University of Pittsburgh- Of the Commonwealth System of Higher Education

Myths about Hunter-Gatherers

Author(s): Carol R. Ember
Source: Ethnology, Vol. 17, No. 4 (Oct., 1978), pp. 439-448
Published by: University of Pittsburgh- Of the Commonwealth System of Higher Education
Stable URL: http://www.jstor.org/stable/3773193 .
Accessed: 22/06/2014 15:00

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

University of Pittsburgh- Of the Commonwealth System of Higher Education is collaborating with JSTOR to
digitize, preserve and extend access to Ethnology.

http://www.jstor.org

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 MythsAboutHunter-Gatherers

CarolR. Ember
HunterCollegeof the City
of NewYork
University

With the re-emergence of evolutionist theorizingin the last few decades,

anthropologists appear to beinterested nowingeneraliting aboutthefeatures of
societiesat differentlevelsof complexity. Particular interestin the hunter-
gathererwayof life appearsto be associated withthe beliefthatthe typical
characteristicsof recenthunter-gatherers weretypicalalso in the Paleolithic.
AlthoughI takeissuewiththisbeliefbelow?it is particularly disturbing that
statements abouttypicalhunter-gatherers aremadeandapparently accepted in
the absenceof systematic surveys of hunter-gatherers to assesswhatcharacter-
isticsare actuallytypical.(AlthoughLee's[I968] surveyis an apparent ex-
ception,thereareseriousproblems withthesampling procedure heemployed, as
I shalldiscussbelow.)Systematic surveys arerelatively easyto make,nowthat
we havea numberof pre-coded databanks(e.g., the Ethnogrogpluic Atlas).
Althoughthesedatabanksarenotrandom samples of theuniverse of described
societies,theyarelargecollections of datawhicharenotlikelyto be severely
biased,becausetheyhavenot beenconstructed witha particular hypothesis in
mind.Manyanthropologists haveanunfortunate habitof tendingto generalize
fromtheirfieldwork experience to the restof the world,or fromthe ethno-
graphlesthey have on theirown bookshelves. Suchsamples,however,are
notoriously biasedby interest,andmayleadto faultygeneralizations.
The mainpurposeof this paperis to show,on the basisof a worldwide
survey,thatthreecurrent conceptions ofwhathunter-gatherers aretypically like
appearto be erroneous. Theseconceptions arethathunter-gatherers aretypi-
callybilocal,thatgathering is themostimportant subsistence activityin hunter-
gatherer economies (andtherelatedideathatwomencontribute morethanmen
to the economy), andthathunter-gatherers aretypically peaceful.
Overthe yearsthe viewof thetypicalresidence patternof hunter-gatherers
appearsto havechanged.The earlierview was that hunter-gatherers were
typicallypatrilocal (Radcliffe-Brown I930; Service I962, I966; Williams I968).
Butnowhunter-gatherers seemto beviewedastypically bilocal.Thischangein
viewmayhaveoccurred becausea numberof fieldworkers (DamasI969; Hiatt
I968; Leacock I969; Meggitt I962) havechallenged theearlierview,orbecause
descriptionsof the !KungBushmen andtheMbuti(bothof whichhavebilocal
residence)havebeenwidelycited,or becausesomeanthropologists (Eggan
1968; Lee I976; Lee andDeVoreI968) haveexplicitly suggested thathunter-
gatherers aretypically bilocal.

