See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/259482723

Antiquity of Communal Pronghorn Hunting in the North-Central Great Basin

Article  in  American Antiquity · October 2009

DOI: 10.1017/S0002731600049027

CITATIONS

28READS

251

2 authors, including:

Some of the authors of this publication are also working on these related projects:

Investigation of Open-Air Western Stemmed Sites, Harney Basin, Oregon View project

Communal game traps View project

Bryan S. Hockett

BLM - The Bureau of Land Management

72 PUBLICATIONS   2,043 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Bryan S. Hockett on 30 December 2013.

The user has requested enhancement of the downloaded file.

https://www.researchgate.net/publication/259482723_Antiquity_of_Communal_Pronghorn_Hunting_in_the_North-Central_Great_Basin?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_2&_esc=publicationCoverPdfhttps://www.researchgate.net/publication/259482723_Antiquity_of_Communal_Pronghorn_Hunting_in_the_North-Central_Great_Basin?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_3&_esc=publicationCoverPdfhttps://www.researchgate.net/project/Investigation-of-Open-Air-Western-Stemmed-Sites-Harney-Basin-Oregon?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_9&_esc=publicationCoverPdfhttps://www.researchgate.net/project/Communal-game-traps?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_9&_esc=publicationCoverPdfhttps://www.researchgate.net/?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_1&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Bryan-Hockett?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_4&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Bryan-Hockett?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_5&_esc=publicationCoverPdfhttps://www.researchgate.net/institution/BLM-The_Bureau_of_Land_Management?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_6&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Bryan-Hockett?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_7&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Bryan-Hockett?enrichId=rgreq-adeab0f84dd54d91dfc6fcad9a909e87-XXX&enrichSource=Y292ZXJQYWdlOzI1OTQ4MjcyMztBUzoxMDMxMjE3MDM0MDc2MTZAMTQwMTU5NzUzODY1Nw%3D%3D&el=1_x_10&_esc=publicationCoverPdf

CHockett and Murphyl

Lubinski (1997, 1999her of culturally and nhorn bonebed assemblcontaining> 5 MNIhuman-caused morUepisode, and a singlecommunal kilt (LubitPoint site located in socis and Widman (199the discovery of nun258) and butcheredlocated within a topolalong a major pronghbination of large numtopographic context Iidiscussions of ancientthe Great Basin (McC

Following the dischorn traps or corrals in(Murphy and Frarnpto:(1992) reported on thin northeastern NevaPoint Site and associasite was discovered dalong an existing precorded during thisand ephemeral in natuwas unique —more Uiconcentrated in a rel:mately 50 m by 50 msisted of Humboldt p3,000 years ago. Peterelied on projectile pcsit&s overall uniquenscatter probably repreor killing area of a coiof the kind known totand elsewhere in the Ctoric times.”

Despite this evid(1997:90) suggestedof projectile points swresent time-averagedby a mobile huntinggatherer bands took atunkies as they encosround:’ At the same tithe potential of equifiprojectile point clusancient communal kill

ANTIQUITY OF COMMUNAL PRONGHORN HUNTINGIN THE NORTH-CENTRAL GREAT BASIN

Bryan Hockett and Timothy W. Murphy

Communal hunting of small game such as hares has probably occurred for 10)000 years in the Great Basin. Ethnohistoricaccounts of the nineteenth century indicate that indigenous peoples communally hunted large game (e.g., pwnghorn. mountain sheep. deen bison) across much ofwestern North America including the Plains, desert Southwest, California, and GreatBasin subregions, during and immediately preceding the contact era. Research in the Plains subregion suggests that communal large game hunting occurred there prior to the adoption of the bow-and.arrow between Ca. 1.500 and 2,000 yearsago, and in fact may have occurred as early as 9,000 to 10,000 years ago. Nineteenth- and early twentieth-century ethnohistoric accounts suggest that communal pronghorn (Antilocapra americana) hunts involving the construction ofa corralwith associated wings were utilized by many Great Basin peoples at the time of historic contact. This paper asks: (1) didcommunal pronghorn hunts occur prior to the Protohistoric Period (before ca. 600 ‘4C B.Rj in the north-central GreatBasin? (2) if so, how ancient is this practice? and (3) did the methods or behaviors of the participants of these communalhunts vary through time? Detailed analysis of sites containing dozens, and in many cases, hundreds ofprojecrilepoints thaipredate ca. 600 “C BR found in or near existing juniper branch corrals and wings suggest that communal pronghorn hunt.ing has occurred for at least 4.000 to 5,000 years in the north-central Great Bashj. Furthe, behavioral variability is seenthrough time in the material remains of these communal hunts, with earlier (Middle Archaic) communal kills characterizedby greater use of local toolstone sources, gearing-up just prior to the kill, and perhaps a greater reliance on shooting thetrapped pronghorn rather than clubbing compared to P,vtohistoric com,nunal kills.

Co;nunales de caza de caza mayot como el berrendo (Antilocapra americana) que supongan Ia construcciOn de un corraljuntocon las alas Se utilizaron por muchos pueblos Gran Cuenca, en el momento histdrico de contacto. Este documento buscaresponder a Ia antiguedad de este comportamiento en Ia Gran Cuenca. Los datos sugieren que la caza comunal berrendo seha producido par lo menos 4.000 a 5.000 años en Ia parte norte-centroi tie Gran Cue nca,

QId and New World ethnographies bothrecord examples of communal trapping oflarge game animals by indigenous societies (e.g., Curtis 1924; Hill 1938, 1982; Kluckhohn et at 1971; Legge and Rowley-Conwy 1987;Lindblom 1935; Saad 2005; Stephen 1936; Steward 1938). In western North America, and in particular the Plains subregion, archaeologists haveextended the communal capture of multiple largegame animals by various forms of trapping into theEarly Holocene. perhaps as early as 9,000—10,000years ago (Prison 1987; 2004; Prison et al. 1986;Lubinski 1999, 2000; Miller et al. 1999). In theGreat Basin subregion, there is amp’e archaeolog

ical evidence for the communal capture of prong-horn (Antilocapra americana), deer (Odocoileushemionus), and mountain sheep (Ovis canadensis)during Protohistoric times (ca. 600—150 years ago)(e.g., Arkush 1986, 2007; Hockett 2005; McGuireand Hatoff 1988; Murphey 1980; Murphy andFrampton 1986; Raymond 1982).

The origin of communal large game hunting inthe Great Basin, however, remains uncertain. Oneof the primary hurdles archaeologists face is identifying distinctive archaeological traces that distinguish ancient communal trapping behaviorsfrom other forms of hunting such as ambushing bylone hunters. For the North American Plains,

Bryan Hockettu Bureau of Land Management. Elko District Office. 3900 East Idaho Street, Elko, Nevada 89801(Bryan_Ilockett@nv.blrn.gov)Timothy W. Murphy IBureau of Land Management, Elko District Office, 3900 East Idaho Street, Elko, Nevada 89801(Tim_Murphy@nv.blm.gov)

American Antiquity 74(4). 2009, pp. 708—734Copyright ©2009 by the Society for American Archaeology

708

‘4TING

-eat Basin. Ethnohistoric(e.g., prangharn, ,noun

‘St. California. and Greategion suggests that corni. 1,500 and 2,000 yearstwentieth-century ethnoconstruction of ci corralThis paper asks: (J) didthe north-central Great

,ants of these communalafprojectile points that

nmunal pronghorn hunti’ioral variability is seensitnal kills characterizedreliance on shooting the

tcciOn de un corraljuntaEste documento busca

lza coniunal berrendo Se

al capture of prong), deer (Odocoileusip (Ovis conadensis)600—150 years ago)

dcett 2005; McGuire1980; Murphy and82).rge game hunting inains uncertain. One)Iogists face is idenical traces that distrapping behaviorsich as ambushing byAmerican Plains,

Nevada 89801

Elko, Nevada 89801

[Hockett and Murphyl

Lubinski (1997, 1999, 2000) investigated a number of culturally and naturally accumulated prong-horn bonebed assemblages. He concluded that sitescontaining> 5 MN! pronghorn with evidence forhuman-caused mortality, a single depositionalepisode, and a single mortality event suggested acommunal kill (Lubinski 2000). At the TrappersPoint site located in southwestern Wyoming, Francis and Widman (1999) noted that in addition tothe discovery of numerous projectile points (N =258) and butchered pronghorn bone, this site islocated within a topographically constricted zonealong a major pronghom migration route. A combination of large numbers of projectile points andtopographic context has also factored in previousdiscussions of ancient large game trapping sites inthe Great Basin (McGuire and Hatoff 1988).

Following the discovery of numerous prong-horn traps or corrals in the north-central Great Basin(Murphy and Frampton 1986), Petersen and Stearns(1992) reported on the Clover Valley site locatedin northeastern Nevada. Similar to the TrappersPoint Site and associated survey, the Clover Valleysite was discovered during a CRM-driven surveyalong an existing paved highway. Most sitesrecorded during this survey were relatively smalland ephemeral in nature. But the Clover Valley sitewas unique —more than 250 projectile points wereconcentrated in a relatively small area (approximately 50 m by 50 m). The majority of these consisted of Humboldt points manufactured prior to3,000 years ago. Petersen and Steams (1992:92)relied on projectile point damage patterns and thesite’s overall uniqueness to argue “that the pointscatter probably represents the site of the ‘corral’or killing area of a corral-and-wings antelope trapof the kind known to have existed in Clover Valleyand elsewhere in the Great Basin during ethnohistoric times.”

Despite this evidence, Zeanah and Elston(1997:90) suggested that sites with large numbersof projectile points such as Clover Valley may represent time-averaged “lithic scatters . . . producedby a mobile hunting strategy. in which hunter-gatherer bands took advantage of hunting opportunities as they encountered them on a foraginground.” At the same time, they still acknowledgedthe potential of equifinality, or the fact that theseprojectile point clusters may indeed representancient communal kills. One issue that Zeanah and

Elston (1997) found particularly troubling aboutsites such as Clover Valley was the fact that projectile point manufacture and repair occurred there,and if, as the Great Basin ethnographic record indicates, communal hunts involved long-range planning and “fandangos” following the kill, thenhunters should have been fully geared-up prior tothe hunt, making on-the-spot point manufactureunlikely.

