Reflections on the Future of Sociology

Author(s): Andrew Abbott

Source: Contemporary Sociology, Vol. 29, No. 2 (Mar., 2000), pp. 296-300
Published by: American Sociological Association
Stable URL: http://www.jstor.org/stable/2654383 .
Accessed: 22/01/2015 11:22

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.

American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to
Contemporary Sociology.

http://www.jstor.org

This content downloaded from 128.235.251.160 on Thu, 22 Jan 2015 11:22:21 AM

All use subject to JSTOR Terms and Conditions
296 Symposia

assignedtoitbythescience ofmarkets.
Itwould Dowsett,GaryW. 1996.Practicing
Desire:Homosexual
be a specialscienceofmarketfailures,
explain- Sex in theEra ofAIDS. Stanford, CA: Stanford
ingthepoor, thedeviants,
theunsuccessful,
and UniversityPress.
theodd.As a democratic science, has Kuletz. Valerie. 1999. The Tainted Desert:
sociology
Environmental
andSocialRuinintheAmerican
West.
thepotential tobecomea strategic
starting
point NewYork:Routledge.
forthecreation ofa newworldculture. Wadsworth,Yoland. 1983. Do It Yourself Social
Research.Melbourne:VCOSS and Allen &
References Unwin.
Connell,R. W. 1997."Whyis ClassicalTheory
Classical?"
AmericanJournalof Sociology102:
1511-57.

Reflectionson the Futureof Sociology

ANDREWABBOTR
Universityof Chicago

Therearetwoaspects tothefuture ofsociology: Thesystem works because department facul-

socialstructural andintellectual.On onehand, tymembers, intheir capacity as agents fortheir
weaskwhether sociology willsurviveas a disci- universities, guarantee thesurvival oftheir disci-
plineatall;ontheother, weaskwhether there plinesbyhiring almost exclusively from those
arenewideasto inhabit thatstructureifitsur- disciplines. Departments arethusthestructure
vives. bywhich academic disciplineswiththousands of
Socialstructurally, wecanthink aboutsoci- members control many oftheresources ofinsti-
ology's future as unfoldingin twocontexts: on tutions withmillions ofstudents. Thewholesys-
onehand,within thesystem ofuniversity edu- temisdriven bytheundergraduate curriculum,
cation;on theother, within thatofacademic which fornearly 100years (sincethephenome-
disciplines. (I realizethattherearesociologistsnonofdepartments emerged) has beenorga-
incommercial orgovernmental contexts, butin nizedintomajors, a curricularform towhich no
theUnitedStatessociology isdominated byan serious alternative haseverbeenpresented and
academic labormarket andthetraining institu-whichisoften littlemorethana focused setof
tionsthatsupport thatlabormarket.) ulstrltutlon requlrements.
.. .. . .

TheAmerican systemofacademic disciplines Eventhesizeofgraduate programs isstrongly

is a peculiar andpeculiarly resilient
structure.driven bymajors, sincea crucial iflatent func-
Current appearances to the contrary, it has tionofmostgraduate programs is to generate
endured withlittle substantial
change fornearly cheapteaching laborso thatuniversities don't
100years. (Interdisciplinarityturnsouttobe as havetohireso many regular faculty.As oneof
old as disciplinarity; the Social Science thelong-standing inhabitants ofthiscurious sys-
Research Councilwas founded in the early tem,sociology is largely safeas longas the
1920sto promote interdisciplinarity,
lessthan underlying structure holds.As longas majors
20 years after thefounding oftheASA.) The exist,sociology departments willexist.They
extraordinary stabilityofthedisciplinarysystem maygetcombined withanthropology orsocial
arises from thebasket-weave structure
ofacade- workdepartments occasionally,buttheywon't
mia,whichis organized intodepartments that disappear altogether. Asforthemajor system as
arebothpiecesofuniversities andpiecesofdis- a whole, itseemsunlikely todisappear; creden-
ciplines. No single universitycanradically mod- tialism isstronger thaneverinhigher education,
ify its departmental structurewithout andwhataremajors ifnotcredentials? (Indeed,
undercutting theemployability ofitsPh.D.s.At double majors areontherise, apparently forcre-
the sametime,no singlediscipline can be dentialist reasons.) Butthestrong drift toward
destroyed unlessa largenumber ofuniversitiesvocationalism, propelled bystudent demand, has
decide simultaneously togetridofit.Evenifone itsdangers forsociology. In thefirst instance,
discipline weredestroyed, the system would vocationalism meansstudents willcontinue to
endure; theotherdisciplines wouldjustfillin choosesociology majors, sincetherearejobs
thegap. thatthey think a sociology degree isgoodfor.

