Settlement patterns and political geography of the Saladoid and Ostionoid peoples of south-central Puerto Rico

Material Information

Settlement patterns and political geography of the Saladoid and Ostionoid peoples of south-central Puerto Rico an exploration of prehistoric social complexity on a regional level
Torres, Joshua M
Publication Date:
Physical Description:
xii, 151, CLII-CLXXVII leaves : illustrations ; 28 cm


Subjects / Keywords:
Land settlement patterns, Prehistoric -- Puerto Rico ( lcsh )
Indians of the West Indies -- Social conditions -- Puerto Rico ( lcsh )
Prehistoric peoples -- Puerto Rico ( lcsh )
Social evolution ( lcsh )
Indians of the West Indies -- Social conditions ( fast )
Land settlement patterns, Prehistoric ( fast )
Prehistoric peoples ( fast )
Social evolution ( fast )
Puerto Rico ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves CLXVII-CLXXVII).
General Note:
Department of Anthropology
Statement of Responsibility:
by Joshua M. Torres.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
48804881 ( OCLC )
LD1190.L43 2001m .T67 ( lcc )

Full Text
Joshua M. Torres
B.A., University of Colorado at Denver, 1998
A thesis presented to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts

This thesis for the Master of Arts
degree by
Joshua Marion Torres
has been approved

Torres, Joshua M. (M.A., Anthropology)
Settlement Patterns and Political Geography of the Saladoid and Ostionoid Peoples of
South Central Puerto Rico: An Exploration of Prehistoric Social Complexity on a
Regional Level
Thesis directed by Professor Tammy Stone
The rise of social complexity and the centralization of formalized power, in
prehistoric contexts, represent a fascinating and important area of archaeological
research. Not only does this research elucidate the processes responsible for social
change, but also contributes to our understanding of how societies form dynamic
multi-faceted social organizations. The processes responsible for these changes, as
well as the concept of social complexity have been explained from a variety of
theoretical perspectives within anthropological research contexts. However, the
utilization of objective and operational frameworks within which to examine social
complexity is often neglected.
The research presented here focuses on a practical approach for exploring
social complexity on a regional level. To facilitate this endeavor, the study here
utilizes settlement patterning to elucidate the dynamics of prehistoric social
organization for south central Puerto Rico. Emphasis is placed on three distinct time
periods within the ceramic age of Puerto Rican prehistory existing between 300 B.C.
and 1492 A.D. In the pursuit of elucidating the regional socio-political dynamic of
the pre-Columbian peoples of south central Puerto Rico, settlement pattern oriented
methods are employed through the use of Geographic Information Systems (GIS)
technology to examine the spatial relationships of archaeological site distributions
based on discrete site characteristics and aspects of the regional environmental
setting. In doing so, the present study not only provides a practical approach for
exploring social complexity but also facilitates current understanding regarding the
socio-political development and organization of these peoples.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.

There are many people to thank who assisted with this research. First, I would like to
acknowledge and thank Hugh Tostenson, Miguel Bonini and Aida Belen of the San
Juan State Historic Preservation Office for being extremely receptive to my research
interests and providing me with the majority of the data utilized for this study. I
would also like to thank the staff at the Institute of Puerto Rican Culture for also
assisting in the acquisition of site data. Special thanks to Dr. Tammy Stone, Dr.
David Tracer, and Dr. John Brett of the University of Colorado at Denver for their
patience, guidance and advice during the course of this study. Also special thanks to
Dr. Lee Newsome for making herself available for advice and as a source of
information. Finally, I would like to thank Dr. Antonio Curet for introducing me to
the world of Caribbean archaeology and without whom this study would not have
been possible.

I dedicate this thesis to my wife, son and family for their love and support

1. Introduction..........................................................1
1.1 Scope.................................................................3
2. Theoretical DiscussionThe Complexity of Complexity...................6
2.1 Conceptualization for the Examination of Social Complexity...............14
2.1.1 Differentiation and Integration....................................15
2.1.2 Boundedness and Scale..............................................18
2.2 Social Complexity and Prehistoric Puerto Rico......................22
3. Environmental Context................................................23
3.1 Physiographic Zones of South Central Puerto Rico.....................25
3.2 Subsistence of the Prehistoric Inhabitants of South Central Puerto Rico.30
4. Cultural Context: A Discussion of the Prehistory of Puerto Rico..........34
4.1 Period II -The Saladoid Series.......................................40
4.2 Period III -The Elenan and Ostionan Ostionoid Series....................43
4.3 Period IV-The Chican Ostionoid......................................48
5. Research Design and Methodology......................................50
5.1 Settlement Pattern Methodology.......................................51

5.1.2 Geographic Information Systems in Archaeological
Research Contexts...................................................55
5.2 Data................................................................57
5.2.1 Site Data...........................................................57
5.2.2 Topographic Data....................................................62
5.2.3 Soils...............................................................64
5.2.4 Physiographic Regions...............................................67
5.3 Methodology of Data Analysis..........................................68
5.3.1 Examination of Environmental Variables..............................70
5.3.2 Examination of Socio-Political Variables............................75
6. Analyses and Discussion................................................78
6.1 Environmental ContextsResource Use, Settlement, and Change Through
6.1.1 Land Use, Economic Boundaries, and Settlement Morphology...........94
6.1.2 Site Catchments and Soil Productivity..............................102
6.2 Regional Integration...............................................Ill
6.2.1 Modeling Regional Polities.........................................112
6.2.2 Intra-polity Dynamics..............................................118
6.3 Site Distributions and Regional Social Complexity....................128
6.3.1 Period II Site Distributions.......................................129
6.3.2 Period in Site Distributions.......................................133
6.3.3 Period IV Site Distributions.......................................139

7. ConclusionPrehistoric Social Complexity and South Central Puerto
7.1 Implications for the Archaeological Research of Puerto Rico........148
A. Soil Data.........................................................CLII
B. Metadata (GIS Reference Information)...............................CLV
C. Site Dataset......................................................CLVI
D. Site Proximity Matrix for Villages by Period......................CLXV

1.1 South-central Puerto Rico..................................................5
2.1 Conceptual model of social complexity proposed by Service (1962)..........8
2.2 The theoretical relationship between differentiation and integration.....17
3.1 Middle America and the Antilles..........................................24
3.2 Physiographic zones of south central Puerto Rico.........................27
4.1 Geographic distribution of Ostionoid ceramic subseries...................44
5.1 DEM Data and the extent of the study region..............................64
5.2 Soils dataset............................................................65
5.3 Physiographic zone data..................................................68
6.1 Site Type Frequencies based on Periods...................................81
6.2 Period II Site Distributions............................................83
6.3 Period III Site Distributions...........................................84
6.4 Period IV Site Distributions.............................................85
6.5 Distribution of Period II sites and physiographic zones..................89
6.6 Distribution of Period III sites and physiographic zones.................90
6.7 Distribution of Period IV sites and physiographic zones..................91
6.8 Site Distributions and Physiographic Zones...............................92

6.9 Period II site type distributions and physiographic zones...............92
6.10 Period III site type distributions and physiographic zones...............93
6.11 Period IV site type distributions and physiographic zones................93
6.12 Period II site cost catchments.........................................99
6.13 Period III site cost catchments........................................100
6.14 PIV site cost catchments................................................101
6.15 Mean percent of area of physiographic zone types occupying 5 km cost
catchments for all periods...............................................102
6.16 5 km cost catchment mean size through time.............................103
6.17 Period II agricultural potential of 1 km village catchments............108
6.18 Period III agricultural potential of 1 km village catchments............109
6.19 Period IV agricultural potential of 1 km village catchments.............110
6.20 Mean agricultural production potential of 1 km cost catchments through
6.21 Period II Settlement Clusters...........................................115
6.22 Period III Settlement Clusters..........................................116
6.23 Period IV Settlement Clusters...........................................117
6.24 Period II Integrative Networks..........................................120
6.25 Period III Eastern Integrative Networks.................................121
6.26 Period III Central Integrative Networks.................................122
6.27 Period III Western Integrative Networks.................................123
6.28 Period IV Eastern Integrative Networks..................................124

6.29 Period IV Central Integrative Networks....................................125
6.30 Period IV Western Integrative Networks....................................126

3.1 Physiographic Zones........................................................32
4.1 Temporal and Cultural Model for Prehistoric Puerto Rico....................39
5.1 Attribute Information of Data Table Created from information in the SHPO data
5.2 Soil scaling system and relative values based on Soil Survey of the Ponce
5.3 Elevation ranges of physiographic zones.....................................67
5.4 Slope ranges and cost values................................................73
6.1 Listing of all sites used in the study by period............................81
6.2 Results of ANOVA for catchment sizes through time..........................104
6.3 Results of ANOVA for distances between sites through time

1. Introduction
The rise of social complexity and the contexts within which formally
stratified prehistoric societies developed are important topics of discussion in
current archaeological research. Investigations of this nature contribute to our
understanding of the processes responsible for the growth of complex social
systems, their influence on social organization and behavior, and the long-term
effects these aspects have on the cultural development of a given society. In this
respect, archaeologists are given an opportunity to explore the maturity of social
complexity through an examination of material remains, which reflect patterns of
past human behavior. As the processes responsible for social change affect
behavior through time, archaeologists are perhaps best suited for this endeavor as
the evidence for their interpretations are conducive to diachronic examination.
The present study is an exploration of social complexity on a regional level.
Specifically, the goals of the study are (1) to critically examine the concept of social
complexity and its use within archaeological research contexts and (2) to provide a
regional case study from south central Puerto Rico that explores the development of
regional socio-political organization through time. As vehicle for this effort,
Geographical Information Systems (GIS) are employed to provide an in depth
settlement pattern analysis for the region. Through an examination of

environmental and archaeological site data, with respect to time, the morphology of
the settlement system and its implications for regional social development,
organization and complexity become salient.
Although complexity is an important area of archaeological research, its
conceptualization and application is often inconsistent and neglected within
individual research frameworks. This is primarily due to the ambiguous nature of
the term complexity and the dynamic nature of society and culture to which the
concept is applied. Further, problems with the use of this concept stem from a
tendency within anthropological research contexts to utilize complexity to qualify
levels of social organization as opposed to exploring the ways in which a social
system is articulated and behavioral patterns that result from it. To address this
issue, it is important to consider what is meant by complexity and how it may be
qualified for the examination of prehistoric societies.
The organization of human societies at any given state in time may be seen
as the result of socio-political and environmental variables that have conditioned its
existence to that point (Boas 2000 [1920]: 134-141). Understanding how these
variables have influenced a society, are important dimensions for qualifying the
nature of complexity for a given social system. The examination of social
complexity presented here, focuses upon four important factors that provide a

framework for looking at social organization on a regional level: (1) scale, (2)
boundedness, (3) integration, and (4) differentiation.
1.1 Scope
The development of complex social systems and the centralization of
formalized power are currently of considerable interest to researchers focusing on
the prehistory of Puerto Rico (e.g. Curet 1996,1992b; Oliver 1992; Siegel 1996,
1989). Fueling this interest are ethnohistoric documents that clearly depict a
complex, socially stratified society that existed upon the arrival of Columbus (Las
Casas 1992; Pan£ 1999) and the rich archaeological record of the island that
elusively chronicles its maturation. Current conceptualizations derived from
archaeological evidence promote the notion that these societies developed in situ in
the Greater Antilles. Yet, many aspects of the structure and the processes
responsible for the development of these prehistoric socio-political systems are still
not thoroughly comprehended.
Exploration of these developments focuses on the south central portion of
Puerto Rico (Fig. 1.1). The region represents roughly 1,341 km2 of land. The center
of the study area contains some of the earliest public ceremonial architecture in the
Caribbean at the ceremonial center of Tibes (Gonzalez Col6n 1984; Rouse 1992).
Long-term occupation of the region coupled with changes in material culture

through time suggests the development of a formally stratified, complex regional
socio-political organization between 300 and 900 A.D.
To qualify paradigmatic conceptualizations necessary for the objective approach
of regional socio-political organization within the present research framework, the
proceeding chapter focuses on a theoretical discussion of social complexity.
Chapters 3 and 4 briefly explore the environmental and cultural contexts of the
island to establish the settings within which the study takes place and provide a
basis for the data used in the analysis. Chapter 5 deals with methodological
considerations of data acquisition and analysis within the study. Further, this
chapter offers a short discussion of settlement pattern studies and GIS applications
within archaeological research contexts to provide a basis for the analysis. Chapter
6 discusses the analysis and interpretations based on the results. Finally, Chapter 7
consists of concluding remarks regarding the study and provides suggestions for
future investigations.

fpi Contemporary Urban Areas
Ii Municipality "Boundaries
Figure 1.1: South-central Puerto Rico.
10 Kilometers

