Artifact Geofact Issues
The artifactual character of the Calico lithic assemblage has been questioned (Haynes 1973; Payen 1982a, 1982b; Taylor and Payen 1979; Duvall and Venner 1979). Haynes (1973) postulated that rock fracturing by tectonic stresses, weather, rock-on-rock percussion in streams and mudflows, pressure retouch of buried cobbles, and successive generations of flake removal and separation from cores through cycles of erosion and redeposition could have occurred during deposition of the alluvial deposits at Calico and produced specimens indistinguishable from artifacts.
In fact, such mechanisms do not normally cause artifact-like fracturing. This is especially true with regard to the small, delicate flaking seen on light-duty tools such as burins, gravers, becs, denticulates, and reamers. Studies indicate that stream transport abrades and rounds rocks quickly; it does not dislodge artifact-like flakes by percussion. Streams are capable of generating only about 10 percent of the force needed to dislodge significant numbers of percussion flakes; forces in mudflows are even lower due to viscosity of the matrix, which acts as a buffer. If streams produced pseudo-artifacts, alluvial fans and dry streambeds would be littered with such specimens. Clearly, they are not. The only phenomena known to produce significant amounts of percussive flaking (and occasional pseudo-artifacts) are high-energy storm conditions on rocky beaches and certain types of rock falls from cliffs. Neither of these produces multiple flake scars with sharp aretes preserved between. Cliff-face rock falls produce conchoidal flake scars only in siltstone, which is far too friable to be useful as tool material.
Significant points bearing on the artifact/geofact issue at Calico were noted by Simpson (1999), and are rephrased and supplemented here as follows:
Artifact lithology. Although the Yermo fan contains diversified rock types of varied quality, more than 95 % of identified tools and diagnostic flakes are composed of siliceous chalcedony, chert, and agate. This indicates intelligent selection; natural forces are not selective.
Non-random distributions. Pits and trenches sampling a tiny fraction of the Yermo Fan reveal that debitage, percussion-generated flakes, and crude chopping, scraping, cutting, and boring tools are present in discrete clusters, both vertically and horizontally, within the fan deposits.
The clustered material, primarily debitage, has not been transported by water action, as demonstrated by the abundance of sharp-edged flakes that are often paper-thin. Both the flake size distribution and the absence of rounding on flake edges indicate that the Calico flakes remain in place at the site of production, and cannot have been generated or redeposited by natural forces.
Rows of clean edge-forming flake scars often terminate distally in a weathered cortex that preserves delicate microstructures. Such weathering structures, resulting from hydration, could not have survived natural transport and random battering sufficient to detach sizeable flakes.
Important attributes of controlled percussive flaking in the Yermo material includes platforms, acute platform angles, bulbs, bulb scars (eraillures), force lines, and compression rings (ripples). Natural impacts do not generate the force to produce such features in durable siliceous rocks.
Flaking of the Calico material is not random, but is characterized by repeated regular patterns, indicating craftsmanship. It is very significant that the excavations have revealed diagnostic percussion flakes and multiple examples of diverse tools that duplicate Old World Paleolithic artifacts, including hand axes, choppers, scrapers, blades, burins, gravers, and rotational tools.
Many large and small tools have denticulate margins that resemble saw blades. No natural process creates regularly-spaced toothed margins, which on large tools require bifacial flaking. Natural processes quickly remove such teeth by blunting or fracture.
Statistical analysis of the Calico flakes indicates that their attributes match those of flakes produced by experimental percussive knapping of siliceous cobbles gathered from surfaces adjacent to the Calico dig. These attributes include presence of distinct force bulbs (26.1% vs 24.3%); acute platform angles (94.3% vs 95.5%); presence of two or more dorsal face facets demonstrating serial flake removals from a core (46.7% vs 41%); high percentage of interior flakes (70.6% vs 61.1%); size range between 15 and 35 mm (88% vs 79%).
Wear patterns indicate that tools were utilized as well as created at the Calico site. While the Yermo Fan appears to have been primarily a knapping station, it was also a work area.
