What Is Political Theory/Philosophy?

Mark E. Warren

PS: Political Science and Politics, Vol. 22, No. 3. (Sep., 1989), pp. 606-612.

Stable URL:
http://links.jstor.org/sici?sici=1049-0965%28198909%2922%3A3%3C606%3AWIPT%3E2.0.CO%3B2-O

PS: Political Science and Politics is currently published by American Political Science Association.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/apsa.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Wed Nov 28 17:22:53 2007
Features
What I s Political
Theory/Philosophy?
Mark E. Warren
Georgetown University
The subdiscipline of political theory and
politrcal philosophy continues t o suffer
some misunderstanding within political
science as a whole. Notwithstanding its
renaissance in the last decade, political
theorylphilosophy is still too often charac-
terized in terms that obscure its roles and
functions within the discipline. Political
theorylphilosophy is often referred to as
"normative theory" as a way of distin-
guishing its concerns with values from the
"empirical theory" and research of politi-
cal science proper. Where the concerns of
political theorylphilosophy are not just
normative, political scientists often charac-
terize it as "speculative theory" to distin-
guish it from "empirical theory" that can
be confirmed or refuted by reference to
observable data. The terms of such a dis-
tinction suggest, of course, that theories
grounded in certainties can and ought t o
replace the speculative approaches of tra-
MARK E. WARREN
ditional political thought. And because
political theorylphilosophy relies heavily on
the history of political thought, many see it
as part of history and the humanities
rather than political scienceinteresting
and necessary for any culturally literate
person, but fundamentally distinct from
contemporary political research.
Terminological distinctions such as these
solidified in an era in which behavioralist
agendas shaped subdisciplinary bounda-
ries, and they reflect the influence that
neo-positivist views of explanation once
had within the discipline. Since that time,
however, we have developed more
sophisticated understandings of how
theory and philosophy relate t o empirical
research. These developments have in
turn affected the way we understand the
explanatory concerns of political science,
while also expanding political theorylphi-
losophy beyond its traditional boundaries.
New assessments of the relation be-
tween theorylphilosophy and explanation
have in part been stimulated by close at-
tention t o philosophy of science over the
last couple of decades. The behavioralist
agenda deserves some credit for this at-
tention, since it sought criteria of scientific
authority in positivist epistemology-that
is, the view that explanatory meaning
depends entirely on reference to observ-
able~.Positivism, however, has long been
superseded by other accounts of explana-
tion. Whatever their many differences, all
schools of philosophy of science agree that
explanatory meanings are underdeter-
mined by observables. Explanation
depends t o a much greater extent than
the positivists appreciated on conceptual
relations and assumptions internal t o
theory, and thus on the kind of systematic
conceptual analysis traditionally practiced
by political theorylphilosophy. For this
reason political theories and philosophies
have what I shall call a meaning-constitutive
function within all explanations. This func-
tion is increasingly recognized within politi-
cal science, and this is no doubt one reason
why the post-behavioralera has coincided
with a resurgence of interest in political
theorylphilosophy.
At the same time, political theorylphi-
losophy has changed dramatically in the
last couple of decades. Whereas it used t o
PS: Political Science G Politics
 What Is Political TheoryIPhilosophy?

be mostly the history of political thought, atory, all explanatory theories involve, and
today it includes a mixture of conceptual, are partly determined by, the philosophi-
linguistic, and normative analysis, "grand" cal presuppositions that are essential to
theories of society and polltics (which are their explanatory power--an issue t o
which I shall return.
Accordingly, I suggest that we reserve
the term political theory (in contrast t o
political philosophy) for those d~mensionsof
conceptual schemes that select and orga-
nize informat~onabout the political world
for explanatory purposes-for example,
neo-marxlst theories of the state or
+.Gu
rational choice models of decis~onmaking.
Political scientists have always appreciated
that theories such as these suggest signifi-
comlng back t o life in new forms), and phi-
cant problems and hypotheses, as well as
losophy of soclal science. As a subdisci-
pline, it is broader, more eclectic, more provide common languages and conceptu-
al tools. What has changed in the post-
sophisticated, and more sensitive to ex-
behavioral era is our appreciation of the
planatory concerns than it was just two meaning-constitutivefunctions of theories.
decades ago. But we have been less suc-
It is now an accepted tenet of the philos-
cessful in developing alternatives t o the
ophy of social science that explanatory
neo-positivist terms we use to character-
meanings of terms are interdependent
ize the relations between political theory1
philosophy and explanation, and thus the within a body of theory. There are, of
role of political theorylphilosophy within course, many different versions of how
political science. Our failure leaves us sub- this occurs and what its implications are.
ject to the tacit blinders of terminological
distinctions. The distinctions I offer here as
alternatives to those with a neo-positivist
genesis suggest one way of more accurate-
ly depicting the functions of political
theorylphilosophy. They also suggest why
explanations of the political world neces-
sarily involve the diversity of theoretical
and philosophical concerns that we are
now seeing within the discipline.

