7/30/2019 Schiffer 96 http://slidepdf.com/reader/full/schiffer-96 1/21 Society for American Archaeology Some Relationships between Behavioral and Evolutionary Archaeologies Author(s): Michael Brian Schiffer Reviewed work(s): Source: American Antiquity, Vol. 61, No. 4 (Oct., 1996), pp. 643-662 Published by: Society for American Archaeology Stable URL: http://www.jstor.org/stable/282009 . Accessed: 20/05/2012 15:07 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 [email protected]. Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to American Antiquity. http://www.jstor.org
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 1/21
Society for American Archaeology
Some Relationships between Behavioral and Evolutionary ArchaeologiesAuthor(s): Michael Brian SchifferReviewed work(s):Source: American Antiquity, Vol. 61, No. 4 (Oct., 1996), pp. 643-662Published by: Society for American ArchaeologyStable URL: http://www.jstor.org/stable/282009 .
Accessed: 20/05/2012 15:07
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 [email protected].
Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to
SOME RELATIONSHIPS BETWEEN BEHAVIORALAND EVOLUTIONARY ARCHAEOLOGIES
MichaelBrianSchiffer
Diversity in archaeology's social theories is desirable, butfactioning of the discipline into antagonistic, paradigm-based
camps undermines he scientific enterprise.In order topromote effortsat buildingbridges betweendifferenttheoreticalpro-
grams, this paper examines relationships between behavioral archaeology and evolutionary (selectionist) archaeology.Potential commonground is broughtto light, incompatibilitiesare critically examined,andpossible synergies are explored.It is concluded that there is nofundamentalreason why these twoprograms cannot work in concert to achieve the goal of
explaining behavioral (or evolutionary) change in human societies. Whetherproductive relationships can be established
between otherprograms remains to be determinedon a case-by-case basis.
La diversidad de teorias sociales en arqueologia es necesaria,pero la partici6n de la disciplina enfacciones antagonisticasbasadas en paradigmassabotea la empresa cientifica. Parapromover esfuerzosdestinados a crearpuentes entreprogramasteoricos distintos, este articulo examina las relaciones entre arqueologia conductualy arqueologia evolucionista (selec-
cionista). Se resaltan las areas comunes,se examinancriticamente las incompatibilidades,y se exploranposibles sinergias.Se concluyeque no existen razonesfundamentalespor las que estos programasno puedan trabajaren conciertopara alcan-
zar la meta de explicar cambio conductual (o evolutivo)en sociedades humanas. Laposibilidad de que se establezcan rela-
ciones productivasentre otrosprogramas queda por determinarsecaso por caso.
Scienceconsists n groupingacts so thatgen-eral laws or conclusionsmay be drawn romthem
-Charles Darwin, The Autobiography ofCharlesDarwin, 1809-1882, p. 70
Since the early 1970s, processual archaeol-
ogy's social theory (sensu Schiffer 1988)hassufferedmany ndignitiesat thehandsof
critics (for useful discussions, see Lamberg-
Karlovsky 1989; Preucel 1991; Trigger 1989;
Yoffee and Sherratt1993). Even so, processual-ism-albeit now in manyvarieties-remains well
entrenched n everydaypractice,andit is doubtful
that any competingprogramwill be able to dis-
lodge it and achieve a comparable position ofdominance. For the foreseeable future, then, a
multitudeof minorityprograms, ncludingbehav-
ioral,evolutionary, ognitive,andMarxistarchae-
ologies, will struggle to win followers and alterdisciplinary practice (see discussions in
Wandsnider 992).As a result,archaeologistsmayneed to become accustomed to an abundanceof
seemingly incompatible ocial theories,contribut-
ing to the "thousandarchaeologies" previouslywelcomed(Schiffer1988:479).Althoughdiversityin social theory is desirable(Knapp 1996), divi-
sion of archaeology into antagonistic camps,
seemingly ncapableof engagingeach other n dis-
cussions of substantive issues, diminishes the
integrityof thedisciplineas a scientificenterprise.An alternativepatternof communicationcan
be envisioned. Instead of caricaturing,misrepre-
senting, or summarilydismissingrivalprograms,
archaeologists might make a diligent effort to
understand each other's social theory-maybeeven build some intellectual bridges. Perhapsbecause their positions on significant issues of
Michael Brian Schiffer * Departmentof Anthropology,Universityof Arizona,Tucson,AZ 85721
AmericanAntiquity, 61(4), 1996, pp. 643-662.
Copyright? by the Society forAmericanArchaeology
643
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 3/21
AMERICANANTIQUITY
social theory lie between today's theoretical
extremes, some behavioral archaeologists are
eager to hold discussions with practitionersof
other programs (Walker et al. 1995:8); manypapers in ExpandingArchaeology (Skibo et al.
