Mobility

 

 Journal of Anthropological Archaeology 28 (2009) 382–396

Contents lists available at at   ScienceDirect

 Journal of Anthropological Archaeology journal homepage:  www.elsevier.com/locate/jaa

Climate change, uncertainty and prehistoric hunter–gatherer mobility Christopher Morgan

*

Utah State University, Anthropology, 0730 Old Main Hill, Logan, UT 84322-0730, United States

a r t i c l e

i n f o

 Article history: Received 12 February 2009 Revision Revisi on received received 12 June 2009 Available online 5 September 2009 Keywords: Mobility Hunter–gatherer

Western Mono Sierra Nevada Foraging California Climate change Uncertainty Nearest-neighbor Variance-to-mean

a b s t r a c t

The onset of Little Ice Age conditions in California’s Sierra Nevada mountains resulted in increased tempor poral al and spatia spatiall var variab iabili ility, ty, and hence hence uncert uncertain ainty ty regard regarding ing the distri distribut bution ion and produc production tion of  resources targeted by its inhabitants, the Western Mono. The Mono responded with a risk-averse strategy composed of lowland winter population aggregation supported by logistical forays and seasonal residential dispersals to the high country, both ways of averaging variance in environmental productivity. These patterns patter ns were reconstruc reconstructed ted using surface archaeology archaeology,, GIS, and two straightfor straightforward ward spatia spatiall statistics, statistics, nearest-neighbor and variance-to-mean ratios, that combined provide a robust, objective picture of population aggregation and dispersal and the scale of these phenomena in different environments and seasons. These divers diverse e strate strategies gies conform to expectations expectations regarding regarding the best ways for hunter–gath hunter–gatherers erers to cope with uncer uncertaint tainty, y, parti particular cularly ly in mounta mountain in enviro environment nments. s. Despit Despite e this, the reside residential ntially ly mobil mobile e aspect of the pattern is rare in mountains and probably the result of historical connections between the Mono and Great Basin groups employing similar behaviors. Ultimately, this research suggests that climate change and environmental variability condition risk-averse, satisficing economic behaviors focused more on security than optimization, implying that pronounced environmental variability runs counter to economic intensification and its association with the evolution of more complex societies.    2009 Elsevier Inc. All rights reserved.

Introduction

Mobility Mobi lity is argu arguably ably the prin principa cipall conce concern rn of arch archaeol aeologist ogistss focusing on hunter–gatherers (Kelly, (Kelly, 1992, 1998). 1998). How foragers moved about and exploited prehistoric landscapes is cited as key to unde understa rstandin nding, g, amon among g othe otherr thin things, gs, preh prehistor istoric ic subsiste subsistence nce ((Moore, 1998), ), trade (and Yellin et al., 1996 1996), complexity Keeley, 1998 Keeley, 1988; Price (Yellin Brown, 1985 198 5),), sociocultural and even gender roles, power relationships, and evolutionary trajectories (Hawkes ( Hawkes et al., 1989;; McGu 1989 McGuire ire and Hilde Hildebran brandt, dt, 2005 2005;; Suro Surovell vell,, 2000 2000). ). But the problem for archaeologists is that movement is a transitory, abstract phenomenon rarely leaving direct material evidence (Close, (Close, 2000:49). 2000:49 ). So reconstructing prehistoric mobility is really a middle-range problem (e.g., Binford, (e.g.,  Binford, 1977 1977)) subsuming the secondary challenges of identifying relevant archaeological proxies for movement and eluc elucidat idating ing mean meaningfu ingfull patt patterns erns within and betw between een these proxies. This stud study y incor incorpora porates tes expe expectat ctations ions deri derived ved from ecolo ecological gical theory and two relatively straightforward statistical techniques that together provide a clear picture of the adaptive nature of late prehistoric mobility in California’s Sierra Nevada. Nevada. It operates on the perspective that though movement, especially pedestrian hunter– gatherer movement, usually leaves no trace, stopping does, with *

 Fax: +1 435 797 1240. E-mail address:  [email protected]

0278-4165/$ - see front matter    2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jaa.2009.07.004 doi:10.1016/j.jaa.2009.07.004

evidence of stops between moves determined by the distribution of bedrock processing features. The statistical methods it employs, the nearest-neighbor statistic and variance-to-mean ratios, measure dispersal dispersal and scale of dispe dispersal rsal,, resp respecti ectively vely,, of stopping stopping places in different ecological zones in different seasons. Combined, these provide a robust, objective picture of how and why people used late Holocene Sierran landscapes techniques whose gic sho should uld be cle clear ar to an anyon yone e wit with h using a ba basic sic und unders ersta tandi nding ng loof  descriptive statistics. More importantly, the study identifies how population size, resource characterist characteristics ics and climatic variability condition hunter–gatherer mobility. Ultimately this research describes a case where different forms of mobility were used to cope with pronounced environmental uncertainty resulting from late Holocene Holoc ene climate change change,, the implicat implications ions of which which show how hunter–gatherer economic behaviors evolve to cope with spatial and temporal variability in the distribution of mountain resources and, more importantly, uncertainty resulting from climate change and climatic variability. Conceptualizing mobility 

At its most fundamental level, mobility is way of bringing consumers to resources and is perhaps the simplest way of averaging temporal and spatial variations in resource productivity (Halstead ( Halstead and O’Shea, 1989:3; Woodburn, Woodburn, 1980 1980). ). From a basic ecological perspective, mobility is critical to understanding how people manage

 

C. Morgan / Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

 

383

relationship relation shipss betw between een popu populati lations ons and reso resource urcess and how they cope with uncertainty with regard to where and when resources become available (Kelly, (Kelly, 1992). 1992). Binford  Binford (1980), (1980), of course, codified thinking along these lines in the forager–collector model, correlating residential and logistical mobility with specific environments, defined by effective temperature. Derivations of this model (e.g., Ames,, 2002 2002;; Kell Kelly, y, 198 1985; 5; Pren Prentiss tiss and Chat Chatters ters,, 2003 2003)) typically Ames equate residential residential mobility with smaller groups living in aseasonal, homogenous environments were resources are patchy in neither time nor space, such as the tropics, or where resources are partic-

lder and Lesbenefits of accessing a new resource patch (Winterha (Winterhalder lie, 2002). 2002). Environmental variability thus forces foraging decision makers to manage risk by coping with the probabilities of different 1990). ). In this outcomes that may result from their move (Low, (Low, 1990 light, though residential mobility can solve the immediate problem of temp temporal oral and/ and/or or spatial spatial reso resource urce depressi depressions ons when return ratess are relat rate relativel ively y high high,, it also enta entails ils coping with the risk of  moving, movi ng, elem elements ents exa exacerb cerbate ated d whe when n envi environm ronmenta entall vari variabil ability ity d with movis also high (Goland, (Goland, 1991). 1991). Because  risks  associate  associated ing increase under more variable environmental conditions, ex-

ularly dispersed. In contrast, logistical mobility is modeled as affiliated with larger populations living in seasonal climates, where resource reso urcess are patchy in time and/or space space.. Patc Patchine hiness ss requires requires intensifie inte nsified d sett settleme lement nt and subs subsiste istence nce beh behavior aviorss like logistica logisticall moves and the division of labor to provision larger groups. Actual behaviors, of course, rarely meet precise definitions of either form of mobility, often taking on aspects of the multiple behavioral options available in the logistical–residential continuum (e.g.,  (e.g.,   Krist, 2001). 2001 ). The behavioral and semantic distinction between logistical and residential mobility, however, continues to be a useful heuristic device (e.g., Fitzhugh (e.g.,  Fitzhugh and Habu, 2002). 2002 ). Beyond basic ecological context, many attempts to understand why and when hunter–gatherers choose one form of mobility over another employ cost–benefit models to ascertain when it is more efficient to transport resources back to a central place residence (i.e., be logistically mobile) and when it is more efficient to move residential location (i.e., be residentially mobile) to reduce logistical travel and resource transport costs (Kelly, (Kelly, 1990, 1998 1998). ). These models predict optimal foraging radii around central places, beyond which it is more efficient to move camp rather than spend more time travelin traveling g and tran transport sporting ing reso resource urcess in a logis logistica ticall round. The most simple of these models predicts that the maximum size of this radius (in the Great Basin, 37–812 km) is determined min ed by the the point point at which which the calori caloricc cost cost of transp transport orting ing resources equals the caloric content of the resources being trans1989). This ultimately means the type ported ( Jones and Madsen, 1989). and quality of resources (or the habitat containing these resources) determines, to a large extent, the size of foraging radii and the distance at which it becomes more efficient to move residence (e.g., 2009). More complex models argue that it is more efficient Grove, 2009). to logistically transport unprocessed resources in small foraging radii (typically between 1.5 and 3.6 km for tree nuts like acorn and piñon) and to field process resources in much larger radii (as much as an absurdly large 1065 km for tree nuts) because field Bettprocessing removes low-yield low-yield bulk and increases load utility ((Bettinger et al., 1997; Barlow and Metcalfe, 1996). 1996 ). The implications of  this type of modeling are that logistical procurement is usually quite costly relative to residential procurement and is only more efficient when resources are abundant enough and foraging radii small enough to reduce combined foraging and travel costs ( Bettinger et al., 1997:897), 1997:897), or when population pressure reduces resi2000:12– :12–13 13). ). But But dentia den tiall pro procur cureme ement nt re retur turn n rates rates (Zeanah, Zeanah, 2000 residential moves can also be costly and are only more efficient than logistical procurement procurement when diet breadth is narrow, or broader-spectrum, lower yield resources like nuts and seeds are abundantt eno dan enough ugh to yie yield ld ret return urn rates rates above above tho those se of log logist istica icall procurement (Zeanah, (Zeanah, 2002:242 2002:242). ). Ultimately, this perspective argues that people should opt for residential over logistical procurement when transport costs exceed the one-way travel threshold (i.e., the foraging radius) of key resources, determined mainly by

tr trem eme e en envi viro ronm nmen enta tall va vari riab abil ilit ity y coul could d co conc ncei eiva vabl bly y favo favorr increased sedentism supported by logistical forays. Alternatively, random search strategies and larger catchments have been found to gene generate rate optima optimall solut solutions ions to findin finding g diffu diffuse se or randomly randomly distributed resources, especially when information on resource dish and tr tribu ibutio tions ns and abund abundanc ance e is poo poorr or absen absentt (Armswort Armsworth Roughgarden, Roughgar den, 2003; Brantingha Brantingham, m, 2006). 2006 ). This suggests a corresponding spond ing pattern of high reside residentia ntiall mobil mobility ity may be the best way of coping with unpredictable circumstances. Finally,   Goland (1991:110)   ar argues gues that unpr unpredict edictable able circu circumsta mstances nces require require ‘‘flexibl ‘‘fle xible e stra strategi tegies,” es,” implying implying that incor incorpora porating ting logis logistica ticall and residential mobility types may be the best way of coping with extreme environmental variability. The arch archaeol aeology ogy of hunter– hunter–gath gathere erers rs in moun mountain tain envi environronments sheds additional light on these problems. Almost without exception exce ption,, moun mountain tain envi environm ronments ents are seen as high highly ly seas seasonal onal and and ma margi rginal nal in terms terms of resou resource rce pro produc ductiv tivity ity (Aldenderfer 2006;; but see Walsh, 2006 see  Walsh, 2005; Walsh et al., 2006 2006). ). Hunter–gatherer Hunter–gatherer adaptati adap tations ons to thes these e condi condition tionss cons conseque equentl ntly y tend to focus on male-dominated logistical hunting, a pattern seen in North America’s Rocky Mountains (Bender (Bender and Wright, 1988; Wright et al., 1980), 1980 ), Sierra Nevada (McGuire (McGuire et al., 2007; Stevens, 2005), 2005), and Gr Grea eatt Ba Basi sin n (Bett Bettinge inger, r, 199 1991; 1; Thom Thomas, as, 198 1982 2); in the the Ande Andess Aldenderfer, r, 1998, 1999; Rick, 1980); 1980); and on the Tibetan and Ethi(Aldenderfe opian Plateaus (Aldenderfer, (Aldenderfer, 2006 2006). ). Only in rare circumstances do hunter–gatherers hunter–g atherers build residential structures indicating some form  Thomas, 1982), of residential residential mobility (e.g., (e.g., Thomas, 1982), a phenomenon that appearss to be conditi pear conditioned oned by popu populati lation on pressure pressure,, at least in the Great Basin (Bettinger, (Bettinger, 1991; Zeanah, 2000 2000). ).   Aldenderfer (2006) explains these patterns by arguing that because mountain environments are marginal, seasonal, patchy, and uncertain with regard to resource productivity, people should be risk-averse, choosing to employ empl oy litt little le resi resident dential ial mobil mobility ity and supp support ort themselv themselves es with logistica logis ticall fora forays ys in smal smalll catc catchmen hments. ts. Ulti Ultimate mately ly risk risk-ave -aversio rsion n under these conditions is akin to economic satisficing, ensuring minimum targeted requirements for subsistence rather than maximizing return relative to labor, this conditioned mainly by the high costs of gathering sufficient information upon which to make Bordley ey and LiCalzi, 2000; Simon Simon,, 1957 1957). ). In optimal opti mal deci decisions sions ((Bordl other words, risk-aversion is a way of hedging one’s bets that at least   one  behavior (or mobility option) will solve the problem of  meeting meet ing subsiste subsistence nce targ targets ets when good infor informat mation ion regarding regarding probable decision outcomes are hard to come by, a situation exacerbated by increased environmental variability.

