HOHOKAM TO AKIMEL OODHAM:

OBSIDIAN ACQUISITION AT THE HISTORIC PERIOD SACATE

SITE (GR-909), GILA RIVER INDIAN COMMUNITY, ARIZONA

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

Chris R. Loendorf, Craig M. Fertelmes, and Barnaby V. Lewis

Obsidian provenance studies within the Phoenix Basin of south-central Arizona have become increasingly comprehensive
during the last four decades. As a result, broad regional and temporal trends have been defined regarding Preclassic (ca.
A.D. 6501150) and Classic period (ca. A.D. 11501450) socioeconomic interactions in the Hohokam core area. However,
Historic period patterns are still poorly understood, and these data are essential for understanding the relationship between
the prehistoric Hohokam and historical Akimel OOdham. The association between the Akimel OOdham and Hohokam
has been debated since Euro-Americans first visited the area in the late 1600s, yet this issue is still not fully resolved. This
article presents analyses of historical obsidian from the Sacate site that suggest that long-term trends in cultural patterns
within the Phoenix Basin continued unbroken into the Historic period. These continuities provide another line of evidence
that the Akimel OOdham are the direct cultural descendants of the Hohokam.

Los estudios sobre las fuentes de origen de la obsidiana dentro de la Cuenca de Phoenix en la regin sud-central de Arizona
han llegado a ser cada vez ms completos durante las ltimas cuatro dcadas. Como resultado, amplios patrones espaciotemporales han sido definidos con respecto a los periodos Pre-Clsico (ca. d.C. 6501150) y Clsico (ca. d.C. 11501450)
destacando las interacciones socioeconmicas en el ncleo de la cultura Hohokam. Sin embargo, las interacciones durante
el perodo Histrico todava no son claras, y estos datos son esenciales para comprender la relacin entre los Hohokam prehistricos y los Akimel OOdham histricos. La asociacin entre los OOdham y los Hohokam ha sido debatida desde que los
primeros Euroamericanos visitaron la regin al final de los 1600; no obstante, este tema todava no est resuelto completamente. Este trabajo resume los resultados del anlisis de obsidiana histrica del sitio Sacate, lo que sugiere que a largo plazo
los patrones culturales dentro de la Cuenca de Phoenix continuaron intactos durante el perodo Histrico. Esta continuidad
es fuerte evidencia de que los Akimel OOdham son los descendientes de los Hohokam.

his study examines Akimel OOdham (i.e.,

Pima) socioeconomic interactions through
analyses of Protohistoric (ca. A.D.
15001693) and Historic period (A.D. 16941962)
obsidian acquisition patterns at the Sacate site (GR909) in south-central Arizona. Until recently,
comparatively few archaeological data were available from the Akimel OOdham settlement area
along the Middle Gila River portion of the Hohokam core area in the Phoenix Basin (Figure 1).
However, as part of the Pima-Maricopa Irrigation
Project, the Gila River Indian Community (GRIC)

created a Cultural Resource Management Program

that has conducted a full-coverage archaeological
survey of nearly 57,000 ha within the community (Darling 2011; Darling et al. 2004; Loendorf
2012; Loendorf and Rice 2004; Ravesloot 2007;
Waters and Ravesloot 2000, 2001; Wells 2006;
Wells et al. 2004; Woodson 2010). These investigations included recording efforts at Sacate, an
extensive historical village located on the south side
of the Middle Gila near the center of the GRIC.
Prehistoric remains are uncommon at Sacate,
and therefore it is possible to study indigenously

Chris R. Loendorf Cultural Resources Management Program, Gila River Indian Community, Sacaton, AZ 85147
(chris.loendorf@gric.nsn.us)
Craig M. Fertelmes Cultural Resources Management Program, Gila River Indian Community, Sacaton, AZ 85147
(craig.fertelmes@gric.nsn.us)
Barnaby V. Lewis Tribal Historic Preservation Office, Gila River Indian Community, Sacaton, AZ 85147
(barnaby.lewis@gric.nsn.us)
American Antiquity 78(2), 2013, pp. 266284
Copyright 2013 by the Society for American Archaeology
266

267

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

OBSIDIAN ACQUISITION AT THE SACATE SITE

Figure 1. Map showing the location of Sacate (GR-909), physiographic features, Late Historic tribal territories (following Ortiz 1983), and historically mapped village locations (following Ezell 1961; Wilson, forthcoming).

produced historic artifacts in a setting that largely lacks intermixing of earlier materials.
Despite repeated attempts to obtain firearms, the
historical Akimel OOdham possessed few guns.
Instead, bows and arrows were commonly used until the end of the nineteenth century (Bancroft
1886:520; Grossman 1873:416; Mason 1894;
Russell 1908:111). In contrast to many eastern and

Plains groups, Native Americans along the Middle Gila rarely employed metal arrowpoints, and
they continued to make projectile tips from stone,
including obsidian, until roughly A.D. 1880 (Bancroft 1886; Ferg and Tessman 1997:259261;
Grossman 1873). Written descriptions of Akimel
OOdham settlement locations, technological traditions, socioeconomic interactions, and subsistence

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

268

AMERICAN ANTIQUITY

practices began in the late 1600s and continued over

the course of the Historic period (DeJong 2009,
2011; Ezell 1961; Fewkes 1912; Hackenberg
1974; Russell 1908; Seymour 2011; Wells 2006;
Wilson, forthcoming). Thus, the situation along the
Middle Gila offers an important opportunity to
compare spatial and temporal patterning among obsidian artifacts with historically documented socioeconomic interactions, projectile technology, and
other cultural practices (Loendorf 2012).
For multiple reasons, obsidian has properties that
are ideally suited for studying exchange and interaction in southern Arizona (Ballenger and Hall
2011; Bayman 1995; Fertelmes et al. 2012; Loendorf 2010, 2012; Marshall 2002; Mitchell and Foster 2011; Mitchell and Shackley 1995; Peterson et
al. 1997; Rice et al. 1998; Shackley 1988, 1990,
1995, 2005). First, raw materials necessary for manufacturing retouched tools are uncommon in the
Phoenix Basin, and obsidian is a desirable stone for
producing arrowpoints. Second, because obsidian
does not occur naturally within the Phoenix Basin,
all of the material at archaeological sites must have
been culturally transported from elsewhere. Third,
obsidian sources are generally abundant and localized deposits. Fourth, obsidian has geochemical properties that allow source locations to be objectively identified with a high degree of precision.
It is therefore possible to examine the nature of exchange interactions between the core area and surrounding populations, as well as relationships
among communities within the Phoenix Basin.
Obsidian source provenance studies are also an
important component of prehistoric archaeological research throughout the Southwest and other
parts of North America, including the Pacific
Northwest, California, the Great Basin, and
Mesoamerica. Although archaeologists have devoted considerable effort to analyzing precontact
assemblages, obsidian from Historic period contexts has rarely been considered (but see Millhauser
et al. 2011; Silliman 2005). This lack of research
does not result from an absence of obsidian at historical indigenous sites but, instead, occurs because
archaeologists who study postcontact sites generally
have not considered the research potential of these
remains (Silliman 2005:75). The following analyses explore the broader implications of these artifacts, and this investigation illustrates how obsidian data can be used to improve our under-

[Vol. 78, No. 2, 2013

standing of associations among prehistoric and historical indigenes.

The relationship between the late prehistoric inhabitants of the Middle Gila (i.e., Classic Hohokam) and the Akimel OOdham has been debated
since Spanish missionaries first visited the Phoenix
Basin in the late 1600s (Fewkes 1912:33; Russell
1908). Despite centuries of speculation and argument, the association between the precontact and
postcontact populations remains unresolved (Ezell
1983:149150; Gilpin and Phillips 1998:2843;
Seymour 2011; Wells 2006). Data from the Protohistoric and Early Historic periods (A.D.
16941820) are critical for assessing the Hohokam
collapse as well as the continuum of populations.
However, Akimel OOdham archaeological remains from this time are largely restricted to the
GRIC, where until recently little research has occurred (Wells 2006:12).
Relationships among the Hohokam and contemporary groups are of considerable importance
to the Akimel OOdham, and this issue has modern sociopolitical ramifications in a region where
highly contested water rights are based on prior usage (DeJong 2011:1721). While the Akimel
OOdham continue to maintain detailed oral traditions regarding their past, these social memories
have been extensively ignored or misunderstood.
Not only does this matter affect these peoples perceptions of their past, it also impacts their lives in
other ways. These include issues of cultural patrimony and repatriation cases where Akimel
OOdham perspectives are still sometimes undervalued or dismissed.
Ethnohistorical and
Ethnographic Background

The following section presents descriptions of projectile design and hunting practices, which are employed to infer functional aspects of historical projectiles. These observations suggest that historical flaked-stone arrowpoints were largely if not
exclusively produced for use in warfare. Next, intergroup conflicts that substantially impacted
Akimel OOdham socioeconomic interactions are
described. Finally, historical descriptions of exchange relationships are used to generate expectations for obsidian acquisition patterns at
Sacate.