1995), for example, initiatepromisingdialogues
(e.g., McGuire 1995; Nielsen 1995; O'Brien and
Holland 1995b; Orser 1995; Senior 1995; Wylie
1995). Similarly, n the context of historicalcase
studies,I have alreadyexploitedcommongroundbetween behavioral and postprocessual archae-
ologies for explaining instances of behavioral
change pertaining to portable radios (Schiffer
1991) and early electric automobiles(Schifferet
al. 1994). Surprisingly,both studies also haveevolutionary eatures n thattheytreat,diachroni-
cally, the interplaybetween variationand selec-
tion processes (more on this below). Although
handling unconventional subjects, the portableradioandautomobileresearchesnonethelessraise
hopesthatadditionalworkalongtheselines mightlead eventually to a more integrateddiscipline
(see also Duke 1995).An explicit experiment n communication,his
paperexaminessomerelationships etweenbehav-
ioral and evolutionary archaeologies (both aredefinedbelow).That theremaybe important ela-
tionshipsbetweenthese two programshas already
been suggestedby severalevolutionists(O'BrienandHolland1995a:178-179, 193, 1995b;O'Brien
ingly numerousandsignificant.Second,thepapertreats he assumptions f evolutionary rchaeologythatseem incompatiblewith behavioralprinciples.It is suggested hatsomeevolutionary ositionsare
ology, see Schiffer [1995c]; for a concise codifi-
cation of fundamentalenets,
from whichmuchof
this section is adapted,see Schiffer [1995d]; for
recent developments, see Longacre and Skibo
[1994] and Skibo et al. [1995].) In the following
synopsis, I highlight the tenets of behavioral
archaeologymost relevant for engaging the evo-
lutionaryprogram.As the name implies, behavioralists ay stress
on studyingbehavior:whatpeople actuallydid or
do (Nielsen 1995;Reid 1995;Walkeret al. 1995).The basic units of behaviorareactivities,defined
as the interactionbetweenelements (e.g., human,artifact, animal), at least one of which is an
energy source (Rathjeand Schiffer 1982:Chapter3; Schiffer1992; Schiffer and Skibo 1997).Thus,artifacts are an integralpart of humanactivities,from a marriageceremonyto rebuildinga diesel
engine. After all, a white wedding gown is as
essential to a traditional church wedding as
wrenches are to tearing down an engine; in the
absence of either, crucial interactions would be
adverselyaffected.
Theartifacts andevenpeople) takingpart n anactivityhave,by virtueof theirmaterialcomposi-tion and form, specific properties hat affect their
suitability or interactingn particularways.These
activity-specific capabilitiesare knownas perfor-mance characteristics nd can pertain o any kindof interaction-mechanical, thermal,visual, etc.
(Nielsen 1995;SchifferandSkibo 1987, 1997).Activities are usually carriedout, often recur-
rently, by people in behavioral components-a
society'sunits of organization.Behavioralcompo-
nents, such as householdsand communities,con-sist of people, places, and artifacts(RathjeandSchiffer1982:Chapter; Schiffer1992:Chapter ).
Because each kind of artifact tends toward
uniqueness n its set of properties ndperformancecharacteristics, here are few true "functionallyequivalent" artifact types (Schiffer 1979).Activities,whichvary n energysources,otherele-
ments, and interactionpatterns,also usually lackexact functionalequivalents Schiffer1979).Thus,when one kind of artifactreplaces another n an
activity or when one kind of activity replacesanother n a behavioral omponent,"disjunctions"
are created that can initiate further change
processes. As a result, much behavioralchangeoccurs n response o the (often unintended) onse-
ties, entail compromises among performancecharacteristics and interactions (McGuire and
Schiffer 1983; Schiffer and Skibo 1987);patternsin these compromisesare influencedby specificbehavioral actors of lifewayandsocial organiza-tion (McGuireand Schiffer 1983; Schiffer 1992;Schiffer and Skibo 1987, 1997). These compro-
mises can be illuminated, n the case of artifacttypes, by use of performancematrices,which fur-
nish explicit comparisons n relationto activity-
specific performance characteristics (Schiffer1995b;Schifferand Skibo 1987).
Farfrombeing autonomous, ndividualactivi-
ties in a behavioralsystem (e.g., a society) are
connected,directlyor indirectly, o all otheractiv-ities by movements of people and artifacts. The
structured elationshipsbetween activities estab-
lish the causal pathwaysalong which behavioral
changestravel(Schiffer 1979; Schifferand Skibo1997). Change processes can also restructure
relationshipsbetweenactivities.
The life historyof artifacts andof people) is a
favored framework for organizing behavioral
studies (Rathje and Schiffer 1982:Chapter4;Schifferand Skibo 1997;Walker1995;Walkeret
al. 1995). Life histories based on groups ofrelatedactivities or processes (e.g., procurement,manufacture, nduse) are knownas flow models
(Schiffer 1972, 1976), whereas those focused on
individualactivitiesaretermedbehavioralchains(Schiffer 1975b, 1976). Developmental cycles(Goody 1971), anotherorganizing ramework, rethe definablestages of existence in the life histo-ries of behavioral components (Rathje and
3. In the explanationof variationand change,culture is not treated as a causal agent (Braun
1991:427; O'Brien and Holland 1992:37;
Ramenofsky1995:137;Schiffer 1995a;Walkeret
al. 1995:2-4). The adjective "cultural"merelymeans learned, as in cultural transmission;
human,as in culturalphenomena;or societal, as
in cross-culturalregularity.Selectionists should
take no exception to this statement, but some
behavioralistsmightbe rather eluctant ojettison
anthropology'smost sacred cow.
4. The phenomenologicalworld of interest is
variationand change in human behavioror soci-
eties. Despite Dunnell's (1980:48) occasional
acknowledgment hat "humanbehavior" s "theprincipalsubjectmatter," ome selectionists maycontestthis tenet,which is expressed n systemic-context terms, because they believe that their
equalsvelocity times time),does conformto thesecriteriabecause, in fact, it is not a scientific law
or theory at all, but a mathematical aw (on the
distinction, see Salmon 1982). The position that
theory is immune to empirical falsification
(Dunnell 1982:16) manifestly contradictsscien-
tific practice;as Sober(1984:82) notes, "theoret-
ical claims ought to be testable." The most
abstractand general theories, of course, may be
difficult to test--even today,Einstein'stheory of
generalrelativity s still undergoingevaluation-
yet they are potentially testable (Nagel 1961).Finally,I note thatthe conceptof"true by defini-tion"is itself problematic e.g., Sober 1984:62).
In view of the contentionthattheories aredef-
initionally true, it is surprisingthat theories arealso said to have a substantialempiricalcontent.