2. Theoretical DiscussionThe Complexity of
The term complexity, as utilized within research contexts is somewhat
ambiguous. This can be seen as a function of its definition and the lack of
qualification of the term within theoretical frameworks. As the term implies, the
concept insists on an examination of constituent parts, interrelated ideas and
activities that constitute its whole. However, in practice approaches to
understanding social complexity are often difficult to define and quantify.
According to Hannerez:
The term complex may in itself be about as intellectually attractive as
the word messy, but one of its virtues in this context is precisely its
sober insistence that we should think twice before accepting any simple
characterization of the cultures in question in terms of some single
essence (Hannerez 1992:6-7).
The concept of social complexity within archaeological theory stems from
contributions to the rapidly growing body of anthropological knowledge in the
1940s and 1950s. A major contribution from this time can be seen as a direct result
of neo-evolutionaiy thought advocated by anthropologists Leslie White (1949,
1959) and Julian Steward (1955). For White, social organization was a product of
how individuals harnessed energy and the technology associated with it (1943).
White utilized this approach to understand generalized developments in human
social organization. According to Qrlove, White was more concerned with the

broad details of evolution than with specific adaptations... he also directed little
attention to the influence of environment on particular cultures. Instead he
emphasized levels of energy use as the determinant of cultural evolution
In the 1950s, Julian Steward proposed that cultures had potential multiple
trajectories of development and that not all cultures would evolve in the same way.
Stewards study promoted the examination of the relationships between certain
features of the environment and the traits a culture possessed within bounded
ecological contexts. Within the environment, Steward emphasized the quality,
quantity, and distribution of resources. Stewards examination of culture focused
on technology, economic arrangements, social organization, and demography.
Further, Steward stressed the fact that the environment influenced only certain
elements of a culture.
These conceptualizations led anthropologists to consider stages of social
development and levels of social complexity among contemporary and prehistoric
societies. In consideration of these issues, Elman Service devised a four-fold
classification system of social development and complexity in his seminal 1962
publication, Primitive Social Organization.1 His typology, consisting of bands,
tribes, chiefdoms and states, represents a hierarchical organization to denote varying
levels of social complexity (Figure 2.1). A given societys place within this

framework is dependent on: (1) population, (2) social organization, (3) economic
organization, (4) settlement patterning, (5) religious organization, (6) architecture,
and (7) integration (Service 1971 [1962]).
Stages of Social Development
Bands Tribes Chiefdoms States

Less Complex Social Complexity More Complex

Figure 2.1: Conceptual model of social complexity proposed by Service (1971 [ 1962]).
Arguably all of these elements hold important clues to the nature of
complexity and the dynamics of social organization within a given society.
However, the application of these characteristics to denote specific levels of cultural
complexity has been criticized as being more arbitrarily than objectively construed
(Feinman and Neitzel 1984; Plog 1974; Yoffee 1993). Problematically, a
typological approach to social complexity does not adequately provide for variation
within these levels of social organization nor does it address transitions between
these categories. In this sense, the application of categorical terms provides a false

sense of knowledge in terras of social development and organization. According to
Blanton and colleagues:
As these dimensions vary independently to some extent, especially
when one considers separately institutions [or structures]organized
for different purposes, we have found it difficult to use the taxonomies
often employed in cultural evolutionary studies. These taxonomies
attempt to define the level of cultural complexity of societies in terms
of specific and static kinds of institutional arrangements.... (Blanton, et al.
Although problematic, traditional conceptualizations of social complexity
have stimulated an approach for (1) attempting to discuss the variation among
different societies and (2) the range of social complexity among all societies. Yet
perhaps the primary detriment to understanding the development of complex social
organizations resides in the term itself and the lack of infrastructure applied to it
within archaeological frameworks. Problematically, this deficiency inhibits the
ability to gauge observations that ultimately influence interpretations regarding
social behavior.
Arguably, all societies are complex in their own right. Yet in what ways can
we discuss complexity and the range of human behavioral patterning within a
relative context? Suggestively, comprehension of any behavioral patterns should
stem from an understanding of the dynamic nature of the relationships between the
elements of a society that are manifested as a result of varying forms of socio-
political organization. In this sense, the theoretical approach to understanding

social organization should not ask "how complex is a society?" but rather "how is it
complex?" (Nelson 1995:614). Understanding complexity in these terms assists in
the provision of an increased flexibility and objectivity in the examination of
prehistoric social organization.
To alleviate these problems in understanding social complexity, our
attention should perhaps focus on the trajectories of social development, the
articulation of social systems and the factors that influence them. Examination of
these factors assist in eliciting the causal relationships and developments in social
organization for a given society at a given point in time. Such an approach
welcomes variation in exploring social development and organization and is better
suited to understand the nature of complexity as a flexible, relativistic concept.
According to Yoffee:
Rather, one must imagine that many different evolutionary trajectories
can exist and that not all known human societies fall on the progressive
steps of a social evolutionary ladder. [Importantly] At what point and
why can one Trajectory be differentiated from another? (1993:72).
To provide a relative context within which to discuss social organization, it
is necessary to furnish a median or basis for variation in human behavioral
patterning. It is suggested here that this median is the distinctive organizational
dynamic and the resultant forms of behavior that are manifested in the differences
between egalitarian and formally stratified societies. Specifically, once
status/power is formalized an increased number of social mechanisms and structures

are necessary to catalyze and reinforce its acquisition and perpetuation. Although
some would argue that even egalitarian type societies do not display equal
distributions of power (e.g. Fried 1960) it is the multiplicity of internal social
mechanisms related to formalized power structures of non-egalitarian societies that
make them complex in a very different way than those that are not.2 These
fundamental differences in complexity may be qualified as the increased
multiplicity of structures and meanings that characterize and are necessary to
reinforce social organization. In this study, relative differences in social complexity
are gauged by societies that represent more egalitarian organization to those that
possess centralized power and its formalized institutionalization.
Archaeological research regarding the maturation of complex societies
typically emphasizes one of two paradigmatic models that that influence trajectories
of social change which are either based on ecological or socio-political catalysts
(Brumfiel and Earle 1987).3 The ecological or adaptationist perspective is often
characterized by a homeostatic approach to cultural systems and a deterministic
perception of the environment (Binford 1965; Service 1971 [1962]; Service and
Sahlins 1960). From this perspective, population pressure and resource imbalances,
constitute the primary force for the organization of centralized authority and the
development of social stratification (Caneiro 1967,1970). This perspective sees
organized decision-making (i.e. the centralization of authority) resultant from

problems oriented to subsistence acquisition and distribution. According to
Brumfiel and Earl:
In this model, political elites are assumed to intervene in the economy;
in fact, the ability of political leaders to organize a more effective
subsistence economy is considered raison d'etre of powerful leaders.
Powerful, centralized leadership is seen as developing in environmental
and demographic contexts where effective management is either necessary
or especially beneficial (1987:2).
Although presenting a cogent argument, such environmental determinism
fails to explain socio-political motivations for culture change and the perpetuation
of developed social differentiation in terms other than increased production. The
model is sufficient for explaining the development of social complexity in terms of
environmental and economic production, yet is limited in its ability to address the
influence of socio-political and ideological variables. To address change in terms
other than the environment, socio-political perspectives discuss developments in
organization as the result of the acquisition and manipulation of power.
From this perspective, the linear development in cultural systems is rejected
(Blanton et al. 1993,1996; Yoffee 1993) as is determinism dictated by the
environment (Arnold 1992). According to Johnson, Although subsistence stress
may well have been a critical factor in particular cases, it is unlikely to have been a
universal source of stress leading to increasing social complexity (Johnson
1982:390). Political models of social change substantiate the ability of political
agents, in complex societies, to gain and maintain control of the masses independent

from the procurement of subsistence resources and population pressures. This is
supported through the concept of self-aggrandizement or agency in the acquisition
of social and political power (e.g. see Arnold 1992; Clark and Blake 1994; Spencer
1993) .
Integrating ecological and socio-political perspectives has been specifically
addressed in recent anthropological literature (e.g. Kottak 1999; Greenberg and Park
1994) . Elements of the New Ecology take into consideration the socio-political
variables associated with resource control and characteristics of local, regional and
extra-regional environments. Although this perspective is primarily applied to
contemporary societies on a global level, the main concept it promotes possesses
considerable potential to assist archaeological theoretical frameworks. Specifically,
its contribution to understanding social complexity is its use of ecological and social
variables for understanding social organization and change through time. To
provide a more holistic perspective of social development and organizational
transitions relevant to social complexity, it is suggested here that these variables
(environmental and social) are mutually influential to some degree.4 Arnold states,
Environmental or social stresses, political opportunism, and elite
control... constitute key elements in the emergence of complexity (Arnold
1992:62). In this sense transitions in social organization are the result of responses
to stimuli within (i.e. social) or external to (i.e. ecologic) the social system.

It is suggested here that social organization is a dynamic and constantly
changing entity that operates within structural constraints of a given social system.
This boundedness may be conceived of as the morphological constructs of a social
system based on social and environmental contexts.5 Morphologically, subsequent
changes in the social system are (in ways) guided by the organization preceding it.6
Within these bounds social systems maintain the ability to change and perpetuate
the functions of society. According to Blanton and colleagues Such deviations
from the norm are especially likely when new or extreme conditions are
encountered, and changed behavior becomes necessary to maintain a traditional way
of life (or at least its most desired aspects) (Blanton et al. 1993:19).
2.1 Conceptualization for the Examination of Social Complexity
To provide a dialog conducive to understanding social behavior, in terms of
organization and change, it is necessary to examine the factors that contribute to the
complexity of a given society. Blanton and colleagues have provided a framework
whose primary focus has been to negotiate ordered and comprehensible units for the
analysis of the development of social organization based on political contexts in the
traditional Maya region (1993). The core constituents of this model consist of
scale, integration, complexity, and boundedness. These core features that they
attribute to understanding social change and organization are related as similarly
ranked conceptual issues for examination. Although their framework is good, the

relationships of these features to one another does not assist explaining the nature of
complexity as a unique concept for understanding developments in social
organization. This is because these factors are viewed, from Blanton and colleagues
perspective, as mutually exclusive elements relating to generalized
conceptualizations of social organization.
This study suggests that in terms of social organization as a whole-scale,
integration and boundedness are aspects relating directly to the theoretical
conceptualization of social complexity. Further, the concept of differentiation
(neglected in their model) is utilized here in direct relation with integration.
Application of these elements for exploring social complexity is not meant to
suggest that their usage is appropriate in all research contexts. Rather, the
utilization of these elements offers and example for qualifying complexity relative
to a given research context in order to provide continuity between theory, method
and the interpretations resulting from them for examining prehistoric social
2.1.1 Differentiation and Integration
Theoretical frameworks in anthropology generally measure complexity by
such things as the size of a society, the number and distinctiveness of its parts, the
variety of specialized social roles that the system incorporates, the number of
distinct social personalities present, and the variety of mechanisms for organizing

these into a coherent functioning whole (Tainter 1996). Hence, a society that has
more sub-groups and social roles, more networks among groups and individuals,
more horizontal and vertical controls, higher flow of information, greater
centralization of information, more specialization, and greater interdependence of
parts will inherently be organized much differently than one that has fewer
components of this nature (Blanton et al 1993; Brumfiel and Earl 1987; Earle 1997;
Hannerez 1992 and Tainter 1996). According to Hannerez:
Particular concepts can be seen as more complex when they have a
greater number of denotations and connotations; a cultural whole
which, can be derived from a smaller number of fundamental principles
is less complex than one which can only be described in terms of a larger
number of such principles (Hannerez 1992:8).
For this reason, greater complexity means increased differentiation of the
structural elements within a society (e.g. economic, religious, political).
Subsequently, this differentiation of elements will require more effective integration
of the systems constituent components. Therefore, two key components of social
complexity may be seen as the dimensions of the relationship between
differentiation and integration (Figure 2.3). Although it has been proposed that these
elements may be present without stratification, the degree to which they are
interrelated within a given society represents the key element to understanding
complexity in social systems. According to Blanton and colleagues:
In a society with little overall integration, the units constituting it are
highly self-sufficient. Higher levels of integration imply more connections

among units. Depending on the kind of component units, the connections
are established as flows of material, energy, information, or people. The
greater the flow through interconnecting channels, the greater the
interdependence (Blanton et al. 1993:16-17).
Figure 2.2: The theoretical relationship between differentiation and integration.
As the differentiation and integration of elements within a society
intensifies, the development of centralized leadership becomes necessary for the
system to function. The centralization of organized decision making leads to more
effective ways of potentially solidifying relationships and promoting cohesion of the
system (Johnson 1982). Hence, the primary binding element in the relationship
between integration and differentiation is the centralization of political authority.
Although it has been suggested that complexity exists without centrality (Blanton et
al. 1996, Mcguire an Saitta 1996), it is suggested here that societies with formally

centralized socio-political power structures are fundamentally different than those
without it According to Earle:
Complexity is a problematic concept, but I wish to emphasize one
critically important element of complexity, namely, centrality. Within
complex societies, centralized systems are conduits, organized around
central nodes, through which flow goods, materials, information, decisions,
and power. The structural relationships enable individuals or groups to
exert a degree of power from their centralized positions (Earle 1997:208).
The concepts of integration and differentiation present a cogent means for
understanding the articulation of social systems in terms of its constituent power,
social and economic structures. However, when examining the maturity of complex
social organizations it is important to consider external social and environmental
factors through time. Doing so, supplies a context within which to understand
differentiation and integration within the system, but well as the universe within
which a society exists at a given point in time.
2.1.2 Boundedness and Scale
To achieve a holistic perspective for examining complexity it is necessary to
establish the universe within which the society exists and the level on which it
operates. The concepts of boundedness and scale are factors that assist in this
understanding. Although these concepts are related in many ways, they represent
unique variables that constitute and delimit the realm of complex social
organization and the morphological constructs (both physical and social) within
which they operate (Blanton et al. 1993; Feinman 1998). Although Wolf (1997)

and Appadurai (1996) suggest that societies are not (and never were) bounded
discrete entities, it is useful to delimit the extent of socio-political interaction in
order to understand how complex societies developed and interacted.
Blanton and colleagues have suggested that boundedness relates to how
the population of a social system interacts with other populations outside its
boundaries, in exchanges of energy, materials, people, genes and information
(1993:18). In this sense, boundedness represents the extent to which a social system
interacts with populations outside its direct sphere of influence. To add to this
concept, boundedness also refers to the internal morphological structure of a social
system that restricts the behavior of individuals operating within it. In this way,
boundedness represents both external and internal constructs that influence social
organization reflected in geographically defined areas of social and ecological
interaction. Within these bounds social-systems operate on varying levels ranging
from individual interaction on a local level to group interaction on a regional level.
Boundedness also refers to the socio-political dynamics of a system inherent
through the processes of history. Sensibly, the developments of a society at any
given moment are the result of events that came before it. Assuming this is true,
social development is morphologically constrained by the dynamic of a societys
established organization and behavioral patterns. This is not to say that complex
social organizations develop within a closed system or that geographic and cultural

boundaries are clearly discemable. Instead, our notion of boundedness represents
elastic physical and social realms that may be real or imagined and change through
time. In this sense, boundedness reflects the elasticity of a social system and the
limits within which social and environmental variables influence social change.
Examination of geographical boundedness and resource availability is
crucial for understanding the influence of the environment on the socio-political
organization of a system (Flannery 1976). To clarify, the environment is not
necessarily a determinant of social development but it defines the range of choices,
which societies and the individuals within them make to construct their social
realities. Subsequently, the results of these actions are evident in the relationships
within die structures of the social system on varying levels. Boundedness in this
study is utilized to provide a regional basis in order to establish an environmental
and social context within which societies construct their socio-political realities.
For example, research by Kirch (1984) and Keegan (1995) have demonstrated that
island societies are bounded within environmentally constrained contexts that affect
social interaction and the trajectories of social development. According to Clark
and Blake:
Considerations of the environment should acknowledge actors with
conventional perceptions and constructions of their world in symbolic
interaction with other people and objects. In short, Nature (including
resources, physical features and concepts of space and distance) is subject
to interpretive shifts and even manipulation by interested individuals within
a given social system (1994:18)