Unknown to both Petersen and Stearns (1992)and Zeanah and Elston (1997), however, the CloverValley site is situated near the center of the greatest concentration ofjuniperbranch corrals and largeprojectile point concentrations (defined as morethan 20 points each) known from the entire GreatBasin subregion (Hockett 2005). In addition, siniilar to Trappers Point these sites (collectively calledthe Spruce Mountain Trap Complex area [SMTCarea]) are located along a north-south migrationcorridor for pronghorn. The Clover Valley site isalso located within a topographic constriction alongthis route. As discussed in further detail below,however, both existing wooden corrals and chronologically earlier Middle Archaic-aged projectilepoint concentrations are directly associated withone another to the north and south of this constriction. Importantly, a major source of artifact-quality chert outcrops less than one mile from thecenter of the constricted killing zone, and artifactquality basalts, rhyolites, siltstones, and argilitescan be procured within a few hours walk. Finally,our reading of the ethnographic literature on communal pronghom trapping suggests that there wasin fact a variety of methods and personnel used inthe communal trapping of pronghom by differentsocieties, such that a single archaeological signature resulting from behavioral decisions aboutwhether or not to gear-up before the hunt or to manufacture projectile points next to the killing placecannot be assumed. In other words, different methods and motivations behind the decision to communally trap pronghorn may result in differentarchaeological signatures of material remains.Given the relatively large number of existing corrals, projectile point concentrations, and the timedepth involved at the SMTC area (5,000—6,000years), these data are prime for the exploration ofissues related to the origins and methods of com

munal large game trapping in the Great Basin.Hockett (2005) previously reported on the

ANTIQUITY OF COMMUNAL PRONGHORN HUNTING 709

710

SMTC area corrals and point concentrations in ageneral discussion of the possible time-depth ofcommunal pronghorn trapping or surrounding inthe north-central Great Basin. Below we providethe details necessary to expand and further refinethis issue, as well as to suggest that a variety ofmethods and motivations may have been employedthrough time in communally capturing pronghomin this region. As any good scientific investigationin the historical sciences should, we utilize a combination of inductive and deductive methods toapproach the problem.

Communal Pronghorn Trapping:The Ethnographic Evidence

Lubinski (1997, 1999) previously reported onethnographic evidence for communal pronghonihunts in western North America. Here we wish tohighlight two facts: (1) human behavior associatedwith communal pronghorn hunts varied betweensocieties, and (2) there is no reason to believe aprion that the behaviors recorded ethnographicallyalso guided the motivations and behaviors of similar communal hunting episodes recorded in thearchaeological record. The ethnographic record isreplete with accounts of pronghorn procurement byboth individuals and groups of hunters. Thesedemonstrate not only that there were a number ofmethods employed to successfully hunt pronghom,but also that these variations in hunting methodologies could have resulted in very similar archaeological traces (equifinality).

Pronghom habitually congregate into largerherds and migrate between summer and winter pastures (Sundstromet al. 1973). This makes them susceptible to predation along migration routes that arepredictable in both space and time. In the GreatBasin, modem and ancient migration routes can beexpected to primarily occur north-south rather thaneast-west due to the abundance of north-southtrending mountain ranges.

Individuals or small groups of hunters usuallypracticed methods such as stalking, ambushing,and luring animals to the hunter. Disguises madeof brush orpronghorn hides with the head and hornsattached were often used in stalking (Steward 1941,1943). Hunters sometimes lured animals withinbow range by waving a flag or other object thatcaught the animals’ curiosity. Ambushes involved

[Vol. 74, No. 4, 2009

hunters hidden along game trails, at water sources,or at other locations frequented by the animals.Hunters might be hidden in blinds, pits in theground, or be wearing disguises. V-fences were attimes employed to funnel the game toa hiddenhunter. Ambushing could be done by individualsor involve both a hidden hunter and drivers.Hunters also used nooses, disguised deadfall pits,or spears placed in trails to catch pronghorn (Steward 1943). All of these methods could have resultedin the loss of a relatively small number of projectile points at individual locales, masking the potential variability in hunting behaviors employed.

The preceding methods usually resulted in theacquisition of one or a few animals. Taking of largenumbers of pronghorn required multiple participants and typically was a communal affair. Theterm “communal” will be used here to simply referto the aggregation of multiple families or groupsof individuals in the procurement of specificresources in space and time. These aggregationsmay have included all family members workingcooperatively (men, women, children, grandparentsetc.), and have taken many forms such as adult menonly, adult men and women only, and the like.

The context within which communal pronghornhunts took place suggests that these events mayhave led to the creation of different archaeologicaltraces from those briefly described above. One ofthe most obvious differences is the greater numberof projectile points that may have been lost or discarded at communal kill spots compared to individuals stalking or ambushing game. As noted,communal hunts could have involved the gathering of entire bands or family groups in which allmembers participated in the hunt in some manner.In these circumstances, all members of individualgroups would move near the location of the communal hunt. Individual members may have playeda variety of roles in the communal hunt based onage, sex, and status, such as serving as shaman orleader, participating in constructing the corral andwings, serving as a scout, assisting in driving theanimals toward the enclosure, ensuring the animalsdid not escape through the sides of the wings orback through the corral entrance, and being designated as a shooter or dubber. Carcasses likelywould have been processed and probably consumedat or near the kill, particularly if the groups stayedfor an extended period of time to complete impor

[Iockett and Murphy]

tant social events suchmatchmaking. If indivieled relatively long disand if lithic raw materavailable at the kill siteably would have gearecIn these cases discardethe kill spot should betoolstone materials.

In other cases, co:involved only adult ifterm, task-specific bcnear the kill location,casses being transpowhere the remainder olIn these circumstancescompleted the taskssources were located alsite itself, many of thebeen manufactured juof locally available rn:

These two simple e:others that could be dtial variability related t’relations, toolstone prfacturing patterns andan archaeological recsubsistence-related taof multiple large gainisurprising to find thattmunal pronghom hutethnographically. Thismunal pronghorn huntypes: relay, surroundwhere hunters wouldmals until they readoccurred on foot, amThe Achomawi of Calito chase down herds(Curtis 1924:140).

Surrounding genetviduals encircling a hwere isolated or concspace, and then shoowere within the hunteaccounts of surrounchsoit is possible that thiquently during postccica. William of Rubnaccounts of a surroun

AMERICAN AN11QUn’Y

[Vol. 74, No. 4, 2009 Hockelt and Murphy] ANTIQUITY OF COMMUNAL PRONGHORN HUNTING 711

uls, at water sources,ted by the animals.

i blinds, pits in the;es. V-fences were ate game to a hiddendone by individualsiunter and drivers.•guised deadfall pits,ch pronghom (Stews could have resultedII number of projec

masking the potenaviors employed.ually resulted in thetrials. Taking of largered multiple particiunmunal affair. TheI here to simply refere families or groups.rement of specificThese aggregations

y members workingalidren, grandparentsms such as adult meniily, and the like.ommunal pronghornat these events mayèrent archaeologicalribed above. One ofs the greater numberave been lost or diss compared to mdiag game. As noted,involved the gather-groups in which allunt in some manner.mbers of individuallocation of the comers may have playednunal hunt based onerving as shaman oracting the corral andisting in driving theensuring the animalsides of the wings orice, and being desiger. Carcasses likely1 probably consumedif the groups stayedto complete impor

tant social events such as enhancing alliances andmatchmaking. If individual groups or families traveled relatively long distances to reach the trap site.and if lithic raw material sources were not readilyavailable at the kill site, designated shooters probably would have gearedup prior to joining the hunt.In these cases discarded or lost projectile points atthe kill spot should be made primarily of nonlocaltoolstone materials.

In other cases, communal hunts could haveinvolved only adult men. In these cases shorter-term, task-specific locales may have been creatednear the kill location, with all or part of the carcasses being transported back to “base camps”where the remainder of the group members resided.In these circumstances only adult men would havecompleted the tasks listed above. If toolstonesources were located along the journey or at the trapsite itself, many of the projectile points may havebeen manufactured just prior to the kill, and madeof locally available materials.

These two simple examples (and there are manyothers that could be derived) highlight the potential variability related to social organization, humanrelations, toolstone procurement, and tool manufacturing patterns and the subsequent creation ofan archaeological record resulting from a similarsubsistence-related task: the communal corrallingof multiple large game animals. It is therefore notsurprising to find that behavioral variability in communal pronghorn hunting has been documentedethnographically. This literature suggests that communal pronghorn hunts consisted of at least threetypes: relay, surrounding, and corralling. Relays,where hunters would take tums pursuing the animals until they reached a point of exhaustion,occurred on foot, and postcontact on horseback.TheAchomawi of California used therelay methodto chase down herds of pronghorn on snowshoes(Curtis 1924:140).

Surrounding generally involved multiple individuals encircling a herd of pronghorn until theywere isolated or concentrated within a restrictedspace, and then shooting the animals when theywere within the hunters range. Most ethnographicaccounts of surrounds mention the use of horses,so it is possible that this method was used more frequently during postcontact times in North America. William of Rubruck gave one of the earliestaccounts of a surround in the 1250s. William was

a Franciscan monk who travelled through Mongolterritory between 1253—1255, and upon his returntrip wrote an ethnography of the peoples he encountered. He described the Mongol’s use of the communal surrounding technique:

So it is that they procure a large pan of theirfood by the chase. When they want to chasewild animals, they gather together in a greatmultitude and surround the district in whichthey know the game to be, and gradually theycome closer to each other till they have shutup the game in among them as in an enclosure,and then they shoot them with their arrows{Saad 2005:20].