According to Dunnell (1989:44), a discipline'sgrand heorymust"generate ts own data"and theunits specified in the theory must be "directlymeasurable in the phenomenological world"
(Dunnell 1982:7, 1994:34). A view more gener-
ally held in science is that the entities, mecha-nisms, or processes postulatedby a theory haveno immediateempiricalcontent,fortheyare often
unobservable.To operationalizeor test a theory,one employs appropriateunits and instrumentsfor measuring the variables that it implicates.These instruments(and the rules of correspon-dence that link them to the theory) involve stillother laws and theories. For example, in testingthe kinetic theoryof gases, which involves invisi-ble entities called molecules, one has to measure
a gas's temperature.Temperaturecan be mea-sured,for example, with a mercurythermometeror infraredthermometry;n eithercase the instru-
ment'soperation s based on principlesother than
the kinetic theory. In addition, from the kinetic
theory one can deduce neithermercury'scoeffi-
cient of thermalexpansion
nor agas's
infrared
emission spectrum. n short,scientific theoriesdonot articulate immediately with the empiricalworld(Sober 1984:73);interposedbetween theo-
(Nagel 1961;Tschauner1996).Failure to appreciatethe complex, principle-
rich apparatus hat links theories to observations
can lead to unhelpfuladvice on building archae-
ological theory and models. To wit, Dunnell(1980:88) asserts that a theory's variables "can-
not be defined in behavioralterms."Going fur-
ther he claims, "If archaeologists are going to
employ evolutionarytheory,they must rewrite itin terms of variables that are empirical in the
[archaeological] record" (Dunnell 1980:88). Ifthis recipe were followed literally,then a theorywould be precluded from implicating mecha-nisms andprocesses, which aredecidedlybehav-
ioral phenomena unobservable in the static
archaeological record. At best, archaeologistswould be reduced to crafting relationshipsbetweenmeasurementson sherds,chips of stone,and so on-scarcely the stuff of theory (but see
O'Brien and Holland 1992); at worst,archaeolo-
gists would be operatingin the murkyworld of
merged systemic andarchaeologicalcontexts(onthenecessity of keepingthese contextsconceptu-ally and analytically distinct, see Reid 1985,
1995). In the final analysis, evolutionarytheoryitself cannot be rewritten n archaeological-con-
text terms:sherdswere not partof anyone'sphe-notype (unless reused), yet a cooking pot-asystemic-context entity operationalizedthroughbehavioral nference-was.
The selectionist prohibition against framingmodels andtheories in behavioral erms is out of
step with practice in modern evolutionarybiol-
ogy. In the latterdiscipline, theories and models
incorporatebehavioralvariableson a wide rangeof systemic phenomena, such as predator-preyinteractions,mating patterns,foraging behavior,
and maintenanceof territories,none of which isempirical in the paleontological and paleoenvi-ronmental ecords.Itcouldnotbe otherwise;after
Schiffer] 649
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 9/21
AMERICAN ANTIQUITY
all, "behaviors at once causeandconsequenceof
evolution" Plotkin 1988:8).In accordwith evolutionarybiologists, and in
contrast o the selectionistposition,
behavioralists
maintain that theories, models, and explana-tions-even those offeredby selectionists-must
be framed n behavioral erms.Ironically, upportfor this claim can be found in Dunnell's (1989)own effort to build an evolutionarymodel. In that
paperhe seeks to explainthe selectionfor "waste"
behavior, specifically the Woodland mortuarycults of the easternUnited Statesthat left behind
obtrusiveburialmounds. The variablesand para-meters of the model include "mean carrying
capacity,""no change or difference in subsis-tence," "populations n equilibriumat different
sizes," "shortfalls in productivity,""intensifica-
tion" of subsistence, "waste-typebehavior,"and
"mortuary ult"(Dunnell1989:48).Tomy knowl-
edge, surveyarchaeologistsnever encountera car-
rying capacity or a waste-typebehavior,nor do
excavators uncover a productivityshortfall or a
mortuarycult. In every instance,these variables
and parametersare more-or-lessbehavioral(i.e.,
systemiccontext),not-as Dunnell(1980:88)pre-
scribes-written interms"thatareempirical n therecord."Other selectionist models invoke vari-
ables and parameters hat are equally systemic
(e.g., GravesandLadefoged1995;Leonard1989;
Leonard and Reed 1993; O'Brien et al. 1994).
Despiterhetoric o the contrary, electionistsbuild
models almostas behavioralas behavioralists.
Becausetheoriesandmodels are formulated n
behavioralterms, the investigator s obligatedto
forge links, through measurement,to empiricalunits in the archaeologicalrecord.This linkage
process is called inference, and it brings us to asecond incompatibility.
On Behavioral Inference
Many selectionists deny the need for behavioral
inferenceanddenigrateas unscientificthe activi-
ties of archaeologistswho reconstructthe past.These efforts are labeled "reconstructionism,"
which Dunnell (1978a:194) elevates to the status
of a paradigm,attributing ts full flowering to
processual and behavioral archaeologies.
According to Dunnell (1978a:195), "behavioralreconstructions.. cannotbe the foundationof an
approach hatprofesses to be both scientific and
evolutionary."Behavioral reconstruction is not
science (Dunnell 1980:78, 1982:20, 1989:43,
1992a:87), but a purveyor of "just-so stories"
(Dunnell 1982:20)that should be abandoned
(Dunnell 1989:45).The decisive dismissal of inference contradicts
many thoughtfulworks in archaeologicalepiste-
mology (e.g., Fritz 1972; Patrik 1985; Salmon
1982; Schiffer 1976; Sullivan 1978; Wylie 1985)and also ignores the countless, well-established
inferences that have greatlyenhancedthe under-
standing of past societies worldwide. What is
more, modern evolutionary biology could not
exist without paleoenvironmentaland paleoeco-
logical reconstruction,not to mention a host ofinferencesaboutthe behaviorof particular axa.
