Communicative Ethics and
Current Controversies 10
Practical Pllilosophy
Seyla Benhabib


The essays in this volume present the first English collection
on the p'rogram of a "comITIunic3r.ive'.' "discourse" ,ethics
and document the livel y controversy this Idea has led to III the
last decade. Like the Explanation vs. Understanding contro-
versy, the dispute concerning communicative ethics is informed
both by the Anglo-American and Conllnental tradillons of
thought and retiects a provocative ll1teraCllon between the two.
Communicative ethics, as formulated by Karl-Otto Apel and
Jiirgen Habermas, has been inAuenced by the work of such
nloral philosophers as Kurl lla ,er. Alan (,e\Vlnh, H. M. Hare,
Marcus Singer, and Stephen Toul!l1ln on moral reasonmg and
universali zabi lity in ethics.' Above all, however, It IS m Joh:1
Rawls 's T1(": O- Kalltiall cOllstructivislll and Lawrcll tc Kohlbcrg s
cognitive-dcvcloplllelllaimoralthcury that Apel alld Habermas
have found the must kindred projects of moral philosophy 111
the Anglo-American world.' .. . .
The central insight of commumcallve or discourse ethiCS
derives from modern theories 'if autonomy and of the SOCIal
contract, as articulated by John Locke, .Iean Jacques Rousseau,
and in particular by Immanuel Kant. Only those norms and
normative institutional arrangements are valid, It IS c1allned,
which individuals can or would freely consent to as a resultof
engaging in certain argumentative practices .. Apel
that such argumentative practices can be descnbed as an Ideal
community of COllllllunication" (dir iri('(1./{'
lIItillsci/{/(i) , wh ile Hailerm;!s call s Ihelll "pr;! cllcal cl1SCOlirses.

iloth agree, however, that such practices are the only plausible
procedure in the light of which we can think of the Kantian
principle o[ "universalizability" in ethics today. Instead of ask-
ing what an individual moral agent could or would will, without
self-contradiction, to be a uni v':rsal maxim [or all , one asks:
,,·,that norms or institutions would the members of an ideal or
real communication community agree to as representing their
common interests after engaging in a special kind of argumen-
tation or conversation? The procedural model of an argumen-
tative praxis replaces the silent thought-experiment enjoined
by the Kantian universalizabilit)· test.
These essays appear at a point when the mood concerning
neo-Kantian, procedural, and formalistic ethical theories on
both sides of the ocean is probably best captured by the follow-
ing statement of Stanley Hauerwas and Alasdair Maclntyre:
This is not the first time that ethics has been fashionable. And history
suggests that in those. periods when a social order becomes uneasy
and even alarmed about the weakening of its moral bonds and the
poverty of its moral inheritance and turns for aid LO the moral phi-
losopher and theologian, it may nol find these disciplines fluuri.c;i1 i!1 g
in such a wa }' as to be able to make available the kind of Ill oral
reRection and theory which the culture actually needs. indeed 011
occasion it may be that the ver), causes which have led to the impov-
erishment of moral cxpericnce and the weakening of mor",] bonds
",j]] also thclllsel\'es hLlVC conlriLmlcd Lv I.he forlllaLioll of kind ur
moral theology and philosophy which are unabl e LO provide the
needed resoLlrces . .'1
If this statCIJlCllt call ue viewed as a accurate indi catioll
of the Zeitgeist concerning ethical theor), today, as I believe is
tbe case, t hen thi s certainl ), does not bode well for yet another
program of ethi cal uni versali sm and formalism. Such ethical
formalism is cunsidered a part of the Enlightenmelll project
of rationalism and of the political project of liberali sm, and it
is argued that precisel y these intellectual and political legacies
are an aspect, if not the main cause, of the contemporary crisis.
If communi cative or discourse ethics is to be at all credible,
therefore, it must be able to meet the kind of challenges posed
by Macintyre and Hauerwas.
In thi s afterword I would like to acknowledge this challenge
and anlicip;!te a set of objections alld criticisms which can be
Seyla Ben habib
pressed against communi cative ethics from a standpoint which
I will· roughl y describe as "neo-Aristotelian" and "neo-Hege-
li an." Since Aristotle's criticism of Plato's theory of the good
and of the ideal state in his Nic01nacheatl Ethics and Politics,' and
since Hegel' S critique of Kantian ethics in his vari ous writings,'
formali st and universalist ethical theories have been conunu-
ously chall enged in the name of some concrete hi storical-ethical
communit y or, in Hegelian language, of some SzuiIchlw./ .. In
fact, Apel and Habermas admit that one cannot Ignore the
lessons of Hegel's critique of Kantian morality· Whet her they
have successfully integrated these lessons II1t o communIcati ve
et hi cs, however, is worth examining more cl osely.
In recent discussions "neo-ArisLOtelianism" has been used to
refer to three, not always clearl y di stinguished, strands of social
analysis and phil osophical argumentation. Panicularl y in the
German context, this term has been identifi ed .with a neocon-
servative social diagn osis of the problems of late-capitalist so-
cieties.' Such societies are viewed as suffering from a loss of
moral and almost civilizational orientation, caused by excessive
individualism, libertarianism, and the general temerity of lib-
eralism when faced with the task of establi shing fundamental
values. Neither capitalist economic and societal modernization
nor technological changes are seen as basic causes of the cur-
rent crisis; instead pol iti cal liberalism and moral plurahsm are
regarded as the chief causes of this situation. From R obert
Spaema nn to Allan Bloom, thi s position has found vIgorous
exponents today.
The term "nco-Aristotelian" is also frequently used to des-
ignate the position of thinkers like Alasdair MacInt yre, Mi chael
Sandel, Charles Taylor, and Mi chael Walzer, who lament the
decline of moral and political communi ties in contemporary
societies.!'! Unlike the Il coconscr";Il,ivcs. the "COlllmUnilarian"
neo-Ari sLOteli ans arc criti cal of cOlllempQrary ca pitalism and
lechnol ogy. The recovery of "community" need not only or
even necessarily mean the recovery of some f undamentaizst
value-scheme; rather, communities can be reconstituted by the
reassertion of democrati c control over the runaway megastruc-
tures of modern capital alld technology. T he communitarians
share with ncoconscrv;zt i"es the belief, however, that the for-
.. ,
malist, ahistorical, and indi vidualistic legacies of Enlightenmelll
thinking have been historicall y impli ca ted in developments
which have led to the decline of community as a way of life.
Panicularly today, they argue, this Enlightenment legacy so
constri cts our imaginati on and impoverishes our moral vocab-
ul ary that we cannot even conceptualize sol uti ons to the current
crisis which wou ld transcend the "rights-entitiemelll-distribu-
tive justice" trinit y of politicalliberali slll .
Finall y, "neo-ArisLOteliani sm" refers to a herllleneutical
philosophical et hi cs, centered around the Aristotelian under-
standing of phTon.sis. Hans-Georg Gadamer was the first to
turn to Aristotle's model of phro71.csis as a form of contextually
embedded and situalionall y sensiti ve judgment of particul ars'"
Gadamer so powerfull y synthesized Aristotle's ethi cal theory
and Hegel's critique of Kant that after his work the two strands
of argumentation became almost indi stinguishabl e. From Ar-
istotl e's critique of Plato, Gadamer extricated the model of a
situationall y sensitive · practical reason, always functioning
against the background of the shared ethical understanding of
a communit y."! From Hegel 's critique of Kant, Gadamer bor-
rowed the insight that all formalism presupposes a context that
it abstracts from and that there is no formal ethics whi ch does
not have some material presuppositions concerning the self
and social instituti ons, II Just as there can be no understa nding
whi ch is not situated in some historical context, so there can
he no "moral standpoint" wh ich would not he dependent "1'011
a shared ethos, be it that of the modern state. The Ka lltiall
moral point of view is only intelli gible in li ght of the revolutions
of modernity and the establishment of freedom as a principle
of the modern world.
These three strands of a neoconservative social diagnosis, a
politi cs of community, and a phil osophi cal ethics of a hi stori-
call y informed practical reason, form the core elements of the
col1lemporary neo-Aristotelian position. Here I shall be con-
cerned with neo-A ristotelianism less as a social diagnosis or as
a political philosophy but more as a philosophi cal ethics.
Let me now formu late a series of objections to communica-
ti ve ethi cs. Some version of these has been voicecl by thinkers
inspired by Aristotle and Hegel against Kantian-typc ethi cal
Seyla Be nhabi b
theori es at some point or another. My goal will be to show that
these objecti ons have not succeeded in deli vering a coujJ de
grace '(a blow of mercy) to a dial ogicall y reformulated univer-
salist ethi cal theory. A serious exchan ge between such a UI1l-
versalist ethical theory, which su fTers neither from the
met hodological individualism nor from the ahistori cism of
traditional Kanti an ethics, and a hermeneuti call y inspired neo-
Aristotelianism can lead us to see that some traditional oppo-
sitions and excl usions in moral philosophy are no longer con-
vincing. Such oppositions as between universalism and
histori city, an ethics of principle and judgment in context, or
ethical cogniti on and moral motivati on, within the confines of
whi ch much recent discussion has run, are no longer compel-
ling. Just as it is not the case that there can be no historically
informed ethi cal uni versalism, it is equall y not the case that all
neo-Aristotel ianism must defend a conservative theory of com-
munal ethi cs. Here I am concerned to indicate how such false
oppositions can be transformed in to a more fruitful set of
contentions between two types of ethi cal theorizing whi ch have
marked the Western philosophical tradition since its beginnings
in Socrates's challenge to the Sophists and his condemnation
to death by the city of Athens.
I Skepticism Toward the Principle of Universalizability, Is
it at Best Inconsistent and at Worst Empty?