This content downloaded from 128.235.251.160 on Thu, 22 Jan 2015 11:22:21 AM

All use subject to JSTOR Terms and Conditions
 Symposia 297

Buttotheextent thatthereisexpansion in Thatsociology is,more thananyother social

explicitlyvocational degrees aimedatthose jobs science, a fairly loosecollection ofthings has
(BAsinsocialwork, criminology, public policy, other important implications forthediscipline's
and communications, forexample), sociologyfuture, to whichI cannotdevotemuchtime
willbeintrouble. Thisisparticularly truewhen here.On one hand,it givesthe discipline
theseprograms expand outside sociology depart- tremendous resilience and,moreover, a great
ments, rather thanwithin themas certificateability to shelter revolutionary work.On the
programs. Buteventhelatter casepresents diffi- other, itmeans thatthebadwork insociology is
cultiestotheextent thatthetailsbegintowag theworst inthesocialsciences andthat,as we
thedog.Notethatit doesn'tmatter thatthe havejustseen,thediscipline canbenickel-and-
wholeprocessis based on a questionabledimedto death.So the riseof substantive
premisethatthecollegemajorhas a strong majors, thedrift oforganizations studiesback
connection with whata student doesasanadult. intobusiness schools, thepressure ofcredential-
It iswhatthecollegestudents think thatmat- ismandsimilar processes: Theseall havetheir
ters,notwhatlifewillteachthemlater. worrisome aspects forsociology, worries perhaps
Moreover, ifonehasrunan undergraduate stronger thanforanyother socialscientificdis-
sociology program, as I have,one knowsthat clpllne. . , .

newmajors focused onsubstantive socialgroups Allofthisstands instriking reliefwhenone

(American studies, women'sstudies,ethnic putssociology in itsinternational context.For
studies,andso on) cutdirectly intosociology'selsewhere the majorsystem and the idea of
constituency, whatever onemaythink ofthese departments area postwar borrowing from the
newfields on substantive grounds. At present,UnitedStates,andforthatreasonmuchless
mostofthemdon'thavewidespread Ph.D.pro- firmly entrenched. Undergraduate curricula are
grams. Ifthey did,sociology wouldbe inserious structured in otherwaysand therearemany
trouble,forthey would thennothavetodrawso more free-standing research sociologistsininsti-
many oftheir faculty from sociology.(Thesame tutes liketheMaxPlanck InstitutesofGermany
thing,ofcourse, holdstrueforvocational fields orinfunded research networks liketheCentre
likecriminology, communications, andso on, Nationale deRecherche Scientifique ofFrance.
many ofwhicharewellalongtoward supplyingManymoresociologists bothacademicand
theirownfaculty ona "disciplinary" basis.) nonacademic arecloselytiedto government
In bothcases vocational majorsandsub- agencies, andhencemorelikely to undertake
stantivemajors sociologyis particularlywell-funded andsocially consequential minister-
endangered, among thesocialsciences, because ialresearch. (France's recent citizenshippolicies
itisorganized around an archipelago ofempiri- andschoolreforms havetheir origin inministe-
calquestions: raceandethnicity, work andoccu- rially sponsored sociological work, forexample.)
pations,stratification,population, urban studies, Thismeans thatthecourse ofsociology overseas
organizations, and so on. It is not organizedmaybe quitedifferent fromits coursein
arounda method, as is anthropology around America.
ethnography; noraround a theoreticalsystem, as Atthesametime, itiscertain thatU.S.soci-
iseconomics around choicemodels; noraround ology, likeall academic disciplinesinAmerica,
a concept, as ispolitical science around power. willbecomemoreandmoretiedup withthe
Sociology's methods andtheories area grabbag emerging international discipline, particularly
(thepolitephrase is "multi-paradigm science") withsociology inEurope. Thecoming together
whencompared withtherelative consistency of ofthesetwodifferent sociological culturesmay
anyoftheother socialsciences. Thus,wheneverproduce surprising results. The Europeans are
anisland ofthearchipelago decidestobecome a livingthrough thevastsocialexperiment of
separate principality,sociology hasno obvious European unification, anextraordinary opportu-
wayofretaining dominion. Inthetypical mega- nity forsocialresearch inrapidly changing soci-
bookstore, thesections onwomenXs studies and eties.Withboththenational governments and
African-American studiesare nowfarbigger theEU government muchmore deeply commit-
thanthoseforsociology, eventhough 25 years tedtofinancing socialresearch thanistheU.S.
ago(andeventoday) muchofthework inboth government (whichhassteadily cutitssupport
areaswas/is being donebypeoplewithsociologyforsociology overthelastdecade),European
Ph.D.s. sociology maywellbegintotaketheleadinthe