Within the social and geographical boundaries of a given social system,
organization operates on varying scales of complexity. Generally, when scale is
referenced in context with social complexity one generally considers the size of the
population of the social system in question (e.g. Feinman 1998). Scale in this study
also refers to the size of the society and the extent of the inhabited geographical
domain (Blanton et al. 1993; Renfrew and Bahn 1996). In terms of the size of the
society, and the area in which it occupies, scale is important to consider because it
holds various implications for impacts on the environment, which influence the
nature of social interaction and trajectories of development. From this perspective,
the scale of a social system directly influences the differentiation and integration of
social structures with a given social system relative to its size. It is a common
observation in anthropology that larger societies tend to be more complex than
smaller ones, and both size and complexity are normally considered to be major
axes of variability in social evolution (Johnson 1982:389).
The scale of a society increases only to the point where the environment or
social mechanisms of control fail to provide for the system (Blanton et al. 1993;
Service 1971 [1962]). Therefore, scale is related to the idea of boundedness in that
the scale in which a society operates does so within discrete environmental and
social limitations.

2.2 Social Complexity and Prehistoric Puerto Rico
Exploration of the development of complex social organization of south-
central Puerto Rico offers a vast potential for further understanding the prehistory
the region and the social organization of the people who lived there. Current
evidence suggests broad changes in social organization based on material culture
between 600 and 1200 A.D. (Garrow et al. 1995; Lundberg 1985; Rodriguez 1985).
Shifts in social organization are suggested to be a result of the centralization of
socio-political authority based on control over symbols and potentially ceremonial
power (Curet 1996; Oliver 1992; Siegel 1989, 1996, 1999).
The exploration of social complexity for south central Puerto Rico,
presented in this study, addresses transitions in social organization on a regional
level. Eliciting the nature of how prehistoric social organization matured in
complexity will be examined in terms of, the morphological constructs (i.e. bounds)
within which it operated and the nature of integration and differentiation of
settlements in the region. Settlement pattern analysis of the region with respect to
environmental and social variables responsible for site distributions will assist in
our understanding of how these peoples were complex and the scale of socio-
political organization. Yet prior to conducting an analysis it is critical to provide
environmental and social contexts within which to conduct the study.

3. Environmental Context
The smallest and most eastern island of the Grater Antilles. Puerto Rico
serves as one of the barriers between the waters of the Caribbean Sea and the
Atlantic Ocean. The island is nearly rectangular in shape running 179 km (111
miles) from east to west and 58 km (36 miles) from north to south or about 9,104
km2 (3515 mi.2) (Figure 3.1). Puerto Rico is located in the tropical zone of the
world and is characterized by sub-tropical vegetation and climate (Ewell and
Whitmore 1973). The temperature of the island varies little between the warmest
and coldest months. Mean temperatures range between roughly 26 and 32 Celsius
during the warmest and coolest months respectively. In the drier south and
southwest temperatures may reach 32 Celsius for more than 200 days of the year.
Freezing temperatures are unknown to Puerto Rico although some of the
mountainous areas experience temperatures below 10 Celsius.
For the most part, the island is mountainous (roughly 60% of total land
cover) surrounded by a low-lying coastal plain. The mountains consist of highly
folded, faulted and eroded volcanic and sedimentary rocks. These formations form
the core of the Cordillera Central. The Cordillera Central is an east-west trending
mountain range that runs from Luquillo and Cayey on the east to Rincon in the
west. The mountains represent a formidable topographic boundary that divides the

island north and south. Jagged peaks and steep slopes characterize the mountain
region, the highest point on the island is approximately 1338 m above sea level
(Cerro la Puntita). The northern side of the mountain range receives more rain and
is steeper than the south. In the mountainous, northern interior exists the karst
region. This area is essentially a plateau with elevations ranging between 30 and
250 meters. The limestone hills that characterize this area are pocketed with caves
that have been eroded by water through time.
Figure 3.1: Middle America and the Antilles.

In terms of natural boundaries, the south central portion of the island is
defined by the peaks of the Cordillera Central, running east to west in the north, and
the waters of the Caribbean Sea to the south. The region falls within the
Subtropical Dry Forest life zone region of the island and is in general relatively arid
in comparison to the rest of the island (Ewell and Whitmore 1973:72).
Puerto Ricos central location in the northern Caribbean, together with its
variations in elevation, rainfall, and soil, have served to stimulate the development
of a variety of plant and animal life. Ewell and Whitmore (1973) have devised a
classificatory system for the life zones of Puerto Rico based on the Holridge
model (1947). These life zones represent distinct geographical units, which consist
of relatively homogenous ecological characteristics in terms of flora, climate and
geology. According to their scheme, there are six subtropical life zones represented
in Puerto Rico, The units are ones of broad bioclimatic similarity and each may
encompass several forest associations, land uses, and successional stages (Ewell
and Whitmore 1973:8). Although generalized, Ewell and Whitmores classification
system provides a baseline for ascertaining the nature of local environmental
3.1 Physiographic Zones of South Central
Puerto Rico
To consider social complexity in terms of settlement
generalized locations of subsistence resources, it is necessary
distribution and
to examine the

ecological contexts of the region. Previous studies of prehistoric peoples of south
central Puerto Rico have focused on broad environmental regularities to provide a
context of resource distribution and site location on a regional level (Lundberg
1985; Rodriguez 1985). Since a fine scale environmental model of prehistoric
resource distribution does not currently exist, broad scale emphasis on regional
resource patterning examines trends that represent large areas of relatively
homogenous ecologic zones. Examination of areas representing generalized
ecologic continuity offer a clearer picture of prehistoric resource distribution than
more specific characteristics that are likely to have changed through time.
In this study the model of general terrestrial resource distribution is based
primarily on soil distribution. Soil types for the region have been identified and the
spatial extent of their distribution has been delimited for the Ponce region
(Gierobolini 1979). Surface deposition for the region represents roughly 90 types
and subtypes (Appendix A). As soils are the product of climate, plants, animals,
parent material, relief, and time they represent an important source of information
for understating discrete ecologic habitats and topographic characteristics of a given
The study area has been divided into 4 distinct physiographic regions:
coastal plain, foothills, uplands and the upland/mountainous region (Figure 3.2,
Table 3.1). The physiographic divisions are based loosely on the U.S. Department

of Agriculture soil survey conducted for the Ponce region (Gierobolini 1979) and
have been generated by elevation ranges from the coast. Previous archaeological
research has examined the region in similar physiographic units as those presented
here with minor variation (Lundberg 1985; Rodriguez 1985).
Physiographic Zones
| | Coastal Plains
E2 Uplands
10 0 10 Kilometers
Figure 3.2: Physiographic zones of south-central Puerto Rico.
The first region considered is the coastal plain, which extends from the
southern coastline to the base of the semiarid foothills and in some places to upland
areas. This region is characterized by a relatively topographically level plain that
ranges between 0 and 5 degrees slope. The coastal plain extends from the coast (sea
level) to approximately 70 m above sea level. The coastal plains are primarily

composed of sand, loam and clayey soils that are slightly alkaline (Gierbolini 1979).
The vegetation of this region consists of grasses and shrubs suited for the dry
environment of the southern plains. This area also encompasses mangroves and
marsh areas,
The second physiographic region consists of the foothills. This zone
constitutes a transition between the coastal plains and the upland region of the study
area. The foothills are immediately surrounded by steep to moderately steep slopes
that lead to an upland type environment. Slopes of the foothill zone range between
10 and 15 degrees. The elevations, which delimit this region, begin roughly at
about 70 m and extend to approximately 225 m above sea level. This area is
characterized by gently sloped hills that neither attains the height nor the ruggedness
of the uplands yet possesses stepper slopes than the coastal plain. Topographically,
this zone denotes the beginning of discrete river valleys that penetrate to the upland
In this zone the difference in flow and volume of streams and rivers
throughout the year result in a large amount of coarse surface material deposited in
the floodplains. Between approximately 50 meters and 100 meters above sea level
these areas are characterized by river terraces in the low lying areas near the stream
or river resource. At approximately 100 meters above sea level in many places the

river terraces extend from the edge of the river wash area and demark the beginning
ascent that leads to uplands and higher moister areas of the region.
In general, the soils for this region consist of stony, clayey loam. Limestone
outcrops are common in this zone specifically, along the southern coast and in the
western portion of the study region. The limestone areas are characteristic of
steeply dissected plateau regions that denote shifts in geologic formations from the
coastal region to the foothills and upland areas. These areas are typically covered in
The upland zone for the study region consists of the steep slopes south of the
Cordillera Central. The majority of this area consists of slopes that are between 20
and 30 degrees. Elevations for this region range from approximately 225 to 450 m
above sea level. The potential for landslides in this region is high based on the
steep inclines, rainfall and shallow soils. Portions of the upland region fall within
Ewell and Whitmores Subtropical Moist Forest classification (Ewell and Whitmore
1973). This area is densely vegetated and plant life is composed of fast growing
light demanding species of shrubs and trees. Timber production is suited to many
of these upland soils. Cultivation of domesticates is difficult due to slope and
erosion. However, coffee has been planted in this region with some success
(Gierbolini 1979).

Based on steep slopes, undulating topography and variations in elevations,
the distinction between upland and mountainous regions is difficult to discern. The
upland/mountainous zone accounts for the highest elevations in the region. The
area possesses slopes over 30 degrees. Elevations of this area extend from 450 to
approximately 1,300 meters above sea level. This region is wetter than the foothill
and upland regions of the southern slope. The peaks of the Cordillera Central,
slopes are steep and are extremely susceptible to landslides. As rainfall is high and
temperatures consistently moderate, this area provides habitat to a variety of
3.2 Subsistence of the Prehistoric Inhabitants
of South Central Puerto Rico
In terms of prehistoric subsistence patterns, the physiographic regions within
the study area attempt to represent range of potential subsistence resources from
zone to zone. Assuming that resource locations were important influence on the
location and development of communities, understanding the distribution of these
resources is important for understanding the subsistence economies and settlement
morphology of the prehistoric people of the region.
The prehistoric inhabitants of Puerto Rico incorporated a wide variety of
plant and animal resources into their subsistence economy (Peterson 1997; Reitz et
al. 1996). The south central portion of the island displays abundant evidence of
coastal resources apparent in the distribution of culturally deposited shell middens

at various sites through out the region. Beyond coastal resources, it is difficult to
speculate the extent of specific plant or animal communities based on the poor
preservation in the archaeological record. However, based on historic documents at
the time of contact and zooarchaeological evidence, a wide variety of terrestrial
resources were utilized in the subsistence economy including birds, lizards, rodents,
turtles, and snakes (Peterson 1997; Robinson et al. 1985:60).
Ethnohistoric evidence suggests a dependence on the use of the root crop
manioc as a major subsistence resource. Material remains support this supposition
through the proliferation of food processing and cooking artifacts at many sites
throughout the study region. Therefore, fertile soils must have been of some
importance to the prehistoric inhabitants for this portion of the island. However, the
utilization and intensity to which these resources were exploited through time has
been a matter of considerable debate (deFrance 1989; Keegan 1989; Siegel 1993).