Political Theory

To begin with, it is useful to recall a pre-

positivist distinction between theoretical
and philosophical problems: although
closely interrelated, we need t o recognize
their differences so as not t o confuse philo-
sophical issues with those of explanatory
theory. In the way I shall use the terms
here, theories are about things that empiri-
cally exist, even if these th~ngsare them-
selves ideas, values, and theories that are
part of the political world. Philosoph~cal
concerns have to do with conceptual pre-
suppositions and judgments that are em-
bedded in explanatory theories. Although
philosophical analysis is not directly explan-
At the very least, however, the point im-
plies that in add~tionto care in specify-
Ing empirical referents (a legacy of behav-
ioralism), political scientists need t o
develop a greater awareness of how their
theories constitute thelr problems and
even their find~ngs.Explanation, as always,
requires that we distinguish empirical from
theoretical questions--something posit~v-
ists rightly insisted upon. But it also re-
quires us to Interrelate both dimensions of

September 1989
Features
meaning-a task pos~tivistsfailed to recog-
nize as a problem because they held that
the meanings of theoretical terms are re-
ducible to their emp~ricalreferents.
A first step In recognlzlng the relat~ve
autonomy of theory IS to not~cethat we
rarely use theor~es as representat~onal
"maps" of the pol~t~cal world at all, even
though this metaphor dom~natescommon
understand~ngs More often, the explana-
tory powers of theor~esare ~nd~rect In a
way that prov~desthem w ~ t ha meanlng-
const~tut~ve d~mens~onwe use them as
l ~ m ~ t ~ cases
ng and counterfactuals t o
reduce the complex~ty of the pol~t~cal
world so ~tm~ghtbecome a d~screteobject
of study. We dec~det o reduce complex~ty
In one way rather than another for reasons
that are, more often than not, normat~ve
This IS the way ~t should be, since th~s1s
how we characterize certa~nd~mens~ons of
the world as problematic--say, i t s effi-
clency, just~ce,d~str~but~on of power, or
v~olence-and thus worthy of further In-
vestlgatlon. Models of democracy work In
th~sway, as do rat~onalcho~cemodels,
models of organ~zat~onalstructure,
theor~esof development, confl~ct,world
systems, and so on The normat~vecom-
ponent of lhm~t~ng cases and counterfac-
tuals IS a problem only ~fone uses them as
~f theor~es were representat~onalmaps
rather than meaning-const~tut~ve dec~s~ons
of a community of pol~t~cal scient~sts.One
can avo~dthese m~stakesonly by making
the theor~esthemselves Into objects of
study Only then can one defend the con-
st~tut~ve meanlngs and normat~ve~mplica-
t~onsthat ~nev~tably follow from them
with interpretation t o include ord~nary
A second role of pol~t~cal theory
In ex- language philosophy, semiot~cs,phenome-
planation whose importance IS ~ncreasingly
recognized is that it deals w ~ t hthe concep-
tual coherence of interpretative schemes
that (empirically) enter into polltical ac-
tions through actors' understandings and
uses of the terms of polltical d~scourse.Ex-
amples would be analyses of political ~deol-
ogies, cultures, rhetoric, diplomatic docu-
ments, as well as everyday discourses as
they come t o bear on the political world.
To take a simple example, the act of vot-
ing is caused in part by the way actors
understand the normative significance and
political function of voting w~thina demo-
cratic system. For this reason, the intell~gl-
bility of voting requires something more
than observation: it requires (conceptual)
interpretation of an emp~rically-existing
universe of discourse in which are embed-
ded understandings about democratic rep-
resentation-or indeed, whatever other
discourse happens to impact on voting.
These "textual elements of polit~csare
"
an intrinsic part of the political world
because they conceptually orient individ-
uals
.- towdrd
-- collective decisionmaking.
They have a causal force that is irreduc~ble
to perceptions, attitudes, opinions, or
norms that have been transformed into
discrete bits of data because actors are in-
fluenced by the internal coherence of their
conceptual universe. This dimension of
polltical life is accessible only through inter-
pretations of meaning structures, that is,
through the methods traditionally em-
ployed in the humanities. Many political
theorists have been meet~ngthis challenge
by expanding their traditional concerns
PS: Political Science G Politics