This content downloaded from 128.235.251.160 on Thu, 22 Jan 2015 11:22:21 AM

All use subject to JSTOR Terms and Conditions
298 Symposia

areaofbread-and-butter empirical work thathas formations ofthelasttwodecades. Itistheprob-

beenAmerica's strength forseveraldecades. lemofsimple datasetsize.As many researchers
Thiswilloccurparticularly if,as seemsto be know(aboveall,thoseworking incommercial
presently happening, moreandmoreEuropean venues), wehavecometoa pointwhere there is
sociology getswritten inEnglish andhencecan- nolimit onthesizeofpotential datasets.Bysize
notbeignored byAmericans whodon'treadfor- hereI do notmeanmerely sizein numbers of
eign languages.(All the same,academic cases,although that,too,isvirtually limitlessat
publishing remains anastonishingly localaffair. thispoint. Rather,I meansizeinterms ofnum-
Ifoneweretojudgebytheshelves ofBlackwell'sbersofvariables andnumbers ofrepeated mea-
bookstore inOxford, onewouldthink sociologysures ofthosevariables. Storage space,longthe
a largelyBritish discipline,
despite thecommon limiting factor,nolonger presents anyproblem;
language.) thepowers ofdata-gathering technology canbe
In sum,internationalization wouldseemto allowed freerein.
bethemainsocialstructural eventinsociology's As moreand morebehavior is conducted
future.IntheUnited States,
there willbemuch electronically, moreand morethings can be
talkof interdisciplinarity, butin practice the measured moreandmoreoften. Datasetsnow
majorsystem willrunon andon,departments exist,
forexample, thathavestockprices across
willproduce andhireinPh.D.sinsociology, and an entire market everyminute forsubstantial
thediscipline willlosea fewsubstantive fields periods.
Forsocialdata,thelinking upofgov-
hereandtheretoother majors anddisciplines.ernment databasesviauniqueidentifying num-
Withgoodluck,itsstructural incohesion will bersmakesthe equivalently comprehensive
makeittheplacewhere thesocialscience ofthe socialdatasetsmerely a matter oftime, oncethe
futureemerges. Butthemainstructural eventof privacy concerns getshouldered aside,as they
thenearfuture forU.S. sociology willbe the nodoubtwill.Suchdatasetsarealready begin-
contactwiththenewlyrobust sociologies of ningto becomeavailable, linking educational
Europe, withtheirheavyfunding, theirquasi- systems, justicesystems,mental healthsystems,
experimental analytic object,theirpowerfulandfamily servicesandother socialwelfare sys-
socialdatabases, andtheirrelativefreedom from temsintocomprehensive single databases.The