439

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
E \E E _l x14;;^,, _= _ _, k ftv {s y ae f k _

440 ETHNOLOGY

TABLE I
ResidenceAmongHunter-Gatherers
- ' 42 'i - == ( ) ( 2) ( 3) -i A
Residence Total Hunter- Omitting Equestrians Omitting Equestrians
Gatherer Sample and High Fishers

Patrilocal 112 (62%) 95 (64%) 51 (56%)

(P or V)

Bilocal 28 (16%) 24 116%) 2Q (22%)

(B, C, D)

Matrilocal 29 (16%) 19 (13%) 18 (20%)

(M or U)

Avunculocal 6 ( 3%) 6 ( 4%) 1 ( 1%)

(A)

NeolQcal 4 ( 2%) 4 ( 3%) 1 ( 1%)

(N)
TOTALS 179 148 91

However,the datain the Ethnogrogphic Atlasdo not-supportthe currentview

of residenceamonghunter-gatherers. ColumnI of Table I showsthe residence
patterns(from column I6 of the AtZas)of all those societiesin the Atlas
summary(MurdockI967) that dependalmostentirelyon gathering hunting,
andfishing(a zeroin columnI0 [herding]andin columnII|agriculture]of the
Atlas). By far,the predominant formof residenceamongthissampleof hunter-
gatherersis patrilocality(62 per cent)*Bilocalitysharesa distantsecondplace
Wlth matRllOCallty (I6 per cent each).
To thosewho mightobjectthattheAtlassampleof hunter-gatherers includes
somesocietiesthat mightnot be "typical"of the past(becausetheyhad horses
or becausethey had an unusuallylarge dependenceon fishing and aquatic
resources),I retabulatedthe frequencydistributionfor residenceomittingthe
equestrianhunters(as tndicatedby an E in column39 of theAtlas see column
2 of Table I) and alsoomittingthe casesthat had approximately 50 percent or
moredependenceon marineresources(5 or morein column9 of theAtlasee
column3 of Table I). But evenwith theseomissions,the resultsdo not change
substantially.Patrilocalityis still much more typicalthan bilocalityamong
hunter-gatherers .
Along with changing attitudesabout the "typical"residencepatternof
hunter-gatherersthere has also been an apparentshift towardthinkingof
hunter-gatherers not so muchas huntersbut rathermoreas gatherers.Earlier
writersseemedto emphasizethe importance of hunting;perhapsbecauseit was
viewedasgivingriseat leastpartlyto patrilocality (StewardI955; Servicex962).
But perhapsbecausegatheringis moreimportantamongthe t KungandMbuti,
and perhapsbecauseLee's (I968) surveysuggestedit, gatheringseemsto be
viewednow as the mostimportantsubsistenceactivityamonghunter-gatherers.
A relatedidea (becausewomengenerallydo the gathering)is that women
typicallycontributesubstantiallymore caloriesthan men do to subsistence
(DeVore and KonnerI974: I20-I22) .
The resultsof the presentsurveydo not supportthe view that gatheringis

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 MYTHSABOUTHUNTER-GATHERERS
44I

generallythe most importantor even the more importantsubsistenceactivity,

nordoesthe surveysupportthe relatedviewthatwomencontributesubstantially
more to subsistencethan men. The Atlas has informationon the caloric
importanceof gathering,hunting,and fishing(columns7-9, respectively).As
Table 2 shows, most societiesin the sample (77 per cent) have gathering
contributingless than half the calories.(These proportionsremainessentially
unchangedif equestrianhuntersare omitted see the secondcolumnof the
table.) And, as Table 3 shows,the somewhatweakerassertionthatgatheringis
moreimportantthan anyother activity(i.e., has a highernumberthan either
huntingor fishing)is alsonot supported.If anything,fishingtypicallyseemsto
be the moreimportantactivity.(This patternstill obtainsif equestrianhunters
are omitted.)Finally,whenAtlvs dataareusedto calculatea sexualdivisionof
laborscore(followingthe procedureemployedin an earlierstudy[C. R. Ember
I975: 202]),2 it is clearlymen,not women,thattypicallycontributesubstantially
moreto primarysubsistence(see Table4). Men predominatein subsistencein
83 per cent of the samplecases,while in only 8 per cent do womencontribute
more than the men.3 (Men still typicallypredominatewhen equestriansand
high hshersare omitted see columns2 and 3 of Table 4).
One puzzlingissue remains.Why do the presentresultsdiffer from Lee's
(I968) survey?His surveysuggestedthatgatheringwastypicallymoreimportant
than anyotheractivity;the presentsurveysuggeststhatfishingis. I suggestthat
the discrepancymay lie in his samplingprocedureand his alterationof the
classification of "gathering."AlthoughLee also used the Ethnographic Atlas,
he substantiallyreducedthe proportionof North Americancases(but not the
proportionsof casesfromother areasof the world). Since80 per cent of the
hunter-gatherers in the Atlas are in North America,and since the North
Americancasesgenerallyhavehuntingandfishingmoreimportantthangather-
ing, Lee's reductionof the proportionof North Americancases inflatesthe
apparentimportanceof gathering.Lee alsoreclassifiedshellfishingfrom"fish-
ing" to "gatherin," which also inflatesthe relativesubsistenceimportanceof
gathering(versusfishing).Thus,the way he selectedhis sampleandthe way he
classifiesshellfishingmake for a distortedview of hunter-gatherers, in my
opinion.If we wish to assesswhatis typicalof hunter-gatherers,I do notsee any