Hill (1938) specifically notes the use of a surround by the Navajo, but he notes that surroundswere less common, less ritualistic, and may haveoccurred more commonly after the adoption ofhorses during the contact era. In a surround, thepronghom

were surrounded on horseback . . . at leasttwenty men were necessary. The circle wascontracted to a diameter of a hundred to a hundred and fifty yards, then the hunters beganshooting the antelope with bows and arrows[Hill 1938:145].

Given this latter scenario, we might expect to finda number of complete and broken projectile pointswithin a 50—100 m zone atthe surround orkilispot.

Egan (1917) and the Nw York Times (1895)represent two accounts in which communal prong-horn corralling were witnessed firsthand ordescribed shortly after they occurred. Egan witnessed a successful communal pronghom corralling by the Goshute of eastern NevadalwesternUtah, and the New York limes reported on a similar event for the Navajo inArizona. Steward (1938)and Hill (1938) provide some of the more detailedsecondhand accounts of communal pronghom corralling. The following summary is based primarilyon these four works,

Most, but not all, of the ethnographic accountswe reviewed report the use of a “leader” or“shaman” to guide a successful communal prong-horn corralling. In all cases, this leader was male.The descriptions of communal pronghorn corralling in the Great Basin by Howard Egan andJulian Steward suggest that, in some cases, both

712

men and women participated in the construction ofthe corral and wings, as well as in maintaining theherd inside the structure. Men were generally sentout as scouts, to begin the driving of the herd towardthe corral-and-wing structure, and in shooting theanimals.

W. W. Hill’s (1938) and the New York Thiies’(1895) descriptions of communal pronghorn drives conducted by the Navajo are two of the mostdetailed ever written. These descriptions differfromthose of Egan and Steward in that only men participated in the initial journey to the trap site andthe actual construction of the corral itself, TI womenwerepresent they participated in cooking and campduties but were excluded from the actual preparation and construction of the corral and wings, aswell as in the drive itself. Hill specifically notes that20—50 men were required to complete a successful communal pronghom drive. Hill (1938: 149)states that several days of travel and ceremonywould commence before construction began. Acorral took an average of five days to complete, andconstruction of the wings took an additional 2—3days. This is potentially significant informationbecause it suggests that a communal pronghornhunt took up to two weeks to complete. This suggests that prior gearing-up in terms of projectilepoint manufacture was probably not critical ifsources of artifact-quality raw material were locatedalong the journey from the base camp to the trapsite or if a quarry was located near the trap itself.

Additional information provided by these lattertwo accounts that are not found in sources such asSteward (1938) includes: (1) the juniper and pinetrees used in the construction of the corral wereburned down prior to theft use as construction material; (2) if horses were used to drive the pronghominto the corral, men immediately shot the animalsas they entered the structure; if the pronghom weredriven on foot, then the men rested before shooting the animals inside the corral; (3) the animalswere not skinned inside the corral where they hadbeen killed; they were taken outside of the structure to be butchered; (4) following skinning, themeat was taken to the various camps or structuresbuilt by the hunting party prior to the drive; (5) ifinsufficient numbers of animals were captured during the first drive, then another drive took place thefollowing day; (6) the meat was carried back to basecamps on wooden frames—each man could carry

[Vol. 74, No. 4, 2009

three dismembered pronghom carcasses; and (7)heads and horns were ceremonially left behind atthe kill site in the crotch and base of trees.

The New York Times’ description, as well asmore recent twentieth—century accounts of the corralling of pronghorn by various wildlife agenciesfor transplanting purposes, provide additional information for studying the archaeological remains ofcommunal corrals. For examples, recent photos ofcorralled pronghorn show that they bunch togetherin a tight circle within the confines of the larger corral (e.g., Yellowstone Digital Slide, File http://www.nps.gov/archive/yel]Islidefile/mammalslpronghomantelopefPage-2.htm). Indeed, the New YorkTimes (1895) reported that “The shouts and yellsof the Indians at length so completely terrify thepoor animals that they stand trembling and apparently unconscious of the Indians, who nowapproach and kill them with clubs, hatchets, stones,or anything at hand:’

Furthermore, these accounts help explain thetopographic position of existing corrals and projectile point concentrations (or possible ancient,now-degraded corrals) on the landscape. A common feature noted by archaeologists who haverecorded a number of corrals in the Great Basin isthe fact that the corral walls often sharply angle atthe point of contact between the beginning of thewing and the corral entrance. In addition, the backside of the corral itself (opposite the entrance) isoften located on the opposite side of a hill or ridge:Standing at the entrance, this gives the impressionthat the narrow passageway into the corral “opens-up” into wide open space, rather than the impression of an enclosed structure. Corrals constructedin this manner represent carefully planned use ofthe landscape in general and micro-topography inparticular based on prior knowledge of pronghornbehavior because these animals become “spooked”if they can see the walls and backside of the structure while being coaxed through the entrance. Notethe references to the Navajo’s use of the landscapein constructing a corral in the following passagesfrom the New York Thnes:

This corral was built in 1890 by the order ofthe old chief, Gano Mumcho, and his son,Many Horses. A place was selected where aslight hill was found upon the border of a wideprairie.

Hockett and Murphyl

A close pen, circiacross, was firstbthe hill so that thenot be seen from

After being closecannot turn back,suit, and they mahill, which to thening between the]But, alas! for thenclosed in a stouin a circle, never tilimes 1895].

In the late 1930sandpish departmentappropriate methocpronghorn using wilar to what Native Amany millennia earl

In 1937, Paul Rusthe New MexicoFish, invented a rplant herds of prctrapping method,:regular pronghorcowboys on horsthe trap. The firstwest rodeo seen

‘ rebounding off thback into the dricboys roping them.in few if any capflby a hidden trapherd once past ancovered with tarp[Santa Fe Guidin;

Archaeological Rattaric Literature: Ba

Ethnographic accothunting suggest thamay or may not cortile points. Few or nclost within the conflimals were clubbedaccounts suggest thshot, and in these caprojectile points I

AMERICAN ANTIQUITY

Hockett and Murphyl

A close pen, circular in form, about 100 feetacross, was first built against the steep side ofthe bill so that the tops of the posts used couldnot be seen from the opposite side.

After being closed in upon they [pronghorn]cannot turn back, as the drivers are in hot pursuit, and they make a dash for the top of thehill, which to them appears to be the only opening between the horrid lines of brush fence.But, alas! for them, they find themselvesenclosed in a stout pen, around which they runin a circle, never trying to jump over [New YorkTimes 1895].

In the late 1930s and early 1940s, State Gameand Fish departments learned through thai and errorappropriate methods of communally capturingpronghom using wing and corral structures simi1w to what Native Americans had known and usedmany millennia earlier:

In 1937, Paul Russell, awlldlifebiologistwiththe New Mexico Department of Game andFish, invented a method of trapping to transplant herds of pronghom. Using a wild horsetrapping method, he set up a winged trap alongregular pronghom routes and used cars andcowboys on horseback to ease the herds intothe trap. The first few attempts ended in a wildwest rodeo scene with the terrified herdsrebounding off the back of the trap, sprinting,back into the drivers, wrecking on cars, cow

boys roping them, and general melee resultingin few if any captures. The system was revisedby a hidden trap door to be closed behind theherd once past and the walls of the trap beingcovered with tarps for sight proof and padding[Santa Fe Guiding Company 2004).

Archaeological Ramifications of the Ethnohistoric Literature: Building Working Hypotheses

Ethnographic accounts of communal pronghornhunting suggest that ancient corrals or kill spotsmay or may not contain large numbers of projectile points. Few or no lithic anifacts may have beenlost within the confines of corrals in which the animals were clubbed. Nevertheless, ethnographic

accounts suggest that the animals were sometimes

shot, and in these cases both complete and broken

projectile points probably would have been

713

deposited and lost inside a corral or within the confines of a surround. Concentrations of broken projectile points should be located primarily within thecorral itself; it is unlikely pronghorn were shotwhile being driven between the wings toward theenclosure. These projectile points may be locatedat the entrance to a corral if the hunters chose toshoot the animals as they entered, or if the animalswere shot as they attempted to escape back through

the entrance once within the corra]. Projectile pointsmay also encompass a much broader area within

the corral. Given the fact that individual corralsmeasure between 100—500 m in diameter, and thefact that pronghorn bunch together when trappedinside a corral, broken projectile points could beexpected to be concentrated at the spot where theherd was bunched together, as well as additionalspots where individuals or smaller bunches mayhave stood. Indeed, clusters of Protohistoric-aged

Desert Side-Notched points associated with existing corrals in the SMTC area are sometimes locatedat the entrance, while other times they are spreadthroughout the inside of the enclosure (see belowfor details).

Depending on where the animals were shotinside a corral, projectile point concentrations may

be located in different micro-topographic zones.Projectile points may be clustered in,a relatively

flat topographic zone or at the base of a hill, ridge,or knoll if the animals were shot at or near theentrance. Alternatively, they may be located atop ahill, ridge, or knoll if the animals were dispatchedtoward the back of the enclosure.

There may be no definitive characteristics thatdistinguish surrounds from ancient corrals. Both

may contain large numbers of similar-styled projectile points concentrated either at the base or atop

a ridge, hill, or knoll. However, if the majority ofancçnt kill spots in the SMTC area represent sur

rounds rather than corrals, then we might expect

them to have a rather random distribution throughout the research area. This is the case because the

herds of pronghorn would not have been directedto a single, predetermined place on the landscape,

as would be the case for corral construction. Thus,

there would be no reason to expect a tendency ofassociation between existing corrals and kill spotsresulting from surrounding events.

