Journal of Anthropological Archaeology 40 (2015) 196–212

Contents lists available at ScienceDirect

Journal of Anthropological Archaeology

journal homepage: www.elsevier.com/locate/jaa

Inequality, demography, and variability among early complex societies

in Central Pacific Panama
C. Adam Berrey
Department of Anthropology, 3302 WWPH, University of Pittsburgh, Pittsburgh, PA 15260, USA

a r t i c l e i n f o a b s t r a c t

Article history: The early complex societies of Central Pacific Panama have long been recognized by anthropologists for
Received 12 December 2014 their strong levels of social hierarchy. Such hierarchy is apparent in the ethnohistoric texts of the 16th
Revision received 14 August 2015 century, and in elaborate burial assemblages of the Late Ceramic II period (AD 700–1522). Not surpris-
ingly, those regions in which hierarchy is most apparent have been the main focus of archaeological
research, while those in which inequalities were apparently weaker have received less attention. These
Keywords: latter regions, however, are also a vital part of understanding hierarchical development—they can bring
Early complex societies
into sharper relief the factors that gave way to hierarchical organization in some regions, as well as those
Chiefdoms
Social inequality
factors that may have discouraged it from developing in others. Regional settlement data from the Río
Settlement analysis Parita and Río Tonosí valleys provide an opportunity to explore this issue. These data suggest that the
Regional demography hierarchical variability observed between these regions may have arisen as a result of differential levels
Environmental risk of regional population growth and environmental risk, which gave way to different patterns of land use
Local interaction and structures of local interaction. These different interaction structures facilitated different sorts of
Panama activities and inter-household relationships, leading to markedly different forms of social organization.
! 2015 Elsevier Inc. All rights reserved.

1. Introduction Despite the highly varied forms of social organization that

In recent decades one major objective of early complex society
research has been to understand the immense amount of variabil-
ity that exists in the way early complex societies developed (e.g.
Drennan and Peterson, 2006; Earle, 1997; Kirch, 1984; Peterson
and Drennan, 2012). Particularly important to this endeavor has
been an attempt to understand the highly variable role played by
social inequality, and the forces that shaped early inequality in
such a variety of different ways. Why was it, for instance, that
some societies came to be organized according to very strong hier-
archical principles, while others exhibit little in the way of status
differentiation? Moreover, why among those in which inequalities
emerged did factors such as economic control, specialization,
ritual, and warfare play such highly variable roles? Archaeologists
have approached these questions from a number of different van-
tage points, resulting in a diverse range of models aimed at under-
standing the various contexts, forms of behavior, and sets of
activities through which early inequalities came to develop (e.g.
Beck, 2003; Blanton et al., 1996; Earle, 1997; Renfrew, 1974;
Sanders and Webster, 1978; Scarborough and Burnside, 2010).
E-mail address: cab10@pitt.edu
developed among early complex societies, though, one need not
scan the globe for such variation to become apparent. Within the
limits of what is often referred to as the Intermediate Area a great
deal of variation can be clearly observed (Fig. 1). Toward the end of
the first millennium AD ritual and ideology were the primary basis
for social power in the Alto Magdalena of Colombia (Drennan,
2000; Drennan and Peterson, 2006), whereas in the western llanos
of Venezuela such power was more firmly connected to warfare
and agricultural production (Redmond et al. 1999; Spencer and
Redmond 1992, 1998). Social inequalities were highly developed
in both of these areas (and conspicuous in the form of monumental
architecture), but were much weaker (or at least much less appar-
ent) in many other parts of the Intermediate Area.
As small as the Intermediate Area is, the geographic scope
across which variation can be observed could be narrowed even
further. As one homes in on increasingly smaller scales one begins
to approach areas within which societies would not only have had
regular contact with one another, but would have shared impor-
tant sociocultural roots and traditions. These areas, or what here-
after are referred to as macroregions (so as to distinguish them
from the distinct trajectories of regional development that exist
within them; cf. Kowalewski, 2004), represent particularly inter-
esting contexts in which to explore the variable pathways of early