Loendorf et al.]

OBSIDIAN ACQUISITION AT THE SACATE SITE

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

Akimel OOdham Projectile Design

Contrary to a common assumption, a stone projectile tip is not necessary to balance the shaft.
Instead, prehistoric and historical projectiles commonly had organic tips such as bone, antler, or
wood (Mason 1894). In a cross-cultural study of
over 100 societies, Ellis (1997) found that different types of projectile tips were employed for separate purposes. Stone points were almost invariably associated with hunting large game animals
(> 40 kg) and/or warfare, while organic tips were
far more commonly employed in small-game (<
40 kg) hunting.
By the 1800s, the Akimel OOdham rarely hunted large animals (Ezell 1961; Rea 1998; Russell
1908). Prehistoric sites below 800 m elevation in
southern Arizona generally have low incidences
of large faunal remains, while habitations above
this elevation commonly have evidence for greater
reliance on ungulates (Szuter 1991). The nearest
extensive areas above 800 m are roughly 30 km
away from the Middle Gila in the mountains to the
north and east, but these locations were occupied
by the Nn (i.e., Apache) and Gn (i.e., Yavapai) during the Historic period (see Figure 1).
The Akimel OOdham instead primarily hunted locally available small game, and the arrows they
used for this purpose lacked stone tips. Arrows designed for use in warfare, however, had flaked-stone
points attached (Bancroft 1886:520; Grossman
1873:416; Mason 1894; Russell 1908). For example, Bancroft said that the Akimel OOdham wing
their war arrows with three feathers and point them
with flint, while for hunting purposes they have only
two feathers and wooden points (1886:520).

Conflict

While the nature and intensity of warfare varied

over time and space, conflict was endemic among
Southwestern Historic populations (Basso 2004;
Burns 2012; Ezell 1961; Jacoby 2008; Kroeber and
Fontana 1986; Rice 2001; Russell 1908; Shaw
1994:1014; Spier 1933; Webb 1959:2225).
The Nn and Gn regularly raided Akimel OOdham villages along the Middle Gila. Although individual attacks were generally minor, conflict with
these groups caused the aggregation of Akimel
OOdham populations, which resulted in the
abandonment of extensive former habitation areas

269

(Ezell 1961; Hackenberg 1974). These conflicts effectively circumscribed Akimel OOdham socioeconomic interactions, because intervening
Nn and Gn populations restricted trade with
other groups, including Pueblo populations to the
north and east (Russell 1908; see Figure 1).
Socioeconomic Interactions

By the nineteenth century the Pee Posh (i.e., Maricopa) and Tohono OOdham (i.e., Papago) were
the primary indigenous exchange partners for the
Akimel OOdham (Russell 1908:93). Prior to
roughly A.D. 1830, Pee Posh from the Gila Bend
area came at harvest time to exchange goods with
the Akimel OOdham (Russell 1908:93). After the
Pee Posh moved from farther west along the Gila
to the area immediately adjacent to the Akimel
OOdham villages, the Tohono OOdham were
their primary nonlocal exchange partners. Although
the Tohono OOdham lived in more arid desert environments to the south, in addition to salt they also
brought food and a wide variety of other items for
exchange (Russell 1908:93; Webb 1959:65).
The Akimel OOdham also sold items to or
bartered with Mexican populations to the south,
and by the 1850s they traded or sold large
amounts of goods with settlers and others who
traveled through the area (DeJong 2009, 2011).
The Akimel OOdham also sent well-armed
bands through the Apache cordon to trade at the
Spanish and Mexican settlements of Sonora (Russell 1908:94). In addition, an annual fair was held
along the Middle Gila that was attended by multitudes of people from the towns and presidios of
Tucson, Tubac, Santa Cruz, and the San Ignacio
River (Ezell 1961:30).
For much of the Historic period, the items the
Akimel OOdham produced for trade consisted
largely of baskets and woven goodsbelts, hair
bands, and above all else cotton blankets (Ezell
1961:29). By the 1850s they also grew large surpluses of commodities that they sold or bartered
to the U.S. military, and previously they were able
to provide food to other communities in times of
need (DeJong 2009; Ezell 1961). These socioeconomic interactions were not exclusively based
on kinship ties or simple reciprocal relationships,
and Russell listed the late 1800s exchange rates as
follows:

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

270

AMERICAN ANTIQUITY

[Vol. 78, No. 2, 2013

Figure 2. Map showing feature locations, cemeteries, and site segments at Sacate (GR-909).

A gourd was equivalent to a basket; a metate,

a small shell necklace, or the combination of
a basket and a blanket and a strand of blue glass
beads was equivalent to a horse; a string of blue
glass beads 4 yards long was equivalent to a
bag of paint; and a basket full of beans or corn
to a cooking pot [1908:92].

While the Akimel OOdham used Euro-American glass beads for exchange, they expressed
strong preferences for particular types of ornaments, especially those made from marine shell and
stone. For example, Bartlett observed that some
Akimel OOdham had
long strings of sea-shells or parts of shells,
which are highly prized. I tried to buy some of
them; but the only man at all disposed to sell
asked me five dollars [equivalent to roughly
$130 today] or a pair of blankets for a few
strings, a price so extravagant that I declined
to make the purchase [1854:231].

Glass beads had relatively limited worth by the second half of the nineteenth century, and the value
of items changed over the course of the Historic
period based on factors of supply and demand
along the Middle Gila (Ezell 1961:31). Audubon,
for example, observed in A.D. 1850 that the extravagance of the Americans who have passed
through has made it difficult for anyone to make
reasonable bargains with either Pimos [Akimel
OOdham] or Maricopas [Pee Posh] (1906:150).
Audubon goes on to say:

Many who came to trade had already made up

their minds only to do so for some particular
article, and in those cases it was not of the least
avail to offer anything else. . . . Jewelry had
no value to them, fancy beads were worthless,
stone beads however they traded for eagerly,
but we had none [1906:150].

In aggregate, these observations suggest that the

Akimel OOdham chose to adopt Euro-American
goods that fit within previously existing socioeconomic interaction patterns along the Middle Gila.
Furthermore, until the Akimel OOdham became
more fully incorporated into the American economic system in the 1900s, Euro-Americans and
the Akimel OOdham also valued goods differently,
such that items considered valuable by the former
had little worth to the latter and vice versa.
Sacate Site Description

Sacate (GR-909) is a roughly 3-km-long and .8km-wide habitation where 204 features, including
ki-kuh (traditional Akimel OOdham round houses), other areas with structural remains, extensive
midden deposits, and three cemeteries were identified (Figure 2). Obsidian artifacts analyzed in this
study were derived from systematic surface collections undertaken by GRIC Cultural Resource
Management Program field crews between March
and August 2000. All identified diagnostic remains,
including any obsidian or projectile points, were
collected. A total of 184 quantitative units, which

Loendorf et al.]

OBSIDIAN ACQUISITION AT THE SACATE SITE

271

Table 1. Absolute Frequencies of Pottery Sherds Recovered from Sacate by Time Period, Type, and Segment of Site.