TABLE 2
Importanceof GatheringAmong Hunter-Gatherers
(1) (2)
Importance of Gathering Total Hunter-Gatherer Sample Omitting Equestrians

Contributes more than half 18 (10%) 15 (10%)

the calories (6 or more)

Contributes approximately 24 (13%) 24 (16%)

half the calories (5)

Contributes less than half 138 (77%) 109 (74%)

the calories (O - 4)

TOTALS 180 148

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
_ lW1911
rl- v*- -;V=fv ]XAN--\ff .1 I's0v_,
D -- \7 vX z srEbX 4)
VC _ _- * =L-203-Sif2KL t@k-- A=T J t \ L An; \\ \-I#xC\E;s WE :J0:Xa-= 'Vx\sx;i(wsirds

442 ETHNOLOGY
TABLE3
of Gathering,
TheRelativeImportance Hunting,andFishing
AmongHunter-Gatherers

Relative Importance of (1 ) (2)

Subsistence Activities Total Hunter-Gatherer Sample Qmitting Equestrians

Gathering more important 54 (3Q%) 51 (34t)

than hunting or fishing

Hunting more important 45 (25t) 21 (14t)

than ga ther i ng or f i sh; ng

tishing more tmportant 69 (38t) 64 (43t)

than gather i rig or hunt i ng

Gathering and hunting 4 ( 2t) 4 ( 3t)

co-dom I nan t

Gather i ng and f i sh Jng 4 ( 2t) 4 ( 3t)

co-dofn 1nant

Hunt I ng and f i sh I ng 4 ( 2t) 4 ( 3t)

co-danlnant

Total s 180 148

TABLE 4
Divisionof LaborAmongHunter-Gatherers

(1) (2) (3)

Diarision of Labor Total
in Primary Hunter-Gatherer Omitting Equestrians Omitting Equestrians
Subsistence Sample and High Fishers

Women contrihute 13 ( 8%) 10 ( 8%) 10 (13%)

more than men
(0-17)

Women contribute 14 ( 9e/o) 14 (11%) 13 (17%;)

about the same
as men ( 18-22 1
Men contribute 134 (83%) 107 (82%) 54 (7a%)
more than
women (23-40)

TOTALS 161 131 77

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 443
MYTHSABOUTHUNTER-GATHERERS

