Beck, C - Rocks Are Heavy

8/12/2019 Beck, C - Rocks Are Heavy

1/27

Rocks are heavy: transport costs andPaleoarchaic quarry behavior in the Great Basin

Charlotte Beck, a, * Amanda K. Taylor, b George T. Jones, a

Cynthia M. Fadem, c Caitlyn R. Cook, d and Sara A. Millward e

a Department of Anthropology, Hamilton College, Clinton, NY 13323, USAb

197 Cleveland Drive, Croton-on-Hudson, NY 10520, USAc Department of Anthropology, Washington State University, Pullman, WA 99164, USAd 1115 Morris Avenue, Point Pleasant, NJ 08742, USA

e 90 Butler Lane, Mohnton, PA 19540, USA

Received 15 January 2002; revision received 14 May 2002; accepted 16 May 2002

Abstract

Central place foraging models are used to investigate assemblage variability at two Paleoarchaic (ter-minal Pleistocene/early Holocene) dacite quarries in the central and eastern Great Basin. Our analysesfocus specically on biface reduction and how varying degrees of reduction relate to the costs of trans-porting the resulting products upon departing the quarry. Our results suggest that when the distance to betraveled to a residential base is great, reduction will proceed further at the quarry than if the residentialbase is fairly close. Further, a residential site assemblage will consist of bifaces at later stages of reductionthan its associated quarry.

2002 Elsevier Science (USA). All rights reserved.

Keywords: Quarry; Paleoarchaic; Great Basin; Central place foraging; Biface reduction; Mobility; Transport costs;Resource utility; Toolstone; Dacite

Archaeological study of quarry workshops hasbeen sporadic at best, even though they have beenof interest to archaeologists for over 100 years(e.g., Holmes, 1891, 1897, 1919). This lack of treatment is likely due to the many problems thatseem to be inherent in dealing with these sites(Singer, 1984). Their areal extent alone, often

many hectares, can be daunting, but added to thisfactor are the vast numbers of artifacts (thousandsor even millions) that are generally present (Singerand Ericson, 1977). Ericson (1984, p. 2) reectsthat while the investigation of quarries should be aprimary focus in the study of peoples who reliedupon stone tools, prehistoric archaeologists haveseemingly avoided analyses of such sites due totechnical and methodological limitations im-posed by a shattered, overlapping, sometimesshallow, nondiagnostic, undatable, unattractive,redundant, and at times voluminous material re-cord. Essentially, because lithic procurementsites are often used over long periods of time by

Journal of Anthropological Archaeology 21 (2002) 481507

www.academicpress.com

* Corresponding author.

E-mail addresses: [email protected] (C. Beck),[email protected] (A.K. Taylor), [email protected] (G.T. Jones), [email protected] (C.M.Fadem), [email protected] (C.R. Cook),[email protected] (S.A. Millward).

0278-4165/02/$ - see front matter 2002 Elsevier Science (USA). All rights reserved.PII : S0278-4165(02)00007-7
http://mail%20to:%[email protected]/http://mail%20to:%[email protected]/


2/27

different groups of people, quarry records arefrequently characterized by mixed technologiesand artifact types.

Although quarries often are complex recordsof human behavior, both temporally and spa-tially, they represent an important aspect of pre-historic land use and adaptation. Quarries aresources of lithic raw material, referred to here astoolstone, a signicant component of most pre-historic technologies. The acquisition of toolstonethus assumes a theoretical importance on a levelwith that of food and other resources and must beconsidered as such if the nature of prehistorichuman behavior is to be fully understood.

Quarry assemblages can exhibit considerablevariation (e.g., Gramly, 1980; Holmes, 1897;

Johnson, 1984; Raab et al., 1979), although theexplanations for this variation differ. Holmes(1897), for example, related various kinds of quarry assemblages to differences in environmen-tal setting. Raab et al. (1979) argue that the vari-ation in assemblages from quarries reects thenumber and kinds of different activities performed.These explanations, of course, are not necessarilymutually exclusive. Johnson (1984, p. 225) statesthat the activities represented in a quarry siteassemblage depend on how the site articulates with

the rest of the prehistoric settlement system,which includes the environmental context of thequarry as well as the activities performed there.

In this paper, we address the issue of quarryassemblage variability as it relates to the overallsettlement and subsistence strategy. Specically,we examine the costs and benets of particularquarry behaviors in relation to the cost of trans-porting the resulting products upon departure, andhow these behaviors are reected in particularquarry assemblages. We explore the conditionsunder which it is benecial to invest more effort inreduction and manufacture at the quarry andthose under which it is better to transport rawmaterial to a residential site for this purpose. Weuse as an example two Great Basin dacite quarries 1

and their associated residential sites 2 from easternand central Nevada (Fig. 1). All four sites wereutilized almost exclusively by populations duringthe terminal Pleistocene and early Holocene, re-ferred to here as the Paleoarchaic (Beck and Jones,1997, p. 162).

The Great Basin Paleoarchaic

The earliest well-documented occupation of the Great Basin begins about 11,50011,000 rcyB.P., about the same time substantial Paleoindianrecords appear in other parts of North America.In the Great Basin, however, this early occupationis represented by an archaeological record com-

prised mainly of lithic artifacts resting on thesurface, and it is thus difficult to date precisely,but we do know that it persists until ca. 80007500rcy B.P. Although uted points are present in theGreat Basin, they have yet to be dated and theycommonly occur with large stemmed point formsthat are known to have also been present in theterminal Pleistocene. These latter forms, however,persist until the mid-Holocene, or ca. 7500 rcyB.P. Thus, there is little to distinguish the earliestGreat Basin records in the terminal Pleistocene

from those that followed in the early Holocene(and are generally termed Archaic ) (Simms,1988). For this reason, following Willig (1989), weuse the term Paleoarchaic to refer to the GreatBasin terminal Pleistocene/early Holocene ar-chaeological record (see Beck and Jones, 1997).

Despite more than six decades of research,archaeologists still do not know a great deal aboutPaleoarchaic subsistence. A realistic, if general,picture can be drawn from site location data, thepaleoenvironmental record, and the meager butsignicant subsistence records from dry rockshel-ters and open sites (Beck and Jones, 1997). Pale-oarchaic sites typically occupy high relictlandforms, including Pleistocene lake features(e.g., spits, beaches) as well as alluvial terraces, invalley bottoms; sites in upland areas are rare, apattern markedly contrasting with the mid- andlate Holocene record. It seems clear, based onthese landform associations, that Paleoarchaic

1 Butler and May (1984, p. 185) dene true quarryingas actual excavation to locate and remove the rawmaterial from its geological matrix. Such quarries arerelatively rare in the Desert West (but see Elston andRaven (1992); more often, toolstone occurs in cobble/

boulder form on the surface. Here we use the termquarry in the more general sense, to refer to areas of prehistoric toolstone procurement, whether that pro-curement involves actual extraction from rock or sedi-ment matrix or simply cobble testing and collection.

2 The term residential site is used here in the sense

meant by Binford (1980) at which the entire groupcamped, presumably for some extended period of time,and where a variety of activities took place. These sitesgenerally exhibit a wide-range of artifactual forms usedin these different activities.

482 C. Beck et al. / Journal of Anthropological Archaeology 21 (2002) 481507


3/27

groups established residential/activity sites adja-cent to wetlands, i.e., shallow lakes, marshes,springs and seeps, and active streams. This set-tlement pattern persists throughout the span of Paleoarchaic occupation despite evidence of sub-stantial climatic changes. Following a relativelymoist Younger Dryas (ca.11,20010,100 rcy B.P.[Madsen, 1999, p. 77]), when paleoenvironmentalrecords indicate that substantial surface water wasavailable, valleys became increasingly dryer dur-ing the early Holocene (Grayson, 1993), althoughmoisture apparently remained high relative tomodern conditions. With drying and presumablya parallel loss of preferred resources, Paleoarchaicdiet breadth appears to have expanded, focusingincreasingly on smaller mammals, birds, and sh(Beck and Jones, 1997). Finally, near the close of the early Holocene, increased and substantialplant use is signaled by widespread appearance of groundstone artifacts.

Paleoarchaic mobility appears to have beensubstantial and wide-ranging. For example, in thecentral Great Basin populations traveled along anorth-south trajectory from the northeastern tosouthern sections of Nevada, a distance of more

than 450 km (Jones et al., 2002). Similar patternsare emerging from the western Great Basin, alsoindicating wide-ranging movement in a north-south direction (Graf, 2001, pp. 126133). Giventhese large territorial ranges, raw material pro-curement was most likely embedded in the Pale-oarchaic subsistence schedule. As littlearchaeological evidence other than stone toolsremains of this system, however, lithic procure-ment and manufacturing waste are the principleclues to Paleoarchaic subsistence strategies andother aspects of their lifeways.

The Paleoarchaic toolkit shares many similar-ities with that of Paleoindians in the Plains andSouthwest; it includes various kinds of scrapers,gravers, notched tools, wedges, hammerstones,and abraders (Beck and Jones, 1997). An addi-tional component of this toolkit is the crescent(Fig. 2E), which rarely occurs in Paleoindiantoolkits outside the Great Basin (but see Daugh-erty, 1956, pp. 247249; Frison and Bradley, 1999,p. 34). The primary component of the Paleoar-chaic toolkit, however, is the large stemmed point(Fig. 2AD). These points, which may havefunctioned both as weapon tips on thrusting

Fig. 1. Map showing locations mentioned in the text.

C. Beck et al. / Journal of Anthropological Archaeology 21 (2002) 481507 483


4/27

spears and as knives (Beck and Jones, 1993, 1997),and the ake debris from their manufacture,

dominate most Paleoarchaic assemblages. In mostof the Great Basin, these points are made pri-marily of ne-grained volcanics, including ande-sites, dacites, and basalts (Basgall, 1993a,b, 2000;Beck and Jones, 1990b); in the northern GreatBasin they are more often made of obsidian(Amick, 1995), which is readily available in thatarea. They are rarely, however, made from chert.These points represent the nal product of a bi-face reduction system that is standardized to suchan extent that it is readily identiable, even inassemblages containing later material (Fig. 3).