nology, hermeneutics, and interpretive
sociology. A less obvious examp!e of an in-
terpretive approach is rational choice
theory: it views the political world as
made up of conceptually coherent inten-
tions that formulate instrumentally rational
action or~entations.What dist~nguishesra-
tional choice theory from other Interpreta-
tive approaches is a confusion about the
status of its methods: ~tconstitutes the do-
main of intentions by means of axiorcs
rather than treatlng intentionality as an In-
terpretive problem. Thls confusion has
allowed rational choice theory t o inherit
positivist expectations for a definitive
"empirical theory" that would displace
traditional political theory.
Political Philosophy
We mlght reserve the term pol~trcalphi-
losophy (as opposed t o pol~t~cal theory) for
concerns that are not lmmedlately explan-
atory. Polltlcal phllosophy typically In-
volves questlons havlng t o do wlth the
conceptual presupposltlons of theoretical
orlentatlons, as well as questlons of judg-
ment about truth and value A common
misconception In poiltlcal science IS that
most questlons of polltlcal phllosophy are
about normatlve Issues; hence ~ t soverly
narrow ldentlfication wlth "normat~ve
theory." In fact, the problems of polrt~cal
phllosophy fall into three dlstlnct, although
interrelated, categories of analysls and
judgment, only one of whlch IS normatlve.
Ontological questions: Some of these
problems are ontological The term ontol-
ogy refers t o the sclence that investigates
September 1989
What Is Political TheorylPhilosophy?
the nature and fundamental properties of
reality. We no longer believe we can carry
out such investigations because we no
longer presume intellectual access t o real-
ity as such. For good reason ontology as a
"science" has fallen out of favor. None-
theless, the term has been resurrected in
polltical philosophy for sllghtly different
purposes: whether reality is knowable or
not, we unavoidably make general as-
sumptions about the nature of the reality
we are investigating. These might properly
be called ontolog~calassumptions because
they are logically prior t o any explanatiori
and serve as its conditions of possibility.
Although necessary, they cannot be em-
pirically investigated because they charac-
terize general properties of the world
we seek to investigate. Most ontological
assumptions are so fundamental that we
take them for granted, not understanding
how they frame, select, and limit possibili-
ties for explanation and judgment. All em-
pirical research presupposes that some set
of features of the human condition--such
as consciousness, language, scarcity, tem-
porality, causal determination, lawlike
regularity--defines a political world as a
knowable object of study. Decisions t o In-
clude some features of the human condi-
tion and exclude others are not trivial:
they constitute disciplinary domains by
defining objects of explanation. For exam-
ple, the behavioralist claim that only
observables that can be ordered into logi-
cal associations count as a (knowable) part
of political reality produces one kind of dis-
ciplinary domaln. The rational choice pre-
supposition that politics is an effect of in-
strumentally rational actions produces a
second kind of domain. The Weberian
focus on intentional actions molded by dis-
Features
tinctive cultural possibilities produces a
third. Those who see social and organiza-
tional structures as irreducible t o behavior
or intentional action produce still another
definition of what features of the human
cond~tionmakes the political world possi-
ble.
Each of these ontological assumptions
about the nature of the political world
guides and limits what is t o count as an
"explanation. " For example, different an-
swers t o the (ontological) question, "Is
voting a behavior, an action, or a struc-
tured manifestation of social interaction?"
will dictate different theoretical approach-
es and criteria of adequacy. Voting behav-
Ion simply need t o be observed and their
regularities theoretically identified. Acts of
voting need t o be observed and concep-
tually understood as part of a cultural sys-
tem within which "voting" involves an
assignment of meaning by the actor in a
way that partly accounts for the act. A
structural analysis would require that one
postulate non-observable entities-a
"class structure" or a "state," for exam-
ple-that influence both behavior and in-
tentional orientation. Ontological decisions
such as these relate closely t o the explana-
tory status one gives t o concepts. Are
there really things called "institutions" or
"class structures" or "cultural systems"
or "states"4r are these s~mplyintellec-
tually convenient ways of spec~fy~ng dif-
ferent k~ndsof actions, the~rsituations,
their effects, and their normative slgnifi-
cance? Are linguistic structures reducible
t o the ind~vidualswho use language?Are
soc~alregularities law-l~keor rule-like? If
law-like, do we mean something that
possesses the necessity of physical laws?If
rule-l~ke,then what 1s their causal status In
pol~t~cal life glven that rules can be broken?
Many methodolog~caldebates about the
explanatory powers of behavioralism, ra-
tional choice theory, structuralism, herme-
neutlcs, and the l~keturn on these klnds of
questlons
Ontolog~cal decisions determine not
only domain and crlterla of explanatory
adequacy, but also the way one concep-
tualizes the normative possibilities of
pol~t~cs For example, if one conceptualizes
the polit~calworld as be~ngmade up only
of behaviors, or only of instrumental ac-
t~ons,then one has excluded by ontologl-
cal fiat the causal force and transformat~ve
possibilities of language and interaction
This in turn will limit the horizons of politi-
cal possibility without analysis or justifica-
tion
Epistemological questions: A second kind
of question in political philosophy is episte-
rnolog~cal.Such questions have t o do with
the authority of theories with respect t o
the world they purport t o explain. An-
swers t o epistemological questions often
follow from different ontological positions.
Take the example of voting: if voting is a
behavior, then the authority of a theoreti-
cal statement depends on its references to
observations. This is consistent with posi-
tivist epistemology. If, however, voting is
an action, then observations underdeter-
mine explanation. One must also under-
stand "voting" as part of an interpretive
field within which the intelligibility of the
behavior depends in part on the actor's
understanding of democracy. This presup-
poses an interpretive (or "humanistic")
component t o methodology, since the
political scientist must interpret the inter-
pretive field which partly constitutes the
object of explanation. From the point of
view of positivists, interpretive methodol-
ogies are not "scientific" precisely because
interpretations cannot be authoritatively
PS: Political Science G Politics