Physiographic Region AREA (km2) % of the Total Study Region Major Soil Association Major Vegetation Types Lundberg (1985) Rodriguez (1985)
Coastal Plains 374 28% Constancia- Jacaquas-San Anton association Bucida buceras, Hippomane mancinella, Terminalia cattapa, Cocos nucifera, Morinda citrifolia, Dodonea viscose, Castela erecta, Chrysobalams icacao, Coccoloba uvifera, Suriana maritime (Ewell and Whitmore 1973; Gierbolini 1979). Coastal Plains Littoral, Coastal Plains
Foothills 421.8 31% Aguilita- Tuque & Fratemidad- Paso Seco associations Busera simaruba, Caphalocereus royenu, Bucida buceras, Tamarindus indica, Prosopis juliflora Limestone Upland, Intermedia te Foothills
Table 3.1: Physiographic zones.

Table 3.1 (Cont.)
Physiographic Region AREA (km2) % Of the Total Study Region Major Soil Association Major Vegetation Types Lundberg (1985) Rodriguez (1985)
Uplands 259.3 20% Aguilita- Tuque & Caguabo- Mucara- Quebrada associations Prestoa Montana, Dendropanax arborea, Manilkara bidentata and Guarea giddonia Foothills Foothills/ Cordillera
Mountains/ Uplands 277.9 21% Caguabo- Mucara- Quebrada & Humatas- Maricao-Los Guineos associations Prestoa Montana, Dendropanax arborea, Manilkara bidentata and Guarea giddonia (Ewell and Whitmore 1973; Gierbolini 1979). Foothills (Portions) Cordillera
Table 3.1: Physiographic zones.

4. Cultural Contexts:
A Discussion of the Prehistory of Puerto Rico
In consideration of the development of complex societies in Puerto Rico, it
is interesting to note the developmental trajectories suggested by current
archaeological evidence characteristics of the social system of the indigenous
peoples at the time of Columbus arrival in 1492. At the time of Spanish contact,
the Taino peoples of Puerto Rico possessed an extensive array of expressive forms
in stone, ceramic, wood, dance, music and language (Deegan 1988; Las Casas 1992;
Oliver 1992; Pane 1999; Rouse 1992). These inventive and dynamic forms of
cultural expression may be seen as a reflection of the complex socio-political
organization of the Taino.
At the time of European contact, the Spaniards document the Taino socio-
political network for the island as being composed of a series of cacicazgos
(chiefdoms) under the leadership of caciques (chiefs). The role of the cacique, in
terms of political, economic and religious power has been discussed rather
extensively in the archaeological literature (Moscoso 1978, 1981; Oliver 1992;
Redmond and Spencer 1992; Wilson 1992). It is currently accepted that the
caciques assumed a multiplicity of political, symbolic, and ceremonial roles serving
as intermediaries between the human world and the supernatural realm (Curet 1996;
Oliver 1992; Siegel 1996). Based on documentation provided by the Spanish,

caciques were instrumental in organizing economic production and establishing the
division of labor. These organized activities reflect economic production that is
typically viewed in terms of the collection and storage of surplus under the chiefs
direction (Curet 1996; Moscoso 1981; Rouse 1992; Wilson 1992).
In teims of settlement organization, sites appear to have been hierarchical in
nature composed of two or three tiers and oriented in proximity to regional centers
characterized by monumental architecture (Oliver 1992; Redmond and Spencer
1992; Siegel 1989,1996,1999). Settlements appear to have ranged from single
structure hamlets to large villages whose domestic structures were accompanied
by clearly delineated ball courts and or central plazas (Garrow et al. 1995; Oliver
1992; Rouse 1992). Large communities are noted to have had populations in the
thousands (Redmond and Spencer 1992).
In terms of social organization, it has been suggested that the Taino society
was highly stratified and differentiated (Curet 1992a; Oliver 1992; Rouse 1992;
Siegel 1999:210; Wilson 1992). However, the manifestations of this differentiation
in the form of regional polities and their development in the periods prior to Spanish
conquest are still unclear. Evidence recorded by Spanish chroniclers point to a
multi-strata social system characterized by the unequal distribution of wealth and
power (Las Casas 1992; Pand 1999). Yet, even though the processes responsible for
the development of complex social organization in prehistoric Puerto Rico are not

completely understood there is considerable archaeological evidence that alludes to
the nature of the socio-political organization of these past societies. Therefore, a
discussion of the archaeological record, in terms of material culture and its relation
to behavioral patterns, provides the historical constructs within which social
organization and change were manifested.
The currently accepted model, for the spatial and temporal distribution of
prehistoric cultural traits in the Caribbean, was developed by Irving Rouse (Cruxent
and Rouse 1958; Rouse 1939,1952,1981, 1992). Due to the tropical climate that
characterizes the island, very little organic material preserves through time unless it
is carbonized. The remaining material culture of prehistoric Puerto Rico is
composed of durable materials such as lithics, shell, bone, and most importantly
ceramics. The durability and profuse distribution of ceramic materials on the island
have made it an ideal unit of study. Hence, the primary unit constituting Rouses
model is ceramic style. Rouse defines ceramic style as the Sum total of a peoples
wares and modesa site unit (Rouse 1992:175). For Rouse, style is the overall
complex of modes that spatially and temporally delineate assemblages. In this
sense, Rouse utilizes similarities in style within discrete spatial and temporal
contexts to discuss cultural units. A complex of ceramic modes recurring from site
to site (i.e. the style) is also conceptualized by Rouse as representing an
archaeological culture.... (Garrow et al. 1995:12).

The spatial and temporal model derived by Rouse is a product of the modal
analysis of archaeological assemblages to determine ceramic style associated with
specific cultural units (Rouse 1965). Assemblages are typically named after the site
for a given style (e.g. Cuevas). Hence, Cuevas can refer (within Rouses
framework) not only to a cultural unit and its peoples, but also to a specific temporal
context within which the style exists. In this sense, style represents a fine-grained
level for discussing the spatial and temporal characteristics of cultural units.
Rouse recognized that ceramic styles, and hence cultural units, displayed
common characteristics through a shared number of ceramic modes. These
similarities in shared modes suggest a degree of relatedness that required a clearly
defined unit of classification. This unit has been termed series. According to
Rouse, series can be defined as Cultural complexes or ceramic styles, together with
the peoples and cultures they define, that are known to have descended from a
common ancestor.... (Rouse 1992:184). The notion of series is then seen as the
similarities among a set of styles and the changes of these styles through time
(Rouse 1986).
In bridging the fine scale of temporal and social units represented by style,
and the coarse grained socio-temporal taxonomic unit of series, Rouse
conceptualized a division of series into subseries. According to Rouse, subseries is
defined as the division of a series consisting of styles or cultures that share a

common ancestor (Rouse 1992:184). In doing this, Rouse essentially
acknowledged certain ceramic styles (representing cultural units) displayed more
commonalities among themselves than other groups within the same series. In
essence, the subseries classification addresses these commonalities within the series
as sharing derived traits from the same ancestral unit.
Finally, Rouses framework can be broken down into temporal units known
as periods. Periods are essentially temporal units that relate to major changes in
ceramic characteristics that may be extrapolated to cultural traits and social
development (Rouse 1992:106). There are four periods in Puerto Rico (I-IV) that
are subsequently divided into two temporal components representing earlier and
later portions of the period respectively. Importantly, the period classification
approximates temporal groupings relating to subseries and style. Therefore, the use
of period to orient general discussions of social context is a useful socio-
chronological unit for examining broad trends and social changes through time.
Rouses framework has undergone considerable modification in the years
since its formal inception to accommodate current archaeological findings and their
associated implications to the temporal and spatial distribution of cultural units. It
presently provides researchers of the Caribbean (and Puerto Rico in particular) a
means of exploring the temporal and spatial distributions of cultural groups and
social developments through time. Yet, some archaeologists conducting research

within the region have problems adapting this framework based on theoretical
assumptions of settlement and the implications associated with them. Despite any
potential theoretical problems that may exist with this model, it is accepted and
utilized in the research presented here and provides the basis for the exploration of
the spatial and temporal distribution of ceramic age sites within the study region.
Table 4.1 presents this socio-temporal model for Puerto Rico that is utilized in the
present study.7
Series Subseries Complex/Style Period Date
West East
Saladoid Cedrosan Hacienda Grande Hacienda Grande* Ila ca. 300 B.C.-400 A.D.
La Hueca** ca. 300 B.C.-300 A.D.
Cuevas lib ca. 400-600 A.D.
Ostionoid Ostionan (West) Elenan (East) Early (Pure) Monserrate Ilia ca. 600-900 A.D.
Late (Modified) Santa Elena Illb ca. 900- 1200 A.D.
Chican Capa/Boca Chica Esperanza IVa ca. 1200- 1500 A.D.
In the later portion of Period Ila.
Huecan Saladoid Subseries.
Table 4.1: Temporal and cultural model for prehistoric Puerto Rico (adopted from Curet 1996;
Garrow et al. 1995; Rouse 1992).
To facilitate the discussion of what is known regarding the development of
social complexity of prehistoric south-central Puerto Rico, it is important to focus
on specific temporal contexts and the associated characteristics of the cultural units
affiliated with them. The following discussion briefly examines several important

characteristics of these prehistoric cultural units through the use of periods for the
temporal classification. Specifically, background for ceramic age cultural groups is
examined with an emphasis on the settlement patterning, ceramic style, village and
household organization, mortuary practices and religion between Period Ha and
Period IVa (300 B.C.-1492 A.D.).
4.1 Period II -The Saladoid Series
Period II (ca. 300 B.C.-A.D. 600) is characterized by the material culture of
the first agricultural and ceramic groups to migrate into the Greater Antilles. The
Cedrosan Saladoid peoples are thought to have originated at the mouth of the
Orinoco River Valley in northeast Venezuela and migrated through the Lesser
Antilles, into the Greater Antilles, and to Puerto Rico (Rouse 1952,1965b, 1981,
1983,1992). Evidence for the Saladoid series in Puerto Rico was initially found by
a collector from Ponce and documented by Rainey in the 1930s based on
investigations at the site of Canas in southern Puerto Rico (1940).
The Saladoid ceramic styles display an extensive use of stylistic decorations
incorporating various zoomorphic and anthropomorphic figures (Curet 1996,1997;
Rouse 1992). The ceramics of this series are of high quality and represent the first
ceramic tradition present in the Caribbean. In general, Saladoid ceramics are hard,
relatively thin, well-fired, and of fine paste (Curet 1996:117). Differences in the
Early Saladoid and Cuevas ceramic styles (i.e. temporally reflected in Periods Ila

and lib) show a decrease in the utilization of decorative attributes (Curet 1996;
Garrow et al. 1995). This decrease in stylistic quality is continued in Period HI.
Importantly, this stylistic change has been hypothesized to be the result of changes
in ideological structures based on access to and use of ritualistic and or socio-
politically charged symbols (Curet 1992). From this perspective, Period II not only
represents the beginning of ceramic making communities in Puerto Rico but also a
time of "settling in" that forms a foundation for the social reorganization that which
is evident for Period III.
Current settlement pattern data for south-central Puerto Rico during Period
II shows evidence that sites are primarily confined to the coastal plains with limited
penetration into alluvial river valleys (Lundberg 1985; Rodriguez 1985). However,
some Hacienda Grande style components have been found at the sites of Tibes and
Hemendez Colon (Curet 2000 personal communication). Settlement studies
conducted by Lundberg (1985) as a result of investigations at the site of El Bronce.
suggest that only in the later portion of this period do sites begin to appear further
inland. No Hacienda Grande sites have been documented more than 8 km inland
during this time. In general, south-central sites during Period II consisted of
somewhat large independent villages located in close proximity to estuarine
environments and the southern coastline (Lundberg 1985). However, Cuevas sites

display a slight shift further inland following the course of river drainages upland
(Lundberg 1985: Rodriguez 1985).
On the level of the community, Saladoid village organization suggests
grouping of communal houses, in a horseshoe or semicircular pattern. This
horseshoe configuration is organized around a central open area devoid of cultural
materia] that often functions also as burial grounds (Curet and Oliver 1998;
Gonzalez-Colon 1984; Oliver 1992; Siegel 1996, 1999). It has been suggested that
this area represents communal space for both village level religious ceremonies and
daily activities alike (Curet and Oliver 1998; Oliver 1992:7; Siegel 1989,1996,
1999). This pattern seems to imply more egalitarian style organization if we
consider the central location of burials and the lack in number and quality of status
items associated with them during Period II.
The presence (as well as size and quantity) of specialized ceremonial space
appears to develop form these central areas in Period III and has been used in
several studies to indicate emergent social complexity and the formalization of
political centralization among the prehistoric cultures of Puerto Rico (Alegria 1983;
Lundberg 1985; Oliver 1992; Siegel 1989). However, the role of ceremonial space
in relation to socio-political units is not clearly evident at this time (Lundberg 1985;
Rodriguez 1992). To the best of current knowledge, the peoples of the Saladoid

series did not construct monumental architecture in the form of segregated
ceremonial space.
Arguably, this lack of differentiated space, in conjunction with the
communal dynamic of village organization and centralization of burials may
represent a non formally centralized socio-political infrastructure associated with
behaviors linked to specialized ceremonies and the spaces within which they
transpire (Oliver 1992; Siegel 1989). Should these elements at the level of the
community be extrapolated to the region, the Saladoid series in Period II points to a
decentralized socio-political organization based on communal housing, centralized
burial patterns, relatively equal access to symbolism (as represented on ceramic
wares), personalized religious artifacts and the absence of monumental ceremonial
architecture. However, as the Saladoid peoples settled in they began to formulate
new forms social organization whose inception can has been speculated to develop
during Period lib.
4.2 Period III-The Elenan and Ostionan Ostionoid Series
The distribution of Elenan and Ostionan (c.a. 600-1200 A.D.) ceramics is
used by Rouse to denote boundaries of ceramic styles of the Ostionoid series whose
patterns of ceramic distribution display less variation between islands than within
them. For Ostionoid ceramic styles, Rouse defines the Vieques Sound Area, which
included the Virgin Islands and eastern portion of Puerto Rico, and the Mona