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

Period
Historic
Historic
Historic
Historic
Historic
Historic
Historic

Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric
Prehistoric

Ceramic Type
Red on Buff
Red on Brown
Plain
Red
Black on Red
Papago Green Glaze
Indet. Yellow Ware
Subtotal

Plain/Smudged
Sacaton Red on Buff
Snaketown Red on Buff
Gila Butte Red on Buff
Casa Grande Red on Buff
Santa Cruz Red on Buff
Gila Polychrome
Tanque Verde Red on Brown
Red/Smudged
Cibola White Ware
Tonto Polychrome
Indet. Corrugated Plain
Mogollon Brown
Subtotal

TOTAL

consisted of 3.14-m2 areas where all artifacts were

recovered, were also collected (Wells et al. 2004).
Darling (2011) describes a process whereby
Akimel OOdham villages drift over time, and
analyses of nonindigenous artifacts and projectile
point data from Sacate support this model (Darling et al. 2004; Loendorf 2012:8389; Spier
1933:22). Temporal shifts among segments of the
site appear to have resulted from historical and environmental events such as immigration, movement
of the Gila River channel, flooding, the construction
of transportation routes, and other Euro-American
interactions (Darling 2011). Although the precise
dating of site segments is uncertain, there is considerable evidence that occupation of Sacate
spans the Protohistoric to Modern periods, and substantial differences in the artifact collections and
feature types in site segments suggest relative differences in the age of habitation.
Indigenous ceramic data suggest that Sacate
largely postdates the prehistoric period, as all precontact types combined make up only 1 percent of
the collection (Table 1). Similarly, projectile
points from the site are also predominately historic
types (Table 2). Based on an analysis of early his-

Central
3659
1588
1534
1165
176
1
1
8124

Expansion
2965
1403
1556
769
129
3
6825

Western
2724
1197
746
394
132
3
5196

TOTAL
9348
4188
3836
2328
437
4
4
20145

8200

6913

5236

20349

42
8
1
5
8
1
6
3
1
1
76

30
25
9
8
13
2
1
88

4
4
3
7
1
5
10
4
1
1
40

46
42
29
21
16
15
13
10
7
2
1
1
1
204

torical records, Wilson (forthcoming) concludes

that Sacate was documented by the Jesuit missionary Father Kino, who was the first Euro-American to visit the Phoenix Basin (see also Ezell
1961). This village was labeled Soacin on
Kinos 16951696 map of the area. Wilson argues
that the synonymy of this name suggests that it corresponds with a village called Sudac-sson or Sutaquisn (Where the water comes up) that was
recorded by Franciscan missionaries in the eighteenth century. As can be seen in Figure 1, the
recorded positions for this village all roughly correspond with the location of Sacate. It is improbable that this area was settled immediately before
Kinos visits, suggesting that Sacate was occupied
at least prior to the mid-1600s.
The relative ages of occupation in different site
segments can in part be inferred from variation in
feature types and nonlocal artifacts. The oldest nonindigenous remains were recovered from the central segment and consist of one-piece metal buttons that were manufactured from 1750 to 1812.
Other nonlocal artifacts suggest that this segment
was occupied through at least the late 1800s. In
contrast to the central segment, ki-kuh are less com-

272

AMERICAN ANTIQUITY

Table 2. Absolute Frequencies of Arrowpoints Recovered from Sacate by Time Period, Style, and Segment of Site.

Period
Historic
Historic

Arrow Point Type

Straight Base Triangular
U-shaped Base Triangular
Subtotal

Classic
Classic
Classic
Classic
Indeterminate
Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

[Vol. 78, No. 2, 2013

Bulbous Base
Concave Base Triangular
Long Triangular
Thin Triangular
Subtotal
Indeterminate

TOTAL

Central
14
13
27
1
4
5
7

39

mon in the expansion segment. A change to other structural types occurred in the 1800s, but kikuh were used until at least the 1930s (Sayles
1938:61). The comparatively low incidence of kikuh in the expansion segment suggests that it was
first inhabited more recently than the central segment. Nonindigenous artifacts from the expansion
segment include military buttons and black glass,
including one example with an improved pontil scar, suggesting that this segment was used prior to the 1880s. Nonindigenous artifacts from the
western segment differ substantially from those collected in the rest of the site, and the incidence of
glass and nonindigenous ceramics as well as
metal items is higher, suggesting that this area has
the most recent occupation (Randolph et al.
2002). Sacate has not been abandoned, and scattered houses as well as modern housing subdivisions occur in the area. Furthermore, the Akimel
OOdham continue to use the cemetery in the western site segment.
Sacate Projectile Point Analysis

Almost 100 arrowpoints were recovered from

GR-909 (see Table 2). All of these artifacts are small
isosceles triangular forms that lack notches or serration (Figure 3). Historic point types that have been
defined within southern Arizona are largely triangular forms that lack notches, but there is significant variation in blade shape, degree of basal concavity, and particularly serration (Justice 2002; Loendorf 2012; Loendorf and Rice 2004; Seymour 2011;
Vint 2005; Wells 2006:17). Although 7 percent of
the arrowpoints from Sacate were classified as Clas-

Expansion
9
12
21
-

12

33

Western
5
13
18
1
1
2

24

TOTAL
28
38
66
1
1
4
1
7

23

96

sic period types, similar unnotched and unserrated

points were produced during the Classic and Historic periods, and these artifacts are possibly misclassified historical remains (Loendorf and Rice
2004:60; Ravesloot and Whittlesey 1987:96).
While the similar morphological characteristics of
Classic and Historic points complicate the separation of these types, this consistency in design (especially the absence of serration) is also evidence
for continuity in technological practices from prehistory to history (Loendorf 2012:101106). Detailed
description of the typological classification system,
metric and attribute data, and images of all points
from Sacate are available in Loendorf and Rice 2004.
The incidence of obsidian artifacts varies
across Sacate, and the highest concentration occurs in the western segment, which was most recently inhabited (Table 3). These observations suggest that increased obsidian use occurred over time,
which is reflected in the material type proportions
for points by site segment (Table 4). This apparent diachronic trend is supported by the fact that
arrowpoint densities are relatively consistent
across the site, which suggests that patterning in
these data did not result from sampling bias, formation, or postdepositional processes.
This trend toward greater reliance on obsidian
for point production appears to be part of a longerterm process that began well before the Historic period. Along the Middle Gila, Archaic points were
only occasionally made from obsidian (< 2 percent),
but use of this material became common by the Classic period (Loendorf and Rice 2004:48). Similar
tendencies for an increase in obsidian use over time
have been documented within Hohokam collections

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

Loendorf et al.]

OBSIDIAN ACQUISITION AT THE SACATE SITE

273

Figure 3. Arrowpoints from Sacate (GR-909): rightmost top two are man-made glass; rightmost bottom three are obsidian.

from elsewhere in southern Arizona (Ballenger and

Hall 2011:146148; Bayman and Shackley 1999;
Fertelmes et al. 2012; Marshall 2002; Peterson et
al. 1997). This continuation of diachronic prehistoric patterns is also evidence for cultural continuity between the Hohokam and Akimel OOdham.
Obsidian Source Analysis Methods

A total of 131 obsidian artifacts from Sacate were

subjected to energy-dispersive X-ray fluores-

cence (EDXRF) analysis. EDXRF analyses of 13

samples were completed by the Archaeological
XRF Laboratory, University of California, Berkeley, using a Spectrace/ThermoNoran QuanX spectrometer (Shackley and Daehnke 2004). The remaining 118 artifacts were analyzed by the GRIC
Material Sciences Laboratory (MSL) using a
Bruker Tracer III-V portable spectrometer.1
In order to compare Berkeley and Tracer III-V
compositional results, GRIC-MSL reanalyzed 105
obsidian artifacts that were previously character-

Table 3. Absolute Frequencies and Densities (n/ha) for Total Obsidian and Arrowpoints from Sacate by Segment of Site.

Site Segment
Central
Expansion
Western
TOTAL

Obsidian
Count
51
54
105
210

Arrow
Point Count
39
28
29
96

Hectares
53
50
48
151

Obsidian
Density
0.97
1.07
2.17
1.39

Arrow
Point Density
0.74
0.56
0.60
0.63

Table 4. Absolute and Relative Frequencies of Arrowpoints from Sacate by Raw Material and Segment of Site.

Site Segment
Central
Expansion
Western
TOTAL

Basalt
64%
57%
17%
46

Obsidian
10%
21%
38%
21

Chert Chalcedonay
13%
5%
11%
7%
38%
0%
19
4

Rhyolite
3%
4%
3%
3

Glass
3%
0%
3%
2

Quartzite
3%
0%
0%
1

TOTAL
39
28
29
96

AMERICAN ANTIQUITY

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

274

[Vol. 78, No. 2, 2013

Figure 4. Elemental concentration results for the Berkeley and Gila River Indian Community Material Sciences
Laboratory analyses. Includes only sources with more than five examples.

ized by the Archaeological XRF Laboratory as part

of several prior investigations (Loendorf and Fertelmes 2012). GRIC-MSL independently determined source locations for these artifacts, which coincided with prior results in 104 of the 105 cases.
Although differences exist in the elemental concentrations calculated by GRIC-MSL and Berkeley, they are not sufficient to affect the accuracy of
source assignments (Figure 4). These data demonstrate that the precision of the Tracer III-V is sufficient to consistently differentiate Southwestern obsidian sources (cf. Millhauser et al. 2011:3141).
Nonetheless, additional interlaboratory, interinstrument, and intermethod comparisons are necessary to fully assess the accuracy of concentration
results produced by the Bruker calibration software
(but see Speakman and Shackley 2013:1438).
Sacate EDXRF Results

