examination
of the curves indicates that there is both an overall trend of growth, but that each of the
three room blocks also grows in a slightly different way. All three of the room blocks
follow an overall pattern of rapid growth with a subsequent period of slow growth, or no
growth at all. Room Blocks 1 and 2 exhibit the fastest growth and are largely complete by
58
Construction Phase 6 or 7. Room Block 3 exhibits a more gradual course of development
in which construction increases at a fairly moderate rate until the end of the sequence.
*10-
100-
©
cr
m
a
3
z
Room Block 1
Room Block 2
s
6
Room Block 3
20-
10*
2
3
4
5
€
7
S
9
10
11
12
13
(4
CONSTRUCTION PHASES
Figure 17.
Relative Growth Curves: Room Blocks 1, 2 and 3
The composite growth curve shown in Figure 18 provides a look at the aggregate
relative growth sequence for the three room blocks of the Main Pueblo (cf. CiolekTorrello 1978, Figure 6). As was the case with the three individual room blocks, a pattern
of relatively rapid growth is illustrated followed by a period of slowed growth after
Construction Phase 6. Once again, the composite relative growth curve for the core area
reflects past reconstructions of site growth (Ciolek-Torrello 1978; Longacre 1975, 1976;
59
Reid 1973, 1989; Reid and Shimada 1982). However, as discussed above, relative site
growth is only one step in determining the actual site growth. Only the application of
tree-ring dates yields the temporal resolution necessary to build an actual growth
sequence.
300-1
275-
250-
(/>
3
o
OL
b
200-
CK
UJ
CD
§
150-
z
£
(=
<
3
2
3
75*
50-
1
2
4
5
6
7
8
9
10
12
13
14
CONSTRUCTION PHASE
Figure 18.
Composite Relative Growth Curve
Absolute Dates for Growth
A test of the growth model at Grasshopper Pueblo requires that each of the room
additions be assigned an absolute temporal position. Given the relatively short span of
60
occupation at Grasshopper, at a time of such dramatic population increase in the region
(Longacre 1975, 1976; Reid 1973, 1989), the necessary temporal resolution can only be
obtained through tree-ring dating. In the best of all situations, it is preferable to have a
large number of tree-ring dates from each of the rooms added (Dean 1969:9-10).
However, because Grasshopper is an open air site and the tree-ring remains are buried
beneath a deep layer of rubble and alluvium, it is not possible to sample every room at the
site. The sample size necessary to adequately test the rooms for datable material is one
variable that needs to be addressed.
Table 5: Sampling Chart for Construction Units
Number of 50% Sample 40% Sample 30% Sample 20% Sample 10% Sample
of Unit
of Unit
of Unit
Rooms in Unit
of Unit
of Unit
6
21
11
8
2
4
10
6
19
8
4
2
9
18
7
5
4
2
9
7
5
17
3
2
8
16
6
5
3
2
15
8
6
5
3
2
14
13
12
11
10
9
8
7
6
5
4
3
2
1
7
7
6
6
5
5
4
4
3
3
2
2
1
1
6
5
5
4
4
4
3
3
2
2
2
1
1
1
4
4
4
3
3
3
2
2
2
2
1
1
1
1
3
3
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
If the basic assumptions related to the construction unit are correct, then all rooms
in a construction unit are absolutely contemporaneous in terms of the behavior associated
with their construction. In light of this, a small representative sample of 20% of the rooms
61
in each construction unit would yield the information necessary to order construction units
in absolute time. Table 5 summarizes the number of rooms that must contain datable
material in each construction unit size class based on several sample sizes. To obtain a
20% sample from the 287 ground floor rooms of the core room blocks requires that 57
rooms, stratified by means of individual construction units, yield datable tree-ring remains
(Table 5). Although a small stratified sample of the Main Pueblo would adequately date
the construction phases at Grasshopper, there are several issues that must be considered.
The difficulties in applying tree-ring dates obtained from Grasshopper to the
construction of room-sets arise from numerous sources including the problem of dead
wood (Dean 1969, SchifFer 1987), remodeling events, reuse of timbers and any number of
other cultural or natural processes. In addition, in order for the sequence to benefit from
the high resolution of tree-ring dating, the majority of tree-ring dates must be cutting
dates; otherwise, only a range of time can be applied to any given construction unit (Dean
1969). Ultimately, the sample that must be obtained is one in which each individual
construction unit yields a representative number of datable specimens, so that what is
being sampled is not the room spaces themselves but rather the dendrochronological
contents of those spaces. If, for example, a room contains no datable wood remains, then
another must be selected in order to fulfill the requirements of the sampling strategy.
However, even with complete and thorough sampling, the randomness of the processes
that affect the preservation of wood remains (i.e., burning of a room) makes recovery of
datable materials difficult.
This sampling strategy would be the next step in testing the growth sequence of
Grasshopper Pueblo. Fortunately, a large assemblage of tree-ring specimens has already
been obtained from the site (Dean and Robinson 1982; Graves 1986, 1991). However,
the context and quality of these samples is far from ideal in terms of the strategy outlined
62
above. Because of this, it is necessary to expand upon Wilcox's discussion of construction
units (Wilcox 1975, 1982) to formulate several principles for applying Graves' revised
cutting-date estimates to the construction phases at Grasshopper.
Grasshopper Tree-Ring Dates
The remainder of this analysis combines the cutting-date estimates devised by
Graves to Reid's construction phase model for Grasshopper Pueblo. The purpose of this
analysis is three-fold: 1) To present a method for applying dates (or date ranges) to
undated units of construction based on their relationship to dated construction units; 2)
To evaluate the construction phase model as it has been applied at Grasshopper, and 3)
To test the long standing model of rapid growth at the site (Graves et al. 1982; Longacre
1975, 1976; Longacre and Graves 1982; Reid 1973, 1989).
Ceramic dates from Grasshopper firmly place the major occupation of the site in
the 14th century, however, poor preservation of wood remains has hampered efforts to
assign precise dates to most of the construction events at Grasshopper. In spite of the fact
that more than 2000 dendrochronological samples have been collected from Grasshopper
Pueblo, only 164 of these provide a date. Of these, only one sample represents a cutting
date (Dean and Robinson 1982). Unfortunately, the recovery of this sample from Oven 2
increases the probability that it represents dead wood. Because of the lack of cutting
dates and the poor quality of the samples in general, Dean and Robinson (1982), in their
initial analysis of the tree-ring material, could only provide construction dates for two
areas of the pueblo. The earliest major building episode, for which they had adequate
dates, is the roofing of the southern corridor leading into Plaza 1 which was assigned a
date of A.D. 1320 or slightly later (Dean and Robinson 1982:47). The last major event
which could be dated was the conversion of Plaza 3 into the Great Kiva. Although Dean
63
and Robinson could only say that this occurred sometime late in the sequence, Graves
(1991) places the construction of the Great Kiva at A.D. 1347-1360 or later (Table 6).
In his recent analysis of the tree-ring dates from Grasshopper, Graves (1986, 1991)
uses a method for estimating ring loss from dendrochronological samples and for
providing a probable range of construction dates for rooms at Grasshopper. The intent of
Graves' analysis is to provide a means of estimating the gap between the dated event (last
ring) and the reference event (the death of the tree) (Graves 1991:85; see also Dean
1978). Ultimately, his goal is to relate the last dated ring to the target date (room
construction). The method employed by Graves improves upon the technique for
estimating ring loss used by S. Plog (1977, 1980). To account for the shortcomings of
Plog's method, the tree-ring samples were separated by species and only samples from the
same region and general time period were utilized (Graves 1991:89) Graves' samples
consist of material from several pueblo period sites in the mountains of east-central
Arizona, including Grasshopper. By comparison with the estimator group of 38 sites, he
is able to determine that with the exception of Douglas Fir, the Grasshopper series is
typical in the mean number of rings present as well as in terms of the patterns of
prehistoric wood use at the site (Graves 1991.101-112). Combining his technique for the
estimation of ring loss with the more traditional methods of analysis such as date
clustering and date overlap (Bannister 1962; Douglas 1935; Haury 1934), Graves
generates ranges of dates from the tree-ring samples recovered at Grasshopper. The
following pages discuss the methods used and the results obtained from the testing of
Graves' construction date estimates against Reid's construction phase model.
64
Table 6.
Tree-ring Dates and Construction Dates of Rooms from Graves (1991 Table IX)
with Associated Construction Unit Designations
Latest
Associated
Construction Unit
Number of
Samples
1385-1400
1375-1395
1320+ (see Rm 35)
1311-1325
1310-1325
1309-1321 (see Rm 44)
1330-1350
1309-1321+(see Rm 44)
1309-1321
1-1 la
l-4h
l-6a
l-5a
l-3a
l-7b
l-6b
l-6b
l-5b
3
2
1
3
6
3
2
1
2
1347
1311
1288
1319
1313
1320
1346
1298
1342
1332
1366
1353
1313-1323+ (see Rm 26)
1311-1321
1319-1331 (see Rm 23)
1319-1331
1313-1322
1320-1330
1326-1338
1298-1308
1323-1333
1332-1344
1366-1376
1347-1360+
2-5c
2-4b
2-3d
2-3d
2-2d
2-5b
2-5c
2-la
2-2a
2-4a
2-5b
none
5
3
2
6
6
1
5
1
9
5
2
20
211
216
231
246
269
270
274
279
280
438
1313
1323
1305
1228
1343
1229
1302
1308
1373
1331
1313-1323
1323-1333
1305-1317
1307-1316+ (see Rm 269)
1307-1316
1307-1316+ (see Rm 269)
1302-1315
1308-1317
1309-1328
1331-1340
3-7e
3-4d
3-3b
3-la
3-la
3-la
3-3a
3-la
3-2c
3-8b
1
1
6
1
12
2
2
3
4
2
Corridor
1333
1320-1325
none
28
Room
Number
Date
Estimated
Construction Date
1385
1375
1280
1311
1310
1282
1330
1199
1309
Room Block 1
8
11
31
33
35
39
42
43
44
Room Block 2
18
19
21
23
26
145
153
164
183
187
197
Great Kiva
Room Block 3
65
Methodology
The method used in this analysis is based in the concepts of the construction phase
model (Reid 1973) and ideas concerning room-set additions put forth by Wilcox (1975,
1982). The computer drafting techniques and data management programs discussed
above are used in conjunction with these concepts to analyze and display construction data
from Grasshopper Pueblo. After a discussion of the methodology used in assigning
absolute dates to the construction phases at Grasshopper, the results of the analysis will be
presented and discussed both in terms of Graves' dating of the site as well as in relation to
existing reconstructions of community growth.