Passage, comprising western Puerto Rico and eastern Hispaniola (Fig. 4.1). Within
Period III there are significant changes in ceramic styles that appears to digress
from those observed for Period n. Changes in the functional and aesthetic quality
of ceramics are accompanied by new developments in architecture, mortuary
patterns, archaeological site distributions and artifact assemblages. These variations
in material culture and their distribution across the landscape signify developments
in behavioral patterning resulting from the restructuring of social organization from
the previous period.
Figure 4.1: Geographical distribution of Ostionoid ceramic subseries. (Map modified from Curet
The post-Saladoid period ceramic distributions suggest an intensified
process of regionalization and possibly regional socio-political units. Decorative
stylistic variation and the representation of symbols from earlier Saladoid

assemblages become increasingly absent through late PII and PHI. According to
Curet Vessel forms were simplified, and red paint was used less. Most of the
diagnostic decorations were restricted to crude, vertical rectilinear incisions close to
the rims of bowls.... Technologically, the ceramics became softer, thicker, coarser,
and rougher (1996: 119).
Households during Period IU continue to display a similar spatial
configuration to those of Period II. However, the size of the structures appears to
decrease in later times denoting a possible increase in emphasis on the nuclear
family unit (Curet 1992a). As the decrease in household size is continued into
Period Illb, the increased importance on the nuclear household domestic unit is
maintained (Curet 1992a; Curet and Oliver 1998:231). This shift suggests a
transition in labor organization and subsequently, socio-political centralization
Mortuary practices during the later portion of Period III represent a shift
from those exhibited in Period II. Centralized burials within plaza contexts cease by
the end of Period IH and an emphasis on interments within domestic units becomes
apparent. Importantly this shift in mortuary patterns coupled with a decrease in
size and focus on the domestic unit suggests a decline in an emphasis on group
solidarity and greater emphasis on the household unit (Curet and Oliver 1998).
Critically, these changes point to what appear to be an increase in status

differentiation based on the organization of households in a hierarchical fashion in
relation to the central plaza (Curet 1992, 1996; Curet and Oliver 1998).
In south-central Puerto Rico, Ostionoid period sites dramatically increase in
frequency from Period II (Lundberg 1985; Rodriguez 1985). In addition to
frequency, other changes in sites are recognized based on size and potential
function. One major change in village organization during Period III is the
delineation of communal space by stone lined plazas and ball courts (bateys). The
presence of this spatial delineation has been thought to be a representation of social
reorganization manifested in centralization of social, economic, religious and
political authority (Curet 1992b, 1996; Oliver 1992; Siegel 1989,1999).
Importantly, formation of these structures implies an increase in behaviors that
would necessitate mechanisms of integration. The sites of Caracoles, El Bronce,
and the Ceremonial Center of Tibes, located in the Ponce region, are excellent
examples of this development between Period lib and Period Illb.
The south-central portion of Puerto Rico shows evidence of extended
habitation and relatively intensive use between 600 and 1200 A.D. (Lundberg
1985). During this period of time there was a proliferation of monumental
architecture specifically within the Ponce region. As monumental architecture can
be associated with the formalization of ritual and the centralization of power
(Alegria 1983; Oliver 1992; Siegel 1989, 1999; Rouse 1992; Wilson 1992) it is

suggested that the manifestations of complex regional socio-political organization
are pronounced where there is evidence of long term occupational sequences. To
analogize to the complex regional organization of Mesoamerica Flannery states,
Such complex regional patterns began growing from the moment the first village
was founded in their area. They evolved through time, as the original village gave
rise to daughter communities which were smaller, or larger, or more specialized, or
less specialized, and which maintained their ties to the parent community
Extrapolation of these patterns to a regional level suggests a dramatic shift in
social organization implied from archaeological evidence of Period II.
Suggestively, the transition denotes an increase in forms of social structures
necessary to maintain organizational stability as a result of centralization.
Archaeological sites in the south-central part of the island potentially best
demonstrate this transitional period based on the presence of long-term habitation
sites and changes in material culture present at them through time. Importantly,
developments in social organization and political structure during this period
represent the basis for increased complexity that becomes engrained during the
subsequent period and evident upon the arrival of Columbus.

43 Period IV -The Cbiean Ostionoid
The last three hundred years of prehistoric occupation in Puerto Rico and the
Virgin Islands may be traced to the Taino peoples. Period IV ceramic distributions,
defined by the Chican Ostionoid styles, follow boundaries established by the Elenan
and Ostionan Ostionoid. In the Mona Passage, the Capa style predominates and in
the eastern area of the Vieques Sound the Esperanza style is predominant.
In the study area, available settlement pattern data for period IV, settlements
in the south-central portion of the island appear to dissipate and new sites develop in
areas of the mountainous interior of the island (Lundberg 1985; Oliver 1992). One
example of this type of major settlement system that appears to develop during this
period can be seen at the base of the northern slopes oriented around the Ceremonial
Center of Caguana in Utuado (Oliver 1992). Monumental architecture of this time
is at its highest frequency with major ceremonial centers having little or no evidence
of extended habitation (Alegria 1983; Oliver 1992). With respect to this, it appears
that villages in the mountainous regions were distributed in close proximity to these
centers and that their use was for religious and political use on the localized regional
level (Oliver 1992). Due to the nature of site clustering and the increase of
delineated ceremonial space at local village sites, a hierarchical political structure of
regional social polities seems highly probable (Oliver 1992; Siegel 1989,1999).

Although sites in the south central portion of the island during Chican
disappear, the distribution of sites in the region may not necessarily indicate large-
scale abandonment as previously suggested (Rouse 1952), but rather a massive
reorganization of the socio-political network for this region (Lundberg 1985). As
the development of dynamic socio-political systems are evident in other areas of the
island during this time (e.g. the Caguana region), changes observed for the south-
central region require further investigation. A thorough understanding of why
these shifts transpired is currently underrepresented in the current literature.
In general, artifactual and settlement pattern data suggest that Chican
Ostionoid possessed an even greater elaboration and centralization of socio-political
and religious elements from previous periods. This increase has been directly
attributed to the development of complex chiefdoms on the island. Of the many
explanations that exist for social change through time, external conquest, ecologic
crisis, invention or adoption of new technologies can sometimes explain the
maturation of social complexity. However, the causes of these changes on the south
coast between Period II to Period IV are more difficult to ascertain.

5. Research Design and Methodology
In 1985, Emily Lundberg conducted a settlement pattern investigation of
south-central Puerto Rico as a contribution to the research investigations at the El
Bronce site located in the contemporary municipality of Ponce (Robinson et al.
1985:L1-L33). In her analysis, Lundberg presents several questions for
approaching the regional distribution of sites for south-central Puerto Rico.
According to Lundberg:
The questions under consideration at this level include the following:
(1) Are there any noticeable environmental regularities in the locations
of settlements? (2) Were there different types of settlements within the
region? (3) Do differences take hierarchical form with respect to size or
importance? (4) What is the evidence for changes through time? (Lundberg
It is with respect to Lundbergs initial regional investigations of the south
central portion of the island that the present study builds. Specifically, this analysis
seeks to address the following elements: (1) what is the morphology of the
settlement system through time, (2) how is it influenced by the environment in
terms of resource distribution, (3) how are sites within the settlement system
interrelated based on location and potential importance, and (4) what do the
distributions based on these factors say about socio-political organization and the
development of social complexity for the region. It is proposed here that through a
diachronic examination of these elements an increased understanding of changes in

social complexity for this region may be acquired. Interpretations based on this
examination will address to varying degrees our understanding of social complexity
in terms of scale, boundedness, integration and differentiation.
To facilitate the analysis of settlement systems within this research context,
settlement pattern methods are used in conjunction with GIS technology. GIS was
selected in this capacity as it permits the exploration, management and analysis of
spatial data in a versatile and efficient fashion. According to Maschner, Among
the most productive avenues of inquiry in GIS-based research are the areas of
graphical representation, exploratory data analysis, and spatial statistics (1996:5).
To promote an understanding of these methods and establish a basis for the
examination of environmental and site data it is important to briefly discuss
settlement pattern analysis and subsequently the recent proliferation of GIS
technology within archaeological research contexts.
5.1 Settlement Pattern Methodology
The spatial analysis of archaeological sites as distributed across a regional
landscape, or settlement patterns, has been an important aspect of archaeological
research since the early 1950s (Willey 1953). Settlement pattern analysis is a
critical methodological facet of archaeological research as it examines the ways in
which humans utilized and distributed themselves across a given geographical
landscape. Settlement pattern analysis not only allows archaeologists to reflect on

the relationship of sites to the natural environment, but also the structure and form
of socio-political boundaries of a given settlement system. In this sense,
settlement pattern analysis supply a means of examining the manifestations of
regional social organization and change through time as exhibited by the
distribution of sites and their place within the natural and social spaces they occupy.
According to Wiley, Because settlement patterns, are to a large extent, directly
shaped by widely held cultural needs, they offer a strategic point for the functional
interpretation of archaeological cultures (Wiley 1953:270).
The basis for the analysis of settlement systems lies in discerning patterns in
the spatial distribution of archaeological sites. The primary assumption associated
with such analysis stems from the notion that human behavior is not random and
settlement systems will conform to a logical and meaningful utilization of the
landscape. According to Hodder and Orton, Just as behavior itself is not random,
human beings do not behave randomly in space; that is, they do not use the
landscape randomly (Hodder and Orton 1976). This non-random use of space is
manifested as a result of human behavior, influenced by social and environmental
factors through time, whose imprint is left on a landscape in the form of material
culture. For instance, environmental considerations of site locations and types may
be the result of food resources, topography, fresh water supply, ease of movement,
transportation, and the availability of other natural resources. Factors that are

culturally based may include defensibility of an area, proximity to other members of
the society, and the cultural perceptions of the landscape.
One type of ecological approach to settlement studies is based on locational
modeling that attempts to predict the extent and intensity of land use as well as the
range within which resource acquisition activities were located. Typically these
applications assume a least cost perspective in which humans are seen as situating
their activities in such ways as to conserve the amount of energy needed to access or
redistribute critical resources. For example, ecotone models predict that humans
will locate themselves on the borders between ecological zones, thereby minimizing
the effort needed to collect varied resources from each zone. Subsequently, these
predictions are compared with actual settlement data to evaluate the influences of
economic behavior on site distributions.
A second ecological application of settlement patterns is catchment analysis,
in which estimates of a sites range of available resources is calculated based on a
hypothesized economic range around a site. This technique is especially effective
when comparing available environmental resources to actual floral and faunal data
at a given site. Typically, site catchments are considered to be circular, but other
more irregular shapes have been proposed in an attempt to compensate for elevation
or topography. With this information, archaeologists can begin to study the

subsistence strategies of the inhabitants of sites within a region, compare them, and
attempt to understand the economic decisions behind site location.
The political organization of past societies also can be derived from
settlement data. Central place models examine the spatial organization of sites to
identify hierarchical networks and interaction between settlement systems. The
expectation is that major centers will be equally spaced from one another and
surrounded by a nested hierarchy of increasingly smaller sites. Rank size analysis,
which is derived from central-place theory, evaluates the intensity of centralization
in a settlement system by assuming that the degree to which a site is dominant is
reflected in its size, architecture and or some other important attribute relative to
sites of the same region and time.
Although these tools do not represent the full spectrum of methods used in
settlement pattern analysis, they are some of the more frequently employed. The
study presented here uses some of these tools aided through the use of GIS to assist
in the recognition of spatial patterns in site distributions across the landscape
through time. Application of these methods promotes the archaeological
investigation of regional social complexity through an understanding of the spatial
relationships between sites within discrete environmental and social settings.