Because selectionist models are actually
expressed in behavioralterms, they do requireinference (see also Wylie 1995:207-208).Evidence that supportsthis latter claim can be
foundin Dunnell's 1989) ownexplanatory ketch
dealing with "waste" behavior. When treating
specifics of the Woodlandcase, he is forced to
operationalize he model by means of behavioral
inferences. Dunnell does not himself construct
these inferences from archaeological evidence,but relies mainlyon reconstructionsprofferedbyotherinvestigators,as in "thenonagriculturalta-
tus of the Woodlandassociatedwith the mortuarycult" (1989:49). Sometimes, without any evi-
dence or argument,he makes assertions about
pastbehavior,such as that the mortuary ult "fre-
quently entails the laborious construction of
earthenmoundsandthemanufacture nddisposalof vast quantitiesof goods, many of which are
costly imports" (1989:48). Dunnell's study of
waste is indicative:every selectionist model canbe shownto rest, implicitlyor explicitly,on a net-
work of behavioral inferences (for other exam-
ples, see Graves and Ladefoged 1995; Leonard
1989; Leonard and Reed 1993; O'Brien and
Holland1992;O'Brienet al. 1994).
Despitethe selectionists'anti-reconstructionist
rhetoric, evaluation of their models patently
requiresbehavioral nference,and in this respect
evolutionary archaeology does not differ from
behavioral,processual,or postprocessualarchae-
ologies (on the latter'sdependenceon behavioralinference, see Duke [1995], Saitta [1992], and
tion between their pronouncementsand practiceon inference (Watsonet al. 1984:255-256), evo-
lutionaryarchaeologistscannotcrediblymaintainany longer that reconstruction s defective and
unscientific. Indeed, until selectionists take
archaeological inference seriously by treating it
explicitly and adoptingmodernmethodology, he
goals of theirprogramwould appear o be out of
reach. Fortunately, n several recent selectionist
papers,there is modest movementtowardexplic-
itly countenancing behavioral inference (e.g.,Jones et al. 1995; O'Brien and Holland 1995b;Teltser1995b).
Closely relatedto the claims that theoriesandmodels should not be built in behavioralterms
and that reconstructionof past behavior is to be
avoided is the selectionist position on functional
principlesand actualisticstudies,a thirdpoint of
incompatibility.
Functional Principles and Actualist Studies
Selectionists downplay the importanceof func-
tional principles n specific explanationsand dis-
regard he actualistic sources of these principles.
Dunnell notes thatecology and evolution are dis-tinct bodies of theory,the former concernedwith
functionalrelationshipsandmechanism, he latter
with historical explanations (Dunnell 1980:36;
1982:12). Both bodies of theory, he does grant,"arecommonly intermingled n particularexpla-nations"(Dunnell 1980:36, cf. 39). Indeed,spe-cific evolutionary explanations in biology are
utterly dependent on functional and behavioral
principles supplied by actualisticstudies, includ-
investigatorshave begun to find roles for experi-mental and ethnoarchaeologicalfindings (e.g.,Dunnell 1995:42; Dunnell and Feathers 1991;
Dunnell et al. 1994; Maxwell 1995; Neff
1992:150; O'Brien and Holland 1990:60,
1995:184;O'Brienet al. 1994). Moreover, n rec-
ognizing thatperformancecharacteristics f arti-
fact types affect their replicative success in
particular electiveenvironments, few selection-ists themselvesarebeginningto carryout experi-ments (e.g., Dunnell and Feathers1991; O'Brien
et al. 1994). Perhapsone day soon selectionist
practicewill includeeven ethnoarchaeology.
Discussion
As originally formulated,the evolutionary pro-
gram containsa numberof tenets that contradict
importantparts of behavioral (and processual)
archaeologies.Try as I might, however,I fail to
appreciatewhy anyof these ideas is necessaryforan evolutionary archaeology. Nothing in
Darwinian heory perse norin modernevolution-
arybiology precludes framing heories and mod-
els in behavioralterms, prohibitsmaking infer-
ences aboutthe past, or rules out the conduct of
actualistic,nomothetic studies. These views in
fact distance selectionism from productiveresearch strategies, practiced by behavioralists
and evolutionary biologists alike, that could
improvethe evolutionaryprogram.Moreover,by
maintaining enets that behavioralistsandproces-sualists regard as clearly wrong, selectionists
alienatethe very archaeologistswho ought to be
most receptive to their insights on evolution.
Selectionistsmay be well advised to cast off the
conceptualbaggagethathas so burdened he pro-
gram(see also Wylie 1995).
Seeking Synergies
Building on the common groundbetween evolu-
tionaryand behavioralarchaeologies,andexploit-
ing ideas from both programs,I now attemptto
indicate,with examples,how the studyof behav-
ioral(orevolutionary) hangemightbe enhanced.
Clarifying Selection Processes
In using the term "naturalselection," Darwin
called attentionto a kind of selection differentfromthatpracticed, orexample,by plantand ani-
mal breeders. In natural selection, competing
organisms are winnowed by environmental ac-
tors, such as predatorsandtemperature xtremes.