its caloric yield relative toalso weight. But hunter–gatherers face variability, or rather nonnorma-

eastern edge of California and an area occupied in late prehistoric and ethnographic times by a group known as the Western Mono (Fig. 1 1). ). Mono populations were small (usually no more than 39 people to a group) and widely dispersed, with families and kin occupyin occu pying g smal small, l, polit political ically ly auto autonomo nomous us hamlets hamlets alon along g or near the San Joaquin Joaquin,, Kings, and Kaw Kaweah eah rive rivers rs (Kroe Kroeber, ber, 192 1925 5). Hamlets

tive resource return rates (Winterhalder, ( Winterhalder, 1980). 1980). Variations in environmental productivity result in uncertainty for hunter–gatherers deciding whether or not to move: a group or individual leaving one place risks that the costs of moving will not be offset by the

Ethnographic and paleoenvironmental setting 

The focus of this study is the late Holocene in California’s southwestern Sierra Nevada, a 4000 m high mountain range along the

 

384

 

C. Morgan/ Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

Fig. 1.   Location of the study area; Mono groups are in bold italics.

tended to cluster around springs, streams and flats along canyon margins immediately below winter snowline (Gayton, ( Gayton, 1948; Gayton,, n.d ton n.d.; .; Gif Giffor ford, d, 193 1932; 2; Gif Giffor ford, d, n.d n.d..). Subsi Subsiste stence nce was was bas based ed mainly on deer (Odocoileus hemionus), salmon (Oncorhynchus sp.) and especially acorn. Acorn, particularly black oak ( Quercus kelloggii) acorn was a portion of nearly every meal and is still a form of Mono identity (Lee, ( Lee, 1998; McCarthy, 1993). 1993). Evidence of acorn processing is ubiquitous in the area, with outcrops of granite bedrock often pocked with bedrock mortars (BRM) used to reduce acorn into flour, especially in hamlet locations (Aginsky, ( Aginsky, 1943; Driver, 1937; 1937; Giff Gifford, ord, 197 1971; 1; Hindes, Hindes, 196 1962 2). Hamlets Hamlets wer were e suppor supported ted in large part by dried, but otherwise unprocessed acorn cached in 5 km radii centered on settlement areas (Morgan, ( Morgan, 2008). 2008). Like most hunter–gatherers living in mountain environments, the Mono exploited a patchy resource base (Aldenderfer, (Aldenderfer, 2006 2006). ). Mountain resources are generally patchier than in adjoining valleys and other physiographic provinces because of the effect orogr grap aphi hicc pr prec ecip ipit itat atio ion n has has bi biot otic ic zone zone di dist stri ribu buttio ion n an and d composition. Put simply, there tends to be much more precipitation at altitude, with growing seasons constrained by temperature

tween 1000 and 4000 m elevation (Table (Table 1 1). ). To the southwest, in the adjoining San Joaquin Valley, he maps only two (Valley oak savanna and San Joaquin saltbush) over the same transect orientatio tion n and dista distance nce.. Thi Thiss indica indicate tes, s, at lea least st for this this regio region, n, tha thatt mountain environments are on the order of 3.5 times more diverse (i.e., (i.e., ‘‘pat ‘‘patchy” chy”)) than valley sett settings. ings. Though each zone cert certainl ainly y contains smaller resource patches (e.g., oak groves, lithic sources, and water), as a whole they comprise large patches that are fundamentally distinct yet relatively nearby adjacent, but very different patches containing very different resources. To simpl simplify ify the anal analyses yses contai contained ned in this paper paper,, the southwestern Sierra Nevada is modeled as containing three main ecozones subs subsumin uming g Küch Küchler’s ler’s seven biot biotic ic zones zones,, each producin producing g distinct disti nct reso resource urcess at diffe different rent elevati elevations ons and at diffe different rent times of the year (Table (Table 1; 1; Fig. 2). 2). The lower montane forest  is  is below winter snowline (1400 m) and contains blue oak ( Quercus douglassii) parklands, chaparral, and major streams; it yields abundant, seasonal runs of fish, winter deer-hunting opportunities, and spring and summer grass seeds, berries berries and acorn. The  montane forest  (between 1400 and 2100 m) is composed mostly of conifers (Pinus

and snowpack high altitudes. This situation results in in themouncompositionally andatelevationally discrete biotic zones found tains, tain s, each compr compresse essed d into relative relatively ly smal smalll horiz horizonta ontall spac spaces es (Holdridge, 1967; Leemans et al., 1996). 1996). To illustrate this, on the west slope of southern Sierra Nevada,  Küchler (1977) (1977) maps  maps seven distinct biotic zones across a 40 linear km, NE–SW transect be-

keland   Abies   sp.) andelevations, stands of acorn-producing black oak (Q.deerloggii ). Like lower the montane forest afforded hunting in the spring, summer, and fall and a rich fall acorn harvest, which was cached in dispersed locations and stored within 2008). ). The third is a resource-poor  subalpine–alhamlets (Morgan, (Morgan, 2008  pine  zone above 2100 m.

 

C. Morgan / Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

 

385

 Table 1

Study area ecozones.

*

*

Ecozone

Biotic zone

Elevation

Lower montane

Blue-oak grey pine forest Chaparral Sierra yellow pine forest

Below 1000 m 1000–1350 m 1000–1400 m

     

Quercus douglassii, Pinus sabiniana Adenostoma fasciculatum, Arctostaphylos sp.,  Ceanothus  sp. P. ponderosa

Montane forest

Sierra montane forest

1400–2100 m

 

Abies concolor, P. lambertiana, P. ponderosa, Q. kelloggii

Subalpine

Upper montane-subalpine forest Northern jeffrey pine forest Alpine

2100–3300 m 2000–2400 m Above 3300 m

   

Abies magnifica, P. contorta P. jeffreyi Limited vegetation, e.g.,  Draba oligosperma, Erigonum ovalifolium

Dominant vegetation

After Küchler After (1977).  Küchler (1977).

Because the Mono occupied the western slope of the Sierra Nevada in ethnographic times and late prehistory (i.e., since about 600 BP and perhaps earlier), they faced not only patchy resource distribu dist ribution tions, s, but also the effe effects cts of substant substantial ial late Holocen Holocene e climatic clim atic change (Gayt Gayton, on, 1948 1948;; Giff Gifford, ord, 193 1932; 2; Kroe Kroeber, ber, 1959 1959;; Morgan, 2006). 2006). Paleoclimatic and paleoenvironmental data indicate climatic variability and disequilibrium were the norm for at least the last 2000 years, especially during the Medieval Climatic 3). ). The MCA was Anomaly (MCA) and Little Ice Age (LIA) (Fig. (Fig. 3 characte char acterize rized d by at leas leastt two extr extreme eme and pers persiste istent nt droug droughts hts interspersed intersper sed by wetter periods between about 1300 and 650 calBP. The LIA was characterized by cooler conditions and glacial advance between about 650 and 150 calBP (Graumlich ( Graumlich and Lloyd, 1996; Graumlich, 1993; Hughes and Brown, 1992). 1992). This means the Mono

occupied the area during a climatic regime dominated by disequilibrium, variability and a substantial shift within this regime: the transition to and dominance of LIA conditions. Abundant research on global warming’s effect on North American forests models how biotic composition should change during warm/dry war m/dry to cool/ cool/wet wet condi condition tionss (and vice vers versa). a). Qualitat Qualitativel ively y (and unfortunately not quantitatively) LIA conditions are modeled as resulting in oak communities rapidly contracting, the density of  oak in mont montane ane forests decreas decreasing, ing, and the dist distincti inction on between between elevatio elev ation-d n-deter etermine mined d biot biotic ic zones beco becoming ming more pron pronounce ounced d due to the constriction of plant ranges, the development of distinct alpine and subalpine communities, and more pronounced seasonality associated with increased snowfall (Campell (Campell and McAndrews McAndrews,, 1993; Peterson, 1998; Woolfenden, 1996). 1996). This means that during

Fig. 2.   Study area ecozones.

 

386

 

C. Morgan/ Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

Medieval Medi eval Climati Climatic c Anomaly Anomaly Little Ice Ice Age Age

? Temperature:

Holocene Mean

?

Precipitation:

Historic Mean

Drought: Glacial Advance: Date Dat e B.P B. P.

2000

1500

1000

500

0

Fig. 3.   Idealized late Holocene Sierra Nevada paleoclimate synthesis.

the LIA the distribution of resources essential to Mono subsistence became increasingly constrained to patches exploitable for only short periods each year. For example, grassland and lower-elevation nut-producing trees like blue oak were constrained to elevations tions bel below ow 10 1000 00 m (Allen Allen and Bre Breshe shear ars, s, 19 1998 98). ). Sug Sugar ar pin pine e (Pinu Pinuss lambe lambertian rtiana a) and bl black ack oak we were re con constr strain ained ed betwe between en 1000 and 2000 m, in lower elevations by water budgets and at higher elevations by snowpack and shorter growing seasons (Mill(Miller et al., 2004; Urban and Miller, 1996). 1996 ). Resource-poor, subalpine red fir ( Abies magnifica) and lodgepole pine (Pinus contorta) forests expanded at the expense of more xerically-adapted montane forests containing sugar pine and black oak (Körner, ( Körner, 1998; Overpeck et al al., ., 19 1990 90). ). Final Finally, ly, tre treelin eline e elev elevatio ations ns migrated migrated down downslope slope,, increasi incr easing ng the geogr geographi aphicc ext extent ent of reso resource urce-poor -poor alpine biot biotic ic communities (Scuderi, (Scuderi, 1987) 1987) (Fig. 4 4). ). The result of these transformations was a significant change in

of key resources to specific biotic zones severely curtailed their availability in both time and space, in the first case by limiting them to very space-specific resource patches and in the second by increasing the effects of seasonality on resource productivity. This latter phenomenon is expressed in more pronounced variability (in both amplitude and frequency) of masting (i.e., the tendency of nut bearing trees to periodically produce ‘‘bumper” yields interspersed by years of poor or absent production) during colder and Sork, k, 200 2002; 2; Koe Koenig nig an and d Kn Knops ops,, wetter wett er condi condition tionss (Kelly Kelly and Sor 2005; McKone et al., 1998) 1998)  (Fig.  ( Fig. 5 5). ). Overall, LIA conditions led to pronounce prono unced d vari variabil ability ity of key reso resource urce produ production ction in spat spatiall ially y and temporally discrete resource patches, with no guarantee of  adequate production in any given year. Distances between productive resource patches increased as well because the gentle western slope (avg. slope 4) of the Sierra Nevada results in substantial horizontal distance between elevationally-discret elevationally-discrete e ecozones. Com-

resource avai resource availabi lability lity,, acce accessib ssibilit ility y and distr distribut ibution. ion. Incr Increase eased d snowfall snow fall lowe lowered red aver average age year yearly ly snow snowline line,, redu reducing cing acce access ss to and limiting the produ productiv ctivity ity of middle middle and high elev elevatio ation n resources like black oak acorn and sugar pine nut. The constriction

bined, increa bined, increased sed pat patchi chines nesss and incre increase ased d varian variance ce in the the environment’s production of key resources resulted in considerable uncertainty regarding resource availability from year to year and from location to location.

TIME

 Alpine

 Alpine

Subalpine Resource Rich Ecozones:

During the MCA, key resources (nut-producing Quercus douglassii, Q. kelloggii  and  and Pinus lambertiana) are modeled as endemic to higher elevations and multiple ecozones.