5.1.1 Geographical Information Systems in
Archaeological Research Contexts
A GIS is essentially a spatially referenceable database. Data in a GIS is
typically represented by thematic features on a map, which have a direct link to a
particular record in an associated database. Hence, GIS data are organized into
thematic layers or coverages of these spatially represented features. The layers can
include such features as contours, hydrology, roads, soil types, vegetation types, and
archaeological sites. These thematic data layers may be used in combination to
provide a spatial representation of a given landscape or used in conjunction with
one another to perform various types of spatial analysis. One of the more important
features of a GIS is the ability to construct new data layers from those already
associated with maps. These might include slope or aspect from a contour data
A GIS is capable of holding information on the location and attributes of
each spatial feature represented in a specific map layer. Each of these features is
stored based on an identification attribute. Associated with a given record for any
specific feature (and characterized by a unique identification attribute) can be
several other attributes such as name, resource type, or material. Thematic data
layers organize data in vector format, as points, lines, and polygons, or they may be
stored in a grid composed of cells, or raster format. A grid data layer representing

vegetation zones would consist of a grid of cells with each cell being coded for a
given vegetation type.
The utilization of GIS within archaeological research contexts has proven to
be a powerful analytical tool for examining, analyzing and interpreting spatial data
(Maschner 1996). As much of what archaeologists find and record has an
associated spatial component (e.g. the location of a site within a region, the location
of features at a site, the location of a given artifact within a cultural deposit etc.),
GIS provides an opportunity to critically examine data within their spatial contexts.
Maps in this sense, represent not only a medium to convey spatial relationships
between sets of features but also a means by which to critically examine these
features in real world space. According to Hodder and Orton, Any map is in a
sense, an attempt at quantification. It provides the empirical evidence on which
some theory is built (Hodder and Orton 1976).
The potential of GIS to help understand how ancient people related to their
landscape and created complex social organizations within in them is tremendous.
Ancient people understood how the land could best meet their needs and located
their settlements accordingly. As these needs are a result of environmental and
social factors the distributions of sites should reflect some logical distribution that is
congruent with the extent of these influences. The spatial analysis for this study

was executed through the use of ESRIs (Environmental Systems Research
Institutes) ArcView GIS software.
5.2 Data
The data for this research consists of GIS thematic layers and their associated
databases. The data has been acquired from both state and federal government
agencies. There are three primary GIS layers that form the core base data for the
analysis: (1) digital elevation models (from the USGS), (2) soils data (from the
NRCS), and (3) archaeological site locations (Puerto Rico SHPO). Subsequent
datasets were derived from these to facilitate certain aspects of the analysis. For
example the slope and physiographic regions data layers were created from the
digital elevation models. Information regarding the creation, intended use, spatial
resolution and attribute definitions of the core GIS datasets may be found within the
appropriate metadata files associated with them (Appendix B).9
5.2.1 Site Data
The location of archaeological sites was acquired from the Puerto Rico State
Historic Preservation Office (SHPO) as a digital shapefile. The shapefile represents
a point thematic data layer displaying the location of all prehistoric archaeological
sites within the region. The archaeological site data set was digitized from maps
housed at the SHPO and the Institute of Puerto Rican Culture (ICP). Attributes
associated with the spatial information were taken directly from the site form

documentation housed at the SHPO and the ICP. Site form data was entered into an
Access database and subsequently linked to the spatial information. The attribute
fields of the Access database mimic the standard site forms.
To facilitate the use of the site data within the GIS, information from the
SHPO was used in conjunction with other publications (e.g. Alegria 1983; Garrow
et al. 1995; Lundberg 1985; Rodriguez 1985) to create a separate and
comprehensive data table reflecting important characteristics of the sites for the
region. This data table contains the primary site data for this analysis and will
hopefully be expanded by future research (Appendix C). This data table is linked to
the tabular data associated with the spatial features in the SHPO site shapefile.
Through the construction of this dataset, and linking it to the site data layer, it is
possible to efficiently run a variety of spatial analysis. The attributes of this table
and their definitions can be seen in Table 5.1
Attribute Definition
Site Number Puerto Rico SHPO site number designation. This number is similar to the Smithsonian trinomial configuration. (This is the linking variable to the spatial data set).
Site Name Name of the site.
Period Da, lib, Ilia, EQb and IV *(Each period and sub-period represents a separate attribute field in the data table). Presence/absence of these temporal components at the site.
Table 5.1: Attribute information of the data table created from information in the SHPO site files and
published resources.

Attribute Definition
Saladoid Component Presence of Cedrosan or Huecan ceramics at the site.
Ostionoid Component Presence of Elenan or Ostionan ceramics at the site.
Chican component Presence of Chican ceramics at the site .
Monumental Architecture/No evidence of long-term domestic habitation. Predominant features at the site consist of monumental ceremonial architecture in the form of stone rows or clearly defined ball courts and or plazas. No evidence of long- term domestic habitation at the site reported.
Village/Monumental Architecture Present Evidence of long-term domestic habitation present or highly probable. At least one ceremonial architectural feature is present.
Village/Monumental Architecture Absent Evidence of long-term domestic habitation present or highly probable. No evidence of ceremonial architecture present.
Hamlet/Limited Activity Area Small domestic site or limited activity area. Site shows no evidence of long term and or substantial domestic habitation. No ceremonial features present.
Area Area of the site in m2 (when available).
Number of Ball Courts/Plazas Number of ceremonial structures at the site in the form of stone lined plazas or bateys.
Area of Ball Courts/Plazas Area of ceremonial features at the site in m2 (when available).
Site Type Concatenation of the site type fields. Used for querying in the GIS.
Table 5.1: Attribute information of the data table created from information in the SHPO site files and
published resources.
The data presented in this table is a full representation of all documented
archaeological sites in the region. It must be stated that many sites may exist for
which there is no formal documentation. Further, the data presented is problematic

in terms of the subjectivity inherent in all archaeological data recording and the
nature of site discovery based on reconnaissance survey. Additionally, the spatial
extent of many sites cannot be confidently defined. This is partially due to the
condition of extant sites as well as the intensity to which certain areas of the study
region have been investigated. Further, the data in the table constructed here may
not be fully representative of all listed characteristics of sites as data for the table
was taken from the site forms and a limited number of research publications.
Therefore, many of the classifications for the sites should be viewed as tentative
until further investigations are conducted or existing documentation is elaborated.
The dataset for the region consists of a total of 96 documented ceramic age
sites with relatively good spatial and temporal information. Sites used within the
analysis are concentrated on occupational sequences temporally attributed to the
ceramic age (i.e. Period Da-Period IVa). Chronological assessment of sites was
derived from site forms, Lundberg (1985) and Rodriguez (1985). Since ceramics
form the basis of socio-temporal investigations, the lack of stylistic differentiation
for all sites poses obvious problems for precise socio-cultural and temporal
assessment. While it is fortunate that some sites possess distinction in ceramic
styles and sub-period stylistic/temporal data-the majority do not To alleviate this
dilemma the analysis chronologically focuses on the major period divisions as
defined by Rouse (1992).

To facilitate an in depth analysis pertinent to understanding the settlement
system of the region, sites are segregated into four types that suggest differences
both in terms of function, intensity and extensiveness of use. Although problematic,
differences in size, architectural features, and the extent of material remains at sites
throughout the region were compelling. Optimistically, the variety of sites
presented here facilitates an understanding of the settlement system rather than
hindering it. The types consist of (1) ceremonial centers with no domestic
habitation features, (2) village sites with ceremonial features, (3) village sites
without ceremonial features and (4) hamlet/limited activity areas.10
To clarify, sites that possess evidence of substantial, long-term domestic
habitation were classified as villages. Evidence which supports long-term domestic
habitation is substantiated by significant deposits of food remains and ceramic
material. Sites lacking substantial deposits or suggesting the presence of two or
fewer domestic structures have been typed as hamlets/limited activity areas.
Hamlets/limited activity areas are applied here to denote limited site use relevant to
function based on the extent and types of artifact deposits and not necessarily time.
Ceremonial architecture in the form of ball courts and plazas have been
identified for many sites in the study area. Sites that possess both evidence of
monumental architecture and long-term domestic habitation (with the potential of
more than one household) are classified as villages with ceremonial features.

However, some sites within the region possess ceremonial features but lack
evidence of long-term domestic habitation. These sites are classified as ceremonial
sites without habitational features (e.g. Garrow et al. 1995). The presence and
number (and when available area) of monumental ceremonial features at sites were
recorded to promote potential ranking of sites for the region as previously
conducted for other parts of the island (Oliver 1992; Siegel 1989).
The purpose in the creation of site typologies within this study is to
recognize variation in terms of potential function and use through time. It is
assumed that by examining the distribution of types through time a firmer of
regional complexity and social organization will be acquired. Ambitiously, this
data offers more intensive consideration of aspects relevant to the prehistory of the
region as it allows us to examine potential differences in site function and the
implied behaviors associated with them. Typically, such analysis has only focused
on major ceremonial centers for very broad geographic areas that ignore smaller
sites (e.g. Siegel 1996,1999).
5.2.2 Topographic Data
The U.S. Geological Survey Digital Elevation Models (DEM) were acquired
from the USGS via download over the Internet. The DEMs provide a digital three
dimensional (i.e. x, y and z) representation of the standard 7.5-minute USGS
quadrangle. This data set is grided data in raster format at a scale of 1:20,000 with

each cell within the dataset representing the elevation of a 30 x 30 meter cell. As
the data set represents the spatial extent of topographical features over a continuous
surface it is possible to derive other information for the area (e.g. slope).
For this study, a total of 15 DEMs were merged together to form a
contiguous topographical surface for the region. These DEMs coincide with the
following U.S. Geological Survey 7.5 minute quadrangles (or portions thereof) for
the region: (1) Ponce, (2) Pefiuelas (3) Santa Isabel, (4) Punta Chuchara, (5) Playa
de Ponce, (6) Adjuntas, (7) Jayuya, (8) Salinas, (9) Yauco, (10) Guayanilla, (11)
Villalba, (12) Coamo, (13) Orocovis, (14) Barranquitas and, (15) Ciales (Figure
5.1). As the Cordillera Central is a formidable natural topographic boundary
between the northern and southern portions of the island, this data set was clipped
to produce a geographical representation of all areas from the southern slope to the

i lan-n
m19 *
H i IsHlra-m
r^l no H4
m 177.2(0
GS 207.240
342 407
m| 408 489
651 -758
H| 150-1312
Figure 5.1: DEM data and extent of the study region.
5.23 Soils
The next layer to be considered is a vector dataset that represents the
distribution of soil types throughout the region (Figure 5.2). The dataset was
produced by the Natural Resource Conservation Service (NRCS, formerly Soil
Conservation Service) and compiled by quadrangle by the Puerto Rico Department
of Natural and Environmental Resources (PDNER). Soil data created by the NRCS
for this portion of the island is derived from the Soil Survey of the Ponce Region
(Gierboloni 1979). The data is composed of over 5,000 polygons of various sizes
denoting the location and extent of soil types within the region. The tabular data

associated spatial features is composed of the area, perimeter and soil type for each
Figure 5.2: Soils dataset.
The tabular data associated with the soil spatial data set was modified so that
the agricultural potential of a given soil type is represented (Appendix A). This
modification to the dataset was executed by adding a column called Agricultural
Potential to the tabular data and populating the attribute field with a numeric value

acquired by cross referencing the productivity potential of a given soil type
referenced in the Soil Survey of the Ponce Area (Gierbolini 1979).
The classification system associated with the soil survey data is based on the
physical and chemical properties of a given soil and the nature of the slope on
which the type is typically located. The agricultural potential classification system
in the soil survey manual is represented by a scale of 1-8 (1 being best, 8 being
worst) that denotes the incremental increase in the limitations of soil usage for any
given type. The system for this study utilizes a scale of 1-5, with 1 being worst and
5 being best, and is based on the soil survey manual criteria. In this sense the
classification system presented here, in comparison to the survey manual system,
was inverted, truncated and simplified (Table 5.3). Missing data were attributed 1 as
a given soil type lacking data will at least have this value.
This Study Gierboloni (1979: 47-49)
5 (Best) 1 (Few Limitations)
4 (Good) 2 (Moderate Limitations)
3 (Average) 3 (Moderate Limitations with special conservation practices)
2 (Fair) 4 (Severe Limitations that require intensive special conservation practices)
1 (Worst) 5-8 (Severe Limitations)
Table 5.2: Soil scaling system and relative values >ased on Soil Survey of the Ponce Area (1979:47-

5.2,4 Physiographic Zones
The physiographic regions data (discussed in Chapter 3) was created through
selecting elevation values from the DEM. Elevation ranges were selected from the
DEM to represent general areas of ecologic continuity as elevation, slope, soil type
and vegetation are all related to varying degrees (Gierbolini 1979). These elevation
ranges represent close approximations of geologic and ecologic zones within the
study area (Table 5.4). Each set of ranges was saved as polygons representing each
physiographic zone (Figure 5.4). Each polygon theme for each zone was
manipulated (using the ArcView dissolve feature) to create contiguous polygonal
boundaries for each zone. These maps were then merged to create a polygon
shapefile and a grid data set. The scale of the data is 1:20,000.
Elevation Zone
0-70 Coastal Plains
70-225 Foothills
225-450 Uplands
450-1300 Upland/Mountains
Table 5.3: Elevation ranges of physiographic zones.

Figure 5.3: Physiographic zone data.
5.3 Methodology of Data Analysis
Methodologically, this analysis approaches the exploration of regional social
complexity with attention to both environmental and sociopolitical elements. This
approach is not meant to suggest that these aspects are mutually exclusive in their
roles in the development of complex socio-political organizations. To the contrary,
it is advocated here that in conjunction both supply a more holistic context for
understanding social organization and complexity and the character of the regional
settlement system. As in Lundbergs analysis this approach is done not to indicate
a theoretical incompatibility but rather to find out how far each may be pursued on

its own merits and to make apparent those specific areas in which current data
deficiencies prevent progress (Lundberg 1985:L4).
The utilization of GIS in this study facilitates the analysis of archaeological
site distributions with respect to environmental and socio-political contexts. In this
capacity, the GIS serves as an analytical tool for exploratory data analysis.
Exploratory data analysis (EDA) has been defined by Tukey (1977, 1980) as the
problem driven exploration of data. Although EDA may lead to hypothesis, its
primary purpose is for exploring trends in the data at hand. The methodological
approach presented here follows this perspective of EDA. Through this inductive
approach it is not only possible to explore the information at hand in a new and
creative way but it also promotes an objective examination of the data that is
currently available.
As the analysis presented here respects the spatial elements of features
across the natural landscape the element of time is also treated with similar
importance. Controlling for time throughout the analysis is critical for modeling
site interaction with the environment and other sites. Hence, examination of the
data is approached in a diachronic fashion with specific emphasis placed on the
distribution of archaeological sites for chronological periods defined by the ceramic

5.3.1 Examination of Environmental Variables
In general terms, the analyses of site distributions for each period (i.e. PH,
Pill and PIV) are explored (1) with respect to physiographic zones and (2) in
relation to resource distribution based on catchment areas. Examination of site
distributions in this way assists in recognizing any broad ranging trends in resource
preferences and the locations of community settlements through time. In doing so,
it is possible to establish a spatial context within which sites are distributed based on
topographic and ecological factors.
The results of this analysis are briefly compared to previous investigations in
the region to see if there are differences or similarities in previously observed
patterns in distribution (i.e. Lundberg 1985; Rodriguez 1985). One element of this
analysis that has not been previously conducted in depth is the examination of site
distributions in relation to physiographic zones through catchment analysis.
Examination of site catchment areas further our understanding of subsistence
economy, resource exploitation and to a large extent the integrative relationships
between sites within the settlement system. Traditionally, site catchments are
typically determined in one of two ways: simple Euclidian distance, travel time, and
natural boundaries.