Alternatively, if specific places on the landscape

have served for millennia as appropriate locations

AN’rIQtJrry OF COMMUNAL PRONGRORN HUNTING[Vol. 74, No. 4, 2009

)rn carcasses; and (7)ioniaJ)y left behind atI base of trees.escription, as well asry accounts of the corious wildlife agenciesovide additional inforaeological remains ofiples, recent photos ofat they bunch togetherifines of the larger cortal Slide, File http://idefile/mammals/pronIndeed, the New York‘The shouts and yellsompletely terrify thetrembling and apparIndians, who now

:lubs, hatchets, stones,

unts help explain the;ting corrals and pro-(or possible ancient,

le landscape. A comaeologists who havein the Great Basin is

often sharply angle atthe beginning of theIn addition, the backosite the entrance) isside of a hill or ridge.gives the impression

nto the corral “opensither than the impres

Corrals constructedefully planned use ofmicro-topography in

iwledge of pronghornils become “spooked”backside of the strucigh the entrance. Notes use of the landscape.e following passages

90 by the order oficho, and his son,s selected where ahe border of a wide

714

for corralling herds of pronghorn due to their topographic position, then we may indeed find an association between existing corrals and more ancientprojectile point concentrations. Given the fact thatwing and corral structures need to be designed toaccommodate specific pronghorn behaviors, wemight expect corrals to have been constructed insimilar places on the landscape through time providing that migration routes remained relativelyconsistent. Put another way, consistency in prong-horn migration routes and a primary concern forutilizing appropriate topography to ensure a successful hunt should lead to similar distributionsamong existing corrals that have survived due totheir relatively young age (along with age-appropriate projectile points if the animals wereshot rather than clubbed) and ancient corrals inwhich the only remaining traces would be projectile point concentrations, associated lithic artifacts,and perhaps bonebeds.

If the pronghorn carcasses were taken outsideof the corral for butchering, then campsites containing burned and butchered bone would belocated some distance away from the actual kill site.However, these campsites are not likely to havebeen located far from the kill, TI the carcasses werebutchered andlor cooked inside the corral, then thiswould suggest behavior different from thatdescribed etimographicaily for groups such as theNavajo, where this behavior was forbidden on symbolic grounds.

Tfpronghorn were shot inside a corral, kill spotsshould be dominated by high tool to flake ratios,clusters of single-type points, and relatively largenumbers of broken points in the form of midsections and tips, although both complete and basalpoint fragments should also occut Equifinality isoften an issue in the analysis of archaeological patterns, and the study of potential ancient communalpronghorn corrals is no exception. Nevertheless, asothers have previously suggested (e.g., Petersenand Steams 1992), sites created away from theactual kill spots where broken projectile pointswere unbound and discarded from retrieved fore-shafts should be heavily dominated by point bases,with far fewer tips, midsections, and completepoints compared to the kill spots themselves. Thus,our analysis of the SMTC point clusters includespercentages of complete, basal, midsection, and tipportions. In addition, the location of the SMTC

[Vol. 74, No. 4, 2009

point clusters were mapped in relation to existingcorrals to determine if they were associated on thesame or similar landform features.

Finally, we also analyzed the raw material usedto manufacture the projectile points recovered fromthe corrals and point concentrations. We comparedthese data with the known distribution of toolstonequarries in the region to help determine the degreeto which hunters geared-up prior to entering theSMTC area to hunt pronghom.

Corrals and Point Concentrations in theSMTC Area—General Introduction

A general description of the location, topography,and vegetation found within the SMTC area can befound in Hockeu (2005). Here we note that theSMTC area currently ejkcompasses approximately15,000 acres of land surVè3’ed at < 30 m spacing(Figure 1). The area is characterized by flat valleyfloors dissected by dozens of ephemeral drainages.These drainages create dozens of low-lying fingerridges. These ridges tend to run north-south in thesouthern zone of the SMTC area, and east-west inthe central and northern regions. The area is relatively open, but it constricts near the northern endof the surveyed zonejust west of a steep bluff (Figure 1). This zone is known as the “hourglass”because the ground flattens out, or “opens up” tothe north and south of the constriction. It is boundedon the west and east by the uplands ofValley Mountain and Spruce Mountain, respectively. The lowest elevations are covered by a near-monocultureof big sagebrush, while the remainder is coveredmainly by big sagebrush-Utahjuniper habitat. Thejuniper trees served as the primary building material for the corrals. Some of the existing corral wallsrun directly into standing live juniper trees, suggesting that these were alive when the corrals werebuilt.

Petersen and Stearns (1992) recorded the firstpoint concentration (Clover Valley) in the SMTCarea. Subsequent to this survey, we (Hockett andMurphy 1993) recorded the first corral and the second projectile point concentration (Knoll site) 15years ago. The remainder of the corrals and pointconcentrations were recorded between 2000 and2004 during a series of surveys overseen by Hockett. Approximately two dozen corrals have beenrecorded outside of the SMTC area as well, but no

Phase

Eagle RockMaggie CreekJames CreekSouth Fork

other place in the entis known to have a siiand point concentratigeographic location.

Hockett (2005) dearea as a site containifragments. This is abecause the three clu(DSN) points that pafound directly associing corrals, only numHowever, 66 DSN pCobre Trap located tThis definition, therethese four clustersancient communal p:had decayed or bumtheless, it must be enDSN points recordednumber because maiedly buried and remwhile the SMTC killtan between 20 andtotals also represent iexcavation at four ofcultural deposits inclwith fewer than 20 siresent ancient comuadhere to the conservanalysis that followskill spots maybe dis

Previous attemptscorrals have not beenSMTC area are madeand juniper has notsistent and reliable trlems become especiaas these that may notold. As a result, weing corrals are no oldrepresents the final (Ithe north-central Camerican contact (Ta

AMERICAN ANTIQUITY Hoelcett and Murphy]

Table 1. Mid’

other place in the entire north-central Great Basinis known to have a similar concentration of corralsand point concentrations within a relatively smallgeographic location.

Hockett (2005) defined a kill spot in the SMTCarea as a site containing 20 or more points or pointfragments. This is a very conservative definitionbecause the three clusters of Desert Side-Notched(DSN) points that postdate 600 14C B.P. and werefound directly associated with three separate existing corrals, only numberbetween 12 and 14 points.However, 66 DSN points have been found at theCobre Trap located to the north of the study area.This definition, therefore, would exclude three ofthese four clusters of DSN points as places ofancient communal pronghorn corrals if the woodhad decayed or burned prior to discovery. Nevertheless, it must be emphasized that the number ofDSN points recorded at these corrals is antinimumnumber because many more points are undoubtedly buried and remain unrecorded. Additionally,while the SMTC kill spots defined below all contain between 20 and 256 projectile points, thesetotals also represent minimum values because-testexcavation at four of these has demonstrated buriedcultural deposits including projectile points. Siteswith fewer than 20 surface points may in fact represent ancient communal kills. Nevertheless weadhere to the conservative figure of 20 points in ouranalysis that follows, recognizing that additionalkill spots may be discerned in the future.

Previous attempts to tree ring date the existingcorrals have not been successful. The corrals in theSMTC area are made ofjuniper branches and logs,and juniper has not been shown to produce consistent and reliable tree ring dates. Old wood problems become especially significant in features suchas these that may not be more than a few centuriesold. As a result, we assume that most of the existing corrals are no older than 600—700 years, whichrepresents the final (Eagle Rock) cultural phase ofthe north-central Great Basin prior to Euro-american contact (Table 1). Pre-600 ‘4C B .P. point

concentrations that may represent more ancient,now-degraded corrals contain Gateciff/Humboldt(ca. 5,000—3,500 14C B.PJ, Ellco (ca. 3,500—1,400‘4C B.E), and Eastgate/Rose Spring (ca. 1,400—600‘4C B,P.) projectile points (Table 1).

The SMTC area contains at least 13 juniperbranch corrals and 13 projectile point concentrations (Tables 2 and 3). Hockett (2005) referred toall point concentrations as “kill spots,” and theseclusters of projectile points could indeed representthe actual place where pronghorn were shot. Alternatively, some of these could be places near theactual kill site where point rehafting/repair tookplace. Below we distinguish between the two byproviding details of point breakage patterns. These13 point concentrations are therefore interpreted aseither the place where the animals were killed (killspots) or as places where rehaftinglretooling wasthe primary activity (point clusters).

The majority of the point concentrations in theSMTC area are located within or next to existingcorrals. These sites are not only located on the samefinger ridges as existing corrals, but also on thesame spot on these ridges. This is discussed andillustrated in further detail below.

In addition to the 13 point concentrations, twoother sites (Hourglass Overlook and HourglassAmbush) in the SMTC area contain relatively largenumbers of points, although their geographic position on the landscape and/or debitage analysis allbut preclude them from representing ancient corrals (Table 3). These latter two sites are used as comparisons to the 13 point concentrations that mayrepresent ancient corrals.

In order to further enhance our comparisonsbetween sites, projectile point fragment data arepresented for an additional six sites that containlarge numbers of projectile points but are locatedoutside the SMTC area proper (fable 4). Thesesites include: (1) The Cobre Trap is another existing corral associated with numerous projectilepoints north of the SMTC area; (2) Town Creek isa site in which foreshafts were retrieved from a kill

Hocketi and Murphyl ANTIQUITY OF COMMUNAL PRONGHORN HUNTING

Table I. Middle and Late Archaic Projectile Point Chronology frnm the North-Central Great Basin.