It has been suggested by a number of scholarsthat bifacial technologies, such as those used byPaleoindians of North America, serve well in sit-uations of high mobility and where a range of toolneeds would select for exibility (Goodyear, 1989;Kelly, 1988, 719720, 2001, 6768; Kelly andTodd, 1988, p. 237; Nelson, 1991, pp. 6676;Torrence, 1989, p. 63). Bifacial cores, while de-signed to enable production of a bifacial tool likea projectile point, constitute serviceable tools inthemselves and provide for akes of predictablemorphologies to serve as expedient tools or blanksfor further reduction. Bifacial technologies are notrepresented universally among mobile popula-tions and are poorly known in ethnographic

accounts (Kelly, 1988, 2001). Still, as Hiscock(1994, p. 277) suggests, where they evolved, bifa-

cial technologies seem to have represented a re-sponse to the uneven distribution of lithic sourcesacross the landscape (see also, Andrefsky, 1994;Bamforth, 1986, pp. 2324; Kelly, 2001, pp. 68 72; Kuhn, 1994, p. 40), serving as a tactic to re-duce the risk of being caught unprepared, eitherwithout the proper tools or without a sufficientsupply of toolstone.

It is certainly the case that bifaces served as thecentral component of the transportable toolkit of Paleoarchaic peoples of the Great Basin, as theydid for Paleoindians elsewhere in North America.This is evidenced not only by the prevalence of bifaces in Paleoarchaic assemblages but also bythe fact that the majority of akes with platformsin these assemblages are biface reduction akes. 3

In our work in eastern Nevada, for example, wehave collected nearly 18,000 unmodied akes

Fig. 2. Paleoarchaic long stemmed points (AD) and crescent (E). Types are as follows: (A) Cougar Mountain; (B)heavily resharpened Cougar Mountain; (C) Lake Mohave; (D) Parman; (E) Buttery Crescent (see Beck and Jones,1997 for original type denition and citations).

3 A biface reduction ake is dened here as any akeremoved from a bifacial core. These akes generallyhave: (1) a faceted platform, (2) an acute external

platform angle, (3) a lip at the intersection of theplatform and ventral surface, and (4) a concave dorsalsurface/arched longitudinal axis. This class encompassesthe more specic category of bifacial thinning akes (seeWhittaker, 1994, pp. 185187).



5/27

from 15 surface sites. Only 5687 (32.2%) of theseakes have platforms but nearly 80% of these

(n 4415) are biface reduction akes.4

In our studies, we have also noted that Pale-oarchaic biface assemblages vary considerablywith respect to the extent of reduction undertakenat different locales (Beck and Jones, 1988, 1990b,1993, 1994b). In some cases, primarily middle-to-late reduction debris is predominant while inothers, only early-to-middle reduction is evident.Here we investigate the variation in biface as-semblages with respect to the extent of reductionundertaken at different locales using the predic-tions of central place foraging models, which ex-plain variation in the eld processing of resourcesas a function of travel time. With respect to lithic

assemblages, variation in the reduction stagesrepresented in different assemblages, especially

those from quarries, is explained by the antici-pated travel distance and the cost of transportingtoolstone.

Central place foraging models

In 1980, Binford introduced the notions of residential and logistical mobility in relation tohow foragers move across the landscape and howthey exploit resources. Residentially mobile for-agers generally forage on a daily basis from theirresidential base, returning to that base at the endof the day; the residential base is moved as re-sources within the foraging radius are depleted.Logistical mobility, on the other hand, involvestravel to procurement sites that are too far fromthe residential base to return the same day; inthese cases small subgroups travel from the resi-dential base to these locations and may remainthere for several days or weeks. Whether orga-nized residentially or logistically, foragers musttransport resources from procurement sites backto their residential base. Although the need totransport resources from distant procurementsites has long been recognized as a condition-ing variable in butchering and eld processing

Fig. 3. Biface reduction sequence represented in Great Basin Paleoarchaic sites. Denitions for biface stages (Table 2)are based on Callahan (1979).

4 It is true, as Ahler (1989b, p. 87) notes, that akesmight appear to be technology-specic, such as bifacialthinning akes and bipolar akes, can be produced bymultiple techniques. Ahler s experimental results pre-sented in Table 1 (Ahler, 1989b, p. 88), however, showthat of the 91 biface thinning akes (a more speciccategory than our biface reduction akes), 88 (96.7%)

were produced through biface edging and thinning,resulting in an error of 3.3%. If we reduce the totalnumber of biface reduction akes by 3.3% (from 4415 to4269), our assemblage is still overwhelmingly dominatedby these akes (75.1%).



6/27

techniques (e.g., White, 1953, 1954), more recentlyarchaeologists have attempted to quantify thecosts of resource transport and how these costsmay affect the types and quantities of food re-

mains found in archaeological sites (e.g., Barlowand Metcalf, 1996; Bettinger et al., 1997; Binford,1978; Bird and Bliege Bird, 1997; Drennan,1984a,b; Jones and Madsen, 1989; Metcalfe andBarlow, 1992; Metcalf and Jones, 1988; O Connellet al., 1988, 1990; Rhode, 1990). These studieshave demonstrated that eld processing reducesthe amount of unusable material, thereby reduc-ing transport costs of the load being carried backto the residential base. As a result, low-utilityportions of a resource, such as nut hulls or certain

animal parts, may be rare or absent at the resi-dential site. The important implication is, of course, that the type and quantity of resourcesrepresented in the archaeological record of theresidential base may not bear a direct relationshipto the importance of those resources in the diet(Bird and Bliege Bird, 1997, pp. 5253).

Transport costs have been addressed usingcentral place foraging models (e.g., Orians andPearson, 1979; Stephens and Krebs, 1986), pri-marily with respect to differential transport of animal body parts (e.g., Binford, 1978; Metcalf and Jones, 1988; O Connell et al., 1988, 1990), butmore recently for other resources as well (Barlowand Metcalf, 1996; Bettinger et al., 1997; Metcalfeand Barlow, 1992). Central place models positthat, holding technology and eld conditionsconstant, optimal foraging behavior changes as afunction of increasing round-trip travel time fromcentral place to foraging location and back(Bettinger et al., 1997, p. 888). Generally, as thedistance between central place and foraging lo-cation increases, eld processing of resources be-comes more cost-effective. The most familiar of these models, and probably the best suited to ar-chaeological use (Bettinger et al., 1997, p. 888), isthat developed by Metcalfe and Barlow (1992)

and Barlow and Metcalf (1996). This model ex-plores the trade-offs that foragers face whentransporting resources comprised of componentsof different utility over long distances (Bird and

Bliege Bird, 1997, p. 44). In the simplest case, aresource package is comprised of only twocomponents, one of which has low or no utility.Metcalfe and Barlow (1992, p. 342) offer nuts asan example, which consist of nutmeat containedin a valueless hull. A forager must decide whetherremoving the hulls at the procurement site iseconomically more benecial than transportingunshelled nuts back to the residential base. Thetrade-off here is between the time spent processingin the eld and the cost of transporting the un-

processed resource from the eld site. If, for ex-ample, the hull equals half of the nut volume, thentwice as much useful nutmeat can be transportedin each load if the hulls are removed beforetransport, decreasing the amount of travel timeexpended per unit of useful material; on the otherhand, eld processing increases the amount of time expended per unit of useful material at theprocurement site (Bettinger et al., 1997, p. 888). If the distance to be traveled is great, the benet of eld processing is obvious in that fewer trips willhave to be made, but if the distance is short, itmay be more cost-effective to make the extra tripsand process the nuts at the residential base.

The primary assumption of central place for-aging models is that foragers will make economi-cally efficient decisions concerning eld processingand transport costs; but implied in this assumptionis that either processing time in camp has no cost,or no processing in camp is required for con-sumption or use (Metcalfe and Barlow, 1992,p. 345; see also, Bettinger et al., 1997, p. 888).Metcalfe and Barlow (1992, p. 345) go on to say,however, that for estimating the order in whichparts should be culled during eld processing(for resource packages consisting of more thantwo components), and the relative (emphasis in

Table 1Total number of dacite bifaces analyzed at each of the four sites studied

Site name Site type Location No. of bifaces

Cowboy Rest Creek-Loc. 1 Quarry Grass Valley 58

Cowboy Rest Creek-Loc. 2 Quarry Grass Valley 613Knudtsen-Locality 1 Residential Grass Valley 215Knudtsen-Locality 2 Residential Grass Valley 656Little Smoky Quarry-Locality 1 Quarry Little Smoky Valley 130Little Smoky Quarry-Locality 2 Quarry Little Smoky Valley 200Limestone Peak Locality 1 Residential Jakes Valley 54



7/27

original) travel times at which they should be re-moved and discarded, it is only necessary to assumethat processing time in camp is less costly per unitthan eld processing time, or less time is requiredto process in camp than in the eld, or both.

The decision of whether or not to spend timeeld processing a resource depends on (1) the eldprocessing time for that resource; (2) the gain inresource utility due to eld processing; and, (3) thedistance to the central place (Bettinger et al., 1997,p. 888). Fig. 4 presents a model of the two-com-ponent resource discussed above in which eldprocessing is a discrete process; that is, the com-ponent of low utility is removed in the eld, in-creasing the overall utility of the transported load.In exploring their model, Metcalfe and Barlow use

a scale of proportional utility where utility isscaled from 1.0 to 0.0; 1.0 indicates the highestpossible utility, 0.0 no utility at all (Metcalfe andBarlow, 1992, p. 354). For example, if nuts arecomposed of 30% nutmeat and 70% unusableshell, the nutmeat has a utility of 1.0 and the shellhas a utility of 0.0. Therefore, an unprocessedload has a proportional utility of 0.3, while that of a processed load is 1.0.

The utility function U t shown in Fig. 4 rep-resents the relationship between the cost of eldprocessing a resource and the increased utility of aload of that resource due to processing; an in-crease in eld processing time from x0 to x1 willincrease resource utility from y 0 to y 1 . The slope of this function predicts the travel time, z 1 , (and thustravel distance) at which eld processing becomescost-effective (Bird and Bliege Bird, 1997, p. 44).There is an inverse relationship between the ben-et derived from eld processing a resource andthe travel time at which eld processing that re-source becomes worthwhile (Metcalfe and Bar-low, 1992, p. 347).