Table 5 lists obsidian source proportions for the 131

artifacts that were subjected to EDXRF analysis.
Figure 5 shows locations for the obsidian quarries
that occur in the sample, as well as other known
deposits. The straight-line distance of sources alone
is a poor predictor of obsidian use (Figure 6; Pearsons r = .42; p = .23), suggesting that other factors also conditioned procurement patterns at
Sacate. Obsidian from 10 locations was identified

in the sample, but 92 percent of the material was

derived from just four sources. Sauceda obsidian,
which is located to the south, is the most common
type and accounts for nearly 65 percent of the collection. Vulture obsidian is the next most common
type, and the incidence of this material is higher
than has been previously documented along the
Middle Gila (Loendorf 2012:107114). Los
Vidrios, which is located to the south of Sauceda,
is the third most common stone and is also more
frequent than previously found in GRIC prehistoric
contexts (Loendorf 2012). Sand Tanks, also to the
south, is the fourth most common obsidian, and despite its close proximity to the Hohokam core area,
this material rarely occurs in previously documented prehistoric assemblages, including those
from the GRIC (Loendorf 2012:107; Shackley and
Tucker 2001).
Akimel OOdham
Socioeconomic Interactions

Sacate data suggest that a general trend toward

greater utilization of materials located to the
south of the Middle Gila occurred over the course
of the Historic period, and 93 percent of the collection from the western site segment is from southern sources (Sauceda, Sand Tanks, Los Vidrios, and
Los Sitios; Figure 7). Materials from the north and

Loendorf et al.]

OBSIDIAN ACQUISITION AT THE SACATE SITE

Table 5. Absolute and Relative Frequencies of Obsidian Sources Identified from Sacate by Segment of Site.

Source
Government Mountain
Superior
Gwynn Canyon?
Antelope Wells
Los Sitios del Aqua
Los Vidrios
Sauceda
Sand Tanks
Tank Mountains
Vulture

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

TOTAL

Direction
N
E
E
E
S
S
S
S
W
W

Central Segment
5%
0%
0%
0%
0%
10%
67%
8%
0%
10%
41

east of Sacate, including the nearby Superior obsidian, are rare or absent. Sources to the west, including the Tank Mountains and Vulture, decline
in the western segment, which has the most recent
remains. This possible decrease in materials from
the west in the latter part of the Historic period
matches the observation that the Pee Posh, who formerly lived in this direction, had immigrated to the
Middle Gila by this time. The patterning in these
data is thus consistent with the ethnohistorically
and ethnographically documented socioeconom-

Expansion Segment
4%
2%
2%
2%
2%
4%
52%
11%
4%
19%
50

Western Segment
0%
0%
0%
0%
2%
11%
82%
0%
3%
5%
40

275

TOTAL
4
1
1
1
1
10
85
9
3
16
131

ic interactions described above, which suggest that

the Akimel OOdham primarily obtained obsidian through exchange. Other observations suggest
that these interactions largely involved raw materials rather than completed arrowpoints.
For instance, the patterning in Sacate obsidian
debris and completed points suggests that raw materials were generally acquired rather than finished
products (Table 6). If projectile points commonly were deposited as completed objects, then material type frequencies for points should differ from

Figure 5. Obsidian locations in the Southwest and archaeological sites with comparative data. Source areas are shaded
based on their overall proportion in the Sacate collection.

AMERICAN ANTIQUITY

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

276

[Vol. 78, No. 2, 2013

Figure 6. Scatterplot of obsidian proportions and source distances from Sacate (GR-909).

those for manufacturing debris, and waste from

point production should not be present (Bayman
1995:50). However, the most common types
(Sauceda and Vulture) do not differ significantly
(Fishers Exact Test p = .5), and obsidian in various stages of reduction was recovered, including
unmodified nodules, cores, shatter, primary flakes,
secondary flakes, and tertiary flakes.
Furthermore, obsidian points are small portions
of complex systems (including the arrow, bow, and
archer) that must be tuned to effectively function
(Cotterell and Kamminga 1992; Loendorf 2012).
Therefore, it is unlikely that completed projectile
points or arrows were regularly exchanged, and in-

stead it is more probable that raw materials necessary for point manufacture were traded. Points
must be the correct size for projectile shafts, which
in turn need to be the proper draw length and stiffness for a given bow and archer. Moreover, arrows
of different masses will have different points of impact when fired from the same bow, and without
some form of standardization in the manufacturing process, projectiles will be inaccurate (Mason
1894:660). Consequently, customized arrows of
consistent sizes were carefully produced to match
the body size of individuals, and arrows or points
are not freely interchangeable among bows or
archers (Burns 1916; Rea 2007a; Russell 1908:96).

1
0.9

Central

0.8

Expansion

0.7

Western

0.6
0.5
0.4
0.3
0.2
0.1
0

Figure 7. Obsidian frequencies by direction of source and segment of Sacate (GR-909).

South

West

North

East

Loendorf et al.]

Table 6. Absolute and Relative Frequencies of Completed Points and Debitage

by Obsidian Sources Identified from Sacate and Segment of Site.

Source
Sauceda Mountains
Vulture
Sand Tanks
Los Vidrios
Government Mountain
Tank Mountains
Antelope Wells
Superior
Gwynn Canyon?
Los Sitios del Aqua
Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

OBSIDIAN ACQUISITION AT THE SACATE SITE

Debitage
70
64%
15
14%
9
8%
7
6%
3
3%
3
3%
1
1%
1
1%
1
1%
0
0%

Projectile
15
71%
1
5%
0
0%
3
14%
1
5%
0
0%
0
0%
0
0%
0
0%
1
5%

277

TOTAL
85
16
9
10
4
3
1
1
1
1

TOTAL
110
100%
21
100%
131
Source: Data derived from Fertelmes et al. 2012; Loendorf 2012; Loendorf and Fertelmes 2011; Marshall 2002; Mitchell and
Foster 2011; Mitchell and Shackley 1995; Peterson et al. 1997; Rice et al. 1998; and Shackley et al. 2011).

Previous researchers have suggested that prehistoric populations commonly exchanged finished
projectile points (Bayman 1995; Bayman and
Shackley 1999; Shackley 2005). Investigators have
also argued that the presence of completed points
made of materials that do not occur in manufacturing debris is evidence for trade (Bayman 1995;
Peterson et al. 1997). Another possibility that could
account for these points is that they were deposited
as the result of conflict. Although this potential formation process is not generally considered by prehistorians, numerous attacks by nonlocal groups
were documented in the Akimel OOdham calendar
stick records (Russell 1908). As an example, a Pee
Posh village located approximately 3 km west of
Sacate was attacked, looted, and burned by the
Quechan on August 31, 1857 (Kroeber and
Fontana 1986). Hypothetically, this event should
have resulted in the deposition of completed arrowpoints. However, neither of these mechanisms appears to have been a primary source for
points from GR-909. The observations presented
in this section do not rule out the possibility that
points were deposited through warfare or trade and
instead only suggest that the majority of the analyzed collection was deposited as a result of various activities undertaken by Sacate residents.
Prehistoric Hohokam Socioeconomic
Interaction Patterns

In order to further compare both spatial and temporal variation in obsidian use at Sacate, Table 7

presents source proportions for collections from

the Hohokam region of southern Arizona. As a geographical reference point, the sites and obsidian
sources are organized based on distance from
Sacate.
At prehistoric sites in southern Arizona, the direction of the source has a greater effect on raw material utilization, and obsidian proportions are weakly correlated with source distances (Bayman
1995:49; Bayman and Shackley 1999; Loendorf
2012:107115; Mitchell and Shackley 1995:299;
Rice et al. 1998). By the Late Classic (ca. A.D.
13201450), obsidian frequencies differ significantly between some adjacent areas, such as the
Middle Gila and Lower Salt rivers, suggesting that
different Hohokam communities maintained separate trade contacts during this time (Loendorf
2010; Simon and Gosser 2001). This variation suggests that Late Classic populations were not
closely economically integrated across the Hohokam region of southern Arizona (Simon and
Gosser 2001). The patterning in obsidian acquisition, instead, suggests that the strongest socioeconomic ties among Classic period communities
were between sites on the same rivers, and exchanging food for items such as obsidian provided a mechanism for redistributing water-dependent
resources (Loendorf 2012:113).
As can be seen in Table 7, temporal patterns
are also apparent in the obsidian from prehistoric
sites. These data suggest that diachronic trends
identified at Sacate began before the advent of the
Historic period (Figure 8). For example, the ten-