http://dx.doi.org/10.1016/j.jaa.2015.08.001
0278-4165/! 2015 Elsevier Inc. All rights reserved.
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
Fig. 1. Culture areas, or macroregions, of the Intermediate Area discussed in the
text.
complex society development. Within these contexts such varia-
tion is not simply the consequence of different historical trajecto-
ries, but of divergent evolution sparked by specific forces of social
change (e.g. Flannery and Marcus, eds. 1983; Kirch, 1984; Linares
and Ranere, eds., 1980).
Central Pacific Panama (Fig. 1) represents one part of the Inter-
mediate Area that encompasses considerable variation of this sort.
Such variation is most apparent during the Late Ceramic II period
(AD 700–1522), which marks the height of sociopolitical develop-
ment in the macroregion.
2. Inequality and variability in Central Pacific Panama
For many years much of what was known about the early com-
plex societies of Central Panama was based largely on the rich and
vivid descriptions that were left by the Spanish in ethnohistoric
accounts (e.g. Andagoya, 1865; Jopling, ed., 1994; de Las Casas,
1986). These accounts document the existence of relatively large
and powerful chiefdoms (Fig. 2), notably those of Natá, Escoria,
and Parita (Helms, 1979:56), organized on the basis of strongly
hierarchical principles and engaged in a seemingly wide range of
specialized activities. Though high-ranking social positions were
often ascribed at birth (Helms, 1979:23–28), warfare was an
important avenue of social mobility and source of chiefly authority
(Helms, 1979:13, 31–37; de Oviedo, 1959:28–29). Interregional
exchange was also important to chiefly power, which, combined
with warfare, helped fuel the political economy and ideology on
which that power relied (Helms, 1979). While craft specialization
and resource control are also documented in ethnohistoric texts
(Helms, 1979:14–15, 57, 1994), Helms (1979) argues that warfare,
exchange, and, above all else, ideology were the mainstays of social
power among Panama’s 16th-century chiefdoms.
Social inequality seems to have manifest in various arenas of
social life, but nowhere was it expressed more extravagantly than
during elaborate funerary rituals. One oft-cited expression of this
inequality was documented at the funeral of Chief Parita (Cooke
197
Fig. 2. Ethnohistoric chiefdoms (gray capital letters) and archaeological sites in
Central Pacific Panama. The regional survey zones of the Río Parita and Río Tonosí
valleys are shown in gray.
et al., 2003:120; Flannery and Marcus, 2012:221–222; Haller,
2008a:3; Lothrop, 1937:46). Ethnohistoric sources indicate that
Parita was decked head-to-toe in gold ornamentation, and buried
with attendants and competing chiefs that were offered as sacri-
ficed victims (see Lothrop, 1937). A funerary investment such as
this was obviously not afforded to everyone, and is a clear indica-
tion of strong inequality during the 16th century. Such inequality
was nothing new to the indigenous societies of Central Panama,
however, as comparable levels appear to have emerged some
700 years beforehand, at the onset of the Late Ceramic II period
(AD 700–900).
2.1. Social inequality during the Late Ceramic II period
The very lavish graves recovered from Sitio Conte (Fig. 2) have
long been recognized as a prehispanic manifestation of the social
power that existed in Central Panama during the 16th century
(Drennan, 1995:323; Flannery and Marcus, 2012:219–222;
Linares, 1977:72). Not only were these graves stocked with the
most elaborate funerary assemblages throughout the macroregion,
but they stand out as being some of the most elaborate burials seen
among early complex societies around the world (Drennan et al.,
2010). The most elaborate of these graves (Burial 11) contained a
total of 23 adult individuals, mostly males, one of whom is believed
to have been the principal figure for which the others were likely
sacrificed. This individual was centrally located among the other
bodies (see Hearne and Sharer, eds., 1992:9; Lothrop, 1937:50,
Fig. 31), and was associated with many of the graves more notable
gold offerings (Drennan et al., 2010:47–48). In total these offerings
included 3496 beads; 233 ear rods; 87 bells; 31 medallions or pen-
dants; 17 chisels; 13 plaques; 10 cuffs, wristlets, or anklets; and 6
nose ornaments. Additional offerings (those not made of gold)
included at least 152 polychrome and 159 monochrome ceramic
vessels, 1548 stone projectile points, 168 stone celts, 4 agate
198 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
pendants, 68 bone points, 100 shark teeth, 321 canine teeth, 176
rabbit teeth, and as many as 205 stingray spines (Briggs,
1989:199–203).
Though Burial 11 was the most elaborate grave assemblage
recovered from Sitio Conte, many others were also very lavishly
stocked (see Briggs, 1989). A cluster analysis of these assemblages
exhibits a very marked hierarchical structure (Briggs, 1989), much
like those of other early complex societies where strong social hier-
archies are known to have developed (Peebles and Kus, 1977).
Comparable assemblages have recently been recovered from the
nearby site of El Caño as well (on the opposite side of the Río
Grande; Fig. 2), which seem to correspond to this same period of
time (AD 700–900; Mayo and Mayo, 2013; see Williams, 2012).
Based on the artifacts recovered from these graves warfare seems
to have been an important element of social power in the Río Grande
valley. Gold armor, various forms of weaponry, and additional
bodies (presumably sacrificed victims) were often interred with
elite individuals (see Briggs, 1989; Lothrop, 1937; Williams, 2012),
and high-status burials were almost exclusively adult males
(Briggs, 1989:72–75; Cooke et al., 2003:124). These patterns are
consistent with sculptural portrayals of ‘‘decapitated” individuals
with their hands bound behind their backs (Mayo et al., 2010), and
with what archaeologists have identified as defensive architecture
in the northern reaches of this region (Mayo et al., 2007).
It is also clear that lavish funerary rituals were a primary arena
in which status distinctions were expressed. This is apparent not
only in the extravagant funerary assemblages that were interred
with elite individuals, but in the formal ritual architecture that
was constructed at each site. Such architecture included stone
altars, basalt stone columns, and, in the case of El Caño, cobble-
paved pathways and small earthen mounds (see Lothrop
1937:39–43; Mayo and Mayo, 2013; Mayo et al., 2007, 2010;
Mojica et al., 2007). The large stockpiles of finely crafted items that
were recovered from these sites (which included gold ornaments,
shell jewelry, and other prestige goods; see Hearne and Sharer,
eds. 1992; Williams, 2012) suggest that craft specialization and
exchange were also important elements of social power, but there
is currently little that can be said about the organization of these
activities in the Río Grande valley (but see Harrison and
Beaubien, 2010).
Despite the marked levels of social power that are evident in
this region between AD 700 and 900, such power seems to have
been relatively short-lived. In what some have argued marks an
episode of chiefly cycling in Central Panama (Menzies and Haller,
2012a), it was around AD 900, as population levels dropped and
inequalities began to wane in the Río Grande valley, that social
inequalities in the Río Parita valley began to reach new heights.
It was at this time that small earthen burial mounds began being
constructed at the central place of El Hatillo (Fig. 2), which were
Fig. 3. Layout of small earthen burial mounds at the sites of El Hatillo (left) and Guaniquito Abajo (right). Redrawn from Bull (1965:32, Fig. 1) and Ichon (1980:394, Fig. 129).
used for the interment of elite individuals. A total of 13 mounds
were ultimately constructed, and by AD 1300 they had come to
delimit a formal plaza area within which ceremonies were presum-
ably held (Fig. 3). Detailed inventories are unfortunately lacking for
many of the graves that were excavated (due to looting and ama-
teur excavation; see Haller, 2008a:93–95), but those which exist
document the presence of relatively elaborate funerary assem-
blages. Mound 2, for instance, contained the remains of three adult
males who were associated with some 100 monochrome and poly-
chrome ceramic vessels; two metal ornaments; 1102 shell beads
and 29 shell ornaments; 7 jadeite beads; 29 beads and ornaments
made of bone; seven animal teeth; two quartz crystals and one
dark, unidentified stone (Bull, 1965:36–38; see Menzies and
Haller, 2012a:457). Approximately 30 well-crafted gold ornaments
were recovered from another grave within these mounds (Biese,
1967; see Bray, 1992:44–45; Cooke, 2004:277; Haller,
2008a:101), while yet another appears to have included a gold hel-
met (Roosevelt, 1979:81), an item which at Sitio Conte was
reserved for high-ranking individuals (Briggs, 1989:138).
While the grave assemblages recovered from these mounds
were not as elaborate as those from Sitio Conte or El Caño they
are to date some of the most elaborate anywhere in Central
Panama outside of the Río Grande valley, and attest to relatively
marked status disparities at the site of El Hatillo. Such disparities
were also evident in matters of daily life, as high-status house-
holds appear to have enjoyed a better diet and greater access
to socially valued goods than other members of their community
(Menzies, 2009). This is evident in the distribution of certain arti-
facts and ecofacts within El Hatillo, which indicate that elites had
greater access to deer meat, higher quality cuts of meat, poly-
chrome ceramics, and bone ornaments and beads (Menzies,
2009).
As in the Río Grande valley artifacts recovered from elite graves
at El Hatillo suggest that warfare was an important aspect of social
inequality. Elaborate weaponry (Cooke, 2004:277–278), gold-
plated armor (Roosevelt, 1979:81), ornaments bearing ‘‘warrior”
iconography (Bray, 1992:45), and jewelry made from human
remains (Haller, 2008a:107–109) were among the items interred
in high-status burials, which again were predominantly, if not
exclusively, those of adult males (Haller, 2008a:93). Elites were
also more actively engaged in certain craft activities and non-
local exchange, which included the production of shell ornaments
and the importation of non-local ceramics (Locascio, 2010;
Menzies, 2009; Menzies and Haller, 2012b); by AD 1100 they
had begun hosting communal feasts as well (Locascio, 2010; see
Menzies, 2009:114). These feasts may sometimes have been held
in conjunction with elaborate mortuary rituals that are evident
at El Hatillo, in the form of well-stocked funerary assemblages
and formalized ritual space (see above).
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
2.2. A tradition of aggrandizement
Archaeological and ethnohistoric data allude to important dif-
ferences between the early complex societies of the Río Parita
and Río Grande valleys during the Late Ceramic II period, and with
the ethnohistoric chiefdoms of the 16th century. These data, how-
ever, also attest to important overarching similarities. To begin
with, it is clear that inequality in each of these contexts was
expressed through rather ostentatious displays of status upon the
death of elite individuals, which took the form of elaborate funer-
ary rituals for which formal space was often constructed. It is also
clear that some form of warfare or violence was an important
activity in each of these contexts, as was some degree of craft spe-
cialization and interregional exchange. There is much about the
organization of these economic activities that remains poorly
understood, but in each case they seem to have emphasized the
production and movement of important prestige items.
Taken together these activities are consistent with a mode of
inequality that was rooted in conspicuous aggrandizing behavior,
or what some might refer to as network (Blanton et al., 1996) or
individualizing (Renfrew, 1974) modes of social hierarchy. Similar
forms of inequality have been documented among early complex
societies in other parts of the world, including the Valley of Oaxaca
between 1150 and 500 BC (Marcus and Flannery, 1996), and the
Philippines between the 10th and 16th centuries AD (Junker,
1999). Inequality in these regions was rooted in a wide range of
activities that are often associated with aggrandizing behavior,
including public ritual, craft production, interregional exchange,
and warfare.
These same elements of social power helped underwrite
inequalities in Central Pacific Panama, and did so in different
regions and different periods of time. Though these regions are
not widely separated they nonetheless represent relatively distinct
trajectories of early complex society development. A ‘‘tradition of
aggrandizement” thus seems to have developed in Central Pacific
Panama, which emerged at the onset of the Late Ceramic II period
(ca. AD 700) and persisted until the 16th century. Not only did this
tradition characterize the nature of inequality among the most
powerful prehispanic societies throughout the macroregion, but
also among the most powerful chiefdoms during the ethnohistoric
period.
While the fact that such a tradition ever came to emerge is an
interesting issue in and of itself, what is equally interesting is the
fact that it did not emerge everywhere. There are other parts of
Central Pacific Panama where prehispanic societies also developed,
but that, while sharing sociocultural characteristics with those of
the Río Grande and Río Parita valleys, do not exhibit the same fea-
tures of organization that were common to these areas. Archaeo-
logical evidence suggests that the Río Tonosí valley represents
one such region (Fig. 2).
2.3. Variability in early complex society development
As in the Río Grande and Río Parita valleys, small earthen burial
mounds began being constructed in the Río Tonosí valley during
the Late Ceramic II period. Nowhere were these mounds more
numerous than at the site of Guaniquito Abajo (Fig. 2). In contrast
to the mounds that were constructed in other regions, however,
those in the Río Tonosí valley do not seem to have been used for
the ostentatious display of status on the part of elite individuals.
The most elaborate grave assemblage recovered from the mounds
at Guaniquito Abajo, for instance, contained only one gold bead
and one gold pendant, two greenstone pendants, one unworked
nodule of jade and another of quartz, one incised bone, one fos-
silized shark tooth, and only 36 monochrome and polychrome
ceramic vessels, many of which Ichon (1980) refers to as having
199
been of ‘‘poor quality” (1980:402). The grave assemblages recov-
ered from these mounds thus seem to have been much more mod-
est than those at Sitio Conte, El Caño, or even El Hatillo, and are in
fact much more reminiscent of those from earlier graves in the Río
Tonosí valley—graves which are generally taken to represent egal-
itarian organization (Briggs, 1989). Also lacking from these assem-
blages is any indication of warfare or violence, or of any significant
degree of craft specialization or interregional exchange (e.g. Ichon,
1980:402). Unlike the formal plaza areas that were constructed in
the Río Grande and Río Parita valleys, the mounds that were con-
structed in the Río Tonosí valley exhibit no such delineation of
space (Ichon, 1980). Lacking such space topography would have
presented an impediment to any large-scale social gathering
(Fig. 3).
The mortuary record of the Río Tonosí valley thus suggests that
societies in this region developed rather differently than in other
areas of Central Pacific Panama. In contrast to the more pro-
nounced status differentiation that is evident in the Río Grande
and Río Parita valleys, social inequality in the Río Tonosí valley
appears to have been less developed. Though concerns have been
raised about inferring inequality from the material remains of mor-
tuary practice (Pearson, 1982; Ucko, 1969), there is good reason to
believe that the mortuary variability observed in Central Panama is
indeed representative of differences in the degree of inequality
that developed in different regions. To being with, as Briggs
(1989) has demonstrated with the graves from Sitio Conte, status
in this context was expressed through an additive process
(1989:138), whereby the higher the status of the interred individ-
ual, the larger and more diverse the grave assemblage was (cf.
Peebles and Kus, 1977). While there may exist many contexts in
which inequality and other aspects of social identity are not
reflected in mortuary treatment, the Río Grande valley does not
appear to have been one of them. If this was one of the many cus-
toms shared among the early complex societies of Central Pacific
Panama (which included shared ceramic styles, iconography, and
the use of small earthen burial mounds), then the most likely con-
