57 THE DISTRIBUTION AND FUNCTION OF BEDROCK MORTARS IN CALIFORNIA Alice B. Francisco 59 This study of the distribution of bedrock mortars, in confining its attention to their occurrence in California, adopts limits which are drawn by availability of data rather than by the natural distribution of the feature studied. Although the extensive use of bedrock mortars by North American Indians in ethnographic times has apparently been limnited to California, the Great Basin, and the Southwest (Driver and Massey 1957: 237), the archaeological distribution is much wider. While most common in California, they probably occur with varying frequency in most areas of the United States. Their occurrence in the Mesoamerican region seems infrequent (Armillas 1950: 120), but they have been reported at one site in the Honduras (Strong 1948: 101) and at several places in Mexico (Armillas 1950: 119, 120). In South America, they are the most outstanding archaeological remain in parts of the Sierra de Cordoba and the Sierra de San Luis (De Aparicio 1946). Outside the New World, occurrence of the feature is less clear. Armillas suggests similar mortars occur in Indonesia, but a check of his sources (a work by Robert von Heine-Geldern, which cites material from Th. a Th. Van Der Hoop 1932) reveals that the mortars described are not bedrock mortars, but single- or multiple- holed mortars worked in hewn, flat stones. Although distribution of bedrock mortars in areas other than California is not within the intended scope of this study, a casual survey of the literature would indicate the feature may be lacking outside the New World. Absence of reports, of course, would be of questionable significance. It is possible that in many areas bedrock mortars might not be considered a category worth noting. As a primary source of data, this study has made use of information collected from the site survey reports on file at the Archaeological Research Facility of the University of California during the summer of 1964. While these files provided a wealth of material, certain limitations of the present study are inherent in the records as they exist. These limitations are: (1) All 58 counties are not equally represented by site survey reports. While Table 1 is intended primarily to show the exact relationship between the number of sites surveyed and the number where bedrock mortars are present, the absolute number of sites recorded for each county may be seen in that table. The significance of the pro- portion of bedrock mortar sites should be judged in the light of that number. (2) Not all parts of each county have been equally well surveyed, and it is entirely possible that the total picture of patterning of distribution has been slanted by a tendency to survey certain types of terrain more thoroughly than others. (3) Although, with the reservations noted above, the percentages of sites re- porting bedrock mortars are probably a fairly accurate reflection of reality, they probably incorporate some error caused by omissions in the site survey reports. Forms were not always fully filled out. In some cases, the occurrence of bedrock mortars was not listed as a feature on the form itself, but was shown on an accompanying sketch 60 map. It is reasonable to expect that some of the many surveyors who have contributed to the files of the Facility may not have included notation of any kind concerning this particular feature. It is equally reasonable to assume that in some cases the occurrence of the feature was totally overlooked. (4) Although many site survey reports made a distinction, there was a ten- dency to lump together bedrock mortars and bedrock metates. For purposes of this study, no effort has been made to distinguish between what should be regarded as two types of remains. It should be noted, however, that there probably exists in the Facility records enough information to assist an analysis of the relationship of these two categories to each other and to other grinding tools. For the sake of convenience, this paper will speak simply of "bedrock mortars. " Although in some cases a "metate" classification might clearly have been more suitable, in most cases the grinding areas on the rock outcrops were noted as round or nearly round holes. In California, the depth to which mortars are ground into bedrock may range from a barely perceptible one-half inch or so to around 15??. 