Hohokam palettes from the Gila Bend region

Citation metadata

Date: Summer-Autumn 2010
From: Journal of the Southwest(Vol. 52)
Publisher: University of Arizona
Document Type: Article
Length: 2,606 words

Document controls

Main content

Full Text: 

Palettes, generally found in Hohokam archacological sites dating to the Colonial through the Sedentary Periods (AD 750-1150), have been studied by a number of researchers over the past eight decades. Much progress has been made toward understanding how they were made, how they were used, and what they meant to the Hohokam people, but much about them remains unknown. Although the best-studied specimens come from the area around Phoenix, many have been found in regions peripheral to the Phoenix Basin, such as the Gila Bend area to the southwest.

The excavations by Norton Allen and others at Gila Bcnd-area sites like Gatlin unearthed several dozen palettes, and the transfer of their private collections to museums has made it possible to analyze and compare those palettes to ones found elsewhere. This allows archaeologists to better understand the relationship between peripheral settlements, like those in the Gila Bend area, and sites in the Hohokam core.

The first archaeologist to define and systematically study palettes was Emil W. Haury. In 1934 and 1935, he and other researchers from the Gila Pueblo Foundation excavated at Snaketown, on the Gila River Indian Reservation south of Phoenix. In his discussion of palettes from the site, Haury (1937:124) proposed that these artifacts, which had been found mainly in cremations, evolved from simple stone slabs. He pointed out that unlike figurines, mosaic plaques, and copper bells, palettes have no equivalent in Mesoamerica. Instead, he inferred that this artifact type had developed in the Salt-Gila Basin (1937:125).

Haury excavated at Snaketown again in 1964 and 1955, and published those findings in 1976. His newer study of palettes revisited his previous research and added new evidence. Because the 1964-65 excavations were not focused on cremations (Haury 1976:286), he recovered significantly fewer palettes than he had in 1934-35. However, the relative frequencies of palettes in established phases deviated little from those presented in his earlier study. Haury re-evaluated his earlier study, finding that the vast majority of his early conclusions required only minor revisions. Nevertheless, while conceding that palettes "still remain something of an enigma" (1976:288), he went on to write that

The long-standing use of the trait and an evident local developmental sequence suggest ... that [the palette] was a Hohokam creation. Yet, the feeling lingers that the inspiration lies somewhere to the south, perhaps because the older examples were made of the harder materials, the shaping of which had been mastered by the stone craftsmen of ancient Mexico. But more importantly, the idea of incense burning, and the concept of sacrifice are southern in terms of the Southwest, and the palettes are connected with these two patterns. On a more specific form level of comparison, the treatment of the human figure as a palette during the Classic Period in Guerrero ... and during the Colonial Period of the Hohokam ... is certainly more than accidental. (Haury 1976:288-89)

Decades later, Jenny Adams (2002) observed that there is much variation among palettes with respect to form and function. This variation has led to a great deal of confusion when it comes to identifying them. One problem is that the definition of a palette is very imprecise, meaning that even the most crudely shaped small piece of ground stone can be termed a palette. Three researchers have however greatly helped in resolving the definitional problem. Stephen H. Lekson (1990, 1991) has contributed the most to the documentation of Hohokam palettes. When he began his research project in the late 1980s, only the palettes from Snaketown had been looked at in any great detail. The ultimate goal of Lekson's study was to track spatial and temporal variability in palette decoration and form. He set out to make a scale, plan-view drawing of any palette he could find, which resulted in renderings of more than one thousand palettes--almost every one that had been unearthed up to that time. He decided to draw the palettes because the faint, intricate border designs often found on palettes generally do not show up well in photographs (see figure 1).

Using Lekson's drawings, Karl Anna Krueger (1993) made the first major attempt at a synthetic analysis of Hohokam palettes by qualitatively looking at aspects of their physical form, design, and possible uses and social functions. Building on Krucger's study, I applied statistical methods in an attempt to synthesize all known previous research on palettes and explore their production, distribution, and use-lives. I also proposed a new name (tablet) for the artifact class, along with a more specific definition:

The term "palette" does not accurately describe the many and varied uses of this artifact class within Hohokam society. "Palette" implies painting, or at the very least pigment processing. Only a handful of palettes have trace amounts of pigment on them (1-2%), and less than one-quarter (15%) of all palettes show any signs of grinding, much less pigment processing, in their basins. Therefore, calling these artifacts "palettes" is a misnomer. I suggest that "tablet" be used instead. Not only is it more accurate with respect to general artifact morphology (thin, flat, and rectangular), but it is also more neutral with respect to implied use, which allows for a broader range of interpretations by archaeologists (White 2004:70).