Problems and Assumptions
Two major problems are inherent in attempts to assign absolute dates to
Grasshopper. The first of these is the lack of cutting dates from the site (see above).
However, the use of Graves' date estimates provides a solution to this problem. Although
the rooms analyzed are not dated to individual years, they are assigned a tighter range of
time than was available in the original formulation of Reid's growth model (Reid 1973;
Reid and Shimada 1982). The second problem in assigning dates to construction phases is
the fact that not all rooms yield tree-ring dates. Because the preservation of wood
remains is greatly enhanced by the entirely random event of burning, the extensive
excavations at the site were unable to fulfill the requirements of the sampling strategy
outlined above and did not provide a dated room from each of the construction units.
However, models concerning site growth and room set additions (Wilcox 1975, 1982)
provide the basis for a number of principles that allow undated clusters to be assigned a
range of dates based on their relationship to surrounding dated units.
66
Before discussing these principles in detail, it is necessary to note some of the
assumptions that must be made prior to such an analysis. The first of these is directly
related to the construction unit. Because a construction unit is defined as a single,
absolutely contemporaneous set of rooms sharing a common, continually bonded wall
(Reid 1973; Wilcox 1975, 1982), it can be assumed that dating one room in a construction
unit provides a construction date for all the rooms in the unit. The second assumption is
more specific to the present analysis. The method used by Graves for estimating cutting
dates is assumed to accurately reflect a range of time in which the actual construction of a
room occurred. Since a reevaluation of Graves' methods is beyond the scope of this
analysis, all of the construction estimates provided by Graves are applied as given. The
final assumption is related to the construction phase model which presupposes that all
construction units of a single phase in all three room blocks of the Main Pueblo are
contemporaneous. This assumption also forms the first step in the overall methodology.
Methods
Given the distinction between relative and actual growth discussed above, the
assumption of contemporaneity is a necessary starting point for applying temporal
information to construction units that cannot otherwise be assigned absolute dates.
However, individual construction units, for which temporal information is available but
which do not fit within the temporal range of the construction phase to which they belong,
are removed from that phase and placed in the temporal period in which they are dated.
This is done in order to increase the precision with which the actual growth sequence is
developed.
67
Figure 19. Construction Units at Grasshopper Pueblo: Main Ruin
68
The next step in applying Graves' cutting-date estimates to the construction phases
at Grasshopper is to assign each of the individual construction units its own sub-phase
designation. Figure 19 shows the breakdown of the core room blocks by construction
phase with each of the sub-phase designations provided. To do this, each construction
unit is given a three-part name. The first part of the classification, followed by a dash,
designates the room block in which the unit occurs. The second number refers to the
construction phase (Figures 3-16) to which the unit belongs. Finally, the letter appellation
represents the single unit referent which distinguishes the unit from all the others of the
same phase within the room block. This classification system gives each unit a designation
that not only facilitates discussion and manipulation of the individual units in a data base
but which also contains information as to what room block and phase the unit belongs
(Figure 19, Appendix B).
The next stage of the analysis was to assign the date ranges provided by Graves to
the construction units from which the dated rooms came. Out of a total of 127 discrete
construction units, only 23 (18%) could be directly dated by the application of Graves'
cutting-date estimates (Figure 20), excluding the Great Kiva and the southern corridor.
As Figure 20 illustrates, there are an essentially equal number of dated clusters in each unit
(RBI and RB2 = 8 each, RB3 = 7) however, Room Block 1 suffers from a lack of dates in
the central section while no dates are available for the outermost rooms of any of the core
room blocks. The lack of dated samples from rooms on the outer margins of the three
core room blocks creates numerous problems in assigning terminal dates to large portions
of all three of the core room blocks, but especially Room Blocks 1 and 3.
DATED ROOMS AND CONSTRUCTION UNITS AT GRASSHOPPER PUEBLO:
MAIN RUIN
PLAZA
PLAZA
•-
q
GREAT
-HI
RB-2
I
0
1
METERS
1
1
10
20
(
LEGEND:
H
Dated Rooms
Xyffh
Construction Units Directly
Miy Doted
L . ..
by Association with Dotea Rooms
|
|
Indirectly Dated Construction Units
j
j
Undated Construction Units
jggyj
(see Table 5 )
Other important Doted Areos
70
To date the remaining 104 undated construction units necessitates that several
definitions and principles be devised for establishing the temporal relationship between the
spatially related construction units. Many of the principles are obvious in their conception
and reflect basic common sense. However, to provide consistency in the application of
tree-ring dates to construction units, it is necessary to formalize them into a set of eight
axioms that can be applied to all of the units equally and without bias.
Definitions
1. Terminal Date - The last possible date that a construction unit could have been
built.
2. Initial Date - The earliest possible date that a construction unit could have been
built.
3. Dependent Construction Unit - Any construction unit that depends on the prior
existence of one or more other units for its construction. This definition is crucial
in assigning dates to units for which there are not dates. A dated dependent unit
can provide a terminal date for all of the units on which it is dependent (Figure
21a).
4. Independent Construction Unit - Any construction unit that is added
independently of other units. An independent unit provides an initial date for all
units that are dependent upon it if other information is not available.
Considerations of dependence and independence are not necessarily predetermined
by the nature of a construction unit (i.e. whether it is a core unit or not) but are
more related to the locations of units in relation to one another in the sequence of
additions (Figure 21a and b).
Principles
1. Given the construction unit model, one dated room in a construction unit dates
that unit.
2. Any room or unit abutted to a dated unit assumes the initial date of the unit it
abuts as its earliest possible construction date (Figure 21c).
3. Any dated construction unit abutted to a non-dated construction unit provides
the non-dated unit with a terminal construction date, i.e. the latest date in the
added construction unit (Figure 2Id).
71
4. The construction date range for any non-dated unit between two dated ones
assumes the initial date of the one it abuts and the terminal date of the one that
abuts to it (Figure 21e).
5. The coincidence of the initial date from a construction unit with the terminal
date of a construction unit that is abutted to it dictates that the construction date
for both units can be only the single date that the two units have in common
(Figure 2 If).
6. For any dated unit in such a position that it is situated between two other dated
units, but which does not fit in the sequence, the date is assumed to be anomalous
and is considered to represent either remodeling or reuse depending on the
direction in which the date is in error (Figure 21g).
7. Construction units that are separated from dated units that are dependent on
them are assigned the terminal date of the dependent units if no dated units are in
between them regardless of the distance or number of abutments (Figure 21h).
8. Date ranges are never modified to include dates outside of the range specified
by Graves 1986, 1991; if dates cannot be modified to fit within this range, the
dates are considered anomalous.
Results
This section presents the results of the analysis. First, a discussion of the growth
of each room block is presented, followed by a discussion of the overall trends observed in
the growth of Grasshopper Pueblo.
Room Block 1
As Figure 20 illustrates, Room Block 1 is the least adequately dated of the three
core room blocks. The northern part of the room block is well dated while rooms in the
southern portion of the room block are: 1) not well dated, and 2) seem to be late in the
construction sequence. Figure 20 also indicates the major obstacle in assigning dates to
rooms in the central section of the room block. The middle of Room Block 1 is highly
72
problematic in that there are no dated rooms to provide terminal dates for the units to the
east and south of Construction Unit l-2a (Figure 19).
2
2
Schematic Representotion:
Principles for Doling Construction Units
Dependent CU
2a
Independent CU
1
a.
1
CU ?0 oddfd independently Irom CU ?
Situaltanof Independency
Independent end Dependent CcmUuclion Unrl?
3
A.D. 1321-1330
(2)
2
A.D 1KB-:
A.D. 1309-1315
(2)
0)
[ • Conduction Unit
Undoled Construction Unit
E 5
AO 1321-1330
(4)
3
AO 1321-1330
2
2
A.D. 1313-1323
A.0 1323
(3)
2
AO 1345-1360-
- Anomokws Dote
Remodeling or Reuse
(2)
0)
AO. *-1330
I.
Prtodple 5
q Pimcfc 6
Figure 21. Schematic Diagram
Illustration of Principles for Dating Undated Construction Units
The best evidence for the establishment date of Room Block 1 comes from the
northern portion of the room block. Unfortunately, the large core unit (Construction Unit
1-la) does not itself contain any dated rooms. The founding date for Room Block 1 is
based on the addition of Construction Unit l-3a, which has a date of A.D. 1310-1321.
Construction Unit l-3a is a key addition in terms of estimating the founding of Room
Block 1. Based on Principle 3, a terminal construction date of A.D. 1321 is assigned to
the core unit as well as Construction Unit l-2a, which the addition of Construction Unit 13a is dependent upon. However, because the addition of Construction Unit l-3a is
separated from the core unit by one construction phase and since the other two room
blocks have founding date ranges earlier than A.D. 1310-1321, the founding date for
Room Block 1 is estimated at A.D. 1305 to A.D. 1321. Establishment of the southern
part of Room Block 1 is difficult to estimate because the earliest dependent date for
construction is A.D. 1375.
From Construction Unit 1-la, growth in Room Block 1 proceeds west and
northward with most of the rooms to the north being added by A.D. 1321 and having
initial dates of A.D. 1310 or earlier. However, since the central portion of the room block
is not well dated, there is no way of accurately assessing the growth trajectory south of
Construction Unit l-2a. Growth in the southern section of Room Block 1 appears to be
different from growth in all other parts of the main ruin (see Figures 3-16). The pattern of
growth reflects what looks to be infilling of the space between the two southern core
construction units by single room additions. Although growth in the southern section of
Room Block 1 appears different than other growth patterns seen in the main pueblo, the
rate of growth for Room Block 1 reflects traditional hypotheses for growth at
Grasshopper (Ciolek-Torrello 1978; Longacre 1975, 1976; Reid 1973, 1989).