Hegel had criticized the Kantian f(lrImlia, "Act onl y on that
maxim throu gh which you can at the saJll e time will that it
should become a uni versal law," o n numerous occasions as
being inconsistent at best and empt y at worst." Hegel argued
that the test alone whether or not a maxim could be univer-
sali zed could not determine its mora,l rightness, As he pointed
out in hi s earl y essay on Natuml Law, whether or not I should
return deposits entrusted to me is answered in the affi rmative
by Kant with the argument that it would be self-contradictory
to will that deposits should not exist. T he youn g Hegel answers
that there is no contradiction in willing a situation in which
deposits and propert), do not exist, unl ess of course we make
some ot her assu mptions abou t human needs, scarce resources,
distributive justi ce, and the like. Out of the pure form of the
moral law alone, no concrete maxims of acti on can follow and
if they do, it is because other unidentifi ed premises have been
smuggled into the argument
l 3
In view of this Hegelian critique, whi ch continues to inAu-
ence discussions of Kantian ethics even today, J< the response
of Kantian moral theorists has been twofold : first, some have
accepted Hegel's critique that the formal procedure of univ-
ersali zability can yield no determina te test of the rightness of
maxims; they admit that one must presuppose some minimall y
shared conception of human goods and desires as goals of
action, and must test principles of action against this back-
ground. T his line of response has weakened the Kantian dis-
tinction between autonomy and heteronomy by accepting that
the goals of action may be dictated by contingent features of
human nature rather than by the di ctates of pure practical
reason alone. J ohn Rawls's list of "basic goods, " which rational
agents are supposed to' want whatever else they also want, is
the best example of the introduction of material assumpti ons
about human desires into the universali zability argument. The
test of universali zability is not about whether we want these
goods but rather about the moral principles guiding their even-
tual dist ri bution. " Other Kanti an moral phil osophers, and
most notabl y among them, Onora O'Neill and Ala n Gewirth,
have refused to jettison the pure Kanti an program, and have
att empted to expand the principl e of the noncontradi ction of
maxims by looking more closely at the formal /t!atu.>·cs of rational
a.ctiun. O'Neill , for exampl e, distinguishes between "conceptual
inconsistency" and "volitional inconsistency" in order to differ-
enliate alTIong lypes of incoherence in aClion.l{; "The non-
uni versali zed maxim," she writes, "embodies a conceptual con-
tradiction onl y if it aims at achieving mutually incompatible
objectives and so cannot under any circumstances be acted on
with success."17 Volitional inconsi slenc)" by contrasl, occurs
when a rati onal agent violates what O'Neill names "Principles
of Rational I ntending. "" Appl ying universali zabili ty to maxims
of action both to test their conceptual consistency and their
volitional consistenc), avoids, according to O'Neill, "the dismal
choice between tri viality and implausibl e ri gorism,"IY In a sim-
Scyla Benhabib
ilar vein, Alan Gewirth expands an the idea af the "ratianal
canditians a f actian" in such a way as to. generate nantrivial
and intersubjectivcly binding maxims af maral actian from
Bath strategies have prablems: in the first case, by allawing
material presuppasitians abaut human nature and loto
the picture, one ru ns the risk of weakemng the dlsllnCllon
between Kantian and other types of utilitarian or Anstotelian
moral theori es. The result is a certain eclecticism in the struc-
ture af the theary. The secand pasitian runs a different dan-
ger: by facusi ng exclusively an the canditians af ratlanal
intending or act ing, as O'Nei ll and Gewirth do., one can lase
sight af the questian af intersubjective.maral validity. After all,
the Kantian principle af universali zablli ty IS farmulated 10 0.1'-
der to generate ma rally binding maxims af action whi ch all
can recognize. As Alasdair MacIntyre shows III hiS sharp cn-
tique of Gewirth, fram the premise that I as a rallanal agent
require certain conditions af actian to be fulfilled, 11 can never
follaw that yau have an obligation nO! to hmder me from en-
joying these conditions." The grounds far thiS obli gatlan are
left unclear; but it was precisely such graunds that the umv-
ersali zabili ty requirement was intended to produce. Put 10
terms which are thase af Ape! and Habermas, the analYSIS af
the rational structure of action far a single agent praduces an
egologica.l moral theary which cannot justify intersubjective
moral validity. Instcad of asking what I as a slOgle rallanal
maral agent can intend or wi ll to be." universal maxim for all
without cantradiction, the cammunl ca llve ethiCist asks: what
pri nciples of actian can we all recagnize or agree to as being
valid if we engage in practical discourse ar a mutual search far
justificat ion?
With this reformulation, is defined as an
intersubjective pracedure of argu mentatian, to attain
communicative agreement. This reformulatian bnngs With H
several significant shifts: instead of thinking o.f universalizabll-
ity as a test af nOll.contmdiction, we think of umversali zabllity as
a test af communicative agreement. We do. nat search far what
wauld be nonself-contradictof)' but rather far what would be
mutually acceptable for all. Furthermore, there is also a shift
fram the madel of the goal-oriented or strategic action of a
single agent intending a specifi c autcame to the madel of com-
municative action whi ch is speech and action to be shared with
What has been gained th rough thi s reformulatian such as to
counter the Hegelian objection? Have we nat simply pushed
the prablem from ane pracedure a nta another? Instead af
deriving maral principles fram same pracedure af conceptual
ar valitlanal coherence, do we not simply derive them now
fram aur definitian of the conversa tianal situatian' Theorists
can canstruct or design canversatians to yield certain outcomes:
the precanditians af canversatian may guarantee that certain
outcames will resul1.
In an earlier article I farmulated this
prablem as foll aws: either models of practical discourse ar the
Ideal communicatian community are defi ned so. minimall y as
to be tnvlalm their Implicatl ans ar there are mare contraversial
substantive premises guiding their design, and which do. nat
belong amang the miniinal conditions defining the argumen-
tallon slt uallon, III whICh case they are inconsistenl.
We are
back to the "dismal cha ice" (O'Neill ) between tri viality or
IIl conslslency.
I naw believe that the way aut of this dilemma is to opt far
a strong and passlbly cantraversial canstructian of the conver-
satianal model which wauld na net heless be able to avoid the
charges af dagmatism andlor circularity." My thinking is as
follaws: what Habermas has previously namcd the condi tions
af an "i,deal speech situation," and which in the essay "Dis-
course Et hICS: Notes 011 a Program of Philosoph ical Justifi ca-
lIall" are call ed the "universal and necessary communicative
pres uppositions of argumentative speech,"" entail , in my apin-
lOn, strong ethical assumptions. They require of us: ( I ) that
we recognize the right of all beings capable of speech and
action to. be parllClpallls 111 the maral canversatian-I will call
this tlze princil)le of ullivenal moml respect; (2) these conditi a ns
further stipulate that within such.conversations each has the
same symmetrical ri ghts to variaus speech acts, to initi ate new
. tapics, to ask for reflecti on abaut the presuppositians of the
conversatlan, etc. Let me call this the principle of egalitarian
reciproCity. The very presuppasitians of the argumentatian sit-
Se)'la Bcnhabib
uation then have a normative content that precedes the moral
argument itself. But can one then reall y avoid the charges of
circularity and dogmatism'
One of the central disagreements between Apel and Haber-
mas concerns precisely this iss ue of the j ustifi cation of the
constraints of the moral conversation. A pel mamtalllS that:
If, on the o ne hand, a presu ppositi o n cannot be in a.r-
gumentation without actual and If,
on the other hand, it Ollnot be dedll cll vely grounded Without formal-
logical pctitio principii , then it belongs 10 those transcendental-prag-
matic presuppositions of argumentati on that one must (al-
ready) have accepted, if the language game of argumentallon IS to
be mcaningfu1.
'For Apel, the principle that all beings capable of speech and
action are potential members of the same communICa ti on com-
munity with me, and that they deserve equal and symmetrical
treatment are twO such condi tions.
In view of this Apelian strategy of fundamental grounding
or Letztbegrundung, Habermas argues that such a strong JustJ-
li cation of communicative ethi cs c"nnOl succeed and mal' not
even be necessary. Rat her than viewing the normative con-
straints of the ideal communication communi ty as being "dis-
closable" via an act of transcendental self-reAection, Habermas
argues thal we view the m as "universal pragmatic presupposi-
tions" of speech acts corres ponding to th e know-how of com-
petent "moral" agents at the postconventional stage. But as
Thomas McCarthy has pointed out there IS no ulll vocal de-
scription of the "know-how" of moral aCLOrs wh0
!,ave reache?
the postconventional stage of moral reasOl1ln g. ' Habermas s
description of th is know- how is one among many others lIke
those of John Rawls and Lawrence Kohl berg. At the stage of
poslconventional nl0ra1 reasoning, unl versahza-
bility, and impartialit y, under some are all aspects
of the moral point of vie\\', but the real point of phIlosophICal
contemi on is the acceptable or adequatc description of these
forma l constraints. The appeal to moral psychology and de-
velopment brings no exe mpti on from the j ustifi catory process.
Lawrence Kohlber g was wron g in thinkin g that the "ought"
' .
can be deduced from the "is. " The formal structure of postcon-
ventional moral reasoning all ows a number of substantive
moral interpretations, and these imerpretations always take
place by presupposing a hermeneutic horizon.
As opposed to Ape!'s strategy of Letztbeg'rundung and Haber-
mas's strategy of a "weak transcendental argument," based on
the rational recons tructi on of competencies, I would like to
plead for a "historicall y self-conscious universalism." The prin-
cipl es of universal res pect and egali tarian reciprocity are our
philosophical clarifi cation of the constituents of th e moral point
of view from within the normative hermeneutic horizon of
modernity. These principles are neither the only allowable in-
terpretation of the formal constituents of the competency of
postconventional moral actors nor are they unequivocal tran-
scendental pres uppositions whi ch every rational agent, upon
deep. reAection, must concede to. T hese principl es are arrived ·
at by a process of "reAecti ve equili brium" in Rawlsian terms,
whereby one, as a philosopher, analyzes, refines, and judges
culturall y defin ed moral intuitions in li ght of articulated phi lo-
sophi cal principles. What one arri ves at the end of such a
process of reAective equili brium is a "thi ck description" of the
moral pres uppositions of the cultu ral hori zon of modernit y.
At one level, of course, the intuiti ve idea behind universalistic
ethi cs is ver y ancient, and corresponds to the "Golden Rule"
of the tradition-"Do unto others as you would have others do
unto you." Uni versalizability enjoins that we rcverse perspec-
tives among members of a "moral communit y"; it asks us LO
judge from the other's poi nt of vi ew. Such reversibility is es-
sential to the ties of reciprocity that bind human communities
together. All human communities deline some "significant oth-
ers" in relat ion to whi ch reversibi li t y and reciprocity must be
exercised-be they members of my kin group, my tribe, my
cIty-state, my nation, my co.rel igionists. What distinguishes
"modern" from "premodern" versions of uni versalistic ethical
theories is the assumption of the former that the moral com-
munity is coextensive with all beings capable of speech and
action, and potentiall y with all of humanity. In this sense, com-
muni cative ethi cs sets up a model of moral conversation among
members of a modern ethical communi ty, for whom the theo-

I" ,.
Sc)' la Bcnhabib
logical and ontological basis of the inequality among humans
. has been radicall y placed into question.
This is not an admission of dogmatism in favor o[ modernity,
for even this "dogma" o[ modernity, if you wish, can be chal-
lenged within the moral conversation itself. The racist, the
sexist, or the bigot can challenge the principle of universal
moral respect and egali tarian reciprocity within the moral con-
versat ion, but if they want to establi sh that their position is
ri ght not simply because it is mighty, they must convince with
argument that this is so. The presuppositions of the moral
conversation can be chall enged within the conversation itself,
but if they are altogether suspended or violated then mi ght,
violence, coercion, and suppression follow. One thus avoids the
charge of circul arity: by allowing that the presuppositions of
the moral conversation can be chall enged within the conver-
sation itself, they are placed within the purview of questioning.
But insofar as they are pragmatic rules necessary to keep the
moral conversation going, we can onl y bracket them in order
to chall enge them but we cannot sus pend them altogether. The
shoe is reall y on the ot her foot. It is up to the critic of such
egalitarian universalism to show, with good grounds, why some
individuals should be effectively excl uded from the moral
Of course, our moral and political world is more character-
ized by struggles unto death among moral opponents than by
a conversation among them. This admission reveals the frag-
ili ty of the moral point of vi ew in a world of power and vio-
lence, but this is not an admi ssion of irrelevance. Political
ideologies as well as more subtle forms of cul tural hegemony
have always sought to make plausible the continuation of vio-
lence and power to those who most suffe red from their con-
·seq uences. When such ideology and hegemon), no longer serve
to justify such r elations, then struggles unto death fo r moral
recognition can follow. As a critical social theorist, the philos-
opher is concerned with the unmasking of such mechanisms
of continuing political ideology and cultural hegemony; as a
moral theorist, the philosopher, has one .central task: to clarify
and just if), those normati ve standards in the light of which
such social criti cism is exercised.