Today he term"cultural election" s in vogue to
designate processes internalto a humanpopula-tion (e.g., Durham 1991:165), whereas natural
selection is retained for selective agents in the
naturalenvironment.Evolutionaryarchaeologistshavediscussedthenecessityof a conceptlike cul-
turalselection,butno consensushas emerged forvarious viewpoints, see Braun 1995:132-133;
Dunnell 1980:53, 63, 1989:41;LeonardandJones
1987:211; O'Brien and Holland 1992:45, 48,
1995:178;Rindos 1984, 1989).In considering cultural selection, Dunnell
evinces skepticismwhile asking,
Is therea point n human volutionaryistoryat which electiondoesbecomeaninternalul-turalmatterndependentf environmentalon-straint ndchangebecomes ransformational?thinknot,ormoreprecisely, think t wouldbe
1980:62).Construingvariety-generating rocessesin this mannerhas, I suggest, hampered he study
of behavioral volution.Like modern evolutionary biologists (e.g.,Nitecki 1990), archaeologists houldregardvaria-
tionand ts sourcesas subjectseminentlyworthyof
explanation.Variationn a populationat one pointintime is a consequenceof bothpriorselectionand
variety-generating rocesses (e.g., inventionand
borrowing).Study of the latter is clearly crucial,forthe creationof new variants n culturalpopula-tions occurs commonly and sometimes at highrates.Because selection operateson variation, he
state of variationat one point in time immediatelyconstrains he outcome of selection (cf. Dawkins
1982:42-46; Neff 1992:147). Thus, one cannot
explainevolutionary hange n specific cases with-
out documenting and accounting for large and
rapidchangesin the availablevariation.New vari-
ants can arise throughan expansionof inventiveactivities in existing behavioral components,through heproliferation f behavioral omponents
undertakingnventiveactivities,or both.
An argumentcan be made thatvariety-gener-
ating processes work in patternedways. Someprocesses, forexample,cause inventionsof a par-ticularkind to clustermarkedly n time andspace.Indeed,the historicaland archaeologicalrecordsfurnishevidence of ratherdramatic-and some-times predictable-increases in the conduct of
variety-generation s that information(as matteror energy) coming from changed conditions in
selective contexts can stimulate an increase ininventiveactivitiesof behavioralcomponentsandcan foster the creationof new behavioralcompo-
nents. This process, which is illustrated n some
detail below, can be called stimulatedvariation
(compare to Neff's [1992:146] discussion of
"directedvariation"and Braun's[1991:431] con-
cept of "directed nnovation").Unlike "directed
mutation" Dawkins 1982:112) and"guidedvari-
ation"(Boyd and Richerson 1985:94-98), which
more than flirt with Lamarckianmechanisms of
change, the process of stimulatedvariation n no
way obviates selection; after all, every variant
producedduringan instanceof stimulatedvaria-
tion can be selectedagainst.Selection thusretains
its Darwinian role, but variety-generationbecomes central o evolutionary nquiry, he study
of its mechanismsandprocesses far fromtrivial.Theprocessof stimulatedvariation anhelp us
to reconcileadaptationist nd selectionistviews of
behavioral hange.In theadaptationistramework,a changingnatural nvironment r a growingpop-ulationexertsselectivepressures hatareperceivedas stresses or problemsby a cultural ystem'spar-
ticipants e.g., Hill 1977).Problem-solvingbehav-iors usually lead to an appropriate esponse,and
theresult s a new adaptation.As selectionistshave
pointedout, however, he adaptationistmodelpre-
sumes thatpeople enjoy a certainomniscienceinidentifyingsignificantproblemsandin forecastingthe future(e.g., Braun 1991:428, 1995:129;Neff
1992:146;see also Schiffer1979, 1992:Chapter).
Essentiallytransformational, daptationist xpla-nations conflate processes of variety-generationandvariety-selectione.g., Jonesetal. 1995:17-18;Rindos 1989:3).
An appreciationor stimulatedvariationallows
us to assignproblem-solvingtsproperrole in evo-
lution, that of producing new variants (Braun
1995:129;Rindos 1989:13-15). Variation reatedby this process (and certainlythere are others-
Basalla [1988]) is not directedby futureadaptiveneeds,but is shapedby contemporaneoushenom-ena intheselectiveenvironment,uchas anartifactthatdoes notreachbehaviorally ignificantperfor-mancelevels in an activityor an activitythat doesnot effectivelyplay its role in a behavioral ompo-nent. If stimulatedvariationhappensto furnish avariant hatbecomes fixed very rapidly, he entire
evolutionaryprocess may be so telescopedthat it
appears ransformational-when n fact it was not.Thetelescopingof evolutionaryprocessesis exac-
erbated,of course, by the low resolutionof most
Schiffer] 655
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 15/21
AMERICANANTIQUITY
archaeological hronologies,which makes it diffi-
cult to discernrapidexpansionsandcontractions f
variation.Whennotutterlyoverlooked, uch varia-
tion islikely
to be dismissed as mere noise that
complicatestypology-construction.Aware of this
problem,archaeologists houldnowdiligentlyseek
the fine-grainedbehavioralvariation,predictedbythe process of stimulatedvariation, hat our cul-
ture-historicalunits so relentlessly obscure (seealso Plog 1974).
Figure 1. Changes in the frequency of companies in the United States that manufactured vacuum-tube radios for the
home market, 1920-1955. Based on data in Grinder (1995), each company's period of radio production was estimated
from the time range of annual models listed.
manyfirms eagerly embracedhome-radiomanu-
facture;by 1949, however, the vast majorityof
newcomers had failed in the radio business.An even more dramatic nstanceof stimulated
variationduring commercializationcomes from
the history of U.S. portable radios (Schiffer
1991).Duringthe GreatDepression,battery-pow-ered sets that were designed to be easily carried
aroundwere rare n the marketplace. n 1939 and
1940, however,there was a phenomenal ump in
the variety of portable radio models offered to
consumers(Figure2). The cause of this burst of
commercialization activity was situated in the
extended selective context of radios and radiocompanies. To wit, with the intensification of
warfareactivities in Europe,radio makers saw an
opportunity o offer a type of productthat could
allow Americans to hear war news anywhere.It
should be noted that this episode of stimulated
variation involved the commercializationactivi-
ties of establishedradio-makingcompanies;new
firms were not founded to produce portableradios(see Figure 1).