74
Based on the Construction Phase Model, and assuming contemporaneity of the
construction units of each phase for which absolute temporal information is not available,
the rate of growth in Room Block 1 appears to be rapid until about A.D. 1325 when
growth tapers off considerably and continues to decline until around A.D. 1385 after
which no construction dates are available. Overall the lack of sufficient dates for much of
Room Block 1 necessitates that the actual growth reconstruction presented here rely
heavily on the assumption that all construction units of a single phase are
contemporaneous (see Chapter 3). Finally, the late dates for the southern portion of
Room Block 1 may indicate a later occupation of this part of the site. This may affect the
distribution of dates in terms of the main occupation of Grasshopper.
The completion date for Room Block 1 is estimated to have been between A.D.
1375 and 1400. Room Block 1 is the only portion of the Main Ruin to yield a date later
than A.D. 1350. The southern section of the room block consists of several single room
additions that are dependent on Construction Unit l-4h which dates to A.D. 1375-1395.
While the late date for this construction unit is the basis for assigning such a late
completion date to Room Block 1, it must be pointed out that this section of the room
block may have been reoccupied late in the occupation of Grasshopper. The dating of
Construction Unit 1 -4h is based on two dates from Room 11, both of which are w dates
and one (1331) is thought to be the result of reuse (Graves 1991:Table IX).
Consequently, an alternative interpretation is that Construction Unit l-4h is constructed
earlier than Graves indicates and the tree-ring samples recovered from this unit represent
either late remodeling or a reoccupation of this portion of the site. A similar explanation is
75
offered for Room 8 (Table 6) which is also dated late by Graves (A.D. 1385-1400). The
argument for later re-use of the southern section of Room Block 1 is further supported by
tree-ring samples from Room 100, which Dean and Robinson (1982) indicate represent a
unique assemblage of wood types in comparison to the other rooms at the site. The use of
non-typical wood species for construction may result from deforestation of the area which
in turn dictated that new types of material be used for construction (Dean and Robinson
1982:47).
Room Block 2 and the Great Kiva
The founding of Room Block 2, based on the available evidence occurs earlier than
either of the other two room blocks of the main pueblo. The 21 room core construction
unit (2-la) is dated by Graves to have been constructed sometime between A.D. 1298 and
1308. Although this is based on a single date from one room, the early founding date is
further supported by the addition of Construction Unit 2-4b, which is added on to two
intervening construction phases and has a date range of A.D. 1311-1321 (Table 6). For
the southern core construction unit (2-lb) no dates are available that would yield an initial
construction date. However, the addition of Construction Unit 2-2d to the north sets the
terminal date for the completion of Construction Unit 2-lb at A.D. 1322. The roofing of
the southern corridor by 1325 provides additional evidence for completion of
Construction Unit 2-lb by this time (Principle 3). Thus, based on an assumption of
contemporaneity of construction phases, and strengthened by the closing of the southern
corridor, the southern core unit is also given an early founding date (Figure 22).
76
Growth in Room Block 2 proceeds from the south of Construction Unit 2-la to
surround the space later to become the Great Kiva and to incorporate Construction Unit
2-lb and the additions to the north of it. The direction of growth implies that the
demarcation and closing off of the Great Kiva was perhaps a driving force behind the
constructions south of the initial founding of the room block. Additions to the north of
Construction Unit 2-la appear to have occurred slightly later than those to the south of
the core unit and probably represent the closing off of Plaza II sometime around A.D.
1340 (see Room Block 3).
The rate of growth for Room Block 2 is similar to that seen in Room Block 1
except that more construction dates for Room Block 2 yield a more accurate depiction of
actual growth for the room block. Most of the construction units in Room Block 2 are
large additions and many have terminal dates. Additionally, there are no additions to
Room Block 2 after Construction Phase 7 in which a three room construction unit is
added north of the southern corridor (Figure 19). Thus, assuming that Construction Phase
7 occurred relatively early in the sequence but sometime after A.D. 1325, all of the
construction in Room Block 2 was likely complete prior to A.D. 1350. The only serious
dating problem that occurs is with the rooms to the far southern end of the room block,
which have no available terminal dates. Although these rooms (Construction Units 2-2e,
2-3f, 2-4d and 2-5e) are assumed to be contemporaneous with the other construction units
of their given phases it is possible that they represent additions after the major growth
period at the site. However, since temporal information is not available for these rooms,
they are included in Growth Period 1 (Figure 22).
ACTUAL GROWTH AT GRASSHOPPER PUEBLO:
MAIN RUIN
- A.O. 1385 - Lolnl Dote
for Construction Timber
(see Groves 1991)
PlOTa • - Greot Krvc Conversion
A D 1320-1360+
Southern Corridor Roofed
A.D *323-1375
Room Block 1 "\
A.D. 1305-1321
Room 0loc* 3 I
A O 1307-13»5 Moundi^ 0o'«
Room Block 2
AD 1298-1308
aj
A.D. 1?98vv - Earliest Dote
for Construction Timber
(see Craves 1991)
LEGEND:
•
•
Growth Period 3 - A O. 1350 - A.D
(Reid's 'Abandonment Phase")
1375/U00
Growth Period 2 - A.D. 1325 - A.D. 1350
(Reid's "Dispersion Phose")
Growth Period 1 - A.D 1298 - AD. 1325
{Reid's 'Expansion Phose": Reid 1973; Reid and Shimodo 19B2)
\
78
Because of its relationship to Room Block 2, the dating of the Great Kiva is
included here in order to round out the discussion of Room Block 2. If the earlier date for
Construction Unit 2-5b is valid (see Table 6), then conversion of Plaza 3 into the Great
Kiva could have occurred any time after A D. 1326 (see Construction Unit 2-5c, Figure
19, Table 6). The late construction date proposed by Graves is based on four dates out of
a total of 20 dated specimens from the Great Kiva. Although the context of the four late
samples is not provided by Graves, it is suggested that the late dates for the Great Kiva
and the late dates from Room 197 may represent a possible remodeling episode sometime
between A.D. 1347 and 1376. Based on the available temporal information, the
conversion of the Great Kiva is given a range of construction between A.D. 1326 and
A.D. 1360+. Additional tree-ring evidence further suggests that the Great Kiva was still
in active use perhaps as late as A.D. 1376 when corroborating evidence for remodeling in
Room 197 suggests that the late dates for the Great Kiva may also represent a late
remodeling episode.
Finally, the completion date for Room Block 2 is difficult to asses due to a lack of
terminal dates for construction in the margins of the room block. The latest dates for
construction in Room Block 2 come from the Great Kiva and Room 197 in Construction
Unit 2-5b which also contains an earlier dated room (Table 6). However, these dates most
likely represent a remodeling episode and are discounted as true construction dates (see
below). The latest dates that appear to represent true room construction occur to the
north of the central core construction unit. Construction Unit 2-4a has a date range of
A.D. 1332-1344 which places the construction of this particular unit and the one that
abuts it (Construction Unit 2-5a) after the period of major growth defined by the current
analysis (Figure 22). The dating of Construction Units 2-4a and 2-5a provide an ideal
example of the breaking apart of construction phases discussed in Chapter 3. While the
79
remaining construction units of these two construction phases are assumed to be
contemporaneous (and occurring before A.D. 1325/1330), the availability of temporal
information enables the two construction units in question to be removed from their
positions in the relative growth reconstruction and to be placed in the proper position in
the actual growth sequence.
Room Block 3 and the Southern Corridor
The best dated single construction unit in the entire site is the central core unit in
Room Block 3 (Construction Unit 3-la). Four of the five rooms in the unit are dated by
18 tree-ring samples (Table 6). The range of construction for the core construction unit
(Construction Unit 3-1 a) falls between A.D. 1307-1315, which sets the founding date for
Room Block 3. Additions to Construction Unit 3-1 a, which have dates of A.D. 13091327 (Construction Unit 3-2c) and A.D. 1307-1315 (Construction Unit 3-3a), further
support the building of Construction Unit 1-3a after A.D. 1300 and prior to A.D. 1315
(Principle 3). Subsequent additions to the early units to the east, west, and to the north of
Construction Unit 1-3 a are difficult to date since there are no dated dependent units until
the addition of Construction Unit 3-8b (A.D. 1331-1340) to the dependent units east of
the Construction Unit 1-3 a (Figure 19). The dating of construction in the northeastern
portion of the room block suggests that Plaza II was fully demarcated sometime between
A.D. 1330-1340, while the date for the final configuration of Plaza II cannot be assigned
absolute dates.
In spite of the problems in dating the northern section of Room Block 3, the
southern portion of the room block is the most tightly dated of the entire site. As was the
case in the dating of Room Block 2, the closing of the corridor is a crucial event in dating
the southern section of Room Block 3. The closure date of A.D. 1325 for the southern
corridor (Dean and Robinson 1982; Graves 1991; Reid 1973) sets the terminal date for
80
Construction Unit 3-5h (Figure 19; Principle 3). This unit immediately abuts a
construction unit (3-4d) dating from A.D. 1323-1333, thus providing a possible two year
range of construction for Construction Unit 3-4d (A.D. 1323-1325) (Principle 3).
Construction Unit 3-7e, which is abutted directly to Construction Unit 3-4d, has a date
range of A.D. 1313-1323 (Table 6, Figure 19). However, because Construction Unit 3-7e
could not have existed prior to the construction of Construction Unit 3-4d (A.D. 13231325), its date range is narrowed to a one year span, in A.D. 1323. In other words, the
coincidence of the terminal date for Construction Unit 3-7e with the initial date of
Construction Unit 3-4d dictates that A.D. 1323 represents the construction date of not
only Construction Unit 3-7e but also Construction Unit 3-4d (see also Construction Unit
3-5f and Construction Unit 3-6e; Principle 5).
Past reconstructions of Grasshopper's growth have placed the roofing of the
southern corridor at around A.D. 1320 (Dean and Robinson 1982:47; Reid 1973,
1989:83; Reid and Shimada 1982:17). Graves' evaluation of the tree-ring dates further
supports this interpretation by suggesting that the corridor was roofed sometime between
A.D. 1320 and A.D. 1325 (Graves 1991). Given the preceding discussion of growth in
the southern part of Room Block 3, the present analysis provides a more precise
estimation of the closure date of the southern corridor by placing the construction of the
roof to within a three year span between A.D. 1323-1325. It is proposed that the
corridor could not have existed before A.D. 1323 and the tree-ring dates indicate that it
could not have been constructed later than A.D. 1325.