clusion to draw from these data is that social inequality in the Río
Tonosí valley was less developed than in other regions.
Even if the individuals interred within mounds at Guaniquito
Abajo did hold some sort of elevated status, which Ichon
(1980:405) suggests they did not, it is clear that such status
did not merit the material investment that was afforded elites
in other regions. There is also not much evidence for any of the
activities that were associated with elite behavior in other parts
of the macroregion, but this cannot be solely explained by the
seemingly undeveloped nature of inequality in the Río Tonosí
valley. Craft specialization can arise irrespective of strong hierar-
chical principles (e.g. Spielmann, 2002), and warfare has been
documented among groups with only incipient levels of social
inequality (e.g. Berndt, 1964; Podolefsky, 1984; Sahlins, 1961).
These aspects of organization nonetheless do not seem to have
developed among early complex societies in the Río Tonosí val-
ley. While it is true that this region has not been subject to the
same amount of intensive, local-scale research as other areas of
Central Panama (notably the Río Parita valley), the fundamental
differences that are addressed above can all be deduced from
mortuary data. These data are common to all of the regions being
discussed.
Understanding the sorts of variability that can be observed
between the developmental trajectories of the Río Grande and
Río Parita valleys, on one hand, and the Río Tonosí valley, on the
other, represents a major objective of early complex society
research. Because populations in these regions likely share a com-
mon ancestral origin (cf. Ichon, 1980) dating back to the initial
occupation of Central Panama (ca. 11,000 BC; Cooke, 2005:136–
137; Cooke et al., 2013), this area provides an excellent context
200 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
in which to home in on the forces of social change that led to such
variability. The research that follows attempts to do so by compar-
ing the regional trajectories of the Río Tonosí and Río Partia valleys
(Berrey 2014a,b; Haller, 2008a,b), focusing on patterns of regional
settlement demography. Specifically, this study takes aim at com-
paring regional population densities, the scale and compactness (or
structure) of local communities, and patterns of population and
resource distribution. These aspects of regional settlement demog-
raphy each represent a distinct axis of variability along which
these two regions may (or may not) differ. Understanding how
each region falls relative to one another along these axes, as well
as how their positions along one axis compare to their positions
along another, can provide valuable insight into the nature of the
variability discussed above. The ways in which these aspects of
demography are evaluated are described in the following section,
followed by a discussion of how they relate to the hierarchical vari-
ability observed.
3. Field and analytical methods
Both the Río Tonosí and Río Parita valleys were subject to sys-
tematic, full-coverage regional survey (see Berrey, 2014a:25–29;
Haller, 2008a:27–33). In both regions field teams walked across
the landscape at fixed intervals not exceeding 50 m, and artifacts
were collected within areas no larger than 1 ha; any distribution
of artifacts exceeding this area was thus subdivided into multiple
collection lots. Though both surface collections and shovel probes
were used in each of these two surveys, shovel probes were far
more common in the Río Tonosí valley, owing to the dense vegeta-
tion that characterized much of this region at the time survey was
conducted. Surface conditions were far more favorable in the Río
Parita valley, and of the 50 shovel probes that were excavated,
none of them yielded cultural remains (Haller, 2008a:31, 2008b);
this in contrast to the 1557 that were excavated in the Tonosi
region, 339 of which generated cultural material, accounting for
just over one-quarter of all the collections made (Berrey,
2014a:25–29). In the end a total of 50 contiguous km2 were sur-
veyed in the Río Tonosí valley, while a total of 104 km2, divided
between two separate survey zones, were covered in the Río Parita
valley (see Haller, 2008a:21). This comparison focuses on the 90
contiguous km2 that constitute the Lower Survey Zone, within
which El Hatillo is located.
For both regions demographic estimates were based on the area
and density of ceramic sherd scatters, which were taken as a proxy
for the area and density of human occupation (Berrey, 2014a:31–
34; Haller, 2008a:39–43). These proxies were then converted into
absolute population estimates. In the Río Parita valley this was
done using a combined index of occupied area and sherd density,
as well as ethnohistoric data from the site of Natá, whose residen-
tial density was taken to approximate that of El Hatillo’s during its
demographic height (see Haller, 2008a:127–129). The degree to
which the area-density index of individual collection lots deviated
from that of El Hatillo’s then served as the basis on which to assign
them absolute population estimates. In the Río Tonosí valley abso-
Fig. 4. Ceramic chronologies for the Río Tonosí and Río Parita valleys during the Late Ceramic Period (200 BC–AD 1522). Those phases which represent the focus of this
comparison are highlighted in gray.
lute population estimates were made by establishing a correlation
between sherd densities and residential densities from multiple
sites in the Isthmo-Colombian zone. The best-fit line of this corre-
lation then served as the basis on which to assign residential den-
sities to individual collection lots, based on the sherd density
recovered from each.
Comparing population estimates for the site of El Hatillo using
both of these approaches later revealed that they are not directly
compatible (Berrey, 2014a:74–78; see Drennan and Dai,
2010:459–461). An adjustment factor thus had to be used, which
resulted in increasing the original population estimates made for
the Río Parita valley (Berrey, 2014a:75–77). These original popula-
tion estimates were then further modified to account for those
areas of prehispanic occupation that could not be assigned to
specific periods of time (and that consequently did not factor into
the demographic estimates of this region). Such areas are the result
of the relatively low proportion of sherds that can be identified
using the ceramic chronology of the Río Parita valley (and of other
regions in Central Panama; Fig. 4), which led to areas of occupation
for which no sherds could be assigned chronologically (Berrey,
2014a:74–75). These modifications were made for the sole purpose
of making reliable comparisons between these two regions, and
should otherwise not be taken as a substitute for the original
demographic estimates of the Río Parita valley (see Haller,
2008a:129, Table 5.3).
The demographic estimates for these two regions were used to
assess the scale and structure of local communities. Such commu-
nities were delineated on the basis of regional settlement data
using a technique developed by Peterson and Drennan (2005). This
technique depicts the density of human settlement in a region as a
topographic surface, across which the elevation, or z-value, corre-
sponds to the relative density of human occupation. In an
unsmoothed surface large, nucleated local communities rise to cre-
ate tall, sharp peaks that are clearly discernible from smaller, more
dispersed forms of settlement. The contours of these peaks can be
used to group multiple collection lots into distinct local communi-
ties (Peterson and Drennan, 2005:9–11), though the delineation of
such communities is not always possible. There are a growing
number of regions in the Intermediate Area where settlement
was characterized by relatively large and continuous patches of
dispersed farmsteads (Cuéllar 2009; Drennan, ed., 2006), making
the identification of distinct local communities virtually impossi-
ble. This technique can also discern such continuous and dispersed
local interaction structures (Peterson and Drennan, 2005:23–27;
see Berrey, 2013), along with a range of variability that can exist
between these extremes. A mathematical smoothing of these sur-
faces permits the identification of larger networks of human inter-
action, or what have sometimes been referred to as supra-local
communities (Peterson and Drennan, 2005:11–15).
Assessing the relationship between population and resources
was based on the distribution of settlement and soils of different
agricultural productivity. The soil classifications used here are
those that were used in the original survey of each respective
region (see Haller, 2008a:125–127; Ichon, 1980:12–14). Both of
these classifications resulted in four types of soils within each
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
respective survey zone, which are here ranked from Class I (most
productive) to Class IV (least productive). These soil rankings are
relative, and should not be considered absolute in any way (i.e.
Class I soils are not necessarily twice as productive as Class II soils,
etc.). Nor are the soil rankings of each region directly comparable;
Class II soils in the Río Parita valley are not necessarily as produc-
tive as Class II soils in the Río Tonosí valley. In fact, soil classifica-
tions across Central Panama suggest that soils in the Río Tonosí
valley are generally more productive than those of the Río Parita
valley (ANAM, 2011:37; see Berrey, 2014a:91, , Fig. 4.4). Though
occupants of both regions would have had alternative subsistence
resources readily at their disposal (including riverine, terrestrial,
and, in the case of the Río Parita valley, marine resources), agricul-
ture seems to have become the mainstay of subsistence in Central
Panama by at least AD 250. By this time maize had undergone mor-
phological changes that increased its productivity (Cooke et al.,
2007:578; Piperno, 2006:156; Piperno and Pearsall, 1998), and
bone isotope analysis indicates that, though subsistence continued
to revolve around a wide range of both wild and domestic
resources, maize consumption in Central Panama increased consid-
erably in the first few centuries AD (Norr, 1991, 1995; see Cooke
and Ranere, 1992). It was also at this time that settlement in many
regions began to shift further inland (i.e. away from coastal
resources), including the Río Tonosí and Río Parita valleys (Berrey
2014a; Haller, 2008a; Ichon, 1980; see Isaza and Ilean, 2007).
4. Regional demography and early complex society
development
Prehispanic occupation first emerged in the Río Parita valley by
at least 5000 BC (Haller, 2008a:47–51). It was not until roughly
5000 years later, though, as a sedentary lifeway became wide-
spread across much of Central Panama (ca. 250 BC–AD 200), that
human occupation first spread into the Río Tonosí valley (Ichon,
1980). Despite this much longer history of human occupation,
which included the growth of La Mula-Sarigua (Haller, 2008a;
Hansell, 1988), it does not seem to have been of much consequence
in explaining the heightened levels of inequality that developed in
the Río Parita valley. By AD 250 La Mula-Sarigua had been all but
abandoned (Haller, 2008a), and patterns of regional settlement
and social organization were strikingly similar in both of these
regions. From AD 200 to 500 regional population densities were
relatively low (Fig. 5), and people scattered themselves widely
across the landscape among farmsteads and small hamlets. Social
organization at this time was highly egalitarian (see Briggs,
1989; Menzies, 2009). The majority of people in each region opted
to live among the moderately productive Class II soils, though
some households settled in less productive zones. One such group
of households emerged at the site of El Hatillo in the Río Parita val-
ley (Haller, 2008a; Menzies, 2009), whereas another emerged at
Guaniquito Abajo in the Río Tonosí valley (Berrey, 2014a). This
phase marks the founding of these two settlements, both of which
consisted of roughly 5–10 households during this period of time.
The period from AD 200 to 500 thus seems to have ‘leveled the
playing field’ between these two regions in terms of social
development. Sometime after AD 500, though, their developmental
trajectories began to diverge. Given that marked differences in
social organization had already developed by at least AD 1000, it
is between AD 500 and 1000 that the factors underlying these dif-
ferences must ultimately be sought. In the Río Tonosí valley this
period corresponds to the phase known as La Cañaza (Briggs,
1989; Cooke and Sánchez, 2003; Ichon, 1980), but greater chrono-
logical precision in the Río Parita valley divides this time frame
into multiple periods. For purposes here, it is the Cubitá and Conte
201
Fig. 5. Regional population densities in the Río Tonosí and Río Parita valleys. The
range between minimum and maximum estimates are shown in gray.
phases (AD 550–700 and AD 700–900, respectively) that are the
focus of this comparison (see Fig. 4).
From AD 500 to 1000 both the Río Parita and Río Tonosí valleys
underwent substantial levels of regional population growth. Such
growth, however, was far greater in the Río Tonosí valley (Fig. 5).
As population levels rose households in this region came to dis-
tribute themselves in a dispersed fashion across much of the regio-
nal landscape (Fig. 6), with the vast majority of them living at an
average density of 5–10 people/ha. Though slightly greater degrees
of nucleation occurred at the site of Montevideo (with an esti-
mated density of 10–15 people/ha), here inter-household spacing
would not have been much different than in other parts of the
region.
Montevideo was located among the alluvial spurs that flanked
the southern edge of the Río Tonosí floodplain, across which settle-
ment was relatively continuous across large areas, precluding the
delineation of distinct local communities. Such communities are
slightly easier to identify in the northern part of the survey zone,
though most of them are very small and consist of no more than
just a few households. The one exception to this pattern is the
community of Guaniquito Abajo, which from AD 500 to 1000 grew
to a population of approximately 150 people. The limits of this
community are easily identifiable due to the relatively vacant
patches of landscape that surrounded it, but with an estimated
density of 5–10 people/ha it was just as dispersed as the more con-
tinuous swaths of settlement that existed further south.
The majority of households in the Río Parita valley also lived
among dispersed farmsteads and hamlets, but here a substantial
proportion, or about one-third of the regional population, clustered
themselves into more nucleated settlements. Nowhere was such
nucleation more pronounced than at the community of El Hatillo
(Fig. 7). From AD 550 to 700 this community grew from a small
hamlet of some five to ten households into a relatively large, nucle-
ated village of approximately 450 people. By far the largest local
community in the Río Parita valley, El Hatillo served as a central
place around which a much larger network of supra-local interac-
tion was organized (a role El Hatillo would retain throughout the
remainder of the region’s prehispanic sequence; see Haller, 2008a).
From AD 550 to 700 the population of El Hatillo was distributed
across some 19 ha of landscape, resulting in an average residential
202 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
Fig. 6. Map and unsmoothed surface of regional settlement patterns in the Río Tonosí valley between AD 500 and 1000.
Fig. 7. Map and unsmoothed surface of regional settlement patterns in the Río Parita valley between AD 550 and 700. The patterns illustrated here are also representative of
regional settlement patterns during the subsequent period (AD 700–900).
density of approximately 25 people/ha. During the subsequent
200 years (AD 700–900) this local community became slightly
more compact; although the local population remained relatively
stable (or even declined slightly), the majority of households
(about 80% of them) lived within half of the settlement’s overall
area. This core zone of the community measured roughly 10 ha,
where residential densities were approximately 30 people/ha on
average.
Patterns of local interaction at El Hatillo would thus have been
considerably different than those that existed anywhere in the Río
Tonosí valley. For instance, of the approximately 300 people that
lived in the core zone of El Hatillo from AD 700 to 900, each house-
hold would have lived only 40 m (on average) from its nearest
neighbor, and would not have lived more than 500 m from any
other household in this part of the community (Fig. 8). Local inter-
action in the Río Tonosí valley, however, was structured rather dif-
ferently. At an average residential density of 5–10 people/ha there
would only have been about 25% as many households (or about 75