justificationfor an a priori downward weighting of one areaof theworld.If

mosthunter-gatherers thathavebeendescribed are in NorthAmerica,why
shouldtheybeunderrepresented? I donotthinkit wouldevenbefairto saythat
the environment of theNorthAmerican casesis probably lessrepresentative of
the past,because,if anything, the"marginality" of theenvironments of many
recenthunter-gatherers in theOldWorldmaymakethemlessinstructive about
Paleolithic conditions thantheNorthAmerican casesare.
Finally,I wishto address myselfto oneotherviewof hunter-gatherers thatI
havereasonto believeis erroneousnamely,theviewthathunter-gatherers are
relativelypeaceful(LeeandDeVoreI968: 9; ServiceI966: 60; StewardI968:
334; Turnbull I968: 34I).4 Ratings of frequencyof warfare wereobtained from
a previousstudyof 50 hunter-gatherer societies(C. R. EmberI975).5 If we
tabulatethe warfaredatafromthatworld-wide sampleof hunter-gatherers
(excluding thosefewcasesthathada littleherdingor agriculture), 64 percent
hadwarfareoccurring at leastonceeverytwoyears,26 percenthadwarfare
somewhat lessoften,andonlyI0 percent(including the !Kung)wereratedas
havingno or rarewarfare(seeTable5). Evenif we excludeequestrian hunters
(indicated by an "Eq")andthosewith 50 percentor moredependence on
fishing(indicatedby an "F"), warfareis rarefor only I2 per centof the
remaining hunter-gatherers. In sum,hunter-gatherers couldhardlybedescribed
as peaceful.
The discrepancy betweenLee's(I968) surveyandthepresentoneraisesthe
possibilitythatwhatis typicalof hunter-gatherers mayvaryconsiderably by
geographical region.Afterall, if Lee'sexclusion of manyof theNorthAmeri-
canhunter-gatherers enhanced theapparent importance of gathering,thenthe
NorthAmerican hunter-gatherers maybequitedifferent fromthoseof theother
areas.To seeif thisis the case,I havetabulated thedistribution of residence,
divisionof labor,andsubsistence foreachof thefivemajorgeographical areas
of the worldin whichhunter-gatherers arefound(seeTable6).
The datain Table6 showthatpatrilocality is everywhere more"typical" of
hunter-gatherers thanbilocality. Patrilocalresidence is foundin themajority of
the hunter-gatherer societiesin sub-Saharan Africa,the insularPacific,and
NorthAmerica. Andevenin Eurasia andSouthAmerica whereonly33percent
of the hunter-gatherers are patrilocal,and matrilocalresidenceis equally
frequent,
bilocality is stilllesscommon. Asidefroma fewcasesthatarescattered
aroundthe world(!Kung,Mbuti,Andamanese, Alacaluf,andAweikoma),
bilocalityis foundcommonly onlyin a few cultureareas:mostof the Pomo-
Yukicasesin California arebilocalandmany(butnotmost)of theEskimoand
GreatBasincasesarebilocal.
If patrilocalresidence is the mosttypicalformof residence amongrecent
hunter-gatherers, we needto explainits prevalence. In a studyof residential
variationamonghunter-gatherers (C. R. EmberI975), I testeda numberof
theoriesaboutthe conditions thatmightfavorpatrilocal residence.The study
foundnoempirical support forthenotionthattheimportance of menin hunting
(Steward I955; Service I962) or the importance of menin war(Murdock I949;
ServiceI962) favorpatrilocal residence.Hunter-gatherer witha greater
societies
dependence on huntingor with frequentwarfareareno morelikelyto have
patrilocal
residence thanotherhuntingandgathering societies.Butsupport was

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 444 ETHNOLOGY

TABLE5
WarfareFrequencyAmongHunter-Gatherers*

Andamanese (Eh1) S . Ute (Nd2 ) Eq

Hore t han once
Murngi n ( I d2) Kutena; (Nd7) Eq
eve ry two yea rs
Tiwi ( Od3) Coeur D'Alene (Ndl4) Eq

Aleut (Na9) F Yavapa i (Nd66)

SekanI (Na28) Gros Vent re (Ne | ) Eq

Yurok (Nb4) f Cananche (Ne3) Eq

Bel lacoola (Nb9) F Crow (Ne4 ) Eq

Squamish (Nbl3) F Tehuelche (594) Eq

Klallam (Nbl6} F Bororo (Sil)

Maidu (Nc12) Awei koma (Sj 3)

Less frequent Dorobo (Aa2} E. Pomo (Nc18)