Montane Forest

Chaparral

Subalpine Montane Forest Chaparral

Foothill

Medieval C limatic Anomal y, 1300-650 B.P.

Foothill

Resource Rich Ecozones:

During the LIA, key resources areconstrained modeled as becoming to specific ecozones and lower elevations.

Little Ic e Age, 650-150 B. P.

Fig. 4.   Graphic representation of modeled changes in ecozone composition and distribution between the MCA and LIA.

 

C. Morgan / Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

mean

Year 

1

2

3

4

5

6

7

8

9

 Acorn Yield: Warm/Dry Conditions

mean

Year 

1

2

3

4

5

6

7

8

9

 Acorn Yield: Cool/Wet Conditions

Fig. 5.  Idealized representation of climate-induced masting variability.

Modeling Mono mobility 

These conditions result in a mixed set of predictions for Mono mobility mobi lity.. At the most basi basicc ecolo ecologica gicall leve level, l, Mono populat populations ions were relatively small and the resources they exploited were patchy and dispersed across the broad western slope of the Sierra Nevada. These conditions tend to favor residential mobility, with populations mapping onto resource patches as they become productive. But the environment the Mono exploited was also highly seasonal, marked mark ed by subs substant tantial ial reso resource urce depr depressi essions ons in wint winter er mont months. hs. These conditions suggest the Mono should occupy low-elevation, below snowline residential bases near ecotones maximizing exposure to multiple highly productive resource patches. Here, they would woul d supp support ort them themselv selves es with logistical logistical fora foraging ging,, hunt hunting ing and fishing forays, with storage offsetting winter resource shortfalls. Above snowline resources, however, were also critical to the Mono and available for a short period of time each year. Berries produce in the summer, deer move to higher elevations in the summer and oak and grasses produce harvestable seeds in the fall. Resource distribution here, however, is more dispersed and homogenous than below snowline (SNEP, (SNEP, 1996). 1996). Hence the main limit to acquiring resources above snowline is time compression, where a fundamentally homogenous resource base is exploitable for a short period of  time. The goal in this environment is simply group sustenance rather than storing enough food to get through the winter, as it is below snowline. Since above snowline the group is not focused on storing food and and is trying to e exploit xploit what is essentially a homogenouss reso enou resource urce macr macro-pat o-patch, ch, resi resident dential ial mobi mobility lity is favo favored. red. Resident Resi dential ial mobi mobility lity allows grou groups ps to map onto dispersed dispersed resources sour ces and to move to equa equally lly attract attractive ive areas areas once resour resources ces are are exh exhaus auste ted, d, me meani aning ng tha thatt combin combined ed wit with h low pop popula ulatio tion n density, dens ity, abov above e snow snowline line settlem settlement ent shou should ld focu focuss on resident residential ial procurement. Betting tinger er et al. Cost–ben Cost –benefit efit model modelss supp support ort this asse assertio rtion. n.   Bet (1997:895)   indicate indicate that the oneone-way way travel thre threshold shold (beyond which caloric costs outweigh the benefits of resource transport) for butisunprocessed oak the key for the the dried Mon Mono, o, 3.6 3.67 7 km, a black pre predic dictio tion nacorn, corrob corrobora orated ted resource by   Morgan (2008:254) who (2008:254)  who identifies a mean 3.4 km black oak acorn-focused foraging radius around lower elevation winter settlements. This means that if the Mono wanted to exploit resources, especially black oak acorn in the above-snowline montane forest, the vast

 

387

majority of which is well more than 3.5 km from lowland winter settlements (the shortest distance from winter settlements to the lower montane forest boundary varies from 0.6 to 3.2 km, averaging 1.6 km; the distance from winter settlements to any given random point (n = 113 113,760 ,760)) in the mont montane ane forest, forest, however, however, vari varies es from 0.6 to 47.8 km, averagi averaging ng 18.6 km), it is more efficie efficient nt to move residence to the montane forest than transport unprocessed acorn back to a winter village in the lower montane forest. Pre-travel processing could conceivabl conceivably y increase acorn load utility enough to offset greater travel costs, but would have the effect of producing meal extremely prone to spoilage and thus not well suited to storing as an overwintering strategy. Predictions are more variable with regard to managing risk and coping with uncertainty. More variable conditions result in greater uncertainty (due to poor or absent information) and potential risk (due to increased chances of resource failure) associated with residential moves. Because of this, logistical procurement and longrange logistical logistical hunting thus appea appearr to be tthe he default in mountain mountains, s, and we might expect this to be the case for the Mono. But variable conditions can also result in potentially great greater er rewards. For example, substantial substantially ly grea greater ter returns on fora foraging ging would be possi possible ble when variations in environmental productivity result in increased acorn masting and/or overall greater biotic productivity, meaning residential residenti al moves might be used to exploit montane resources outside hamlet logistical radii when such conditions prevail. Variable conditions also favor random moves as a way of optimizing resource encounter rates (Brantingha (Brantingham, m, 2006 2006), ), a situation that might also favor correspondingly correspondingly random residentia residentiall moves. Similarly, exploitation of larger portions of the landscape has been modeled as an optimal solution to subsistence in fluctuating environments (Armswort Armsworth h and Roughgarden, 2003), 2003), a situation also favoring residential mobility due, once again, to the costs of transporting key resources more than 3.6 km back to settlements. Thus, between occasional, but unpredictable increased chances of reward favoring random-dispersed residential moves and a general climatic-ecological context favoring risk-averse behaviors like logistical procu cure reme ment nt,, it thus thus ap appe pear arss that that Go Gola land nd’s ’s argu argume ment nt that that unpredictable circumstances favor flexible strategies is applicable in the Mono case and that multip multiple le mobi mobility lity and proc procurem urement ent strategi stra tegies es shou should ld be empl employed oyed to cope with this variab variabilit ility. y. In sum, ecological context, black oak acorn return rates, and unpredictable dict able circ circumst umstance ancess sugg suggest est that effe effectiv ctive e hunt hunter–g er–gathe atherer rer exploitation of the late Holocene Sierra Nevada required a combination of logistical mobility below snowline and residential mobility above above sno snowli wline, ne, leavin leaving g the pro proble blem m of ide identi ntify fying ing the presence, absence, and extent of these behaviors.

 Analyses

Archaeological approaches to reconstructing mobility have historically centered on settlement pattern analysis using site types and diachronic variations in their distribution across landscapes to iden identify tify seasona seasonal, l, year yearly, ly, and chro chronolog nology-sp y-specifi ecificc chan changes ges in Bettinger, 1977; Mortensen, 1972; Thomas, 1973; Wil use ( (Bettinger, land ley, 195 1953 3). Spatial Spatial stat statistic isticss have been used almost through throughout out the history of sett settleme lement nt patt pattern ern stud studies ies to provi provide de obje objectiv ctive e measures meas ures of artifact artifact,, site and feat feature ure dens density ity and dist distribu ribution tions, s, typically using nearest-neighbor and goodness-of-fit tests to measure degrees degrees of associat association ion with environ environment mental al and othe otherr vari vari-abless thou able thought ght to cond condition ition settlement settlem entand patt patterni erning ng 1987; (Attw Attwell ell and Fletcher, 1987; Bettinger, 1979; Gould Yellen, Hodder and Hassell, 1971; Pinder et al., 1979; Washburn, 1974; 1974 ; but see Voorrips and O’Shea, 1987). 1987). This approach was largely abandoned (until recently) due to a perceived misapplication of some statistical methods and a reconceptualization of space by post-proces-

 

388

 

C. Morgan/ Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

sualists and proce sualists processua ssualist listss alik alike e (Baxter, Baxter, 2008 2008;; Kant Kantner, ner, 2008 2008;; Thomas, 1976; 1976; but see Kowalewski, see  Kowalewski, 2008; Stanish, 2003 2003). ). Recent studies, studi es, howe however, ver, have returned returned to sett settleme lement nt patt pattern ern anal analysis ysis 2008), ), seve several ral (e.g.,   Burke, 2006; Jochim, 2006; Underhill et al., 2008 incorpora incor porating ting GIS to pred predict ict and anal analyze yze prehistor prehistoric ic sett settleme lement nt and mob mobili ility ty ( Jones, 2007; Miller and Barton, 2008; Morgan, 2008)) and others employing spatial statistics to measure spatial 2008 patterni patt erning ng (Bevan Bevan and Con Conoll olly, y, 200 2004; 4; Fle Fletch tcher, er, 200 2008; 8; Pre Premo, mo, 2004; 200 4; Zha Zhang ng et al. al.,, 200 2007 7). The latter latter ten tend d tow towar ard d usi using ng mor more e sophisticated statistics (e.g., Ripley’s K, K-means analysis, kernel density estimates) that cope with some of the perceived shortm, 199 1993; 3; Conn Connolly olly and coming com ingss of earli earlier er mea measur sures es (Blankhol Blankholm, Lake, Lak e, 20 2006 06). ). Though Though metho methodol dologi ogical cally ly strong strong,, pro proble blems ms wit with h these thes e latt latter er appr approach oaches es center center on thei theirr opacity opacity to rese research archers ers without with out advance advanced d tra training ining in stat statistic isticss and their failu failure re to link theory theo ry and pred predictiv ictive e mode models ls with spat spatial ial arch archaeol aeologic ogical al data data.. The following analyses make these linkages using straightforward statistical techniques that combined provide a robust picture of  spatial point patterning and how these patterns relate to the ecology of mobility oriented hunter–gatherer adaptive systems. The premise behind these analyses is that bedrock milling stations are precise, in situ indicators of both the location and intensity of Mono resident residential ial choi choice ce and logi logistica sticall field processi processing ng 1985). ). The use of bedrock milling stations (Haney, 1992; White, 1985 as direct measures of Mono processing and as proxy measures of  Mono settlement settlement is pred predicat icated ed on the assumption assumption that balanobalanophagy (acorn eating) was the basis of the Mono economy and that processing acorn in bedrock milling stations was a fundamental attribute of this economy. It is also based on the fact that though the Mono may or may not have used every milling station in the study area over the course of their tenure in the Sierra Nevada, they certainly knew of most their locations, reusing and revisiting milling milli ng stat stations ions that may have bee been n manu manufact factured ured by earl earlier ier 1984). ). This assumption is based on the fact inhabitants ( Jackson, 1984 that markers of Mono ethnicity are found throughout the study area,, almo area almost st inva invariab riably ly close closely ly associat associated ed with milli milling ng stat stations, ions, indicating they travelled to and used nearly every milling station in their territory (McCarthy, (McCarthy, 1993; Morgan, 2006 2006). ). Because milling stations are so closely allied with Mono economics and settlement and were repeatedly reused in the course of any given year or season, a cumulative sample is used to approximate the norm of Mono behaviors for the period of time they most clearly occupied the study area (i.e., the last 600 years, during the LIA), a method bene-

 Table 2

fiting the large samples developed forartifact the study. Finally, the use offrom stationary features, rather than say scatters, avoids some of the problems associated with performing landscape-scale analyses with only the arguably arbitrary conception of what comprises a site (e.g.,  Dunnell, 1992) 1992) and with taphonomic problems associate assoc iated d with site forma formation tion and pres preserva ervation tion (see   Zvelebil et al., 1992:196–197). 1992:196–197). The geographic distribution of milling stations speaks directly to residential choice and logistical field processing. Though mortars are unequivocally associated with field processing, they are also associated with habitation and residence. Larger, more intensively and repeatedly occupied sites contain the greatest number of BRMs (i.e.,   P14) and are usually associat associated ed with substa substantia ntiall middens, midde ns, hous house e pit depr depressi essions ons and othe otherr resi resident dential ial indi indicato cators rs 2006). ). They (Gifford, 1932; McCarthy et al., 1985:307; Morgan, 2006 thus stand as proxy for residential choice, and by inference, residential mobility. Fewer milling surfaces (i.e., <14) are associated

2 or less transect spacing) covering surveys over 555(i.e., km2 those of thewith 162615 kmm study area performed over the last 60 years for timber sales and other large projects. Survey coverage was appr approxim oximatel ately y 34% more more-or-or-less less equally equally dist distribu ributed ted acro across ss study area ecozones (Table (Table 3 3), ), with smaller portions of the subalpine zone covered mainly due to extremely steep slopes (i.e., greater than 30 ). This large sample proportion and relatively consistent coverage between ecozones is believed to accurately reflect milling station distribution in the study area and across ecozones. The first analysis uses nearest-neighbor statistics statistics to determine the distribution of processing sites in different ecozones as a way of ascertaining seasonal seasonal variabil variability ity in popul populatio ation n clust clusterin ering g and dispe dispersal rsal,, these thes e indic indicatin ating g seas seasonal onal vari variatio ations ns in resi resident dential ial mobi mobility lity.. The second analysis uses variance-to-mean ratios as a way of objec-

with logistical stations used solely for field processing and as temporary campsites (Hindes, (Hindes, 1962; Jackson, 1984; TCR/ACRS, 1984); 1984 ); they thus stand proxy for logistical mobility. This functional distinction allows for reconstructions of Mono logistical and residential moves and recognition of different site types based primarily on BRM counts (Table (Table 2 2). ).