" .

:.. :

, .
.; ..
34 I
Let us return once more to the Hegeli an objecti on: can a
ulllversahst ethIcal theory, which views universalizability in eth-
ICS as a moral governed by certain procedural
cons tralllts, aVOId the dIsmal choice" (O'Neill ) betwee t"
r . . n nVI-
a Ity or IIlconslstency? Hegel's critique assumes but does not
clanfy a dlstlllcllon between universalizability as a procedure
for tesllng and ul1lversalizability as a procedure for generating
maxIms. As a procedure for testing the intersubjective validity
of moral pnnClples and norms of action, communi cative ethi cs
IS neIther tnvlal nor IIlconsistenl' as a proced ure r .
I' d . . I .' lor generatmg
va I pnnclp es of aCll on, the model of moral conversation is
a necessary but insufficient test case that requires, in any given
IIlstance, adequate contextualization. In other words we can
say of a course of act ion, the principle of which has the
of conversati onal ulllversalizability, that it is morally per-
mISSIble, but also assert that it wasthe wrong thing to do under
the CIrcumstances., The uOlversahzabllity test should produce
standards of what IS morally permissible and impermissible in
general; however, such tests are by no means sufficient to
establI sh what IS morall y meritorious in any given context.
Habermas formul ates the test of universalizability thus:
... unl ess the consequences and side effects which the general ob-
of a norm can be expected LO have for the
sausfacuon of the mterests of each i,U/ividual can be {, I d b
J ree:y accepte y
What we are asking is not whether from this procedure the
moral theonst can deduce concrete moral pri nciples guiding
aCllon. The adoption of "all contents," writes Habermas "no
matter how fundamental the aClion norm involved ma;, be,
must be made dependent on r eal discourses (or ad vocator),
dIscourses conducted as substit utes for them.}"
Even If tillS principle.of uni versali zabili ty is not intended to
generate concrete pnnClples or norms of action, can it serve as
a test procedure for determining what is morall y permissibl e
and ImpermI ssIble? As a test procedure "U" e ..
. nJ olll s us to en-
gage 111 a counterfacLUal though t-experiment in which we enter
IIlto conversati on WIth all who would be potentiall), affected by
our actions. Let us consider some standard moral maxims to
Se)'la Benhabib
assess what has been gained by this reformulation. Take the
example used by Kant, "deposits once made must be returned
for otherwise there would be no property." The relevant ques-
tion is: does the principle "there ought to be property" satisfy
the test that "the consequences and side effects which the gen-
eral observance of a controversial norm can be expected to
I have for the satisfaction of the interests of each individual be
freely accepted by all?" The answer is that both the existence
of properly relations and its opposite can be adopted as a
coll ective maxim of thei r actions by moral actors, if the con-
sequences of such arrangements for the satisfaction of the
interests of each can be freely accepted by all. In other words,
the existence or nonexistence of property relations cannot be
determined via a moral deduction. Contrary to what Kant
assumed, as long as they serve the satisfaction of the interests
of each individual and this can be freely accepled by all, nu-
merous forms of property arrangements are morally permis-
sible. Kant was wrong in attempting to generate a categoncal
imperative to uphold property relations; what is at stake is not
property as such but other moral values like general welfare
and the correct mode of dispensing of scarce resources. To
this extent, the universalizability procedure in communicative
ethi cs upholds Hegel'S critique of Kant.
Yet, as formulated by Habermas, "u" also leads to morall y
disturbing and counterintuitive consequences. Take the
maxim, "Do not inflict unnecessary suffering." Whether or not
we are to inflict unnecessary sufl'c ring is to be determined by
whether the cOllse4uences and si de effects wh ich the general
observance of a controversial norm can be expected to have
for the satisfaction of the interests of each individual can be
freely accepted by all. Can we imagine a situation in which it
would be in the interests of each individual and freely accepted
by them that they would be not only perpetrators but receivers
of unnecessary suffering? The answer to this question appears
to depend on an equivocation concerning "interests." Suppose
there are masochists and sadists among us who interpret their
interests as consisting precisely in the opportunity to inflict and
receive such suffering. Are we ready to say that under these
conditi ons Nem;nclII laede ceases to be a morall y valid principle?
In other words, what appears to be the virtue of "U" in the
property example, i.e., its indeterminacy, is its weakness in the
second case. But the least that a universali st ethical theory
ought to do is to cover the same ground as what Kant had
described as "negative duties," i.e., duties not to violate the
rights of humanity in oneself and in others. Yet "U" does not
appear to do this.
I believe the difficulty is that Habermas has given "U" such
a consequentialist formulation that hi s theory is now subject to
the kinds of arguments that deontological rights theorists have
always successfull y brought against utilitarians. Without some
stronger constraints about how we are to inlerprel"U," we run
the ris k of regressing behind the achievements of Kant 's moral
philosophy. T he categorical imperative proves as morall y im-
permissibl e what Kant names "negative duties": not to lie, not
to harm, not to cheat, or otherwise violate the dignity of the
human moral person. Positi ve moral duties cannot be d educed
from the universalizability test alone but require contextual
moral judgment in their concretization.'o I have suggested
above that the communi cati ve ethi cs version of "U" must like-
wise deliver criteria for distingui shing among the morall y per-
missible and the morall y impermissible; nonetheless, this
distincti on alone does not yield adequate criteria of the morally
right or virtuous or appropriate action under cuncrete
Albrecht Wellmer, Agnes Heller, and Otfriec\ Hiiffe have all
recelllly ex pressed stronger criticisms of' comllluni ca tive ethi cs:
evell as a test procedure for what is illlersubjectivc\ y permis-
sible, "U," they argue, is either too indeterminate or too COI11-
plex or too counterfactual. In Hell er 's sharp formulation: "Put
bluntly, if we look to moral philosophy for guidance in our
anions here ~ l n c l now, we cannot obtain an}' positive g uidance
from the Habermasian version of the categorical imperative.
Rather, what we could get is a substantive limitation placed on
our intellectual intuitions: we, as indi viduals, should only claim
universal validity for those moral norms which we can assume
would be accepted by everyone as valid in an ideal situation of
symmetric reciprocity."" Albrecht Well mer writes: "If we in-
terpret 'u' as an expli cation of our preunderstandi ng of moral
Se),la Ilenhabib
validity, then this means that in our moral convictions and in
OIH moral judgments, only such judgments must be involved
that the consequences and side effects which the general ob-
servance of a specifi c norm would have for each indi vi dual
could be Jreely (zwangslos) accepted by all . This, however, so it
.appears to me, would make justified moral judgment a total
chimera (ein Ding der Unmuglichlwit).""
Heller argues that the Habermasian theory cannot be saved
for it is in effect a theory of "legitimation rather than one of
validation."" Wellmer recommends that we interpret the ideals
of "rational consent" or "agreement" as regulative principles,
but that in the solution of Teal moral problems under real moral
conditions, we can "onl y think of what the reasonable person or
. those competent judges or those affected by our actions would
say if they were sufficientl y reasonable, good willing and com-
petent in judgmerit. "" I think Well mer's response weakens the
distincti on between justifI cati on and contextualization. While I
agree that such contextualization is absolutely crucial for moral
judgment in real situations, I think hi s response makes the test
of the validity of moral judgment a matter of l,hronesll alone.
I am interested in seeing whether there is anything at all, any
guidelines, in the procedure of discourse ethi cs that could place
a "suustanti ve limi tation on our intell ectual intuition, " in the
way of necessary but insufficient criteria. Heller considers the
placing of such limi tati ons alone too minimal an achi evement
for moral theory. In my opini on, however, it would be quite
sufficient for it universali st moral theory which is self-conscious
aboul the historical hori zon of modernity within whi ch it is
situated, if it succeeded in placing such a substanti ve limitat ion
on our intuitions.
I want to suggest that "U" is actually redundant in Haber-
mas's theoq' and that it adds liu.le bUI consequentiali st confu-
sion to the basic premise of discourse et hi cs. "D" states that
onl y those norms can claim to be vali d that meet (or could
meet) with the approval of all concerned in their capacity as
participants in a practical discourse. "D," together with those
rules of argument governing discourses, and the normative
content of whi ch I summari zed as the principles of universal
moral respect and egali tarian reciprocity, are in my view quite
adequate to serve as the onl y uni ver sali zability test.
The chief difference between my proposal and Habermas's
is that for him "U" has the effect of guaranteeing consensus.
vVithout having their interests violated, all could freely consent
to some moral content. But the di ffi culty with consent theories
is as old as Roussea u's dictum-"On les forcera d'etre li bre."
Consent alone can never be a cri terion of anything, neither of
truth nor of moral validi ty; rat her, it is always the rationali ty
of the proced ure for attaining agreement which is of philo-
sophical interest. We must inter pret consent not as an end-goal
but as a process for the cooperative generation of truth or
validity. The core intui tion behind modern universalizability
procedures is not that everybody coul d or would agree to the
same set of principles, but that these principles have been
adopted as a result of a procedu re, whet her of moral reasoning
or of publi c debate, which we are ready to deem "reasonable
and fair. " It is not rhe result of the process of moral judgment
alone that counts but the process for the attainment of such
juugment which plays a role ill its validity and, I woulu say,
moral worth. Consent is a misleading term fo r capturing the
core idea behind communicative et hi cs: namely, the processual
generation of reasonabl e agreement about moral principles via
an open-ended moral conversation. I t is my c,"im that this core
intuiti on, together with an interpretation of the normati ve con-
straints of argument in light of the principles of universal
respect and ega litarian reciprocit y, arc suffi cient to accomplish
what "U" was intended to accompli sh, but onl y at the price of
conseq uentialist confusion.
Let us return once more to the pr inciple, "Do not infli ct
unnecessary suffering" to tcst this claim. " According to my
formula , we are to imagine whether if I and all those whose
actions would affect me and by whose acti ons I would be af-
fected were to engage in a moral conversation, governed b),
the procedural constrain ts ofuniversal respect and egalitarian
reciprocit y, we could adopt this as a principle of action. By
adopting the inflicti on of unnecessar y suffering as a norm of
action, however, we would in effect be undermining the very
idea of a moral di alogue in the first place. But it would be
Seyl;, Ben habib
abs urd to want to adopt as ,·ali d or correct a principle of
action-the infliction of arbitrary suffering-such as would im-
pair or jeopardi ze the very possibil ity of an ongoi ng conver-
sation among us. Since such ongoing moral conversation
involves sustaining relations of uni versal respect and egal itar-
ian reciprocity, if we all were to engage in the inRiction of
unnecessary suffering among ourselves we would undermine
the very basis of our ongoing moral relationship. In this sense,
universali zability is not onl y a fonnal j;rocedw·e but involves the
utopian project ion of a way of life as wel l.
There is an interesting consequence here: when we shift the
burden of the moral test in communi cative ethics from consen-
sus to the idea of an ongoing moral conversation, we begin to
ask not what all would or coul d agree to as a result of practical
discourses to be morall y permissible or impermissible, but what
would be allowed and perhaps even necessary from the stand-
point of continuing and sustaining the practice of the moral
conversation among us. T he emphasis now is less· on ralional
agreemenl, but more on sustaining those normat ive practices
and moral relationships within whi ch reasoned agreement as a
wa)' of life can Aourish and continue.