1 There is a certain amount of discussion in the literature concerning the depth at which California bedrock mortars were abandoned. Much of the discussion tends to ignore field evidence. Even such experienced field workers as Elsasser and Bennyhoff seem to have been curiously influenced by one particular sentence from Barrett and Gifford's study of Miwok culture, which states simply, "These cuppings [mortar holes] were abandoned after they had worn to a depth of perhaps five inches" (Barrett and Gifford 1933: 143). Bennyhoff uses this report as evidence that "to some extent the depths of a group of mortar holes would therefore give some indication of the length of occupation of a par- ticular site" (Bennyhoff 1956: 12). Elsasser repeats Bennyhoff's suggestion, adding, "Some data are at hand as to the point at which an individual mortar hole was abandoned (at 5 inch depth)?? (Elsasser 1960: 40). The single report of Miwok norms should not, obviously, be generalized too far in either space or time. The immediate possibility that even within a given area such norms may have varied during different periods is supported by Price's description of the Washo: ??The author' s survey of historic and prehistoric bedrock mortar sites indicates that recent Washo prefer a more shallow basin than was used previously. This was also suggested by a Washo woman, Bertha Holbrook, who said that the California Indians used much deeper bedrock mortars. Most of the recent mortar holes range from one inch to 2. 5 inches in depth; prehistoric mortars in Washo territory range up to 7 inches in depth" (Price 1963: 89). Another aspect of this discussion in the literature which seems to be curiously divorced from the evidence can be seen in those works which deal with the reasons prompting abandonment at a given depth, whatever that depth may be taken to be. There seems to be wide agreement that practical reasons, connected with the physical charac- teristics of the material being ground, governed the norms of depth. Fenenga writes: "Large, oily seeds (acorns, pinenuts, and buckeyes) could be ground only in the shallow 61 pits because otherwise they congeal into a pasty mass" (Fenenga 1952: 342). Gifford, in the year before he noted the 5-inch norm which has been so frequently cited, wrote In a study of the Northfork Mono: "A mortar hole three Inches or less in depth was employed for acorns. If deeper, the meal was likely to become 'grease, ' that is, to pack solidly in the bottom" (Gifford 1932: 24). Among the Tubatulabal, Voegelin (1938: 17) found that mortar holes were "enlarged with use until at the depth of 10 Inches they became unusable. " Although different grinding techniques might be a factor, an arbitrary level of depth is unlikely to affect the tendency of acorn meal to pack into a solid mass. Except in those areas where the limited availability of suitable rock dictated a need for working the holes to a deeper than average depth, the depth of mortar holes may have been deternined by nothing more than habit and custom. Deep or shallow, these mortar holes have been noted in many types of rock in California -- granite, steatite, lava, sandstone, "very poor conglomerate,," limestone -- Indeed, one is even reported in obsidian! The lack of a consistent norm for depth of abandonment, the varying characteristics of the rock Into which they were worked, and the possibility of variation in patterns of use combine to cast doubts on the utility of depth of mortar holes as a measurement of length of occupation of a particular site. Although the bedrock mortar in California is generally thought of as being closely tied to the use of acorns, some bedrock mortar sites occur outside the range of the oak. In acorn-using territories, such mortars were also used for grinding other food substances. Although a minor use, the grinding of materials other than foodstuffs should be mentioned. A number of small bedrock mortars found in association with painted petroglyphs were probably paint mortars. Fenenga mentions the use of bedrock mortars for pounding clay (Fenenga 1952: 352), but the source he cites (Gayton 1929: 245) makes it clear that such pounding was done only on the flat surface of a bedrock outcrop in which mortars were also present, and that "clay was never put in mortar holes in which food substances