What are the necessary and sufficient, objective and testable criteria, which can be employed to determine whether a particular lithic collection is indeed the result of man’s craftsmanship? Leland Patterson developed one very productive approach (Patterson 1983). This approach focuses on quantifiable definitive morphological characteristics and technological attributes. The attributes considered include the presence or absence of force bulbs, bulb scars, and ripple lines; the condition of the striking platform, whether intact, crushed or missing; the angle of the striking platform; and the striking platform type, whether single, multiple faceted, or with remaining cortex. Also examined is the dorsal face of the striking platform for the presence or absence of small facets and number of major facets on the dorsal surface. Flake size in terms of square millimeters and flake thickness are also recorded. Finally, the flake is categorized as to type, whether primary (with cortex on entire dorsal surface), secondary (with cortex on part of the dorsal surface), or interior. By these criteria, the Calico specimens are indisputably artifacts.
More than 20 years ago two quantitative studies concluded that the Calico material does not include human artifacts. However, these studies are flawed by inappropriate methods, inclusion of inappropriate material, and comparisons with the comparatively refined Paleo-Indian tool kit rather than Paleolithic artifacts, which are the true analogues of the Calico material.
Payen (1982) studied flake scar angles as traits for distinguishing artifacts from geofacts. He tested a method developed by Barnes (1939) who had compared frequency of obtuse angles on eoliths, natural fractures, and artifacts. Barnes found obtuse angles on 72% of eoliths, 75% of natural fractures, and 18% of artifacts and concluded that the flaked tools of an industry may be considered to be of human origin if not more than 25% of the angles of scar-platform are obtuse (90 degrees or more) (Barnes 1939:111).
Payen’s study lacks explicit hypotheses, specification of statistical methods to be used in their testing, and a statement of assumptions about the nature of the statistical distributions involved. It is unclear whether Payen assumed all artifacts in his study to be samples from the same population, or if each separate assemblage was assumed to represent a different population. The same ambiguity applies to different types of geofacts. The possibility of differences in population variances must be addressed before differences between mean values of attribute measurements can be studied whether with analysis of variance or Student’s t-test.
Hypothesis testing can only proceed if an investigator assumes certain things about nature, and through sampling, obtains data from the real world. If the data so acquired is sufficiently at odds with the assumption, there is cause to abandon that assumption. If the data are not compellingly different from what is to be expected by assumption, there is no reason for the investigator to abandon his original hypothesis. Failing to reject does not make a very strong case in favor of the original assumption because it amounts to begging the question; rejection, however, is considered a sound basis for disbelieving it.
Payen measured all flake angles on each Calico specimen in his sample. It is conceivable that flake scars were confused with striking platforms. Flaking from one side of a specimen can often remove earlier platform areas on the other side. Angles between two flake scars are different from angles between platforms and derivative flake scars.
Payen compared mean angle values for Calico specimens with those on specimens selected as representing controlled and uncontrolled fracture. He found that “statistically, there is no significant difference between the sample of alleged tools and the uncontrolled fracture series.” (Payen 1982:200). Payen’s conclusion, however, does not follow unambiguously from his data: his controlled and uncontrolled fracture series are not distinct populations. Neither Payen, nor Barnes, has established a single trait criterion for distinguishing artifacts from geofacts.
Duvall and Venner (1979:462) examined a sample of Calico artifacts and concluded they were form-selected examples of naturally flaked rocks. Their assessment was based on variances in seven attributes (length, width, thickness, flake angle, medial axis angle, lateral edge angle, and distal edge angle), and comparison with comparable attributes on specimens in eight Paleo-Indian collections reported by Wilmsen (1970). Wilmsen was concerned with differences in tool technologies and functions, not with distinguishing artifacts from geofacts.
Instead of establishing one null hypothesis, Duvall and Venner list three non-equivalent statements. They fail to demonstrate that any of these or the statistical techniques used in their testing bears on the question of the artifactuality. In the end, they illustrate only that the examined Calico tools are not from the same population as the non-bifacial tools and utilized flakes from certain Paleo-Indian sites.
Both of these studies compared Calico specimens, putatively dated at older than 100,000 years, with Paleo-Indian artifacts tens of thousands of years younger than the Calico objects could have been. It seems obvious that a truly fair analysis would compare the Calico material with artifacts contemporaneous in age with the Calico specimens–i.e., the Old World Paleolithic: Acheulean, Clactonian, etc.