In testing this model among the Meriam of theTorres Strait Islands, Australia, Bird and Bliege

Bird (1997) nd good agreement between thepredictions of the model and their observations.In their study, Bird and Bliege Bird observedprocurement and processing of two-componentresources: several species of shellsh comprised of high-utility esh and low/no-utility shell. Theynd that, in general, the Meriam eld process thelow utility components of shellsh for transportin a manner that maximizes the rate at which

Fig. 4. Central Place Foraging Model describing the relationship between eld processing and resource utility. In thismodel a resource has two components. Removal of the low utility component incurs time costs while increasing theutility of the resource for transport. An increase in eld processing time from x0 to x1 will increase resource utility from y 0to y 1 . The utility function U t represents the relationship between the cost of eld processing a resource and the in-creased utility of a load of that resource due to processing. The slope of this function predicts the travel time, z 1 , at whicheld processing becomes cost-effective. (After Metcalfe and Barlow (1992, 346).



8/27

they deliver edible esh to a central locale (Birdand Bliege Bird, 1997, p. 52). They also observethat shellsh species that are relatively difficult toeld process, for which eld processing adds littleto the proportion of usable meat, or are collectednear the residential base or temporary camp, werenot eld processed but transported whole (Birdand Bliege Bird, 1997, p. 52).

As Metcalfe and Barlow (1992, p. 348) note,however, utility functions for some resourcesmay be continuous and differentiable, as de-picted in Fig. 5. The question here is not whethera low-utility component of a resource is expectedto be transported, but rather how much of it willbe transported (Metcalfe and Barlow, 1992, p.348). As is the case in Fig. 4, the slope of a line

tangent to the curve in Fig. 5 predicts the traveltime at which a certain level of eld processingbecomes cost-effective. This model predicts that if foragers must travel to point B, then they willprocess the resource to a greater degree in the eldthan if they were traveling to point A.

The procurement of food resources, however,rarely occurs in the absence of technology (Gray-son and Cannon, 1999, p. 146) and thus the costs

and benets of raw material acquisition, process-ing, and transport must gure into the overallevaluation of a particular foraging strategy. AsKelly (2001, p. 68) notes, to spend time acquiringand processing toolstone at the quarry, a foragermust give up the opportunity to do something else,which can exact a cost. In presenting their model,Metcalfe and Barlow (1992, p. 341) suggest that thesame trade-off between eld processing and trans-port that had been evident for some time regardingbutchering and eld processing prey should beevident for other resources procured in packagesthat include useful and useless (or less useful)parts. Although Metcalfe and Barlow concen-trate their efforts on food resources, they note themodel s applicability to toolstone, a packaged

resource (1992:352). Several researchers have be-gun to explore the use of the Metcalfe and Barlowmodel to investigate procurement, processing, andtransport costs of lithic toolstone (e.g., Elston,1990, 1992a, pp. 153174, 775801; Kelly, 2001,pp. 6576). We draw on these studies below in ouruse of this model in an attempt to explain vari-ability in Paleoarchaic biface assemblages atquarry and associated residential sites.

Fig. 5. Central Place Foraging Model in which the utility function U t is not discrete but continuous and differentiable.

This function describes a case of diminishing returns in which the gain in utility per increment of processing decreaseswith increased eld processing time. The slope of a line tangent to the curve predicts the travel time at which a certainlevel of eld processing is cost-effective. Points A and B represent two hypothetical residential camps. (After Metcalfeand Barlow, 1992, 349).



9/27

Modeling biface reduction and transport costs

Fig. 5 shows a utility function that increaseswith increasing processing effort but at a de-creasing rate. Biface reduction has been describedas a continuous process (e.g., Ingbar et al., 1989;Magne, 1985; Muto, 1971; Raab et al., 1979;Shott, 1996b; Sullivan and Rozen, 1985) andpresumably is well modeled by such a utilityfunction. In practical terms, however, most ar-chaeologists segment this continuum into a seriesof stages (e.g., Bloomer et al., 1992; Boldurian,1990; Bradbury and Carr, 1999; Johnson, 1981,1984, 1989; Magne and Pokotylo, 1981; Nami,1999); it is certainly the case that, prehistorically,mobile populations often reduced bifaces to vari-ous degrees short of the nished product fortransport (Pecora, 2001, p. 175). The stages of reduction can thus be represented as points alongthe utility curve x0 ; x1 ; . . . ; xn, as depicted in Fig.6. The slope of a line tangent to the curve at anyof these points predicts the travel time, andtherefore travel distance, at which eld processingto that particular stage is worthwhile.

Using an optimal load size (that is, the maxi-mum volume or weight that can be economically

carried by an individual [Bettinger et al., 1997, p.892; Jones and Madsen, 1989, pp. 529530]), thetravel time at which eld processing a resourcewill increase its utility for transport can be cal-culated (see Bettinger et al., 1997; Metcalfe andBarlow, 1992 for mathematical details). The datanecessary for this calculation include procurementtime for a load of unprocessed resources, the timerequired to procure and process a load of re-sources, the utility of the unprocessed load, andthe utility of the processed load. The utility of plant and animal food resources is measured incalories per unit weight or volume, 5 while traveltime, procurement time, processing time, andoptimal load size are estimated on the basis of

Fig. 6. Central Place Foraging Model describing biface reduction revised from the continuous and differentiable modeldepicted in Fig. 5. In this case, eld processing is a discrete, multi-step process. The x-axis to the right represents discretebiface stages; the x-axis to the left depicts travel distance, which is implied by travel time (see Bettinger et al., 1997; Birdand Bliege Bird, 1997; Jones and Madsen, 1989). A line tangent to the curve predicts the travel distance at which re-duction of a biface to a particular stage is cost-effective. Points AD represent hypothetical residential bases.

5 That the only (or even primary) payoff of all foodresources is in energy has been challenged by a numberof researchers, especially with respect to meat andhunting (e.g., Bliege Bird et al., 2001; Hawkes et al.,2001) but the issue is debated (e.g., Gurven et al., 2000;

Hill and Kaplan, 1993; Kaplan et al., 2000; see also,Comments following Hawkes et al., 2001). We do notenter this debate here, but accept the assumption of central foraging models that energy is a signicantbenet of food resources.



10/27

ethnographic descriptions or actualistic studies(e.g., Barlow and Metcalf, 1996; Bettinger et al.,1997; Binford, 1978; Bird and Bliege Bird, 1997;Bunn et al., 1988; Jones and Madsen, 1989;O Connell et al., 1988, 1990; Rhode, 1990; seeGrayson and Cannon (1999) for a discussion of the difficulties involved in arriving at these andother measures used in optimal foraging models).Comparable estimates for the acquisition andprocessing of toolstone, however, are not avail-able. Ethnographic observations, for instance,rarely include information concerning the pro-curement and processing of toolstone at a quarry,especially procurement and processing times.Some work has been done on the amount of toolstone used by an individual over the period of

a year (e.g., Gould, 1977; Hayden, 1977), but nodata of which we are aware exist on optimal loadsize for an individual procurement event. As Kelly(2001, p. 67) notes, understanding the costs re-lated to toolstone procurement, processing, andtransport are hindered by the fact that there areno peoples left who make extensive use of stonetools. Whereas the hunting and butchering of animals as well as differential transport of animalparts can still be observed among some hunter-gatherer groups (although in a modern context),

the data concerning these activities with respect totoolstone will have to be generated experimen-tally.

Elston and his colleagues (Carambelas andRaven, 1991; Elston, 1992a) conducted severalactualistic experiments using toolstone from theTosawihi quarries in northeastern Nevada. To theextent possible, we use the results of these exper-iments in our discussions below. These results,however, do not give us the necessary data tocalculate specic travel times for toolstone typesat which eld processing becomes cost-effective.We must, instead, examine archaeological cases tosee if they conform to the qualitative predictions(Kaplan and Hill, 1992) of the model in Fig. 6,using the experimental results as supportive data.We begin by examining if our four assemblagesare appropriately modeled by the same utilityfunction.

Toolstone utility

In simplest terms, the measure of toolstoneutility would seem to be the amount of usablestone in a core, but dening what is usable stone isproblematic. Bettinger et al. (1997, p. 888), in a

hypothetical example, suggest that in the manu-facture of a biface, half of the core ends up aswaste (and thus unusable) material; therefore araw cobble has half the utility per unit weight asthe nal product, the biface. The Tosawihi ex-periments suggest that as much as 98% of theextracted raw material ends up as waste (Elston,1992a, p. 787). In reality, however, toolstoneutility is likely somewhat more complex thanBettinger et al. (1997) suggest. Sources may bevariable in their utility depending upon the formof different technological organizations (Ingbar,1994, p. 55), and thus toolstone utility is related toa number of different factors.

First, some of what is termed waste, in fact,may be usable material and serve as expedient

tools or blanks for additional reduction (Elston,1992a, p. 787). Thus the utility of an unakedcobble of toolstone relates to the intention for thenal product(s), that is, whether the cobble isbeing reduced only to produce blanks, to producea nished biface for a particular function, or is tobe reduced to a certain stage for transport. In thelatter case, the biface may serve as a core fromwhich akes can be derived for use as informaltools or as blanks, but also can itself be used as isor further reduced into a formal tool.

Second, there is not necessarily a one-to-onerelationship between core and biface. Dependingupon the toolstone package size, i.e., boulderversus cobble, that package may produce manybiface blanks or may actually be the biface blank.Therefore, toolstone utility will vary according tosize of the toolstone package.

A third factor that may affect toolstone utilityconcerns the functional requirements of the toolbeing manufactured. Some tools require particu-lar kinds of raw material. Archaeological pat-terns, for instance, suggest that a preference wasshown for high quality cherts when manufactur-ing uted points, although they were sometimesmade from obsidian as well (Beck and Jones,1990b; Goodyear, 1989; Kelly, 1988; Kelly andTodd, 1988). These tools, however, were nevermade from ne-grained volcanics, such as an-desite, dacite, or basalt (Beck and Jones, 1990b,1997). Given this apparent functional require-ment, an artisan will choose to travel a longdistance to obtain suitable raw material and passup unsuitable materials closer to hand. Thus,utility in this case is related to the tool beingmanufactured; that is, dacite will never have highutility when the goal is the manufacture of utedpoints.