Site types and defining characteristics.

Mobili Mob ility ty type type

BRM (n)

Site type

Associated features

Res esid ide ent ntia iall

25+ 14–2 14–24 4

Pr Prin incc ip ipal c amp amp// hamlet Su Subs bsid idia iary ry camp camp

Midden, housepit, artifact scatter Ar Arti tifa fact ct scat scatte ter, r, midd midden en (rare)

5–1 –13 3 1–4 1–4

Te Temp mpor orar ary y cam amp p Proc Proces essi sing ng stat statio ions ns

L it ithi hicc sca sca tt tt er er Lith Lithic ic scat scatte terr (rar (rare) e)

L ogi ogist stic ical al

Milling station geographic distribution also speaks directly to seasonality of occupation. Unlike many lower elevation settings where seasonality is reconstructed, reconstructed, with some difficulty, using faunal and floral indicators (e.g., (e.g., Adams  Adams and Bohrer, 1998 1998), ), movement patterns and settlement in the Sierra Nevada were strongly condiwinter  snowpack,, a fact corroborated by ethnographic tioned by  by   winter snowpack  Gifford (1932:17 (1932:17))  writes that the Mono moved, ‘‘Annually sources. Gifford sources. from lower winter to higher summer residences which had too much snow in the winter and vice versa.” Multiple ethnographic sources attest to the location of lower-elevation residences, all below average winter snowline (1400 m) in the lower montane forest. Residences here were occupied mainly in the winter, where the ground was free of snow, and where people subsisted mainly on stored acorn gathered in the fall (Gayton, ( Gayton, 1948; Gifford, 1932; Merriam, 1955). 1955). Ethnogra Ethnographic phic docum documenta entation tion of highland highland resi resi-dence (i.e., in the montane forest and subalpine zones, above winter snowline) is poor (the preceding quotation is among the few references to highland settlement). Residential sites here, though occasionally accessible during the winter during brief thaws interspersing sper sing winter storms woul would d have usually been buri buried ed unde underr more than a meter or more of snow for 4 months or more each winter, precluding occupation in an area where snow-free habitation sites were available in many cases less than 10 km downslope. Further, biotic productivity in higher elevations is markedly constrained during winter months (i.e., December–March) and migrator tory y game game like like dee deerr win winter ter in the low cou county nty,, leavin leaving g little little incentive incen tive to occup occupy y high higher er elev elevatio ations. ns. Resi Resident dential ial occu occupati pation on above winter snowline was thus mainly a spring-fall phenomenon constrained by ecological circumstances. Based on these parameters, analyses focus on the distribution of  processing features in a roughly 30 km 2 study area in the San Joaquin River watershed (Figs. (Figs. 1 and 2 2). ). They use data from intensive

 Table 3

Intensive survey coverage, by ecozone.

Ecozone Lower montane Montane Subalpine Total

Area (km2 )

Area Area survey surveyed ed (km2)

Survey Survey cov covera erage ge (%)

407.19 369.96 848.25

165.59 148.33 237.57

40.67 40.09 28.00

1626.48

551.49

33.91

 

C. Morgan / Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

tively measuring the scale at which processing locales are clustered, this speaking directly to size of site catchments and degrees of seasonal population aggregation and dispersal. Nearest-neighbor 

The nearest-neighbor statistic (NN) provides a more objective measure of spatial point patterning than raw density data because it does not rely on arbitrary units of analysis like quadrats. It is derived measuring the linear distance between everyofdata point and itsbynext nearest-neighbor and dividing the mean observed distances (dobs) by expected mean distances (dran) between the same number of randomly distributed points. The   dran value is one-half the square root of study area size ( a), divided by number of points (n)  (  (Clark Clark and Evans, 1954 1954). ). The resulting formula, still p  used (Diggle, (Diggle, 2003; Durand et al., 1992), 1992 ), is: NN =  dobs/0.5 (a/n). Study area boundaries, however, can disallow measurements between points and limit measurements between bounded points, rather rather than ones distr distribut ibuted ed in infini infinite te spac space, e, a requisit requisite e for achi achieveving truly random point distributions (Pinder ( Pinder et al., 1979 1979). ). Ebdon  Ebdon (1976) accounted (1976)  accounted for this  boundary effect  with   with a correction coeffip  cient,  C  (0.497  (0.497 + 0.127 (a/n)), replacing the 0.5 value in the original formula. Using GIS to meas measure ure distance distancess betw between een multiple sets of computer-generated random points is an alternative way of generatin generating g   dran. Regardles Regardlesss of method, method, NN values less than 1.00 indicate clustering, clustering, values greater than 1.00 dispersal, and values near 1.00 random distribu distribution tions. s. NN valu values es wer were e gene generate rated d using all three methods and the nearest-neigh nearest-neighbor/event– bor/event–event event dis2002). ). The analysis generate generated d tances extension for ArcGis (Sawada, ( Sawada, 2002 NN values at multiple scales, from a gross analysis of all study area processing sites, to analyses of the site types identified in Table in  Table 2 correlat corr elated ed with the thre three e prin principal cipal stud study y area ecozo ecozones nes (e.g (e.g.,., Fi Fig. g. 6). Results

NN values are consistent regardless of method, though values derived from generated random points are slightly lower than for4). ). To simplify discussion, the following uses mulae values (Table (Table 4 values generated with the correction coefficient formula developed by by Ebdon  Ebdon (1976). (1976). In the lower montane forest, the NN statistic for all processing sites is 0.79, indicating slight clustering (Fig. ( Fig. 7). 7). Site type NN values range from slightly clustered at 0.77 for logistical sites, to nearly random at 0.92 for residential sites. Overall, sites in the lower montane forest are slightly clustered, indicating winter population aggregation below snowline. In the montane forest, NN values indicate random and even dispersed settlement, varying from 1.04 for all processing sites to 1.37 for residential sites. These values show populations mapping onto dispersed resources when the area is clear of snow. Values for subalpine/high elevation residentia den tiall sit sites es ar are e highly highly va varia riable ble,, a resul resultt of small small sam sample ple siz size e (n = 12). They range from random (NN = 1.00 for subsidiary camps) to clustered (NN = 0.40 for principal camps). Low NN values in the subalpine zone indicate clustering due to association with transHindes des,, 19 1959 59;; Sny Snyder der,, 20 2001 01). ). Site Sitess he here re are are Sierr Sierran an trails trails (Hin constrained to the only passable portions of the landscape in these settings: creek and river canyons and alpine passes (Fig. ( Fig. 8) 8)  (  (MorMorgan, 2006). 2006). Together these these data indicate low lowland land winter population aggregation, montane forest spring and summer population dispersal, and high residential mobility associated with trans-Sierra trade and travel.  Variance-to-mean  Variance-to-mea n

Variance-to-mean Varianceto-mean ratios (VMR) provide an objective measure of  the the sca scale le of point point pat patter tern n dis distr tribu ibutio tions ns and pro provid vide e a clear clearer er

 

389

picture of the size of territories affiliated with different sites and site types. VMR also cope with problems associated with simple spatial analyses that measure density within arbitrary units (e.g., quadrats) of analysis that skew interpretations of clustering, dispersal and the scale at which these phenomena occur simply as a function of quadrat size. VMR are simply the ratio of the variance divided by the mean density of data points per quadrat. Variance, a measure of the spread of the distribution, changes relative to quadrat size. As quadrat size approaches the scale of patterning in the point distribution, the histogram showing the number of  points per quadrat becomes bimodal, with some quadrats containing many points and others containing very few or none (Ebert, (Ebert, 1992:191). 1992:191 ). Here, variance is high relative to the mean; this is the scale of patterning. The scale of patterning in the data is more meaningful than density distribution because it indicates at what scale the points in the distribution are actually clustered. Methods

VMR were determined using ArcGis software. BRM geographic distribution distribut ion was entered into a spreadsheet using site UTM coordinates as  x  and  y  coordinates for BRM location. At sites with more than one BRM, northing and easting values were estimated by either adding or subtracting one meter per BRM from each UTM value, so that at no site were BRMs more than 100 m from the UTM marking the center point of the site. Though not precisely locating each BRM, these are certainly within the margin of error of hand-held GPSlocations units and the map measurements from which whic h UTMs were origina originally lly deri derived. ved. More importan importantly, tly, thes these e methods quantify every study area BRM without relying on the somewhat arbitrary site definitions used by the multiple studies contributing to the database. The resulting data were subdivided by eco ecozon zone. e. Six grids grids,, 0.0 0.05, 5, 0.10, 0.10, 0.25, 0.25, 0.5 0.50, 0, 1.00, 1.00, 2.50, 2.50, and 2 5.00 km , were generated and superimposed over the study area and then queried to determin determine e the number of mort mortars ars in each quadrat (e.g., Fig. (e.g.,  Fig. 8 8). ). This resulted in a database of the density of  mortar mor tarss per qua quadra dratt in eac each h eco ecozon zone e at six scales scales of an analy alysis sis;; VMR were derived from these data (Table (Table 5). 5). Results

Results indicate substantial differences in settlement and processing behaviors cessing behaviors by ecozo ecozone. ne. The VMR in the lower montan montane e ecozone peaks at 1 km. The VMR in the montane forest and alpine zones peaks at 2.5 km (Fig. ( Fig. 9 9). ). Together, these data indicate that BRMs are clustered in 1 km2 areas in the lower elevations of the study area, the area containing Mono winter settlements. Above snowline snow line,, mort mortars ars clus cluster ter in 2.5 km2 area areas, s, indic indicatin ating g greater greater dispersal of processing sites in montane forest and subalpine ecozones. Together, scales of patterning indicate population clustering below snowline in winter and dispersal above snowline in spring, summer and fall, with dispersed summer settlements settlements and processing stations stations patt patterne erned d at a scal scale e 2.5 times greater than winter camps, hamlets, and processing stations. Synthesis and conclusion

The onset of LIA conditions some 600 years ago resulted in a set of circumstance circumstancess particular particularly ly challenging for hunter–ga hunter–gatherers therers:: how to best average increased variance in temporal and spatial resource sour ce distr distribut ibutions ions.. Anal Analyses yses of processin processing g site and proc processin essing g surface distributions as proxy measures of Mono mobility indicate a multifaceted approach to solving these problems. NN values indicate clustering, dispersal, and clustering once again of processing sites in lower montane, montane forest, and subalpine ecozones,

 

390

 

C. Morgan/ Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

Logistical Processing Site Lower Montane Forest Euclidian Path to Nearest Neighbor 

0 1.5 3

6 Km

North Fig. 6.   Distribution of logistical processing sites and nearest-neighbor distances, lower montane forest.

respectively. This distribution is partly predicated, of course, on correlation between local environmental variables and settlement choice (Premo, (Premo, 2004 2004). ). Previous research in the area indicates that site locat locations, ions, espe especial cially ly proce processin ssing g site loca locations tions,, are stro strongly ngly associated with chaparral and coniferous forest vegetation, stream terraces terr aces and midsl midslope ope land landform forms, s, and prox proximit imity y to pere perennia nniall

montane resources, and subalpine–alpine travel along narrow trail corridors corr idors (again, (again, indi indicate cated d by clus clusteri tering ng alon along g trav travel el corr corridors idors). ). Interest Inte restingl ingly, y, thes these e resu results lts indic indicate ate that process processing ing site sitess in the montane forest are distributed in a random-dispersed pattern, suggesting that at these landscape scales behavior determines settlement as much as does correlation with localized environmental

streams and trans-S streams trans-Sierr ierran an trav travel el corr corridor idorss (Crist, 1981; Jackson, 1984, 1988; Hull and Mundy, 1985; Pilgram, 1987). 1987). The current results, however, go beyond this kind of determinism and clearly show how populations used different ecozones on a seasonal basis: lowland lowla nd popul populatio ation n aggr aggregat egation ion (clus (clusteri tering) ng) below below snow snowline line in winter, wint er, dispersa dispersall in sprin spring g to map onto seas seasonal onal mid-ele mid-elevati vation on

variables. VMR indicate the scale of patterning within this system, with BRMs clustered in areas 2.5 times smaller in the lower montane forest than in montane forest and subalpine zones. This indicates not only substantial lowland settlement clustering, but also substantially smaller logistical catchments in the lower montane forest. Together, these data describe seasonal, intensive, semi-sed-

 

 

C. Morgan / Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

391

 Table 4

Processing site NN values by ecozone.