verified by referring t o theory-indepen-
dent observations. This is why positivists
deny a (knowable) causal status t o inter-
pretive phenomena in the politlcal world,
and seek t o reduce them t o lawlike regu-
larities. In contrast, much political
philosophy today has t o do with articulat-
ing postpositivist accounts of the authority
of social scientific interpretations of inter-
pretatlve phenomena. Recent critical
theory, for example, focuses on criteria of
authority implicit in the intersubjective
dimensions of action.
Normative questions: It is well recognized
that political philosophy deals with a third
category of questions, those having t o do
with normative judgment. The classical
form of questioning in political philosophy
occurs when problems of explanatory
judgment can be distinguished from those
of normative judgment. Thus, assuming
that the relevant aspects of a political do-
main are known, how are they t o be
judged?What are the criteria of judgment,
and how are they related t o fundamental
human values? What modes of political
organ~zationwould maximize these values?
Although normative judgments are logi-
cally dlstinct from explanatory concerns,
they are also closely related by the pecu-
liar way that political science is defined as a
discipline. Whatever other disagreements
there might be about domain, we usually
call something "political" if ~tconcerns col-
lective decisionmaking, where the ques-
tion "what ought we t o do?" is always im-
pliclt. Political science is unique among the
social sciences in that its domain is precon-
stituted by normative questions. Stated
otherwise, such questions are intrinsic to
the possibility of a political science. This
logical formulation has an empirical corre-
late: individuals act politically when they
are normatively oriented toward collec-
tive problems. Political philosophers artic-
ulate thls dlmension of politics when they
reconstruct and analyze the normative
dlmension of political discourse.
Normatlve and explanatory goals are
closely related in other ways as well. As
suggested, we often assign significance to
political research according to normative
judgments. For example, studylng voting
September 1989
What Is Political TheorylPhilosophy?
presupposes that voting is problematic,
which in turn presupposes a normative
theory of liberal-democracy within whlch
voting is a central and definitive element.
In contrast, Marxlst theories downplay
voting because of the normative judgment
that political democracy is diminished in
value without economic democracy; this is
partly why Marxists are interested In the
effects of economic structures on politics.
Somewhat less obvious is the way that
different normative tendencies and possi-
bilitles follow from ontological decisions
that select for some kinds of applications
and exclude others For example, if one
constitutes the politlcal universe as made
up of behaviors, the form of knowledge
one produces will lack connections t o in-
tentlonal and linguist~cphenomena Lack-
ing these connections, it will be relatively
useless-r at best insufficlentfor In-
creasing lndlvidual capacities for choice and
self-direct~on What behavioral research
can be used for is behavlor modification
as, for example, in campalgn use of opin-
ion survey research to tailor medla images
for deslred responses. But such applica-
tions are technocratic rather than demo-
cratlc. Because behavioral forms of knowl-
edge can be more easlly put t o such uses
(rather than, say, locatlng conditions of
public discourse) they produce a blas
toward technocracy and away from
democracy
Features