11/27

Finally, the quality of the raw material willaffect toolstone utility. Fracture mechanics, evenwithin material types (e.g., chert), can vary con-siderably and thus affect search, procurement, andprocessing costs. The Tosawihi experiments sug-gest that assaying and processing costs increasesubstantially for low quality toolstone (Elston,1992a, p. 787; see also, Kelly, 2001, p. 69)

It is important, however, to distinguish be-tween the utility of the toolstone and the utility of the tool . A number of authors have addressed thelatter issue (e.g., Elston, 1992b, pp. 4042; Gouldand Saggers, 1985, pp. 129134; Kelly, 2001, pp.6872; Kuhn, 1994; Shott, 1996a, pp. 270273),suggesting various ways in which this variablemight be quantied, from the number of useful

strokes of a tool per unit weight or volume of usedstone (Gould and Saggers, 1985, p. 131) to thepotential for producing fresh edges combined withthe length of those edges (Kuhn, 1994, p. 429).Shott (1996a, p. 270) suggests employing twomeasures, maximum utility, which is equal to theaverage maximum use-life of specimens in a toolclass, and realized utility, the actual period of useof an individual specimen. There is an importantdifference, however, between the resource (e.g.,pinon nuts, toolstone) and the use of the resource

(e.g., food, tools). When considering the acquisi-tion of toolstone at a quarry factors such as thenumber of strokes that are eventually accom-plished with a tool are not relevant; what is rele-vant is simply optimizing the amount of usefulstone that can be economically transported fromthe quarry. Given the complexities mentionedabove, then, how might this quantity be mea-sured?

The factors outlined above that likely have aneffect on toolstone utility are:1. the intended intermediate and nal tool mor-

phologies;2. the size of the toolstone package;3. the functional requirements of the intended

product; and,4. the quality of the toolstone.It is obvious from these factors that different kindsof toolstone (e.g., chert versus dacite) can havedifferent utilities and thus a single currency isnot appropriate for all. By holding the type of toolstone constant (i.e., only considering dacite),then two sources of variation are largely elimi-nated, the functional requirements of the intendedproduct and toolstone quality. As stated earlier,Paleoarchaic stemmed points are almost alwaysmanufactured from ne-grained volcanic tool-

stones, except in those areas where obsidian isplentiful; they are rarely made from chert. In thecase considered here, a comparison is made be-tween two dacite quarries where material is of comparable quality and where the manufacturingtrajectory leads to a single nal product largestemmed points; thus the qualities that make chertattractive for certain tools are irrelevant.

The problem caused by difference in toolstonepackage size can be approached through experi-mental studies that focus on nding the averagenumber of blanks of a certain size that can beproduced from a single package. Consideringquarries where the package size is similar, as is thecase for the two dacite quarries considered below,eliminates this problem altogether. We are left,

then, with the most difficult source of variation, of course, the intended product(s) of toolstone re-duction.

Intent in prehistoric lithic reduction is oftenunknowable. As stated above, however, the re-duction sequence for Paleoarchaic stemmedpoints is highly standardized (Fig. 3) and identi-able even in temporally mixed assemblages. Thisbiface trajectory dominates the assemblages atboth of the dacite quarries considered here; ex-pended and broken fragments of nished stem-

med points are also present, representingspecimens broken when very nearly completed orexhausted pieces made of another source materialthat were discarded. Therefore we are condentthat the toolstone at both quarries can be modeledusing the same utility function.

The question remains, however, is the bifacereduction sequence considered here validly mod-eled by the diminishing returns utility curve assuggested above? Flintknapping is fundamen-tally a reductive process (and as such,)placespredictable and repetitive size constraints on thebyproducts produced (Ahler, 1989b, p. 89; seealso, Ahler, 1989a, pp. 205210). As reductioncontinues, the maximum possible size as well asthe average size of the ake byproducts shoulddecrease progressively as the tool itself becomesprogressively smaller (Ahler, 1989b, p. 89; seealso, Ahler, 1989a, pp. 205210; Stahle and Dunn,1982, p. 86). Newcomer (1971) demonstrated thisrelationship experimentally over 30 years ago, andsubsequent studies have supported his conclusions(e.g., Ingbar et al., 1989, p. 124; Pecora, 2001, pp.179180; Stahle and Dunn, 1982, 1984). It is truethat small akes will likely dominate numericallythroughout the reduction process; however, indi-vidual large akes will weigh considerably more



12/27

than individual small akes and thus will domi-nate the average ake weight in stages where theyare most numerous (Ahler, 1989b, p. 90; Stahleand Dunn, 1982, pp. 8687).

The reduction in ake weight, however, is notlinear but curvilinear (e.g., Ahler, 1989b; New-comer, 1971, p. 92; Stahle and Dunn, 1982, p. 89),with weight diminishing quickly during the earli-est stages of reduction, but slowing considerablyas reduction proceeds. Taking the simplest case,that a single cobble is used to produce a singlebiface, then toolstone utility can theoretically bemeasured in terms of the weight of the nishedbiface (e.g., Elston, 1992a, p. 791). Using theBettinger et al. (1997) example from above, that50% of the core ends up as waste, then toolstone

utility should increase with the removal of eachwaste ake. Because the akes removed in initialreduction will be, on the average, of the greatestweight, then the increase in utility per unit of weight will be greatest with initial reduction, anddecline fairly quickly in conjunction with akeweight, just as suggested by the diminishing re-turns utility curve in Fig. 5. This relationship issupprted by the Tosawihi experiments (e.g., El-ston, 1992a, p. 792).

In sum, central place foraging models provide

explanations for certain kinds of variation in re-source procurement and processing. Specically,whether or not a forager chooses to process aresource in the eld to increase the utility of thatresource for transport depends on (1) eld pro-cessing time, (2) the increase in resource utility dueto processing, and (3) the distance to the centralplace. The models predict that there is a traveltime for each resource beyond which it is prot-able to spend time eld processing to eliminatelow-utility components. As our discussion sug-gests, biface reduction at the two Paleoarchaicquarries considered here can validly be modeledby the diminishing returns utility curve in Fig. 5;analytically, however, it is more useful to segmentthis continuum into a set of stages, as modeled inFig. 6.

Although we cannot provide specic traveltimes for the cases presented here, we can use themodel in Fig. 6 to predict directional tendencies(Kaplan and Hill, 1992) in the assemblages fromthe two quarry and residential site pairs in whichthe distance between the members of each pair issubstantially different. In one case the quarry isonly 9 km from the residential site while in theother case, this distance is 60 km. The Tosawihiexperiments suggest that when the distance be-

tween quarry and residential site is less than10 km, high proportions of early-stage bifaces anddebitage are expected at the quarry, but if thisdistance is greater than 10 km, a greater propor-tion of middle-to-late stages will dominate (El-ston, 1992a, p. 798). Therefore, we would expectconsiderable more reduction of bifaces at thequarry in the latter case than in the former.

The central Great Basin database

The four assemblages examined in this analysiswere collected as part of eldwork done by Beckand Jones over the last 15 years in eastern andcentral Nevada (e.g., Beck and Jones, 1988,

1990a,b, 1993, 1994a,b; Jones and Beck, 1990;Jones et al., 1997; Jones et al., 1996; Jones et al.,2003; Huckleberry et al., 2001). The work wasinitiated with the purpose of studying Paleoar-chaic land use in these areas.

The Cowboy Rest Creek Quarry is located inGrass Valley, central Nevada (Fig. 1). This quarryoccurs on an extensive alluvial fan containingabundant large cobbles. Data were collected fromtwo areas of high artifact density, the rst (Lo-cality 1) in 1999 and the second (Locality 2) in

2001. At Locality 1, all bifaces within a 50 50 marea were collected with point provenience usingan electronic total station; debitage and otherartifacts were collected from four randomly se-lected grids within this 2500m 2 area. Severalthousand artifacts were collected from this quar-ry, but currently, total artifact counts are avail-able only for bifaces ( n 58).

At Locality 2, which lies approximately 1 kmnorthwest of Locality 1, analysis was done in theeld. All bifaces as well as all artifacts of chert(n 47) and obsidian ( n 4) and two possiblehammerstones within an approximately 150250 m area were agged. Each artifact was given apoint provenience and in-eld analysis was com-pleted for each of the 613 dacite bifaces.

Lying approximately 9 km to the northeast of Cowboy Rest Creek Quarry is the Knudtsen Site,which occurs on an east-west oriented spitreaching out onto the valley oor. This site isrepresented by pockets of artifacts extending al-most the entire length of the spit, perhaps for3 km. Artifact density is highest at the eastern endof the spit and this is where we focused our work.Two separate localities, termed Knudtsen 1 andKnudtsen 2, were collected in 1999. All bifaces atKnudtsen 1 ( n 215), an area of 200 140 m,



13/27

were given point provenience while a 10% sampleof the area was randomly selected for grid col-lection.

Knudtsen 2 was a dense artifact concentrationthat contained an extraordinary number of arti-facts within a 60 100 m area. Again, bifaces(n 656) were collected with point proveniencewhile all other artifacts within the 6000 m 2 area(n 59; 806) were collected within 2 2 m units.

Given the proximity of Knudtsen to CowboyRest Creek Quarry, we have made the assumptionthat these bifaces are made from Cowboy RestCreek dacite. Upon reviewing the ethnographicliterature, Kelly (1995, p. 133) suggests that a 20 30 km round-trip is the maximum distance thatforagers will travel in daily food collection efforts.