The overall growth pattern of Room Block 3 follows that of the two previously
discussed room blocks (Figures 22 and 23). The rate of growth is extremely rapid in the
early stages of construction followed by a continuously more gradual rate of growth until
abandonment. Additions to Construction Unit 3-la proceed very rapidly to the south of
81
the construction unit, finally closing the southern entrance into the community between
A.D. 1323 and 1325. The core units in the northeast section of the room block
(Construction Units 3-lb and 3-lc; Figure 19) that bound Plaza II have no initial temporal
information but are assumed to have been constructed contemporaneously with
Construction Unit 3-la. Rooms added onto these construction units are terminally dated
by the addition of Construction Unit 3-8b, which has a date range from A.D. 1331-1340.
Thus, coupled with similar dates from Room Block 2 (see above), the completion of Plaza
II probably occurred sometime prior to A.D. 1340 or 1350. Finally, the enclosed space
between Room Blocks 2 and 3 known as Plaza I was fully enclosed by no later than A.D.
1325 (Graves 1991; Reid 1973). While the final completion of Plaza I is difficult to
estimate because of a lack of terminal dates for the construction units that are added on
the inside of the plaza, it probably occurred around the same time as the completion of
Plaza II, given that the same construction phases are being added into both spaces (Figure
19).
As was the case in Room Blocks 1 and 2, a complete lack of temporal information
in the marginal areas of the room block makes the completion date for Room Block 3
difficult to estimate. Furthermore, unlike the seven total construction phases of Room
Block 2, there are 14 construction phases in Room Block 3 and a large section of the
room block lacks terminal dates. To the west of Construction Unit 3-la, there are no
dependent dated units and a terminal date of A.D. 1375 is given for all rooms west of
Construction Unit 3-3a (Figures 19 and 20). The northwestern corner of Room Block 3 is
particularly problematic because Construction Phases 4 through 14 are represented yet no
room in this portion of the room block yields a date of any kind. Once again, the
construction phase model is applied in order to provide temporal placement for rooms in
this section of Room Block 3. It is assumed that units of a particular construction phase
were added contemporaneously with all other additions of the same construction phase as
these construction units cannot be assigned temporal information by any other means.
Discussion
Dating Construction
Combining Graves' cutting-date estimates with Reid's construction phase model at
Grasshopper allows for pueblo growth to be verified and refined in terms of the discussion
presented in Chapter 3. While the actual growth reconstruction presented here (see
Figures 22 through 24) is by no means a true actual growth reconstruction (Chapter 3),
several overall trends can be seen. There is strong evidence for rapid early development
prior to A.D. 1325 from all three of the room blocks in the main pueblo. By this time, the
southern corridor is roofed and Plaza I is fully demarcated. Additionally, there is evidence
to suggest that the Great Kiva may have been constructed in Plaza III before A.D. 1330.
The definition of Plaza II by architectural units seems to have occurred somewhat later in
the sequence, but was probably complete by A.D. 1350. The completion dates for the
three room blocks are difficult to estimate due to a lack of terminal dates. The completion
date of A.D. 1375 used in this analysis is a rough estimate based on three lines of
evidence. First, the early constructions at Grasshopper all occurred within a 25-year
period which is well dated (see above). The second line of evidence comes from the
construction units around Plaza II, which (when temporal information is available) all date
in the A.D. 1330 to 1340 range. This suggests that a second 20- to 25-year growth period
occurred beginning before A.D. 1330 and ending with the completion of Plaza II
sometime after A.D. 1340. Finally, the late construction dates from the southern end of
room block one indicate that some construction activity was occurring at around A.D.
1375, which indicates a third, roughly 25-year period, of community development.
83
Relative and Actual Growth
At the completion of the analysis, the revised relative growth curves for
Grasshopper were plotted including the temporal information obtained from the tree-ring
dates. Figures 23 and 24 illustrate the overall pattern and rate of growth at Grasshopper,
first by room block and then for the entire core of the site. The growth curves illustrate
the rapidity of community development at Grasshopper in the first half of the 14th century.
When compared with the relative curves drawn without tree-ring dates (Figures 17 and
18), it is clear that the temporal information causes the curves to compress laterally with
the major growth of the site occurring between 1300 and 1325. Differences in intrasite
growth are presented in Figure 23, which illustrates the fact that while the three room
blocks grow at similar rates, there are subtle differences between the ways in which each
develops through time. Room Block 2 appears to grow very rapidly and then levels off
relatively early in the sequence, while Room Block 3 displays the most gradual rate of
growth. By contrast, Room Block 1 represents the middle range between the two. The
only factor that may be influencing this picture somewhat is the lack of closing dates for
most of the rooms from Room Blocks 1 and 3, while Room Block 2's overall growth
sequence is far more complete.
In building the revised chronology for the main pueblo, the discussion of relative
and actual growth in Chapter 3 played an important part. The construction phase model
with its assumptions of equal intervals for construction phases and contemporaneity of all
units of a single phase was used as a base line for adding the revised tree-ring date
estimates provided by Graves (1991). The principles outlined above were also utilized in
developing an actual growth reconstruction.
84
Growth Pgflod 1
Growth Period 3
105
100
95
90
85
BO
75
70
65
Block
60
Room Block
55
Room Block
50
45
40
30
25
20
15
10
5
o
S
o
8
YEARS A.O
Figure 23.
Actual Growth Curves: Room Blocks 1, 2 and 3
Growth Period 2
Growth Period 1
Growth Period 3
290
2B0
270
250
230
220
210
200
190
180
170
160
ISO
140
130
120
>10
100
90
80
70
60
50
40
30
20
10
0
o
o
s s
o
&
r»
Figure 24. Composite Actual Growth Curve
86
As discussed in Chapter 3, contemporaneity was assumed for all units in a
construction phase, and construction phases were dated based on temporal information for
any construction units in that phase. In most cases, when more than one construction unit
in a phase could be dated, the dates for each construction unit overlapped and thus
reinforced the dating of the phase. By applying the principles for dating individual
construction units discussed above to entire construction phases, it was possible to further
refine the actual growth sequence. Following Principle 3, all construction phases abutted
by dated construction phases were assigned the terminal dates of the added phases. If
individual construction units in that phase could be dated independently and the logical
sequence of construction additions was not violated, those construction units were
removed from the phase in question and placed in the proper position in the actual growth
sequence.
For example, in Room Block 2, Construction Phases 1 - 5 are all well dated to
before A.D. 1330. However, Construction Unit 2-4a dates after A.D. 1330 (Table 6) and
is not included in Growth Period 1 with the rest of Construction Phase 4. Additionally,
Construction Unit 2-5a is dated after A.D. 1330 (Principle 2) and also is not included with
the rest of its construction phase in Growth Period 1. In this case, the addition of
Construction Units 2-4a and 2-5a were not terminally dated by any of the construction
units that were dated earlier, and removing them temporally from Growth Period 1 did not
violate the logical sequence of relative growth. This example illustrates the means by
which relative growth reconstructions can be fashioned into actual growth
reconstructions. It also points out how construction phases can begin to break down with
the addition of temporal information.
To apply the terminology discussed in Chapter 3, the application of Graves'
revised construction dates to Reid's construction phases moves the growth sequence for
87
Grasshopper to a point along the continuum between true relative growth and true actual
growth. Although a true actual growth reconstruction can never be developed for
Grasshopper, this analysis has indicated that certain parts of the site can be precisely dated
in spite of the complete lack of cutting dates
Pueblo Growth
After combining the revised construction dates with the construction phases from
Grasshopper, it became apparent that no significant modifications to the overall site
chronology were necessary. Figure 22 shows the chronology based on the addition of
Graves' cutting-date estimates to the construction phases. Comparison of the chronology
presented here with that devised by Reid (Reid 1973; Reid and Shimada 1982) illustrates
that the two reconstructions are essentially the same with the exception of the A.D. 1275
establishment date, which is based on evidence for a previous occupation underlying the
14th century constructions (Reid 1989). The earliest dated core unit at Grasshopper is
Construction Unit 2-1 a in Room Block 2, which is assigned a construction date estimate
of A.D. 1298 - 1308 (Table 6, Figure 22). Dates for the other core units, whether directly
dated or dated by dependent construction units, all occur after A.D. 1300 (see
Construction Unit 3-la and Construction Unit 1-la, also Appendix B) while the remaining
core units cannot be assigned an initial construction date.
The chronology presented here divides growth at Grasshopper into three growth
periods spanning roughly 25 years each. Growth Period 1, which is equivalent to Reid's
Expansion Phase (A.D. 1300-1330), dates from A.D. 1298 to A.D. 1325/1330 and
includes Construction Phases 1-5 (with the exception of those individual construction
units for which contrary temporal information is available, see above). Growth Period 2,
comparable to Reid's Dispersion Phase (A.D. 1340 to Abandonment Phase) dates between
A.D. 1325/1330 to A.D. 1345/1355 and consists of Construction Phases 6-10 (again
88
with the exception of those units with contrary absolute temporal information). Finally
Growth Period 3 is roughly synonymous with Reid's Abandonment Phase (Dispersion
Phase to A.D. 1400) and dates from A.D 1340/1350 to A.D. 1375/1400 (see above).
Figure 22 illustrates both the current chronology as well as provides a map of the
individual rooms that are associated with each growth period. Examination of Figure 22
makes it clear that early growth at Grasshopper was very rapid and to some extent guided
by a notion of the centrality of public space. The actual growth reconstruction presented
here fits in well both with Graves' new tree-ring dates (Graves 1986, 1991), and also with
Reid's community growth model which sees a period of rapid expansion followed by a
period of slowed growth and eventual abandonment of the community (Reid 1973, 1989;
Reid and Shimada 1982).
Summary and Conclusions
Not only is architecture an artifact produced to fulfill the demands of providing
and maintaining shelter, it also serves to structure the daily interactions of the people who
build and maintain an architectural space. In the early period of southwestern
archaeology, architectural analysis was equivalent in importance to the study of other
artifactual classes. Subsequent to this early period, however, architecture began to lose
some of its importance as an analytical category and became more of a classificatory
device important to the identification of cultural and temporal differences. Foreshadowed
by the early work of C. Mindeleff (1900), Prudden (1903, 1918) and Steward (1937), and
coinciding with the rise of processual archaeology in the 1960's, the focus shifted
somewhat to a view of architecture as an indicator of social relationships. As such, a
community was viewed as a composite of a number of different social groups each
represented by some type of architectural unit. Based on the idea that the added units of
89
construction in a building sequence are representative of a some sort of domestic unit,
several models were developed for determining social relationships and community
growth.