2 The Right and the Good
Sympathetic criti cs of communicative ethics have persistently
pointed out that this project formu lates more a model of jJolil-
icallegilimac)' than one of 1I/.ora.! va.lidil)'. To ask whether certain
nonnat ive institu tional arrangements wuuld or could be freely
adopted by all as being in thei r common interests, it is argued,
is precisely to continue the central id ea of the modern nat ural
ri ght and social contract traditions from Locke and Rousseau
to Kant.
Whi le many agree that such a principl e of rational
consent is fundamental to the modern ideas of democratic
legitimacy and justice, equall y many contest that it can serve as
a moral procedure that would be reJ e,·ant in guiding individual
action and judgment.
I have argued above that on my interpretation, the basic
principl e of discourse ethi cs together with the normative con-
straillts of argullIcnt ;. ni ol1 (all serve a ~ "substa nti ve tests" of our
moral intuitions. Furthermore, if we do not want to jettison
the distinction between contextua.lizalion and justification in et hi cs
altogether, we can sti ll preserve the model of a moral cOIl\'er-
sation taking place under the constraints of discourse as a
limiting test for our intuitions of the morall y permissible and
Impermiss ibl e. Clearly then whether discourse ethi cs is a model
of legitimacy or one of moral validity will depend on what
IInplicatlons and usefulness we think this model has for guiding
IIldl vldual moral action and judgment. Precisely because I
thlllk that It can have such impli cati ons when interpreted
properly, I also want to suggest that, at thi s stage of the debate,
the criti cs' arguments are not convincing.
Whereas some critics of discourse ethi cs want to rega rd it as
a program of political legitimacy rather than as one of moral
validi ty, others of a more n eo-Aristoteli an persuasion argue
that no pnnClples of legmmacy can be formulated without
presupposing some substanti ve theory of the good life. Quite
III lill e With Hegel's critique of Kant, these contemporar y Ar-
Istotehans and espeCIally communi tarian critics of li berali sm
maintain that the very idea of a minima.l-univenalist ethi c, whi ch
would be supposedly "neutral " vis-a-vis the multipli cit y of eth-
Ical life- forms, IS untenable. Charl es Taylor's objections to com-
municati ve ethics have foll owed this line of argument" I want
to name this the issue of the "right" vs. the "good."
. From the outset, however, we must di stin guish between the
hberal-communitarian version of this controversy on the one
hand, and the controversy as it appli es to conllnuni cative ethi cs
on the other. The first controversy conce rns whether li beral
principles of justice, as formulated by J ohn Rawls and Ronald
Dworkin in particular, are "neutra l," in the. sense of all owing
the coexistence of man y forms of life in th e polity, or whether
these pnnclples both pres uppose and privil ege a specific way
of hfe-Iet us sayan individualist one, centered around the
virtues of the rule of law a t the expense of solidarity, of privacy
at the expense of communi ty, aI1d of justice at the expense of
fnendship. While liberals continue to aspire to such neutrality,
COllllll ullitanans IIlSISt on the Illusory <jualIl y of their search.'"
This debate between liberals and communitarians cannot be
simply extended to communicative ethi cs, for the obvioll s rea-
Scyltl llcnhabib
son that neit her Apel nor Haber mas have developed a nor-
malive theory of justice out of communi cative ethics, alt hough
communicative ethics has definite institutional impli cations (see
section 3 below). When appli ed to communi cative ethics, the
issue of the "right"' vs. the "good" concerns nOt so much the
aUeged neutrality or non-neutralit y of principles of j ustice, as
it does the very basis of the distinction between 'justice" and
the "good life" within ethi ca l theory itself. .
The defense ofa deontologica l outl ook in Habermas·s theory
takes a different form than what we encounter in Rawl's Theory
oj j ustice'9 Whereas Rawls distinguishes between justice as the
basic virtue of a social sys tem and the domain of moral theory
at large in whi ch a full theory of the good is at work,' o Haber-
mas is committed to the stronger claim that after the transition
to modernity and the destructi on of the teleological world view,
moral theory in fact can onl y be deontological and must focus
on questions of justice. Following Kohlberg, he insists that this
is not merely a hi stori call y contingent evolution, but that 'j udg-
ments of justice" do indeed constitute the hard core of all moral
judgments. Habermas writes: "Such an ethi c ... st.yli zes ques-
ti ons of the good li fe , and of the good life together into questions
oj justice, in order to render practical questi ons accessibl e to
cogniti ve processing by wa y of thi s abstracti on."" It is not that
deontol ogy describes a kind of moral theor)" juxtaposed to a
teleological one; for Habermas, deontological judgments about
justice and rights claims defin e the moral domain insofar as
wc can say anything cogniti ve1y mea ningful about this.
How can we in fact defend the thesis that judgments of
justice and right constitute the moral doma.ill? I can see two
distinct arguments in Habermas's work on thi s issue. First,
Habermas assumes that only judgments of justice possess a
clearl y discerni ble formal structure and th us can be studied
along an evolutionary model." J udgllJents concerning the
good life are amorphous and do not lend themselves to the
same kind of formal stud)". But of cou rse t.hi s observation, far
from juslifying the r estrin ion of the l1Ioral domain LO matters
of justice, could also lead to the concl usion that one needed to
develop a less formalistic cthi cal theory. This is a view which
h<ls bccn successfull y defended by Bernard Wi lli ams in hi s
Ethics and the Limit.s oj Philosoph), and by Charles Taylor in
various anicles.<i3
Second, Habermas maintains that the evoluti on of j udgments
of justice is intimately tied to the evolu tion of self-other rel a-
tions. Judgments of justi ce reAect various conceptions of self-
other relations, whi ch is to say, th at the formation of self-
identit y and moral j udgments concerning justice are intimatel y
linked. This is because justice is the social virtue par
excell ence.
Again, however, it can be objected that the evolu tion of self-
ot her relations must also be accompani ed by the development
of self-u nderstandi ng and self-evaluation, and if j ustice is the
sum of othcr-,·egm·ding virtues par excell ence, this sti ll does not
preclude the consideration of self-regarding virtues and their
signifi cance for moral theory. If one understands Habermas's
defense of deontological ethi cs as a claim concerning the ajJ- .
'JrojJri.ate object domain of moral theory, then I can see no plau-
sible arguments in favor of such a restri ctive view of what moral
theory can hope to accomplish.
I concur then with communitarian critics of d eontology like
Bernard Williams, Charles Taylor, and Michael Sandel onl y to
the ex tent that viewing justice as the center of morality unne-
cessaril y restricts the domain of moral theory, thus di storting
the nature of our moral experiences. But a uni versalist and
communicative model of ethi cs need not be so strongly con-
st rued. Such a theory can be understood as d e fending a ··weak"
deontology; th is means thal valid moral norms mUSI be able 10
stand the test of discursi ve justifi cation. Since practical dis-
courses do not theoreticall y predefine the domain of moral
debate and since individuals do not have to abstract from their
everyday attach ments and beliefs when lhey begin argumen-
tati0l1, however, we can accept that not onl y matters of justi ce
but those of the good life as well will become thematized in
practi cal discourses. A model of communicati ve et hi cs, whi ch
vi ews 1110ral theory as a theory of argumentati on, need not
restri ct itself to questions of justice. I see no reason as to wh y
qu esti ons of" the good life as well cannOI become subj ect matters
of practical discourses. It ma), very well be that di scourses will
not yield conce ptions of the good life eq uall y acceptable to all;
Se)' la Benhabib
yet there is a di ffere nce bet ween assuming a priori that certain
mallers are ques ti ons of the good lire and therefore inappro-
priate matters of moral argument, and between assuming that
a moral communit y will establi sh a line between individual
conceptions of the good to be pursued freely and shared norms
and values to be cul tivated coll ectivel y. It is crucial that we view
our concepti ons of the good life as matters about whi ch inter-
subjective debate is possible, even if intersubj ective consensus,
let al one legislati on, in these areas remains undesirable. How-
ever, onl y th rough such argumentative processes can we draw
the line between iss ues of justice and of the good life in an
epistemi cally plausible manner, whil e rendering our concep-
ti ons of the good life accessibl e to moral renecti on and moral
transformati on.
Of course, thi s is a far weaker result than may be preferred
by a strong teleologist like Alasdair Maclntyre bu t it remains
for such a teleologist to show that under conditions of mod-
ernity one can indeed formulate and defend a uni vocal con-
cepti on of the human good . So far Habermas is right: under
conditions of modernity and subsequent to the difrc rellliation
of th e value spheres of science, aesthetics, jurisprudence, reli-
gion, and morals we can no longer formulate an overarching
vision of the human good. Indeed, as Alasdair Maclntyre's
definiti on of the good life, namel )" "the life spent in seeking
the good life for man"45 very well reveals. as moderns we have
to li ve with vari eti es of goodness. Whether the good life is to
be fulfill ed as an Afri can famine reli ef fighter, a Warsaw ghetto
resistant, a Mother Teresa, or a Rosa Luxemburg ethi cal theory
cannot prejudge; at the most modern moral theor y provides
us with some ver y general criteri a by whi ch to assess our in-
tuiti ons abou t the basic validi ty of cen ain courses of acti on and
the int egrity of certain kinds of I regard neither the
plurali ty and vari ety of goodness with whi ch we have to li ve in
a disenchanted uni verse nor t he loss of cen aint )' in moral
theor), to be a cause of di stress. Under condi tions of value
diffe renti ati o n we have to conceive of reason no t in t.he image
of a homogeneous, transparent glass sphere into whi ch we can
fit all our cogniti ve and \'alue commi tments, but more as the
35 1
Afle rword
light shed by bits and pi eces of di spersed crystals whose con-
tours shine out from under th e rubble.
3 On the Distinction Between Justice, Morality, and Politics
The neo-Aristotelian and neo-Hegelian insistence on the cen-
trality of a shared ethos or of a concrete Silllichkeil in the
conceptualization and resolution of. moral qu estions, has un-
avoidable impli cations in the domain of political action as well.
If this shared ethos and this Siulichkeil are viewed not primarily
as the unavoidable hermeneuti cal horizon over and against
whi ch moral questions and problems can be formul ated , but if
they are considered the normati ve standard in li ght of which
to assess individ ual actions, then morality becomes su bordi-
nated to the collective ethos of a community.
As the young Hegel wistfull y wrote of the poli s,
As freemen the Greeks Romans obeyed laws laid down by them-
selves, obeyed men whom they had themselves appointed to offi ce,
waged wars on which they had themselves decided, gave their prop-
crt)', exhausted their passions, and s;lcrifl ced their lives by tho usands
fo r an end whi ch was thei r own ... In public as in and
domestic life, every individual was a free man, one who li ved b), hi s
own laws. The idea (Idee) of hi s country or of hi s state was the
invisible and hi gher reality for whi ch he strove, whi ch impell ed him
to effort; it was the final end of his world or in his eyes the final end
of the world, an end whi ch he found manifested in the reali ties of hi s
dail y life or whi ch he himself cooperated in manifesting and
Undoubtedl y, thi s idealization of the Greek polis has to be
viewed today more in light of German romantic attitudes to-
ward Greek antiquit y than judged as a historicall y accurate
depicti on of Greek society. As the mature Hegel himself rec-
ogni zed, the ri ghts of subjective welfare and conscience are
among the constituents of the moral freedom of the individual ,
and the individuals' pursuits of vari ous conceptions of the good
can never be wholl y integrated within a concrete ethical totalit y.