Sem*ng (£33) Sanpol I (Nd4) F

Nootka (Nbit) f Whtte Knife tFJd43)

Tubatulabal (Nc2} Shivwlts (Nd52)

No or rare warfare 'Kung (Aal} Yahgan (SgJ) F

Pekangcku (Na34)

These data cone from C. R. £rrber 'v1975). The few cascs of huntcr-gathercrs with

a little herding or agriculture were excluded from th;s table. Atlas numbers

appear in parentheses; the 5ymbol F refcrs to hunter-gatherers that are highly

J¢pendent on fishing (approximately 50 percent or more); the stenbol £q refers to

hunter-gatherers than have horses.

foundforthe theory(M. EmberandC. R. EmberI97I) thatinternal warfare

favorspatrilocal residence(since familiesmaywantto havetheirsonsnearbyfor
protection), or, in the presence of purelyexternalwarfare,a male-dominant
divisionof laborfavorspatrilocal residence.Therewasempirical support forthe
traditionalnotion(MurdockI949) that sexualdivisionof laborby itself
determines residence.(Whydivisionof laborpredicts residenceamonghunter-
gatherersbutnotin samples of societies complexity
at alllevelsof cultural [see
M. EmberandC. R. EmberI97I andDivaleI974] iS something of a puzzle.)

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 MYTHSABOUTHUNTER-GATHERERS
445

TABLE6
Residence,Subsistence,Divisionof LaborAmongHunter-Gatherers
by Area of the World*

Sub-Saharan East Insular North So. s Cen.

Africa Eurasia Pacific America America

Residence

Patr;local 3 ( 60t) 2 ( 33t) 7 (tOOt) 98 ( 65t) 3 ( 33t)

B;local 2 ( 408) 1 ( }7t) 0 ( 0t) 2} ( 14t) 2 ( 22t)

Matri local O ( 0t) 2 ( 33t) 0 ( 02) 24 ( 16t) 3 ( 33t)

Other O ( Q) I ( 1 7t) 0 ( °t) 8 ( 576) 1 ( 118)

Importance of Gather; ng

tNore than hal f the calories 3 ( 60t) 0 t ot) 2 t 25t) 12 ( 8%) 1 ( 208)

Flalf the calories 1 ( 20t) 0 ( 0;t) 2 ( 258) 20 ( 13t) 1 ( 20t)

Less than hal f the calories 1 ( 20t) 6 (IQOt) 4 ( 50t) 120 ( 79t) 8 ( 80t)

Re 1at; ve Importance of

Subsistence Activ; ties

Gathering most important 3 ( 60t) 2 ( 33t3 4 ( 50t) 43 ( 28t) 2 ( 20t)

Hunting most important 1 ( 20t) 1 ( 17t) } ( 12t) 38 ( 25t) 5 ( 50t)

F;shing most important ° 1 0t) 2 ( 33t) 1 t 12t) 63 ( 41t) 3 ( 30t)

Co-dominant combinations I ( 20t) 1 ( 17t) 2 ( 25t3 8 ( 5t) 0 ( 0t)

Division of Labor in Subsistence

Women contribute more than men 3 ( 60t) o ( Ot) 4 ( 57t) 6 ( 4t) 0 ( 0t)

slomen and men about equal I ( 20t) 1 ( 17z) 1 ( 14t) 9 ( 7t) 2 ( 25t)

Men coneribute more than *somen I ( 20t) 5 ( 83t) 2 ( 28t) 120 ( 89t) 6 ( 75t6)

*The first letter of tt7e Atlas identification number provides the basis for the geographical

breakdc>wn shown above.