The round-trip distance between Knudtsen andCowboy Rest Creek Quarry (18 km) lies belowthis maximum, suggesting the quarry lies withinthe foraging range of Knudtsen and that most of the dacite from this site was carried there from theCowboy Rest Creek Quarry. In addition, theCowboy Rest Creek dacite is very uniform andexhibits a distinctive red weathering surface, bothof which are evident in the dacite bifaces fromKnudtsen, adding to our condence that the ma-terial at this site is from the Cowboy Rest Creek

Quarry. Finally, the exhausted state of obsidianbifacial tools at Knudtsen indicates that peoplearrived at that site with largely depleted toolkits inneed of refurbishment and Cowboy Rest CreekQuarry was the nearest at hand.

The Little Smoky Quarry is located in thesoutheastern portion of Little Smoky Valley ineastern Nevada (Fig. 1) and was rst described byPrice (1989). This quarry also coincides with anextensive alluvial fan containing dacite cobbles.Collection was made in a 480 m 2 area (Locality 1),selected because of its high artifact density (Beckand Jones, 1994b). A total of 309 artifacts werecollected, 130 of which are bifaces. A second area(Locality 2) was examined in 2001; as in the caseof Cowboy Rest Creek Quarry, Locality 2 analysiswas done in the eld and the artifacts were notcollected. This area was also chosen because of high artifact density and lies adjacent to the pre-vious collection area. All bifaces within a 1800 m 2

area were analyzed. Of the 210 bifaces analyzed,200 are made of dacite; the remaining 10 are madefrom obsidian ( n 7) and chert ( n 3).

Although extensive systematic archaeologicalsurveys have not been conducted in northernLittle Smoky Valley, the work that has beencompleted in the valley has not identied any

Paleoarchaic site comparable to the KnudtsenSite. A site approximately 15 km north of LittleSmoky Quarry, Black Point, exhibits a smallPaleoarchaic assemblage but is dominated by lateArchaic artifacts (Price and Johnston, 1988).While there are a number of Paleoarchaic sites tothe southeast in Railroad Valley (Zancanella,1988), which lies about 65 km from Little SmokyQuarry, we have not analyzed the raw materialsfrom these sites. A good quality dacite source,however, occurs on the Duckwater Indian Reser-vation, which lies between Railroad Valley andLittle Smoky Quarry; we suspect the majority of the Railroad Valley assemblages are comprised of this material.

The closest substantial Paleoarchaic site to

Little Smoky Quarry is Limestone Peak Locality 1(henceforth referred to as Limestone Peak-L1),located ca. 60 km to the east of Little SmokyQuarry in southwestern Jakes Valley (Fig. 1). In1991 a total of 6932 artifacts were collected fromthe site surface. Although this assemblage con-tains 268 ne-grained volcanic bifaces, only thosethat could be identied as having come from LittleSmoky Quarry ( n 54) were used in the presentanalysis. Little Smoky Quarry dacite exhibits adistinctive pattern of large phenocrysts (ca. 1 mm

in diameter), spaced about 510 mm apart. Noneof the other 20 dacite and andesite sources wehave examined from this region and characterizedgeochemically (Jones et al., 1997) possess thismorphology, and therefore we are condent of our sample selection.

A total of 1925 dacite bifaces from four siteswere examined in this analysis. Table 1 shows thenumber of bifaces represented in each of the siteassemblages.

Analytical protocol

Our arguments center on the extent of bifacereduction undertaken at each of the quarries andresidential sites. For this evaluation, we employtwo measures of reduction stage: a traditionalstage classication based on that devised byCallahan (1979) and a biface thinning index pro-posed by Johnson (1981). Traditional stage clas-sications are often criticized as problematicbecause they segment what are believed to betheoretical continua (Teltser, 1991, p. 366). Thereis some indication, however, that there are some-times empirical disjunctions in the reductioncontinuum, especially as the knapper changes



14/27

from hard to soft hammer percussion (see Pecora,2001), but these disjunctions do not necessarilyoccur at the same point in every case. Therefore

stage assignment can still be somewhat arbitrary.The advantage of stage classications is that theyare generally based on multiple criteria and thusassignment of an artifact to a stage is not based ona single variable.

The Johnson Thinning Index has the advan-tage of measuring reduction along a continuumbut is based on only two criteria: weight and planarea. These two independent approaches, how-ever, complement one another and act to limitbias that may be introduced by simply using one

or the other.

Stage classication

We began our laboratory analyses by at-tempting to determine the dominant patterns of reduction for the quarry and associated site as-semblages, identifying each biface according to itsstage of manufacture. In theory, a biface is pro-cessed through several stages of reduction beforeit reaches the form of the nal product in ourcase, a stemmed point. The classication used herecreates four stages of reduction (Fig. 3, Table 2),modied from those of Callahan (1979, pp. 10 11). Stage denitions are based on form, numberand shape of ake scars, edge sinuosity, andthickness.

Each of the 1925 bifaces in the sample wasexamined by two analysts, and these results werecompared to test the reliability of the classica-tion. Analysts consistently agreed on stage as-signments for the Cowboy Rest Creek Quarrybifaces, but there was some disagreement regard-ing assignments for the other assemblages. In re-viewing the results it was evident that one analystrelied more heavily on edge sinuosity to makestage assignments while the second analyst em-

phasized aking patterns. Indeed, these featuresdo not precisely change in tandem (that is, sinu-osity may decrease more quickly than symmetry

in one case while the reverse may be true in an-other). As a consequence, even though each ana-lyst agreed on the assignment of those bifacesexhibiting the modal character of each stage, they judged some bifaces differently. To correct for thedifferences, the assignments of each biface fromeach assemblage were averaged, creating a set of intermediate stages and increasing the number of classes to seven. 6

The Johnson Thinning Index

As an alternative to stage classication, John-son (1981) developed a thinning index (JTI) foridentication of the trend in biface reduction fromearly-stage to late-stage manufacture. The index,which is computed as the ratio between weightand plan surface area, is based on the hypothesisthat a manufacturer would maximize bifacesurface area while minimizing thickness; thus,early-stage bifaces are considerably thicker thanlater-stage bifaces. As Johnson (1981, p. 13)points out, however, small, early stage bifaces

Table 2Biface stage determinants (after Callahan, 1979)

Stage Features

0 Large biface with irregular shape and low symmetry; few very widely and/or variably spaced

ake scars; very wide edge offset (very sinuous); very thick and irregular cross-section1 Large biface with irregular shape; widely and/or variably shaped akes; wide offset; thick andirregular cross-section

2 Large biface with semi-regular and symmetrical shape; closely and/or semi-regularly spacedake scars; edge offset moderate; cross-section semi-regular

3 Regular, symmetrical biface; closely and/or quite regularly spaced ake scars (pressure akingoften present); offset close (little edge sinuosity); cross-section thin and regular; later edge grindingevident on haft

6 It would be inappropriate to use an average of twoordinal categories for statistical comparison. For exam-ple, if a teacher is evaluated on a scale of 16, it isinappropriate for his/her superior to suggest his/herteaching is better or worse because his/her average ishigher or lower than another s. In the present case,however, we are simply using the average to designate anadditional category in between the originally denedstages. In no case were the assignments made by the twoanalysts more than one stage apart (e.g., Stage 1 versus

Stage 2) and thus the averaged category (Stage 1.5)indicates only that the biface has been reduced to agreater degree than Stage 1 but to a lesser degree thanStage 2. Statistical comparisons are not made betweenbiface stages.



15/27


16/27

Fig. 7 shows the distribution of biface stageclasses at both quarries and illustrates how eachquarry differs from the other. The Little SmokyQuarry assemblage is more heavily weighted to-wards middle and late-stage bifaces than either of the Cowboy Rest Creek Quarry samples. As Table4 shows, nearly two-thirds (64.3%) of the bifacesat Cowboy Rest Creek Quarry fall in the rstthree reduction stages while only a third (33.6%)of Little Smoky Quarry bifaces represent stages 0 1. This pattern is also evident in the distribution of the JTI (Fig. 8). JTI values decrease with in-

creasing reduction and, as Table 5 shows, theLittle Smoky Quarry mean for this variable ismuch smaller than that at Cowboy Rest CreekQuarry. These differences are conrmed by sta-tistical comparison (Table 6).

It seems clear from these results that Paleoar-chaic knappers staged lithic tool manufacturedifferently at these two workshop localities, car-rying reduction further at Little Smoky Quarrythan at Cowboy Rest Creek Quarry. It followsthat their companion residential sites should ex-hibit complementary reduction signatures. That

Table 3Statistical tests between different collection samples at the two quarries and at the Knudtsen site

Comparison Separate variances t test of JTI KS test biface stage

t df p p

Little Smoky Quarry, Localities 1 and 2 ) 1.693 227.8 0.092 0.936

Cowboy Rest Creek Quarry, Localities 1 and 2 ) 2.090 103.4 0.039 0.223Knudtsen 1 and Knudtsen 2 2.027 268.4 0.044 0.930

Fig. 7. Prole summary graphs of biface stages represented in the Little Smoky Quarry, Localities 1 and 2 (LSQ-l1,LSQ-l2) and Cowboy Rest Creek Quarry, Localities 1 and 2 (CRCQ-L1, CRCQ-L2) assemblages.



17/27

is, the biface assemblage at each residential siteshould be dominated by later-stage products thanits associated quarry (Elston, 1992b; Johnson,1989, pp. 2022; Metcalfe and Barlow, 1992, p.352). It does not necessarily follow, however, that

Knudtsen and Limestone Peak-L1 will differ fromone another in the same way as their respectivequarries differ. The assemblage composition atthese two sites will be governed as much by theon-site activities and the anticipated travel dis-

Fig. 8. Histograms of Johnson Thinning Index (JTI) values for bifaces in the Little Smoky Quarry, Localities 1 and 2(LSQ-L1, LSQ-L2) and Cowboy Rest Creek Quarry, Localities 1 and 2 (CRCQ-L1, CRCQ-L2) assemblages.