Site type

Site count

Mean observed distance

Lower montane forest  Residential 95 Principal camp 44 Subsidiary camp 51 Logistical 128 Temporary camp 77 Processing station 5 51 1

Mean random distance

Expected (traditional formula)

Expected (coefficient formula)

NN (random points)

NN (traditional formula)

NN (coefficient formula)

1001.70 1689.00 1370.90 721.80 937.40 1417.10

1281.80 1901.80 1520.90 1086.90 1371.90 1697.00

1035.16 1521.05 1412.81 891.79 1149.80 1412.81

1082.90 1628.41 1504.84 926.48 1209.47 1504.84

0.78 0.88 0.90 0.66 0.68 0.83

0.96 1.11 0.97 0.80 0.81 1.00

0.92 1.03 0.91 0.77 0.77 0.94

All processing sites

223

548.50

716.30

675.64

694.57

0.76

0.81

0.79

Montane forest  Residential Principal camp Subsidiary camp Logistical Temporary camp Processing station All processing sites

28 10 18 84 37 47 47 112

2716.80 5794.30 3103.50 1234.10 2287.60 1539.40 993.40

2506.90 5479.00 2484.40 1384.60 2175.40 1863.60 1141.80

1817.47 3041.21 2266.79 1049.32 1581.05 1402.81 908.73

1981.05 3511.52 2524.60 1101.18 1703.60 1498.34 946.90

1.08 1.05 1.24 0.89 1.05 0.82 0.87

1.49 1.90 1.36 1.17 1.44 1.09 1.09

1.37 1.65 1.22 1.12 1.34 1.02 1.04

Subalpine Residential Principal camp Subsidiary camp Logistical Temporary camp Processing station All processing sites

12 4 8 72 20 5 52 2 84

4552.70 3663.80 6082.00 1216.70 2544.90 1611.10 1 11 192.60

5911.70 14631.30 6224.70 1899.10 4269.70 2091.00 1844.50

4203.78 7281.17 5148.56 1716.18 3256.24 2019.43 1588.88

4795.03 9086.90 6042.38 1808.63 3606.58 2149.58 1667.41

0.77 0.25 0.97 0.64 0.59 0.77 0.64

1.08 0.50 1.18 0.70 0.78 0.79 0.75

0.94 0.40 1.00 0.67 0.70 0.75 0.71

entary logistical exploitation of lower montane settings and substantial residential mobility in the montane forest in spring, summer and fall. This mixed mobility pattern conforms to expectations regarding the most effective ways to average pronounced spatial and temporal resource variability. Increased sedentism, few residential moves and logistical mobility are ways of averaging temporal resource variance vari ance in seas seasonal onal sett settings ings,, part particula icularly rly whe where re reso resource urcess are abundant and diverse, as they are in the lower montane forest. Winter Win ter Mono sett settleme lement nt patt patterns erns confo conform rm to this expectat expectation: ion: clustered site distributions below snowline indicate seasonal population aggregations relying on logistical forays and cached and stored stor ed foods foodstuff tuffss to comp compensa ensate te for wint winter er reso resource urce shor shortfal tfalls ls analysis ysis of Mono logisti logistical cal mobility (see   Morgan (2008)   for an anal and storage behaviors). Conversely, when low population densities face extreme environmental conditions and highly unpredictable but homogenous resource bases, mapping onto resources is mod-

eled as the most effective way to average resource shortfalls, particularl ticu larly y when these moves are beyond the dista distance nce at whic which h resources can be efficiently moved back to camp. This is the behavior seen in the random-dispersed settlements and camps in the montane forest. This pattern recalls the most efficient way to optimize encounter encounter rate ratess with diffuse or ran randomly domly distributed distributed resource sou rces, s, esp especi eciall ally y wh when en for forek eknow nowle ledge dge of envir environm onmen ental tal productivity and resource distribution is poor or absent (Branting( Brantingham, 2006), 2006), a situation exacerbated by LIA induced uncertainty. Subalpine mobility appears to be conditioned less by subsistence and more by travel across the Sierran crest along narrow travel corridors. What makes these findings particularly interesting is the fact that these patterns appear to have changed substantially in the last 1000 years or so. Though comparable data are not yet available at the resolution presented in this study, a fairly substantial body of  literature indicates that from about 3500–1350 calBP the western

1.6 Lower Montane Forest

1.4

1.2    t   a    t    S  

   N    N

Montane Forest

1

0.8 Subalpine

0.6

0.4  All Processing Processing Sites

Logistical Sites

Residential Sites

Fig. 7.   NN values, by site type and ecozone.

 

392

 

C. Morgan/ Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

Fig. 8.  Map Showing 2.5 sq. km quadrats, ecozones, and processing sites.

slope of the southern southern and cent central ral Sier Sierra ra Nevad Nevada a saw more inte intensive nsive use and occupation of large, semi-permanent villages in lowland, foothill footh ill sett settings ings (i.e., (i.e., belo below w 1000 m) and longlong-rang range e logis logistica ticall hunting in the highlands (Moratto, (Moratto, 1972; Moratto et al., 1978; Stevens, 2003). 2003). Data from excavations in the study area point to aban-

idea that high higher er elev elevatio ations ns are indeed esse essentia ntiall compo component nentss of  montane hunter–gather lifeways (e.g., Aldenderfer, (e.g.,  Aldenderfer, 1999; Wright et al., 1980). 1980). The way the Mono exploited the Sierra Nevada during the LIA, however, however, was pred predicat icated ed only in part on logistical logistical mobi mobility lity,, in contrast to the strategies seen in most mountain environments.

donmen don ments ts of som some e mon montan tane e for forest est locale localess be betw tween een 105 1050 0 and 775 calBP, during the MCA, and very intensive use of these same settings by the Mono during the LIA (Caputo, (Caputo, 1994; Goldberg and Moratto, 1984; Jackson, 1983; Jackson and Holson, 1984; Jones and Origer, 2001). 2001). Use of high elevations intensified after 1350 calBP, with larger, seasonal residential sites being occupied, particula larl rly y af afte terr 65 650 0 BP an and d the the on onse sett of th the e LI LIA A (Ste Stevens vens,, 2002 2002;; Stevens, 2005). 2005). So prior to the LIA, adaptive patterns were more typical typi cal of seas seasonal onal,, and espe especiall cially y moun mountain tain envi environm ronments ents,, focus focused ed on logistical procurement and long-range logistical hunting in the highlands. But during the LIA, the strategy described in the current study developed deve loped,, with more intensi intensive ve use of highl highland and setting settingss and a mixed mixe d logis logistica tical–re l–reside sidentia ntiall stra strategy tegy that inte intensiv nsively ely expl exploite oited d the diverse environments of the western Sierra Nevada. From an ecological perspective, these mobility patterns are clearly effective means of coping with the constraints posed by mountain environments. They averaged temporal variance in resource availability in lower elevations by winter population fusion, logistical mobility, and storage. They averaged pronounced spatial variability of resource productivity with spring, summer and fall population dispersal pers al to expl exploit oit reso resource urcess in mont montane ane forests forests,, suppo supportin rting g the

Residential mobility in higher altitude settings might appear surprising during the LIA, given shorter growing seasons and limits on mobility due to increased snowfall and even glacial advance. This pattern is expected, however, due to the distances between higher elevation resource patches and larger winter settlements: it is easier to move people to resources than to move resources to larger, centralized settlements, mainly because they are well beyond the 3.4 km foraging radius of hamlets in the lower montane ecozone. In any event, these patterns support the assertion that climatic conditions are, due to their effect on habitat quality and resource distribution, the basic limiting factors conditioning hunter–gatherer mobility (Grove, (Grove, 2009:7 2009:7). ). These conclusions also speak to the nature of hunter–gatherer response to risk and uncertainty. That the LIA favored  Goland’s (1991) ‘‘flexible (1991)  ‘‘flexible strategies” is clearly exhibited in the multifaceted Mono mobility pattern. Here, diverse mobility options are a riskaverse behavior ensuring solution to any number of spatial and temporal resource fluctuations fluctuations and failures, clearly a way of coping with uncertainty, particularly of the kind found in montane settings. Mono residential mobility, however, is anomalous as far as most montane foragers go, perhaps a behavior derived from their cultural and linguistic affiliates in the western Great Basin, the Nu-

 

C. Morgan / Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

 

393

 Table 5

BRM density and VMR by ecozone.

Quadrat size (km2)

Quadrats (n)

Mean

Variance

VMR  

Lower montane forest  0.05 0.10 0.25 0.50 1.00 2.50 5.00

168,771 42,184 6739 1699 424 69 17

0.026 0.103 0.645 2.560 10.259 63.043 255.882

0.554 2.862 23.251 135.220 1141.098 1135.893 277.981

21.308 27.786 36.048 52.820 111.229 18.018 1. 1 .086

Montane forest  0.05 0.10 0.25 0.50 1.00 2.50 5.00

157,370 39,370 6278 1569 387 64 18

0.007 0.029 0.183 0.731 2.964 17.922 63.722

0.103 0.448 3.863 16 1 6.647 72 72.237 1832.055 51 51.887

14.714 15.448 21.109 22.773 24.371 102.224 0 0..814

Subalpine 0.05 0.10 0.25 0.50 1.00 2.50 5.00

344,942 86,197 13,810 3435 855 140 33

0.002 0.007 0.041 0.166 0.667 4.071 17.273

0.018 0.091 0.600 2.613 1 10 0.603 137.182 343.687

9.000 13.000 14.634 15.741 15.897 33.697 19.897

mic-spea micspeaking king Shoshone Shoshone and Paiu Paiute, te, each of whom establi established shed high-altitude residential bases in the very late Holocene (in the Alta Toquima and White Mountains, respectively) (Fig. (Fig. 10 10). ). Ultimately,, this study elicits some fundamenta Ultimately fundamentall observations and questions regarding the ecology and evolution of hunter–gather responses to environmental variability over larger spans of time. Though linking climate change to cultural dynamics can be problematic (see Anderson (see  Anderson et al., 2007:12–18), 2007:12–18 ), the current study sugge gests sts tha thatt the the shi shift ft from from MCA MCA con condit dition ionss (a dro drough ughtt-pro prone ne,, stressing period associated with substantia substantiall cultural disruptions in California and beyond [e.g., Jones [e.g.,  Jones et al., 1999 1999]) ]) to LIA conditions (a period often regarded as one of regional environmental ameliorization) brought about resource stress due to increased snowfall and more unpr unpredic edictab table le reso resource urce produ productivi ctivity ty in Cali Californ fornia’s ia’s mountains, suggesting good times are only good (and bad times bad) relative to specific ecological context. The study also suggests

Fig. 10.  Map showing the distribution of Numic-speaking peoples (hatched) and evidence of high-altitude residential mobility (See evidence of (See above-mentioned references for further information). information).

that hype hypervar rvariabl iable e clim climatic atic condi condition tionss may favo favorr risk reducing reducing strategies, a possibility that has important implications for human socioeconomic evolution. For example, if the LIA is something of a small-scale small-scal e analog for hypervariable hypervariable Pleistocene climatic conditionss domin tion dominatin ating g so much of huma human n evolu evolution tion and preh prehistor istory y  Grafenstein et al., 1999), (e.g., Grafenstein (e.g., 1999), it follows that Pleistocene conditions may have also favored risk-averse behaviors. This could conceivably explain remarkably conservative Pleistocene behaviors like East Asia Asian n core core/flak /flake e (or flak flake/sha e/shatter tter)) tech technolo nologies gies,, whic which h per-

120 Lower Montane Forest

Montane Forest

100

80    R    M 60    V

40

20

Subalpine

0 0

1.00

2.50 Quadrat Size (sq. km) Fig. 9.   Bedrock mortar VMR by ecozone.

5.00

 

394

 