The spli t of ethi cal life into the famil y, civil society, and the
state under conditi ons of modernit y also means that potentiall y
the dictates of individual conscience and welfare on the one
Seyla Benhabib
hand and the claims of institutions, li ke the famil y, market,
and the state, can always clash. In a famous passage of the
Philosophy of Right Hegel defended the rights of Anabaptists
and Quakers to refuse military service in the modern state on
the g,-ounds that the state is strong enough to all ow for dissenL
without crumbling in the face of il." However, both in his
theory of representative institutions and even more so in his
reAections on war and world history, Hegel made the "self-
preservation" of the universal the normative goal to which mo-
rality had to be subordinated. Politics, understood as the sphere
governed by the d ictates of the self-preservati on and the wel -
fare of coll ectivities, is juxtaposed by the mature Hegel to the
"abstract cosmopoli tanism" and "universalism" of KanLian
ethi cs.
In contemporary debates one can recognize this Hegelian
antecedenL in twO charges which are frequently leveled against
communicative ethi cs: First, communicative ethi cs is said to
lead to anti-institutionalist and fundamentally anarchisti c con-
sequences in political life'S; second, communicative ethics is
said to be "moralistic" to the point of complete utopianism in
the domain of politics. Imagine conducting a practi cal dis-
course on malleI'S of international relations, state security,
maybe even banking and fiscal policy under the constraints of
an id eal speech situation! The strategic and instrumental re-
lation of the parties to each other is so fundamentally consti-
tutive of these macroinstitut ions of political life that the kind
of moralistic utopianism advocated by partisans of discourse
ethics, so argues the political reali st, would only result in con-
fus ion and insecurit y. In the domain of poli tics realism, enli ght-
ened by an ethics of responsibility, in the Weberian sense, is
the best approach (see Herman Liibbe's essay in this volume) .
In the face of the charge of anti-institutionalism it must be
said that the discourse ethi cs is not a theory of instituti ons,
alt hough it has institutional im plications. Whether we interpret
them as principles of legitimacy or as principles of moral valid-
it y neither uD" nor "U" can yield a concrete theory of institu-
tions. but they have instituti onal impl ical iuns . .<HI Institutionalist
thinkers like Liibbe and Niklas Luhmann maintain that up-
holding any concrete instilliti ons to the demands of such ra-
tional consensus would make life impossible. Within the
constrai nts of institutions, decision procedures, limited by
space and time and scarce resources, must be respected. To
hope for the rational consensus of all under these circum-
stances would paralyze institutional life to the point of a
This obj ection is justifi ed, but it confuses levels: the discourse
theory does not develop a positive model of functioning insti-
tuti ons, which after all will always be subj ect to time-space
const raints as well as to those of scarce resources and person-
nel. The di scourse theory develops a normative and critical
criteri on by which to judge existing institutional arrangements,
insofar as these current arrangements suppress a "generaliza-
ble interesl." T his appeal to the "suppressed generali zable in-
terest" need not be read along Rousseaui an lines.'o In complex
societies, it is doubtful that there could be a definition and
specifi cation of the suppressed generalizable interest which
would meet with the consent of all. But one can use this cri-
terion as a critical yardstick by whi ch to uncover the under-
representation, the excl usion and sil encing of certain kinds of
inLerests. In other words, it is not so much the identifi cation
of the "general interest" whi ch is at stake, as the uncovering
of those partial interests which represent themselves as if they
were general. The assumption is that institutions can function
as channels of illegitimate exclusion and sil enci ng, and the task
of a critical discourse theory is to devel op a moral presumption
in favor of the radical democratization of such processes.
What instituti onalists neglect is that power is not only a social
resource to be distributed, say like bread or automobil es. It is
also a sociocultural grid of interpretation and communi cation.
Public dialogue is not external. to but constitu tive of power
relati ons : paraphrasi ng Nancy Fraser, there are officiall y rec-
ogni zed vocabul aries in which one can press claims; idioms for
interpreting and communicating one's needs; establi shed nar-
rative conventi ons for constructing individual and coll ective
identities; paradi gms of argumentati on accepted as authorita-
tive in adjudicating confli cting claims; the repertory of avai lable
rhetorical devices, and the like'l These constitute the "meta-
politics of instituti onal dialogue," and as a critical theorist, one
Sc}' la Benh;tbib
is interested in idelllifying those social relations, power struc-
tures, and sociocultural grids of communication and interpre-
tation at the present wh ich limi t the identi ty of the panies to
the dial ogue, whi ch set the agenda for what is considered
appropriate or inappropriate malter for institutional debate,
and whi ch sanctify the speech of some over those of others as
being the language of the public.
Cenainly thi s is not the only point of view from which to
understand and evaluate institutions: justice, efficiency, stabil-
ity, and predictability are also relevant criteri a. To assume
though that all discourses of legitimacy are counterproductive
or anarch istic is to disguise political authoritarianism as a post-
Enlightenment critique of the Enli ghtenment.
In his essay, "Is the Ideal Communication Community a
Utopia?," Karl Olto-Apel deals extensively wi th the question
of the utopian cOlllent and implications of communicative eth-
ics (in this volume). In his view, it would be utopian in the
negative sense of extreme irrelevance to· demand that all in-
stances of strategic acti on, whether individual or coll ective, be
governed by the nor ms of communicative acti on, aimed at
achieving mutual understanding and reciproci ty. Nonethel ess,
it is both a moral and a politi cal question to ask what the limits
of individual and coll ective strategic action are, and to reflect
on how to mediate between the requ irements of self- interest
on the one hand and the moral principles of mutual and
cooperative understand ing on the other. Once we restate the
problem in thi s fashion, a whole r'lI1!(e of interesting con-
siderations begin to emerge. The stark opposition between
political utopianism and poli tical realism is softened. Com-
municative et hi cs anticipates nonviolent strategies of confli ct
resol uti on as well as encouraging cooperative and associative
methods of problem solvi ng. It is. a maller of political imagi-
nation as well as coll ective fantas y to project institutions, prac-
tices, and ways of life which promote nonviolent confli ct
resol uti on strategies and associative problem solving methods.
Far from being utopian in the sense of being irrelevant, in a
world of complete interdependence among peoples and na-
ti ons, in whi ch the alternatives are between nonviolent collab-
oration all d !lucle;,r anl1ihilati on. COllllllllllicalivc el hics
suppl y our minds with just the r ight dose of fa ntasy such as to
think beyond the old oppositions of utopia or realism, contain-
ment or confli ct. Then, as today, we still can sa)" "L'imaginat ion
au pouvoir'"
4 On The Problem of Moral Motivation and Character
A major weakness of cognitive and procedural ist ethical theo-
ri es since Kant has been their reductionist treatment of the
emotional and affecti ve bases of moral judgment and conduct.
Twentiet h-century neo-Kantian ethical theories have by and
large rej ected Kant's dual istic moral psychology, and his re-
pressive treatment of sensualit y and the emotions, all the "'hil e
retaining the distinction between "action done from the motive
of dut y" and "self-regarding actions." Nevertheless, th is rej ec-
ti on of the Kantian treatment of the emotional and affective
basis of ethi cs has not meant paying renewed attention to these
issues. In recent years, it has been philosophers like Amelie
Rony, Martha Nussbaum, Annette Baier, and Lawrence Blum
on this side of the ocean and Ursula Wolff in German y, as well
as feminist moral theorists like Virginia Held and Sara Rud-
di ck, who have developed a ri ch and signifi cant body of work,
analyzing moral emotions and moral character
' Does the ne-
glect of these iss ues by advocates of communi cative ethi cs so
far point not just to a weak spot in the theory but maybe to a
blind spot altogether?
I would like to suggest that vcr)' often ethi cal cogni tivislll
has been confused with ethi cal ra ti onali sm, and the neglect or
the affective and emotive bases of ethi cs is a res ult of the
narrow "rationalism" of most neo-Kanti an theori es. By "ethical
cognitivism" J understand the view that ethical judgments and
principles have a cogni tively aniculable kernel, that they are
neither mere statements of preference nor mere statements of
taste bUl that they imply validity claims. These claims can be
stated as: "X is ri ght," where by X is meant a principle o f act ion
or a moral judgment, meaning "J can justif), to you with good
grounds why one ought to respect, uphold, agree with X." In
this sense, ethi cal cognitivism is to ethi cal
that red uces Sti ch principles andj udglllcms to an "I will " which
Sc)' Ja Benhabib
cannot be further questioned. Eth ical cogl1lllvlsm is also op-
posed to ethi cal emotivism that conAates statements like "Child
molesting is wrong" with claims like "I like Haagen-Dasz ice
By "ethical rationalism," by contrast, I mean a theoretical
position which views momi judgments as the core of moral the-
ory, and which neglects that the moral self is not a moral
geometer but an embodied, finite, suffering, and emotive
being. We are not born rational but we acquire rationality
through contingent processes of sociali zation and identity for-
mation. Neo-Aristotelians as well as feminist theorists in recent
years have argued that we are children before we are adults,
and that as human children we can only survive and develop
within networks of dependence with ot hers, and that these
networks of dependence constitute the "moral bonds" that con-
tinue to bind us even as moral adults. In Virginia Held's words,
by ignoring the genealogy of the moral self and the develop-
ment of the moral person out of a network of dependencies,
universalist theorists often view the moral agent as the auton-
omous, adu ll male head of household, transacting in the mar-
ket place or in the polit y with like others" Since Rousseau the
demand has been to make ''I'homme'' whole .again, either by
making him wholly a "Burgher" or by making him a "citoyen."
This "rationalist" bias of universali st theori es in the Kantian
tradition has at least two consequences: first, by ignoring or
rather by abstracting away from the embedded , contingent,
and fin ite aspects of human beings, these theori es are bli nd to
the variety and richness as well as signifIcance of emotional
and moral devel opment. These are viewed as processes pre-
ceding the "genealogy" of the adult moral self; they seem to
constitute the murky and shadowy background out of which
the li ght of reason emerges.
Second, the neglect of the contingem beginnings of moral
personality and character also leads to a distorted vision of
certain human relationships and of their moml lexlw·e, precisely
beGlllSe universalist and proccduralisl et hi caltheorisls confuse
the moral ideal of autonomy with the vis ion of the self "as a
mushroom" (Hobbes).'" Far from bei ng a description of the
"moral pOilll of view," state of nature abstractions as well as
visions of the "original position" are projections of the ideal of
moral autonomy which only reAect the experience of the male
head of household. But let us proceed cautiously here: I am
nol arguing that a trul y universalist articul ation of the moral
point of view, one that includes the experiences of women and
children, mothers and sisters, as well as brothers and fathers
is not possible. The gender-blindness of much modern and
contemporary universalist theory, in my opinion, does not com-
promise moral universalism as such, it only shows the need to
judge universalism against its own ideals and to force it to
make clear its own unjustified assumptions.
Current constructions of the "moral point of view" so lopsid-
edly privilege either the homo economicus or the homo /Joiilicus
that they exclude all familial and other personal relat ions of
dependence from their purview. Whil e to become an autono-
mous· adult means asserting one's independence vis-a-vis t h e s ~
rel ations, the process of moral maturation need not be viewed
along the fi ctive model of the nineteenth-century boy who
leaves home to become "a self-made man" out "yonder" in the
wide, wi ld world. Moral autonomy can also be understood as
growth and change, sustained by a network of relationships.