715

Phase Date (“C BP) Point Style (s)Eagle Rock 600— 150 B? Desert Side-Notched; CottonwoodMaggie Creek 1.400 — 600 BP Eastgaie; Rose SpringJames Creek 3,500— 1,400 BP ElkoSouth Fork 5.000 — 3.500 BP Gatecliff; Humboldt

[Vol. 74, No. 4, 2009

in relation to existingwere associated on theattires.the raw material usedpoints recovered fromtrations. We comparedstribution of toolstonedetermine the degreeprior to entering the

centrations in theal Introduction

location, topography,the SMTC area can be-lere we note that theipasses approximatelyred at < 30 m spacingtctenzed by flat valley[ephemeral drainages.ns of low-lying fingerrun north-south iii thearea, and east-west inions. The area is relanear the northern end

st of a steep bluff (Fig-n as the “hourglass”out, or “opens up” to

istriction. It is bounded,lands ofValley Mounrespectively. The low

a near-monocultureremainder is covered

ahjuniper habitat. The•rimary building mate-he existing corral wallsive juniper trees, sugwhen the corrals were

)92) recorded the firstValley) in the SMTC

rvey, we (Hockett andfirst corral and the sectration (Knoll site) 15f the corrals and pointed between 2000 andeys overseen by Hock-zen corrals have beenrc area as well, but no

spot and subsequently unhafted and dropped there.Located directly north of the SMTC area, a SouthFork Phase corral may have been built and successfully used nearby; (3) Ander Wright and Palisade Canyon are bluff sites overlooking majorperennial streams (Marys River and HumboldtRiver, respectively). Both sites represent placeswhere large game was probably ambushed belowthe bluff and foreshafts were retrieved and unhaftedthere. At Ander Wright (South Fork Phase), thegame was probably deer or mountain sheep,although excavations failed to recover faunalremains. At Palisade Canyon (Maggie CreekPhase), bison were brought back to this base campfor processing; and (4) Point Blank Hill (MaggieCreek Phase) and Santa Fe (James Creek Phase)are sites located atop small hills or ridges about 60miles north of the SMTC area. Mountain sheepwere probably the targets, and these animals wereprobably either ambushed or trapped along established migration routes very close to these sites. Nofaunal remains were found at the Santa Fe site, butPoint Blank Hill is a site in which points wereunhafted from retrieved foreshafts, and mountainsheep were brought there for processing (Schroedl1995).

Topographic Placement

Figures 1—3 show that the existing corrals and pointconcentrations are geographically associated withone another. Figure 1 illustrates that the corrals andpoint concentrations cluster in two regions withinthe survey block: a northern region, which consistsof nine corrals and nine point concentrations, anda southern region consisting of four corrals and

four point concentrations. The southern cluster isnot located in any special topographic position suchas a narrow constriction. Nevertheless, this clusterof corrals and point concentrations is located at thejunaure of a flat valley that extends to the southand the beginning of numerous narrow fingerridgesdissected by ephemeral drainages to the north. It istherefore likely that hunters took advantage of thisterrain, as well as consistent pronghorn migrationsacross this spot to build south-facing corrals.Entrances and wings were constructed on the lowerterrain, and the back-ends were concealed on theopposite sides of finger ridges. Hunters probablyhave taken advantage of this place to corral or surround pronghom since the South Fork Phase (ca.3,500—5,000 ‘4C B.R), as the Mizpah Chute, Mizpah Valley, and Spruce Knoll sites (Figure 2) allcontain large concentrations of Catecliff and Humboldt projectile points.

The “Hourglass Constriction” (Figure 1) onlycontains four of the 13 corrals and four of the 13point concentrations. It is also here that the mainsource of artifact-quality chert (Valley Mountainchert) is located (Figure 3). During Protohistorictimes pronghorn were primarily corralled as theymigrated from the north to the south because all ofthe corral openings face northward within the constriction. Based on the presence of hundreds ofHumboldt points at the Clover Valley and ValleyMountain B sites, pronghom have been corralledor surrounded within this constriction since at leastthe South Fork Phase (ci 5,000—3,500 aC BR).

Interestingly, as the terrain opens to the north ofthe topographic constriction, nearly all of the existing corrals face southward, suggesting that prong-

716 AMERICAN ANTIQUITY [Vol. 74, No.4,2009

Table 2, The Corrals or the SMTC Area.

Hackett and Murphy]

Site/FeaLure Type Noteswiseman Trap corral rio points associatedSpruce Well Trap corral no points associatedSprucemont Trap corral no points associatedSpruce Ridge Trap corralikill spot 16 DSN pointsSpruce Pond Trap corral/kilt spots 14 DSN, 6 Elko, 1 EastgateLi2a Jane North Trap corral/kill spots 12 DSN and 23 Elko pointsValley Mountain Trap corrai/kill spots 118 Humboldt and 50 Elko pointsHill Trap corral/kill spot 21 Elko and 1 DSNStorey Trap corral no noints associatedGallegos Trap corral I Elko pointSpruce Knoll Trap corral 1 Gatediff, 1 Humboldt, 2 Elko, 1 Eastgate, I DSNMiapali Trap Complex corrals I DSNPygmy Rabbit Trap corral no points associated

Site/Feature T

Antelope Ridge A ki

Antelope Ridge B or

Spruce Ridge ki

Spruce Pond ki

Sir Spruce 111 pr

Liza Jane North A ki

Liza Jane North B

Valley Mountain A ki

Valley Mountain B

Hill ki

Spruce Knoll or

Mizpah Chute ki

Mizpah Valley ki

Hourglass Ambush ar

Hourglass Overlook to

Table 4.

Site/Feature Type

Cobra A corral!Cobre B corral!

Palisade Canyon butche

Point Blank Hill butche

Santa Fe butche

Ander Wright butche

IC southern cluster isgraphic position suchertheless, this clusterlions is located at theextends to the southnarrow finger ridges

ages to the north. It is30k advantage of thiswnghorn migrations)uth-facing corrals.istructed on the lowerere concealed on thes. Hunters probablylace to corral or sur

outh Fork Phase (ca.Mizpah Chute, Miz1 sites (Figure 2) allif Gateeliff and Hum

ion” (Figure 1) onlyis and fourof the 130 here that the mainrt (Valley MountainDuring Protohistoricrily corralled as they

south because all ofward within the conence of hundreds ofer Valley and Valleyhave been corralled

striction since at least100—3,500 ‘4C B.P.).opens to the north of

nearly all of the existiggesting that prong-

[Vol. 74, No. 4, 2009 AN11QUITY OF COMMUNAL PRONGHORN HUNTING

Table 3. The Projectile Point Concentrations of the SMTC Area.

717

ko,, I Bastgate, I DSN

Hockett and Murphy]

1..

‘4

Site/Feature Type Notes

Antelope Ridge A kill spot 108 points (51 Eastgate); located about 1,0m west of Spruce Well Trapnear end of finger ridge; probable back-end spot of an ancient corral

Antelope Ridge B unhafting 130 points (87 Elko); located lOOm south of Antelope Ridge A site nearend of finger ridge; probable place of unhafting from retrieved foreshaftsshot at nearby corral

Spruce Ridge kill spot 16 DSN points; located at the entrance to the Spruce Ridge Trap; animalsshot as they entered or attempted to back-out of the corral

Spruce Pond kill spot 14 DSN, 6 Elko, 1 Eastgate; located inside Spruce Pond ‘frap; animals shotinside corral

Sir Spruce Ill possible kill spot 29 Gatecliff/Humboldt, 24 Elko, 18 point fragments; located between theSpruce Ridge and Spruce Pond traps; now-degraded corral probably waslocated at or near this spot

Liza Jane North A kill spot 23 EUro, 6 dart point fragments; located at entrance to existing Liza JaneNorth Trap; animals were shot as they entered or attempted to back-out ofthe corral; location has served as a place to corral pronghorn since theJames creek Phase

Liza Jane North B kill spot 12 DSN points; located at or near entrance to Liza Jane North Trap; animalswere shot as they entered or attempted to escape through the entrance

Valley Mountain A kill spot 50 Elko points; located within Valley Mountain Trap. and clustered 50mfrom Valley Mountain B kill spot and 200m from Dover Valley kill spot;location was spot of ancient corral built during the James Creek Phase

Valley Mountain B kill spot 118 Humboldt points; located within Valley Mountain Trap, and clusteredSOm from Valley Mountain A kill spot and 250m from Clover Valley killspot; location of corral built during the South Fork Phase

Hill kill spot 21 Elko, 11 dart point fragments: located directly alongside eastern edge ofSpruce Hill Trap; corrals were built here since at least the James CreekPhase

Spruce Knoll unhafting 60 Humboldt, 49 Elko, 44 point fragments; located about 40Cm south ofSpruce Knoll ‘ftap; Middle Archaic corrals probably built nearby

Mizpah Chute kill spot 136 Gatecliff points; located SOm sooth of Mizpah Trap Complex, 30Cmeast of Pygmy Rabbit Trap, and 200m north of Mizpah Valley kill spot:location of corral built during South Fork Phase

Mizpah Valley kill spot 126 Gacecliff points: located 35Cm southeast of Mizpah Trap Complex andPygmy Rabbit Trap, and 20Cm south of Mizpah Chute kill spot; location ofcorral built during South Fork phase

Hourglass Ambush ambush 25 Elko points; located in foothills directly east of Valley Mountain Trap,and Valley Mountain and Dover Valley kill spots; animals (probably deer)ambushed along probable migration corridor

Hourglass Overlook toolstone reduction 14 Elko points; located directly west of Valley Mountain Trap on lowerslope of the Valley Mountain chert source area; location of the full range oftool production and unhaftiog of retrieved foreshafts shot nearby

TabLe 4. Additional Projectile Point Concentrations from the North-Central Great Basin.

Site/Feature TSpe Notes

Cobre A corral/kill spot 66 DSN points; located within Cobre Trap

Cobre B corral/kill spot 47 Elko points; located within and along edge of existing Cobre Trap;location of James Creek Phase corral

• Palisade Canyon butchery, tool 60 Rose Spring points and point production fragments; located on bluffabove Humboldt River

Point Blank Hill butchery, ambush; 224 Eastgate/Rose Spring points and corral’trap? point fragments; locatedatop small hill: mountain sheep probably ambushed nearby

Santa Fe butchery, ambush 176 Elko points and point fragments; located on small ridge; deer ormountain sheep probably ambushed nearby

Ander Wright butchery, ambush 141 Gatecliff points and point fragments; located on bluff above Marys

River: game probably ambushed nearby

horn were trapped as they migrated from the southto the north. In addition, clusters of South ForkPhase-aged projectile points (Gatecliff and Humboldt) are rarer, found only at the Sir SpniceTll site.The point concentrations here primarily date to theJames Creek Phase and later (ca. post—3.500 ‘4CB.P.).