were ground" (Gayton 1929: 241). Some mortars may on occasion have been used by shamans for preparing medicines. Such use, as well as the general physical similarity between utilitarian bedrock mortars and the cupped rocks known as "baby rocks" or "rain rocks", may possibly account for the occasional report of the "sacredness" of a bedrock mortar. The fact that bedrock mortars frequently are not actually part of a habitation site should not be taken as indication of any ceremonial use, although such a theory has been suggested (Armillas 1950: 123). Choice of location for living sites would clearly be dictated by a number of factors, for example, that suitable rock outcroppings be within some reasonable distance would be all required for their use. Indeed, there may even have been certain advantages to having the community grinding area slightly separated from the habitation site as such. Another circumstance which might lead to question of ceremonial use -- the 62 frequent association in California of bedrock mortars and petroglyphs -- requires no more explanation than that the horizontal, as well as the vertical, surfaces of convenient rock outcroppings were naturally exploited, for both paint and food mortars. We are considering, then, the California distribution of the bedrock mortar, a grinding tool used primarily for the preparation of acorns, but also for the grinding of other food and non-food materials. Site survey reports give us the information presented in Table 1 and in Maps 1 and 2. In addition to organizing information presently available on the archaeological occurrence of the feature, this study has attempted to determine the geographic range of its recent use as reported in ethnographic accounts. Map 3 is the visual representation of the latter type of information. This map can give only an approximate guide to actual occurrence. To evaluate its reliability for any area, it would be advisable to refer to those notes on informants and research procedures which are given in the sources from which the map is drawn. For this purpose, such sources are listed in a separate list of references at the end of the paper. Some tribal areas were, of necessity, reported using only one informant of doubtful reliability. For the Central Coast area, field procedures described make it doubtful that all informants actually denied the presence of bedrock mortars. For the sake of the value of visual communication of pattern, it was necessary to take boundaries for tribal groups which may not actually delimit the use of bedrock mortars. A map of greater truthfulness, but of less useful visual impact, might have simnply shown the locations of the informants who reported on the presence or absence of the feature. Two areas of the map require special note, since there exist conflicting ethno- graphic reports. Gifford, writing on the Kamia, reports the absence of bedrock mortars (Gifford 1931: 28-29); this contradicts Drucker's CED study. Drucker' s information comes from one "half-breed" informant with a Kamia mother; the map follows Gifford's report. For the Kidu-dokado of Surprise Valley, the map shows bedrock mortars as absent, following Stewart (Stewart 1941: 381). It should be noted, however, that Isabel Kelley describes what may be a bedrock metate for that tribe (Kelly 1932: 138). 63 TABLE 1 BEDROCK MORTAR SITES OF CALIFORNIA BY COUNTY County No. Sites No. Bedrock Mortar % Total Sites Reporting Sites Bedrock Mortars Alameda 95 15 15.8 Alpine 65 21 32 3 Amador 106 64 60.4 Butte 159 81 50.9 Calaveras 239 142 59,4 Contra Costa 134 6 4.5 Colusa 19 1 5.3 Del Norte 12 0 0.0 El Dorado 78 36 46.2 Fresno 375 173 46.1 Glenn 98 0 0.0 Humboldt 206 0 0.0 Imperial 66 0 0.0 Inyo 439 44 10.0 Kern 196 62 31.6 Kings 27 1 3.7 Lake 270 24 8.9 Lassen 272 24 8.8 Los Angeles 118 8 14.8 Madera 103 66 64.1 Marin 369 3 .8 Mariposa 251 149 59,4 Mendocino 569 3 .5 Merced 63 5 8.0 Modoc 194 15 7.3 Mono 540 34 6.3 Monterey 370 112 30.3 Napa 267 20 7.5 Nevada 84 17 20.2 Orange 81 2 2.5 Placer 59 21 35.6 Plumas 68 3 4.4 Riverside 110 33 30.0 Sacramento 266 20 7.5 San Benito 15 2 13.3 San Bernardino 322 14 4.3 San Diego 301 53 17.6 San Francisco ?3 0 0.0 64 TABLE 1 (continued) County No. Sites No. Bedrock Mortar % Total Sites Reporting Sites Bedrock Mortars San Joaquin 137 0 0.0 San Luis Obispo 182 31 17.0 San Mateo 97 1 1.0 Santa Barbara 263 8 2.9 Santa Clara 51 10 19.6 Santa Cruz 46 4 8. 