we have not yet recognized a clear pattern of any specific design or style overlapping another. Instead, it appears that in this area a few superimpositions were done during approximately the past thousand years in a seemingly random pattern. Possibly further study will yield information that is of more value for turning these clues into a relative chronology.
Patination
Many petroglyphs along the middle Columbia River and the lower Snake River were made by pecking through the dark patina on the basalt bedrock characteristic of these areas. This patina, sometimes called desert varnish, is a brown to black stain that colors exposed rock surfaces. It occurs most often on stones in hot, arid portions of the world. Some scientists suggest that this patina forms due to chemical weathering and leaching of iron and manganese oxides from the stone, while others hypothesize that airborne microorganisms oxidize these minerals and concentrate them on the rock surface. In either case, the process is a slow one and desert varnish takes considerable time to develop.
When a petroglyph is pecked or carved through the patina on a rock surface, it exposes the lighter colored interior stone and creates a negative image, with the paler petroglyph showing on an otherwise dark background. If conditions for patina development still exist after the petroglyph is made, the newly exposed surfaces gradually begin to acquire the desert varnish. After a long period the design will be repatinated; it will have essentially the same patina as the unaltered rock face. Although repatination of petroglyph designs does not provide an absolute age (since exposure, temperature, humidity, and other factors influence the rate of patina formation), petroglyphs repatinating differently on the same surface are useful for creating a relative chronology. In such cases, those that appear fresher are younger than those that have repatinated to an appearance closer to the original surface of the stone.
Differential patination has been used on petroglyphs near The Dalles and at Buffalo Eddy on the Snake River south of Lewiston, Idaho, to suggest relative chronologies. At The Dalles, petroglyphs in the pit and groove style invariably are heavily patinated and weathered. At two sites, these contrast with much fresher appearing designs pecked at a later time. The heavy repatination of all of these pit and groove petroglyphs is consistent with that noted in the Great Basin area of Utah, Nevada, and California. There the appearance of this art style has been dated between five and seven thousand years ago. While we cannot automatically assign an equally ancient age to the pit and groove petroglyphs at The Dalles, certainly they are the oldest in this locale.
At Buffalo Eddy, Nesbitt described two rock art styles (1968). A naturalistic style shows primarily mountain sheep, deer, and humans wearing horned headdresses, while a “graphic” style is composed of triangles, circles, dots, and lines arranged in geometric patterns. At Buffalo Eddy, the naturalistic petroglyphs are usually repatinated, some very heavily. In contrast, the graphic designs are reported as fresher looking and cut through the patina on the rock surface. In one instance graphic style designs may actually be superimposed on repatinated naturalistic designs. In this case, the naturalistic drawings of men and mountain sheep clearly seem older than the graphic geometric designs.
Weathering
Variations in the weathering of different designs are often used in conjunction with patination studies, but such variations can also be applied to pictographs, which are not affected by patination. Many Columbia Plateau pictographs show weathering differences: some designs at a site will be very “fresh” looking while others will appear somewhat faded or will be partially covered by mineral deposits. Since different artists likely used paints of slightly different composition and color, differential weathering of pictographs is not as reliable an indicator of the passage of time as is differential patination of petroglyphs, but it does serve to indicate that paintings were made at different times.
The lack of evidence of extensive pictograph weathering during the last century also provides a dating clue by suggesting that some of these paintings could be of considerable age. Photographs taken of the Painted Rocks pictograph site on Flathead Lake in western Montana show the paintings to appear as fresh today as they did in 1903. With such minimal weathering, these (and other similar) paintings could be as much as several centuries old.
A Word Regarding Interpretation
Scientifically accurate interpretations of Columbia Plateau rock art are notably scarce. The written material that presents a realistic idea of the richness of this art tradition (and of the cultures of the artists), along with reasonable explanations of some of the reasons why the art was created, consists of a small handful of professionally published journal articles and graduate student research papers. None of these attempts to deal with more than a small part of the Columbia Plateau province, and very few are oriented toward helping the public understand and appreciate this rock art.
This lack of a comprehensive publication describing and interpreting rock art for the lay public is due primarily to scholars’ reluctance to speculate in print about their subject beyond the limits of statistical confidence intervals, competing hypotheses, and regional style definitions. Thus, the only interpretation readily available to the public often consists of far-fetched “translations” of sites or designs—some of which even go so far as to attribute this art to ancient Chinese explorers, lost Celtic monks, or prehistoric spacemen! The result is a dearth of interesting, scientifically accurate information about rock art written for lay readers.
Elsewhere, other authors and I (Dewdney 1964; Hill and Hill 1974; Joyer 1990; Keyser 1990) have shown that rock art can be interpreted for the general public in a way that both educates and entertains, while still observing the basics of scientific accuracy. Throughout this book I continue this effort. In some cases this involves making leaps of faith (albeit minor ones) that a strict scientific treatment would not. For instance, paintings of horsemen or groups of men with bows and arrows found in the extreme eastern part of the Columbia Plateau (in British Columbia and central Idaho) cannot be positively identified as depicting successful raiding parties returning from the northern Plains. No site was so identified by any of the Kutenai, Nez Perce, or Flathead Indian informants used by early ethnographers in the area. These Indians did, however, recall such raids on Blackfeet, Crow, and Cheyenne villages, and among the Nez Perce there are ledger drawings of such war parties. Informants also indicated that important events were sometimes pictured in rock art.
Therefore, putting these three things together—informant recollections of raiding parties, informant indications that important events were recorded in rock art, and the presence of rock art sites showing armed warriors or groups of horsemen—I feel comfortable speculating that some of these sites do, in fact, show war parties going to or returning from raids on Great Plains tribes. Similarly, the identification of bison hunt drawings, “twin” figures, “spirit beings,” and vision quest pictographs is based on logical deductions drawn from ethnographic accounts, comparison with rock art in other areas (world-wide), and my own experience in nearly twenty years of rock art research. Obviously, in a courtroom much of this would be circumstantial evidence and hearsay, but this is usually the only sort of evidence available to archaeologists. It is exactly this circumstantial evidence that enables us to tell the stories about the people who made the artifacts or painted the pictographs!
This book is, therefore, my interpretation and retelling of some of the myriad fascinating stories with which archaeologists entertain one another around field campfires or in the bars at professional conferences. As such, some of these deductions are not “statistically significant,” and some have alternate explanations in the form of “competing hypotheses.” They are, however, good stories, based on the best available scientific information and thousands of hours of analysis, study, and thought. I hope they make you think about the subject, but more than that, I hope you enjoy reading them as much as I enjoyed writing them.
† Long Narrows was the term for part of The Dalles of the Columbia River. A series of basalt gorges formed steep-sided canyons through which the Columbia cascaded; these are now obliterated by the Dalles Dam.
Map 2. The geographical boundary of the Columbia Plateau, which consists of the drainages of the Fraser and Columbia rivers and the Hells Canyon