Niches in Bedrock: Towards Integrating Bedrock Mortars into a Niche Construction Framework

Purpose of Study: This thesis investigates the ability to use bedrock mortar depth to reliably infer resource processing goals of the prehistoric peoples that utilized Browns Valley Ridge, California. The study area is a portion of a Sierra foothill formation at the eastern margin of the Sacramento Valley, with a particularly high density of bedrock mortars. Traditionally, bedrock mortar depth has been assumed to be a result of usewear. The purpose of this study is to analyze the depth patterns of the mortars for evidence of manufacture to desired specifications that can be linked to intended processing goals.

Methods: This study utilizes archival data, field survey, and statistical analysis. Archival data include ethnographic and archaeological evidence of bedrock mortar use. Field methods include the intensive pedestrian survey of the study area and detailed recording of bedrock milling features. Statistical analyses involved Kolmogorov-Smirnoff and Shapiro-Wilks tests for normality; hypothetical frequency distribution comparisons; and data correction and transformation techniques including kernel density estimate, logarithmic transformation, and differential binning. The combination of the three methods create the ability to analyze bedrock mortar depth frequency datasets for meaningful patterns.

Findings: The 1,274-acre study area on Browns Valley Ridge contains over 74 bedrock milling loci comprised of at least 722 milling features on 240 bedrock outcrops. The bedrock mortar depth-frequency distribution of the study area is a highly non-normal, strongly multi-modal, and heavily right-skewed. The distribution does not match the results of hypothetical use-wear patterns, and the depth-frequency distributions of bedrock mortars at the locus and outcrop scales are highly variable. Data correction shows that the modes of the study area distribution are indicative of meaningful data, and not the result of data smoothing or recording errors.

Conclusion: The multi-modality, non-normality, and skew of the depth frequency distribution in the study area show that it is not the result of natural processes or usewear. The robust modes of the distribution display desired depths of those who manufactured them. Given the substantial investment in time and place of bedrock mortar manufacture, it is highly likely that the depth distribution of the study area represents a design indicative of processing goals.