7 Shasta 272 8 2.9 Sierra 38 6 15.8 Siskiyou 321 44 13.7 Solano 48 8 16.7 Sonoma 258 5 1. 9 Stanislaus 133 16 12.0 Sutter 31 5 16.1 Tehama 239 2 .8 Trinity 139 1 .7 Tulare 326 189 58.0 Tuolumne 224 48 21.4 Ventura 24 3 12.5 Yolo 66 2 3. o Yuba 16 3 18.8 CHANNEL ISLANDS Anacapa 1 0 0.0 S. Clemente 6 0 0.0 S. Miguel 51 0 0.0 S. Nicolas 68 0 0.0 Santa Catalina 105 1 1.0 Santa Cruz 18+ 0 0.0 Santa Rosa 155 0 0.0 65 DISCUSSION Map 1 shows clearly a general pattern of distribution. The entire Sierra Nevada, extending from Butte County in the north to Kern County in the south, is an area of generally dense distribution. The frequency with which bedrock mortars are found in the Sierras might have appeared even more pronounced had Tuolumne County been more completely surveyed. Although the absolute number of site surveys for that county appears adequate on the surface, only certain portions of the county are well represented. Beyond the latitudes of the Sierra, bedrock mortars are seen in California only in scattered distribution, except for a cluster of sites at Tule Lake. To the east, in the Basin ranges, density falls off sharply from that noted in the Sierra. To the west of the Sierra, across the Great Valley, only scattered sites occur in the northern Coast Ranges. A somewhat closer pattern of sites in the southern Coast Ranges of Monterey and San Luis Obispo Counties is echoed by stretches of site-dotted coastline. Worthy of remark are the great number of sites on the coast of Monterey County. Because they cluster so closely they cannot be adequately indicated on the map; but Table 1 provides supplementary information. Several areas of the state are virtually devoid of bedrock mortar sites: the Mojave Desert, the Great Central Valley, the Colorado Desert, and most of the northern quarter of the state. Absence of suitable rock outcrops clearly dictates the Valley pattern. Map 3 traces the approximate limits of the Great Valley. Within the boundaries of the Valley, the meaning of ethnographic reports of the "presence" of bedrock mortars must be modi- fied by the facts of geography. (Only one of the sources from which Map 3 was drawn made any note of a lack of availability of rock outcrops.) A comparison of Maps 1 and 3 shows a few suggestive patterns. The areas in Napa and Lake counties which show archaeologic, but not ethnographic, reports of bed- rock mortars may indicate areas where earlier use had been forgotten.2 The same possibility would appear in the more striking cluster of bedrock mortar sites near Tule Lake. Comprising a total of 59 sites (including two sites at which both bedrock mortars and bedrock metates were recorded), this cluster is located just north of a tribal area which reports the use of bedrock mortars ethnographically, and yet for which no archaeo- logical occurrence of the feature is recorded. Some keys to this paradox may lie in the possibility of error in the map's arbitrary boundary lines, in lack of adequate site surveys, or in the reliability of informants. Nevertheless, the cluster -- so distant from any areas of similarly intense distribution farther south in the Sierras -- clearly raises a question, and may be taken to suggest either independent invention of this quite simple feature or the clue to some past movement of population. Less dramatic, but interesting, -is the fact that so many informants in the southern half of the state reported knowledge of the use of the bedrock mortar in areas 66 lacking in archaeological remains. More complete site survey of the area would be required before an explanation of this phenomenon could be sought. If bedrock mortars prove to be more common in these southern regions than present archaeological evidence has indicated, some adjustment of present theories concerning the close association between bedrock mortars and the acorn might be necessary. If no more bedrock mortar sites appear, the ethnographic reports require some explanation. The most interesting possibility raised by the maps, however, concerns the significance of the mortars which appear along the coast of the middle section of the state. The frequency with which these mortars occur in Monterey County is remarked above. Suitable stone is not limited to this one stretch of the state's coast; there would seem to be little unique to