thehardwork ofteaching thatistheforeground equivalentcommercial databases,identifying
ofdailyexperience formostAmerica sociolo- most oftheAmerican populace comprehensive-
gists.Once theywritemostly in English, the lythrough creditcardnumbers andnetworks of
Europeans maywelltaketheleadership thathas purchases, are alreadyin widespread use.
hitherto eludedthem.(Thisis notto portrayBasically, wearebeingconfronted withcontin-
European sociology asbeingofa piece,whichit uous-time, population-level
data.
certainly is not,but merelyto arguethat Thesinglemostimportant challenge facing
Americans willsoonbe unableto disregard it theempirical socialsciencesinthenext50years
thewaythey haveinthepast.) istheproblem offindingpatterns insuchmonu-
The intellectual future of sociology seems mentally detailed data.Andthebluntfactis
muchmoreunclear. Thesingle mostimportantthatsociology is woefullyunprepared to deal
unclarity is so largethatonecanhardly think withthisproblem: Wehaveneither theanalyt-
seriouslyaboutit.Thereislittle question thata ic toolsnortheconceptual imagination neces-
gradual revolution inthenature ofknowledge is sary.Ourstock-in-trade analytic methods were
takingplace:a sloweclipsing ofprint byvisual designed forinvestigating relations between
representation, a movetoward knowledge thatis smallnumbers ofvariables andareuseless for
more experimental andevenaleatory, anexten- large-scale pattern-recognition or,as we have
sivecommodification ofimportant parts ofpre- pejoratively labeledit,datadredging. Ouridea
viouslyesoteric knowledge. Theimplications of ofdatadredging ishaving graduate students run
thesechanges aretoo vastformyreflectionsthousands ofregressionspredicting onevariable
here.Butevenwithout takingaccount ofthis thatwehappen tohavedecided toexamine. At
greatshift inknowledge, itispossibletospecify thatrate,itwouldtakeusa century tobeginto
threeorfour major intellectual
challenges facing think aboutanyoneofthese giant newdatasets.
thediscipline inthenextquarter-century. Norisitjusta matter oframping upexisting
Thefirst intellectualchallenge forthesocio- methods. Wehavetorethink dataanalysis from
logical
future flows from thetechnological trans- theground up.Intheshort run, wearegoing to

This content downloaded from 128.235.251.160 on Thu, 22 Jan 2015 11:22:21 AM

All use subject to JSTOR Terms and Conditions
 Symposia 299

havetojettison theideaofcausality thathasled do almost noneofa's,b's,c's,andd'sthatlate