people) within the same 10 ha of contiguous settlement, each of
which would have resided at an average distance of 75–100 m
from its nearest neighbor. Even at the more densely occupied site
of Montevideo there would only have been about 100 people
within this same area, and it would have required a total of
35 ha of settlement (including Montevideo) to encompass the
same number of households that lived in the core of El Hatillo.
Given the way that settlement was distributed in this part of the
region such an area would have spanned approximately 1.5 km
from one end to the other along its maximum dimension (Fig. 8).
Thus, while the distance that separated households in the Río
Tonosí valley would not have presented a major obstacle to
frequent interaction between neighbors, any sizeable group of
households (such as one the size of El Hatillo’s core) would have
been spread across a relatively large distance. The inconvenience
introduced by that distance would likely have produced a more
attenuated pattern of human interaction.
The local community structure of El Hatillo would thus have
facilitated more intense patters of local interaction than anywhere
in the Río Tonosí valley between AD 500 and 1000. These different
interaction structures may have encouraged households to engage
in different types of activities and inter-household relationships,
and may therefore help explain some of the organizational differ-
ences that can be observed during the Late Ceramic II period.
One such difference relates to the development of craft specializa-
tion. Craft specialization was an important element of inequality at
El Hatillo during the Late Ceramic II period (Menzies and Haller,
2012b), and first began to develop during this period of time
(Menzies, 2009).
4.1. Local interaction and craft specialization
Craft specialization has historically been a relatively difficult
concept to define (Costin, 1991; Hruby and Flad, eds. 2007); it
can be carried out within different contexts, at different scales,
and with different degrees of intensity. It is this latter axis of vari-
ation, however, or what some have referred to as the degree of pro-
ductive differentiation (Drennan and Peterson, 2012:78–79), that
is most relevant for purposes here. On one end of the spectrum this
differentiation can be relatively modest, such as that which charac-
terized the production of shell ornaments along the south coast of
Ecuador (Fig. 1) from AD 800 to 1532 (Martín, 2010; Masucci,
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
Fig. 8. Distribution of 300 people in the most densely occupied parts of the landscape in the Río Parita and Río Tonosí valleys between AD 500 and 1000 (on the left: the core
zone of El Hatillo during the Conte phase (AD 700–900); on the right: Montevideo and surrounding settlement during the La Cañaza phase (AD 500–1000)).
1995). Here households engaged in this activity on a part-time
basis so as to bolster their domestic economy, but nowhere so
intensively as to suggest that it was the principal endeavor of
any given household. The same cannot be said, however, of shell
ornament production at Cerro Juan Díaz, in the Río La Villa valley
of Central Panama (Fig. 2), which lies near the opposite end of
the differentiation spectrum. Between AD 550 and 700 such pro-
duction was carried out in a highly intensive manner, under what
some archaeologists have characterized as workshop-like condi-
tions (Mayo, 2004; Mayo and Cooke, 2005; see Menzies, 2009:38).
While craft specialization was not as intensive at El Hatillo as it
was at Cerro Juan Díaz, it was nonetheless seemingly more inten-
sive than anything that developed in the Río Tonosí valley, where
no evidence for craft specialization has been recovered (see
Berrey, 2014a:58). Though it possible this is due to the lack of
intensive, local scale research in the Río Tonosí valley, preliminary
evidence of specialization at El Hatillo was first identified at the
regional scale (Haller, 2008a). Such specialization was intensive
enough that some specialized households may not have been able
to independently meet all of their subsistence needs, thus leading
to the development of economic interdependence. It is such inter-
dependence that the compact local community structure of El
Hatillo may have helped facilitate.
From AD 700 to 900 elites at El Hatillo were more actively
engaged in the production of shell ornaments and polished stone
axes than other members of the community (Menzies, 2009).
While it is unclear what types of axes elite households were pro-
ducing (either those used for agricultural clearing or smaller vari-
eties used for various craft activities, as was the case later on in the
trajectory; see Menzies and Haller, 2012b), it does seem clear that
they were not producing all of the items they would have required
to carry out basic subsistence tasks. For instance, though elites may
have been the primary producers of certain axes within El Hatillo
(see Menzies, 2009:93–96), they do not seem to have been very
actively engaged in the production of other stone tools, as this
appears to have been concentrated largely among low-status
households (Menzies, 2009:93–94). These items would have been
important for many basic subsistence needs, and would thus have
been required by low- and high-status households alike. Low-
status households may also have been the primary ceramic pro-
ducers within the community (Menzies, 2009:94), and may have
been responsible for certain processing activities that entailed
the use of cutting and scraping tools—tools which do not appear
in elite household contexts (Menzies, 2009:93–94).
These patterns are in some instances plagued by small sample
sizes, but they nonetheless suggest that there was some degree
203
of economic interdependence at El Hatillo between elite and com-
moner households between AD 700 and 900. This interdependence
was likely facilitated by the compact community structure that
existed at El Hatillo, and the relatively frequent patterns of interac-
tion that such a structure would have enabled. To begin with, if
interdependence is to form between households, then there is an
economic advantage of living in close proximity to those with
whom one is interdependent, so as to minimize the cost and bur-
den of travel that is required for the exchange of goods (Drennan
and Haller, 2007:79; Kohler et al., 2007:93). Such proximity also
allows for greater oversight between interdependent households,
which can be advantageous given that interdependence inherently
entails placing a part of one’s livelihood in the hands of another
household.
Not every household at El Hatillo, however, would have
engaged in economic specialization. Within practically any com-
munity in which specialization develops there are bound to exist
ordinary households (e.g. Drennan and Peterson, 2006), which typ-
ically represent the majority of any prehistoric population. Some
such households may have been responsible for providing food
to specialized producers, for which they would have received
important craft items in return. Others simply may not have
engaged much in economic interdependence, and opted to remain
more or less self-sufficient. These households may not have pos-
sessed the characteristics—be it the skill, the reliability, or even
the desire—that are required to engage in, and more importantly
sustain, interdependent relationships. In compact local communi-
ties such as El Hatillo the pool of households that would have been
needed to engage in and sustain any significant level of interde-
pendence would all have lived in close proximity to one another;
amid more dispersed structures of local interaction, however, such
households may have been separated by many hundreds, if not
thousands, of meters. The distance that separated households in
the Río Tonosí valley may thus have discouraged economic inter-
dependence from forming among much more than a small group
of households, which in turn would have discouraged the develop-
ment of any significant level of craft specialization. Though the
desire to engage in interdependence could itself have prompted
households to nucleate, it would seem that there were other fac-
tors in the Río Tonosí valley that, as will be discussed further
below, were more important in determining household location.
Whatever may have discouraged craft specialization from
developing in the Río Tonosí valley, that it was somehow con-
nected to the dispersed structure of local interaction in this region
is supported by data from other early complex society cases. Very
similar patterns have been documented in broader comparisons
204 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
throughout the Intermediate Area (Berrey, 2013; Martín and
Murillo Herrera, 2014), as well as in comparisons of a more global
scale (Drennan and Peterson, 2006, 2008). Among the cases
included in these comparisons levels of productive differentiation
and interdependence were greatest in regions where daily interac-
tion was structured into compact local communities; they were
much less developed among those in which no such communities
existed, and a dispersed structure of local interaction extended
across the entire landscape.
The different structures of local interaction that developed in
these two regions may thus help explain some of the organiza-
tional variability that can be observed between them. While the
dispersed interaction structure of the Río Tonosí valley cannot nec-
essarily explain why no strong inequalities developed in this
region (as will be discussed further below), it does help explain
why one important activity on which inequalities were based in
other parts of Central Panama (craft specialization) did not. Such
specialization seems to have required more intense patterns of
local interaction than what the dispersed settlement structure of
the Río Tonosí valley encouraged, helping set societies in this
region down a different path of social development.
This, of course, does not mean that craft specialization (or
inequalities rooted in specialization) is the inevitable outcome of
nucleated local interaction, but simply that such interaction is an
important part of the equation if any significant level of specializa-
tion or economic differentiation is to develop (Drennan and
Peterson, 2008:383). Specialization and interdependence can
themselves be the catalyst for the nucleation process, but this does
not seem to have been the case in the Río Parita valley. A large
number of households began to nucleate at El Hatillo between
AD 550 and 700, and it was not until the subsequent 200 years
(AD 700–900) that craft specialization began to emerge. This raises
the question as to what might have caused these households to
nucleate in the first place, along with what may have discouraged
households from doing so in the Río Tonosí valley.
4.2. Settlement dispersal and regional demographic pressure
The most likely reason for the dispersed nature of settlement in
the Río Tonosí valley was the desire of households to live directly
on or adjacent to the plots of land that they farmed. Such a dis-
persed, agrarian mode of settlement has been documented ethno-
graphically in a wide range of world regions (e.g. Hunter, 1967;
Netting, 1993; Stone, 1996). It is also argued to have been the
motivation behind settlement dispersal among early complex soci-
eties in in the Tehuacán valley (Drennan and Haller, 2007), and is
widely accepted as the reason for such dispersal among the Classic
Maya (e.g. Drennan, 1988; Netting, 1977; see Lemonnier and
Vannière, 2013; Nondédéo et al., 2013) and other contexts of
low-density urbanism (Fletcher, 2012; Isendahl and Smith, 2013).
That this was also the motivation behind settlement dispersal in
the Río Tonosí valley is plausible given the residential densities
that existed in this region, and the amount of land that this would
have made available for cultivation between neighboring
households.
The amount of land that separates households in dispersed set-
tlement systems can be highly variable. In some cases, such as
among the Kofyar of Nigeria, households can be separated by
hundreds of meters, owing to the multiple hectares of farmland
(3–8 ha on average) that typically surround each dwelling (Stone,
1991, 1992, 1996). In other cases the fields that surround houses
(and thus the distances that separate them) can be somewhat
smaller. Among the Classic Maya of the Río Bec region, for instance,
some household plots were similar in size to those documented
among the Kofyar, but others (the majority, in fact) were notably
smaller, averaging 0.25–1.0 ha in area (Lemonnier and Vannière,
2013). Comparably sized plots have been documented ethnograph-
ically in many tropical regions around the world (Kuman and Nair,
eds., 2006). Though relatively small these plots can still be highly
productive under sustainable forms of intensive cultivation
(Netting, 1993) and make a substantial contribution to household
subsistence.
Based on the residential densities that were estimated for the
Río Tonosí valley, virtually every household in the region would
have had at least 0.5–1.0 ha of land available to it in the landscape
immediately surrounding its residence. Those living among more
isolated farmsteads and hamlets in the northern part of the survey
zone would have had even more land immediately at their dis-
posal, and those living amid the larger swaths of settlement further
south may have had somewhat more as well, given that all of the
settlement identified for this period was not likely contemporane-
ous. Even if it was all contemporaneous, though, ethnographic evi-
dence suggests that families could still have reaped a substantial
portion of their subsistence requirements from the land that
existed between neighboring households (Kuman and Nair, eds.,
2006; Netting, 1993). To the extent that this land fell short of ful-
filling subsistence requirements, households could have engaged
in additional modes of subsistence, such as hunting, fishing, or cul-
tivating additional plots of land at a greater distance from their res-
idences. In this sense agriculture in the Río Tonosí valley may not
have been characterized solely by infield agriculture, but rather by
an infield–outfield system, whereby households intensively culti-
vated the plots of land on which they lived and practiced more
extensive forms of cultivation elsewhere (e.g. Killion, 1990;
Sanders, 1981:362–363). Such a system seems plausible given
the decisions people made about where to live with respect to dif-
ferent soil zones.
If households were to intensively cultivate the plots of land on
which they lived, then it follows that they would opt to live on the
most productive soils available, as these would be the ones best
suited for sustaining intensive cultivation. While the Class I soils
of the Río Tonosí floodplain were the most productive soils in the
region, much of this zone is subject to inundation multiple times
a year, making it relatively undesirable for human habitation. This
would have made Class II soils the most practical soils on which to
live, which is indeed where the majority of households opted to
locate their residences (Fig. 9). Most of these households occupied
the single largest patch of Class II soils available, which accounts
for the disproportionately large amount of settlement in the south-
ern part of the survey zone; this particular patch of soil was home
to roughly 50% of the regional population, despite making up only
17% of the overall regional landscape. Here households would have
been well suited to intensively cultivate the moderately productive
soils on which they lived, and to practice more extensive forms of
cultivation in the nearby alluvial zone, or the vacant patches of
Class II soil that existed between larger swaths of settlement.
The relatively unproductive Class III soils that characterize the
northern part of the survey zone may, similarly, help explain
why settlement in this part of the region was somewhat sparser
and less continuous. Residential densities here suggest that house-
holds were still intensively cultivating the plots of land on which
they lived, but these plots would ultimately have produced lower
yields as a result on the less productive soils on which they were
located. This would have made the outfield component of the
infield–outfield system relatively more important, which may
explain the relatively large areas of unoccupied territory in this
particular part of the region. Though households here would also
have had access to sizeable patches of alluvial soil, the larger, adja-
cent patches of Class III soil would still have been important.
If the dispersed nature of settlement in the Río Tonosí valley
was indeed born from a desire of households to intensively
cultivate the plots of land on which they lived, then this raises
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212 205