In the 1980's the focus of archaeology shifted back to the notion of architecture as
an artifact. Combining this approach with the models of architectural additions and
community growth developed in the 1960's and 1970's enabled the current analysis to view
the individual construction units as distinct artifacts that together make up the overall
assemblage known as Grasshopper Pueblo.
With the rediscovery of the idea of architecture as artifact, it is now necessary to
develop new methods and techniques for analyzing architectural data. One means of
providing a more sophisticated architectural analysis is through the use of computer
technology. Computer aided drafting programs such as AutoCAD® are not only useful
for quantitative analyses which are structured around the measurement of area, distance,
and volume, they also lend themselves to architectural studies that call for the
classification and control of large sets of data. The use of multiple layers, which can be
turned on and off as necessary for analytical and display purposes, allows for the
classification of architectural entities on the map itself. Additionally, studies involving
stylistic comparisons and analyses of structural and engineering characteristics are also
greatly facilitated by the use of computer technology.
For this analysis, AutoCAD® was used primarily as a device for organizing and
classifying the various architectural data presented above. The availability of computer
drafting allowed for two different growth reconstructions to be presented. For this
analysis, growth was conceptualized as being the product of the relationship between time
and behavior. This characteristic of community growth follows a basic set of principles
that can be used to reconstruct a relative growth sequence. The use of trees as
90
construction materials allows temporal information to then be assigned to the relative
growth reconstruction. This dichotomy in the ways in which temporal information is
represented allows for community growth to also be viewed in two ways: relative growth
and actual growth. Reid's construction phase model was used as a means of
reconstructing relative community growth. To this base, Graves' revised construction
dates were applied as a means of developing an actual growth sequence for the main
pueblo at Grasshopper, and for testing Reid's community growth model. Before this could
be done, however, it was necessary to formulate several definitions and principles for
applying absolute dates to undated construction units.
Fourteen construction phases were presented for the main pueblo at Grasshopper.
These consisted of a total of 127 individual construction units made up of 287 ground
floor rooms. Plotting the cumulative curves for relative growth in the main pueblo
indicated a three-part division in the growth sequence. The first period consisted of
Construction Phases 1-5 and represented rapid early growth. This was followed by an
intermediate period made up of Construction Phases 6 and 7 in which a reduction in the
rate of community growth was indicated. Finally, the third relative growth period was
characterized by very slow growth to no growth at all ending with the abandonment of
Grasshopper.
Because adequate temporal information was difficult to retrieve from the site, the
application of Graves' revised cutting-date estimates to the construction phases at
Grasshopper necessitated that the assumptions of the construction phase model be
maintained where adequate temporal information was not available. As a result, the actual
growth reconstruction presented did not approach a true actual growth reconstruction.
However, the addition of Graves' revised cutting-date estimates to Reid's construction
91
phases did allow for several statements to be made concerning the growth of Grasshopper
Pueblo.
The combination of these two data bases indicated that the actual growth
reconstruction closely approximated Reid's model of community growth at Grasshopper.
A comparison of the relative and actual growth curves indicated that the application of
absolute temporal information to the relative sequence caused the actual growth curves to
compress laterally, portraying a more rapid rate of community growth. The application of
Graves' cutting-date estimates to the construction units at Grasshopper supports previous
analyses by indicating a very rapid growth rate for the community. (Ciolek-Torrello 1978;
Graves et al. 1982; Longacre 1975, 1976; Reid 1973, 1989). Furthermore, this analysis
indicates that community growth at Grasshopper Pueblo may have been more rapid than
has been estimated in the past. In fact, Graves proposes that the site was 60 percent
complete by A.D. 1325 (Graves 1991:108). The present evaluation indicated a 73 percent
level of completion for the main pueblo by A.D. 1325/1330. Although this study focused
only on the three large room blocks of the main pueblo and did not consider the outliers or
second story rooms, Graves' reconstruction also did not account for second story rooms.
Furthermore, the addition of outlying room blocks to the sample probably would not
amount to a 13 percent difference given that there is evidence to suggest that some of the
outliers were already under construction prior to A.D. 1330 (see Graves 1991:Table IX).
Finally, it was not necessary to revise the overall site chronology in any significant
way. Growth was broken down into three growth periods, each spanning about 25 years.
Growth Period 1 began around AD. 1300 and ended between A.D. 1325 and 1330.
Growth Period 2 began after A.D. 1325 and lasted until as late as A.D. 1355. The final
growth period, Growth Period 3, began as early as A.D. 1345 and ended with the final
abandonment of the community sometime after A.D. 1375. This revised picture of
92
community growth closely approximates both that put forth by Graves in his reanalysis of
the tree-ring material from Grasshopper (Graves 1986, 1991) as well as ReicT s
developmental growth model (Reid 1973; Reid and Shimada 1982).
The vast amount of construction data gathered by the University of Arizona
Archaeological Field School at Grasshopper has provided both a wealth of information
concerning the processes of pueblo growth as well as a framework for the application and
use of computerized techniques of analysis. Future research on the architecture of
Grasshopper will benefit from this analysis in that a tighter temporal framework will be
available for individual rooms than has been accessible in the past. This analysis has joined
the revised tree-ring dates established by Graves (1986, 1991) to Reid's construction
phase model at Grasshopper Pueblo (Reid 1973). In the course of this endeavor, it has
become clear that the construction phase model as applied at Grasshopper is a useful
analytical construct for estimating pueblo growth. Although the division of units into
phases of typological contemporaneity in itself does not reflect the true rate of actual
village growth (Figures 3-16, cf. Figure 22), the individual construction units in almost all
instances fit well with Graves' tree-ring estimates. The method used in this analysis has
allowed for the chronological placement of undated units of construction based on their
relationship to dated construction units. While exact dates still can not be applied to all of
the construction units, the method outlined above allows for more formalized application
of the principles necessary to solve the chronological problems addressed in this paper.
The methods and principles set forth in the preceding pages, although mainly relevant to
growth at Grasshopper, could be used in estimating architectural growth at other sites.
This method can be used to obtain similar results from extant data bases, or will be useful
as a blueprint for subsequent research designs that attempt to target community growth.
93
APPENDIX A
GRASSHOPPER: MAIN PUEBLO
TREE RING DATES (After Graves 1986)
8
8
8
11
11
18
18
18
18
18
19
19
19
21
21
22
23
23
23
23
23
23
26
26
26
26
26
26
31
33
33
33
35
35
35
35
35
Terminal
Date
1309w
1371w
1385
1333w
1375w
1204w
1206w
1238w
1269w
1347w
1252w
1301w
131 lw
1267w
1288w
1261w
1247w
1301w
131 lw
1312w
1318w
1319w
1300v
1302w
1303w
1308w
1310w
1313w
1280-H-w
1243w
1305++W
131 lw
1229w
1242w
1305w
1305++W
1306w
Species
Ponderosa
Ponderosa
Ponderosa
Juniper
Juniper
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Juniper
Ponderosa
Ponderosa
Ponderosa
Fir
Fir
Ponderosa
Pinyon
Juniper
Juniper
Ponderosa
Ponderosa
Juniper
Juniper
Juniper
Ponderosa
Juniper
Ponderosa
Juniper
Ponderosa
Ponderosa
Stockpiled
or Reused
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Firewood
Remodeled
94
35
39
39
39
42
42
43
44
44
114
145
153
153
153
153
153
164
183
183
183
183
183
183
183
183
183
187
187
187
187
187
197
197
211
216
231
231
231
231
231
231
Terminal
Date
1310w
1181w
1278w
1282w
1329w
1330w
1199w
1304w
1309w
1301w
1320w
1286w
1312w
1312w
1326w
1346w
1298w
1294w
1303w
1305w
1305w
1309w
1319w
1319w
1323w
1342w
1329w
1330w
1332w
1332v
1332v
1237w
1366w
1313w
1323w
1234w
1291w
1298+r
1302w
1303w
1305w
Species
Ponderosa
Ponderosa
Juniper
Juniper
Juniper
Juniper
Juniper
Ponderosa
Fir
Pinyon
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Fir
Ponderosa
Fir
Fir
Fir
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Fir
Fir
Stockpiled
or Reused
Firewood
Remodeled
+
+
+
+
+
+
+
+
+
+
+
+
+
95
Room
Number
246
269
269
269
269
269
269
269
269
269
269
269
269
270
270
274
274
279
279
279
280
280
280
280
438
438
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Terminal
Date
1228++W
1240w
1273w
1284w
1287+w
1293w
1301w
1302+w
1305w
1307w
1325w
1332w
1343w
1198w
1229w
1231w
1302w
1265+w
1303w
1308w
1268w
1297+w
1309w
1373w
1274+w
1331w
1090w
1 lOlw
1129w
1184w
1190w
1240w
1248w
1253w
1271w
1291w
1293w
1304w
1308w
1311w
1314w
Species
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Fir
Ponderosa
Ponderosa
Fir
Fir
Pinyon
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Juniper
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Stockpiled
or Reused
+
+
+
+
+
+
Firewood
Remodeled
+
+
+
+
+
+
?
?