ustodenigrate precisely theanalytic toolsnec- nineteenth-century bookkeepers did,butthey
essary toaddress theproblems ofhugedatasets. areinsomesensestillbookkeepers), where now
We havetogiveupthefutile questforeffectsweignore thediscontinuity. Theproblem ofcat-
"netofother variables" andwallow intheend- egory changewithtimeis something we are
lessmultiplexity ofdata.We haveto entera going tohavetotheorize andembody informal
world ofiterative pattern-recognition, ofsimula- conceptions ofmeasure.
tion,of MonteCarlooptimization. It is a Intellectually, thisproblem ofhistory is the
methodological world thatwilldrawheavily on secondgreatchallenge formodern sociological
computer science, onanalgorithmic andaleato- methodology. Itremains unresolved, despite the
ryapproach to knowledge. And it is a world workof thehistorical sociologists one might
completely foreign toourmethods courses.We haveexpected toaddress it.Muchdistinguished
haveinthepastsimply ignored thevastness of historical workhas been done in sociology
data.Wetalkabout"finding theright variable,"recently. Butithasshiedawayfrom developing
butinreality wehavealways hadthousands of a truly historical theory ofthesocialworld, and
variables tochoosefrom andnosensible wayto hasshown littleconcern forthehistorical nature
makethechoice.Thereresults inourliteratureoftheempirical datathatconcerns so many
theamusing spectacle ofone indicator being SOC10 OglStS.
usedto indicate dozensofdifferent things in Thisshying awaymakes a usefultradition to
dozens ofdifferent articles.In 50 years, people mythird great challenge. A central intellectual
willviewtheseactivities thewaywenowview taskforthediscipline istorebuildgeneral social
thepeoplewhopagedthrough sheaves oftwo- theory. Sociology needsa bignewtheoretical
waycrosstabs. idea. (Aren'tyou boredwithWeher?with
A secondbasicintellectual challenge arises Durkheim? maybe evenwithMarx?) Twothings
from thesheer passage oftime.Wenowpossess nourish great theory: priortheoryandreflection
important andsubstantial socialdatasetsthat abouttheempirical world.Inmyview,wehave
haveexperienced realhistorical changein the today fartoomany theorists
whodonothing but
nature ofthecodedcategories. Notsimple drift thinkaboutpriortheory. We do havedistin-
in theusageofwords, butactualdrift in the guished theorists whohaveemerged invarious
nature ofthethings denoted, in bothcontent subdisciplines, peoplewhotry togivetheoretical
andstructure. Themost familiar examples come shapetoempirical literatures,
oftenaftermaking
from thecensus, where occupational andethnic theirownmajorcontributions to thoselitera-
classifications havechanged withregularity over tures. Indeed, wehavetoofewofthesepeople;
theyears. Astheepochofelectronic datagath- manya subfield coulddo witha revolutionary
ering continues, moreandmoredatawillhave synthesis ortwo.Butevensuchsubdiscipline-
thisquality oftruehistorical change, constitu-generated theory as wedo havedoesnothave
tivechange intheunderlying A cen- theambition
categories. andscopetobecome general social
tralchallenge forthefuture ofallsocialresearchtheory.
is to figure outhowto handlethiscategory Important general theory alwaysgrows outof
changewithout simply sweeping it underthe extensive empirical work;every great sociologi-
g. cal theoristhasbeena datahead. So itis likely
Anexample ofthissweeping under therugis thatournewgeneral theory willcomeoutof
thebrave butwillfully ahistorical efforttocreate responses totheproblems discussed above the
a singlesetofoccupational categories forthe problem ofthemassive datasetandtheproblem
PUMSsamples from1880to 1980,an effortofhistorical change. Butnotonlyoutofthem.
soontobe expanded, withequalanachronism, ForI haveleftouttwoothercrucial aspects in
tothe"historical" statisticsof20 OECD coun- themixofsociology.
tries.
"Bookkeepper" justdoesn't meanthesame Thefirst oftheseis culturemeaning-mak-
thing in1880asin1980.Norcanweevenimag- ingactivity. Thereis a delusion abroadthat
inea listofmicro activities interms ofwhich we meaning-making activity cannotbe theorized
couldcreate time-transcending categories(i.e.,a except through thefoggy fantasies
ofthefamous
bookkeeper iswhoever doesa andb andc andd) French. Thisiswrong. Coming toa serious the-
because thenwewouldignore a hugeamount of oryofmeaninga worked-out theory worthy of
continuity (latetwentieth-century bookkeepersCassireror the earlyBarthesor Kenneth