The efforts of later researchers, building on work done by archaeologists like Haury and offering their own insights, have made it possible to talk about palettes in much more informed and specific ways.


There are fifty known palettes for the entire Gila Bend area. Artifact provenience is problematic at best, which is not uncommon for palettes (White 2004). The site of origin is known for most of them, but next to nothing about the context from which they were removed. The palettes have been recovered from sites of various sizes and ages, including Gatlin, Homestead, Citrus, Tovrea Ranch, Escuela, Komertkwutch, Four Mile, Six Mile, Seven Mile, and Twelve Mile (Lekson 1990). Sixty percent of these palettes are from Gatlin, Homestead, and Twelve Mile, with the remainder spread fairly evenly among the remaining seven sites.

The palettes from the Gila Bend area are primarily housed in three locations: the Arizona State Museum (Norton Allen Collection), the Pomona College Museum of Art (Parker Collection), and the Peabody Museum of Archaeology and Ethnology at Harvard University. I collectively refer to all known palettes collected from sites in the Gila Bend area as the Gila Bend palettes unless a specific site name is mentioned.


I was interested in examining the palettes in terms of shape, how they were made, how they were decorated, how they were used, and when and where they were found. When combined and analyzed, these different facets of palettes provide a wealth of information about the people who owned them and their place in the Hohokam world.

Although some palettes are shaped like humans or animals, most palettes are rectangular and have a border, much like a picture frame. The portion of the palette surrounded by the border is termed the basin. Depending on time period and location, how palettes are decorated varies, but there are some basic design elements worth mentioning, as seen in figure 2.

* End Sculpture: any type of sculptural embellishment that is attached to at least one end (short side) of the palette, and sometimes both ends. Recorded examples include what look like decorated handles on one end, but more common are animal motifs: dogs, birds, snakes, and snake-attacking-bird (White 2004:84).

* Medial Groove: A line carved into the side of a palette, running parallel to the edges of the border and generally all the way around the palette. One groove is common, but multiple ones have been found. Successfully adding a medial groove requires great skill, because it can easily split the stone and ruin the palette.

* Interior/Exterior Notching: Small "tick marks" that are added to the inside or outside edge of the palette border.

A palette is decorated by carving designs into the border. Border designs come in two varieties: border motifs and corner motifs. Border motifs are recurring patterns that run all the way around the border, while corner motifs are distinct designs that show up only in the corners of the border. Corner motifs are mostly limited to basic geometric shapes (twenty-six are known), but border motifs can be quite complex. There are 204 known border motifs, ranging from interlocking geometric patterns to waves, stars, and animals (White 2004:app. A).

Palettes were made mainly during the Colonial through Sedentary Periods (AD 750-1150). The Colonial Period saw the expansion of Hohokam influence outside of the heartland of the Salt-Gila Basin into peripheral areas like Gila Bend, where new communities were being built, while the Sedentary Period saw the focus shift to growing established communities and building stronger links among them through trade networks, ballcourts, and irrigation canals.

Palettes made during the Colonial Period are quite distinct from those made during the Sedentary Period. Colonial palettes are carefully made, vary greatly in size, and have a basic border motif (one or more parallel lines) and no corner motifs. Sedentary palettes are less carefully made and are fairly standard in size, instead focusing heavily on border decoration. In fact, all corner motifs and 197 of the 204 border motifs (97%) are found on Sedentary Period palettes.

In both periods, there was probably a group of specialists who produced palettes. In the Colonial Period, they focused on making a smaller number of elaborate ones, while in the Sedentary Period they made a larger number of simple ones that individual people could have decorated themselves (White 2004:58). Using these criteria, I placed the fifty Gila Bend palettes into one of three temporal categories for analysis: Colonial (n = 7, 14%), Sedentary (n = 37, 74%), and Unknown (n = 6, 12%). Figure 3 shows examples of Colonial and Sedentary palettes from Gila Bend.

Compared to sites in the Salt-Gila Basin, the proportion of Sedentary palettes to Colonial palettes at Gila Bend sites is significantly higher. This is not unexpected, because peripheral sites like Gatlin were being settled in the Colonial Period and expanded in the Sedentary Period (Doyel 2008). What is also unsurprising is that, with one exception, the Gila Bend palettes show no signs of having been used for any particular purpose, apart from being included in a cremation. The one exception is a Colonial palette that exhibits a crust on its basin, which is generally thought to be part of a pyrotechnic display associated with the cremation process (Hawley 1937). Most Sedentary palettes show few signs of use, possibly due to their democratization, which seemingly involved a shift from being restricted ritual objects of high importance to things that anyone could own (White 2004).