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
96
Room
Number
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Corridor
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Great Kiva
Terminal
Date
1315w
1315w
1317w
1318w
1318w
1318w
1318w
1318w
1318w
1319w
1320w
1320w
1333w
1190w
1194w
1194w
1200w
1205w
1208w
1209w
1225w
1226w
1253w
1267w
1272w
1273w
1287w
1309w
1321w
1336w
1347w
1347w
1353w
Species
Ponderosa
Ponderosa
Ponderosa
Fir
Ponderosa
Ponderosa
Fir
Ponderosa
Fir
Fir
Ponderosa
Fir
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Ponderosa
Pinyon
Juniper
Pinyon
Pinyon
Pinyon
Pinyon
Fir
Ponderosa
Fir
Ponderosa
Fir
Pinyon
Stockpiled
or Reused
Firewood
Remodeled
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
97
APPENDIX B
DATES FOR ROOMS IN THE MAIN PUEBLO
2
4
5
6
7
8
10
11
12
13
14
15
16
18
19
20
21
22
23
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Date(s)
1305-1375
1375-1400
1375-1400
1305-1375
1305-1375
1385-1400
1305-1375
1375-1395
1313-1338
1311-1338
1320-1330
1305-1375
1313-1338
1313-1338
1311-1321
1320-1330
1319-1330
1319-1330
1319-1330
1313-1322
1313-1322
1311-1325
1311-1325
1320-1375
1320-1375
1310-1325
1311-1325
1310-1321
1310-1321
1320-1375
1310-1375
1310-1375
1310-1321
1310-1321
1310-1321
1310-1321
1310-1321
Unit
Designation
1-lb
1-12a
l-6e
1-lc
1-lb
1-1 la
1-lb
l-4h
2-5c
2-4c
2-5b
1-lb
2-5c
2-5c
2-4b
2-5b
2-3d
2-3d
2-3d
2-2d
2-2d
l-5a
l-5a
l-6a
l-6a
l-4a
l-5a
l-3a
l-3a
l-6a
l-8a
l-7a
1-7b
l-6b
l-6b
l-6b
l-6b
Associated Rooms
7, 10, 15, 104, 107
none
none
none
2, 10, 15, 104, 107
none
2, 7, 15, 104, 107
97
16, 18, 153, 304
156, 303
20, 143, 145, 147, 162, 163, 197
2, 10, 15, 104, 107
12, 18, 153, 304
12, 16, 153, 304
146, 152, 193
14, 143, 145, 147, 162, 163, 197
22, 23
21, 23
21, 22
27
26
29, 33
28, 33
31,36
30,36
none
28, 29
35
34
30,31
none
none
none
41,42, 43
40, 42, 43
40, 41, 43
40,41,42
98
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
Date(s)
1310-1321
1310-1321
1305-1321
1305-1321
1305-1325
1305-1325
1305-1325
1307-1340
1305-1321
1305-1325
1305-1375
1305-1375
1305-1375
1305-1375
1305-1375
1305-1325
1305-1321
1305-1321
1305-1325
1305-1375
1305-1375
1305-1375
1305-1375
1305-1375
1305-1321
1305-1321
1305-1375
1305-1375
1305-1375
1305-1375
1305-1321
1305-1375
1305-1375
1305-1375
1305-1375
1305-1375
1305-1375
1305-1375
1305-1321
1305-1321
1305-1375
Unit
Designation
1-5b
l-4c
1-1a
1-1a
l-2a
l-2a
l-2a
3-7b
1-1a
l-2b
l-3b
1 -3 b
1-3 b
l-4d
1 -3 b
1-2b
1-la
1-1a
1-2b
1 -3 b
l-4d
l-5d
l-4e
l-4e
1-la
1-la
l-2c
l-3c
l-3c
l-3c
1-la
l-2c
l-4f
l-5e
l-6c
l-7c
l-6d
l-3d
1-la
1-la
l-2d
Associated Rooms
none
none
47, 52, 60,61,68, 69, 74, 82, 83
46, 52, 60, 61, 68, 69, 74, 82, 83
49, 50
48, 50
48, 49
none
46, 47, 60, 61, 68, 69, 74, 82, 83
59, 62
55, 56, 58, 63
54, 56, 58, 63
54, 55, 58, 63
64
54, 55, 56, 63
53, 62
46, 47, 52, 61, 68, 69, 74, 82, 83
46, 47, 52, 60, 68, 69, 74, 82, 83
53, 59
54, 55, 56, 58
57
none
67
66
46, 47, 53, 60, 61, 69, 74, 82, 83
46, 47, 53, 60, 61, 68, 74, 82, 83
75
72, 73
71,73
71, 72
46, 47, 53, 60, 61, 68, 69, 82, 83
70
none
none
none
none
none
none
46, 47, 53, 60, 61, 68, 69, 74, 83
46, 47, 53, 60, 61, 68, 69, 74, 82
85, 86, 87
99
127
128
129
130
131
Date(s)
1305-1375
1305-1375
1305-1375
1305-1375
1305-1375
1305-1400
1375-1400
1305-1395
1305-1395
1305-1395
1375-1400
1375-1395
1375-1400
1375-1400
1375-1400
1375-1400
1375-1400
1305-1400
1305-1375
1375-1400
1375-1400
1305-1375
1375-1400
1310-1375
1310-1375
1310-1375
1310-1375
1310-1375
1310-1375
1298-1375
1298-1375
1298-1375
1298-1375
1298-1375
1298-1375
1298-1322
Unit
Designation
1-2d
1-2d
1 -2d
l-4g
l-4g
l-5f
l-5g
l-3e
l-3e
l-3e
l-5h
l-4h
1-10a
l-9a
l-7d
1 -8b
1 -9b
l-2e
1-1 b
l-10b
1-1 lb
1-lb
1-13a
l-4b
1-4b
l-5c
l-5c
l-4b
l-4b
2-4d
2-5d
2-3 e
2-2e
2-2e
2-2e
2-lb
132
1298-1322
2-lb
133
1298-1322
2-1 b
85
86
87
88
89
90
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
118
119
120
121
122
123
125
126
Associated Rooms
84, 86, 87
84, 85, 87
84, 85, 86
89
88
90
none
94, 95
93, 95
93,94
none
11
none
none
none
none
none
none
2, 7, 10, 15, 107
none
none
2, 7, 10, 15, 104
none
119, 122, 123
118, 122, 123
121
120
118, 119, 123
118, 119, 122
none
none
none
129, 130
128, 130
128, 129
132, 133, 134, 135,
140, 142
131, 133, 134, 135,
140, 142
131, 132, 134, 135,
140, 142
100
Room
134
Date(s)
1298-1322
Unit
Designation
2-lb
135
1298-1322
2-lb
136
1298-1322
2-lb
137
1298-1322
2-lb
138
1298-1322
2-lb
139
1298-1322
2-lb
140
1298-1322
2-lb
141
142
1298-1330
1298-1322
2-2c
2-lb
143
144
145
146
147
148
1320-1330
1320-1375
1320-1330
1311-1321
1320-1330
1298-1308
2-5b
2-6b
2-5b
2-4b
2-5b
2-la
149
1298-1308
2-la
150
1298-1308
2-la
151
152
153
154
155
156
157
158
1298-1321
1311-1321
1313-1338
1311-1338
1311-1321
1311-1321
1311-1321
1298-1321
2-2b
2-4b
2-5c
2-4c
2-3c
2-3c
2-3c
2-2b
Associated Rooms
131, 132, 133, 135, 136, 137, 138, 139,
140, 142
131, 132, 133, 134, 136, 137, 138, 139,
140, 142
131, 132, 133, 134, 135, 137, 138, 139,
140, 142
131, 132, 133, 134, 135, 136, 138, 139,
140, 142
131, 132, 133, 134, 135, 136, 137, 139,
140, 142
131, 132, 133, 134, 135, 136, 137, 138,
140, 142
131, 132, 133, 134, 135, 136, 137, 138,
139, 142
none
131, 132, 133, 134, 135, 136, 137, 138,
139, 140
14, 20, 145, 147, 162, 163, 197
none
14, 20, 143, 147, 162, 163, 197
19, 152, 193
14, 20, 143, 145, 162, 163, 197
149, 150, 159, 160, 161, 164, 165, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 150, 159, 160, 161, 164, 165, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 159, 160, 161, 164, 165, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
158, 171
19, 146, 193
12, 16, 18, 304
13, 303
156, 157
155,157
155, 156
151, 171
159
Date(s)
1298-1308
Unit
Designation
2-la
160
1298-1308
2-la
161
1298-1308
2-la
162
163
164
1320-1330
1320-1330
1298-1308
2-5 b
2-5b
2-la
165
1298-1308
2-la
166
1298-1308
2-la
167
1298-1308
2-la
168
1298-1308
2-la
169
1298-1308
2-la
170
1298-1308
2-la
171
172
173
1298-1321
1323-1375
1323-1375
1298-1308
2-2b
2-3b
2-3b
2-la
Associated Rooms
148, 149, 150, 160, 161, 164, 165, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 161, 164, 165, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 164, 165, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
14, 20, 143, 145, 147, 163, 197
14, 20, 143, 145, 147, 162, 197
148, 149, 150, 159, 160, 161, 165, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 166,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
167, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 168, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 169, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 170, 174, 175, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 174, 175, 176, 177,
178, 179, 180, 181
151, 158
173, 182
172, 182
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 175, 176, 177,
178, 179, 180, 181
102
Unit
175
Date(s)
1298-1308
2-la
176
1298-1308
2-la
177
1298-1308
2-la
178
1298-1308
2-la
179
1298-1308
2-la
180
1298-1308
2-la
181
1298-1308
2-la
182
183
184
185
1323-1375
1323-1333
1323-1333
1323-1333
1323-1333
1332-1344
1332-1344
1323-1344
1323-1344
1332-1375
1332-1375
1311-1321
1320-1330
1323- 1375
1323- 1375
1323- 1325
1323-1325
1323-1375
1323
2-3b
2-2a
2-2a
2-2a
2-2a
2-4a
2-4a
2-3a
2-3a
2-5a
2-5a
2-4b
2-5b
3-10d
3-8e
3-5h
3-5h
3-9f
3-7e
186
187
188
189
190
191
192
193
197
201
202
203
204
205
206
Associated Rooms
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 174, 176, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 174, 175, 177,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 174, 175, 176,
178, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 174, 175, 176,
177, 179, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 174, 175, 176,
177, 178, 180, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 174, 175, 176,
177, 178, 179, 181
148, 149, 150, 159, 160, 161, 164, 165,
166, 167, 168, 169, 170, 174, 175, 176,
177, 178, 179, 180
172, 173
184, 185, 186
183, 185, 186
183, 184, 186
183, 184, 185
188
187
299, 300, 301, 302, 190
299, 300, 301, 302, 189
192
192
19, 146, 152
14, 20, 143, 145, 147, 162, 163
none
none
204, 273, 289
203, 273, 289
none
211,215,218
103
Room
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
Date(s)
1323
1323-1375
1323-1375
1323-1375
1323
1323
1323-1375
1323-1375
1323
1323
1323-1375
1323
1323
1323
1323
1323-1375
1323
1323