Inasmuch as mosthunter-gatherershavewarfare(as notedabove) whichis

usuallyinternalwarfare,and inasmuchas men usuallycontributemoreto
subsistence,the "warfare"theoryandthe"division
of labor"theorymayboth
explainwhy mostethnographically describedhunter-gatherer
societieshave
beenpatrilocal.
Bilocalitymay occuronly in somewhatunusualcircumstances.Service's
(I962) notionthatseveredepopulation
transforms
a previously
unilocalsociety

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 446 ETHNOLOGY

intoa bilocalor multilocalsocietyseems to be generallysupportedAmong

depopulatedagriculturalists (C. R. Emberand M. EmberI972) as well as
amongdepopulated hunter-gatherers(C. R. EmberI975), non-unilocalresi-
denceis signiEcantly morelikelythanunilocality.(Amonghunter-gatherers, the
Andamanese and the Yavapai,for example,weredepopulatedandhad bilocal
residence.)It has also been suggestedthat severeresourcefluctuationand sex-
ratiofluctuationmay favor bilocalityamong hunter-gatherers, since bilocal
residenceprovidesa way of shiftingbandmembershipin responseto changing
conditions(see FordeI947; EgganI966, I968; AndersonI968; StewardI968;
Lee I976; C. R. EmberI975). Thereappearsto be empiricalsupportfor bothof
thesetheories(C. R. EmberI975). Usingan indexof precipitation variabilityas
a measureof resourcefluctuation,the more such variability,the more likely
bilocality(e.g., Aweikoma,Kidutokado,Wappo).And usingtypicalbandsize
as a measureof the likelihoodof sex-ratiofluctuation(the smallerthe band,the
morelikely therewill be chancedeparturesfroma sex ratioof I.OO), I found
that hunter-gatherers with verysmallbands(e.g.?!Kung Mbuti) tend also to
have bilocal residence. Thus an earlierstudy (C. R. EmberI975) and the
presentone suggestthatbilocalresidencemaynot be thatcommonbecausethe
precipitating conditionsmaynot have been so prevalentamongrecenthunter-
gatherers.Indeed, in the Paleolithicand Mesolithic,when many hunter-
gatherersinhabitedless marginalenvironments, bilocalitymayhave been even
less common.
Unlike residence there appearsto be some geographicalvariabilitywith
respectto the importanceof differentsubsistenceactivitiesand the relattve
contributionof menandwomento subsistence. As the readercansee in Table6,
gatheringis generallythe most importantsubsistenceactivityin sub-Saharan
Africaand the insularPacific;gatheringandfishingtie for the mostimportant
activityin eastEurasia;fishingis generallythe mostimportantactivityin North
America;and huntingis generallymost importantin South America.Not
surprisingly, since the importanceof gatheringshowsgeographicalvariability,
so doesthe relative contributionof menandwomento subsistence. Both in sub-
SaharanAfrica and the insularPacific-where gatheringis more important
than anyotheractivity-women typicallycontributemorethanmento subsist-
ence (see Table 6). But in most areasof the world men typicallycontribute
moreto subsistence.Sincemostgeographicalareashave men typicallycontrib-
utingmoreto subsistenceand sincethereareproportionately veryfew societies
with women contributing more than men (only I3 out of I63), the current
notionthat womentypicallycontributemore to subslstencethan men among
hunter-gatherers clearly needs to be revised.Men may generallycontribute
moreto subsistence thanwomenbecausehuntingandfishinggenerallyaccount
for more thanhalf the caloricintakein mosthunter-gatherer societies.
Finally, it is clear that the high proportionof North Americanhunter-
gatherersin theAtlasdoesnot accountfor the generalfindingthatmosthunter-
gatherershave had warfareat leastonceeverytwo years.As Table 5 shows,I4
of the 2I NorthAmericancases(67 percent)fightat leastonceeverytwo years,
which is not significantly differentfromthe percentagein the restof the world
(six out of ten or 60 per cent).