Table 4Biface stages represented in the Little Smoky Quarry and Cowboy Rest Creek Quarry assemblages

Bifacestage

Little SmokyQuarry

Cowboy Rest CreekQuarry, Locality 1

Cowboy Rest CreekQuarry, Locality 2

Cowboy Rest CreekQuarry, Localities 1 and 2

No. % No. % No. % No. %

0 20 6.1 2 3.4 78 12.7 80 11.90.5 12 3.6 8 13.8 76 12.4 84 12.51.0 79 23.9 35 60.3 233 37.9 267 39.91.5 69 20.9 7 12.1 86 14.5 93 13.92.0 82 24.9 4 6.9 103 16.8 107 16.02.5 41 12.4 2 3.4 25 4.8 27 4.03.0 27 8.2 0 0.0 12 2.0 12 1.8

Total 330 100.0 58 100.0 613 100.0 671 100.0



18/27


19/27

Discussion and conclusion

We have presented an analysis of biface as-semblages from two dacite quarries and associ-ated residential sites using the central place

foraging model proposed by Metcalfe and Barlow(1992). Our aim was to evaluate if variability inbiface reduction at these different sites was afunction of transport costs. The model applicationhas been successful, but we realize there is a long

Fig. 9. Prole summary graphs of biface stages represented in the Knudtsen, Localities 1 and 2 (K1, K2) and CowboyRest Creek Quarry, Localities 1 and 2 (CRCQ-L1, CRCQ-L2) assemblages.

Table 7Biface stages represented in all four assemblages

Biface stage Cowboy RestCreek Quarry

Knudtsen Little SmokyQuarry

LimestonePeak-L1

No. % No. % No. % No. %

0 80 11.9 3 0.5 20 6.1 0 0.00.5 84 12.5 9 1.0 12 3.6 3 5.51.0 267 39.9 221 25.4 79 23.9 14 25.91.5 93 13.9 192 22.1 69 20.9 17 31.52.0 107 16.0 231 26.5 82 24.9 13 24.12.5 27 4.0 98 11.3 41 12.4 5 9.33.0 12 1.8 117 13.4 27 8.3 2 3.7

Total 671 100.0 871 100.0 330 100.0 54 100.0



20/27

way to go in this endeavor. To this point, centralplace foraging models have been used almost ex-clusively to investigate the relationship betweenthe processing and transport of food resourcesand have been most successfully applied in eth-nographic situations, where behavior can be ob-served and time budget analysis can be directlyapplied to assess the costs of alternative behav-iors. Their application to decisions regardingtoolstone acquisition are in their infancy.

Central place foraging models are applicable toinstances of economizing behavior. This wouldsuggest that these models are appropriate tostudies of lithic tool manufacture and transport,and the success of our analysis appears to conrmthis claim. In fact, we were successful in spite of the fact that archaeological assemblages representtime-averaged behaviors, which will serve tomask these relationships unless redundant be-haviors yield strong signatures. Still, there may be

Fig. 10. Histograms of Johnson Thinning Index (JTI) values for bifaces in the Knudtsen, localities 1 and 2 (K1, K2) andCowboy Rest Creek Quarry, Localities 1 and 2 (CRCQ-L1, CRCQ-L2) assemblages.

Table 8Johnson Thinning Index statistics for all four assemblages

Statistic Cowboy RestCreek Quarry

Knudtsen Little SmokyQuarry

LimestonePeak-L1

N 671 871 327 54

Minimum 0.15 0.08 0.14 0.51Maximum 27.86 8.39 11.02 3.61Mean 2.94 1.20 2.18 1.38SD 2.20 0.50 1.25 0.48



21/27

Table 9Statistical tests between little smoky and cowboy rest quarries and their associated residential/activity sites

Comparison Separate variances t test of JTI KS testbiface stage

t df p pCowboy Rest Creek Quarry, Locality 1 and Knudtsen 1 ) 8.21 67.6 < 0:001 < 0:001Cowboy Rest Creek Quarry, Locality 1 and Knudtsen 2 ) 9.17 58.3 < 0:001 < 0:001Cowboy Rest Creek Quarry, Locality 2 and Knudtsen 1 ) 16.50 812.7 < 0:001 < 0:001Cowboy Rest Creek Quarry, Locality 2 and Knudtsen 2 ) 19.27 650.2 < 0:001 < 0:001Cowboy Rest Creek Quarry and Knudtsen 1 ) 20.07 722.5 < 0:001 < 0:001Cowboy Rest Creek Quarry and Knudtsen 2 ) 20.38 718.8 < 0:001 < 0:001Cowboy Rest Creek Quarry, Locality 1 and Knudtsen ) 9.01 58.4 < 0:001 < 0:001Cowboy Rest Creek Quarry, Locality 2 and Knudtsen ) 18.98 653.1 < 0:001 < 0:001Cowboy Rest Creek Quarry and Knudtsen ) 20.07 722.5 < 0:001 < 0:001Little Smoky Quarry and Limestone Peak-L1 8.42 196.5 < 0:001 0 :900

Fig. 11. Prole summary graphs of biface stages represented in the Little Smoky Quarry (LSQ) and Limestone PeakLocality 1 (LPL1) assemblages.

Fig. 12. Histograms of Johnson Thinning Index (JTI) values for bifaces in the Little Smoky Quarry (LSQ) andLimestone Peak Locality 1 (LPL1) assemblages.



22/27

questions concerning the suitability of thesemodels for examining behavioral variability re-garding lithic toolstone. This concern, termeddynamic sufficiency by Lewontin (1974, p. 11),regards whether the model contains enough andthe right variables to account for the eld inquestion (Dunnell, 1980, p. 85). Although oursuccess here suggests these models, in fact, do havedynamic sufficiency, the results could be fortu-itous. For future studies the starting point, then, isto continue to evaluate whether the variables inthese models are sufficient, and this can be doneby shifting those variables held constant and thoseallowed to vary.

In the case presented in this paper, for in-stance, we have held toolstone quality, package

size, and intended product constant in order toexamine transport distance and its effects on as-semblage composition. Under the right empiricalcircumstances, transport distance could be heldconstant instead, while allowing other factors tovary. Consider, for example, two sources forwhich quality differs, quality being measured interms of isotrophy (i.e., homogeneity), sharpnessof edges, durability, and workability (Elston,1992b, p. 35). The utility curves for these twomaterials will differ. Because of increased assaying

and processing times for the poorer quality ma-terial, it would take more time, and thus effort, toreduce a blank of the same size to, say, a Stage 3biface than it would for the better quality mate-rial. We would predict, then, that biface manu-facture would be carried out to different stages atthese material sources despite both lying equaldistances from residential sites. These costs, atleast in an average way, can be estimated throughrepeated experiments.

Regarding the quantitative measureability of the variables in these models in the archaeologicalrecord, or what Lewontin (1974, p. 11) refers to asempirical sufficiency, there are some serious ob-stacles. To apply the Metcalfe and Barlow modelin its entirety, we must know: (1) the optimal loadsize for a particular type of toolstone; (2) theprocurement time for an unprocessed load; (3) thetime required to procure and process to differentpoints along the reduction continuum for a pro-cessed load; (4) the utility of an unprocessed load;and, (5) the utility of loads processed to differentpoints along the reduction continuum. The To-sawihi experiments (Elston, 1992b) have providedsome initial data on procurement and processingtimes, but these data pertain to quarrying activi-ties that involved considerable effort to excavate

raw material from quarrying pits. At surfacequarries such as those considered here the initialeffort is expended in search and assaying time, andthus experiments must be done that focus specif-ically on these circumstances. Clearly, a great dealof experimental work is needed to derive estimatesof acquisition and processing costs.

The most signicant obstacle to a more rigor-ous, quantitative application of this model, how-ever, lies in establishing quantitative estimates of toolstone utility. Economists use utility as a de-scriptive (emphasis in original) tool: they measureutility by observing what people choose, not byindependent criteria (Stephens and Krebs, 1986,p. 105). In models concerning food resources,utility is more directly measured as caloric return.

Neither option can be applied to measure tool-stone utility, although the proportion of caloricreturn attributable to technology is an appealingheuristic. But in all practical senses, if measure-able at all, toolstone utility as measured by caloricreturn would be subject to widely varying esti-mates and therefore of dubious value. Toolstoneutility, then, probably should be conceived in adifferent manner.

Using the concept of proportional utility asdiscussed above (e.g., Metcalfe and Barlow, 1992,

p. 354), we might assign a nished biface a utilityof 1.0 and the original blank a value less than 1.0,depending upon the amount of waste removed inthe reduction process (see also, Bettinger et al.,1997, p. 888). But, of course, this requires that weknow the weight of the original blanks reduced toform the biface. A more important question, itseems, is what constitutes a nished biface? If theintent is to take reduction to a form such as aprojectile point, and there is no need for the akesproduced in the process, then it is the point thatwill have a utility of 1.0. But if the intent is to onlypartially reduce a biface so that it can serve as acore, then it reaches a maximum utility earlier inthe reduction sequence. However, as numerousauthors who have written about bifacial coretechnology have pointed out, the advantage of such a technology is in both the nished productand in the akes removed. It is precisely this factorthat makes utility so difficult to measure. Whichcomponent contributes the most to utility? Wemay be able to go some distance towards solvingthis problem with studies that begin with cores of different sizes and in which bifaces are manufac-tured with different intents: nished tool, creationof blanks, creation of bifacial cores. Using ameasure of size, such as weight, edge length, or



23/27

area, a utility value can be assigned to the originalblank, depending on the nal product. In onecase, the intended product may be a bifacial core,and thus reduction is taken to a particular point,at which the utility is 1.0. In another case, theintended product may also be a bifacial core, butthe production of blanks as well, and therefore thenumber of akes of a certain size can be counted,their relative contribution to the nal weight de-termining their proportion of the utility.

In this type of procedure, however, the ar-chaeological record must play an important role.In a certain type of reduction trajectory, for ex-ample, the average ake size we must consider forthe manufacture of tools must be determined.That is, if the purpose is to use akes that result

from the manufacture of a bifacial core or projectile point, there should be an optimal size belowwhich a ake is not likely to be used. This canonly be determined empirically, and althoughthese estimates may not be specically correct inevery situation, the uncertainty of such a deter-mination will likely be no greater than the un-certainties involved in estimates of food resourcereturn rates, whether derived ethnographically orexperimentally. As Grayson and Cannon (1999, p.146) point out, both ethnographically and ex-

perimentally derived return rates have an un-known and unknowable relationship to returnrates that characterized the past. Grayson andCannon (1999, p. 145) also point out that one wayaround this problem has been to convert resourcereturn rates to return rate classes; as a result, themodels rely only on ordinal return rate estimates.The same could be done for toolstone utility es-timates or reduction classes, as discussed above.