92

Sacate
Sand
Sauceda
Period Distance (km) Tanks Superior Mts
Historic
N/A
7%
1%
65%
Preclassic
8
1%
60%
22%
Classic
11
5%
8%
63%
Preclassic
19
0%
51%
15%
Classic
22
0%
20%
61%
Classic
35
0%
10%
30%
Classic
35
0%
23%
30%
Preclassic
38
0%
2%
3%
Classic
38
0%
3%
48%
Classic
41
0%
18%
57%
Classic
44
0%
7%
46%
Preclassic
47
0%
95%
1%
Preclassic
66
0%
0%
85%
Classic
84
0%
3%
79%
Preclassic
87
0%
0%
0%
Classic
95
0%
48%
10%
Classic
107
0%
40%
12%
Classic
144
1%
0%
13%

87
Vulture
12%
5%
5%
3%
0%
27%
47%
27%
17%
0%
1%
1%
4%
7%
55%
0%
2%
0%

120
Los
Vidrios
8%
0%
7%
0%
2%
4%
0%
4%
22%
0%
0%
0%
1%
3%
0%
0%
0%
14%

198
Burro
Creek
0%
0%
1%
0%
0%
0%
0%
0%
0%
0%
0%
0%
3%
0%
0%
0%
0%
0%

200

260

267

Tank Government RS Hill/

Mts
Mtn
Sitgreaves
2%
3%
0%
0%
2%
2%
0%
3%
4%
0%
13%
0%
0%
4%
0%
1%
22%
0%
0%
0%
0%
0%
42%
20%
0%
6%
0%
0%
23%
0%
0%
29%
0%
0%
0%
0%
0%
0%
0%
0%
4%
0%
0%
31%
11%
0%
22%
0%
0%
23%
7%
2%
36%
0%

221

284

287

Cow Partridge Mule/ Unknown/

Canyon Creek Antelope
Other
0%
0%
0%
2%
2%
0%
1%
0%
0%
0%
0%
1%
0%
0%
0%
18%
0%
0%
2%
12%
0%
4%
0%
3%
0%
0%
0%
0%
0%
7%
1%
0%
0%
2%
0%
3%
0%
0%
0%
1%
0%
0%
14%
2%
0%
0%
1%
2%
0%
0%
0%
7%
0%
0%
1%
1%
1%
2%
0%
0%
3%
0%
16%
2%
5%
0%
13%
2%
1%
0%
31%
0%

239

Sample
Size
131
299
76
39
51
220
43
73
103
23
137
137
75
67
122
91
99
214

Middle Gila Data

*Average
Preclassic
13
1%
56%
18%
4%
0%
0%
0%
7%
1%
1%
0%
0%
9%
338
*Average
Classic
16
3%
14%
62%
3%
5%
1%
0%
4%
2%
0%
0%
1%
7%
271
*Combined Average Both
15
2%
35%
40%
3%
2%
0%
0%
5%
2%
0%
0%
1%
8%
609
*Nearby sites included in the middle Gila average.
(Data from Fertelmes et al. 2012; Loendorf 2012; Loendorf and Fertelmes 2011; Marshall 2002; Mitchell and Foster 2011; Mitchell and Shackley 1995; Peterson et al. 1997; Rice et al.
1998; Shackley et al. 2011.)

Collection
Sacate
*Snaketown
*ELXP
*GR-522 Locus D
*GR-522 Locus A
Pueblo Grande
Rowley
Los Colinas
Los Colinas
Crismon
Casa Grande
Grewe
Gatlin
Brady Wash
Palo Verde
Tonto Arm
Salt Arm
UIR

48

Sacate Distance (km)

Table 7. Relative Frequencies of Obsidian Artifacts at Southern Arizona Archaeological Sites by Source.

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM
278
AMERICAN ANTIQUITY
[Vol. 78, No. 2, 2013

Loendorf et al.]

OBSIDIAN ACQUISITION AT THE SACATE SITE

279

90%
Preclassic

80%

Classic

70%

Central

60%

Expansion

50%

Western

40%

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

30%
20%
10%
0%
Tank
Mountains

Los Vidrios

Vulture

Government

Sauceda

Sand Tanks

Superior

Figure 8. Obsidian proportions for the seven most common types by time period and segment of Sacate (GR-909). The
Preclassic category includes data from Snaketown and Locus D of GR-522; Classic period data include ELXP and Locus
A of GR-522.

dency for greater reliance on southern Arizona

sources started during the Classic period throughout the Hohokam core area (Fertelmes et al. 2012;
Marshall 2002:132133; Shackley 2005). At the
same time, use of sources located to the north and
east of the Phoenix Basin declined. Superior obsidian, for example, was the most common material (60 percent) identified at the Preclassic
Snaketown site within the GRIC, but this stone
rarely occurs in Classic contexts along the Middle Gila, when Sauceda obsidian is consistently
more abundant (Bayman and Shackley 1999;
Loendorf 2012; Shackley 2005). Shackley (2005)
argues that access to Superior was restricted by
the Salado. However, a similar pattern occurs in
the Tonto Basin, which is considered to be the
heartland of the Salado, where use of Superior obsidian also declined during the Classic period (Rice
et al. 1998). Rice et al. (1998) suggest two explanations for this pattern. First, the source was
depleted. Second, a village appropriated exclusive
use of the material. The first suggestion is improbable because substantial obsidian deposits remain at the source today. While the second is possible, the site that cut off access has not been identified. Another possibility is that hunter-gatherers,
such as the Nn who occupied the source area

during the Historic period, moved into the area and

restricted access (Loendorf 2010).
Finally, these data also suggest that the Akimel
OOdham did not generally acquire points by collecting them at prehistoric sites, as some researchers have suggested (e.g., Russell 1908). If
the Akimel OOdham obtained a substantial proportion of their points from nearby sites, then obsidian frequencies at Sacate should reflect the underlying proportions at prehistoric sites where the
artifacts were collected. However, the frequencies
for prehistoric obsidian data differ substantially
from those for the Sacate sample (see Table 7 and
Figure 8). In particular, the near-complete absence
of Superior obsidian and high incidence of Sauceda obsidian are unlikely to have have occurred if
points were commonly obtained from prehistoric sites (Fishers Exact Test p < .01).
HohokamAkimel OOdham
Continuum Discussion

Until recently, most outside observers who considered the relationship between the Hohokam and
the Akimel OOdham focused on differences between historical ki-kuh and Classic period adobe
structures. Based on this evidence, they concluded

280

AMERICAN ANTIQUITY

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

that the Historic populations were recent migrants

to the Phoenix Basin (Ezell 1961; Fewkes 1912;
Russell 1908; Seymour 2011). However, Emory
is an exception; he wrote:
My own impression, and it is stated so in my
journal, is that the many ruins we saw on the
Gila might well be attributed to Indians of the
races we saw in New Mexico, and on the Gila
itself. I mean by the last, the Pimos [Akimel
OOdham], who might easily have lost the art
of building adobe and mud houses. In all
respects, except their dwellings, they appeared
to be of the same race as the builders of the
numberless houses now level with the ground
of the Gila River [Emory et al. 1848:133].

At the time this conclusion was almost universally rejected, and it was not until the 1960s that the
possibility of a Hohokam and Akimel OOdham
continuum gained more widespread acceptance.
Ezell (1963), for example, examined material cultural traits and found both similarities and differences
between prehistoric and Historic remains. Based on
this review, Ezell (1963:62) provisionally concluded
that the Akimel OOdham were related to the Hohokam. By the 1990s, broader agreement was
reached among Southwestern archaeologists that the
Akimel OOdham are cultural descendants of the
Hohokam (Gilpin and Phillips 1998:117). Despite
this general consensus, some researchers continue
to argue that the Akimel OOdham are recent migrants to the Phoenix Basin (e.g., Rea 2007b), and
ambiguity regarding the nature of their relationship
remains. For example, Hegmon et al. (2008) suggest that although some people remained in the
Phoenix Basin after the Classic period, the association between prehistoric and Historic groups is difficult to trace using archaeological evidence. Most
recently, based on a comprehensive data review, Seymour concludes that historical populations either
replaced the Hohokam, absorbed them, or represent
a modified form of them (2011:296).
However, this analysis has identified continuities between prehistoric and historical material culture that suggest that the Akimel OOdham are the
direct cultural descendants of the Hohokam. Furthermore, many additional lines of evidence also
exist. First, the Akimel OOdham practiced a dispersed rancheria settlement pattern that is similar
to Preclassic Hohokam strategies (Fish 1989:21;