Modern and contemporary constructions of the moral point of
view are like the distorting lens of a camera: if you focus too
badly, the scene in front of you not only becomes murky but
can lose contours altogether and become unrecognizable. Like-
wi se, the construction of those moral procedures whi ch arc \0
act as "substantive limits on our intuitions" must not be so out
of focus that by looking through them, we lose the moral
contours and moral textures of such personal relati onships.
Moral vision is a moral virtue, and moral blindness implies not
necessaril y an evil or unprincipled person, but one who cannot
see the moral texture of the situation confronting him or her"
Since the eighteenth century, ethical rationalism has promoted
a form of moral blindness with respect to the moral experience
and claims of women, children , and other "nonautonomous
others," as well as rough handling the moral texture of the
personal and the familial.
Communicative ethics, in my view, is a form of ethical cog-
nitivism which has so far been presented as a form of ethical
Scyla Bcnh;dJib
rati onalism. Part icularl y the claim, discussed above , that judg-
ments of justi ce constitute the hardcore of all moral theory is
an instance of such rationalism. As 1 have argued above (see
section 2), even from within the constraints of a discourse
theory, this hard distinction between judgments of justice and
those of the good li fe cannot be sustained. Neither can the
privilegin g of moral judgments to the neglect of moral emo-
tions and character. There is a curious inconsistency here. The
theory of communi cative competence devel ops a post-Enli ght-
enment conception of reason and views reason as the contin-
gent acquisition of beings capable of language and action to
articulate and sustain intersubj ective validity claims'· The the-
ory of communicative ethics, however, more often than not
seems to perpetuate the Enlightenment illusions of the rational
moral self as an isolated moral geometer.
If this is so how can I maintain, as I also did in the first part
of this essay,' that the model of a universalist moral dialogue,
envisaged ' in accordance with the formal constraints of dis-
courses, can serve as a defensibl e version of the "moral point
of vi ew"? My answer is that the less we view such discourses
along the model of public fora or courts of appeal, and the
more we understand them as the continuation of ordinary moral
conversations in which we seek to come to terms with and ap-
preciate the others' point of view, the less do we submit to the
distorting lens of procedural uni versalism. To argue that the
counterfactual ideals of reciprocity, equali ty, and the "gentle
force of reason" are implicit in the very st ructures of commu-
ni cative action, is to argue that the "moral point of view" artic-
ulates more precisely those implicit structu res of speech and
action within whi ch human life unfolds. Each time we say to a
chil d, "But what if other kids pushed you into the sand, how
would you feel then?", and each lime we say to a mate, or to
a relative, "But let me see if I understand your point correctly,"
we are engaging in moral conversations of justi fi cation. And if
I am correct that it is the process of such di alogue, conversa-
tion, and mUlual understanding. and no l consensus which is
our goal, discourse theory can represent the moral point of
view without having to invoke the fiction of the homo economicus
or h 01l/.0 fi olilic11..1. To know how to susLain an ongoing human
relationship means to know what it means to be an "1" and a
"me," to know that I am an "other" to you and that likewise,
you are an "I" to yourself but an "other" to me. Hegel had
named this structure that of "r eciprocal recognition." Com-
municative actions are actions through whi ch we sustain such
human relationships and through whi ch we practice the rev-
ersibility of perspectives implicit in adult human relationships.
T he development of this capacity for reversing perspectives
and the development of the capacity to assume the moral point
of view are intimately li nked. In the final analysis, universal-
izability requires us lO practice t he reversibility of standpoints
by extending this lO the vi ewpoint of humani ty. Such a capacity
is essential to being a good partner in a moral conversation,
and is itsel f furthered by the practice of moral conversation.
In conversation, I must know how to listen, I must know how
lO understand your point of view, I must learn to represent lO
myself the world and the other as you see them. If I cannot
listen, if I cannot understand, and if I cannot represent, the
conversation slOpS, devel ops into an argument, or maybe never
gets started. Discourse ethics projects such moral conversations,
in which reciprocal recognition is exercised, onto a utopian
community of humankind. But the abili ty and the willingness
of individ uals to do so begins with the admonition of the parem
to the child: "What if others threw sand in your face or pushed
you into the pool, how would you feel then?" .
5 Judging in Context vs. Principled Rigorism
The last issue I would like to treat in this aft erword is the
problem of phronesis or practical wisdom concerning parti cu-
lars. ArislOtle saw this as the crowning achievement of moral
paideia and character. A common criticism of Kami an-type eth-
ical theories is that they substitute an ethical ri gorism of prin-
ciples for the art of moraljudgment'7 Justifiabl e as this critique
may be, the di scussion concerning moral judgment by either
group of contenders in this debate has not advanced very far.
The metaphor of the "archer hi tting the .mark," the language
of moral insight and blindness, still dominate many recent
treatments of the issue. If we can register a certain impatience
Seybl Hcnhabib
with neo-Aristotelians in this respect, we must also admit that
distinguishing between 'Justification" and "contextualization"
cannot exempt the discourse theorists from analyzing what it
is that we do when we supposedly contextualize moral princi-
ples and how this .activity is related to the work of judging'·
Obviously, there is a difference between the contextual appli-
. cation of a cookbook recipe in the environment of our kitchens,
given the ingredients and the utensi ls we have, and the so-
call ed "co11lcxtuali zation" of moral principles. If the discourse
model is to succeed in acting as "a substantive limi t on our
intuitions" of the morally permissible and impermissible as well
as guiding us in our vision of the morally required, we must
be able to suggest how the procedural model of the moral
conversation developed so far is in volved in the process of
moral judgment.
I would like to suggest that if there are certain moral and
cognitive skills involved in reaching perspicacious, appropriate,
sensitive, and illuminating judgments that they may bear a
"famil y resemblance" to th e conversational skill s and virt ues
invol ved in the ongoing pract ice of moral dial ogue and dis-
course. T here is a cardinal requirement of contextual judg-
ment, which most theorists, from Immanuel Kant to Hannah
Arendt, who have developed the problem of judgment have
suggested, and this is the abi li ty, in Hannah Arendt's words,
for "representative thinking":
The power of judgment rests on a potential agreement with others,
and the thinking process which is active in judging something is not,
like the thought process of pure reasoni ll g. a dialogue between me
and myse lf, uut fi nds itself always and prim<lril)r, even if I am quite
alone in m;:lking lip my mind, in an anticipated communication with
others with whom I know I must finally come to some agreement.
From thi s potential agreement judgment derives its specific validity.
This means, on the olle hand, that such judgment must liberate itsel f
from the "subjective pri\'ate conditi ons," that is, from the idiosyncra-
cies whi ch natural]\, determine the outl ook of each indi vi dual in hi s
privacy and are lc'gilim3te as long as the), are onl y privately held
opinions but whi ch are 110l fit LO enter the market place, and lack all
\'alidit), in the publ ic realm. And thi s enlarged way of thinking, which
as judgmem kll ows how 10 transcend its individual limitat ions. cannot
funCLion in strict isolation or soliLUde; it needs the presence of others
~ 6 1
"in whose place" it must thillk, whose perspectives it must take inlO
consideration, and without whom it never has the opportunit y to
operate at all'9
In Kant 's discovery of the "enlarged mentality" in his theory of
reRective judgment, Arendt saw a model for the kind of in-
tersubjective val idit y which judgments had to be submitted to
in the public realm. J udgment involves the capacity to r epre-
sent to oneself the multiplicity of viewpoints, the variety of
perspectives, the layers of meaning, etc., which constitute a
situati on. This represen tational capacity is crucial for the kind
of sensiti vity to particulars, which most agree is central for
good and perspicacious judgment. The more we can identi fy
the different viewpoints from which a situati on can be inter-
preted and construed, the more we will have sensitivity to the
particularities of the perspectives involved. Put differently,
judgment involves certain "interpretive" and "narrative" skills,
which, in turn, entail the capacity for exercising an "enlarged
mental ity." Thi s "enlarged mentality" corresponds precisely to
the reversibility of perspectives wh ich the discourse theory en-
joins. The link then between a uni versalist model of moral
conversati on and the exercise of judgment is this capacity for
reversing moral perspectives, or what Kant and Arendt name
the "enlarged mentality." Let me suggest in more detail why
the narrative and interpretive skills in volved in judging entail
reversibili ty of moral perspectives.
Moral judgment is crucial
in at least three domains of moral interacti on: the assessment
of one's duties; the assessment of one' s specific cou r se of action
as fulr.lling these duties; and the assessment of oll e's maxims
as embodied, expressed, or revealed in actions.
In the assessment of duties, we are concerned with recog-
nizing a particular situation as being one that calls for a specific
kind of moral duty. How do we know that thi s human situation
call s for the duty of honesty, or the virtue of loyalty or .of
gcnerosity' What is it about a particular human situation that
will allow us to identify it as being of a certain kind? I would
li ke to suggest that here moral judgment is concerned fi rst wit h
the identifi cation of human situations and circumstances as
being "morall y rele\'ant." By "morall y relevant" I mean a sit-
Seyb] lk nhabib
uat ion or circumsta nce so defmed that it wou ld lead us to
recogni ze a prima facie moral dUly LO act in a certain way.
Whi le it is precisely the mark of one who has good moral
judgment that she id entifi es this as being a situat ion of loyalty,
of generosity, of courage, or of integrity, whatever else such
judgmen t takes, it most certainl y must involve the capacit y for
represel1lative thinking or the reversibili ty of moral perspec-
tives. Onl y one who is abl e imaginati vely LO represel1lto herself
the va ri ety and meaning of the human perspecti ves involved
in a situation can also identi fy its moral relevance. For moral
relevance in this comext means understanding the moral de-
scriptions and expectations and interpretati ons that make up
the narrative fabric of a human story.
What about the assessment of one's acti on? Whereas in the
case of assessing moral dut y we as k, in what ways is this situa-
ti on morall y re1evam for me, now we are 'as king, "What is it
that I must do to fu lfi ll my duty to act morall y once I have
recogni zed it ?" In other words, what· r do, whi ch course of
acti on I choose, in volves some inter preti ve abili ty to see my act
under various act d escri pti ons and to anti cipate how, while
action A may be viewed as one of generosity, acti on B may be
vi ewed as one of overbearing soli citude. I must have enough
moral imagination to know the possibl e act descriptions or
narratives in li ght o f whi ch an act embodying a maxim can be
consi dered. Determining the identity of a moral acti on entails
the exercise of moral imagination; this acti vates our capacity
for imagini ng possible narratives and descriptions in li ght of
whi ch our ,lCtions can be understood by others. Again, .such
moral imagination involves representative thi nking, namely,
the capacity to take the standpoint of others involved into
account and to reason from their point of vie\\'.
Finall y, let us look at the concn;tizati on of one's maxim, or
principle of d uty, via a concrete action. There is often a cl ash
between th e moral intenti ons or principles guiding an agent
and the interpreta ti on of thi s by the worl d, once they are
embodi ed in actions. In formul ating moral intenti ons and max-
ims-"I recogni ze that I must be generous now," "Honesty is
always Illy I'0li cy"-we project ourselves. our narrative history,
int o the worl d, and we want to be recogni zed as the doer of
such and such. We identi fy our moral intenti ons and principles
in terms of a narrative of whi ch we ourselves are the author.