C. Morgan/ Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

sisted nearly unchanged for more than 50,000 years (Seong, (Seong, 2004): 2004): it may have been simply too risky to adopt new tools that may (or may not) do a job better than existing technologies. It follows that less clima climatica ticallylly-vari variable able (and (and thus more pred predicta ictable ble)) Holoc Holocene ene conditions may have favored the more optimizing (and arguably behavior viorss like sede sedentis ntism, m, more risky – see   Bettinger 2006:313) 2006:313) beha agriculture and other intensified behaviors affiliated with the evolution lutio n of more complex societ societies ies (e.g (e.g.,.,   Schurr and Schoeninge Schoeninger, r, 1995). 1995 ). Following this logic, the persistence (if not the adoption) of high high-ris -risk, k, but pote potentia ntially lly higher-y higher-yield ield econ economic omic beha behaviors viors would thus be favored either when times were predictable enough and/or when surplus production and storage were great enough to insure against the uncertainty and higher risks of failure associated 1989). ). with such behaviors (e.g., Richerson (e.g.,  Richerson et al. 2001; Halstead, 1989 Finally, if risk-averse behaviors are ways of averaging the multiple possible outcomes of decisions (the number of which increases under variable conditions), they are not necessarily disposed toward maximizing energetic return relative to labor input. Rather, they may also be target-or target-oriented iented (e.g., storing enough food to make it through the winter, ensuring minimum daily caloric requirementss are met, obta ment obtaining ining suffici sufficient ent water or firew firewood.) ood.).. In this way they are geared towards economic satisficing over optimizing, the former an economic strategy that is particularly particularly effective when information on optimal solutions is difficult (i.e., costly) to procure, such as when variable climatic (and perhaps social or other) conditions prevail. If this is the case, then it is conceivable that climate change and variability favor not economic maximizing or even efficiency but economic security. Security, of course, comes with its own costs and rewards, but is clearly unlike the optimizing behaviors assoc associate iated d with the evolu evolution tion of compl complex ex hunt hunter– er–gath gatherer erer,, agricultural, and industrial economies during the Holocene.  Acknowledgments

This research has benefited from guidance by Robert Bettinger, Bruce Winterhalder, Aram Yengoyan, G. James West and Thomas  Jackson as well as excellent criticism by  Journal of Anthropological  Archaeology  reviewers. I also thank Galen Lee for his willingness to share information about Mono use of Sierr Sierran an landsca landscapes. pes. Any errors or omissions, however, are my own. References Adams, Adams, K.R., K.R., Boh Bohrer rer,, V.L V.L., ., 1998. 1998. Archae Archaeobo obotan tanica icall indica indicator torss of season seasonali ality, ty, examples examp les from arid southweste southwestern rn United United States. States. In: Rocek, T.R., Bar-Yosef, Bar-Yosef, O. (Eds.), Seasonality and Sedentism, Archaeological Perspectives from Old and New World Sites. Sites. Peabody Peabody Museum of Archaeology Archaeology and Ethno Ethnology, logy, Harvard University, Cambridge, Massachusetts, pp. 129–142. Agi Aginsk nsky, y, B.W., B.W., 1943. 1943. Cul Cultur ture e elemen elementt distri distribut bution ions: s: XXI XXIV, V, Centra Centrall Sierra Sierra.. Anthropological Records 8, 393–468. Aldenderfe Alden derfer, r, M., 1998. Montane Foragers: Asana and the SouthSouth-centr central al Andean Andean Archaic. University of Iowa Press, Iowa City. Aldenderfe Alden derfer, r, M., 1999. An archaeological archaeological perspective perspective on the human use of cold montane monta ne enviro environment nmentss in Ande Andean an South America. America. Revista Revista de Arque Arqueologi ologi´a Americana, 75–96. Aldenderfe Alden derfer, r, M., 2006. Modelling plateau people peoples: s: the early human use of the world’s high plateaux. World Archaeology 38, 357–370. Allen, C.R., Breshears, Breshears, D.D., 1998. Drought-ind Drought-induced uced shift of forest-wood forest-woodland land ecotone: ecoto ne: rapid lands landscape cape response to climatic climatic variation. Proceedings Proceedings of the National Academy of Sciences 95, 14839–14842. Ames, K.M., 2002. Going by boat: the forager–collector continuum at sea. In: Habu,  J. (Ed.), Beyond Foraging and Collecting: Evolutionary Change in Hunter– gatherer Settlement Systems. Kluwer/Plenum, New York, pp. 19–52. Anderson, D.G., Maasch, K.A., Sandweiss, D.H., Mayewski, P.A., 2007. Climate and culture change, exploring holocene transitions. In: Anderson, D.G., Maasch, K.A., San Sandw dweis eiss, s, D.H. D.H. (Ed (Eds.) s.),, Climat Climate e Change Change and Cul Cultur tural al Dynami Dynamics, cs, A Glo Global bal Perspective on Mid-Holocene Transitions. Academic Press, Amsterdam, pp. 1– 24. Armsworth, P.R., Roughgarden, J.E., 2003. The economic value of ecological stability. Proceedings of the National Academy of Sciences 100 (12), 7147–7151. Attwell, M.R., Fletcher, M., 1987. An analytical technique for investigating spatial relationships. Journal of Archaeological Science 14, 1–11.

Barlow, K.R., Metcalfe, D., 1996. Plant utility indices: two Great Basin examples.  Journal of Archaeological Science 23, 351–371. Baxter, T.S., 2008. Mathematics, statistics and archaeometry: the past 50 years or so. Archaeometry 50, 968–982. Bender, S.J., Wright, G.A., 1988. High-altitude occupations, culture process, and high plains prehistory: retrospect and prospect. American Anthropologist 90, 619– 639. Bettinger, R.L., 1977. Aboriginal human ecology in Owens Valley: prehistoric change in the Great Basin. American Antiquity 42, 3–17. Bettinger, Betti nger, R.L., 1979. Multivar Multivariate iate statistic statistical al analy analysis sis of a regional regional subsis subsistence tence-settlement model for Owens Valley. American Antiquity 44, 455–470. Bettinger, R.L., 1991. Aboriginal occupation at high altitude: alpine villages in the white mountains of California. American Anthropologist 93, 656–679. Bettinger, R.L., 2006. Agriculture, archaeology, and human behavioral ecology. In: Kennett, D.J., Winterhalder, B. (Eds.), Behavioral Ecology and the Transition to Agriculture Agric ulture,, Origins Origins of Human Behavior Behavior and Culture, Culture, vol. 1. University of  California Press, Berkeley, pp. 304–322. Bettinger, R.L., Malhi, R., McCarthy, H., 1997. Central place models of acorn and mussel processing. Journal of Archaeological Science 24, 887–899. Bevan, A., Conolly, J., 2004. GIS, archaeological survey, and landscape archaeology on the Island of Kythera, Greece. Journal of Field Archaeology 29, 123–138. Binford, L.R., 1977. General Introduction. In: Binford, L.R. (Ed.), For Theory Building in Archaeology. Academic Press, New York, pp. 1–10. Binford, Binfor d, L.R., 1980. Willo Willow w smoke and dog’s tails: hunter–gatherer hunter–gatherer settleme settlement nt systems and archaeological site formation. American Antiquity 45 (1), 4–20. Blankholm, H.P., 1993. Presab: a new method of spatial analysis. Helinium 33, 3–19. Bordley, Bordle y, R., LiCalzi, LiCalzi, M., 2000. 2000. Decisi Decision on analysis using targe targets ts inste instead ad of utili utility ty functions. Decisions in Economics and Finance 23, 53–74. Brantingham, P.J., 2006. Measuring forager mobility. Current Anthropology 47, 435– 459. Burke, A., 2006. neanderthal settlement patterns in Crimea: a landscape approach.  Journal of Anthropological Archaeology 25, 510–523. Campell, I.D., McAndrews, J.H., 1993. Forest disequilibrium caused by rapid little ice age cooling. Nature 366, 336–388. Caputo, J.M., 1994. Arrowheads Cast through Time: Archaeological Investigations at the Portuguese Flat Site. M.A. Thesis, Sonoma State University, Department of  Anthropology. Crist, M.K., 1981. Settlement Settlement Patterns Patterns and Locational Locational Strategie Strategies: s: Prehistoric Prehistoric Land Use in the Southern Sierra Nevada and San Joaquin Valley. M.A. Thesis, San Francisco State University. Clark, P.J., Evans, F.C., 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35, 445–453. Close, A.E., 2000. Reconstructing movement in prehistory. Journal of Archaeological Method and Theory 7, 49–77. Connolly, Connol ly, J., Lake, M., 2006. 2006. Geogra Geographica phicall Informatio Information n Systems Systems in Archaeology Archaeology.. Cambridge University Press, Cambridge, U.K. Diggle, P.J., 2003. Statistical Analysis of Spatial Point Patterns. Oxford University Press, New York. Driver,, H.E., 1937. Cultu Driver Culture re eleme element nt distributio distributions: ns: VI, Southern Southern Sierra Nevada. Anthropological Records 1, 53–154. Dunnell, Dunnel l, R.C., 1992. The notio notion n site. In: Rossignal, J., Wands Wandsnider nider,, L. (Eds.), Space, Time, and Archaeological Landscapes. Plenum Press, New York, pp. 21–42. Durand,, S.R., Pippin, L.C., Spennman, D.H.R., 1992. A pragm Durand pragmatic atic applicat application ion of the nearest neighbor statistic. Journal of Field Archaeology 19, 263–274. Ebdon, D., 1976. On the underestimation inherent in the commonly used formulae. Area 8, 391–399. Ebert, J.I., 1992. Distributional Distributional Archaeolo Archaeology. gy. Unive University rsity of New Mexico Press, Albuquerque. Fitzhugh, B., Habu, J., 2002. Beyond foraging and collecting, evolutionary change in hunter–gat hunte r–gatherer herer settlement systems. In: Feinman, Feinman, G.M., Price, T.D. (Eds.), Fundamental Issues in Archaeology. Kluwer Academic/Plenum Publishers, New York. Fletcher, Fletch er, R., 2008. Some spatial analyses of chalcolithi chalcolithicc settlement settlement in Southern Southern Israel. Journal of Archaeological Science 35, 2048–2058. Gayton, Gayto n, A.H., 1948. Yokuts and Weste Western rn Mono Ethnograp Ethnography, hy, II: Northern Northern Foothill Yokuts and Western Mono. University of California Anthropological Records 10. Gayton, Gayto n, A.H., n.d. Notes from Yokuts, Western Western and Eastern Eastern Mono. On File, the Bancroft Library, University of California, Berkeley. Gifford, E.W., 1932. The Northfork Mono. University of California Publications in American Archaeology and Ethnology 31, 15–65. Gifford, Giffo rd, E.W., 1971. Calif California ornia balanoph balanophagy. agy. In: Heizer, Heizer, R.F., Whipple, M.A. (Eds. (Eds.), ), The California Indians, A Source Book. University of California Press, Berkeley, pp. 301–305. Northfork ork Mono Mono Hamlet Hamletss and Camp Camp Sites. Sites. Eth Ethnol nologi ogical cal Gifford,   E.W.,   n.d.   Northf Documents 168. Goland, C., 1991. The ecological context of hunter–gatherer storage: environmental predictability and environmental risk. In: Miracle, P.T., Fisher, L.E., Brown, J., (Eds.), Foragers in Context: Long Term Regional and Historical Perspectives in Hunter–Gatherer Studies, pp. 107–125. Goldbe Goldberg, rg, S.K., S.K., Mor Moratt atto, o, M.J., M.J., 1984. 1984. Archae Archaeolo ologic gical al Invest Investiga igatio tions ns at Balsam Balsam Meadow, Fresno County, California: Data Recovery from Sites 04-FRE-811, 04FRE-812, FRE-8 12, and 04-FRE-818. 04-FRE-818. Report to Southern Southern California California Edison Company, Company, Rosemead, California. Gould, R.A., Yellen, J.E., 1987. Man the hunted: determinants of household spacing in desert and tropical foraging societies. Journal of Anthropological Archaeology 6, 77–103.