[Vol. 78, No. 2, 2013

Seymour 2011:198209). Second, the Akimel

OOdham made red-on-buff pottery, which has
close similarities in manufacturing technique
(i.e., paddle and anvil), clay, and design with prehistoric ceramics (Ezell 1963). Third, in contrast
to other surrounding groups, the Akimel OOdham
were dependent on large-scale irrigation agriculture, and their canal system shares close correspondences with Hohokam agricultural strategies
(Ravesloot et al. 2009; Webb 1959:121126;
Woodson 2010:45). Fourth, most Akimel OOdham believe that they are Hohokam descendants
(Shaw 1994; Webb 1959:53). Fifth, Classic period sites play a prominent role in Akimel OOdham
traditions, and the close similarities between the
archaeological record and these stories are unlikely
to have occurred by coincidence (Loendorf 2012;
Teague 1993).
Finally, rather than resulting from the migration
of outside ethnic groups, the architectural changes
that occurred in the Phoenix Basin are consistent
with responses to variation in the raw materials
available for construction (Loendorf 2012:
123126). Instead of an abrupt disruption in construction techniques, house types that used progressively less wood were built over time from the
Preclassic to Classic periods, which suggests that
the architectural changes were a response to environmental constraints (Abbott and Foster
2003:2630; Ezell 1961:49; Sayles 1938:7980).
Preclassic pithouses were built largely of wood and
frequently had multiple support posts, whereas Late
Classic adobe structures used no wood in the walls
and commonly only had one main roof support post
(Haury 1976:4674). A massive down-cutting event
occurred at the transition between these two periods around A.D. 1150 (Waters and Ravesloot
2000, 2001). This event impacted riparian habitats
along the Gila and Salt rivers, which eliminated
many of the trees in the area (Waters and Ravesloot 2001:292). Macrobotanical evidence also suggests that fewer trees were locally available during the Classic period (e.g., Kwiatkowski 2003:57).
Waters and Ravesloot (2001:292) have suggested that riparian habitats along the Middle Gila
recovered by the Historic period. As a consequence,
more trees were available that could have been used
for construction and other purposes. Historic kikuh are similar to Preclassic pithouses, and they
both require substantially more wood for con-

Loendorf et al.]

OBSIDIAN ACQUISITION AT THE SACATE SITE

struction than adobe structures (Ezell 1963; Sayles

1938; Seymour 2011:97138). These similarities
led Haury to conclude that the Pima house, in my
opinion, represents the retention of the old Hohokam architectural idiom, a not insignificant argument in the favor of HohokamPima continuity (1976:72).

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

Conclusions

Multiple lines of evidence suggest that the Akimel

OOdham primarily obtained obsidian in the form
of raw materials acquired through trade with the Tohono OOdham and Pee Posh. They appear to have
largely employed this obsidian to manufacture
flaked-stone points, which were used to tip arrows
designed for use in warfare. While some other goods
may have been more readily traded among different people, the employment of flaked-stone points
in warfare is expected to have restricted patterns
of exchange for the materials, including obsidian,
necessary to manufacture these weapons. Instead,
it appears that these items were generally only exchanged among more closely allied peoples.
Data from Sacate suggest that prehistoric temporal trends toward greater reliance on obsidian
sources to the south of the Middle Gila culminated during the Historic period, which appears to be
the result of changes in socioeconomic interactions
and the well-documented conflicts that occurred
during this interval. Access to northern and eastern obsidians including the San Francisco volcanics
and Superior sources was restricted by intervening Nn and Gn populations. Meanwhile, alliances between the Tohono OOdham and the Pee
Posh allowed continued access to raw materials to
the south and west. The fact that the decline of obsidian from the north and east begins during the
Classic period suggests that the Nn or similar
groups moved into southern Arizona at this time.
This possibility is supported by recent research on
Nn migration, as well as both Nn and Akimel
OOdham social memories (Loendorf 2012; Seymour 2011).
Examination of Akimel OOdham obsidian use
at Sacate suggests that long-term cultural trends
that began during prehistory continued unabated
after the advent of the Historic period in A.D. 1694.
These include a general pattern of increasing reliance on obsidian for point manufacture and con-

281

sistencies in point design between the Classic and

Historic periods. More important, trends in the interaction patterns within southern Arizona continue
unbroken between the prehistoric and Historic periods. This continuation of socioeconomic relationships and technological traditions provides additional evidence that the Historic Akimel OOdham are the cultural descendants of the prehistoric
Hohokam. This conclusion is supported by multiple lines of evidence including similarities in architecture, subsistence practices, ceramics, and settlement strategies.

Acknowledgments. This research was conducted as part of the

Pima-Maricopa Irrigation Project, which is funded by the Department of the Interior, U.S. Bureau of Reclamation, under
the Tribal Self-Governance Act of 1994 (P.L. 103-413), for the
development of a water delivery system utilizing Central Arizona Project water. This article would not have occurred without the dedication and support of J. Andrew Darling. His efforts resulted in the establishment of the Gila River Indian
Community (GRIC) Material Sciences Laboratory, and he
provided important comments on this research. We recognize
the hard work of the field crew who recorded Sacate. These
investigations were overseen by John Ravesloot, who also developed the GRIC Cultural Resource Management Program.
Brenda Randolph made important contributions to this study.
Bruce Kaiser helped design our obsidian methodology, and his
encyclopedic knowledge of energy-dispersive X-ray fluorescence analysis formed the foundation of our research program
at the GRIC Material Sciences Laboratory. Many thanks go to
Manuel Palacios-Fest for his Spanish translation of the abstract
and M. Kyle Woodson for providing insightful input on the
manuscript. Comments provided by two anonymous reviewers and E. Christian Wells were essential for improving and
clarifying the concepts in this article. Lynn Simon provided invaluable assistance by making maps and completing geographic information system analyses.

References Cited

Abbott, David R., and Michael S. Foster

2003 Site Structure, Chronology, and Population. In Centuries of Decline During the Hohokam Classic Period, edited by David R. Abbott, pp. 2447. University of Arizona
Press, Tucson.
Audubon, John W.
1906 Audubons Western Journal 18491850: Being the Ms.
Record of a Trip from New York to Texas, and an Overland
Journey Through Mexico and Arizona to the Gold Fields
of California. Arthur H. Clark Company, Cleveland.
Ballenger, Jesse A. M., and John D. Hall
2011 Obsidian Procurement, Technology, and Exchange in
the Vicinity of Superior, Arizona. In The U.S. 60 Archaeological Project: Early Agricultural, Formative, and Historical-Period Use of the Upper Queen Creek Region, edited by Robert M. Wegener, Michael P. Heilen, Richard
Ciolek-Torello, and John D. Hall, pp. 137156. Technical
Series, 92. Statistical Research, Tucson.

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

282

AMERICAN ANTIQUITY

Bancroft, Hubert H.
1886 The Native Races of the Pacific States of North
America, Vol. I: Wild Tribes. A. L. Bancroft and Company, San Francisco.
Bartlett, John R.
1854 Personal Narrative of Explorations and Incidents in
Texas, New Mexico, California, Sonora, and Chihuahua.
D. Appleton and Company, London.
Basso, Keith H. (editor)
2004 Western Apache Raiding and Warfare. From the
Notes of Grenville Goodwin. University of Arizona Press,
Tucson.
Bayman, James M.
1995 Rethinking Redistribution in the Archaeological
Record: Obsidian Exchange at the Marana Platform
Mound. Journal of Anthropological Research 51:3763.
Bayman, James M., and M. Steven Shackley
1999 Dynamics of Hohokam Obsidian Circulation in the
North American Southwest. Antiquity 73:836845.
Burns, Michael
1916 Indian Troubles. In History of Arizona: Vol. III, edited by Thomas E. Farish, pp. 286355. Filmer Brothers Electrotype Company, San Francisco.
2012 The Only One Living to Tell: The Autobiography of a
Yavapai Indian. Edited by Gregory McNamee. University of Arizona Press, Tucson.
Cotterell, Brian, and Johan Kamminga
1992 Mechanics of Pre-Industrial Technology. Cambridge
University Press, New York.
Darling, J. Andrew
2011 S-cuk Kavick: Thoughts on Migratory Processes and
the Archaeology of OOdham Migration. In Rethinking Anthropological Perspectives on Migration, edited by Graciela S. Cabana and Jeffrey J. Clark, pp. 6883. University Press of Florida, Gainesville.
Darling, J. Andrew, John C. Ravesloot, and Michael R. Waters
2004 Village Drift and Riverine Settlement: Modeling
Akimel OOdham Land Use. American Anthropologist
106:282295.
DeJong, David H.
2009 Stealing the Gila: The Pima Agricultural Economy and
Water Deprivation, 18481921. University of Arizona Press,
Tucson.
2011 Forced to Abandon Our Fields: The 1914 Clay
Southworth Gila River Pima Interviews. University of Utah
Press, Salt Lake City.
Ellis, Christopher J.
1997 Factors Influencing the Use of Stone Projectile Tips:
An Ethnographic Perspective. In Projectile Technology, edited by Heidi Knecht, pp. 3774. Plenum Press, New York.
Emory, William H., James W. Abert, Philip St. George Cook, and
Abraham R. Johnston
1848 Notes of a Military Reconnaissance from Fort Leavenworth, in Missouri, to San Diego, in California, Including
Parts of the Arkansas Del Norte, and Gila Rivers. 31st U.S.
Congress, 1st Session, Senate Executive Document No. 7
(Serial No. 505). Wendell and Van Benthuysen, Washington, D.C.
Ezell, Paul H.
1961 The Hispanic Acculturation of the Gila River Pimas.
American Anthropologist (N.S.) 63(5), Pt. 2.
1963 Is There a HohokamPima Culture Continuum?
American Antiquity 29:6166.
1983 History of the Pima. In Southwest, edited by A. Ortiz,
pp. 149160. Handbook of North American Indians, Vol.
10, W. C. Sturtevant, general editor, Smithsonian Institution, Washington, D.C.