The alm of political ph~losophyhere, of that we might understand the powers and
course, is t o make such normative judg- l~mitsof different possibilit~es,and gain a
ments into problems that one can treat awareness of our own
cr~t~cal
systemat~cally.At the same tlme, aware-
ness about the interrelat~onsbetween nor- Conclusion
matlve or~entat~onsand research can
guard against "scient~stic"political science The resurgence of political theory and
-that is, research that tacitly confuses philosophy is, in my view, part of the discl-
political or value problems w ~ t hsc~entific pline's recovery from self-~mposedmls-
findings, understandings about the nature of the
polit~calworld, and about how it is possl-
History of Political Thought ble t o know and judge ~ tThe . discipline is
healthier today because these questions
F~nally,a comment may be useful about are no longer marginallzed by an exclusive-
why pol~t~cal theorylph~losophydoes so ly behav~oralagenda. But ~t1s now tlme t o
much of ~ t swork by means of the h~story do away with the term~nologicaldistinc-
of pol~t~cal thought. Class~calsystems of tions inherited from the behavioral era as
pol~t~cal thought-from Plato and Ar~stotle well. These terms suggest indefens~ble
t o Marx and Weber--exempl~fy d~fferent roles for polltical theorylph~losophy,and
k~ndsof answers t o many of the above they continue t o confuse us about what
questions, answers that express central roles, exactly, political theorylph~losophy
stralns In our pol~t~cal culture. Pol~t~cal does play within the d~scipl~ne. The distinc-
sc~ent~sts lnev~tablyrely on the conceptual t~onsand interrelations I offer here--be-
and l~ngu~st~c tools prov~dedby the~rpol~tl- tween theory and philosophy, and be-
cal culture--indeed, ~fonly so that they can tween ontological, epistemological, and
take the commun~cat~on and slgn~ficanceof normative questions--provide one way of
the~rresearch for granted. Cultural aware- replac~ng these terms and explaining these
ness 1s no doubt valuable for ~ t sown sake: roles. They also advocate my own view
how else could we know who the we IS that the current renaissance of political
that 1s defined by a trad~t~on of pol~t~cal theorylphilosophy reflects a growing rigor
d~scourse?But, In add~t~on, study~ngrela- and depth in our understand~ngof political
tlvely comprehens~veand d~scretesystems l~fe
of thought 1s an invaluable means of devel-
oping an awareness about our own pre-
suppos~t~ons and values. The canon of About the Author
pol~t~cal thought 1s close enough so we can
recognlze our own assumptions w~thln~ t ,
but d~stantenough so we can recognlze Mark E Warren IS Ass~stant Professor of
d~scretesets of values, problems, presup- Government at Georgetown Unlverslty. He IS
author of Nietzsche and Poirtical Thought (MIT
posltlons, and m~stakesIt 1s never s~mplya Press, 1988), and currently worklng on a book
questlon of learn~ngand borrow~ngfrom ent~tledDemocracy and the Seif whlch w~ll
past masters, but also one of seelng them examlne relat~onsbetween language, pol~tlcal
as exemplars of the Interdependence of ~nteractlon, and subject~v~tyIn democratic
ph~losophy,theory, and explanat~on,such theory.

PS: Political Science E Politics