Fig. 9. Distribution of settlement among different soil zones in the Río Tonosí valley between AD 500 and 1000.

the question as to what prompted households to pursue such an
intensive form of cultivation in the first place. One important factor
behind this decision may have had to do with the relatively large
number of people that were beginning to occupy the region during
La Cañaza times (AD 500–1000). In the face of rising population
levels and diminishing landscape, households may have felt the
need to lay claim to their own individual plots of land; as the land-
scape filled in and plots became smaller, households would have
had to adopt, in true Boserup-type fashion (Boserup, 1965), more
intensive forms of cultivation. While it is unclear just how much
of a household’s caloric requirements these individual plots may
have provided, it was clearly enough that securing access to such
land was important to subsistence needs. High rates of settlement
continuity were documented for both La Cañaza- and Bijaguales-
phase times (80% and 60%, respectively), which suggests that
households were reluctant to abandon their plots once they had
been established.
That settlement dispersal in the Río Tonosí valley was adopted
so as to pursue intensive cultivation in response to high levels of
regional population growth is also consistent with other archaeo-
logical and ethnographic case studies where similar patterns have
been documented. As Fig. 10 illustrates this mode of settlement
tends to occur most frequently in regions characterized by rela-
tively high regional population densities. Though widespread dis-
persal is not the inevitable outcome of high regional population
levels, it is under these types of demographic conditions that such
dispersal tends to occur (see Netting, 1993:263–265; Stone,
1996:101–104). Environmental conditions must also be met to
pursue such a form of intensive cultivation (Stone and Downum,
1999), which may help explain why this particular agrosystem
was not adopted in regions such as Southern Manabí (Ecuador)
or Chifeng (China), where regional population densities were com-
parable to those of dispersed settlement systems (Fig. 10). As will
be discussed further below, this may also explain why, even if
regional population densities in the Río Parita valley had been
comparable to those of the Río Tonosí valley, settlement dispersal
and intensive household cultivation may not have been a viable
option.
Many households in the Río Parita valley would have had access
to the amount of land that they required in the landscape immedi-
ately surrounding their houses, but the same cannot be said for
those that lived at El Hatillo. Based on the residential densities that
were estimated for this community (see above) each household
would have had approximately 1500–2000 m2 of land available
to it before encroaching on the plots of their neighbors (though
those residing on the periphery of the community would obviously
have had more). Plots of this size fall on the extreme low end of the
spectrum when compared to the sizes of those documented
ethnographically in many tropical zones (see Kuman and Nair,
eds., 2006). Residents at El Hatillo may thus have been able to reap
some subsistence goods from the land that immediately sur-
rounded their houses, but the yield would have been considerably
less than that of households in the Río Tonosí valley—not only were
household plots at El Hatillo much smaller than those in the Río
Tonosí valley, but they were located on some of the least produc-
tive soils in the region.
Despite the abundance of more productive land throughout the
rest of the survey zone, the occupants of El Hatillo opted to live

Fig. 10. Regional population densities from ethnographic cases and a sample of early complex societies around the world (ethnographic cases are marked by an asterisk).
Regions exhibiting dispersed settlement systems are hatched in gray. Dotted lines indicate the average population density of those regions in different parts of the world. Data
from Bandy (2005), Berrey (2014a), Chifeng (2011), Cuéllar (2009), Drennan ed. (2006), Drennan and Dai (2010), Hunter (1967), Kowalewski et al. (2009), Martín (2010),
Murillo Herrera (2011), Peterson et al. (2014), and Stone (1996).
206 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212

Fig. 11. Distribution of settlement among different soil zones in the Río Parita valley between AD 550 and 900. The map on the left corresponds to the Cubitá phase (AD 550–
700).

among the less productive Class III soils (Fig. 11). These soils make Table 1
Annual rainfall (in mm) in the towns of Parita (Río Parita valley)
up less than 10% of the overall regional landscape, but were home
and Tonosí (Río Tonosí valley). Data from INEC (2010).
to approximately 50% of the regional population, roughly half of
which was attributable to the community of El Hatillo. The major- Year Parita Tonosi
ity of household subsistence requirements at El Hatillo would thus 2001 676.5 1152.1
likely have been met by practicing more extensive forms of cultiva- 2002 433.9 1214.0
2003 1183.8 1686.5
tion in the landscape surrounding this local community (see Haller,
2004 945.9 1497.8
2008a:121–127). These more extensive forms of cultivation would 2005 1049.5 1583.4
have required more land per household to meet subsistence 2006 809.6 1580.7
requirements, and would thus have been facilitated by the lower 2007 1291.0 2109.1
regional population densities that existed in this region. El Hatillo, 2008 1508.8 2070.1
2009 1171.9 1508.8
in particular, would have required a large amount of land to sus-
2010 1900.3 2032.3
tain its local inhabitants; as Haller (2008a:171–173) has noted,
Mean 1097.1 1643.5
some households here would have had to maintain fields located
SD 420.3 337.4
at least 3 km away from their residences. CV 0.383 0.205
Farming practices for the occupants of El Hatillo would thus
have been somewhat different than for those that lived in the Río
Tonosí valley, entailing different patterns of land use and, poten-
tially, household labor organization. These differences may have of the Río Parita survey zone (see Haller, 2008a:21, Fig. 2.1), that
existed since the very beginning of El Hatillo’s occupation, and of Tonosí is located roughly 10 km closer to the coast from where
may have helped prompt households to nucleate there in the first the Río Tonosí survey was conducted. It thus receives less rainfall
place. than the regional survey zone (ANAM, 2011:31; see Berrey,
2014a:90, Fig. 4.1), but nonetheless provides a reliable indication
of inter-annual variability in the Río Tonosí valley.
4.3. Settlement nucleation and agricultural risk
The degree of inter-annual variation in rainfall, or the coeffi-
cient of variation (CV), in the Río Parita valley is almost double that
Although occupants of the Río Parita valley may not have been
of the Río Tonosí valley (Table 1). The CV for the Río Parita valley, in
faced with the same sorts of demographic pressures as those of the
fact, is even higher (in some cases substantially) than CV’s docu-
Río Tonosí valley, this does not mean that they were not presented
mented in the Valley of Oaxaca (Table 2)—where inter-annual vari-
with subsistence challenges of their own (e.g. Haller, 2008a:23).
ation in rainfall and other environmental conditions are argued to
The environment of the Río Parita survey zone is considerably drier
have posed a significant risk to subsistence pursuits, shaping the
than that of the Río Tonosí valley, and is in fact one of the driest
development of early complex societies (Sanders and Webster,
areas in all of Panama (ANAM, 2011); it receives much less annual
1978). Though the Río Parita valley receives somewhat more rain-
rainfall than the Río Tonosí valley (1000–1300 mm and 2100–
fall than does the Valley of Oaxaca, which receives about 700 mm
2700 mm, respectively) and experiences slightly higher rates of
annually (Kowalewski et al., 1989:8), it is clear that in some years
evapotranspiration (ANAM, 2011:31; see Berrey, 2014a:90–91,
it receives well below the Oaxaca average (Table 1). The CV for the
Figs. 42 and 43). The soils of the Río Parita survey zone also have
Río Tonosí valley is also in the range of those documented for Oax-
greater limitations with respect to cultivation than do those of
aca, but such variation would not have been as detrimental given
the Río Tonosí valley (ANAM, 2011:37; see Berrey, 2014a:91,
Fig. 4.4), increasing the restraints on agricultural pursuits.
In addition to being drier and having poorer soils than the Río Table 2
Tonosí valley, environmental conditions in the Río Parita valley Average annual rainfall (in mm) and coefficients of variation (CV) for three separate
locales in the Valley of Oaxaca. Data from Kowalewski (1980:154).
are also more variable from year to year. This further increases
the risks associated with agriculture production; the greater the Locale Range Mean SD CV
degree of inter-annual variability, the greater the degree of uncer- Oaxaca City 1941–1962a 597.4 91.9 0.154
tainty that households will meet all of their subsistence needs. Tlacolula 1926–1968b 529.7 160.7 0.303
Such variability is illustrated in Table 1, which presents the total Tlalixtac de Cabrera 1968–1977 788.6 137.0 0.174