[Vol. 78, No. 2, 2013

Ferg, Alan, and Norm Tessman

1997 Two Archival Case Studies in Western Apache and
Yavapai Archaeology. In Vanishing River: Landscapes and
Lives of the Lower Verde Valley, the Lower Verde Archaeological Project, edited by Stephanie M. Whittlesey,
Richard Ciolek-Torello, and Jeffrey H. Altschul, pp.
214279. SRI Press, Tucson.
Fertelmes, Craig M., David R. Abbott, and M. Steven Shackley
2012 Obsidian Source Characterization at Las Colinas: Shifting Exchange Patterns During the Hohokam Sedentary to
Classic Transition. Kiva 77:281311.
Fewkes, Jesse W.
1912 Casa Grande, Arizona. In Twenty-Eighth Report of the
Bureau of American Ethnology, pp. 25180. U.S. Government Printing Office, Washington, D.C.
Fish, Paul R.
1989 The Hohokam: 1,000 Years of Prehistory in the
Sonoran Desert. In Dynamics of Southwest Prehistory, edited by Linda. S. Cordell and George J. Gumerman, pp.
1963. Smithsonian Institution Press, Washington, D.C.
Gilpin, Dennis, and David A. Phillips
1998 The Prehistoric to Historic Transition Period in Arizona, Circa A.D. 1519 to 1692. Report submitted to the Arizona State Historic Preservation Office, Arizona State Parks
Board, Phoenix.
Glascock, Michael G., and Jeffrey R. Ferguson
2012 Report on the Analysis of Obsidian Source Samples by
Multiple Analytical Methods. Archaeometry Lab, University of Missouri, Research Reactor Center, Columbia.
Grossman, Frank E.
1873 The Pima Indians of Arizona. In Annual Report of the
Board of Regents of the Smithsonian Institution Showing
the Operations, Expenditures, and Condition of the Institution for the Year 1871, pp. 407419. U.S. Government
Printing Office, Washington, D.C.
Hackenberg, Robert A.
1974 Aboriginal Land Use and Occupancy of the Pima-Maricopa Indians, Vol. II. Garland Publishing Inc., New York.
Haury, Emil W.
1976 The Hohokam: Desert Farmers and Craftsmen
Excavations at Snaketown. University of Arizona Press, Tucson.
Hegmon, Michelle, Matthew A. Peeples, Ann P. Kinzig,
Stephanie Kulow, Cathryn M. Meegan, and Margaret C.
Nelson
2008 Social Transformation and Its Human Costs in the Prehispanic U.S. Southwest. American Anthropologist
110:313324.
Jacoby, Karl
2008 Shadows at Dawn: A Borderlands Massacre and the
Violence of History. Penguin Press, New York.
Justice, Noel D.
2002 Stone Age Spear and Arrow Points of the Southwestern United States. Indiana University Press, Bloomington.
Kroeber, Clifton B., and Bernard Fontana
1986 Massacre on the GilaAn Account of the Last Major
Battle Between American Indians, with Reflections on the
Origin of War. University of Arizona Press, Tucson.
Kwiatkowski, Scott M.
2003 Evidence for Subsistence Problems. In Centuries of Decline During the Hohokam Classic Period at Pueblo
Grande, edited by David R. Abbott, pp. 4869. University of Arizona Press, Tucson.
Loendorf, Chris
2010 Regional and Temporal Variation in Obsidian Use Within the Hohokam Region. Journal of Arizona Archaeology
1:4759.

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

Loendorf et al.]

OBSIDIAN ACQUISITION AT THE SACATE SITE

2012 The HohokamAkimel Oodham Continuum: Sociocultural Dynamics and Projectile Point Design in the
Phoenix Basin, Arizona. Gila River Indian Community Anthropology Research Papers No. 5. University of Arizona
Press, Tucson.
Loendorf, Chris, and Craig M. Fertelmes
2011 EDXRF Source Provenance Analysis of Obsidian Artifacts from University Indian Ruin (AZ BB:9:33 [ASM]),
Tucson, Arizona. pXRF Technical Report 2011-04. Gila River Indian Community, Material Sciences Laboratory,
Sacaton, Arizona.
2012 Pima-Maricopa Irrigation Project Obsidian Sourcing
Analysis: Methodology and Research Design. P-MIP
Technical Report 2010-06. Gila River Indian Community Material Sciences Laboratory, Sacaton, Arizona.
Loendorf, Chris, and Glen E. Rice
2004 Projectile Point Typology: Gila River Indian Community, Arizona. Gila River Indian Community Anthropology Research Papers No. 2. University of Arizona Press,
Tucson.
Marshall, John T.
2002 Obsidian and the Northern Periphery: Tool Manufacture,
Source Distribution, and Patterns of Interaction. In Phoenix
Basin to Perry Mesa: Rethinking the Northern Periphery,
edited by Mark R. Hackbarth, Kelly Hays-Gilpin, and Lynn
Neal, pp. 121138. Arizona Archaeologist No. 34. Arizona
Archaeological Society, Phoenix.
Mason, Otis T.
1894 North American Bows, Arrows, and Quivers. Government Printing Office, Washington, D.C.
Millhauser, John K., Enrique Rodrguez-Algra, and Michael D.
Glascock
2011 Testing the Accuracy of Portable X-Ray Fluorescence
to Study Aztec and Colonial Obsidian Supply at Xaltocan,
Mexico. Journal of Archaeological Science 38:31413152.
Mitchell, Douglas R., and Michael S. Foster
2011 Salt, Seashells, and Shiny Stones: Prehistoric Hohokam
Resource Exploitation in the Papagueria and Northern Gulf
of California. Journal of Arizona Archaeology 1:176184.
Mitchell, Douglas R., and M. Steven Shackley
1995 Classic Period Hohokam Obsidian Studies in Southern Arizona. Journal of Field Archaeology 22:291304.
Ortiz, Alfonzo
1983 Key to Tribal Territories (map). In Southwest, edited
by A. Ortiz, p. IX. Handbook of North American Indians,
Vol. 10, W. C. Sturtevant, general editor, Smithsonian Institution, Washington, D.C.
Peterson, Jane D., Douglas R. Mitchell, and M. Steven Shackley
1997 The Social and Economic Contexts of Lithic Procurement: Obsidian from Classic-Period Hohokam Sites.
American Antiquity 62:231259.
Randolph, Brenda G., J. Andrew Darling, Chris Loendorf, and
Bonny Rockette
2002 Historic Piman Structure and the Evolution of the Sacate
Site (GR-909), Gila River Indian Community. In Visible Archaeology on the Gila River Indian Reservation: Papers
Presented at the 67th Annual Meeting of the Society for
American Archaeology, March 2124, 2002, Denver,
compiled by Tom Herrschaft, pp. 166195. P-MIP Report
No. 21. Gila River Indian Community Cultural Resource
Management Program, Sacaton, Arizona.
Ravesloot, John C.
2007 Changing Views of Snaketown in a Larger Landscape.
In The Hohokam Millennium, edited by Suzanne K. Fish
and Paul R. Fish, pp. 9198. School for Advanced Research
Press, Santa Fe, New Mexico.