This narrative also anticipates the meaning that such proj ecti on
mayor will have in the eyes of others. Assessing one's moral
intentions and maxims, therefor e, requi res understanding the
narrati ve history of the self who is the actor; this understanding
exhibits both self-knowl edge and knowledge of onesel f as
vi ewed by others. The narrative capacity for proj ecting a course
of acti on, whi ch exhibits and embodies our moral intenti ons
and maxims, requi res sensitivity to the many perspectives and
interpretations in li ght of whi ch our narrative and personal
story will be construed . This means once more that reversi bility
of perspecti ves or the capacity for re presentative thinking are
cemral in such formulations.
What I have suggested so far is that if we vi ew discourses as
a procedural model of conversations in whi ch we exercise rev-
ersibili ty of perspectives either by actuall y listening to' all in-
volved or by representing to ourselves imaginati vely the many
perspecti ves of those involved, then this procedure is also an
aspect of the skills of moral imaginati on and moral narrative
whi ch good judgment involves whatever else it mi ght involve.
do not therefore see a gulf between moral intuition guided
by an egalitarian and universali st model of moral conversati on
and the exercise of contextual judgment. Qui te to the contrary,
the kinds of interpr etive and na rrative skill s I di scussed above
can also be easil y used for "amo ral " purposes.
The exercise of good judgment can also mea n manipul ating
people-presumably good ad ministrators, poli ticians, thera-
pists, social workers, and even teachers of young children all
exercise "good judgment," not always for the sake of moral
reciprocity or with respect to enhancing the moral integrity of
the one about whom such judgment is exercised. Moral judg-
ment alone is not the totality of moral virtue. Here as well we
need a "substanti ve limit" on our intuitions: onl y judgment
guided by the principl es of uni versal moral res pect and reci -
procit y is "good" moral judgme nt, in the sense of being ethi-
call y right. Judgments whi ch are not limited by s uch principles
may be "brilliant," "ri ght on the mark," "perspicacious, " but
also immoral or amoral. Saying this, however, is not to say that
Sc)'I" Uenhabib
in a fragmented universe of value we arc never in the situation
of juggling moral principl es against other political, artistic, and
administrative ends. Kamian theories have paid lillie allemi on
to this "fragmemation of value," and to the consequences which
the fine LUning and balancing of our moral commitmems with
other value commitments have for the conduct of our li ves.
Here, we reach a frontier where moral theory flows into a
larger theory of value, and I would say, into culture at large.
Morality is a central domain in the universe of values which
define cui LUres, and it is cultures which supply the motivational
pallerns and symboli c interpretations in li ght of which individ-
ual s think of narrative histories, project their visions of the
good li fe, il1lerpret their needs, and the like. Moral theory
finds this material , so to speak, "given. " T.hus, moral theory is
limited on the one hand by the macroinstiLUtions of a polity,
politics, administration, and the market, within the limits of
which choices concerning justice are made. On the ot her hand,
moral theory is limi ted by culture, its repertory of interpreta-
tions of the good li fe, personality, and socialization patlerns.
These two domains form the larger ethi cal context of which
moralit y is always but an aspect. Yet the relation between mo-
rality and this larger ethical context is not what neo-Aristote-
li ans and the young Hegel would like us to think it is. Under
conditions of modernity, as the old Hegel knew, the moral
poil1l of view always judges the instilUtions of which it is a part;
and the modern individual exercises autonomy in distancing
him or herself from the given cultural interpretation of social
roles, needs, and conccptions of t.he good life. III this sense the
dispute between discourse theori sts and neo-Aristotelians and
neo-Hegelians is at its heart a dispute about modernity; it is a
dispute about whether modern moral theory si nce Kant has
been an accompl ice in the process of disintegration of person-
alit y and the fragmentation of whi ch is said to be our
general condition today.· ' My intervention in this debate in-
tended to show that, j udged from within the confines of moral
theoq', and without delving il1lo this larger iss ue about mod-
ernit y and its discontems, the debate between neo-Aristoteli-
ans/neo- Hegelians and discourse theorists is still very much
cominuing. Although it is too tritc to think that all philosoph-
ical debates lead to good endin gs, my own personal sense at
this stage is that this confrontation has invigorated rather than
weakened contemporary moral theory.
J would like 10 thank my colleagues Ke nneth Baynes and Dick Howard fo r their
illuminating criticisms of an earlier draft. A shoner vers ion of Ihis essay appeared in
Tlu Philosophical ForI/in. special doubl e issue on "Hermenellti cs in Ethics and Social
Theory," cd. by Michael Kell y. \'01. 21. nos. 1-2 (Fall-Willler 1989-90), pp. 1-32.
1. See Kurt Baier, Till' Moral Poi111 of Vinn. abridged ed. (New York: Random Housc,
1965); Alan Gewirlh, Reason and Moralit), (Chicago: University of Chicago Press. 1978);
H. M. Hare, Fr'udom and RI'tL'iD11 (Oxford: Ox ford University Press. 1963); Marcus
Singer, Gelll'roiiwbiiit)' in Ethics. An Ello)' in the Logic of Ethics with the Rudimrnu of a
System of Moral PhiloJoph), (New York: Alfred Knopf, 1961 ); Stephen Toulmin, Tilt
Piau of ReaJon i11 Ethics (Cambridge: Cambridge Uni\lersit), Press, 1953),
2. SceJohn Rawl s,A TJltory' ofJwtict, 2d printing (Cambridgc. MA: Han'ard Uni\lcrsity
Prcss , 1972); John Rawl s. "Kalllian ConstTlJctivism in Moral Philosophy: The Dewey
Memorial LccLUres 1980," Journal of Philosoph)'. 77 (Seplcmbcr 1980). pp. 515-572;
Lawrence Kohlbcrg, EJ.Soyi on Moral Drodopmnlt, vol. I and TJu Psychology of Moral
Droelopmt1lt, vol. 2 (San Francisco: Harper and Row, 1984) .
3. Alasdair Macint yre and SLanlc)' Haucrwas, Revisions (Notre Dame: Uni\lersity of
Notre Damc Press, 1983). p. vii . For a general discussion of this COnlext, see also Fred
Dallmayr's "Introduction," this volume.
4. Ariswlle, in Tht Bruu Works of Aristotlt, cd. and trans. by Richard McKeon (New
York: Random House, 1945).
5. For Hegel's early critique of Kalll , see 'The Spirit of Christianity and its Fate," in
G. W. F. Hegel , Earl)' Theological Wrilingl, T. M. Knox. trans. (Philadelphia: University
of Pennsylvania Press. 1971 ), pp. 182-302; C. w. F. Hegel , Htgel'l Pht1IOmen%g)' of
Spirit , trans. by A. V. Mill er (Oxford: Clarendon Press, 1977). ch. 6, scction C;
Phi/o.w/lll)' of Irtms. hy T. M. Knox (Oxford: Oxford Press. 197:\ ). 10,
Additiull. pp. 3!IIT.; II cgel, Scirllrr a/uJgi,. /\' . V, Miller, 0 ':1115. (Ncw )'m·k: Ilum:lni lics
J1ress, 1969) , pp. 133f('
6. Karl-OliO Allel . "Kant, Hegel und das aktueJle Problem der normativen Grundlagen
von Moral und Recht," in DiJkurs ulld Vel'mlllJlorlUlIg (Frankfurt : Suhrkamp, 1988). pp.
69- 103: Apel, "Kann der postkantische Stal1dpunkl noch einmal in substantielle Sin-
li chkeil aufgcllobt=n werden? . . " in ibid .. pr. 10:\-154:Jurgcn Habcrmas . "Moralital
und Sitt li chkeit. Treffen Hegels Einwande.gegen Kant auch auf die Diskurselhik zu?"
in Af oroilliit uud Siu/i,hktil. Dru Pl'OblnT! Hegtl.$ ul1d die Disli1mtlhik. ed. by W. Kuhlmann
(Frankfurt : Suhl'kamp. 1986), pp. 16-38.
7. Herbert Schnadelbach. "Was ist NcoaristotelisnlUs?" in Moralittit und Sittlichktil , W.
Kuhlmann . ed .. pp. 38-64 ; English trans. as "Whm is Nco-ArisIOlciianism?" in Praxu
lutemaliol/al, vol. 7, no. 3/4 (October-January 1987), pp. 225-238.
H. For an excell ent su.-ve)' of the \'(ll'ioll s stnmds of IIcn-Ari stOl clianis lIl in cOlHcmpo-
rar)' discuss ions. and in particllbr for thl' seriOliS dilTerences between German ;md
Sc)' la Bcnhabib
Anglophune nco·ArislOlciian tTench . sce Maurizio I'asscrin d'Enlrb'cs. "Aristotle or
Burke? Some COTnlllents 011 H. Schnadclbach's ' \Vhat is Nco-Aristotcliani sm?'," in
Praxu 'nlf'nltll;rmal. \'01. 7 .. nos. 3/4 (DClobeT 1987-Januar), 1988). pp. 238-246. I
discuss communi tarian ph ilosophies in "Aut onomy. Modcrnit), and Community. An
Exchange BCl\veen Commu nitarianism and Critical Social Theory," in Zwischt llbtt r(l(.h-
IUlIgt" 1m Proll'S! da Aufkliinmg, ed. b)' A. Honnclh. T. A. McCanh)'. Claus arret and
Albrecht Wctlmcr ( F"3nkrun: Suhrkmap. 19S8). pp. 373-395.
9. H,3l1s-Ccorg Gadamcr. Truth alld Method (New York: Seabury Press, 1975).
10. Gadamcr, "Hermeneutics as I' r;:lclica1 Philosophy," in Rm.!on ;" Iht Agt oJ ScittllU,
tram •. by Frederick G. Lawrence (Ca11lbridge, MA: The MIT Press, 1981). pp. 88-
113. I have not incl\lded Hannah Arelllh's work ul1der this GHegor iz<ltion. because in
matlers of moral as opposed (Q political philosoph)' Arendt remained a Kantian
thinker. I deal with some aspects of this admittedly not generall), shared interpretat ion
of Hannah Arendt 's ",'ork in my "J udgment and the Moral Foundations of Politics in
Hannah Arendt's Thought." i n PoLitical Thttory', vol. 16, no. I (February 1988), pp. 29-
II . See Hans·Ceorg Gadamer. "Hegel's Philosophy and il.!i Aftereffects until Today,"
and "The Heri tage of Hegel ," in RttQ.Jon in tht Agt of Sciuu:e, pp. 2 1-38 and 38-69;
and Gadamer, Httgel'j Dialutic. trans. by P. Christopher Smith (New Haven: Yale
University Press, 1976).
12. I. Kalil, Cnmdlttgtmg dt:r MttlaphyjVf tkr Sillen, trdll5. by H. J. Pat on as Tlu Moral
Law (London: Hutchinson, 1953), p. 421.
13. G. W. F. Il egcl, Nnt llml Law, trans. b)' T. M. Knox ,md illtroti . by H. B. Acton
(Philadelphia : Uiliversilr of Pennsylvania Press. 1975), pp. 77-78.
14. For some "ecelll considerations on Hegel's critique of Kantian elhics, see Jonalhan
Lear. "Moral Objectivity." in Objulivil)' and Cullumi Divt!rgt!nCt, ed. by S. C. Brown
(Cambridge: Cambridge University Press, 1984). pp. 153- 171.
15. The Kalltian principle or universaliz.abilit y does nOl, of course, dictate any specifiC
content to the of justice; rather, it is operative in the construction of lhe
"original position," as the privileged moral vantage point rrom which to enter illlo
dclil>crations about mailers of justice. See Rawls, A TJuory' of passim.