In the end the study presented here may seemoverly simplistic. It does, however, demonstratethat central place foraging models can be useful intrying to understand variability in lithic assem-blages that has by default been interpreted asfunctional differences in sites. In spite of theproblems, they see with the implementation of optimal foraging models to archaeological cir-cumstances, Grayson and Cannon (1999, p. 150)state that foraging theory provides the best, if not the only, means currently available to ar-chaeologists for examining interactions betweenpeople and their environments within an evolu-tionary framework. We are sure that many ar-chaeologists would dispute this bold claim.Nevertheless, foraging theory in general, andcentral place foraging models in particular, offerfruitful insights into the reasons for assemblage

variability. What we have done here is the easypart; the hard part is yet to come.

Acknowledgments

An earlier version of the paper was presentedas a poster at the 66th Annual Meeting of theSociety for American Archaeology in New Or-leans, LA. We would like to thank Michael Can-non, Donald Grayson, David Madsen, and twoanonymous reviewers for their helpful commentson previous versions of the paper.

References cited

Ahler, S.A., 1989a. Experimental knapping with KRFand Midcontinent Cherts: overview and applications.In: Amick, D.S., Mauldin, R.P. (Eds.), Experimentsin Lithic Technology. BAR International Series 528.British Archaeological Reports, Oxford, pp. 199 234.

Ahler, S.A., 1989b. Mass Analysis of aking debris:studying the forest rather than the tree. In: Henry,D.O., Odell, G.H. (Eds.), Alternative Approaches toLithic Analysis. Archaeological Papers of the An-thropological Association no. 1, pp. 85118.

Amick, D.S., 1995. Raw material selection patternsamong paleoindian tools from the Black Rockdesert, Nevada. Current Research in the Pleistocene12, 5557.

Andrefsky Jr., W., 1994. Raw-material availability andthe organization of technology. American Antiquity59, 2134.

Bamforth, D.B., 1986. Technological efficiency and toolcuration. American Antiquity 51, 3850.

Barlow, K.R., Metcalf, D., 1996. Plant utility indices:two Great Basin examples. Journal of Archaeolog-ical Science 23, 351371.

Basgall, M.E., 1993a. Archaeology of the Nelson Basinand Adjacent Areas, Fort Irwin, San BernardinoCounty, CA. Prepared for the US Army Corps of Engineers, Los Angeles District, by Far WesternAnthropological Research Group, Davis CA.

Basgall, M.E., 1993b. Early Holocene Prehistory of theNorth-Central Mojave Desert. Unpublished Ph.D.Dissertation, Department of Anthropology, Univer-sity of California, Davis.

Basgall, M.E., 2000. Patterns of toolstone use in late-Pleistocene/early-Holocene assemblages of the Mo- jave desert. Current Research in the Pleistocene 17,46.

Beck, C., Jones, G.T., 1988. Western Pluvial LakesTradition occupation in Butte valley, Nevada. In:Willig, J.A., Aikens, C.M., Fagan, J.L. (Eds.), EarlyHuman Occupation in Far Western North America:



24/27

The Clovis-Archaic Interface. Nevada State MuseumAnthropological Papers no. 21, pp. 273301.

Beck, C., Jones, G.T., 1990a. Late Pleistocene/earlyHolocene occupation in Butte valley, Eastern Ne-vada: three season s work. Journal of California and

Great Basin Anthropology 12, 231261.Beck, C., Jones, G.T., 1990b. Toolstone selection and

lithic technology in early Great Basin prehistory.Journal of Field Archaeology 17, 283299.

Beck, C., Jones, G.T., 1993. The multi-purpose functionof Great Basin stemmed series points. CurrentResearch in the Pleistocene 10, 5254.

Beck, C., Jones, G.T., 1994a. Dating surface assem-blages using obsidian hydration. In: Beck, C. (Ed.),Dating in Surface and Exposed Contexts. Universityof New Mexico Press, Albuquerque, pp. 4776.

Beck, C., Jones, G.T., 1994b. On-site analysis as analternative to collection. American Antiquity 59,304315.

Beck, C., Jones, G.T., 1997. The terminal Pleistocene/early Holocene archaeology of the Great Basin.Journal of World Prehistory 11, 161236.

Bettinger, R.L., Malhi, R., McCarthy, H., 1997. Centralplace models of acorn and mussel processing. Jour-nal of Archaeological Science 24, 887899.

Binford, L.R., 1978. Nunamiut Ethnoarchaeology. Ac-ademic Press, New York.

Binford, L.R., 1980. Willow smoke and dogs tails:hunter-gatherer settlement systems and archaeolog-ical site formation. American Antiquity 45, 420.

Bird, D.W., Bliege Bird, R.L., 1997. ContemporaryShellsh gathering strategies among the Meriam of the Torres Strait islands, Australia: testing predic-tions of a central place foraging model. Journal of Archaeological Science 24, 3963.

Bliege Bird, R., Smith, E.A., Bird, D., 2001. The huntinghandicap: costly signaling in male foraging strategies.Behavioral Ecology and Sociobiology 50, 919.

Bloomer, W.W., Ataman, K., Ingbar, E., More, M.W.,1992. Bifaces. In: Elston, R.G., Raven, C. (Eds.),Archaeological Investigations at Tosawihi, A GreatBasin Quarry. Prepared for the Bureau of Land

Management, Elko Resource Area, by Intermoun-tain Research Reports, Silver City, NV, pp. 81131.Boldurian, A.T., 1990. Lithic Technology at the Mitchell

Locality of Blackwater Draw: A Stratied FolsomSite in Eastern New Mexico. Plains Anthropologist,Memoir 24.

Bradbury, A.P., Carr, P.J., 1999. Examining stage andcontinuum models of ake debris analysis: anexperimental approach. Journal of ArchaeologicalScience 26, 105116.

Bunn, H.T., Bartram, L.E., Kroll, E.M., 1988. Variabil-ity in bone assemblage formation from Hadzahunting, scavenging, and carcass processing. Journalof Anthropological Archaeology 7, 412457.

Butler, B.N., May, E.E., 1984. Introduction to Part III:Quarry and Workshop Sites. In: Butler, B.M., May,E.E. (Eds.), Prehistoric Chert Exploitation. Studies

from the Midcontinent. Center for ArchaeologicalInvestigations, Occasional Paper No. 2, p. 185.

Callahan, E., 1979. The Basics of Biface Knapping in theEastern Fluted Point Tradition: A Manual for FlintKnappers and Lithic Analysts. Archaeology of

North America v. 7.Carambelas, K., Raven, S., 1991. Actualistic Quarrying

Experiments at Tosawihi. Paper presented at the56th meeting of the Society for American Archaeol-ogy, New Orleans.

Daugherty, R.D., 1956. Archaeology of the Lind CouleeSite, Washington. Proceedings of the AmericanPhilosophical Society 100, 223278.

Drennan, R.D., 1984a. Long-distance movement of goods in the Mesoamerican Formative and Classic.American Antiquity 49, 2743.

Drennan, R.D., 1984b. Long-distance transport costs inpre-Hispanic Mesoamerica. American Anthropolo-gist 86, 105112.

Dunnell, R.C., 1980. Evolutionary theory and archae-ology. In: Schiffer, M.B. (Ed.), Advances in Archae-ological Method and Theory, vol. 3. Academic Press,New York, pp. 3599.

Elston, R.G., 1990. A Cost-Benet Model of LithicAssemblage Variability. In: Elston, R.G., Budy, E.(Eds.), The Archaeology of James Creek Shelter.University of Utah Anthropological Papers no. 115,pp. 153163.

Elston, R.G., 1992a. Economics and Strategies of LithicProduction at Tosawihi. In: Elston, R.G., Raven, C.(Eds.), Archaeological Investigations at Tosawihi, aGreat Basin Quarry. Prepared for the Bureau of Land Management, Elko Resource Area, by Inter-mountain Research Reports, Silver City, NV, pp.775801.

Elston, R.G., 1992b. Modeling the Economics andOrganization of Lithic Procurement. In: Elston,R.G., Raven, C. (Eds.), Archaeological Investiga-tions at Tosawihi, a Great Basin Quarry. Preparedfor the Bureau of Land Management, Elko ResourceArea, by Intermountain Research Reports, SilverCity, NV, pp. 3147.

Elston, R.G., Raven, C. (Eds.), 1992. ArchaeologicalInvestigations at Tosawihi, a Great Basin Quarry.Prepared for the Bureau of Land Management, ElkoResource Area, Intermountain Research Reports,Silver City, NV.

Ericson, J.E., 1984. Toward the analysis of lithicproduction systems. In: Ericson, J.E., Purdy, B.A.(Eds.), Prehistoric Quarries and Lithic Production.Cambridge University Press, Cambridge, pp. 29.

Frison, G., Bradley, B., 1999. The Fenn Cache. ClovisWeapons and Tools. One Horse Land and CattleCompany, Santa Fe, NM.

Goodyear, A.C., 1989. A hypothesis for the use of cryptocrystalline raw materials among Paleoindiangroups of North America. In: Ellis, C., Lothrop, J.(Eds.), Eastern Paleoindian Lithic Resource Use.Westview Press, Boulder, pp. 110.



25/27

Gould, R.A., 1977. Ethnoarchaeology; or, where domodels come from? In: Wright, R.V.S. (Ed.), StoneTools as Cultural Markers. Australian Institute of Aboriginal Studies, Canberra, pp. 162168.

Gould, R.A., Saggers, S., 1985. Lithic procurement in

central Australia: a closer look at Binford s idea of embeddedness in archaeology. American Antiquity50, 117136.

Graf, K.E., 2001. Paleoindian Technological Provision-ing in the Western Great Basin. Masters Thesis,Department of Anthropology and Ethnic Studies,University of Nevada, Las Vegas.

Gramly, R.M., 1980. Prehistoric industry at the Mt.Jasper mine, Northern New Hampshire. Man in theNortheast 20, 124.