Duringadoption,consumersbuy anduse com-
mercializedproducts. Purchasingactivities exertselectivepressures,and the lattercanbe swift and
sure. In 1953 nearly two million portableradios
were sold in the United States (ElectronicIndustriesAssociation 1970:13)-and every one
contained vacuum tubes. Late in 1954 the firsttransistorportableradiowas commercialized,and
others were rapidly broughtto market.Althoughfor a few years more expensive than their tube
counterparts,ransistor adios had marvelousbat-
tery economy along with the cachet of "moder-
nity."As a result, consumers quickly selected
againstthe tube-basedportablesand,in less than
seven years, only transistor radios remainedon
the market Schiffer 1991).
Surprisingly, he adoption process is also an
important source of variation, as consumersbecome inventors, iterallyplayingwith theirnew
toys. A new productis tried out in old activities
and the possibilities of new activities are
explored; the result is an expansion of activity
variation,which can contribute o stimulatedvari-
ation in processes of invention and commercial-
ization. For example, in the early 1920s, people
exploredthepossibilitiesof home radios,a newlycommercialized product, taking them outdoors
and using them in diverse activity settings, from
motorcycle-policeon the beat to hot-air balloonraces. These experimentswere widely publicizedin radiomagazines and called attention o possi-
Schiffer] 657
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 17/21
AMERICAN ANTIQUITY
120 -
03
0
10.20
CO
CZ
a,
0
CZ
Laa,
U3-
100 -
80 -
60 -
40 -
20 -
1
0 1.1
1920 1930 1940 1950
Figure 2. Changes in the frequency of different portable radio models manufactured and sold in the United States,1920-1955. Based on a ca. 10 percent interval sample of radios listed by Grinder (1995). To be counted as a portableradio, the receiver must use battery or three-way power and be designated as "portable."All portable radios in thisdatabase employ vacuum tubes.
bilities for new kinds of radios.Impressedby the
opportunities, inventors and entrepreneurs
respondedwith newradiodesignsforspecial-pur-pose communicationsgear, some of which were
commercialized Schiffer 1991).
Discussion
This framework f processes in the life historyof
a product type-invention, commercialization,and adoption-allows one to investigate variety-
generationand variety-selectionin a systematicmanner. n historiesof theportableradio(andthe
early electric automobile [Schiffer et al. 1994]),
the complex interactionof variety-generation ndvariety-selectionprocesses has been illuminated.
The intricateconnectionsof variety-generationandvariety-selectionprocessesestablish relation-
shipsbetweenunits of evolution at diverse scales.
Significantly, in the evolution of cultural phe-nomena,selectionprocesses at one scale can cre-
ate variationat others.Forexample,the selection
of inventions for commercializationby financial
andmanufacturingorporationseads to variation
in products available in the wholesale market-
place. Selection of this variation by retailingbehavioralcomponents hen createsproductvari-
ation in the retailmarketplace. n turn,the selec-
tion processes of consumers influence the varia-
tion in household artifact inventories. One can
envision that the latter variation will be actedupon by additionalselection processes, thereby
affecting the differential persistence of house-
holds. These cascading effects and complicated
relationshipsof variety-generatingand variety-
selecting processes ensure that both must be
treated n evolutionaryexplanations.
Conclusion
In an effort to promote constructive dialoguebetween different heoreticalprograms, hispaper
has examined some relationshipsbetween evolu-tionaryand behavioralarchaeologies, ocusingon
potentialcommon ground, ncompatibilities,and
possible synergies.Various assumptions, tenets, and principles
were delineatedthat have some potentialto form
common groundbetween the two programs.The
statements summarizingthis apparentcommon
groundappear o be of some importance.Tenetsof evolutionaryarchaeology hatcannot
be accepted by behavioralistswere scrutinized.It
was argued hatthese incompatibilities tem fromquestionable ssumptionshatarenot, in fact,inte-
gral to the selectionistprogram.Abandonment f
658 [Vol. 61, No. 4, 1996
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 18/21
RELATIONS BETWEEN BEHAVIORALAND EVOLUTIONARYARCHAEOLOGIES
these assumptions would make evolutionary
archaeology compatible,not just with behavioral
archaeology,but also with modern evolutionary
biology.Finally,the possibility that the two programscould enjoy synergies was examined through a
treatment of variety-generation and selection
processesalongwith scales andunitsof selection.
And what of the "thousandarchaeologies"once welcomed Schiffer1988:479)? remain om-
fortablewith a diversityof ideas-even a plethoraof socialtheories.Whatdoes concernme, however,is the prospectof archaeologybecoming perma-
nentlyorganized ntoparadigm-basednclaves hat
exchangeonly epithets.This is an intellectual ux-urythatarchaeology, disciplinewithtoo fewprac-titioners spread over too much subject matter,cannot afford. Perhapsit is time to dispute the
Kuhniandogma that differentparadigms-in this
case theoretical rograms orexplainingvariabilityand changein humanbehaviorand societies-are
conceptuallyncommensurable,nd so theirpracti-tionerscannottakepart n meaningfuldialogues.I
suggestthatthedegreeof compatibility rcomple-
mentaritybetween differentprogramsshould be
determinedon a case-by-casebasis. In the case ofbehavioraland evolutionaryarchaeologies,there
appearo be manyopportunitiesorsubstantive is-
cussions, and even hints that the two programsworking nconcertcouldproducea more ntegratedand successful science. Perhaps he establishmentof aconstructiveworking elationship etweenevo-
lutionaryarchaeologyand behavioralarchaeologycould serve as a model for interactionbetween
other,seeminglyless reconcilable, heoreticalpro-
grams.