The Gila Bend palettes also contain the largest number of unique border motifs outside of the Salt-Gila Basin. Considering the small number of palettes involved (50 out of more than 1,200 studied), that is a significant finding. Palettes from other peripheral areas tend to have only a handful of unique motifs (White 2004). During my original research project, I noticed that there were border motifs that appeared only on palettes from peripheral sites such as Gila Bend, which suggests that individuals at these sites were decorating palettes themselves and not communicating their designs to larger sites in the Salt-Gila Basin. The Gila Bend sites have eighteen such border motifs, more than any other area on the Hohokam periphery, along with one corner motif (figure 4). In addition, when I compared the Gila Bend palettes with those from other peripheral sites, I found that if a motif was found only at a peripheral site, it always occurred in the Gila Bend area as well. This suggests connections, either direct or indirect, among peripheral communities during the Preclassic, and also that information from these areas perhaps traveled along a "cultural beltway" that circumscribed the Hohokam core area.

Even with some data suggesting communication between peripheral sites that did not involve the Hohokam core area, several Gila Bend palettes make it very clear that--at least during the Colonial Period--the region was still closely tied to the core and probably acted more as a consumer of finished goods than an upstart colony looking to express an independent artistic sentiment. This inference is based on the fact that a rare type of palette, known as an "effigy palette" because it is generally shaped like an animal or person, appeared only in the Gila Bend area and in the Hohokam core area. All three examples from Gila Bend, which appear to be from the Colonial Period, have identical counterparts at the site of Snaketown, suggesting that the same craftsperson was involved in the creation of both palettes in each pair. Also, one Gila Bend palette from the Sedentary Period has an end sculpture in the form of large, coiled snakes. The only other known example of this type of sculpture occurred at Snaketown during the same period. All four pairings are shown in figure 5.


While palettes are still largely enigmas, much can be learned about the people who owned them and the cultural system in which they played a part by examining their shape, how they were made, how they were decorated, and where and when they are found in sites. Based on the evidence presented in this article, it appears that the Gila Bend area maintained a strong connection to the Hohokam core area during both the Colonial and Sedentary Periods, though that connection was somewhat stronger in the earlier period. That some border motifs are found on palettes from Gila Bend and other outlying sites, but not in the cultural core, suggests that these peripheral communities established independent relationships with each other, possibly to foster the exchange of ideas and goods outside channels controlled by the core.


I would like to thank Jannelle Weakly, of the Arizona State Museum, for supplying the photographs used in this article.


Adams, Jenny L. 2002 Ground Stone Analysis: A Technological Approach. Salt Lake City: University of Utah Press.

Doyel, David E. 2008 Edge Work: The Late Prehistory of the Gila Bend Frontier. In Fragile Patterns: The Archaeology of the Western Papagueria, edited by Jeffrey H. Altschul and Adrianne G. Rankin, pp. 233-51. Tucson: SRI Press.

Haury, Emil W. 1937 Stone Palettes and Ornaments. In Excavations at Snaketown: Material Culture, by Harold S. Gladwin, Emil W. Haury, E. B. Sayles, and Nora Gladwin, pp. 121-34. Medallion Papers No. 25. Globe, Ariz.: Gila Pueblo Foundation.

1976 The Hohokam: Desert Farmers & Craftsmen: Excavations at Snaketown, 1964-1965. Tucson: University of Arizona Press.

Hawley, F. G. 1937 Chemical Investigation of the Incrustation on Pottery Vessels and Palettes from Snaketown. In Excavations at Snaketown: Material Culture, edited by Harold S. Gladwin, Emil W. Haury, E. B. Sayles, and Nora Gladwin, pp. 282-89. Medallion Papers No. 25. Globe, Arizona: Gila Pueblo Foundation.

Krueger, Kari Anna 1993 Definitive Analysis of Hohokam Stone Palettes. Master's thesis. Department of Anthropology, Northern Illinois University, De Kalb.

Lekson, Stephen H. 1990 Palette Recording. Unpublished manuscript on file with the author.

1991 Ritual and the Hohokam Region. Paper presented at the Fifty-sixth annual meeting of the Society for American Archaeology, New Orleans.

Sedat, David W. 1972 The Parker Collection of Hohokam Artifacts. Master's thesis. Claremont Graduate School, Claremont, Calif.

White, Devin Alan 2004 Hohokam Palettes. Arizona State Museum Archaeological Series No. 196. Tucson: Arizona State Museum.

DEVIN ALAN WHITE holds an MA and PhD in anthropology from the University of Colorado, Boulder. He currently works as a photogrammetric scientist for Integrity Applications Incorporated, in Chantilly, Virginia. He is also a research associate at Crow Canyon Archaeological Center in Cortez, Colorado.

Source Citation

Source Citation   

Gale Document Number: GALE|A244279224