1307-1340
1307-1323
1307-1317
1323-1375
1307-1375
1307-1375
1307-1317
1307-1375
1307-1375
1307-1317
1323-1375
1307-1375
1307-1375
1307-1317
1323-1375
1323-1375
1323-1375
1323-1375
1307-1375
1307-1375
1307-1375
1307-1315
1309-1328
Unit
Designation
3-4d
3-6f
3-10c
3-8d
3-7e
3-4d
3-6f
3-9e
3-7e
3-4d
3-5g
3-7e
3-4d
3-6e
3-4d
3-6d
3-5f
3-4d
3-5e
3-4c
3-3b
3-5d
3-13c
3-5c
3-3b
3-12c
3-10b
3-3b
3-5d
3-1 lb
3-9b
3-2b
3-5d
3-6c
3-7d
3-6f
3-12b
3-10a
3-8a
3-la
3-2c
Associated Rooms
212,216,219, 221,224
213, 242
none
none
206, 215, 218
207,216,219, 221,224
208, 242
none
206,211,218
207,212,219, 221,224
none
206,211, 215
207,212,216, 221,224
none
207,212,216,219, 224
none
none
207, 212, 216, 219, 221
none
none
231,234
235, 239, 260
none
none
227, 234
none
none
227, 231
228, 239, 260
none
none
none
228, 235, 260
none
none
208, 213
none
none
252
269, 270, 278, 279
248, 249, 280
104
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
Date(s)
1309-1328
1309-1328
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1323-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
1307-1375
after 1307
1307-1315
1307-1315
1307-1375
1307-1340
1323-1325
1307-1315
1307-1315
1307-1315
1307-1315
1307-1315
1307-1315
1309-1328
1307-1340
1307-1340
1307-1340
1307-1340
1307-1340
1307-1340
1307-1340
1307-1340
Unit
Designation
3-2c
3-2c
3-13b
3-1 la
3-8a
3-7a
3-5a
3-14a
3-12a
3-9a
3-7a
3-5a
3-5d
3-13a
3-9a
3-6a
3-6a
3-5a
3-5a
3-5a
3-4b
3-la
3-la
3-5a
3-4a
3-5h
3-3a
3-2a
3-3a
3-2a
3-la
3-la
3-2c
3-1 b
3-1 b
3-1 b
3-1 b
3-5b
3-6b
3-6b
3-lc
Associated Rooms
247, 249, 280
247, 248, 280
none
none
245
258
259, 265, 266, 267, 271
none
none
262
253
254, 265, 266, 267, 271
228, 235, 239,
none
257
264
263
254, 259, 266, 267, 271
254, 259, 265, 267, 271
254, 259, 265, 266, 271
none
246, 270, 278, 279
246, 269, 278, 279
254, 259, 265, 266, 267
none
203, 204, 289
276
277
274
275
246, 269, 270, 279
246, 269, 270, 278
247, 248, 249
282, 283, 284
281, 283, 284
281,282, 284
281,282, 283
none
287
286
292, 293, 294, 295
105
Room
289
290
291
292
293
294
295
297
298
299
300
301
302
303
304
305
306
307
308
438
439
440
Corridor
Great
Kiva
Date(s)
1323-1325
1323-1325
1307-1340
1307-1340
1307-1340
1307-1340
1307-1340
1323-1375
1323-1375
1323-1344
1323-1344
1323-1344
1323-1344
1311-1338
1313-1338
1320-1375
1320-1375
1320-1375
1320-1375
1331-1340
1331-1340
1331-1375
1323-1325
1320-1360+
Unit
Designation
3-5h
3-6g
3-7c
3-lc
3-lc
3-lc
3-lc
3-8c
3-9d
2-3 a
2-3 a
2-3 a
2-3a
2-4c
2-5c
2-7a
2-6a
2-7a
2-7a
3-8b
3-8b
3-9c
n/a
n/a
Associated Rooms
203, 204, 273
none
none
288, 293, 294, 295
288, 292, 294, 295
288, 292, 293, 295
288, 292, 293, 294
none
none
300, 301, 302, 189,
299, 301, 302, 189,
299, 300, 302, 189,
299, 300, 301, 189,
13,156
12, 16, 18, 153
307, 308
none
305, 308
305, 307
439
438
none
none
none
190
190
190
190
106
REFERENCES CITED
Bannister, Bryant
1962 The Interpretation of Tree-Ring Dates. American Antiquity 27:508-514.
Binford, Lewis R.
1962 Archaeology as Anthropology. American Antiquity 28(2):217-225.
Bullard, William R., Jr.
1962 The Cerro Colorado Site and Pithouse Architecture in the Southwestern United
States Prior to A.D. 900. Papers of the Peabody Museum of American
Archaeology and Ethnology, vol. 44, no. 2. Cambridge.
Ciolek-Torrello, Richard S.
1978 A Statistical Analysis of Activity Organization: Grasshopper Pueblo, Arizona.
MS, Doctoral Dissertation, Department of Anthropology, University of Arizona,
Tucson. Ann Arbor: University Microfilms.
1985 A Typology of Room Function at Grasshopper Pueblo, Arizona. Journal of
Field Archaeology 12:41-63.
Colton, Harold S. and Hargrave, Lyndon L.
1937 Handbook of Northern Arizona Pottery Wares. Museum of Northern Arizona
Bulletin 11. Flagstaff.
Cushing, Frank Hamilton
1890 Preliminary Notes on the Origin, Working Hypothesis, and Primary Researches
of the Hemenway Southwestern Archaeological Expedition. Congres
International des Americanistes, Compte-rendu de la Septieme Session, pp. 151194, Berlin.
Dean, Jeffrey S.
1969 Chronological Analysis of Tsegi Phase Sites in North-Eastern Arizona. Papers
of the Laboratory of Tree-Ring Research 3. University of Arizona Press, Tucson.
1978 Independent Dating in Archaeological Analysis. In Advances in Archaeological
Method and Theory 1, ed. by Michael B. Schiffer, pp. 223-255. Academic Press,
New York.
1986 Dendrochronology. In Dating and Age Determination of Biological Materials,
ed. by Michael R. Zimmerman and Lawrence Angel, pp. 126-165. Croom Helm,
London
107
Dean, Jeffrey S. and William J. Robinson
1982 Dendrochronology of Grasshopper Pueblo. In "Multidisciplinary Research at
Grasshopper Pueblo, Arizona", ed. by W. A. Longacre, S. J. Holbrook, M. W.
Graves. Anthropological Papers of the University of Arizona 40:46-60. University
of Arizona Press, Tucson.
Doelle, William H.
1985 The Southern Tucson Basin: Rillito-Rincon Subsistence, Settlement, and
Community Structure. In "Proceeding of the 1983 Hohokam Symposium," ed. by
Alfred E. Dittert, Jr., and Donald E. Dove, pp. 183-198. Arizona Archaeological
Society Occasional Paper 2, Phoenix.
Doelle, William H., Frederick W. Huntington, and Henry D. Wallace
1987 Rincon Phase Community Reorganization the Tucson Basin. In The Hohokam
Village: Site Structure and Organization, ed. by David E. Doyel, pp. 71-96.
American Association for the Advancement of Science, Denver.
Douglas, A. E.
1935 Dating Pueblo Bonito and Other Ruins of the Southwest. National Geographic
Society Contributed Technical Papers, 1. Washington, D.C.
Ferguson, T. J.
1992 Structural Change in Historic Zuni Architecture and Society. MS, Paper
Presented at the 57th Annual Meeting of the Society for American Archaeology,
Pittsburg, PA, April 10, 1992.
1993 Historic Zuni Architecture and Society: A Structural Analysis. MS, Doctoral
Dissertation. Department of Anthropology, University of New Mexico.
Ferguson, T. J., Barbara J. Mills, and Calbert Seciwa
1990 Contemporary Zuni Architecture and Society. In Pueblo Style and Regional
Architecture, ed. by Nicholas C. Markovich, Wolfgang F.E. Preiser, and Fred G.
Strum, pp. 103-121. Van Nostrand Reinhold, New York.
Fewkes, Jesse Walter
1907 Excavations at Casa Grande, Arizona, in 1906-1907. Smithsonian
Miscellaneous Collections 50:289-329.
1911 Antiquities of Mesa Verde National Park: ClifFPalace. Bureau of American
Ethnology, Bulletin 51, Smithsonian Institution, Washington.
1912 Casa Grande, Arizona. Bureau of American Ethnology Annual Report 28, pp.
25-180, Smithsonian Institution, Washington.
108
Gilman, Patricia A.
1987 Architecture as Artifact: Pit Structures and Pueblos in the American Southwest.
American Antiquity 52:538-564.
Graves, Michael W.
1986 Room Construction Strategies and Dynamics at Grasshopper Pueblo and Other
Late Prehistoric Sites in East Central Arizona. MS, Paper Presented at the
Fourth Mogollon Conference, Tucson, Arizona, October 1986.
1991 Estimating Ring Loss on Tree-Ring Specimens from East-Central Arizona:
Implications for Prehistoric Pueblo Growth at the Grasshopper Ruin. Journal of
Quantitative Anthropology 3:83-115.
Graves, Michael W., Sally J. Holbrook, and William A. Longacre
1982 Aggregation and Abandonment at Grasshopper Pueblo: Evolutionary Trends in
the Late Prehistory of East-Central Arizona. In "Multidisciplinaiy Research at
Grasshopper Pueblo, Arizona", ed. by W. A. Longacre, S. J. Holbrook, M. W.
Graves. Anthropological Papers of the University of Arizona 40:110-121.
University of Arizona Press, Tucson.
Gregory, David A.
1987 The Morphology of Platform Mounds and the Structure of Classic Period
Hohokam Sites. In The Hohokam Village: Site Structure and Organization, ed.
by David E. Doyel, pp. 183-210.
Haury, Emil W.
1928 The Succession of House Types in the Pueblo Area. MS, Masters Thesis,
University of Arizona, Tucson.
1934 The Canyon Creek Ruin and the Cliff Dwellings of the Sierra Ancha. Medallion
Papers 14. Globe, Arizona: Gila Pueblo.
1936 The Mogollon Culture of Southwestern New Mexico. Medallion Papers 20.
Globe, Arizona: Gila Pueblo.
1985 Mogollon Culture in the Forestdale Valley, East-Central Arizona. University
of Arizona Press, Tucson.
Hegmon, Michelle
1989 Social Integration and Architecture. In "The Architecture of Social Integration
in Prehistoric Pueblos", ed. by William D. Lipe and Michelle Hegmon, pp. 5-14.