 

C. Morgan / Journal of Anthropological Anthropological Archaeology 28 (2009) 382–396 

 

395

von Grafenstein, Grafenstein, U., Erlen Erlenkeuse keuser, r, H., Braur, A., Jouzel, Jouzel, J., Johnso Johnson, n, S.J., 1999. A mideuropean decadal isotope-climate record from 15,000 to 5000 years BP. Science 284, 203–207. Graumlich, L.A., Lloyd, A.H., 1996. Dendroclimatic, Dendroclimatic, ecological, and geomorphological evidence for long-term climatic change in the Sierra Nevada, USA. In: Dean, J.S., Meko, D.M., Swetnam, T.W. (Eds.), (Eds.), Tree Rings Enviro Environment nment and Humani Humanity. ty. Radiocarbon, Tucson, Arizona. Graumlich, Grauml ich, L.J., 1993. A 1000-year 1000-year record of temperatur temperature e and precipitation precipitation in the Sierra Nevada. Quaternary Research 39, 249–255. Grove,, M., 2009. Hunter–gathere Grove Hunter–gathererr movement movement patterns: patterns: causes and constraints constraints..  Journal of Anthropological Archaeology. doi:10.1016/j.jaa.2009.01.003. doi: 10.1016/j.jaa.2009.01.003. Halstead, P., 1989. The economy has a normal surplus: economic stability and social change among early farming communities in Thessaly, Greece. In: Halstead, P., O’S O’Shea heaBad Bad,, J. (Ed (Eds.) s.),, Yea Yearr Eco Econom nomics ics,, Cul Cultur tural al Res Respon ponses ses to Ris Risk k and Uncertainty. Cambridge University Press, Cambridge, pp. 68–80. Halstead,P., Halst ead,P., O’Shea, O’Shea, J., 1989.Introduc 1989.Introduction:culturalresponses tion:culturalresponses to riskand uncertainty uncertainty.. In: Halstead, P., O’Shea, J. (Eds.), Bad Year Economics, Cultural Responses to Risk and Uncertainty. Cambridge University Press, Cambridge, pp. 1–7. Haney,, J.W., 1992. Acorn exploitat Haney exploitation ion in the eastern Sierra Nevada. Journal of  California and Great Basin Anthropology 14, 94–109. Hawkes Haw kes,, K., O’C O’Conn onnell ell,, J.F., J.F., Blu Blurto rton-J n-Jone ones, s, N.G N.G., ., 1989. 1989. Hardwo Hardworki rking ng Hadza Hadza grandmother grand mothers. s. In: Standen, V., Foley, Foley, R. (Eds. (Eds.), ), Comparative Comparative Socioecology Socioecology of  Mammals and Man. Blackwell, London, pp. 341–366. Hindes, M.G., 1959. A report of Indian trails, Huntington Lake region, California. University of California Archaeological Survey Reports 48, 1–15. Hindes, Hinde s, M.G., 1962. The archaeology archaeology of the Huntingt Huntington on Lake region. University University of  California Archaeological Survey Reports, 58. Holdridge, L.R., 1967. Life Zone Ecology. Tropical Science Center, San Jose, Costa Rica. Hodder, I., Hassell, M., 1971. The non-random spacing of Romano-British walled towns. Man 6, 391–407. Hughes, M.K., Brown, P.M., 1992. Drought frequency in central California since 101 B.C. Recorded in giant sequoia tree rings. Climate Dynamics 6, 161–167. Hull, K.L., W.J. Mundy, 1985. The 1984 Yosemite archaeological surveys: the south entrance, Mariposa Grove, Tioga Road, Crane Flat, and Glacier Point Road Areas.

Körner, C., 1998. World Körner, Worldwide wide positi positions ons of alpin alpine e treelines treelines and their causes causes.. In: Beniston, M., Innes, J.L. (Eds.), The Impacts of Climate Variability on Forests. Springer, New York, pp. 221–230. Kowale Kow alewsk wski, i, S.A., S.A., 2008. 2008. Region Regional al settle settlemen mentt patter patterns ns studie studies. s. Jou Journa rnall of  Archaeological Research 16, 225–285. Krist, Jr., F.J., 2001. A Predictive Model of Paleo-Indian Subsistence and Settlement., Ph.D. Dissertation, Department of Anthropology, Michigan State University. Kroeber, A.L., 1925. Handbook of the Indians of California. Dover Publications, New York. Kroeber, A.L., 1959. Ethnographic interpretations 7–11. In: McCowan, T.D., Gifford, E.W., Mandelbaum, D.G., Olson, R.L. (Eds.), University of California Publications in American Archaeology and Ethnology. University of California Press, Berkeley and Los Angeles, pp. 235–307. Küchler, A.W., 1977. The map of the natural vegetation of California. In: Major, M.B.a.J. (Ed.), Terrestrial Vegetation of California. John Wiley & Sons, New York, pp. 909–938. Lee, G.D., 1998. Walking Where We Lived. Lived. Memoi Memoirs rs of a Mono Indian Indian Family. University of Oklahoma Press, Norman. Leeman Leemans, s, R., Cra Cramer mer,, W., Min Minnen nen,, J.G.V. J.G.V.,, 1996. 1996. Predic Predictio tion n of glo global bal biome biome distributio distri bution n using bioclimati bioclimaticc equilibrium equilibrium models. In: Breymeyer, A.I., Hall,  J.M.M.D.O., Agren, G.I. (Eds.), Global Change: Effects on Coniferous Forests and Grasslands,   Scope, vol. 56. John Wiley and Sons, New York, pp. 413–450. Grasslands, Low, B.S., 1990. Human responses to environmental extremeness and uncertainty: a cross-cultural perspective. In: Cashdan, E. (Ed.), Risk and Uncertainty in Tribal and Peasant Economies. Westview Press, Boulder, Colorado, pp. 229–256. Madsen Mad sen,, D.B D.B., ., Rhode, Rhode, D. (Ed (Eds.) s.),, 1994. 1994. Acr Across oss the Wes West: t: Human Human Popula Populatio tion n Movement Movem ent and the Expansion Expansion of the Numa. University University of Utah Press, Salt Lake City. McCarthy, McCar thy, H., Blount, C., Hicks, R., 1985. Appendix Appendix F: a functional functional analysis analysis of  bedrock mortars: western Mono food processing in the Southern Sierra Nevada. In: I. Infote Infotecc Resear Research, ch, I. Theodo Theodorat ratus us Cul Cultur tural al Resear Research ch (Ed (Eds.) s.),, Cul Cultur tural al Resources of the Crane Valley Hydroelectric Project Area, vol. I. Prepared for Pacific Gas and Electric Company, Sonora and Fair Oaks, California, pp. 303–356. McCarthy, H., 1993. A Political Economy of Western Mono Acorn Production, (Ph.D. Dissertation), Anthropology Department, University of California, Davis.

Yosemite Research Center Publications in Anthropology 1.  Jackson, T.L., 1983. Report of Archaeological Test Excavations for FERC Project 67 (CA-FRE-422, FRE-809, FRE-940, FRE-942, FRE-1208, FRE-1209, and FRE-1211). Archaeological Consulting and Research Service. Report to Southern California Edison Company, Rosemead.  Jackson, T.L., 1984. Predictive Model of Prehistoric Settlement Patterning in the So Sout uthe hern rn Sier Sierra ra Ne Neva vada da.. In In:: Th Theo eodo dora ratu tuss Cu Cult ltur ural al Rese Resear arch ch,, I. I.,, an and d Archaeological Consulting and Research Services, I., (Eds.), Cultural Resources Overview Overvi ew of the South Southern ern Sierra Nevad Nevada. a. Prepared for USDA Forest Service, South Central Contracting Office, Bishop, California.  Jackson, T.L., 1988. Amending models of trans-Sierran obsidian tool production and exchange. Journal of California and Great Basin Anthropology 10 (1), 62–72.  Jackson, T.L., Holson, J. 1984. Report of Archaeological Data Recovery Excavations at CA-FRE-1211 CA-FRE -1211.. Submitted Submitted to Southern Southern California California Ediso Edison n Company, Company, Rosemead, Rosemead, California. Califo rnia. Archaeological Archaeological Consulting and Research Research Services, Services, Inc. Santa Cruz, California.  Jochim, M., 2006. Regional perspectives on early mesolithic land use in southwestern Germany. Journal of Anthropological Archaeology 25, 204–212.  Jones, E.L., 2007. Subsistence change, landscape use, and changing site elevation at the Pleistocene–Holocene transition in the Dordogne of Southwestern France.  Journal of Archaeological Science, 344–353.  Jones, K.T., Madsen, D.B., 1989. Calculating the cost of resource transportation: a

McGuire, K.R., Hildebrandt, W.R., 2005. Re-thinking Great Basin foragers, prestige hunting huntin g and costly signaling signaling durin during g the midd middle le archaic archaic period. period. Ameri American can Antiquity 70, 695–712. McGuire, McGui re, K.R., Hilde Hildebrandt brandt,, W.R., Carpent Carpenter, er, K.L., 2007. 2007. Costly signali signaling ng and the ascendance of no-can-do archaeology: a reply to Codding and Jones. American Antiquity 72, 358–365. McKone, M.J., Kelly, D., Lee, W.G., 1998. Effects of climate change on mast-seeding species: frequency of mass flowering and escape from specialist insect seed predators. Global Change Biology 4, 591–596. Merriam, C.H., 1955. The Mono Paiute. In: The Department of Anthropology of the University Univer sity of California California (Ed.), Studi Studies es of Calif California ornia Indians. University University of  California Press, Berkeley and Los Angeles, pp. 71–76. Miller, A., Barton, C.M., 2008. Exploring the land: a comparison of land-use patterns in the middle and upper Paleolithic of the Western Mediterranean. Journal of  Archaeological Science 35, 1427–1437. Miller, C.I., Westfall, R.D., Delany, D.L., King, J.C., Graumlich, L.J., 2004. Response of  subalpine conifers in the Sierra Nevada, California to 20th-century warming and decadal climate variability. Arctic, Antarctic, and Alpine Research 36, 181–200. Moore, K.M., 1998. Measures of mobility and occupational intensity in highland Peru. Peru. In: Rocek, Rocek, T.R., T.R., Bar-Yo Bar-Yosef sef,, O. (Ed (Eds.) s.),, Sea Season sonali ality ty and Sed Sedent entism ism,, Archaeological Perspectives from Old and New World Sites. Peabody Museum of Archaeology and Ethnology, Harvard University, Cambridge, Massachusetts,

Great Basin example. Current Anthropology 30 (4), 529–534.  Jones, T.E., Origer, T.M., 2001. Final Report Archaeological Investigations at CA-FRE217, Camp Sierra, Fresno County, California. Tom Origer & Associates, Rohnert Park, CA. Report submitted to USDA Sierra National Forest.  Jones, T.L., Brown, G.M., Raab, L.M., McVickar, J.L., Spaulding, W.G., Kennett, D.J., York, A., Walker, P.L., 1999. Environmental imperatives reconsidered. Current Anthropology 40, 137–170. Kantner, J., 2008. The archaeology of regions: from discrete analytical toolkit to ubiquitous spatial perspective. Journal of Archaeological Research 16, 37–81. Keeley Keeley,, L.H L.H., ., 1988. 1988. Hunter Hunter–ga –gathe therer rer eco econom nomic ic comple complexit xity y and ‘‘Popu ‘‘Populat lation ion Pressure”: Pressu re”: a cross-cultu cross-cultural ral approach. approach. Journa Journall of Anthr Anthropolo opological gical Archaeolog Archaeology y 7, 373–411. Kelly, D., Sork, V.L., 2002. Mast seeding in perennial plants: why, how, where? Annual Review of Ecology and Systematics 33, 427–447. Kelly, R.L., 1985. Hunter–Gatherer Mobility and Sedentism, A Great Basin Study, (Ph.D. Dissertation), Department of Anthropology, University of Michigan, Ann Arbor. Kelly, R.L., 1990. Marshes and Mobility in the Western Great Basin. In: Janetski, J.C., Madsen, Madse n, D.B. (Eds.), (Eds.), Wetland Wetland Adaptatio Adaptations ns in the GreatBasin, Museum ofPeoples and Cultures Occasional Papers No. 1. Brigham Young University, Provo, Utah. Kelly,, R.L., 1992. Mobility/s Kelly Mobility/sedent edentism: ism: concepts, concepts, archaeologi archaeological cal measures, measures, and effects. Annual Review of Anthropology 21, 43–66.

pp. 181–197. Moratto, M.J., 1972. A Study of Prehistory in the Southern Sierra Nevada Foothills, California. Ph.D. Dissertation. University of Oregon. Moratto, Morat to, M.J., King, T.F., Woolfende Woolfenden, n, W.B., 1978. Archa Archaeolog eology y and California’ California’ss climate. The Journal of California Anthropology 5 (2), 147–162. Morgan, C., 2006. Late Prehistoric Territorial Expansion and Maintenance in the South-Centr SouthCentral al Sierra Sierra Nevad Nevada, a, Calif California, ornia, (Ph.D. (Ph.D. Disser Dissertatio tation), n), Anthropolo Anthropology gy Department, University of California, Davis, Davis, California. Morgan, C., 2008. Reconstructing prehistoric hunter–gatherer foraging radii: a case study from Calif California’ ornia’ss Southern Southern Sierra Nevada. Journal of Archaeolog Archaeological ical Science 35, 247–258. Mortensen, P., 1972. Seasonal camps and early villages in the Zagros. In: Ucko, P.J., Tri Tringh ngham, am, R., Dim Dimble bleby, by, G.W. G.W. (Ed (Eds.), s.), Man Man,, Set Settle tlemen mentt and Urbani Urbanism. sm. Duckworth, London. Overpeck, Overp eck, J.T., Rind, Rind, D., Goldberg, R., 1990. ClimateClimate-induc induced ed changes in forest forest disturbance and vegetation. Nature 342, 51–53. Peterson, D.L., 1998. The impact of climatic extremes on forests: an introduction. In: Beniston, Benist on, M., Innes, Innes, J.L. (Eds.), The Impac Impactt of Climate Variability Variability on Forests. Springer, New York, pp. 191–208. Pilgram, T.K., 1987. Predicting Archaeological Sites from Environmental Variables: A Mathem Mathemati atical cal Mod Model el for the Sierra Sierra Nevada Nevada Foothi Foothills, lls, Califo Californi rnia. a. Bri Britis tish h Archaeological Reports International Series No. 320.