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 447
MYTHSABOUTHUNTER-GATHERERS

AND IMPLICATIONS
SUMMARY
The datapresentedhere suggestthat somecurrentviewsabouthunter-
gatherersmayneedto be revised.Specifically, thedatasuggestthat,contrary to
currentopinion,recenthunter-gatherers patrilocal,
aretypically have
typically
mencontributing moreto subsistence
relatively thanwomen,andtypically have
hadfairlyfrequent warfare.
As notedabove,muchof the interestin the hunter-gatherer wayof life
appearsto be associated with the beliefthattypicalcharacteristicsof recent
hunter-gatherers weretypicalalso in the Paleolithic. But even if we quan-
establish
titatively "normal"
thestatistically patterns
cultural of recenthunter-
gatherers,I takeissuewtththe beliefthatwe areentitledto inferfromthis
information whatculturalpatterns musthavebeentypicalin the distantpast.
We know,forexample, thatthereis substantial amongrecenthunter-
variation
gatherersin residence,subsistence, divisionof labor,and warfare.If these
variationsare the resultof differentcausalconditions, then whathas been
"typical"in recenttimes may only be a statisticalartifactof the recent
prevalence of certaincausalconditions.If wewishto makeinferences aboutthe
Paleolithic,then,I suggestwe needto do twothings.First,we needto discover
whatpredicts variationamongrecenthunter-gatherers.ffAndthen,usingarche-
ologicalindicators,we needto discoverthe pastprevalence of thosepredictors
andtheirpresumed effects.7 in theseefforts,weshallbeable
If we aresuccessful
to drawtnferencesaboutPaleolithichunter-gatherers that are basedupon
systematicevidenceratherthanmerelyplausibleconjecture.

NOTES
I. It should-be noted that more dependenceon huntingdoes not predictpatrilocality among
hunter-gatherers, althoughhigherdependenceupon gatheringdoes predicta tendencytoward
matrilocalityand higherdependenceuponfishingpredictsa tendencytowardpatrilocality(C. R.
EmberI975).
2. The procedure usedto calculatedivisionof laborscoresis now fairlyconventional.Basically,
the non-numerical informationin the Atlas on the degreeto whichmalesandfemalesparticipate
in each of the five primarysubsistenceactivities(columns54, 56, 58, 60, 62) iS assigneda
numericalscorewhich is thenmultipliedby the importance of eachof the subsistenceactivitiesas
iven in the Atlas (o-g in columns7-II) and summedacrossall activities.Numericalscoresare
assignedas followsto the informationon contribution by sex: F= o; G= I; D,E,I,P,O,a dot= 2;
N=3; M=4. Scoresrangefromo-40. A scoreof 40 meansthat the men do all the subsistence
work, 20 meansan equalcontributionof men and women.
3. Omitted from Table 4 are those societiesfor which missingdivisionof labor information
(arbitrarilyassigneda "2" in calculatingthe score),if known,mighthavealteredtheirplacement
in Table 4.
4. Harris(I975), citing data collectedby Divale, has noted that hunter-gatherers have fairly
frequentwarfare.
5. The deEnitionof warfarein this studyfollowedthat of M. Emberand C. R. Ember(I97I:
577).Warfareis definedas fightingbetweentwo or moreterritorialunits(at the communitylevel
on up) as long as there is a group of Eghterson at least one side. If pacificationoccurred,
frequencyof warratingsweremadeas far backas 50yearspriorto the ethnographic presentlisted
in the Atlas. If pacificationoccurredearlierthan 50 yearsback,warfarewas codedas rareor
absent.
6 For some predictiveconditionsof residentialvariationamong recenthunter-gatherers, the
readeris referredagain to C. R. Ember(I975).