109
Occasional Papers of the Crow Canyon Archaeological Center No. 1, Cortez
Colorado.
Henderson, T. Kathleen
1987 Growth of a Hohokam Village. In The Hohokam Village: Site Structure and
Organization, ed. by David E. Doyel, pp. 97-125, Southwestern and Rocky
Mountain Division of the American Association for the Advancement of Science.
Hillier, Bill and Julienne Hanson
1984 The Social Logic of Space. London. Cambridge University Press.
Hunter-Anderson, Rosalind
1977 A Theoretical Approach to the Study of House Form. In For Theory Building
in Archaeology, ed. by L. R. Binford, pp. 287-315. Academic Press, New York.
Jelinek, Arthur J.
1976 Form, Function, and Style in Lithic Analysis. In Cultural Change and
Continuity: Essays in Honor of James Bennett Griffin, ed. by Charles E. Cleland,
pp. 19-33. Academic Press, New York.
Judge, W. James
1991 Chaco: Prehistory and the Regional System. In "Chaco and Hohokam:
Prehistoric Regional Systems in the American Southwest," ed. by Patricia L.
Crown and W. James Judge, pp. 11-30, School of American Research Advanced
Seminar Series. School of American Research Press, Santa Fe.
Kidder, Alfred Vincent
1927 Southwestern Archaeological Conference. Science 68:489-491.
Kidder, Alfred Vincent, and Anna O. Shepard
1936 The Pottery of Pecos (vol. 2). The Glaze-Paint, Culinary and other Wares.
Papers of the Southwest Expedition 7. Phillips Academy, Andover, and Yale
University Press, New Haven.
Lekson. Stephen H.
1986 Great Pueblo Architecture of Chaco Canyon, New Mexico. University of New
Mexico Press, Albuquerque.
1991 Settlement Pattern and the Chaco Region. In "Chaco and Hohokam: Prehistoric
Regional Systems in the American Southwest," ed. by Patricia L. Crown and W.
James Judge, pp. 31-55, School of American Research Advanced Seminar Series.
School of American Research Press, Santa Fe.
110
Lekson, Stephen H., Thomas C. Windes, John R. Stein, and W. James Judge
1988 The Chaco Canyon Community. Scientific American 256(7): 100-109.
Longacre, William A.
1975 Population Dynamics at Grasshopper Pueblo, Arizona. In "Population Studies
in Archaeology and Biological Anthropology: A Symposium," ed. by Alan C.
Swedlund, pp. 71-74. Memoirs of the Society for American Archaeology 30.
1976 Population Dynamics at Grasshopper Pueblo, Arizona. In Demographic
Anthropology: Quantitative Approaches, ed. by Ezra B. W. Zubrow, pp. 169-183.
University of New Mexico Press, Albuquerque.
Longacre, William A., and Michael W. Graves
1982 Multidisciplinary Studies at Grasshopper Pueblo. In "Multidisciplinary Research
at Grasshopper Pueblo, Arizona", ed. by W. A. Longacre, S. J. Holbrook, M. W.
Graves. Anthropological Papers of the University of Arizona 40:1-4. University of
Arizona Press, Tucson.
Longacre, William A. and J. Jefferson Reid
1974 The University of Arizona Archaeological Field School at Grasshopper: Eleven
Years of Multidisciplinary Research and Teaching. The Kiva 40(l-2):3-38.
McGuire, Randall H. and Michael B. Schiffer
1983 A Theory of Architectural Design. Journal of Anthropological Archeology
2:277-303.
Mindeleff, Cosmos
1900 Localization of Tusayan Clans. In 19th Annual Report of the Bureau of
American Ethnology for the Years 1897-1898, Pt. 2, pp. 635-653, Washington.
Mindeleff, Victor
1891 A Study of Pueblo Architecture: Tusayan and Cibola. Annual Report of the
Bureau of American Ethnology 8:3-228. Washington, DC.: Smithsonian
Institution.
Plog, Stephen E.
1977 A Multivariate Approach to the Explanation of Ceramic Variation. Doctoral
Dissertation, University of Michigan, Ann Arbor. Ann Arbor: University
Microfilms.
1980 Stylistic Variation in Prehistoric Ceramics: Design Analysis in the American
Southwest. Cambridge University Press, Cambridge.
Ill
Powers, Robert P., William B. Gillespie, and Stephen H. Lekson
1983 The Outlier Survey: A Regional View of Settlement in the San Juan Basin.
Reports of the Chaco Center, no. 3. National Park Service, Division of Cultural
Research, Albuquerque.
Prudden, T. Mitchel
1903 The Prehistoric Ruins of the San Juan Watershed in Utah, Arizona, Colorado,
and New Mexico. American Anthropologist, n.s., 5(2):224-288.
1918 A Further Study of the Prehistoric Small House Ruins in the San Juan
Watershed. Memoirs of the American Anthropological Association, vol. 5, No. 1
Reed, Erik Keilerman
1948 Western Pueblo Archaeological Complex. El Palacio 55(1):9-15.
1956 Types of Village-Plan Layouts in the Southwest. In "Prehistoric Settlement
Patterns in the New World", ed. by G. R. Willey Viking Fund Publications in
Anthropology 23:11-17. Wenner-Gren Foundation for Anthropological Research,
Inc.
Reid, J. Jefferson
1973 Growth and Response to Stress at Grasshopper Pueblo, Arizona. Doctoral
dissertation, University of Arizona, Tucson. Ann Arbor: University Microfilms.
1989 A Grasshopper Perspective on the Mogollon of the Arizona Mountains. In
Dynamics of Southwestern Prehistory, ed. by G. Gumerman and L. Cordell, pp.
65-97. Smithsonian Institution Press, Washington.
Reid, J. Jefferson, Michael B. Schiffer, and Jeffrey M. Neff
1975 Archaeological Considerations in Intrasite Sampling. In Sampling in
Archaeology, ed. by James W. Mueller, pp. 209-224. University of Arizona Press,
Tucson.
Reid, J. Jefferson and Izumi Shimada
1982 Pueblo Growth at Grasshopper: Methods and Models. In "Multidisciplinary
Research at Grasshopper Pueblo, Arizona", ed. by W. A. Longacre, S. J.
Holbrook, M. W. Graves. Anthropological Papers of the University of Arizona
40:12-18. University of Arizona Press, Tucson.
Rice, Prudence M.
1987 Pottery Analysis: A Sourcebook. University of Chicago Press, Chicago.
112
Rinaldo, John B.
1964 Architectural Details, Carter Ranch Pueblo. In "Chapters in the Prehistory of
Eastern Arizona II," by Paul S. Martin, John B. Rinaldo, William A. Longacre,
Leslie G. Freeman, Jr., James A. Brown, Richard H. Hevly, and M. E. Cooley, pp.
15-58. Fieldiana: Anthropology 55.
Roberts, Frank H. H., Jr.
1935 A Survey of Southwestern Archaeology. American Anthropologist 37(1): 1-35.
Rohn, Arthur
1965 Postulation of Socio-economic Groups from Archaeological Evidence. In
"Contributions of the Wetherill Mesa Archaeological Project," compiled by
Douglas Osborne, pp. 65-69. Society for American Archaeology Memoir 19.
1971 Mug House. Washington: National Park Service.
Roys, Lawrence
1936 Masonry of Lowry Ruin and of the Southwest. In "Lowry Ruin in
Southwestern Colorado," by Paul S. Martin, pp. 115-142. Fieldiana:
Anthropology 23(1).
Scarborough, Robert, and Izumi Shimada
1974 Geological Analysis of Wall Composition at Grasshopper with Behavioral
Implications. The Kiva 40(l-2):49-66.
Schiffer, Michael B.
1987 Formation Processes of the Archaeological Record. University of New Mexico
Press, Albuquerque.
Sires, Earl W.
1983 Archaeological Investigations at Las Fosas (AZ U:15:19): A Classic Period
Settlement on the Gila River. In "Hohokam Archaeology Along the Salt-Gila
Aqueduct, Central Arizona Project, ed. by L. S. Teague and P. S. Crown. Arizona
State Museum Archaeological Series 150(6). Tucson.
1987 Hohokam Architectural Variability and Site Structure During the SedentaryClassic Transition. In The Hohokam Village: Site Structure and Organization,
ed. by David E. Doyel, pp. 171-182, Southwestern and Rocky Mountain Division
of the American Association for the Advancement of Science.
Steward, Julian H.
1937 Ecological Aspects of Southwestern Society. Anthropos 32:87-104.
113
Vivian, R. Gwinn
1970 Aspects of Prehistoric Society in Chaco Canyon, New Mexico. Doctoral
Dissertation, University of Arizona, Tucson. Ann Arbor: University Microfilms.
1990 The Chacoan Prehistory of the San Juan Basin. Academic Press, New York.
Wheat, Joe Ben
1955 Mogollon Culture Prior to A.D. 1000. Memoirs of the American
Anthropological Association 82, Memoirs of the Society for American
Archaeology 10.
Wilcox, David R.
1975 A Strategy for Perceiving Social Groups in Puebloan Sites. In "Chapters in the
Prehistory of Arizona IV". by Paul S. Martin, Ezra B. W. Zubrow, Daniel C.
Bowman, David A. Gregory, John A. Hanson, Michael B. SchifFer, and David R.
Wilcox, pp. 120-159. Fieldiana: Anthropology 65, Field Museum of Natural
History, Chicago.
1982 A Set-Theory Approach to Sampling Pueblos: The Implications of Room-Set
Additions at Grasshopper Pueblo. In "Multidisciplinary Research at Grasshopper
Pueblo, Arizona", ed. by W. A. Longacre, S. J. Holbrook, M. W. Graves.
Anthropological Papers of the University of Arizona 40:19-27. University of
Arizona Press, Tucson.
Wilcox, David R., T. R. McGuire and C. Sternberg
1981 Snaketown Revisited. Arizona State Museum Archaeological Series 155.
Tucson.
Wilcox, David R. and Lynette O. Shenk
1977 The Architecture of Casa Grande and its Interpretation. Arizona State Museum
Archaeological Series 115. Tucson.
Willey, Gordon R. and Jeremy A. Sabloff
1993 A History of American Archaeology. (3rd Edition) W. H. Freeman and
Company, New York.

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project