This content downloaded from 128.235.251.160 on Thu, 22 Jan 2015 11:22:21 AM

All use subject to JSTOR Terms and Conditions
300 Symposia

Burkeiscrucial forsociology.Butthattheory sincetheclassical theorists

wrote.Thepolitical-
willworkonlyifit is melded together witha economy agenda thatpreoccupiedthem hasvir-
socialstructuraltheorycapable ofsustainingthe tuallyvanished from America's publicpolitical
methodological advances I callforabove.Not discourse, withitssleepwalkingdevotiontoeco-
an easyjob,butanybody whowantsto think nomicgrowth. Meliorationis nowbeingpub-
seriouslyabouthowtounderstand modern con- liclydefinedinterms ofmeliorationbygroups, a
sumption hastodo it,as every marketing con- modusoperandi thatmovesus rapidly back
sultantknows. towardDurkheimian corporatism.
Dittoforanybody whowantsto intervene It isnotat allclearwhatsortoftheory will
seriously inpolitics.
Modern politics
isas much seriously bringtogether analysis
ofsocialstruc-
aboutthemanipulation of meanings as it is tureand culture in a waythatcan support
abouttheproblematics ofstructure. To think melioristic argument and willpermit, indeed
politicallyweneedtorecast ourtheoriesofboth evendrive, therevolutionindataanalysis.Butit
culture andstructure. Andsociology as a disci-
willnotbe another rerun,as so manyrecent
plinehasalways beenpoliticalandmeliorist to
itsboots from Albion("MarxistheGalileoof sociological theorieshavebeen.Itwillbea hard-
the socialsciences")Smallforward through fought,obscure vision,notclearevento its
Ernest ("I thinkwhatisgoingon intheSoviet authors. Butitwillgiveusa waytothink about
Unionis quiteinteresting") Burgess through politics,
and culture,and gigantic
data sets,and
labeling theory'sloveofdeviants tothestrongly realhistorical
change. Likeall greattheory, it
left
stance thatdominatesthediscipline'sassoci- willemerge from wrestlingwith the empirical
ationtoday.Anysociology ofthefuture will problemsthathavealways beensociology'scon-
have to make a place as Marx did, as cerns:education, health,occupations, inter-
Durkheim did,as Deweyandtheother pragma-grouprelations, stratification,
and so on. To
tists
did forsocialamelioration. buildthisnewgeneral approach is thedisci-
Thiswillbe difficultbecausethepolitics of pline'sproject
fortheimmediate future.
Therest
melioration seemsto havechangedradicallybelongs tothepast.

Simplicity,Uncertainty,
and the Power of GenerativeTheories
LYNNSMITHLOVIN
Universityof Arizona
I confessthatmyscientific heroisnota sociolo- takenphysicists days,months, orevenyearsto
gist,butphysicistRichard Feynman.' Thereare workout (becausetheyhad to tracethem
manywonderful things aboutthisgreatman's through intricatetimeevolutions).Something
career (many ofwhich I cannot understand), but thathadbeenincredibly complex wasmadesim-
therearea fewfeatures ofhislifeandthoughtple(tophysicists, atleast).Feynman wasfamous
thatI wouldliketouseasguides forhowsociol- formaking thingssimple.His lecturesto an
ogistsshould approachtheir scienceinthetwen- introductory physicsclassat Cal Techarenow
ty-firstcentury.
Feynman wontheNobelPrize required readingforanyprofessional physicist
in 1965for(amongotherthings) developing(Feynman, Leighton,andSands1963),because
Feynman diagrams.Thesediagrams aresimple they aresuchclearstatements ofwhatweknow,
pictures withsquiggly linesthatmanaged to whattheimportant questionsare,andhowsci-
express insimple graphs somekeyprocesses in entists goaboutanswering thosequestions.
He
quantum electrodynamics thathad previouslyalsoisfamous fordumping theO-ringmaterial
inicewater duringtheChallenger investigation,
cutting through moundsof pseudo-scientific
I wantto thankmyhusband, MillerMcPherson,obfuscation witha simpleobservationthatthe
forintroducing
metoFeynman's ideaslongbefore stuffgotbrittle
whenitwascold.
Nova andFeynman's humorous autobiographical
books(Feynman 1985,1988;Sykes1994)made I willspendmostofthisessayarguing fora
hima popularfigure.
Having a spousewhostudies goal ofsimplicityin ourexplanationsofsocial
physicsandmathasa hobby hasshaped myviewof life, andinoutlining a generalapproachthatI
scienceinmanyways reflectedinthisessay. think willleadusthere. Mostoftheexamples

This content downloaded from 128.235.251.160 on Thu, 22 Jan 2015 11:22:21 AM

All use subject to JSTOR Terms and Conditions