16. Onora O·Neill. "Consislcl1c}, in Action ," in Mtlmfily ml(J Vlllllfnnlil)'. cd . hy Nclsnn
T. !'olter and Mark Tillllllom (Dordl't'cht : U. Hcidcll'uhlish illg. IHH5). pp. 159- 186.
17. It,;d .. p. IGR.
18. Ib;d .. p. 169.
19. Ib;d.
20. See Alall Ccwirt h. Rl'tJ.{oll (wd lI1 omlil.)' (Chicago : Universit), of Chicago Press, 1978).
pp. '18-129.
21. Alasdair Aftrr Vjrtur (Notrc D,11l1C: Univcnit}' or Not re Dame Press,
1984), p. 67.
22. Sec Michild W"lzcr, "i\ CI'ili<l ul' or PhiimopiliGl1 (;lIlIvcrs;l(ion." in Thr PhilosolJhical
Forum. vol. xxi . nus. 1-2 (Fail-WillieI' IY89-90), lip. IR2- 197.
After ..... ord
23. S. Bcnhabib, "The Methodological Illusions of Modern Political Theory: The Case
or Rawls and Habermas," in Ntut Ht!fu fur Philosophit! , no. 21 (Spring 1982), pp. 4i-
24. 1 have developed lhis argument more extensively in "Liberal Dialogue \'s. A
Discourse Theory of LegitimaC)'," in Libt:ralism and thtt Moral Lift!, cd. by Nanc), Rosen-
blum (Cambridge, MA: Harvard University Press, 1989) , pp. 143-157.
25. J. Habcrmas, "Diskursethik. Notizen zu cineOl Begrundungsprogramm." in 11101'-
albt!WusSlsttin und /tommunika. tivt!s Harukln (Frankfurt: Suhrkamp. 1983), pp. 96-97;
Engl ish translation in this volume.
26. K.-O. Apcl , "The Problem of Philosophical Fund .. rnental Ground ing in Light of
a Transcendental Pragmatics or Language, " in K. Baynes, J. Bohman. and T. A.
McCarth)'. cds. After Philojophy (Cambridge. MA: The MIT Press, 1987), p. 27i .
27. T. McCarthy, "Rationality and Relati vism. Habcrmas's Overcoming or Hermeneu-
tics," in Thompson and Held, cds. Habt!TmO.S: Critical Dt!balti (Cambridge, MA: The
MIT Press, 1982), p. 74.
28. The meLastatus of such criticism-whether such social criticism needs to be philo-
sophically grounded in some generall y acceplable sys tem or norms or whether it can
be exercised immanently. by internally appealing to, critiquing. or debunking thc
norms or ;l. given culture. community. and grou J>-is what sharply dividcs social
theorists like Habermas ancl Michael Walzer. Given also the large area of substantive
agreement among them upon the need for the radical-democratic reconstruction of
laI c-capita li st societies. it is worth pursuing what stalUs these meta philosophical dis-
agreements-immanent or transcendental; relativist or univers;llisl-have. For Walzer,
sec l uttrprttlatio'l aud Social Criticism (Cambridge, MA: Harvard, 1987).
29. In this volume, p. 90.
30. See Barbara Herman's excellent discussion, "The Practice of Moral Judgment," in
TJuj ounwl of Philosoph), (August 1985). pp. 414-436.
31. Agnes Heller, "The Discourse Ethics of Habermas: Critique and Appraisal ," in
Thuu £lLvnl , no. 10/ 11 ( 1984-B5), pp. 5-17, here p. 7; sec also Albrecht Wellmer.
£ thik Imd Dialog. £It!mttnu drs moraliscJlffl UrteilJ bt!i Kmll Iwd ill dtr (Frank.
rurt : Suhrbmp, 1986) : Otfried Hofre. " Kantian Skepticism Toward the Transcen-
demal Ethics of COllllllunicatioll," in this vlllulIlC.
32. Wdlmer, £thik uud Dialog, p. 63. My translation.
33. Heller, "The Discourse Ethics of Ha?ermas," p. B.
34. Well mel', Elhik urn! Dialog, p. 64 .
35. Well mer also discusses th is principle in ibid., pp. 65fr. Wellmcr's argument is that
sincc the universal adherence to this norm would climinale precisel y lhose cases like
the legilimate right to self·defense and junifi ed punish ment, the discourse ethics is
obli ging us to think of what is morall y right on I)' in relation to counterfaclual ideal
c011(lil ions and not real ones. Wcllmer concludes that the conditions of action suggested
by "U" can properl y be thought of as those appropriate for a "kingdom of ends." But
the fact thaI in actual life we must always make justified exceptions to such general
moral rules ha!i litt le to do with the question whethcr our morallheor), is aule to justir y
Sc)'la Bcnh"hib
what \\'(' intuiti vel y kll Ow to u(' a r ight moral priucipl c. i.e., ill this case nOI to inflict
unnecessary sufferi ng.
36. See Wellmcr. ibid., pp. 121-122; Heller, "The Discourse Ethics of Habermas," p. 9.
3i. Charles Taylor. "Di e Motive ciner Verfahrenselhik," in Moralilii/fwd Siuiic.hkl'il, W.
Kuhlmann , cd .. pp. IOlrr.
38. See Mich;;tel Sandel , "Inlroduclion,"lo Uluralism and its Critic.;, MichaclJ . Sandel,
ed. '(New York: New York Universi ty Press, 1984), pp. 1-13.
39. Part of the di!icussion which follows has appeared in S. Benhabib. "Autonomy.
Modernity and Communit ), : An Exchange Between Communitaria nism and Critical
Social Theor)'," pp. 377-79.
40.John Rawls , A Theo")' of jus/iu, pp. 398f['
41 . J. Habcrmas, "A Repl y to My Critic5," in Habmnas: Cdtical Debates. J. Thomp50n
and D. Held , ed5 .. p. 24 6.
42. Haberma5, "Ego Development and Moral in Communication and the Evo-
Iwion oj Society, tram. by T . McCarthy (Boslon: Beacon Pre5s. 1979). pp. 78ff.
43. See Bernard Williams. Ethics'muJ the Limit..! oj Philolophy (Cambridge. MA: Harvard
University Press. 1985); Charles Taylor, Philruophy and tM Human Scimas. vol. 2 of
Philolophical Papen (Cambridge: Cambridge University Press. 1985). pp. 23-247.
44 . Habcrmas, "Moralbcwusstscin und kommunikativcs Handcln," in MoralbcwlLutst in
und lcommunikotivtJ HaudeJrl. PP' l44ff.
45. MacIntyre, After Virtue, p . 204.
46. Hegel , "The Positivity or the ChriStian Religion," in Earl), Theological Writings, p.
47. Hegel , Philosuphy oj Right, Note to para. 27.0. pp. 168-69.
48. See Robert Spaemann. "Die Utopie del' Herrschaftsfreiheit ," in Merkur, no. 292
(August 1972). pp. 735-752; Niklas Luhmann and J. Haocrmas, Theorie del' Geseloch4Jt
oder Soziallcdlllologi f .Wa.I/fil/t' 1 die S)'.demJoncluHlgr (Frankfurt : Su hl'kamp. 1976).
49. For a provocat ive consideration of the implications or discourse theory ror a critical
thcory of ne\\' social mo\'emcnts in Weslern and Soviet-type societies, see Andrew
Arato and Je'lIl Cohen, Civil Socil!l.v alid Social ThlOry (Cambridge. MA: The MIT Press,
50. I havc de'llt with tht, or the conctpt or t he "suppressed generalizable
intcrest" cXlcnsivcir in Crith/lit, Nonn and Ulopin (New r ork: C.olumbia University
Press. 1986). pp. 3 1 orr.
51 . Nancy FI'aser, "'-O\\'ard a Discourse Ethic of Solidarit y, " Praxis J'lltnilltimlOl, \'01. 5.
no. 4 (janu3t·,. 1986). p. 425.
52. Sec Amclie ROflY, "ColT!,lIunit )' as the Context of Character," part four in Mind
Action. ESjo,rs ill tht Philosoph." of Mi,ui (Boston: Beacon Press, 1988), pp. 27 1-34 J;
Martha Nussbaum, Till Fms.rilil.l' oj Good,u.u (Cambridge: Cambridge Universil ), Press,
1986): Annette Baier, "What do Women Want in Moral Theor),," NOILS , no. 19 (1985),
pp. 53-63, and A. Baier, "J-Iume. The Women's Moral Theorist ?" in Womttl and MoTtlI
Thtor)', ed. by E. F. Killay and Diana T. Meyers (New Jerscy: Rowman and Littlefield,
1987), pp. 37-56; Lawrence Blum, Frie1id.ship, Allruism and Moralit), (London: Routledge
and Kegan Paul : 1980); Ursula wolrr. DaJ Problem del moralu, hm Sol/rns (Berlin and
New York: de GI'U}'ter, 1983); Virginia Held, " Feminism and Moral Theory," in Women
(HId Moral Theory', E. F. Kina)' ami Diana T. Meyers. cds.; Sara Ruddick, Maternal
Thinking (Boston : Beacon Press, 1989).
53. Virginia Held, "Feminism and Moral Theory," pp. 114ff.
54. I have di scus5cd the gender-bias or modern conceptions or autonomy in "The
Generalized and the Concrete Other: The Kohl bcrg-Gilligan Controversy and Moral
T heory," in Women and Moral TheoT)', Killa), and Meyers. cds. , pp. 154-178; reprinted
in Benhabib and Cornell, cds. Feminism aJ Crilique (Minnesota: University or Minnesota
Press, 1987), pp. 77-96.
55. See Amclie Rort)'. "Vinues and the Vicissitudes," in Mind in Action, pp. 3 14ff.
56. See. in particular, Herbert Seh nadelbaeh's refiectiom in "Remarks About Ration-
alit}' and Language." in this volume.
57. For a recent statement of the hermeneutic critique of ethical theory fTom this
point of view, see Ronald BeineI', "Do We Need a Philosophical EthicS? Theory,
Prudence and the Primacy of Ethos?", The Philosophical Forum, vol. xx, no. 3 (Spring
1989), pp. 230fr. See also Alessandro Ferrara for an incisive probing or discourse
ethics from this viewpoilll, "Universal isms: Procedunll, ContexLUalisl and Prudential,"
Philo.laph)' and Social Cdli,ism, vol. 14, nos. 3-4, pp. 243-271.
58. See Habcrmas, "Moralbewusmein und kommunikatives Handeln," pp. 187ff.,
where the work of Norma Haan and Carol Gilligan is discussed; Ape!. "Kann del'
pos tkantische Standpunkt der Moralitiit noch einmal in substanticl1e Sittlichkeit auf-
gehoben we rden?", pp. 103ff.
59. Hannah Arendt, "The Crisis in Culture," in Belwttm Prut and Future. Six ExercutJ
in Political Thought (New York: Meridian, 196 1), pp. 21-22.
60. For a more detailed presentation of the rollowing argument, see S. Benhabib,
and the Moral Found;lIions of Politics in Hannah Arendt 's Thought,"
J' olitirni TllfoT)'. vol. 16. 110. 1 (FcbruOl"), 1988). pp. ::l41T.
61. 1 have dealt with the I)'pes of "es pol1ses 1O modernit ), among contemporary social
theoristS in "Autonomy, and Communi t)'. An Exchange Between Com·
munitarianism and Critical Social Theory."