The natural boundary method determines catchment zones based on natural
regions, such as watershed, mountain range, or a valley. This has the advantage of
providing an easily defined survey boundary, and there are a number of
archaeological surveys that have used this method to establish catchments. The
main problem with this method is that it does not reflect social and economic
boundaries. Since site catchments are really about socio-economic use areas,
catchments based on natural boundaries are less a definition of site catchments than
they are of archaeological survey boundaries.
Euclidian distance catchments are determined by placing a radius of
predefined distance around a site. This concept for modeling site catchments is
based on the assumption that The further the area is from the site, the less it is
likely to be exploited (Viti-Finzi and Higgs 1970:7). This method of determining
site catchments is relatively simple and provides boundaries that are easy to
manage. Based initially on !Kung observations by Lee (1969:61), many catchment
studies have been focused on 5 or 10 km procurement radius. This premise is based
on the spatial context of the site and the resources within a days walking distance.
According to Jarman Human populations are generally only able to exploit
resources that exist within a certain distance of their occupation site, be this a base
camp, cave, village or town (Jarman 1972:706).

Yet as the functionality of sites and their spatial contexts within their
environmental surroundings vary, so too should the use of the landscape. This
differential intensity of resource exploitation has led to the notion of concentric
areas of intensified exploitation, or zonation, surrounding the site (Viti-Finzi Higgs
1970; Flannery 1976:92). The problem with Euclidean distance models is that they
do not take into account the natural terrain in determining site catchments. For
example, if a site is bounded on the east by a flat plain and on the west by a steep
ascent, the true catchment of the site will likely extend farther onto the plain than it
will up the steep ascent.
In order to model the extent of these differential use patterns, and
accommodate the weaknesses inherent in both Euclidian and natural boundary
catchments, this study combines the two through the use of cost distance modeling
in the GIS. Catchments were generated for each period for sites classified as
villages. In this analysis, it is assumed that villages form the nexus of economic
utilization on the landscape. Catchment areas are utilized in an attempt to
recognize patterning that may exist between the location of archaeological sites and
the distribution and natural resources and to model socio-economic boundaries.
Through this it is possible to examine the spatial characteristics of the settlement
system in relation to the environment and assess the effect of these distributions in
terms of socio-political development and organization.

Much like the travel time models, cost distance models measure cost units
across a surface, rather than distance, to calculate a least accumulative cost surface
from a point location or locations. Included in this cost measurement are surface
distance and impedance factors based on characteristics of the natural environment.
The cost distance model differs from simpler models in that it incorporates a z-value
for calculating true distance traveled. In the analysis presented here, the slope data
set derived from the DEMs serves as a cost surface. The slope data set was
reclassified to represent costs for traveling across the surface of the landscape
(Table 5.5). In doing so, catchments are quickly calculated in the GIS to conform to
variations in the topography. The site catchments examined here have been
modeled to the equivalent cost of walking 5 km over a flat surface at 1 km cost
increments (to denote differential land use based on the distance away from the
Slope Range Cost
0-10 1
10-20 2
20-30 3
30-90 4
Table 5.4: Slope ranges and costs.
Catchments are also utilized to examine settlement in terms of physical
space and agricultural productivity. In terms of size, catchment sizes are explored
to suggest influences of territoriality and or environmental pressures that may exist

based on variation in the frequency of sites in the region through time. Second,
analysis of relative agricultural productivity of catchments through time offers a
second line of inquiry regarding one aspect of socio-economic behavior. The focus
of this later analysis based on the area of land occupied by villages and the soil
types within them. Although catchments are modeled to 5 km cost equivalents,
emphasis of agricultural productivity for communities is emphasized at the 1 km
cost boundary. Chilsom (1968) indicates that the majority of a communitys
economic activities occur within 1 km of the site. In this study, the 1 km range
around a given community is considered to be fully exploited (Flanneiy 1976; Viti-
Finzi Higgs 1970). Productivity potential of a 1 km catchment (based on soils) was
calculated by multiplying the agricultural potential value by the area occupied by
the soil type within the 1km cost boundary of a site. Subsequently the values were
summed to achieve an agricultural potential for the village.
The exploration of ceramic age communities in relation to the natural
environment are geared toward understanding the settlement system in terms of site
location, socio-economics and settlement. These factors offer a basis for
elucidating the influences the environment may have had on the development of
socio-political organization. However, as the landscape and its ecological
characteristics influence the location of sites so too must the interaction of

communities across it also be considered. In this way, it is possible to offer a
holistic context for the development of regional socio-political complexity.
5 J 2 Examination of Socio-Political Variables
The development of complex socio-political settlement systems in Puerto
Rico has typically been focused on sites possessing monumental architecture and
large-scale geographic distributions through time (e.g. Siegel 1999). As these
analyses are beneficial for examining broad trends in socio-political organization,
they often focus solely on these types of sites and ignore the morphology of
settlement system within less generalized geographical contexts. This portion of the
analysis attempts to discuss the morphological development and organization of
regional socio-polities and the integration of sites within them.
Based on archaeological evidence within the study region, pronounced
increase in the frequency of ball courts and plazas in the early Ostionoid period
appears to coincide with differences in household size, mortuary patterns, ceramics
and other forms of material culture (Curet and Oliver 1998; Rouse 1992; Siegel
1996,1999). These changes suggestively reflect transitions in behavior associated
with the redevelopment of social organization from earlier forms evident Period n.
The most interesting of these developments, the appearance of formalized
monumental architecture, implies an intensification of communal ritual activity and
differential rank in status (Curet 1996). Further, these structures have been

considered to be an indication of socially integrative mechanisms on both the
community and extra community level (Oliver 1992; Siegel 1999).
As social structures have material correlates in the types of sites distributed
within a settlement system, exploration of villages and the distribution of
monumental architecture are important for understanding regional social
complexity. Therefore, site type frequencies and locations through time should
reflect some aspects of regional organization. In this analysis, regional complexity
is explored by jointly examining site integration through proximity of communities,
differentiation through site type frequencies and, the location and number of
monumental architecture associated with communities.
For the purpose of this analysis, it is suggest that interaction on a regional
level is to a certain degree bound to rules of proximity dictated by catchment
analysis. Therefore, sites that are in close proximity to one another will have a
higher probability of interaction than those that are spaced further apart. For this
reason, further scrutiny of the cost catchments become useful for discussing,
regional polities, and the interrelationships of sites within them. To approach this,
sites will be examined in two ways to discuss socio-political organization and
First, examination of communities looks at cost catchments to denote the
development and morphology of the settlement system. Catchments are utilized in

this way in order to visually recognize possible spheres of interaction between sites
and localized socio-political groups. The utilization of catchments stems from the
assumption that these boundaries approximate the immediate territory associated
with a community. In this way, boundaries that overlap suggest the high potential
for social interaction. In turn, clusters of these boundaries may be used to denote
localized socio-political units. To address the second issue, sites are examined in
terms of proximal distances to one another in order to explore integration and or
competition of sites within the settlement system.
The regional distribution of sites in relation to one another and the relative
frequencies of types are important considerations for exploring the maturation of
regional socio-political complexity. Archaeological research in Puerto Rico has
shown that this aspect is important for examination because the manifestation of
complex settlement systems appears to be congruent with the centralization of
socio-political power (Alegria 1983; Curet 1992b, 1996; Oliver 1992; Rouse 1992;
Siegel 1989,1999). Siegel argues, that this spatial pattern reflects the political
hierarchy that had developed at this time. This assumes that ball courts and
ceremonial plazas are symbols of political consolidation" (1996:236). Hence the
exploration of site distributions through these analyses offer a means for further
consideration of the spatial patterning of communities and the implications of these
distributions on regional socio-political organization.

6. Analyses and Discussion
There are considerable limitations and constraints to any settlement pattern
analysis. These factors may be attributed to aspects pertaining to the definition of
an archaeological site, the typological classifications of sites utilized in settlement
studies, the formation processes that have influenced the current condition of the
environment and the archaeological record, and the nature of site discovery and
their recordation. Further, spatial analysis is problematic in terms of how data are
generated and the accuracy to which information is represented and utilized in a
given analysis. The present study of south central Puerto Rico is not exempt from
the problems associated with these issues.
One of the sources of data bias in this study relates to the processes of site
discovery and the nature of the sample. Sites in this analysis have been discovered
over the course of several decades, mostly through non-systematic surveys
(Lundberg 1985). Because of this, the spatial location and frequency of sites for a
given area may be misleading based on the amount of survey coverage represented
there. In the present analysis, it is assumed that most of the large, long-term
settlements for the periods of interest have probably been documented. However,
interpretations that result from the research presented here must be tempered by

the important question of whether any of those differing densities are meaningful in
terms of actual prehistoric patterns (Lundberg 1985:L5).
The utilization of typological classification to discuss sites is also
problematic. Specifically, site typologies possess the potential to obfuscate
variation within the settlement system through the homogenization of diagnostic
characteristics into subjective types. This problem is further compounded by
potential errors in site classification, within the context of this study, as many sites
on the south coast have not been thoroughly excavated. Further, no site typologies
currently exist for the region and the range of site types in this area is still somewhat
unclear. However, acknowledging variation in types of sites, and possible
behaviors associated with them, promotes the recognition of patterns in social
complexity and change through time. Utilization of types in this fashion will
hopefully catalyze researchers to further consider variation in site types for this
region and the implications of their functional characteristics on social organization.
Another bias in the present study relates to the variation between prehistoric
and contemporary environmental settings. Historic land use patterns coupled with
natural geomorphological process have changed the face of the landscape in this
part of the island. As the GIS datasets represent contemporary contexts, the present
analysis focuses on broad based trends that are conducive to representing a
generalized picture of the prehistoric ecological setting. Although it is seemingly

impossible to recreate the prehistoric environmental context of the region it is
essential to acknowledge the discrepancies that may exist between the former
ecologic settings and the contemporary datasets at hand.
Despite these issues, I believe that the problems in the data do not preclude
their utility for exploring regional social complexity through settlement pattern
analysis. Interpretations of the data flow from the assumption that the spatial
patterning of material culture is a reflection of human behavior, which is influenced
by environmental and social variables. In the present analysis, these patterns are
considered to be manifested through the distribution of sites across the landscape
and their spatial relations to one another through time. Through close scrutiny of
distributional trends within both environmental and socio-political contexts, the
nature of regional social organization and complexity become salient.
The sample consists of 95 sites (Table 6.1). This includes all sites for the
region that date from Periods II through IV.11 As some sites possess evidence of
multi-period occupation their use in the study occurs for each period in which there
is evidence that the site was utilized. Hence, the sum of the number of sites for each
period is different form the total sample. A break down of these sites by type, per
period, can be seen in Figure 6.1.

Period N Site Numbers
PII 21 COl, C03, GN13, GN14, GY1, GY3, GY4, GY 10, GY 16, JD6, POl, P04, P08, POl 1, POl3, POl9, P029, P038, SN10, YA2, YA12
Pill 63 A14, COl, C02, C03, C04, GY1, GY3, GY4, GY16, JD1, JD2, JD6, JD7, PN1, PN3, POl, P02, P03, P05, P07, P08, P09, PO10, POl 1, P012, POM, POl5, POl9, P021, P029, P033, SI4, SI6, SI8, SN2, SN4, SN10, SN11, SN14, SN15, SN16, SN17, SN18, SN21, SN22, SN23, SN25, SN26, SN27, SN28, SN29, SN30, SN31, SN32, SN33, SN34, SN35, VL4, YA2, YA11, SN3, SN24, SN6
PIV 42 JD6, All, AI5, C02, GY1, GY3, GY4, JD5, PN1, PN4, P02, YA2, YA10, POl6, P039, PO10, SN22, SN34, SN36, SN20, SI1, SI2, SI3, S14, GY10, GY11, POl 1, P06, PN5, JD4, JD3, YA9, SN6, SN7, SN10, SN13, SN12, SI7, SN2, SN3, GY5, GY6
Table 6.1: Listing of all sites used in the study by period.
Site Type Distributions Through Time

Architecture Present
B Village/Monumental
Architecture Absent
Hamlet/Limited Activity
P1I Pill PIV
Temporal/Cultural Affiliation
Figure 6.1: Site type frequencies based on period.
A cursory visual examination of site distributions for all periods displays a
propensity for communities to be oriented along major river valleys or at the
confluence of major drainages (6.2,6.3, and 6.4). The distributional pattering of

sites along major river drainages has been observed in other discussions of site
distributions for the region (Garrow et al. 1995; Rodriguez 1985). General
distributional patterning of sites, within the region, display obvious differences in
number, type and location through time. Assuming that these distributions are the
result of a range of socio-economic behaviors that influence site locations, then a
critical examination of the contexts within which sites and types of sites are located
should yield information regarding long term developments in regional socio-
political organization.