Acknowledgments.An earlierversionof this paperwas read at
the Universityof Washingtonin January1994. I thank the
Department f Anthropology, ndespeciallyJulie K. Stein,for
the invitationto speak.AnnetteSchifferhelpedto record the
data for Figures1 and2. I am gratefulto the people who fur-
nishedperceptiveandhelpfulcomments on earlierversions ofthis paper, ncludingStevenL. Kuhn,Vincent M. LaMotta,R.Lee Lyman,PatrickD. Lyons,HectorNeff, Michael J. Shott,James M. Skibo,Julie K. Stein, PatriceA. Teltser,William H.
Walker, LuAnn Wandsnider, and especially Michael J.O'Brien and StephenL. Zegura.I am indebted to PatriceA.
Teltser,visiting
scholar in theLaboratory
of Traditional
Technology,1992-1995, for leading manystimulatingdiscus-sions about evolutionaryarchaeology.I thankMariaNievesZedefio fortranslating he abstract.
References Cited
Aitken,H.
1976 Syntonyand Spark:TheOrigins of Radio. Princeton
UniversityPress, Princeton,New Jersey.
1985 The Continuous Wave: Technologyand AmericanRadio, 1900-1932. Princeton University Press,Princeton,New Jersey.
Basalla,G.1988 The Evolutionof Technology.CambridgeUniversity
Person. Archaeologicaland Ethnological Perspectives,editedby C. Carrand J. E. Neitzel, pp. 123-141. Plenum,New York.
Darwin,C.1958 TheAutobiographyof Charles Darwin, 1809-1882,
editedby N. Barlow.Collins, London.
David,N.
1971 The Fulani Compoundand the Archaeologist.WorldArchaeology3:111-131.
Dawkins,R.1982 The Extended Phenotype. W H. Freeman, San
Francisco.
Deal, M., and M. Hagstrum1995 Ceramic Reuse Behavior Among the Maya and
Wanka: Implications for Archaeology. In ExpandingArchaeology,editedby J. M. Skibo,W.H. Walker, nd A.E. Nielsen, pp. 111-125. Universityof Utah Press, SaltLakeCity.
Clarification, nd Review of theEvidence o Date.Journal
ofArchaeologicalMethodand
Theory2:299-368.
Duke, P.1995 Working Through Theoretical Tensions in
ContemporaryArchaeology:A PracticalAttempt fromSouthwestern Colorado. Journal of ArchaeologicalMethodand Theory2:201-229.
Dunnell,R. C.1978a Style and Function: A FundamentalDichotomy.
AmericanAntiquity43:192-202.1978b Archaeological Potential of Anthropological and
Scientific Models of Function.InArchaeologicalEssaysin Honorof IrvingB. Rouse, pp. 41-73, editedby R. C.Dunnelland E. S. Hall. Mouton,The Hague.
1980 EvolutionaryTheory and Archaeology.Advances in
ArchaeologicalMethod and Theory,vol. 3, editedby M.
B. Schiffer,pp. 35-99. AcademicPress,New York.1982 Science, Social Science, and Common Sense: The
Agonizing Dilemma of ModemArchaeology.JournalofAnthropologicalResearch 38:1-25.
Schiffer] 659
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 19/21
AMERICANANTIQUITY
1989 Aspects of theApplicationof EvolutionaryTheoryin
Archaeology. In Archaeological Thought in America,edited by C. C. Lamberg-Karlovsky, pp. 35-49.
CambridgeUniversityPress,Cambridge.1992a Is a Scientific Archaeology Possible? In
Metaarchaeology, edited by L. Embree, pp. 75-97.Boston Studies in the Philosophyof Science, vol. 147.KluwerAcademic,DordrechtThe Netherlands.
1992b Archaeology and Evolutionary Science. In
Quandariesand Quests:VisionsofArchaeology Future,editedby L. Wandsnider, p. 207-222. OccasionalPaperNo. 20. Center for Archaeological Investigations,SouthernIllinois University,Carbondale.
1995 What Is It That Actually Evolves? In Evolutionary
Archaeology: Methodological Issues, edited by P. A.
Orser,C. E., Jr.1995 Is There a Behavioral Historical Archaeology? In
ExpandingArchaeology, edited by J. M. Skibo, W H.
Walker,and A. E. Nielsen, pp. 187-197. Universityof
UtahPress,Salt LakeCity.Oswalt,W.H.
1976 An Anthropological Analysis of Food-GettingTechnology.Wiley,New York.
Patrik,L. E.1985 Is There an Archaeological Record? Advances in
ArchaeologicalMethod and Theoryvol. 8, editedby M.B. Schiffer,pp. 27-62. AcademicPress,SanDiego.
Plog, F1974 The Study of Prehistoric Change. Academic Press,
New York.
Plotkin,H. C.1988 Behavior and Evolution.In The Role of Behavior in
Evolution,editedby H. C. Plotkin,pp. 1-17. MITPress,
Cambridge,Massachusetts.
Preucel,R. W (editor)1991 Processual and Postprocessual Archaeologies.
MultipleWays f Knowing he Past. OccasionalPaperNo.10. Center for ArchaeologicalInvestigations,SouthernIllinois University,Carbondale.
Ramenofsky,A. F1995 EvolutionaryTheory and Native AmericanArtifact
Change in the Postcontact Period. In EvolutionaryArchaeology: Methodological Issues, edited by P. A.
Teltser, pp. 129-147. University of Arizona Press,Tucson.
Rathje,W L., and M. B. Schiffer1982 Archaeology.HarcourtBraceJovanovich,New York.
Reid,J. J.
1985 FormationProcesses for the PracticalPrehistorian.nStucture and Process in Southeastern Archaeology,editedby R. S. Dickens, Jr.,and H. T. Ward,pp. 11-33.