annual rainfall for each year of a 10-year period (2001–2010) in a

Data missing for eight years of range.
both regions. Whereas the town of Parita falls within the limits b
Data missing for five years of range.
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
the range within which rainfall varies. The driest year of this
10-year period, for instance, still received approximately
1100 mm of rain, and this value would have been even higher
within the limits of the regional survey zone.
It is worth noting that the relatively high CV’s observed for the
Río Parita and Río Tonosí valleys may in part be due to the rela-
tively small sample size from which they were derived. These
CV’s are calculated by dividing the standard deviation of rainfall
in each region by its respective mean, and smaller samples typi-
cally result in larger standard deviations. Despite this potential
problem, however, there are data to suggest that these CV’s can
still be taken as reliable values. These data come from other
regions of Panama for which annual rainfall data are available for
longer periods of time, but whose precipitation patterns could still
satisfactorily be characterized on the basis of a 10-year period.
Table 3 presents the average annual rainfall and CV’s across a
40-year period (1908–1947) for two separate sites in the Panama
Canal Zone (Cristóbal and Balboa Heights). This table also presents
what the mean and CV would be for each of these two sites had
they been calculated for any one of seven distinct 10-year periods
within the larger 40-year time frame. Despite the smaller samples
from which these means and CV’s were derived, they would
nonetheless provide a relatively accurate representation of the
average rainfall and inter-annual variability for these two areas.
Using any of these calculations one would estimate that the mean
annual rainfall for Cristóbal was approximately twice as high as
that of Balboa Heights, and that the CV’s for each site were roughly
similar (with the possible exception of the first decade, in which
the CV for Cristóbal was somewhat higher).
Taken together, then, these data indicate that the Río Parita val-
ley is notably drier, has poorer soils, and experiences greater levels
of inter-annual variability in rainfall than the Río Tonosí valley.
While paleoclimatic data could help confirm these trends for the
prehispanic era, they nonetheless suggest that there would have
been greater levels of agricultural risk for the prehispanic occu-
pants of the Río Parita valley. This risk may not have been as great
as in other parts of the world where it is argued to have played an
important role in early complex society development, such as
highland Mexico (Sanders and Webster, 1978; Spencer, 1993) or
the U.S. southwest (Nicholas and Feinman, 1989; Tainter and
Tainter, eds., 1992), but it may well have been enough to prompt
occupants of the Río Parita valley to engage in certain forms of
risk-minimizing behavior. In western Hawai’i early complex soci-
eties would have had access to a comparably diverse range of
resources and received comparable levels of rainfall as those of
the Río Parita valley, yet they nonetheless adopted various forms
of risk minimization, even in the most productive parts of the
region (Allen, 2004).
There are many forms of risk-minimization that inhabitants of
the Río Parita valley could have adopted so as to cope with the
threat of agricultural risk. In some parts of the world this was
accomplished through infrastructure such as dams and large-
Table 3
Average annual rainfall (in mm) and coefficients of variaiton (CV) for Cristóbal and Balboa Heights. Data from NOAA (2002a,b).
Range Cristóbal Balboa heights
Mean SD CV Mean
1908–1947 3358.5 521.5 0.155 1744.9
1908–1917 3395.3 594.9 0.175 1774.9
1913–1922 3005.3 434.0 0.144 1650.8
1918–1927 3188.6 557.5 0.175 1651.5
1923–1932 3377.8 466.8 0.138 1757.4
1928–1937 3418.7 553.7 0.162 1850.8
1933–1942 3654.5 500.2 0.137 1838.4
1938–1947 3431.2 404.8 0.118 1702.4
207
scale irrigation canals (e.g. Spencer, 1993), but there is no evidence
for such features among the regions of Central Panama. Other com-
mon risk-minimizing strategies include storage, food sharing, and
distributing parcels of farmland across different environmental
zones (Hegmon, 1989; see Halstead and O’Shea, eds., 1989), each
of which would have been a viable option in the Río Parita valley.
Though storage was undoubtedly important to the occupants of
this region, it is the latter two strategies listed that are most rele-
vant for purposes here; these strategies, along with inter-
household cooperation in subsistence activities, may have helped
spur the nucleation process that occurred at El Hatillo.
To the extent that households engaged in food sharing and
cooperation in subsistence activities on a relatively frequent basis,
it would have been advantageous, or even necessary, for them to
live near those households on whom they regularly relied. If
households wished to engage in these strategies with families out-
side of their kinship group, then this could help prompt the forma-
tion of large local communities. Food sharing and other modes of
reciprocal exchange are argued to create highly resilient social net-
works (Reynolds et al., 2003), and this may help explain the high
degree of resiliency exhibited by El Hatillo; once El Hatillo emerged
as a large local village between AD 550 and 700, it persisted as such
without interruption for at least the next 700 years (Haller, 2008a).
The nucleation process at El Hatillo may also have been facili-
tated by the way that farmland was distributed throughout the
region. If such land was distributed among different environmental
zones, so as to guard against shortcomings or catastrophe within
any single one, then this would have discouraged households from
living directly on or adjacent to the plots of land that they farmed.
Such a mode of land distribution has been documented among the
Hopi of the U.S. southwest (Hegmon, 1989) and the Zapotec of
Oaxaca during the early 20th century (Schmieder, 1930). As
Flannery (2002:424) has summarized:
Zapotec families collaborated in clearing land by means of large
work groups, then distributed the land among the families who
participated. Because big work gangs moved from place to
place, a family could wind up owning parcels of land scattered
through several environments. This process not only spread risk
among many families, it also minimized the chances that a local
environmental disaster would damage all of a family’s plant-
ings. As Schmieder points out, it also promoted the growth of
large, permanent villages; since there was no point in moving
one’s house to fields that were scattered over so large an area,
families continued to live within the larger cooperating group.
Households in the Río Tonosí valley would also likely have
cooperated with one another in certain subsistence activities, but
such cooperation may not have been carried out with the same fre-
quency as at El Hatillo. For instance, among the Kofyar of Nigeria,
who pursue intensive forms of cultivation organized at the house-
hold level and exhibit a dispersed pattern of settlement like that
envisioned for the Río Tonosí valley, households regularly
Ratio of means Ratio of CV’s
SD CV (Cristóbal:Balboa Heights) (Cristóbal:Balboa Heights)
243.3 0.139 1.92 1.12
187.4 0.106 1.91 1.65
169.6 0.103 1.82 1.40
229.1 0.139 1.93 1.26
267.2 0.152 1.92 0.91
279.5 0.151 1.85 1.07
278.0 0.151 1.99 0.91
256.0 0.150 2.02 0.79
208 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
cooperate in agricultural activities, particularly during the growing
season (Stone, 1991, 1992, 1996). The amount of time spent work-
ing on another household’s farm, however, is relatively small, and
typically constitutes no more than 20% of a household’s total agri-
cultural labor, with the rest being spent on its own plot of land
(Stone, 1991:347, Table 1). In those contexts where households
cooperated on a more frequent basis, it would have been advanta-
geous to live near those households with whom they regularly
cooperated.
A combination of risk-minimizing strategies may thus have
contributed to the nucleation process that occurred at El Hatillo
between AD 550 and 700, including food sharing, cooperation in
subsistence activities, and distributing parcels of farmland among
different environmental zones. These strategies may well have
been at work in the period preceding El Hatillo’s growth (AD
250–550), but it was not until the subsequent 150 years, as regio-
nal population levels rose, that opportunities for food sharing and
inter-household cooperation outside of one’s own kinship group
became a practical possibility. These new possibilities for interac-
tion led to the growth of new and larger local communities, and
may have led to the development of new forms of social organiza-
tion as well.
4.4. Growth, interaction, and social inequality
Though the nucleation process at El Hatillo may have initially
been sparked by factors relating to subsistence and risk-
minimization, these may not have been the only factors integrating
this community during its initial period of demographic growth
(AD 550–700). It was during this time that social inequalities first
emerged at El Hatillo (Menzies, 2009), and these may also have
come to play an important integrative role. Though inequalities
were only modestly manifest during this initial period, they
nonetheless suggest that there was differential access to certain
socially valued goods. Such differential access became even more
pronounced during the subsequent 200 years (AD 700–900;
Menzies, 2009), and is a clear antecedent to the more marked
levels of inequality that developed during the remainder of the
Late Ceramic II period.
Social inequality at El Hatillo could have arisen from a variety of
factors. One possibility is that it arose in response to agricultural
risk, and the activities that households engaged in so as to buffer
against it (e.g. Halstead and O’Shea, 1982; Spencer, 1993). Inequal-
ities could have arisen, for instance, by particularly large, hard-
working, or otherwise productive households amassing
significant amounts of surplus, thus allowing them to gain leverage
and/or prestige within their respective food-sharing network and
throughout the larger community (e.g. Halstead and O’Shea,
1982; see Kohler et al., 2007:94). They may also have arisen as a
result of increased cooperation in subsistence activities, and the
need for some individuals to coordinate that cooperation (Kohler
et al., 2012).
It is also possible that inequality emerged at El Hatillo as a
result of the dramatic levels of community growth that were
occurring at this time (see above). This growth would have
resulted in greater levels of inter-household conflict, and thus the
need for individuals to mediate such disputes (e.g. Johnson, 1982).
Providing conflict mediation may have afforded certain households
a higher status within the community (Rosenberg, 2009; see
Sahlins, 1958), leading to the modest manifestation of inequality
that is apparent at this time. That these inequalities emerged as
a result of inter-household conflict brought on by El Hatillo’s rapid
growth is consistent with the fact that it was during this time that
the community’s population surpassed an important threshold
that multiple scholars have argued can prompt increased levels
of scalar stress (ca. 150 people; see Kosse, 2001; Feinman,
2011:41, Table 3). It is also consistent with ethnohistoric observa-
tions from Central Panama which document conflict mediation as
an important role of chiefly authority (e.g. Andagoya, 1865:13).
Whatever may have sparked the onset of inequality at El Hatillo
between AD 550 and 700, it is clear that it was successful in spur-
ring a long trajectory of hierarchical development. Not only did
inequality continue to develop here for at least the next 700 years,
but it did so in what seems to have been the same group of house-
holds throughout this entire period of time (see Menzies, 2009).
The forces that initiated such perpetual inequality may have been
quite varied, but some of them seem to have been connected to the
nucleation process that was occurring at El Hatillo, perhaps via
conflict that resulted from local community growth, or risk-
minimizing behavior that both sparked and was facilitated by set-
tlement nucleation.
However, though the processes and activities associated with
settlement nucleation may help explain the rise of inequality at
El Hatillo, the absence of such nucleation does not necessarily
explain why no strong inequalities developed in the Río Tonosí val-
ley. In both the Alto Magdalena and the Quijos region a dispersed
pattern of settlement extended across virtually the entire land-
scape (Cuéllar, 2009; Drennan, ed., 2006; see Berrey, 2013), yet
notable levels of inequality have been documented for each of
these two regions. In the Alto Magdalena during the Regional Clas-
sic period (AD 1–900) they were based largely in ritual activity,
while in the Quijos region both ritual and interregional exchange
helped support clearly marked status distinctions. Such exchange,
however, does not seem to have been based on the movement of
economic goods. As Cuéllar (2009:171) has suggested, ‘‘transac-
tions of a non-commercial nature were at the core of Quijos chiefly
dynamics, and . . .marriage alliances were the driving force behind
the systems of regional and interregional interaction of which the
Quijos chiefdoms were a part”.
The dispersed nature of local interaction in the Río Tonosí valley
therefore cannot explain why no strong level of social inequality
ever developed in this region. It does, however, help explain why
some of the potential factors that helped spark inequality in the
Río Parita valley, along with some of the principal activities that
may have helped sustain it, did not develop, such as conflict medi-
ation brought on by scalar stress, the coordination of cooperative
subsistence activities, and craft specialization. In the Río Partia val-
ley these would all have been facilitated by the compact local com-
munity structure that existed at El Hatillo, and in some cases by
the risker environmental conditions that existed in this region.
Thus, even if inequalities were to have developed in the Río Tonosí
valley, they would likely have been based on different sorts of prin-
ciples than what was common to other parts of Central Panama.
4.5. Subsequent developments
With the rise of El Hatillo the stage was now set for households
to engage in the frequent patterns of interaction that traditionally
facilitate the development of inequality and other complex forms
of social organization. This inequality continued to develop for at
least the next 400 years (from AD 900 to 1300), culminating in
the relatively elaborate graves and stronger levels of household dif-
ferentiation that are evident during the Late Ceramic II period. Dur-
ing this period El Hatillo remained the largest local community in
the Río Parita valley, and continued to serve as an important cen-
tral place around which supra-local interaction was organized
(Fig. 12; see Haller, 2008a). El Hatillo’s population did not change
much during this period of time, nor did the larger regional popu-
lation (Fig. 5; see Haller, 2008a).
It was also during this time that warfare seems to have emerged
as an important element of social inequality in the Río Parita val-
ley. Like the nucleation process warfare may have arisen (at least
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
Fig. 12. Smoothed surfaces of regional settlement in the Río Parita (left) and Río Tonosí (right) valleys (following Peterson and Drennan, 2005). The surface on the left
illustrates the supra-local community centered on El Hatillo during the Parita phase (AD 1100–1300); that on the right illustrates the absence of such a community around
Guaniquito Abajo during the Bijaguales phase (AD 1000–1522).
in part) in response to the relatively high degree of environmental
risk that existed in this region. Ethnohistoric documents indicate
that defending the resources within one’s territory, or attempting
to acquire those from another, were important motivations for
warfare during the 16th century (Linares, 1977:73–74; see
Helms, 1979:33, 1994:57). So too, however, was the prestige that
was gained from success in battle. Not only was such success an
important pathway for status advancement (Helms, 1979:13,
31–37; de Oviedo, 1959:28–29), but it is argued to have been a
critical part of sustaining the ideology on which chiefly power
was based (Helms, 1979; see Earle, 2001a). In this sense warfare
in the Río Parita valley may have been spurred more by sociopolit-
ical factors than by purely economic ones during the Late Ceramic
II period.
In contrast to the Río Parita valley, the regional population of
the Río Tonosí valley continued to grow from AD 1000 to 1522
(Fig. 5). As population levels rose settlement remained highly dis-
persed throughout the region, though notable levels of nucleation
emerged at Guaniquito Abajo (Berrey, 2014a:55). While Guaniq-
uito Abajo likely served as a focal point around which supra-local
interaction was organized (given the numerous small earthen
mounds that were constructed there, which are likely indicative
of ritual functions; see Ichon, 1980), such interaction was struc-
tured rather differently than in the Río Parita valley. Not only
would this interaction have been organized around different sorts
of principles, but these principles do not seem to have had the
same centripetal effect on settlement that those at El Hatillo had.
Unlike the very centralized and clearly discernable supra-local
community that developed around El Hatillo, no such community
can be identified in the Río Tonosí valley (Fig. 12). Instead, con-