H Village/Monumental Archlecture Absent
e Harrilet/Limited Activity Area
Figure 6.2: Period II site distributions.

10 Kilometers
A Monumental Architecture/No Habitation
0 Village/Manumental Architecture Present
a Village/Monumental Archtecture Absent
Hamlet/Limited Activity Area
Figure 6.3: Period III site distributions.

6.1 Environmental ContextsResource Use, Settlement,
and Change through time
To initiate the analysis, sites were examined based on their
distribution within physiographic regions in a fashion similar to Lundberg (1985)
and, to a certain degree, Rodriguez (1985). This portion of the analysis is
operationalized on two analytical levels that seek to (1) explore trends in site
distributions based on broad ecological settings and (2) the implications of these
trends for regional settlement organization. The motivations for these varying
levels of analysis stem from the necessity to provide a spatial and economic context
as well as to recognize any differences or similarities that may exist in the
distribution of sites as observed in previous studies of the region (Lundberg 1985;
Rodriguez 1985).
To explore trends based on ecological contexts, sites were queried in the
GIS to ascertain the physiographic zone locations (Appendix D) and subsequently
these locations were on the regional physiographic base (Figures 6.5,6.6 and 6.7).
Interestingly, differential trends in the ecological contexts of site locations through
time are clearly evident (Figure 6.8). Specifically, sites during Period II appear to
be primarily located on the coastal plain and at the base of the foothill zone (with 37
and 63 percent of the Period II sample represented in these locations respectively).
During Period ID, the overall distribution of sites displays a different trend than that

relatively small sample size for this period (n=21) definitive statements regarding
resource acquisition and land use are difficult to make. What can be said regarding
Period II is that the Saladoid people preferred establishing settlements within the
physiographic contexts of the coastal plains and foothill environmental zones. As
the majority of sites during this period tend to be located in areas of transition
between these two zones it is inferred that settlement locations were selected based
on relatively easy access to multiple ecologic zones to maximize resource
acquisition. These patterns support those previously observed by Lundberg (1985),
and promote the notion that Saladoid populations were opportunistic in terms of
subsistence economy (Siegel 1991).
For Period HI, the general observations made from the distribution of site
types and physiographic zones show preferential land use and an increase in the
number of functional site types. Villages and hamlets appear to be focused in the
foothill zones seconded by settlement locations on the coastal plains. Site
distributions during Period III also expand into the upland regions of the study area.
The presence of habitation sites (albeit few in number) in this region suggests an
increase in the range of acceptable settlement locations, which in turn may indicate
changes in the perceptions of land use during Period HI. Importantly, the utilization
of this zone by communities suggests a diversification in resource utilization and an
expansion of the settlement system in general.

10 Kilometers
Figure 6.5 Distribution of Period II sites and physiographic zones.
9 Villagc/Monumental architecture develops
during Period III
B9 Villagc/Monumental Archtccturc Absent
Hamict/Iinitcd Activity Area
Physiographic Zones
f n Coasts Plains
| ] Foothills

10 Kilometers
A Monumental Architecture/No Habitation
^ Village/Monumental Architecture Present
E] Villaac/Monumental Archtccture Absent
HamKt/Linited Activity Area
Physiographic Zones
Coastal Plains
Uplands/Mottn tains
Figure 6.6: Distribution of Period III sites and physiographic zones.

Full Text
A Monumental Architecture/No Habitation
Village/Monumental Architecture Present
h Village/Monumental Archtecture Absent
Hamlet/Limited Activity Area
Figure 6.4: Period IV site distributions.

observed for Period H. The distribution of Period ID sites show a slight increase in
the number of coastal zone site locations (2%) and an 11 percent decrease in the
number located in the foothill zones. Further, a small portion of the PHI sample
(10%) can be seen distributed in the previously unsettled upland zone. Finally, PIV
is characterized by a relatively high number of sites located on the coastal plains (58
%). PIV sites also show a propensity for foothill locations (22%) and a slight
increase in the number of locations in upland (14%) and mountainous zones (6%).
Consideration of site type distributions based on physiographic zones allows
for further scrutiny of our initial observations. To address this, site type
distributions within physiographic zones were examined for each period (Figure
6.11, 6.12 and 6.13). The zonal distributions and the types of sites located within
them represent an important indicator of settlement change in relation to the
environment through time. Assuming that the distribution of sites in each period
reflects the economic behaviors associated with utilization of the land, at a given
point in time, an examination of site types within zones should denote the nature of
resource extrapolation and the ecological bounds within which communities
developed. Examining trends in site type distributions throughout the region can be
utilized to note continuities or discrepancies in our previous observations.
In contrast to our initial observations, villages during Period II appear to be
equally distributed within the coastal and foothill zones. However, based on the

Mean Area Physiographic Zones within 5km Catchment Areas per
Is l l 58
36 42 42
t- I 12 22 14 r
1 4 ran v*' A 1L.
P1I pin PIV
Coastal Plains 36 42 58
Foothills 63 42 22
Uplands 1 12 14
Mountains 4 6
Coastal Plains
Tempofal/Cultural Affiliation
Figure 6.15: Mean percent area of physiographic zone types occupying 5 km cost catchments for all
6.1.2 Site Catchments and Soil Productivity
Upon initial examination of the mean catchment size for each period,
general trends in the area encompassed by these boundaries through time become
clear (Figures 6.20). The size of Pin catchments shows that the immediate area
associated with a given community averaged approximately 4,971.9 acres. In
comparison to Period II (9,675.0 acres) and Period IV (7,143.4 acres), site
catchments for Period HI appear to be considerably smaller.

In consideration of these issues, an analysis of variance was run on
catchment size. In this ANOVA, significant variation in the changes in size through
time would denote either a potential source of competition from increases in
environmental resource use and decrease in their availability and or an increase in
cooperation through increased contact with neighboring community clusters. To
test this we consider the null hypothesis to show no variation in the size of the
catchments through time, which implies no change in regional interaction based on
available land. Alternatively, significant variation would suggest changes in
regional interaction through either competition and or cooperation based on land
Based on the results of the analysis, there appears to be significant
difference in the catchment sizes through time (Tables 6.2). In conjunction with the
observation of size variations through time this would suggest the consolidation of
socio-economic boundaries. From an economic perspective, less land to exploit
would put pressure on the economic system if populations exceed carrying
capacities. From the socio-political perspective it could be suggested that these
encroaching boundaries represent an increase in territoriality (Siegel 1999:220).
Period N Mean F Sig.
II 10 9675.0 3.811 .028
m 29 6528.0
IV 17 7983.7821
Table 6.2: Results of ANOVA for catchment size through time.

The observed clustering of sites for the region through time have been
previously discussed in this study in terms of the potential influence of the
environment on the location of communities. However, consideration of why
communities appear to cluster independent of the environment is also important.
Typically, clustered patterns of distribution are viewed as the result of either mutual
attraction to central places and or formation process in the development of new
villages form from older parent communities (Flannery 1976:197).
As new communities formed, integrative mechanisms would have been
beneficial to maintain ties with parent villages, or solidify socio-political
relationships between sites. Therefore, an important aspect to consider is the
number of monumental architectural features at sites within each cluster. Sites with
high numbers of these features are assumed to be areas of socio-political power
within the cluster. Further, these features represent mechanisms of integration both
at the community and extra-community level. Hence, the presence of these features
both at the level of the community and the community cluster may be utilized to
speculate socio-political importance of a given site and the level of integration
within a given cluster.

0 4 Kilometers
6.22: Period III socio-political units defined by clusters of 5 km cost
Villagc/Monumental Architecture Absent
2 Villagc/Monumental
^ ^ Architecture Present
| ] Socio-political Units
] 5 km Cost Boundaries

. r
5 Kilometers
B Village/Monumental Architecture Absent
Hamlet/Limited Activity Area
] Socio-political Units
j 5 kmCostBomdarics
A ,
/ ^ / Village Networks
,* \ Hamlct/Activity Networks
Figure 6.27: Period III west integrative networks

Catchment boundaries are less constricted then those observed for the
proceeding period. The relatively moderate soil productivity values within these
catchments suggests that agricultural production was not highly intensive during
this time. Otherwise, we would expect settlements early on in the occupation of the
region to occur in areas characterized by relatively high productivity values. These
items coupled together suggest low levels of regional competition and
environmental pressure.
Small clusters of long-term habitations! sites that do not display evidence of
formalized ceremonial architecture characterize socio-political organization of the
region during this time. However, looking ahead to Period IE, considerations of
where these ceremonial features develop potentially indicate communities of socio-
political importance. During Period H, these clusters can be seen in the central
northeast, central and western portions of the study region. Although the clusters do
not take on any observable regular geometric shape as suggested by traditional
central place models, the proximity of sites within community clusters promotes a
notion of interrelatedness to some degree. Based on observations of the data, the
relatively high level of clustering would seem to be a result of parent/daughter
community relations.
In the central eastern portion of the study region, the sites of Las Flores and
Buenos Aires are located in the foothill zone of the Coamo River Valley

The western portion of the study region is a relatively high-density area
during Period II. The village sites in this cluster consist of Tecla, La Vega, and
Diego Hernandez. Based on the boundaries established by the 5 km cost distance
model, the relationship between the first and last of the three communities show the
least potential for interaction. For this region, it appears that the site of Tecla
represents the potential area of socio-political importance based on its relatively
large settlement size, extensive deposits of material remains, and evidence of
extensive continual occupation from Period II through Period TV (Lundberg 1985;
Moscoso 1978). Although there is no documented evidence of ceremonial features
at the site it has been suggested that such features may have existed but were
In general, it is suggested that Period II is a time of settling down and
settling in for the Saladoid peoples. The socio-political system of Period II appears
to consist of three small community clusters of localized social interaction. In the
central northeast portion of the region, these interactions seem likely to have been
centered on the site of Las Flores based on its substantial size and presence of
formalized ceremonial features that develop in Period HI. In the central portion of
the study area, the site of Tibes appears to be an important site of regional socio-
political interaction as it develops the greatest number of formalized multi ballcourt
and plaza features relatively early during Period HI. Finally, in the east, the site of

The high density of sites in the eastern portion of the study region for Period
in sites is interesting to note. Specifically, the development of densely populated
community clusters in this region points an expansion of burgeoning parent
communities from the eastern portion of the island. Arguably, if settlements were
derived from either the Tibes or Tecla regions, expansion and high-density clusters
would most likely appear to form in these areas first (Flannery 1976). However, the
development form these areas must not be ruled out completely.
Sites during this period are distributed amongst the coastal, foothill and
upland physiographic zones respectively. The distribution of the physiographic
zone types within 5 km cost catchments of villages suggests an equal importance
upon coastal and foothill regions for resource extraction. Notably the equal
representation of these zones within the catchments of communities suggests a
greater maximization of access to variety of resources. This maximization of land
use is not only reflected at the regional level but also at the level of community
Maximization of community production is also suggested by an increase in
site locations in areas of high soil productivity. The implications of this possible
maximization are twofold. First, increased maximization of land during this time
imply reduced resource availability that may have promoted competitive and or
cooperative behaviors. Second, increased maximization may have been the result of

Of particular interest during this period is the development and distribution
of sites with monumental ceremonial architecture that lack evidence of long-teim
habitation. This is important because ceremonial sites without evidence of long-
term domestic habitation suggest more specialized forms of behavior within the
regional socio-political landscape. Interestingly, these ceremonial site types neither
reach the size nor possess the number and elaboration of features represented at
village sites in this part of the island. It has been suggested that these sites represent
localized ceremonial sites to integrate adjacent communities or to perform specific
specialized religious functions for small communities that do not possess these
formalized ceremonial features (Garrow et al. 1995). In support of this supposition,
examination of the community cluster boundaries shows that these sites tend to
appear on or near the edge of village cluster borders. However, these sites are few
in number and their function within the settlement system is not fully understood.
During Period El there are 3 major community clusters. These are located
in the eastern, central and western portions of the study area respectively.
Suggestively, the units in the west and central areas grew from integrated
communities that developed during Period H However, the eastern portion of the
region appears to represent an expansion of settlements from the eastern areas
outside the study area. The eastern cluster is extremely ambiguous due to high site

Camp Santiago P14 and Esperanza are located over 5 km away from the PI6
site. Based on the distance between the sites and the distribution of ceremonial
features at the periphery of the Esperanza site suggests some sort of integrative
network between the sites. However, based on the location of Esperanza and Camp
Santiago PI 2 at the edge of the study area, these sites may be more connected to
sites eastward, outside the study region.
In the central portion of the study area, an increase in sites from the previous
period is also evident. Of note are the Caracoles and Ana Maria sites develop in
close proximity to Tibes. As these sites also possess ceremonial architecture, the
appearance of these villages during this time within the community cluster would
indicate an increase in social integration and social complexity. During Period ID
the development of Tibes numerous ball courts and plazas suggests its importance
as a regional socio-political center. In addition to the potential relationships of sites
discussed for this area during Period It, a number of new sites appear to develop in
the Cerrillios River Valley north of El Bronce. The relationships between these
sites remain unclear (Garrow et al. 1995).
Trends in the decrease of stylistic variability in ceramics during this period
have been viewed as the result of differential access to symbols for centralization of
socio-political power (Curet 1996). Suggestively, this stylistic change might
represent the homogenization of local behavior and the centralization of formalized