Universityof AlabamaPress,University,Alabama.
1995 Four Strategies After Twenty Years: A Return to
Basics. InExpandingArchaeology,ditedby J. M. Skibo,W H.Walker, ndA. E.Nielsen, pp. 15-21. Universityof
UtahPress,Salt LakeCity.Reid,J.J.,M. B. Schiffer,and W. L. Rathje
Sanders,W, J.Parsons,andR. Santley1979 The Basin of Mexico. Ecological Processes in the
Evolutionof a Civilization.AcademicPress,New York.
Schiffer,M. B.1972 Archaeological Context and Systemic Context.
AmericanAntiquity37:148-157.
1975a Archaeology as Behavioral Science. AmericanAnthropologist77:836-848.
1975b BehavioralChainAnalysis:Activities,Organization,and the Use of Space. In Chaptersin the Prehistory ofEasternArizona IV Fieldiana:AnthropologyVol.65, pp.103-119. Field Museum of NaturalHistory,Chicago.
1975c Some Further Comments on Morse's DaltonSettlement Pattern Hypothesis. In The Cache River
Archeological Project. An Experiment in Contract
Archeology, assembled by M. B. Schiffer and J. H.
House, pp. 102-112. Research Series No. 8. Arkansas
ArcheologicalSurvey,Fayetteville.1976 BehavioralArcheology.AcademicPress,New York.1978 Methodological Issues in Ethnoarchaeology. In
Explorations in Ethnoarchaeology, edited by R. A.Gould,pp. 229-247. Universityof New Mexico Press,
Albuquerque.1979 A PreliminaryConsiderationof BehavioralChange.
In Transformations:MathematicalApproacheso Culture
Change, edited by C. Renfrew and K. Cooke, pp.353-368. AcademicPress,New York.
1987 Formation Processes of the ArchaeologicalRecord.
Universityof New Mexico Press,Albuquerque.1988 The Structureof Archaeological Theory. American
Antiquity53:461-485.1991 ThePortableRadio in AmericanLife. Universityof
ArizonaPress,Tucson.1992 TechnologicalPerspectives on Behavioral Change.
Universityof ArizonaPress,Tucson.
1993 CulturalImperativesand ProductDevelopment:TheCase of the Shirt-PocketRadio. Technologyand Culture34:98-113.
1995a A PersonalHistory of BehavioralArchaeology.In
661Schiffer]
7/30/2019 Schiffer 96
http://slidepdf.com/reader/full/schiffer-96 21/21
AMERICANANTIQUITY
Behavioral Archaeology: First Principles, by M. B.
Schiffer,pp. 1-24. Universityof Utah Press, Salt Lake
City.1995b Social Theory and History in Behavioral
Archaeology.In
ExpandingArchaeology,edited
byJ. M.
Skibo, W. H. Walker,and A. E. Nielsen, pp. 22-35.
Universityof UtahPress,Salt LakeCity.1995c Behavioral Archaeology: First Principles.
Universityof UtahPress,Salt LakeCity.1995d The Conceptual Structure of Behavioral
Archaeology. In Behavioral Archaeology: First
Principles, by M. B. Schiffer,pp. 251-253. Universityof
Utah Press,Salt LakeCity.1996 Formation Processes of the Historical and
ArchaeologicalRecords. InLearning rom Things,edited
by W. E. Kingery, pp. 73-80. SmithsonianInstitution
Press,WashingtonD.C.
Schiffer,M. B., T.C. Butts,andK. Grimm
1994 TakingCharge:TheElectric Automobile n America.
Smithsonian nstitutionPress,Washington,D.C.Schiffer,M. B., T. E. Downing,andM. McCarthy
1981 WasteNot, WantNot: An Ethnoarchaeological tudyof Reuse in Tucson, Arizona. In Modern Material
Culture:TheArchaeologyof Us, edited by R. A. Gould
and M. B. Schiffer, pp. 68-86. Academic Press, New
York.
Schiffer,M. B., and J.M. Skibo
1987 TheoryandExperimentn the Studyof Technological
Change.CurrentAnthropology 8:595-622.
1997 The Explanationof Artifact Variability.American
Antiquity62:inpress.Schiffer,M. B., J.M. Skibo,T. C. Boelke,M.A. Neupert,and
M.Aronson
1994 New Perspectives on ExperimentalArchaeology:SurfaceTreatmentsand ThermalResponse of the Clay
CookingPot. AmericanAntiquity59:197-217.
Senior,L. M.
1995 The Estimationof PrehistoricValues: CrackedPot
Ideas in Archaeology.In ExpandingArchaeology,edited
by J. M. Skibo, W. H. Walker,and A. E. Nielsen, pp.92-110. Universityof UtahPress,Salt LakeCity.
Simms, S. R.
1987 BehavioralEcology and Hunter-Gatherer oraging:An Example rom the Great Basin. BAR International
Series 381. BritishArchaeologicalReports,Oxford.
Skibo,J. M.
1992 Pottery Function: A Use-Alteration Perspective.
Plenum,New York.
Skibo,J. M., W.Walker,andA. Nielsen (editors)1995 ExpandingArchaeology.University of Utah Press,
Salt LakeCity.Sober,E.
1984 The Nature of Selection: Evolutionary Theory in
Philosophical Focus. MIT Press, Cambridge,Massachusetts.
Spencer,C. S.1990 On the Tempo and Mode of State Formation:
Neoevolutionism Reconsidered. Journal of
AnthropologicalArchaeology9:1-30.
Sullivan,A. P.1978 Inference and Evidence: A Discussion of the
Conceptual Problems. In Advances in Archaeological
Methodand Theory,vol. 1, editedby M. B. Schiffer,pp.183-222. AcademicPress,NewYork.