cerns of having immediate access to cultivable land seem to have
remained the primary factor behind settlement location, as house-
holds continued to concentrate among the moderately productive
Class II soils that flank the Río Tonosí floodplain (Berrey, 2014a).
The very decentralized nature of supra-local interaction in this
region may also have had to do with the relatively undeveloped
nature of social inequality; lacking any significant degree of hierar-
chical organization, this region would have been without one of
the most effective known mechanisms for large-scale integration.
5. Summary and conclusions
Understanding the development of social inequality has long
been and remains a major objective of early complex society
research (e.g. Price and Feinman, eds., 1995, 2010; Flannery and
Marcus, 2012). Such research has revealed a great deal of variabil-
ity in the way early inequalities developed, and attempts to iden-
tify the numerous factors that underlie such variability. Much
headway has been made in this regard through global-scale com-
parisons of early complex society development (e.g. Drennan
209
et al., 2010; Earle, 1997; Scarborough and Burnside, 2010), but
there is also a great deal to be learned by comparing trajectories
across a much smaller scale. To the extent that this scale encom-
passes only culturally related and interacting groups, it allows us
to home in more closely on the forces of social change and the fac-
tors the lead to organizational variation. Central Pacific Panama
represents one such context, which, like other parts of the Interme-
diate Area, exhibits considerable variation in the degree of inequal-
ity that developed in different regions. While the variability that
can be observed here is well recognized in the archaeological liter-
ature, it is not often the explicit focus of research among
archaeologists.
In Central Panama this observed variation in hierarchical devel-
opment is based largely on differences in mortuary data. As noted
above, there is good reason to believe that these differences are
indeed reflective of different levels of inequality, and this is sup-
ported by regional settlement data from the Río Tonosí valley. Even
if strong inequalities had developed in the Río Tonosí valley, but
were not expressed through conspicuous displays of status on
the part of elite individuals, as might be expected among more cor-
porate (Blanton et al. 1996) or group-oriented (Renfrew 1974)
modes of social hierarchy, then these inequalities should be appar-
ent in alternative lines of evidence. One common manifestation of
inequality in such contexts is differential access to key economic
resources (e.g. Earle, 2001b). Settlement data from the Río Tonosí
valley, however, suggest that no differential access ever existed.
There was no concentration of important goods or craft activities
anywhere in the region, and there was no apparent control over
any productive resources (Berrey, 2014a). These data thus suggest
that the mortuary data from the Río Tonosí valley are representa-
tive of relatively weak social inequalities, as much of the classic
mortuary literature would suggest (Binford, 1971; Saxe, 1970).
What, then, may have led to these variable levels of social
inequality among the early complex societies of Central Pacific
Panama? Regional settlement and environmental data from the
Río Parita and Río Tonosí valleys suggest that some familiar factors
may have been at play, though in some instances they resulted in
somewhat different outcomes than much of the early complex
society literature would lead one to expect. Regional population
growth was one such factor. Though both the Río Parita and Río
Tonosí valleys experienced notable levels of regional population
growth between AD 500 and 1000, growth in the latter region
was more substantial. Much of the classic early complex society lit-
erature suggests this should have resulted in greater levels of
inequality in the Río Tonosí valley (e.g. Carneiro, 1970), particularly
given the rather circumscribed nature of this environment (see
Berrey, 2014a:22), but here regional population growth may actu-
ally have discouraged inequalities from forming (or at least the
activities that helped underwrite inequalities in other parts of Cen-
tral Panama). Rather than adopt new, hierarchical forms of social
210 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
organization to help cope with high regional population densities,
households in the Río Tonosí valley opted to deal with such pres-
sure at the household level by laying claim to their own plots of
land. This decision resulted in a dispersed settlement pattern,
which did not easily facilitate frequent patterns of local interaction
or the sorts of activities and social relationships that traditionally
arise from it.
Differential levels of agricultural risk may have been another
important factor in explaining this variability. Greater levels of risk
in the Río Parita valley may have prompted households to engage
in significantly greater levels of risk-minimizing behavior, which
may have helped spark the onset of social inequality by requiring
some individuals to coordinate collective activities (Kohler et al.,
2012). This behavior may also have helped trigger substantial
degrees of settlement nucleation, which in turn may have been
another factor in creating inequalities in the Río Parita valley. As
nucleation progressed and El Hatillo continued to grow some
households may have increased their status by mediating conflicts
arising from scalar stress. As time went on nucleation would have
facilitated some of the primary activities in which inequality was
rooted, while risk would have provided a valid reason for elites
to initiate conflict with other polities, thorough which they could
advance their social prestige.
Ultimately, then, it may have been a combination of different
factors, including different levels of regional population growth,
different levels of agricultural risk, and the different structures of
local interaction that these factors helped create, that gave way
to the different forms of social organization and levels of inequality
that developed in the Río Tonosí and Río Parita valleys. Though
these factors have emerged as some of the most plausible explana-
tions for such variability by way of comparing regional settlement
and environmental data, they are not offered as the sole explana-
tory factors. Rather, they represent specific points of departure
for subsequent research aimed at different scales and/or different
types of analyses. Research aimed at comparable objectives
throughout the Intermediate Area would provide valuable insight
into the highly variable pathways of early complex society devel-
opment in this particular part of the world, and into the variability
that can be observed at a more global scale.
Acknowledgments
Fieldwork in the Río Tonosí valley was funded by a National
Science Foundation Dissertation Improvement Grant (#1048453).
Regional settlement data for the Río Parita valley was made avail-
able via the Comparative Archaeology Database at the University of
Pittsburgh. Much thanks to Dick Drennan and Christian Peterson
for providing helpful comments on previous drafts of this article,
and to Mike Haller, Marc Bermann, and Olivier de Montmollin,
who provided helpful comments on the dissertation from which
this article stems.
References
Allen, Melinda S., 2004. Bet-hedging strategies, agricultural change, and
unpredictable environments: historical development of dryland agriculture in
Kona, Hawaii. J. Anthropol. Archaeol. 23, 196–224.
Andagoya, Pascual de, 1865. Narrative of the Proceedings of Pedrarias Davila in the
Provinces of Tierra Firme or Castilla del Oro, and the Discovery of the South Sea
and the Coasts of Peru and Nicaragua. Translated by Clements Markham, R. The
Hakluyt Society, London.
Autoridad Nacional del Ambiente (ANAM), 2011. Atlas Ambiental de la República de
Panamá (Primera Versión) (Electronic document, last accessed 08.12.14).
<http://www.anam.gob.pa/images/stories/AtlasAmbiental/index.html>.
Bandy, Matthew S., 2005. New world settlement evidence for a two-stage neolithic
demographic transition. Curr. Anthropol. 46 (S5), S109–S115.
Beck, Robin A., 2003. Consolidation and hierarchy: chiefdom variability in the
Mississippian southeast. Am. Antiq. 68 (4), 641–661.
Berndt, Ronald M., 1964. Warfare in the New Guinea Highlands. Am. Anthropol. 66
(4), 183–203.
Berrey, C. Adam, 2013. Interaction structures and the development of early complex
society in Southern Central America and Northern South America. In: Palumbo,
Scott D., Boada Rivas, Ana María, Locascio, William A., Menzies, Adam C.J. (Eds.),
Multiscalar Approaches to Studying Social Organization and Change in the
Isthmo-Colombian Area. University of Pittsburgh Center for Comparative
Archaeology, Pittsburgh, pp. 1–14.
Berrey, C. Adam, 2014a. Organization and Growth among Early Complex Societies in
Central Pacific Panama. Unpublished Ph.D. dissertation, Department of
Anthropology, University of Pittsburgh, Pittsburgh.
Berrey, C. Adam, 2014b. Río Tonosí Valley Settlement Dataset. Comparative
Archaeology Database, University of Pittsburgh. <http://www.cadb.pitt.edu>.
Biese, Leo P., 1967. The Gold of Parita. Archaeology 20 (5), 202–208.
Binford, Lewis R., 1971. Mortuary practices: their study and their potential.
Memoirs Soc. Am. Archaeol. 25, 6–29.
Blanton, Richard E., Feinman, Gary M., Kowalewski, Stephen A., Peregrine, Peter N.,
1996. A dual-processual theory for the evolution of mesoamerican civilization.
Curr. Anthropol. 37 (1), 1–14.
Boserup, Ester, 1965. The Conditions of Agricultural Growth: The Economics of
Agrarian Change Under Population Pressure. Aldine, New York.
Bray, Warwick, 1992. Sitio Conte metalwork in its pan-American context. In:
Hearne, Pamela, Sharer, Robert J. (Eds.), River of Gold: Precolumbian Treasures
from Sitio Conte, University Museum of Archaeology and Anthropology.
University of Pennsylvania, Philadelphia, pp. 33–46.
Briggs, Peter S., 1989. Art, Death, and Social Order: The Mortuary Arts of Pre-
Conquest Central Panama. BAR International Series 550. British Archaeological
Reports, Oxford.
Bull, Thelma H., 1965. Report on archaeological investigations, Azuero peninsula,
province of Herrera, republic of panama: a preliminary report. Panama
Archaeol. 6, 30–64.
Carneiro, Robert L., 1970. A theory of the origin of the state. Science 169, 733–738.
Chifeng International Collaborative Archaeological Research Project, 2011.
Settlement Patterns in the Chifeng Region. University of Pittsburgh Center for
Comparative Archaeology, Pittsburgh.
Cooke, Richard, 2004. Rich, poor, shaman, child: animals, rank, and status in the
‘Gran Coclé’ culture area of pre-Columbian Panama. In: O’Day, Sharyn Jones,
Van Neer, Wim, Ervynck, Anton (Eds.), Behaviour Behind Bones. The
Zooarchaeology of Ritual, Religion, Status, and Identify. Oxbow Books, Oxford,
pp. 271–284.
Cooke, Richard, 2005. Prehistory of native Americans on the central american land
bridge: colonization, dispersal, and divergence. J. Archaeol. Res. 13 (2), 129–
187.
Cooke, Richard, Ranere, Anthony J., 1992. The origin of wealth and hierarchy in the
central region of Panama (12,000–2000 BP), with observations on its relevance
to the history and phylogeny of chibchan-speaking polities in Panama and
elsewhere. In: Lange, Frederick W. (Ed.), Wealth and Hierarchy in the
Intermediate Area. Dumbarton Oaks, Washington, D.C, pp. 243–316.
Cooke, Richard G., Sánchez, Luis Alberto, 2003. Alain Ichon in Panama (1967-1970):
A Reappraisal of the Tonosí Research Project in the Light of New Research. In:
Arnauld, M. Charlotte, Breton, Alain, Fauvet-Berthelot, Marie-France, Valdés,
Juan Antonio (Eds.), Misceláneas.. en honor a Alain Ichon. Centro Francés de
Estudios Mexicanos y Centroamericanos and Asociación Tikal, Guatemala, pp.
13–26.
Cooke, Richard, Isaza, Ilean, Griggs, John, Desjardins, Benoit, Sánchez, Luís
Alberto, 2003. Who crafted, exchanged, and displayed gold in pre-Columbian
Panama? In: Quilter, Jeffrey, Hoopes, John W. (Eds.), Gold and Power in
Ancient Costa Rica, Panama, and Colombia. Dumbarton Oaks, Washington, D.
C, pp. 91–158.
Cooke, Richard, Jiménez, Máximo, Ranere, Anthony, 2007. Influencias humanas
sobre la vegetación y fauna de vertebrados de Panamá: Actualización de datos
arqueozoológicos y su relación con el paisaje antrópico durante la época
precolombiana. In: Leigh, Egbert G., Jackson, Herre, Jr., Herre, E. Allen, Jackson,
Jeremy B.C., Santos-Granero, Fernando (Eds.), Ecología y Evolución en los
Trópicos. Smithsonian Tropical Research Institute (STRI), Panama City, pp. 563–
593.
Cooke, Richard, Ranere, Anthony, Pearson, George, Dickau, Ruth, 2013. Radiocarbon
chronology of early human settlement on the Isthmus of Panama (13,000–7000
BP) with comments on cultural affinities, environments, subsistence, and
technological change. Quatern. Int. 301, 3–22.
Costin, Cathy Lynne, 1991. Craft specialization: issues in defining, documenting, and
explaining the organization of craft production. Archaeol. Method Theory 3, 1–56.
Cuéllar, Andrea M., 2009. The Quijos Chiefdoms: Social Change and Agriculture in
the Eastern Andes of Ecuador. University of Pittsburgh Memoirs in Latin
American Achaeology, No. 20. University of Pittsburgh Latin American
Archaeology Publications, Pittsburgh.
de Las Casas, Bartolome, 1986. Historia de las Indias. Biblioteca Ayacucho, Caracas.
de Oviedo y Valdés, Gonzalo Fernández, 1959. Natural History of the West Indies.
Translated and edited by Stoudemire, Sterling A. University of North Carolina
Press, Chapel Hill.
Drennan, Robert D., 1988. Household location and compact versus dispersed
settlement in prehispanic Mesoamerica. In: Wilk, Richard R., Ashmore, Wendy.
(Eds.), Household and Community in the Mesoamerican Past. University of New
Mexico Press, Albuquerque, pp. 273–293.
Drennan, Robert D., 1995. Chiefdoms in Northern South America. J. World
Prehistory 9 (3), 301–340.
 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
Drennan, Robert D., 2000. Las Sociedades Prehispánicas del Alto Magdalena.
Instituto Colombiano de Antropología e Historia, Bogotá.
Drennan, Robert D. (Ed.), 2006. Prehispanic Chiefdoms in the Valle de la Plata,
Volume 5: Regional Settlement Patterns. University of Pittsburgh Memoirs in
Latin American Archaeology, No. 16. University of Pittsburgh Latin American
Archaeology Publications, Pittsburgh.
Drennan, Robert D., Dai, Xiangming, 2010. Chiefdoms and states in the Yuncheng
Basin and the Chifeng region: a comparative analysis of settlement systems in
North China. J. Anthropol. Archaeol. 29 (4), 455–468.
Drennan, Robert D., Haller, Mikael J., 2007. The local village community and the
larger political economy: formative and classic interaction patterns in the
Tehuacán valley compared to the valley of Oaxaca and the basin of Mexico. In:
Scarborough, Vernon L., Clark, John E. (Eds.), The Political Economy of Ancient
Mesoamerica: Transformations during the Formative and Classic Periods.
University of New Mexico Press, Albuquerque, pp. 65–81.
Drennan, Robert D., Peterson, Christian E., 2006. Patterned variation in prehistoric
chiefdoms. Proc. Natl. Acad. Sci. 103 (11), 3960–3967.
Drennan, Robert D., Peterson, Christian E., 2008. Centralized communities,
population, and social complexity after sedentarization. In: Bocquet-Appel,
Jean-Pierre, Bar-Yosef, Ofer (Eds.), The Neolithic Demographic Transition and Its
Consequences. Springer, New York, pp. 359–386.
Drennan, Robert D., Peterson, Christian E., 2012. Challenges for comparative study
of early complex societies. In: Smith, Michael E. (Ed.), The Comparative
Archaeology of Complex Societies. Cambridge University Press, Cambridge,
pp. 62–87.
Drennan, Robert D., Peterson, Christian E., Fox, Jake R., 2010. Degrees and kinds of
inequality. In: Price, T. Douglas, Feinman, Gary M. (Eds.), Pathways to Power:
New Perspectives on the Emergence of Social Inequality. Springer, New York,
pp. 45–76.
Earle, Timothy K., 1997. How Chiefs Come to Power: the Political Economy in
Prehistory. Stanford University Press, Stanford.
Earle, Timothy K., 2001a. Institutionalization of chiefdoms: why landscapes are
built. In: Haas, Jonathan (Ed.), From Leaders to Rulers. Kluwer Academic Press/
Plenum Publishers, New York, pp. 105–124.
Earle, Timothy K., 2001b. Economic support of Chaco Canyon Society. Am. Antiq. 66
(1), 26–35.
Feinman, Gary M., 2011. Size, complexity, and organizational variation: a
comparative approach. Cross-Cult. Res. 45 (1), 37–58.
Flannery, Kent V., 2002. The origins of the village revisited: from nuclear to
extended households. Am. Antiq. 67 (3), 417–433.
Flannery, Kent V., Marcus, Joyce (Eds.), 1983. The Cloud People: Divergent Evolution
of the Zapotec and Mixtec Civilizations. Academic Press, New York.
Flannery, Kent V., Marcus, Joyce, 2012. The Creation of Inequality: How Our
Prehistoric Ancestors Set the Stage for Monarchy, Slavery, and Empire. Harvard
University Press, Cambridge, MA.
Fletcher, Roland, 2012. Low-density, agrarian-based urbanism: scale, power, and
ecology. In: Smith, Michael E. (Ed.), The Comparative Archaeology of Complex
Societies. Cambridge University Press, Cambridge, pp. 285–320.
Haller, Mikael J., 2008a. Asiento Viejo and the Development of the Río Parita
Chiefdom, Panama. University of Pittsburgh Memoirs in Latin American
Archaeology, No. 19. University of Pittsburgh Latin American Archaeology
Publications, Pittsburgh.
Haller, Mikael J., 2008b. Asiento Viejo Dataset. Comparative Archaeology Database,
University of Pittsburgh. <http://www.cadb.pitt.edu>.
Halstead, Paul, O’Shea, John, 1982. A friend in need is a friend indeed: social storage
and the origins of social ranking. In: Renfrew, Colin, Shennan, Stephen (Eds.),