Grayson, D.K., 1993. The Desert s Past: A NaturalPrehistory of the Great Basin. Smithsonian Institu-tion, Washington, DC.

Grayson, D.K., Cannon, M.D., 1999. Human paleo-ecology and foraging theory in the Great Basin. In:Beck, C. (Ed.), Models for the Millennium. GreatBasin Anthropology Today. University of UtahPress, Salt Lake City, pp. 141151.

Gurven, M., Hill, K., Hurtado, A., Lyles, R., 2000.Food transfers among Hiwi foragers of Venezuela:tests of reciprocity. Human Ecology 28, 171214.

Hawkes, K., O Connell, J.F., Blurton Jones, N.G., 2001.Hunting and nuclear families. Current Anthropology42, 681695.

Hayden, B., 1977. Stone tool functions in the westerndesert. In: Wright, R.V.S. (Ed.), Stone Tools asCultural Markers. Australian Institute of AboriginalStudies, Canberra, pp. 178188.

Hill, K., Kaplan, H., 1993. On why male foragers huntand share food. Current Anthropology 34, 701 706.

Hiscock, P., 1994. Technological responses to risk inHolocene Australia. Journal of World Prehistory 8,267292.

Holmes, W.H., 1891. Aborignal novaculae quarries inGarland county, Arkansas. American Anthropolo-gist 4, 313316.

Holmes, W.H., 1897. Stone Implements of the Potomac-Chesapeake Tidewater Province. Bureau of Ameri-can Ethnology, Bulletin 15. Smithsonian Institution,Washington, D.C.

Holmes, W.H., 1919. Handbook of Aboriginal Ameri-can Antiquities. Part I. The Lithic Industries. Bureauof American Ethnology Bulletin 60.

Huckleberry, G., Beck, C., Jones, G.T., Holmes, A.,Cannon, M.D., Livingston, S.D., Broughton, J.M.,2001. Terminal Pleistocene/early Holocene environ-mental change at the Sunshine Locality, North-Central Nevada, USA. Quaternary Research 55,303312.

Ingbar, E.E., 1994. Lithic material selection and tech-nological organization. In: Carr, P.J. (Ed.), TheOrganization of North American PrehistoricChipped Stone Tool Technologies. International

Monographs in Prehistory Archaeological Series 7,pp. 4556.

Ingbar, E.E., Larson, M.L., Bradley, B.A., 1989. A non-typological approach to debitage analysis. In: Amick,D.S., Mauldin, R.P. (Eds.), Experiments in Lithic

Technology. BARInternational Series 528,pp. 6799.Johnson, J.K., 1981. Yellow Creek Archaeological

Project v. 2. Tennessee Valley Authority Publicationsin Anthropology no. 2.

Johnson, J.K., 1984. Measuring Prehistoric Quarry SiteActivity in Northeastern Mississippi. In: Butler,B.M., May, E.E. (Eds.), Prehistoric Chert Exploita-tion: Studies from the Midcontinent. Center forArchaeological Investigations Occasional Paper No.2. Southern Illinois University at Carbondale, pp.225235.

Johnson, J.K., 1989. The Utility of Production Trajec-tory Modeling as a framework for regional analysis.In: Henry, D.O., Odell, G.H. (Eds.), AlternativeApproaches to Lithic Analysis Archaeological Pa-pers of the American Anthropological Associationno. 1, pp. 119133.

Jones, G.T., Beck, C., 1990. An obsidian hydrationchronology of late Pleistocene/early Holocene sur-face assemblages from Butte valley, Nevada. Journalof California and Great Basin Anthropology 12, 84 100.

Jones, G.T., Bailey, D.G., Beck, C., 1997. Sourceprovenance of andesite artifacts using non-destruc-tive XRF analysis. Journal of Archaeological Science24, 929943.

Jones, G.T., Beck, C., Jones, E., 2002. Lithic Source Useand Paleoarchaic Foraging Territories in the GreatBasin. American Antiquity, in press.

Jones, G.T., Beck, C., Nials, F.L., Neudorfer, J.J.,Brownholtz, B., Gilbert, H., 1996. Recent archaeo-logical and geoarchaeological investigations at theSunshine Locality, Long valley, Nevada. Journal of California and Great Basin Anthropology 18, 4863.

Jones, K.T., Madsen, D.B., 1989. Calculating the cost of resource transportation: a Great Basin example.Current Anthropology 30, 529534.

Kaplan, H., Hill, K., 1992. The evolutionary ecology of food acquisition. In: Smith, E.A., Winterhalder, B.(Eds.), Evolutionary Ecology and Human Behavior.Aldine de Gruyter, New York, pp. 167201.

Kaplan, H., Hill, K., Lancaster, J., Hurtado, A.M.,2000. A theory of human life history evolution: diet,intelligence, and longevity. Evolutionary Anthropol-ogy 9, 156185.

Kelly, R.L., 1988. The three sides of a biface. AmericanAntiquity 53 (4), 717734.

Kelly, R.L., 1995. The Foraging Spectrum. Diversity inHunterGatherer Lifeways. Smithsonian InstitutionPress, Washington, DC.

Kelly, R.L., 2001. Prehistory of the Carson Desert andStillwater Mountains. Environment, Mobility, andSubsistence in a Great Basin Wetland. University of Utah Anthropological Papers No. 123.



26/27

Kelly, R.L., Todd, L.C., 1988. Coming to the country:early Paleoindian hunting and mobility. AmericanAntiquity 53, 231244.

Kessler, R.A., 2002. Trash: A Study of Great BasinPaleoarchaic Debitage. Poster presented at the 67th

meeting of the Society for American Archaeology,Denver.

Kuhn, S.L., 1994. A formal approach to the design andassembly of mobile toolkits. American Antiquity 59(3), 426442.

Lewontin, R.C., 1974. The Genetic Basis of Evolution-ary Change. Columbia University Press, New York.

Madsen, D.B., 1999. Environmental change during thePleistoceneHolocene transition and its possibleimpact on human populations. In: Beck, C. (Ed.),Models for the Millennium. Great Basin Anthropol-ogy Today. University of Utah Press, Salt Lake City,pp. 7582.

Magne, M.P., 1985. Lithics and Livelihood: Stone ToolTechnologies of Central and Southern Interior B.C.Archaeology Survey of Canada, Mercury SeriesNo.133. Ottawa.

Magne, M.P., Pokotylo, D., 1981. A pilot study inbifacial lithic reduction sequences. Lithic Technology10, 3447.

Metcalfe, D., Barlow, K.R., 1992. A model for exploringthe optimal trade-off between eld processing andtransport. American Anthropologist 94, 340356.

Metcalf, D., Jones, K.T., 1988. A reconsideration of animal body-part utility indices. American Antiquity53, 486504.

Muto, G.R., 1971. A Stage Analysis of the Manufactureof Stone Tools. In: Aikens, C.M. (Ed.), Great BasinAnthropological Conference 1970.: Selected Papers.University of Oregon Anthropological Papers no. 1,pp. 109118.

Nami, H., 1999. The Folsom Biface Reduction Se-quence: evidence from the Lindenmeier Collection.In: Amick, D.S. (Ed.), Folsom Lithic Technology.Explorations in Structure and Variation. Interna-tional Monographs in Prehistory. ArchaeologicalSeries 12, pp. 8297.

Nelson, M.C., 1991. The study of technological organi-zation. In: Schiffer, M.B. (Ed.), ArchaeologicalMethod and Theory, vol. 3. The University of Arizona Press, Tucson, pp. 57100.

Newcomer,M.H.,1971. Somequantitative experiments inhandaxe manufacture. World Archaeology 3, 8594.

O Connell, J.F., Hawkes, K., Blurton Jones, N., 1988.Hadza hunting, butchering, and bone transport andtheir archaeological implications. Journal of Anthro-pological Research 44, 113161.

O Connell, J.F., Hawkes, K., Blurton Jones, N., 1990.Reanalysis of large mammal body part transportamong the Hadza. Journal of Archaeological Science17, 301316.

Orians, G.H., Pearson, N.E., 1979. On the theory of central place foraging. In: Horn, D.J., Mitchell,R.D., Stairs, G.R. (Eds.), Analysis of Ecological

Systems. Ohio State University Press, Columbus, pp.154177.

Pecora, A.M., 2001. Chipped stone tool productionstrategies and lithic debris patterns. In: AndrefskyJr., W. (Ed.), Lithic Debitage. Context, Form,

Meaning. University of Utah Press, Salt Lake City,pp. 173190.

Price, B.A., 1989. Archaeological Survey of Seismic TestLines in Chevron Exploration and Production Ser-vices Pancake Prospect, White Pine, Eureka, andNye Counties, Nevada. Bureau of Land ManagementCultural Resources Report 6-956. Reno, Nevada.

Price, B.A., Johnston, S.E., 1988. A model of latePleistocene and early Holocene adaptation in East-ern Nevada. In: Willig, J.A., Aikens, C.M., Fagan,J.L. (Eds.), Early Human Occupation in Far WesternNorth America: The Clovis-Archaic Interface. Ne-vada State Museum Anthropological Papers 21,Carson City, pp. 231250.

Raab, L.M., Cande, R.F., Stahle, D.W., 1979. Debitagegraphs and archaic settlement patterns in the Ar-kansas Ozarks. Midcontinental Journal of Archae-ology 4, 167182.

Rhode, D., 1990. On Transportation costs of GreatBasin resources: an assessment of the Jones-Madsenmodel. Current Anthropology 31, 413419.

Shott, M., 1996a. An exegesis of the curation concept.Journal of Anthropological Research 52, 259280.

Shott, M., 1996b. Stage versus continuum in the debrisassemblage from production of a uted biface. LithicTechnology 21, 622.

Simms, S.R., 1988. Conceptualizing the Paleoindian andArchaic in the Great Basin. In: Willig, J.A., Aikens,C.M., Fagan, J.L. (Eds.), Early Human Occupationin Far Western North America: The

Date post:	03-Jun-2018
Category:	Documents
Upload:	masterash
View:	216 times
Download:	0 times

Beck, C - Rocks Are Heavy

Documents