the range of foods available in that area to call for utilization of a certain form of grinding tools with an intensity otherwise found only in the Sierras of the state. It is entirely possible that the mortars are no reflection of truly local use. The patterning offers confirmation of a theory previously drawn from other evidence. Pilling, after quoting documents from 1860 and 1900 which mentioned historical summer migrations for Yokut Indians, and noting the presence of shell heaps marking camp sites along the Monterey coast, wrote: "Another interesting implication for the archaeologist is presented by the fact that under this annual migration pattern some sites scattered along the coastal region may easily be the result, not of local inhabitants, but of the annual tourlsts of prehistory" (Pilling 1950: 439). Another summer migration pattern -- this time from Sycamore Creek to Tulare Lake -- is described by Latta (1929: 29). Map 1 would suggest the possibility that the coast of San Luis Obispo County may also hold just such prehistoric tourist sites. The bedrock mortar sites in the Coast Ranges of the middle section of the state occur roughly in the latitudes of areas of coast-line distributions between these areas and the Sierra. It would seem possible that the truly "local" nature of these sites should be questioned also; they may represent the traditional campsites, used regularly over long stretches of time, by migrating Sierra tribes en route to summer seafood. These possibilities are worth noting for two reasons. They raise questions in the interpretation of evidence from the sites involved, and, perhaps more important, they indicate that some modifications may be called for in the generally accepted picture of the California Indian. Although for many areas of the state the traditional picture of isolated tribes tied to their territories may hold, a broad band across the central section of the state may have been an area of rather different patterns. 67 NOTES 1 The range exhibited in California does not Indicate the limits of the form this feature may take: the Kentucky "hominy holes" were "often two or three feet deep, generally six or seven inches in diameter, . . " (Webb and Funkhouser 1929: 701). 2 The ethnographic reports, however, should be evaluated in light of the field methods used and informants consulted. SOURCES OF DATA USED FOR MAP 3 Aginsky, B.-W. 1943 CED: XXIV - Central Sierra. Anthropological Records 8 (4): 393-468. Barnett, H. G. 1937 CED: VII - Oregon Coast. Anthropological Records 1(3): 155-204. Driver, H. E. 1936 Wappo Ethnography. University of California Publications in American Archaeology and Ethnology 36(3): 179-220. 1937 CED: VI - Southern Sierra Nevada. Anthropological Records 1(2): 53-1 54. 1939 CED: X - Northwest California. Anthropological Records 1(6): 297-433. Drucker, P. 1937 CED: V - Southern California. Anthropological Records 1(1): 1-52. 1941 CED: XVII - Yuman-Piman. Anthropological Records 6 (3): 91-230. Essene, F. 1942 CED: XXI - Round Valley. Anthropological Records 8(1): 1-97. Gifford, E. W. 1931 The Kamia of Imperial Valley. Bureau of American Ethnology, Bulletin 97. Gifford, E. W. and S. Klimek 1936 CED: II - Yana. University of California Publications in American Archaeology and Ethnology 37(2): 71-1000. Gifford, E. W. and A. L. Kroeber 1937 CED: IV - Pomo. University of California Publications in American Archaeology and Ethnology 37 (4): 117-254. Harrington, J. P. 1942 CED: XIX - Central California Coast. Anthropological Records 7(1): 1-46. Steward, J. H. 1941 CED: xm - Nevada Shoshone. Anthropological Records 4(2): 209-359. 69 Stewart, O. C. 1941 CED: X1V - Northern Paiute. Anthropological Records 4(3): 361- 446. Voegelin, E. 1942 CED: XX - Northeast California. Anthropological Records 7(2): 47- 251. 70 REFERENCES Armillas, P. 1950 Pozuelos en Penas en el Estado de Guerrero. Mesoamerican Notes 2, Department of Anthropology, Mexico City College, pp. 118-124. Aschmann, H. 1959 Barrett, S. A. 1910 The Central Desert of Baja California: Demography and Ecology. Ibero-Americana 42. University of California Press. The Material Culture of the Klamath Lake and Mooc Indians of Northeastern California and Southern Oregon. UCPAAE 5(4): 239- 292. Barrett, S.A. and E. W. Gifford 1933 Miwok Material Culture. 