283

Ravesloot, John C., J. Andrew Darling, and Michael R. Waters

2009 Hohokam and Pima-Maricopa Irrigation Agriculturalists: Maladaptive or Resilient Societies? In The Archaeology of Environmental Change, edited by Christopher T.
Fisher, J. Brett Hill, and Gary M. Feinman, pp. 232245.
University of Arizona Press, Tucson.
Ravesloot, John C., and Stephanie M. Whittlesey
1987 Inferring the Protohistoric Period in Southern Arizona.
In The Archaeology of the San Xavier Bridge Site (AZ
BB:13:14), Tucson Basin, Southern Arizona, edited by John
C. Ravesloot, pp. 8198. Archaeological Series No. 171.
Arizona State Museum, University of Arizona, Tucson.
Rea, Amadeo M.
1998 Folk Mammalogy of the Northern Pimas. University
of Arizona Press, Tucson.
2007a Wings in the Desert: A Folk Ornithology of the Northern Pimas. University of Arizona Press, Tucson.
2007b The Poorwill in Pima Oral Tradition. Archaeology
Southwest 21(1):7.
Rice, Glen E., Arleyn W. Simon, and Chris Loendorf
1998 Production and Exchange of Economic Goods. In
A Synthesis of Tonto Basin Prehistory: The Roosevelt Archaeology Studies, 1989 to 1998, edited by Glen E. Rice,
pp. 105130. Roosevelt Monograph Series, 12; Anthropological Field Studies, 41. Arizona State University,
Tempe.
Russell, Frank
1908 The Pima Indians. Twenty-Sixth Annual Report of the
Bureau of American Ethnology, 19041905. Smithsonian
Institution, Washington, D.C.
Sayles, E. B.
1938 Houses. In Excavations at Snaketown: Material Culture, pp. 5992. Medallion Papers No. 25. Gila Pueblo,
Globe, Arizona.
Seymour, Deni J.
2011 Where the Earth and Sky Are Sewn Together:
Sobaipuri-OOdham Contexts of Contact and Colonialism.
University of Utah Press, Salt Lake City.
Shackley, M. Steven
1988 Sources of Archaeological Obsidian in the Southwest:
An Archaeological, Petrological, and Geochemical Study.
American Antiquity 53:752772.
1990 Early Hunter-Gatherer Procurement Ranges in the
Southwest: Evidence from Obsidian Geochemistry and Lithic Technology. Unpublished Ph.D. dissertation, Department
of Anthropology, Arizona State University, Tempe.
1995 Sources of Archaeological Obsidian in the Greater
American Southwest: An Update and Quantitative Analysis. American Antiquity 60:531551.
2005 Obsidian: Geology and Archaeology in the North American Southwest. University of Arizona Press, Tucson.
Shackley, M. Steven, and Jon Daehnke
2004 Source Provenance of Obsidian Artifacts from Various Contexts on the Gila River Indian Community Land,
Central Arizona. Berkeley Archaeological XRF Lab, University of California, Berkeley.
Shackley, M. Steven, Elizabeth Eklund, and Caroline Ogasawara
2011 Source Provenance of Obsidian Artifacts from Crismon
Ruin, AZ U:9:173 (ASM), Mesa, Arizona: Appendix G. In
Crismon Ruin: A Hohokam Settlement at the Head of the
Lehi Canal System, edited by T. Kathleen Henderson, pp.
479482. Anthropological Papers No. 44. Center for
Desert Archaeology, Tucson.
Shackley, M. Steven, and David B. Tucker
2001 Limited Prehistoric Procurement of Sand Tank Obsidian. Kiva 66:345374.

Delivered by http://saa.metapress.com
Society for American Archaeology - American Antiquity access (392-89-746)
IP Address: 64.16.57.4
Friday, April 19, 2013 1:20:39 PM

284

AMERICAN ANTIQUITY

Shaw, Anna M.
1994 A Pima Past. University of Arizona Press, Tucson.
Silliman, Stephen W.
2005 Obsidian Studies and the Archaeology of 19th Century California. Journal of Field Archaeology 30:7594.
Simon, Arleyn W., and Dennis C. Gosser
2001 Conflict and Exchange Among the Salado of Tonto
Basin: Warfare Motivation or Alleviation? In Deadly
Landscapes: Case Studies in Prehistoric Warfare, edited
by Glen E. Rice and Stephen A. LeBlanc, pp. 219238. University of Utah Press, Salt Lake City.
Speakman, Robert J., and M. Steven Shackley
2013 Silo Science and Portable XRF in Archaeology: A Response to Frahm. Journal of Archaeological Science
40:14351443.
Spier, Leslie
1933 The Yuman Tribes of the Gila River. University of Chicago Press, Chicago.
Szuter, Christine R.
1991 Hunting by Prehistoric Horticulturists in the American Southwest. Garland Press, New York.
Teague, Lynn S.
1993 Prehistory and the Traditions of the Oodham and Hopi.
Kiva 58:435454.
Vint, James M.
2005 An Instance of Violence Along Cienega Creek: The
Cienega Creek Burials (AZ EE:2:480 [ASM]) and a
Study of Protohistoric Projectile Technology in Southeastern
Arizona. Technical Report No. 2005-101. Center for
Desert Archaeology, Tucson.
Waters, Michael R., and John C. Ravesloot
2000 Late Quaternary Geology of the Middle Gila River, Gila
River Indian Reservation, Arizona. Quaternary Research
45:4957.
2001 Landscape Change and the Cultural Evolution of the
Hohokam Along the Middle Gila River and Other River Valleys in South-Central Arizona. American Antiquity
66:285299.
Webb, George
1959 A Pima Remembers. University of Arizona Press,
Tucson.
Wells, E. Christian
2006 From Hohokam to OOdham. The Protohistoric Occupation of the Middle Gila River Valley, Central Arizona.
Gila River Indian Community, Anthropological Research
Papers No. 3. University of Arizona Press, Tucson.
Wells, E. Christian, Glen E. Rice, and John C. Ravesloot
2004 Peopling Landscapes Between Villages in the Middle
Gila River Valley of Central Arizona. American Antiquity 69:627652.
Wilson, John P.
(Forthcoming) Peoples of the Middle Gila: A Documentary
History of the Pimas and Maricopas, 1500s1945. Gila River Indian Community Archaeological Research Paper
No. 6. University of Arizona Press, Tucson.
Woodson, M. Kyle
2010 The Social Organization of Hohokam Irrigation in the
Middle Gila River Valley, Arizona. Unpublished Ph.D. dissertation, Department of Anthropology, Arizona State
University, Tempe.

Note

[Vol. 78, No. 2, 2013]

1. The primary components of the Tracer III-V are an Xray tube, an X-ray detector, and a pulse processor. The X-ray
tube is equipped with a rhodium target, and it emits a collimated 4.0-mm-diameter beam. The instrument was set to operate at 40 keV and 12 A using an external power supply. A
beam filter composed of 304 m of aluminum, 25 m of titanium, and 152 m of copper was placed between the tube and
the sample. This filter provides efficient detection of the fingerprint elements (i.e., rubidium, strontium, yttrium, zirconium, and niobium) used in obsidian studies (Shackley 1988,
1995, 2005). The detector is a Peltier-cooled SiPIN diode sensor. This device has a resolution of approximately 170 eV
Full Width at Half Maximum for 5.9 keV X-rays (at 1,000
counts per second) in an area of 7.0 mm2. The pulse processer
partitions X-ray data into 2,048 channels (20 eV/channel),
with each channel counting the number of X-ray pulses gathered in that specific energy window during the 180-second
analysis time. These multichannel data are stored by the
S1PXRF software program. Partitioned pulse data were normalized using the inelastic (or Compton) peak of the rhodium
backscatter at 18.219.2 keV. Normalized pulse data were
converted to parts per million values using a calibration program that incorporates elemental results for 40 reference samples that have a wide range of compositional variation. Concentration values for the reference obsidian were determined
by the Archaeometry Laboratory, University of Missouri Research Reactor, using neutron activation analysis, microwave
digestion inductively coupled plasma mass spectrometry, and
laser ablation inductively coupled plasma mass spectrometry
(Glascock and Ferguson 2012). Provenance assignments were
assessed through reference to known source samples including
a K-means cluster analysis of the parts per million values for
rubidium, strontium, yttrium, zirconium, and niobium. To examine group fitness, elemental concentrations were compared
using two-way scatterplots. Statistical analyses were completed in PASW Statistics 18 (Release 18.0.0). Source nomenclature is adopted from Shackley 1988, 1995, 2005 (see also
http://www.swxrflab.net/swobsrcs.htm).

Submitted April 24, 2012; Revised June 14, 2012; Accepted

August 12, 2012.