A closer look at the precise pattcrning of thepoint concentrations in relation to the existing corrals solidifies our proposition that the majority ofthem likely represent places of ancient corral construction. Figure 4 displays a close-up view of theMizpah Chute and Mizpah Valley point concentrations in relation to the Mizpah Trap Complex andthe Pygmy Rabbit Trap. Both of these point concentrations are located below the top of adjacentfinger ridges, which are themselves located toward

the backside of the existing corrals. Given theirposition on the landscape, these projectile pointswere probably shot at pronghorn as the animalsentered the corrals or attemptedto back out of them.Interestingly, both point concentrations containalmost exclusively Gatecliff points.

Figures 5 and 6 display the spatial relationshipsbetween existing corrals and point concentrationsnear the Valley Mountain and the Liza Jane Northtraps within the topographic constriction, Asexpected, the entrance to the Valley Mountain Trap(Figure 5) was built on flat terrain, with the backside of the structure located atop a finger ridge. Boththe Valley Mountain A (James Creek Phase) andValley Mountain B (South Fork Phase) point concentrations are located at the same place on theridge that contains the Valley Mountain trap itself.

718 AMERICAN ANTIQUITY [Vol. 74, No.4,2009

Figure 1. The 15,000-acre sun’ey block that encompasses the SMTC ara Note the two concentrations of corrals/traps(T) and kill spots/point dusters (K), as well as the location of the topographic constriction.

o i 1 N— ,Kaomotsrs A

Hockett and Murphy]

Figure 2. Th

While it is possible thcorrals associated wifaced southward instttogether with the adjgest that pronghomnear the back-ends olPhase.

Figure 6 shows tiPhase) and Liza Janeconcentrations in rel;Trap. In both cases, Iare located at or neaicorral on flat terrain(backside of the cornhorn have been repeatcorrals or attemptedleast the James Creel

One difference bcand Liza Jane Northof the corral at Valleatop a low-lying finof the corral at Lizaatop a much higher 2steepness of the hill abeen less conducivepronghorn toward ththe animals were mcthe entrance. This wocentrations itpresentiEagle Rock phases benear the entrance toatop the knoll.

While it is possible that the ancient, now-degradedcorrals associated with these point concentrationsfaced southward instead of northward, these sites,together with the adjacent Clover Valley site, suggest that pronghorn were repeatedly dispatchednear the back-ends of corrals since the South ForkPhase.

Figure 6 shows the Liza Jane A (Eagle RockPhase) and Liza Jane B (James Creek Phase) pointconcentrations in relation to the Liza Jane NorthTrap. In both cases, the point concentrations hereare located at or near the entrance to the existingcorral on flat terrain rather than on top of the ridge(backside of the corral). This suggests that prong-horn have been repeatedly shot here as they enteredcorrals or attempted to back out of them since atleast the James Creek Phase.

One difference between the Valley Mountainand Liza Jane North localities is that the backsideof the corral at Valley Mountain was constructedatop a low-lying finger ridge, while the backsideof the corral at Liza Jane North was constructedatop a much higher and steeper hill or knoll. Thesteepness of the hill at Liza Jane North might havebeen less conducive or less efficient at funnelingpronghorn toward the back of the corral such thatthe animals were more likely to congregate nearthe entrance. This would account for the point concentrations representing both the James Creek andEagle Rock phases being located on the valley floornear the entrance to the existing corra] rather thanatop the knoll.

Figures 7 and 8 display two point concentrationsassociated with the existing Spruce Ridge Trap andSpruce Pond Trap located north of the constriction.The DSN points associated with the Spruce RidgeTrap (Figure 7) are all located near its entrance, suggesting once again that pronghom were shot asthey entered or as they attempted to back out of thestructure. Similar to Liza Jane North. the backsideof this corral was built atop a fairly steep hill orknoll. All of theprojectile points, regardless of age.associated with the Spruce Pond Trap were foundscattered within the existing corral (Figure 8), suggesting that pronghom were shot inside and nearthe center of the corral. Similar to Valley Mountain, the backside of this corral was constructed ona low-lying finger ridge.

Projectile Point Breakage Patterns

Topographic context strongly suggests that manyof the point concentrations in the SMTC area represent ancient corrals. The next logical question toask is: Do most of them contain relatively high percentages of midsections and tips in relation tobases? The answer is “yes:’ As noted previously.sites located some distance from the actual kill spotwhere projectile points were unhafted fromretrieved forehsafts should contain relatively largenumbers of basal point fragments. These types ofsites previously analyzed from the north-centralGreat Basin include Town Creek (Petersen andSteams 1992), Point Blank Hill (Schroedl 1995),and Ander Wright (Zeanah and Elston 1997). The

Hocketi and Murphy] ANTIQUITY OF COMMUNAL PRONGHORN HUNTING 719

1)Gallegos Trap

Spruce Trap

• Mizpah Trap ComplexSpruru Knoll

Mkpah ChuLoC.Pygmy Rabbit Trap htipah vamkv

N

A750 1,500 3000

Figure 2. The southern cluster of corrals and point concentrations (dots) within the SMTC area

[Vol. 74, No. 4, 2009

rations of corrals/traps

omTals. Given theirse projectile pointsiorn as the animalsIto back out of them.centrations containoints.spatial relationshipsoint concentrationsthe Liza Jane Northc constriction. AstHey Mountain Traprain, with the back-a finger ridge. BothCreek Phase) and

k Phase) point con-same place on the4ountain trap itself.

projectile point fragment assemblages from allthree of these sites are comprised of approximately50—60 percent basal fragments (Table 5). This percentage range can serve as a general guideline forinterpreting similar localities in the SMTC area. Itshould also be noted, however, that sites locatedaway from kills may also contain relatively largenumbers of point tips if large game carcasses werealso brought back to the campsite for butcheringbecause tips can “ride” inside body cavities. Thisappears to be the case at the Palisade Canyon sitelocated on a bluff overlooking the Humboldt River.Bison were brought to this site for processing during the Maggie Creek Phase (Hockett 2007a), andthe largest percentage (nearly 50 percent) of projectile point fragments recovered were in the formof tips (Table 5). Point manufacture also occurred

at this she, and nearly 50,000 pieces of debitagewere recovered, so some tips there may simply represent points broken during the manufacturingprocess.

During the summer of 2007, we also completedtest excavations at the Mizpah Chute, AntelopeRidge A, Antelope Ridge B, and Hourglass Overlook sites in order to augment the projectile pointbreakage pattern analysis. Our excavations not onlyrecovered many more projectile points than wereoriginally recorded on the surface of these sites, butalso revealed broken and burned artiodactyl bonesat the Mizpah Chute, Antelope Ridge A, and theHourglass Overlook sites. In addition, charcoal wasrecovered in sufficient quantities at Antelope RidgeA, Antelope Ridge B, and Hourglass Overlook,which provide corresponding radiocarbon dates.

Figure 4. Aerial view oftthe Mizpah ftap Compl

Debitage was also reThese latter data are dsections that follow.

The values displayfour of the 15 pointRidge B (Figure 9), 5and Hourglass Overkbasal fragments thatCreek, Ander Wright.This is particularly thtwhich displays valuesas Town Creek. The rare rather unique at Sidisplays the highest pof any of the South ISMTC area. At theHumboldt (N = 60, hiand Elko (N = 49, ba

AMERICAN ANTIQUITY Hockett and MLarphyj720

0Wae,nan Trap

7

AtelopeRidgeA SpnacewellTrap.

Antelope Ridge BSpnicetnont Trap

VM CHERT

Spruce RIdge Trap

SirSpnacefllSpruce Pond Trap

Liza Jane North Trap

Hourglass overlookbuah

Valley Mountain Trap Cloverv5u5,.

VMCHERT

Hill Trap

tStorey Trap

No 750 500 3.000

Melers

Figure 3. The northern cluster of corrals and point concentrations (dots) within the SMTC area.Valley Mountain chert source (VM chert) are also shown.

[Vol. 74, No. 4, 2009

Two locations of the

Debitage was also recovered from all four sites.These latter data are discussed in more detail in thesections that follow.

The values displayed in Table 5 show that onlyfour of the 15 point concentrations—AntelopeRidge B (Figure 9), Sir Spruce m, Spruce Knoll,and Hourglass Overlook—exhibit percentages ofbasal fragments that approach those of the TownCreek, Ander Wright, and Point Blank Hill sites.This is particularly the case for Antelope Ridge B,which displays values almost identical to sites suchas Town Creek. The point fragmentation patternsare rather unique at Sir Spruce ifi, however, whichdisplays the highest percentage of complete pointsof any of the South Fork Phase-aged sites in theSMTC area. At the Spruce Knoll site, both theHumboldt (N = 60, basal fragments = 77 percent)and Elko (N = 49, basal fragments = 55 percent)

points consist of relatively high percentages ofbasalfragments. Although there were 44 tips and mid-sections recovered from the Spruce Knoll site thatcould not be definitively classified as either Humboldt or Elko, 60 percent of all the points from thesite represent basal fragments. This site is probably not the spot of an actual kill, but because it issandwiched between the intensive corralling activity surrounding the Mizpah Trap Complex to thesouth and existing corrals to the north (Figure 2),it is likely related to earlier corralling events.

The Hourglass Overlook site is located near thebase of the Valley Mountain chert source in thefoothills of the Valley Mountain range. l’his siteoffers a broad lookout across the corrals and killspots located on the valley bottom below within theconstricted topographic zone. Its location alonewould suggest that it was not the site of an ancient

Hockett and Murphy] ANTIQUITY OP COMMUNAL PRONGFfORN HUNTING 721[vol. 74, No. 4, 2009

tea. Two locations or the

)0 pieces of debitagehere may simply rep-

the manufacturing

7, we also completedah Chute, Antelopeand Hourglass Over-it the projectile pointexcavations not only

tile points than wereface of these sitesbutied artiodactyl bonespe Ridge A, and theddition, charcoal wasles atAntelopeRidgelourglass Overlook,

radiocarbon dates.