Kelly, R.L., 1998. Foraging and sedentism. Kelly, sedentism. In: Rocek, Rocek, T.R., Bar-Yosef, Bar-Yosef, O. (Eds.), Seasonality Season ality and Sedentism: Sedentism: Archaeologica Archaeologicall Perspectiv Perspectives es from Old and New World Wor ld Sites. Sites. Peabod Peabody y Museum Museum of Archae Archaeolo ology gy and Eth Ethnol nology ogy,, Harvar Harvard d University, Cambridge, Massachusetts, pp. 9–21. Koenig, Koeni g, W.D., W.D., Knops, Knops, J.M.H., J.M.H., 2005. The mystery mystery of masting masting in trees. American American Scientist 93, 340–347.

Pinder Pinder,, D., Shimad Shimada, a, I., Gregor Gregory, y, D., 1979. 1979. The neares nearest-n t-neig eighbo hborr statis statistic tic:: archaeologi archae ological cal applicatio application n and new developmen developments. ts. Ameri American can Antiquity Antiquity 44, 430–445. Premo,, L.S., 2004. Local spatial autoc Premo autocorrela orrelation tion statistics statistics quantify multi-scale patterns in distributional data: an example from the Maya lowlands. Journal of Archaeological Science 31, 855–866.

 

396

 

Anthropological Archaeology 28 (2009) 382–396  C. Morgan/ C.  Morgan/ Journal of Anthropological

Prentiss, W.C., Chatters, J.C., 2003. The evolution of collector systems on the Pacific coast of northwest North America. In: Habu, J., Savelle, J.M., Koyama, S., Hongo, H. (Eds.), Hunter–Gatherer Hunter–Gathererss of the North Pacific Rim. National National Museum Museum of  Ethnology, Osaka, pp. 49–80. Price, T.D., Brown, J.A., 1985. Aspects of hunter–gatherer complexity. In: Price, T.D., Brown,, J.A. (Eds.), Prehistoric Brown Prehistoric Hunter–Gatherers Hunter–Gatherers:: The Emergence Emergence of Cultu Cultural ral Complexity. Academic Press, Orlando, Florida, pp. 3–20. Richerson, P.J., Boyd, R., Bettinger, R.L., 2001. Was agriculture impossible during the pleistocene but mandatory during the Holocene? a climate change hypothesis. American Antiquity 66, 387–412. Rick, J.W., 1980. Prehistoric Hunters of the High Andes. Academic Press, New York. Sawada, M., 2002, Nearest Neighbor Analysis/Event–Event Distances. Electronic document <http://arcscripts.esri.com/details.asp?dbid=12227> <http://arcscripts.esri.com/details .asp?dbid=12227> (accessed  (accessed 03.01.09). Sch Schurr urr,, M.R., M.R., Sch Schoen oening inger, er, M.J., M.J., 1995. 1995. Ass Associ ociati ations ons bet betwee ween n agricu agricultu ltural ral intensification and social complexity: an example from the prehistoric Ohio valley. Journal of Anthropological Archaeology 14, 315–339. Seong, C., 2004. Quartzite and vein quartz as lithic raw materials reconsidered: a view from the Korean Paleolithic. Asian Perspectives 43 (1), 73–91. Scuderi, L.A., 1987. Late-Holocene upper timberline variation in the southern Sierra Nevada. Nature 325, 242–244. Sierra Nevada Ecosystems Project (SNEP), 1996. Sierra Nevada Ecosystem Project, Final Report to Congress, Status of the Sierra Nevada Centers for Water and Wildland Resources, University of California, Davis. Simon, H.A., 1957. Models of Man: Social and Rational. Wiley, New York. Snyder, J.W., 2001. Map and Comments on Indian Trails on the Sierra National Forest. Yosemite Research Library, Yosemite National Park. Stanish, C., 2003. A Brief Americanist Perspective on Settlement Archaeology. In: Levanthal, J.K.P.a.R.M. (Ed.), Theory and Practice in Mediterranean Archaeology: Old World and New World Perspectives Perspectives.. Cotsen Institute Institute of Archa Archaeolog eology. y. University of California, Los Angeles, Los Angeles, pp. 161–172. Stevens, Steve ns, N.A., 2002. Prehistoric Prehistoric Use of the Alpine Sierra Nevad Nevada: a: Archaeologi Archaeological cal Investigat Inves tigation ion at Taboose Taboose Pass, Kings Canyon Canyon National National Park, Californ California, ia, M.A. Thesis, Department of Anthropology, California State University, Sacramento, Sacramento, California. Stevens, N.A., 2003. Spatial and temporal patterning of bedrock mortar sites in the

Underhill, A.P., Feinman, G.M., Nicholas, L.M., Fang, H., Luan, F., Yu, H., Cai, F., 2008. Changes Chang es in Regional Regional Settlemen Settlementt Patterns Patterns and the Devel Developmen opmentt of Complex Complex Societ Societies ies in Southe Southeast astern ern Shando Shandong, ng, Chi China. na. Journa Journall of Ant Anthro hropol pologi ogical cal Archaeology 27, 1–29. Urban, D., Miller, C., 1996. Modeling Sierran Forests: Capabilities and Prospectus for Gap Models, Sierra Nevada Ecosystem Project, Final Report to Congress, Status of the Sierra Nevada. Centers for Water and Wildland Resources, University of  California, Davis. Voorrips, A., O’Shea, J.M., 1987. Conditional spatial patterning: beyond the nearest neighbor. American Antiquity 52, 500–521. Walsh, K., 2005. Risk and marginality at high altitudes: new interpretations from fieldwork on the Faravel Plateau, Hautes-Alpes. Antiquity 79, 289–305. Walsh, K., Riche Richer, r, S., Beaulieau, Beaulieau, J.-L.d., 2006. Attitudes to altit altitude: ude: changing meanin mea nings gs and percep perceptio tions ns within within a ‘Margi ‘Marginal nal’’ alpine alpine landsc landscape ape – the integratio integ ration n of paleoecolo paleoecological gical and archaeologi archaeological cal data in a high-altitu high-altitude de landscape in the French Alps. World Archaeology 38, 436–454. Washburn, Washbu rn, J., 1974. Neare Nearest st neigh neighbor bor analysis analysis of Pueblo Pueblo I–III settlement settlement patterns along the Rio Puerco of the East, New Mexico. American Antiquity, 39. White, D.M., 1985. Settlement Pattern Implications of Bedrock Milling Implement Analys Analysis: is: An Exampl Example e fro from m the Kern Kern River, River, Sou Southw thwest est Ant Anthro hropol pologi ogical cal Association Meetings, Chico, California. Willey, G.R., 1953. Prehistoric Settlement Patterns in the Viru Valley, Peru. Bureau of Ethnology Bulletin 155. Smithsonian Institution, Washington, DC. Winterhalder, B., 1980. Environmental analysis in human evolution and adaptation research. Human Ecology 8, 135–170. Winterhalder, B., Leslie, P., 2002. Risk-sensitive fertility, the variance compensation hypothesis. Evolution and Human Behavior 23, 59–82. Woodburn, J., 1980. Hunters and gatherers today and reconstruction of the past. In: Gellner, E. (Ed.), Soviet and Western Anthropology. Duckworth, London, pp. 95– 117. Woolfenden, W.B., 1996. Quaternary vegetation history, Sierra Nevada Ecosystem Project, Final Report to Congress, Status of the Sierra Nevada. Centers for Water and Wildland Resources, University of California, Davis, pp. 47–70. Wri Wright ght,, G.A., G.A., Bender Bender,, S., Reeve, Reeve, S., 1980. 1980. Hig High h cou countr ntry y adapta adaptatio tions. ns. Plains Plains Anthropologist 25, 181–197.

southern Sierra Nevada: a regional exploration. Proceedings of the Society for California Archaeology 16, 175–182. Stevens, N.A., 2005. Changes in prehistoric land use in the alpine Sierra Nevada: a regional region al expl exploratio oration n using temperatur temperature-adj e-adjusted usted obsidian obsidian hydration hydration rates. rates.  Journal of California and Great Basin Anthropology 25, 187–205. Surovell, Surove ll, T.A., 2000. Early Paleoindian Paleoindian women, women, child children, ren, mobility, mobility, and fertility. fertility. American Antiquity 65, 493–508. TCR/ACRS, TCR/A CRS, 1984. Cultural Cultural Resources Resources Overview of the Southern Sierra Nevada. Nevada. Theodo Theodorat ratus us Cul Cultur tural al Res Resear earch, ch, Inc. Inc. and Archae Archaeolo ologic gical al Consul Consultin ting g and Res Resear earch ch Ser Servic vices, es, Inc., Inc., Prepar Prepared ed for USD USDA A Forest Forest Servic Service, e, South South Centra Centrall Contracting Office, Bishop. Thomas, D.H., 1973. An empirical test for Steward’s model of Great Basin settlement patterns. American Antiquity 38, 155–176. Thomas,, D.H., 1976. The awful truth about statistic Thomas statisticss in archaeology archaeology.. American American Antiquity 43, 231–244. Thomas, D.H., 1982. The 1981 Alta Toquima Village Project: A Preliminary Report. Desert Research Institute Social Sciences and Humanities Publications 27.

Yellin, J., Levy, T.E., Rowan, Y.M., 1996. New evidence on prehistoric Yellin, prehistoric trade routes: the obsidian evidence from Gilat, Israel. Journal of Field Archaeology 23, 361– 368. Zeanah, D.W., 2000. Transport Costs, Central Place Foraging, and Hunter–Gatherer Al Alpi pine ne La Land nd Use Use Stra Strate tegi gies es.. In In:: Ma Mads dsen en,, D.B. D.B.,, Me Metc tcal alfe fe,, M.D. M.D. (Eds (Eds.) .),, Intermount Intermountain ain Archaeology Archaeology.. Unive University rsity of Utah Anthropolo Anthropological gical Papers Papers No. 122, Salt Lake City, pp. 1–14. Zeanah, D.W., 2002. Central place foraging and prehistoric pinyon utilization in the Great Basin. In: Fitzhugh, B., Habu, J. (Eds.), Beyond Foraging and Collecting: Evo Evolut lution ionary ary Change Change in Hunter Hunter–Ga –Gathe therer rer Set Settle tlemen mentt System Systems. s. Klu Kluwer wer Academic/Plenum Publishers, New York, pp. 231–256. Zhang, H.Q., Zhao, W.M., Liu, B., 2007. Mathematical Modelling of the Relationship Between Betw een Neolithic Sites and the Rivers In Xi’an (Shaanxi (Shaanxi Province, Province, China). Archaeometry 36, 765–773. Zvelebil, Zveleb il, M., Green, S.W., Mack Macklin, lin, M.G., 1992. Archa Archaeolog eological ical landscape landscapes, s, lithic scatters, and human behavior. In: Rossignal, J., Wandsnider, L. (Eds.), Space, Time, and Archaeological Landscapes. Plenum Press, New York, pp. 193–226.