Ranking, Resource and Exchange: Aspects of the Archaeology of Early European
Society. Cambridge University Press, Cambridge, pp. 92–99.
Halstead, Paul, O’Shea, John (Eds.), 1989. Bad Year Economics: Cultural Responses to
Risk and Uncertainty. Cambridge University Press, Cambridge.
Hansell, Patricia K., 1988. The Rise and Fall of an Early Formative Community: La
Mula-Sarigua, Central Pacific Panama. Unpublished Ph.D. dissertation,
Department of Anthropology, Temple University, Philadelphia.
Harrison, Ainslie, Beaubien, Harriet F., 2010. Bringing context to the Smithsonian
collections of pre-columbian gold from panama through technical examination
and analysis. In: Mardikian, Paul, Chemello, Claudia, Watters, Christopher, Hull,
Peter (Eds.), Metal 2010: Proceedings of the Interim Meeting of the ICOM-CC
Metal Working Group. Clemson University, Clemson, SC, pp. 266–272.
Hearne, Pamela, Sharer, Robert J. (Eds.), 1992. River of Gold: Precolumbian
Treasures from Sitio Conte. University Museum of Archaeology and
Anthropology, University of Pennsylvania, Philadelphia.
Hegmon, Michelle, 1989. Risk reduction and variation in agricultural economies: a
computer simulation of Hopi agriculture. Res. Econ. Anthropol. 11, 89–121.
Helms, Mary W., 1979. Ancient Panama: Chiefs in Search of Power. University of
Texas Press, Austin.
Helms, Mary W., 1994. Chiefdom rivalries, control, and external contacts in lower
Central America. In: Brumfiel, Elizabeth, Fox, John W. (Eds.), Factional
Competition and Political Development in the New World. Cambridge
University Press, Cambridge, pp. 55–60.
Hruby, Zachary X., Flad, Rowan K., (Eds.), 2007. Rethinking Craft Specialization in
Complex Societies: Archaeological Analyses of the Social Meaning of
Production. Archaeological Papers of the American Anthropological
Association, No. 17. American Anthropological Association, Arlington, VA.
Hunter, John M., 1967. The social roots of dispersed settlement in northern Ghana.
Ann. Assoc. Am. Geogr. 57 (2), 338–349.
211
Ichon, Alain, 1980. Archéologie du Sud de la Péninsule d’Azuero, Panama. Etudes
Mesoaamericaines, Serie II. Mission Archeologique et Ethnologique Française au
Mexique, Mexico City.
Instituto Nacional de Estadística y Censo (INEC), 2010. Precipitación Pluvial
Registrada en las Estaciones Meterológicas de la República, Según Provincia,
Comarca Indígena y Estación: Años 2001–10. Electronic document, last
accessed 08.12.14. <https://www.contraloria.gob.pa/INEC/Publicaciones/
Publicaciones.aspx?ID_SUBCATEGORIA=4&ID_PUBLICACION=377&ID_IDIOMA=
1&ID_CATEGORIA=2>.
Isaza Aizpurúa, Ilean Isel, 2007. The Ancestors of Parita: Pre-Columbian Settlement
Patterns in the Lower La Villa River Valley, Azuero Peninsula, Panama.
Unpublished Ph.D. dissertation, Department of Archaeology, Boston
University, Boston.
Isendahl, Christian, Smith, Michael E., 2013. Sustainable agrarian urbanism: the
low-density cities of the mayas and aztecs. Cities 31, 132–143.
Johnson, Gregory A., 1982. Organizational structure and scalar stress. In: Renfrew,
Colin, Rowlands, Michael, Segraves-Whallon, Barbara A. (Eds.), Theory and
Explanation in Archaeology. Academic Press, New York, pp. 389–421.
Jopling, Carol F., 1994. Indios y Negros en Panamá en los Siglos XVI y XVII:
Selecciones de los Documentos del Archivo General de Indias. Centro de
Investigaciones Regionales de Mesoamérica, Antigua.
Junker, Laura Lee, 1999. Raiding, Trading, and Feasting: The Political Economy of
Philippine Chiefdoms. University of Hawai’i Press, Honolulu.
Killion, Thomas W., 1990. Cultivation intensity and residential site structure: an
ethnoarchaeological examination of peasant agriculture in the Sierra de los
Tuxtlas, Veracruz, Mexico. Latin Am. Antiq. 1 (3), 191–215.
Kirch, Patrick V., 1984. The Evolution of the Polynesian Chiefdoms. Cambridge
University Press, Cambridge.
Kohler, Timothy A., David Johnson, C., Varien, Mark, Ortman, Scott, Reynolds,
Robert, Kobti, Ziad, Cowan, Jason, Kolm, Kenneth, Smith, Schaun, Yap, Lorene,
2007. Settlement ecodynamics in the prehispanic central mesa verde region. In:
Kohler, Timothy A., van der Leeuw, Sander. (Eds.), The Model-Based
Archaeology of Socionatural Systems. School for Advanced Research Press,
Santa Fe, pp. 61–104.
Kohler, Timothy A., Denton, Cockburn, Hooper, Paul L., Bocinsky, R. Kyle, Kobti, Ziad,
2012. The coevolution of group size and leadership: an agent-based public
goods model for prehispanic pueblo societies. Adv. Complex Syst. 15 (1/2),
1150007-1–1150007-29.
Kosse, Krisztina, 2001. Some regularities in human group formation and the
evolution of societal complexity. Complexity 6 (1), 60–64.
Kowalewksi, Stephen A., Feinman, Gary M., Laura Finsten, Blanton, Richard E.,
Nicholas, Linda M., 1989. Monte Albán’s Hinterland, Part II: Prehispanic
Settlement Patterns in Tlacolula, Etla, and Ocotlan, the Valley of Oaxaca,
Mexico. Memoirs of the Museum of Anthropology, No. 23. University of
Michigan, Ann Arbor.
Kowalewski, Stephen A., 1980. Population-resource balances in period I of Oaxaca,
Mexico. Am. Antiq. 45 (1), 151–165.
Kowalewski, Stephen A., 2004. The new past: from region to macroregion. Soc. Evol.
History 3 (1), 81–105.
Kowalewski, Stephen A., Balkansky, Andrew K., Stiver, Laura R., Walsh, Thomas J.,
Pluckhahn, John F., Chamblee, Verónica Pérez, Rodríguez, Verenice Y., Espinoza,
Heredia, Smith, Charlotte A., 2009. Origins of the Ñuu: Archaeology in the
Mixteca Alta. University Press of Colorado, Boulder, Mexico.
Kumar, B.M., Nair, P.K.R. (Eds.), 2006. Tropical Homegardens: A Time-Tested
Example of Sustainable Agroforestry. Springer, Dordrecht.
Lemonnier, Eva, Vannière, Boris, 2013. Agrarian features, farmsteads, and
homesteads in the Río Bec Nuclear Zone, Mexico. Ancient Mesoamerica 24
(2), 397–413.
Linares, Olga F., 1977. Ecology and the Arts in Ancient Panama: On the Development
of Social Rank and Symbolism in the Central Provinces. Studies in Pre-
Columbian Art and Archaeology, No. 17. Dumbarton Oaks, Washington, D.C.
Linares, Olga F., Ranere, Anthony J., (Eds.), 1980. Adaptive Radiations in Prehistoric
Panama. Peabody Museum Monographs, No. 5. Peabody Museum of
Archaeology and Ethnology, Harvard University, Cambridge, MA.
Locascio, William A., 2010. Communal Tradition and the Nature of Social Inequality
among the Prehispanic Households of El Hatillo (He-4), Panama. Unpublished
Ph.D. dissertation, Department of Anthropology, University of Pittsburgh,
Pittsburgh.
Lothrop, Samuel K., 1937. Coclé: An Archaeological Study of Central Panama, Part 1.
Memoirs of the Peabody Museum of Archaeology and Ethnology, vol. 7. Peabody
Museum of Archaeology and Ethnology, Harvard University, Cambridge, MA.
Marcus, Joyce, Flannery, Kent V., 1996. Zapotec Civilization: How Urban Society
Evolved in Mexcio’s Oaxaca Valley. Thames and Hudson, London.
Martín, Alexander J., 2010. The domestic economy and its implications for social
complexity: Spondylus craft production in coastal Ecuador. Res. Econ. Anthropol.
30, 111–155.
Martín, Alexander J., Murillo Herrera, Mauricio, 2014. Networks of interaction and
functional interdependence in societies across the intermediate area. J.
Anthropol. Archaeol. 36, 60–71.
Masucci, Maria A., 1995. Marine shell bead production and the role of domestic craft
activities in the economy of the Guangala phase, Southwest Ecuador. Latin Am.
Antiq. 6 (1), 70–84.
Mayo, Julia, 2004. La Industria Prehispánica de Conchas Marinas en ‘‘Gran Coclé”,
Panamá. Unpublished Ph.D. dissertation, Universidad Complutense de Madrid,
Madrid.
212 C.A. Berrey / Journal of Anthropological Archaeology 40 (2015) 196–212
Mayo, Julia, Cooke, Richard, 2005. La Industria Prehispánica de Conchas Marinas en
Gran Coclé, Panamá: Análisis Technológico de los Artefactos de Concha del
Basurero-Taller del Sitio Cerro Juan Díaz, Los Santos, Panamá. Archaeofauna 14,
285–298.
Mayo, Julia, Mayo, Carlos, 2013. El Descubrimiento de un Cementerio de Élite en El
Caño: Indicios de un Patrón Funerario en el Valle de Río Grande, Coclé, Panamá.
Arqueología Iberoamericana 20, 3–27.
Mayo, Julia, Mojica, Alexis, Ruiz, Alberto, Moreno, Enrique, Mayo, Carlos, de Gracia,
Guillermina Itzel, 2007. Estructuras Arquitectónicas Incipientes y Áreas de
Explotación Minera Prehispánica de las Cuencas de los Ríos Grande y Coclé del
Sur, Panamá. Revista Española de Antropología Americana 37 (1), 93–110.
Mayo, Julia, Mayo, Carlos, Karas, Vicky, 2010. La Escultura Precolombina del Área
Intermedia: Aproximación al Estudio Estilístico, Iconográfico y Espacial del
Grupo Escultórico de El Caño. In: Tísoc, Emiliano Meglar, Ciriaco, Reyna Solís,
González Licón, Ernesto (Eds.), Producción de Bienes de Prestigio Ornamentales
y Votivos de la América Antigua, Syllaba Press, Doral, FL, pp. 85–96.
Menzies, Adam C.J., 2009. Craft Specialization and the Emergence of the Chiefly
Central Place Community of He-4 (El Hatillo), Central Panama. Unpublished Ph.D.
dissertation, Department of Anthropology, University of Pittsburgh, Pittsburgh.
Menzies, Adam C.J., Haller, Mikael J., 2012a. A macroregional perspective on chiefly
cycling in the central region of Panama during the Late Ceramic II period (A.D.
700–1522). Latin Am. Antiq. 23 (4), 449–466.
Menzies, Adam C.J., Haller, Mikael J., 2012b. Embedded craft production at the pre-
columbian (AD 900–1522) community of He-4 (El Hatillo, Central Region of
Panama. Can. J. Archaeol. 36 (1), 108–140.
Mojica, Alexis, Mayo, Julia, Mayo, Carlos, Chantada, José Ramón, de Gracia,
Guillermina Itzel, Florsch, Nicolas, 2007. Resultados de las Prospecciones
Magnética y Eléctrica del Yacimiento Arqueológico El Caño (NA-20), Gran
Coclé, Panamá. Revista Española de Antropología Americana 37 (1), 111–126.
Murillo Herrera, Mauricio, 2011. Precolumbian Social Change in San Ramón de
Alajuela, Costa Rica. University of Pittsburgh Memoirs in Latin American
Archaeology, No. 22. University of Pittsburgh Center for Comparative
Archaeology, Pittsburgh.
National Oceanic and Atmospheric Administration, 2002a. Monthly Meteorological
Summary (Balboa Heights, Panama). Electronic document, last accessed
08.12.14. <http://docs.lib.noaa.gov/rescue/data_rescue_panama.html>.
National Oceanic and Atmospheric Administration, 2002b. Monthly Meteorological
Summary (Cristobal, Panama). Electronic document, last accessed 08.12.14.
<http://docs.lib.noaa.gov/rescue/data_rescue_panama.html>.
Netting, Robert McC., 1977. Maya subsistence: mythologies, analogies, possibilities.
In: Adams, Richard E.W. (Ed.), The Origins of Maya Civilization. University of
New Mexico Press, Albuquerque, pp. 299–333.
Netting, Robert McC., 1993. Smallholders, Householders: Farm Families and the
Ecology of Intensive, Sustainable Agriculture. Stanford University Press,
Stanford.
Nicholas, Linda M., Feinman, Gary M., 1989. A regional perspective on Hohokam
irrigation in the lower Salt River Valley, Arizona. In: Upham, Steadman,
Lightfoot, Kent G., Jewett, Roberta A. (Eds.), The Sociopolitical Structure of
Prehistoric Southwestern Societies. Westview Press, Boulder, pp. 199–235.
Nondédéo, Philippe, Arnauld, M. Charlotte, Michelet, Dominique, 2013. Río Bec
Settlement Patterns and Local Sociopolitical Organization. Ancient
Mesoamerica 24 (2), 373–396.
Norr, Lynette Caryl, 1991. Nutritional Consequences of Prehistoric Subsistence
Strategies in Lower Central America. Unpublished Ph.D. dissertation,
Department of Anthropology, University of Illinois at Urbana Champaign,
Urbana.
Norr, Lynette Caryl, 1995. Interpreting dietary maize from bone stable isotopes in
the American tropics: the state of the art. In: Stahl, Peter W. (Ed.), Archaeology
in the Lowland American Tropics: Current Analytical Methods and Applications.
Cambridge University Press, Cambridge, pp. 198–223.
Pearson, Michael Parker, 1982. Mortuary practices, society, and ideology: an
ethnoarchaeological study. In: Hodder, Ian (Ed.), Symbolic and Structural
Archaeology. Cambridge University Press, Cambridge, pp. 99–113.
Peebles, Christopher S., Kus, Susan M., 1977. Some archaeological correlates of
ranked societies. Am. Antiq. 42 (3), 421–448.
Peterson, Christian E., Drennan, Robert D., 2005. Communities, settlements, sites,
and surveys: regional-scale analysis of prehistoric human interaction. Am.
Antiq. 70 (1), 5–30.
Peterson, Christian E., Drennan, Robert D., 2012. Patterned variation in regional
trajectories of community growth. In: Smith, Michael E. (Ed.), The Archaeology
of Complex Societies. Cambridge University Press, Cambridge, pp. 88–137.
Peterson, Christian E., Lu, Xueming, Drennan, Robert D., Zhu, Da, 2014. Hongshan
Regional Organization in the Upper Daling Valley. University of Pittsburgh
Center for Comparative Archaeology, Pittsburgh.
Piperno, Dolores R., 2006. The origins of plant cultivation and domestication in the
neotropics. A behavioral ecological perspective. In: Kennett, Douglas J.,
Winterhalder, Bruce (Eds.), Behavioral Ecology and the Transition to
Agriculture. University of California Press, Berkeley, pp. 137–166.
Piperno, Dolores R., Pearsall, Deborah M., 1998. The Origins of Agriculture in the
Lowland Neotropics. Academic Press, San Diego.
Podolefsky, Aaron, 1984. Contemporary warfare in the New Guinea highlands.
Ethnology 23 (2), 73–87.
Price, T. Douglas, Feinman, Gary M. (Eds.), 1995. Foundations of Social Inequality.
Plenum Publishers, New York.
Price, T. Douglas, Feinman, Gary M. (Eds.), 2010. Pathways to Power: New
Perspectives on the Emergence of Social Inequality. Springer, New York.
Redmond, Elsa M., Spencer, Charles S., Gassón, Rafael A., 1999. A macroregional
view of cycling chiefdoms in the western Venezuelan llanos. In: Bacus, Elisabeth
A., Lucero, Lisa J. (Eds.), Complex Polities in the Ancient Tropical World.
Archaeological Papers of the American Anthropological Associations, No. 9.
American Anthropological Association, Arlington, VA, pp. 109–129.
Renfrew, Colin, 1974. Beyond a subsistence economy: the evolution of social
organization in prehistoric Europe. In: Moore, Charlotte B. (Ed.), Reconstructing
Complex Societies: An Archaeological Colloquium. Supplement to the Bulletin
of the American Schools of Oriental Research, No. 20. American Schools of
Oriental Research, Cambridge, pp. 69–95.
Reynolds, Robert G., Kohler, Timothy A., Kobti, Ziad, 2003. The effects of generalized
reciprocal exchange on the resilience of social networks: an example from the
prehispanic mesa verde region. Comput. Math. Org. Theory 9, 227–254.
Roosevelt, Anna Curtenius, 1979. The goldsmith: the Coclé style of ancient Panama.
In: Roosevelt, Anna Curtenius, Smith, James G.E. (Eds.), The Ancestors: Native
Artisans of the Americas. Museum of the American Indian, New York, pp. 66–
101.
Rosenberg, Michael, 2009. Proximate causation, group selection, and the evolution
of hierarchical human societies: system, process, and pattern. In: Prentiss, Anna
Marie, Kuijt, Ian, Chatters, James C. (Eds.), Macroevolution in Human Prehistory:
Evolutionary Theory and Processual Archaeology. Springer, New York, pp. 23–
49.
Sahlins, Marshall D., 1958. Social Stratification in Polynesia. University of
Washington Press, Seattle.
Sahlins, Marshall D., 1961. The segmentary lineage: an organization of predatory
expansion. Am. Anthropol. 63 (2), 322–345.
Sanders, William T., 1981. Classic maya settlement patterns and ethnographic
analogy. In: Ashmore, Wendy (Ed.), Lowland Maya Settlement Patterns.
University of New Mexico Press, Albuquerque, pp. 351–369.
Sanders, William T., Webster, David, 1978. Unilinealism multilinealism and the
evolution of complex societies. In: Redman, Charles L., Berman, Mary Jane,
Curtin, Edward V., Langhorne Jr., William T., Versaggi, Nina M., Wasner, Jeffery
C. (Eds.), Social Archaeology: Beyond Subsistence and Dating, Academic Press,
New York, pp. 249–302.
Saxe, Arthur A., 1970. Social Dimensions of Mortuary Practice. Unpublished Ph.D.
dissertation, Department of Anthropology, University of Michigan, Ann Arbor.
Scarborough, Vernon L., Burnside, William R., 2010. Complexity and sustainability:
perspectives from the ancient maya and the modern Balinese. Am. Antiq. 75 (2),
327–363.
Schmieder, Oscar, 1930. The Settlements of the Tzapotec and Mije Indians, State of
Oaxaca, Mexico. University of California Publications in Geography, Volume IV.
University of California Press, Berkeley.
Spencer, Charles S., 1993. Human agency, biased transmission, and the cultural
evolution of chiefly authority. J. Anthropol. Archaeol. 12 (1), 41–74.
Spencer, Charles S., Redmond, Elsa M., 1992. Prehispanic chiefdoms of the western
Venezuelan llanos. World Archaeol. 24 (1), 134–157.
Spencer, Charles S., Redmond, Elsa M., 1998. Prehispanic causeways and regional
politics in the llanos of Barinas, Venezuela. Latin Am. Antiq. 9 (2), 95–110.
Spielmann, Katherine A., 2002. Feasting, craft specialization, and the ritual mode of
production in small-scale societies. Am. Anthropol. 104 (1), 195–207.
Stone, Glenn Davis, 1991. Agricultural territories in a dispersed settlement system.
Curr. Anthropol. 32 (3), 343–353.
Stone, Glenn Davis, 1992. Social distance, spatial relations, and agricultural
production among the Kofyar of Namu District, Plateau State, Nigeria. J.
Anthropol. Archaeol. 11 (2), 152–172.
Stone, Glenn Davis, 1996. Settlement Ecology: The Social and Spatial Organization
of Kofyar Agriculture. University of Arizona Press, Tucson.
Stone, Glenn Davis, Downum, Christian E., 1999. Non-Boserupian ecology and
agricultural risk: ethnic politics and land control in the arid southwest. Am.
Anthropol. 101 (1), 113–128.
Tainter, Joseph A., Tainter, Bonnie Bagley (Eds.), 1992. Evolving Complexity and
Environmental Risk in the Prehistoric Southwest. Addison-Wesley, Reading,
MA.
Ucko, Peter J., 1969. Ethnography and archaeological interpretation of funerary
remains. World Archaeol. 1 (2), 262–280.
Williams, A.R., 2012. The golden chiefs of Panama. Nat. Geogr. 221 (1), 66–81.