Figure 4. Aerial view of the Mizpnh Chute (northern) and Mizpah Valley (southern) point concentrations in relation tothe Mizpah &ap Complex and the Pygmy Rabbit frap,

corralling event, and the large percentage of basalprojectile point fragments corroborates this interpretation.

The remainder of the point concentrationsrecorded in the SMTC area exhibit relatively largepercentages of complete points and point midsections and tips, a pattern more consistent with killspot localities (see also Figure 10). Theft topographic context in relation to existing corrals solidifies this interpretation.

The DSN points associated with the corrals tendto exhibit far greater numbers of complete pointscompared to the other point styles (Table 5 andFigure 10). This simply may be due to their smallsize, which led to these points breaking less frequently on impact. In addition, it would have beenmore difficult to locate broken fragments of DSNpoints on the surfaces of these sites during recor

dation. Full-scale excavation at sites such as SpruceRidge, Spruce Pond, and Liza Jane North mayreveal many more broken DSN point fragments. Inany case, between 60—80 percent of the DSN pointsrecorded within existing corrals consisted of complete points and point rnidsections.

The only potential Maggie Creek Phase kill spotfrom the SMTC area is the Antelope Ridge A site.This site overwhelmingly consits of Eastgatepoints, but Elko points (which were exclusivelyfound at the nearby Antelope Ridge B site) werealso present. More than 60 percent of the pointsfrom Antelope Ridge A consisted of completepoints and point midsections and tips. The site islocated atop and near the end of a finger ridge thatwould have served as aprime trapping location. Thenow-degraded corral(s) that were probably constructed here would have faced to the north, with

AMERICAN ANTIQUITY (Vol. 74, No. 4,2009722

FIgureS. Aerial view of the Valley Mountain’&ap and associated point concentrations. The location of the original CloverValley site is shown near the eastern edge of the existing corraL The crosses (Humboldt points) and triangles (Elko points)show internal clustering at the back-end of the existing corral. The lone DSN point (D) is also shown.

Hackett and Murphy)

Figure 6. Aerial view of thepoints; I) = DSN points; R

theft entrances locatedthe ridgetop, with the ton top of the ridge wheered. Both the AnteIoplikely associated with atlocation, with AntelopiMaggie Creek Phase-Iapalimpsest located nearAntelope Ridge B reprether up the ridge where Iretrieved foreshafts outsing the James Creek Ph:

The Valley Mountairsites all appear to be Janwhile the Mizpah Chuteley, Clover Valley, and VSouth Fork Phase kill sassemblages from thesites exhibit betweenpoints and point midsec

their entrances located on the valley floor belowthe ridgetop, with the back of the corrals locatedon top of the ridge where the points were discovered. Both the Antelope Ridge A and B sites arelikely associated with ancient corral building at thislocation, with Antelope Ridge A representing aMaggie Creek Phase-James Creek Phase kill spotpaliinpsest located near the backside of a corral, andAntelope Ridge B representing a place lOOm further up the ridge where points were unhafted fromretrieved foreshafts outside of the corral walls during the James Creek Phase.

The Valley Mountain A, Hill, and Liza Jane Asites all appear to be James Creek Phase kill spots,while the Mizpah Chute (Figure 11), Mizpah Valley, Clover Valley, and Valley Mountain B sites areSouth Fork Phase kill spots. The projectile pointassemblages from the three James Creek Phasesites exhibit between 62—74 percent completepoints and point midsections and tips. The points

from the four South Fork Phase sites exhibitbetween 60—84 percent complete points and pointmidsections and tips. As noted above, all seven ofthese Middle Archaic sites are directly associatedwith existing Protohistoric-aged corrals, suggesting that hunters intermittently utilized the samevalley-and-ridge topography for corralling prong-horn at each of these locations for many millennia.The topographic position of the points belowridgetops on flat terrain at Mizpah Chute and Mizpah Valley suggest that these sites represent placeswhere pronghorn were shot as they entered orattempted to back out of ancient, now-degradedcorrals.

Additional Data Collectedfrom the Kill Spotsand Point Clusters—Radiocarbon Dating,Faunal Remains, and Debitage Analysis

In 2007, we excavated a total of 12 l-x-l-m unitswithin the Mizpah Chute kill spot, seven units each

Hackett and Murphy] ANTIQUITY OF COMMUNAL PRONGHORN HUNTING 723

Figure 6. Aerial view of the Lit. Jane North Ti’ap and associated point concentrations located near its entrance. E = Elkopoints; 1) = DSN points; R = Rose SpringlEastgntc points; P = undetermined point fragment.

We!. 74, No. 4, 2009

ion of the original Cloverid triangtes (Elite points)how,i.

it sites such as Spruceiza Jane North mayN point fragments. Inent of the DSNpoints315 consisted of corndons.Creek Phase kill spotritelope Ridge A site,;onsists of Eastgatech were exclusively

Ridge B site) wereercent of the pointsisisted of completeand tips. The site isof a finger ridge thatrapping location. Thewere probably cond to the north, with

o 2 50 100Melor, N

within the Antelope Ridge A kill spot and B pointcluster, and 10 units within the Hourglass Overlookpoint cluster (Table 6). At Mizpah Chute, in addition to buried Gatecliffpoints, we retrieved 62 largemammal shaft fragments, 12(19 percent) of whichwere burned. Attempts at radiocarbon dating thefaunal remains failed to retrieve enough collagenfor an accurate date. This is not surprising giventhe fact that any organic remains there have beensubjected to erosion in an open-air setting for atleast 4,000 to 6,000 years. A total of 826 pieces ofdebitage was recovered from the Mizpah Chuteexcavations, averaging 43 per square meter. Of thediagnostic flakes, 81 percent were late stage bifacethinning and finishing flakes. iThis includes threenotching flakes. These data suggest that late stagebiface manufacturing including projectile pointproduction occurred at the site. Little use was madeof the local Valley Mountain chert source area

(approximately 2 percent). Instead, other cherts (44percent) and the local argillite (25 percent) weremost common. Obsidian (12 percent), mainly fromthe Browns Bench Obsidian Source Area, was morecommon than at any of the other excavated sites.Overall, debitage analysis suggests that effort wasexpended on point repair or manufacturing prior tothe successful corralling event(s), just followingthe event(s), or both.

The Antelope Ridge A kill spot was sign ficantlyeroded and deflated, yet projectile point fragmentswere retrieved just under the gravelly surface of thesite, as well as two burned artiodactyl bone fragments and 235 pieces of debitage. Debitage density was 34 pieces per square meter, Like MizpahChute, the focus was primarily on late-stage bifaceproduction, with 65 percent of the diagnostic debitage representing late-stage biface thinning andfinishing flakes. However, early-stage biface reduc

Figure 8. Aerial view oftpoints; I) = DSN points; I

tion was relatively corflake was present. Nnant toolstone (63 penChert was common (2Bench) accounted for

At nearby Anteloptcharcoal radiocart2,900—3,400 4C B.P.,Phase age. No faunalthis site. Excavationaverage density of 3Mountain chert domisisting of 84 percent oup another 14 percentand obsidian were repercent), early bifacelate biface reductionof the debitage. The s

724 AMERICAN ANTIQUITY [Vol. 74, No. 4, 2009 Hocketi and Murphy3

Figure 7. Aerial view of the Spruce Ridge lNp mid associated DSN points located near its entrance.

tion was relatively common and only one notchingflake was present. Nonlocal chert was the dominant toolstone (63 percent), whileValley MountainChert was common (29 percent). Obsidian (BrownsBench) accounted for only one percent of the total.

At nearby Antelope Ridge B, one unit containedcharcoal radiocarbon dated between ca.2,900—3,400 14C BR, confirming its James CreekPhase age. No faunal remains were recovered fromthis site. Excavation yielded 248 flakes, with anaverage density of 35 per square meter. ValleyMountain chert dominated the assemblage, consisting of 84 percent of the total. Other cherts madeup another 14 percent, while a few flakes of basaltand obsidian were recovered. Core reduction (18percent), early biface reduction (33 percent) andlate biface reduction (17 percent) made up the bulkof the debitage. The scarcity of finishing flakes (8

percent) and lack of notching flakes suggest thatprojectile point repair/muufacture was either ofminor importance or lacking altogether.

Finally, at the Hourglass Overlook site werecovered charcoal and burned and unbumed large mammal shaft fragments. Charcoal dating between ca.1,900—2,360 ‘4C B.P. confirms its James CreekPhase age. A total of 31 artiodactyl bone fragmentswas recovered, 10 (32 percent) of which wereburned. Importantly, debitage analysis confirmsthat this location is unlikely to have been a kill spot.Rather, a full range of toolstone reduction and toolmanufacture occurred. The quantity and density ofdebitage at Hourglass Overlook dwarfs the amountsfound at the other sites. Waste flakes number 8,651,with an average density of 1,442 pieces per squaremeter, or 33 times as much as at Mizpah Chute.Given that the site is located only a few hundred

ANTIQUITY OF COMMUNAL PRONGHORN HUNTINGHocketi and Murphyj 725

Figure 8. Aerial view of the Spruce Pond &ap and associated projectile point scatter within the corral walls. E = Elkopoints; P = DSN points; R = Rose Spring/Eastgate points.

[Vol. 74, No. 4, 2009

ear its entrance.

istead, other cherts (44lite (25 percent) werepercent), mainly from

Source Area, was moreother excavated sites..iggests that effort wasnianufacturing prior to‘ent(s), just following

1 spot was significantlyjectile point fragmentsgravelly surface of theartiodactyl bone fragbitage. Debitage dene meter. Like Mizpah.ly on late-stage bifaceof the diagnostic deb

biface thinning andrly-stage biface reduc

o 25 50 100Meters

meters from a primary outcropping ofValley Mountain chert, it is no surprise that this material accountsfor 99 percent of the debitage. Emphasis was onbiface reduction although core/flake blank production and final finishing are represented. Whilethe 320 finishing flakes make up only 9 percent ofthe debitage at Hourglass Overlook they are stillmore numerous here compared to the other sites.The presence of finishing flakes and three notching flakes indicates projectile point repair or manufacture . This site was not a kill spot, but probablyplayed a role in the communal hunt as a camp where

gearing-up using local chert took place immediately prior to the hunt and consumption of gameafterwards during the James Creek Phase.