3ulletin of the Public Museum of the City of Milwaukee 2(4): 117-376* Beauchamp, W. T, 1897 Polished Stone Articles Used by the New York Aborigines Before and During European Occupation. Bulletin of the New York State 4(18). Bennyhoff, J.A. 1956 An Appraisal of the Archaeological Resources of Yosemite National Park. UCAS Report No. 34. De Aparicio, F. 1946 The Comechingon and Their Neighbors of the Sierras de Cordoba. Handbook of South American Indians 2: 673-685. U. S. Government Printing Office, Washington. Driver, H.E. and W.C. Massey 1957 Comparative Studies of North American Indians. Transactions of the American Philosophical Society 47(2). Elsasser, A.E. 1960 The Archaeology of the Sierra Nevada in California and Nevada. UCAS Report No. 51. 1962 Indians of Sequoia and Kings Canyon National Parks. Sequoia National History Association, Three Rivers. Gifford, E. W. 1932 The Northfork Mono. UCPAAE 31(2): 15-65. 71 Heizer, R.F. and A.B. Elsasser 1953 Some Archaeological Sites and Cultures of the Central Sierra Nevada. UCAS Report No. 21. Hindes, M.G. 1959 A Report on Indian Sites and Trails, Huntington Lake Region, California. UCAS Report No. 48. Kelly, I. 1932 Ethnography of the Surprise Valley Paiute. UCPAAE 31(3). Latta, F. F. 1929 San Joaquin Primenval, Uncle Jeff's Story, A Tale of a San Joaquin Valley Pioneer and His Life With the Yokuts Indians. Press of Tulare, Calif., Times. Lothrop, S. K. 1933 Pilling, A.R. 1950 Atitlan. Carnegie Institution of Washington, Publication No. 444. The Archaeological Implications of an Annual Coastal Visit for Certain Yokuts Groups. American Anthropologist 52(3): 438-440. Price, J. A. 1963 Washo Prehistory: A Review of Research. University of Utah Anthropological Papers 67: 77-93. Steward, J. H. 1933 Strong, W. D. 1948 Ethnography of the Owens Valley Paiute. UCPAAE 33(3): 233-350. The Archaeology of Honduras. Handbook of South American Indians 4: 71-142. Th. a Th. Van Der Hoop, A. N. J. 1932 Megalithic Remains in South Sumatra. W. J. Thieme & Cie., Zutphen, Netherland. Voegelin, E.W. 1938 Tubatulabal Ethnography. Anthropological Records 2(1): 1-84. Webb, W. S. and 1929 W. P. Funkhouser The So-called "Hominy Holes" of Kentucky. American Anthropologist 31(4): 701-709. 73 Map 1. Location of Archaeological Bedrock Mortar Sites in California. 74 REtCOROED 9RC14EoLa&tChL SITES: J7o RP-PoTcriwe BEOROCv MORTARS .S- 4.59 i 4+4 30-44.9 s s s * . * .S , . . * * * . * * * * * * * . . .s ., ....... . ..... ..... >s . z * . . - z . * * . * - . @ * - \ \ . - . - \ .............. - - * . ? . - -s * v - - - * * * . - * @ ) . . . . . . . . . . . .. . . . . . v * - - - Z Z e - s e - @ - - - - * * * - s v v - * * * - - _ . * * * * - .......... - . * ....... . * . . * . . . . * . . . . . . * . . * . . * . * - - - * - * * @ @ * @ * . -s * B @ @ * @ @ @ @ @ - @ @ * e e *-- - e x .. ............ .......... a , * . * * * - . * . * * - * * * * - . - * * * - . * v - v * * @ - _ , . . - . . . @ . @ . * . * . @ * . - @ * - -t ........................ ................................................................... ^ ......................... { * * * . . . . . . . . . . . . .. . . . . . * . . \ ......................... E , . . .. . .. . . .. . .. .. .. . . .. . ... ......................... .......................................................... K .... ...... w i ........ ...................... ....... ................ # # . . 0 * b ......... * * .......... - . ........................ . * * .. . . . . . . . . . . . .. . . . . . . .. . . . . . , . . . . . . . . . . . . . . . . . . .R ................... , , ,#2 + + ++ ++ r w+; + * + + . +.+ * + + + + + + v + + + ; + + ^ 5+ + s + + + + * v +';+ X +;F f F + *F+ +ws + ;*+++44r's-+'s's +'v' e s -s4ci 555-645 Map 2 . . . . . . . . . . . . . . . Overlay for Map 3: The Great Valley of California. 75 ETHNOGPeP#4C REPOPRS OF r3EORPoc tORTARP3 %M BP. I I - a 0 * 0 *:@@: NO IAm l, .,_ _ I I * ___ I I t ; I I I _ _ l I I\- % - I I ) I I I9 _ _ _ 6 I 1 1 - - - - - -O I,,X _ __E I~~~~ %n + + (I + + 4 ' HtsX./ ~ % - -_' _ _ -, %.0 %W _ %_ _ _ %_ _ w _ M%,'- - _ _ _ __ _ _ %.o%.o%.o %I % .P/ J %~ I o %wod-O ' %* d I %D 9..- t @ _ _ _ _ _ _ _ _ _4 _ + 4 4 * V %_ _ _ *_ %_ _ _ , s O_ %._ %_O_ -O_ %.O_ %. ~~~~ PR?~~~~~~~. %O*a*Io . %#4T04 5U'~~~~~LE Ro..k AV.ML.A%W od ..# %,,L%E %.e _.0 _ _ __ w _ _ _Xl I 4- -. + -\ wS~~~~~~~~~N %.. _ 0 I%o %0 %. %.iv v_Os A EMTs aE Q v _ _ _LOO- 'SUI'TA5L ROL AYt_XLA~~~~~ ___v SLF Map 3