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions
 448 ETHNOLOGY

7. For example,cross-culturalresearchhas suggestedone possiblearcheologicalindicatorof

matrilocalveruspatrilocalresidence(M. EmberI973; Divale I977).
BIBLIOGRAPHY
AndersonJ. M. I968. Commentson the "Analysisof GroupComposition." Manthe Hunter
eds. R. B. Lee and I. DeVore,pp. I53-I55. Chicago.
Damas,D. I969. Characteristics of CentralEskimoBandStructure.Contributions to Anthro-
pology: Band Societies,ed. D. Damas, pp. II6-I38. National Museumsof Canada
Bulletin228.
DeVore, I., and M. J. Konner. I974. Infancyin Hunter-Gatherer Life: An Ethological
Perspective.Ethologyand Psychiatry,ed. N. F. White, pp. II3-I4I. Toronto.
Divale, W. T. I974. Migration,ExternalWarfare, and MatrilocalResidence.Behavior
ScienceResearch9: 75-I33.
I977. Living FloorArea and MaritalResidence:A Replication.BehaviorScience
ResearchI2: I09'II5.
Eggan,F. I966. The AmericanIndian:Perspectives fortheStudyof SocialChange.Chicago.
I968. Commentson "Typologyand Reconstruction." Man the Hunter,eds. R. B.
Lee and I. DeVore, pp. I6I-I62. Chicago.
Ember,C. R. I975. ResidentialVariationamong Hunter^Gatherers. BehaviorScienceRe-
searchI0: I99^ 227.
Ember,C. R., and M. Ember. I972. The ConditionsFavoringMultilocalResidence.South-
westernJournalof Anthropology28: 382-400.
Ember, M. I973. An ArcheologicalIndicatorof MatrilocalVersus PatrilocalResidence.
AmericanAntiquity38: I77- I82.
Ember,M., and C. R. Ember. I97I. The ConditionsFavoringMatrilocalVersusPatrilocal
Residence.AmericanAnthropologist 73: 57I-594.
Forde,C. D. I947. The Anthropological Approachin SocialScienceAdvancement of Science
4: 2I3-224.
Harris,M. I975. Culture,People,Nature:An Introductionto GeneralAnthropology(2nd
ed.). New York.
Hiatt, L. R. I968. Ownershipand Use of Landamongthe AustralianAborigines.Man the
Hunter,eds. R. B. Lee and I. DeVore, pp. 99-I02. Chicago.
Leacock, E. I969. The Montagnais-Naskapi Band. Contributionsto Anthropology:Band
Societies,ed. D. Damas,pp. I-I7. NationalMuseumsof CanadaBulletin228.
Lee, R. B. I968. What HuntersDo for a Living,or, How to MakeOut on ScarceResources.
Man the Hunter,eds. R. B. Lee and I. DeVore, pp. 30-43. Chicago.
I976. !KungSpatiaIOrganization:An EcologicaIand HistoricalPerspective.Kala-
hariHunter-Gatherers: Studiesof the tKungSan and TheirNeighbors,eds. R. B. Lee
andI. DeVore,pp. 73-97. Cambridge.
Lee, R. B., and I. DeVore. I968. Man the Hunter.Chicago.
Meggitt,M. J. I962. DesertPeople:A Studyof the WalbiriAboriginesof CentralAustralia.
Sydney.
Murdock,G. P. I949. SocialStructure.New York.
I967. Ethnographic Atlas: A Summary.Ethnology6: I09-236.
Radeliffe-Brown, A. R. I930. The SocialOrganization of AustralianTribes,I. OceaniaI: 34-
63.
Service,E. R. PrimitiveSocialOrganization:
I962. An Evolutionary Perspective.
New York.
I966. The Hunters.EnglewoodClifEs.
Steward,J. H. I955. Theoryof CultureChange.Urbana.
I968. CausalFactorsand Processesin the Evolutionof Pre-farmingSocieties.Man
the Hunter,eds. R. B. Lee and I. DeVore,pp. 32I-334. Chicago.
Turnbull,C. I968. Commentson "PrimateBehaviorandthe Evolutionof Aggression."Man
the Hunter,eds. R. B. Lee and I. DeVore,p. 34I. Chicago.
Williams,B. J. I968. The Birhorof IndiaandSomeCommentson BandOrganization. Man
the Hunter,eds. R. B. Lee and I. Devore,pp. I26-I3I. Chicago.

This content downloaded from 195.34.79.54 on Sun, 22 Jun 2014 15:00:20 PM

All use subject to JSTOR Terms and Conditions