In sum, projectile point breakage patterns areconnensurate with the topographic context analysis, and these data suggest that the majority of thepoint concentrations recorded in the SMTC area arekill spots. These data strongly suggest that the killspots represent the remains of degraded corrals thatwere constnicted as early as 3,500 ‘4C B.P. andperhaps as early as 5,000 ‘4C WE Repair of brokenprojectile points occurred at some of these kill

726 AMERICAN ANTIQUITY

Table 5. Projectile Point Breakage Patterns from Various Sites Located in the SMTC Area and Elsewherein the North-Central Great Basin.

[Vol. 74, No.4,2009 Hockett and Murphy]

Site Complete Distal’ Midsection Tip Lateral Totals

Eagle Rock PhaseSpruce Ridge 4 (25) 5 (.31) 6 (.38) 1 (06) 0 (0) 16Spruce Pond 8 (.57) 3 (.21) 3 (.21) 0 (0) 0 (0) 14Liza Jane North B 3 (.25) 5 (.42) 4 (.33) 0 (0) 0 (0) 12CobreTrapA 32(48) 9 (.14) 16 (.24) 8 (.12) I (.02) 66TOTALS 47 (.44) 22 (.20) 29 (.27) 9 (.08) 1 (.01) 108

Maggie Creek PhaseAntelope Ridge A 10(09) 40 (.37) 35 (.32) 22(20) 1 (.01) 108Palisade Canyon’ 4 (.07) 4 (.07) 21 (.35) 29 (.48) 2 (.03) 60Point Blank Hill’ 50 (.22) 124 (.55) 18 (.08) 6 (.03) 26 (.12) 224TOTALS 64 (.16) 168 (.43) 74 (.19) 57 (.15) 29 (.07) 392

James Creek PhaseAntelope Ridge B 14 (.11) 70(54) 28 (.22) 15 (.12) 3 (.02) 130Valley Mountain A 7 (.14) 16 (.32) 12 (.24) 12 (.24) 3 (.06) 50Hill 3 (.14) 5 (.24) 7 (.33) 4 (.19) 2 (.10) 21Liza Jane North A 5 (.22) 6 (.26) 10(43) 2 (.09) 0 (0) 23Spruce Knoll A 9 (.18) 27 (.55) 9 (.185 4 (.08) 0 (0) 49Hourglass Overlook 4 (.29) 8 (.57) I (.07) 1 (.07) I (.07) 14Hourglass Ambush 10 (.40) 9 (.36) 2 (.08) 4(16) 0(0) 25Cobre Trap B 3 (.06) 12 (.26) 26 (.55) 6 (.13) 0(0) 47Santa Fe2 Ii (.06) 51 (.29) 32 (.18) 61 (.35) 21 (.12) 176TOTALS 66 (.12) 204 (.38) 127 (.24) 109 (.20) 30 (.06) 535

South Fork PhaseMizpah Chute 14 (.10) 53 (.39) 42 (.31) 26 (.19) 1 (.01) 136Mizpah Valley 1 (.01) 20 (.16) 86 (.68) 19 (.15) 0(0) 126Clover Valley 18 (.07) 90 (.35) 89 (.35) 59 (.23) 0(0) 256Valley Mountain B 5 (.04) 37 (.31) 55 (.47) 20 (.17) 1 (.Ot) 118Spruce Knoll B 10 (.17) 46 (.77) 4 (.06) 0(0) 0(0) 60SIr Spruce iW 19 (.27) 33 (.46) 12 (.17) 6 (.08) I (.01) 7Town Creek 21 (.14) 90 (.58) 30 (.19) 13 (.08) 0 (0) 154AnderWright’ 12 (.09) 65 (.46) 15 (.11) 30 (.21) 19 (.13) 141TOTAlS 100 (.09) 434 (.41) 333 (.31) 173 (.16) 22 (.02) 1,062

Figure 9. Projectile pointssites. Top row: Elko SeriesBottom row: Elko Series g

spots, but this activitymal. Portions of cookediscarded at several of

Raw Material Use anthe SMTC Area for C

Now that we have e

‘“Distal” ends are defined here as projectile point bases, in contrast to point “tips?’2Additional kill locations located in the north-central Great Basin outside of the study area-‘Combines both South Fork (Humboldt and Gateclifi) and James Creek (Etko Series) phase projectile points because themidsections and tips from this site could not be separated into individual phases

0.9

0.8

0.2

0.1

0

Eagle Roc

Figure 10. Composite pertSMTC point concentratior

[Vol. 74, No. 4, 2009

a and Elsewhere

Hockett and Murphy] ANTIQUITY OF COMMUNAL PRONGHORN HUNTING

ral Totals

16I) 14I) 122) 661) 108

727

1) 1083) 60.2) 2247) 392

2) 1305) 50

21) 23

497) 14

2547

2) 1766) 535

Figure 9. Projectile points recovered from the Sir Spruce m (top and middle rows) and Antelope Ridge B (bottom row)sites. Top row: Elko Series points from Sir Spruce HI. Middle row: Humboldt and Gatecliff points from Sir Spruce ilLBottom row: Elko Series points from Antelope Ridge B.

[) 136126

I 256118

p 6071154

3) 1412) 1.062

spots, but this activity appears to have been minimal. Portions of cooked pronghorn carcasses werediscarded at several of the kill spots.

Raw Material Use and Human Movements tothe SMTC Area for Communal Hunting

Now that we have established that communal

pronghom hunting likely occurred in the SMTCarea beginning in the South Fork Phase, we canaddress questions related to the timing of gearingup prior to these communal kills through the analysis of the raw material used to manufacture projectile points. Chert was used more frequently thanany other raw material stone to manufacture the

ile points because the——--4L -_____ —_____

%%

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

_____________

. —-

took place immedinsumption of gameDreek Phase.eakage patterns areaphic context analyI the majority of thentheSMTCareaaresuggest that the kill

legraded corrals that500‘4C B P. and perR Repair of brokensome of these kill

-V

—4--- Distal

0

complete +

Eagle Rock Maggie Creek James Creek South Fork

Figure 10. Composite percentages of projectile point distal ends (bases) versus complete points and point tips from theSMTC point concentration sites.

projectile points found at the SMTC kill spots during the South Fork, James Creek, and Maggie Creekphases (Table 7). The Valley Mountain chert sourceis the only raw material with bedrock outcroppingsavailable within the SMTC area itself. Fortunately,this chert is visually distinctive from other knownchert sources. In general, Valley Mountain chert iseither white or reddish in colot Oranges, yellows.and occasionally brown and green also occur. Thewhite material tends to range from poor to fair quality. It is opaque with a dull luster that can be entirelywhite, mottled, or contain black inclusions. Thereds and other colors are semi-translucent and typ

ically contain white or occasionally white and blackinclusions that give the material a speckled or mottled appearance. Even the pieces that are a solidcolor typically will have an occasional white inclusion.

The second-most commonly utilized raw material stone was obsidian. Obsidian was the prefenedmaterial, howeverjust prior to Euro-american contact during the Eagle Rock Phase. The vast majority of glass points were manufactured of BrownsBench Obsidian (BBO), a visually distinctive glassthat is typically black in color and completelyopaque in appearance (although mahogany or red

728 AMERICAN ANTIQUITY [Vol. 74, No. 4, 2009

Figure 11. Projectile points recovered from the Mizpah Chute kill spot Tbp row: complete Gatecliff points. Middle row:point tips. Bottom row: various Gatecliff points manufactured from local argilhite.

Hockett and Murphy

LFigure 12. Regional viewtoolstone source areas. WIcies for use of different trn

with black inclusions i:the BBO source area iscorner of Nevada, abotthe SMTC area (Figurt

Several source arebasalt, argillite, and silta 60 km radius surrouture 12). The basalt somately 30—60km southargillite source (fine-gr:sometimes banded) is Ithe study area. Siltstonorth. A small percentamanufactured of a tannwe haveyet to locate anartifact quality cobbles:SMTC area.

Table 6. Faunah Remains and Radiocarbon Dates Obtained from Four of the SMTC Area Point Concentrations and TwoHearths Located at the Existing Cobre Trap.

Site: Artiodactyl Remains (N): Radiocarbon Dates (‘4C Bfl:Antelope Ridge A 2 2,150±40 (Beta-235958)Antelope Ridge B 0 2,910 ± 40 (Beta-235959)

3.4 10 ± 40 (beta-235960)Mizpah Chute 62Hourglass Overlook 31 1.890 ± 40 (Beta-235957)

2,360 ± 40 (Beta-235955)2,360 ± 40 (Beta-235956)

Cobre Trap 35 120 ± 40 (Beta-236870)690 ± 40 (Beca-236868)

des for use of different toolstone sources through time.

with black inclusions is also present). The heart ofthe BBO source area is located in the northeasterncorner of Nevada, about 150—200 km northeast ofthe SMTC area (Figure 12).

Several source areas of artifact quality chert,basalt, argillite, and siltstone are also located withina 60 km radius surrounding the SMTC area (Figure 12). The basalt sources are located approximately 30—60 km southeast of the SMTC area. Theargillite source (fine-grained, grayish-blue in color;sometimes banded) is located about 50 km east ofthe study area. Siltstone is available 50—60 kmnorth. A small percentage of projectile points weremanufactured of a tannish colored rhyolite. Whilewe have yet to locate a rhyolite source outcropping,artifact quality cobbles are scattered throughout theSMTC area.

The Valley Mountain chert source and the othersources ofbasalt, argillite, and rhyolite can be considered local raw materials because they are alllocated less than a day