Upload
yorluisguzman
View
216
Download
0
Embed Size (px)
Citation preview
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
1/216
On the Pragmatics of Social Interaction
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
2/216
Studies in Contemporary German Social Thought (partial listing)
Thomas McCarthy, general editor
Karl-Otto Apel, Understanding and Explanation: A Transcendental-Pragmatic Perspective
Seyla Benhabib and Fred Dallmayr, editors, The Communicative Ethics Controversy
James Bohman, Public Deliberation: Pluralism, Complexity, and Democracy
James Bohman and Matthias Lutz-Bachmann, editors, Perpetual Peace: Essays on Kant's
Cosmopolitan IdealCraig Calhoun, editor, Habermas and the Public Sphere
Jean Cohen and Andrew Arato, Civil Society and Political Theory
Maeve Cooke, Language and Reason: A Study of Habermas's Pragmatics
John Forester, editor, Critical Theory and Public Life
Jiirgen Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law
and Democracy
Jiirgen Habermas, Justification and Application: Remarks on Discourse Ethics
Jiirgen Habermas, On the Logic of the Social Sciences
Jiirgen Habermas, The Inclusion of the Other: Studies in Political Theory
Jiirgen Habermas, Moral Consciousness and Communicative Action
Jiirgen Habermas, The New Conservatism: Cultural Criticism and the Historians' Debate
Jiirgen Habermas, The Philosophical Discourse of Modernity: Twelve Lectures
Ji irgen Habermas, Philosophical-Political Profiles
Ji irgen Habermas, Postmetaphysical Thinking: Philosophical Essays
Jiirgen Habermas, On the Pragmatics of Communication
Ji irgen Habermas, On the Pragmatics of Social Interaction: Preliminary Studies in the Theoryof Communicative Action
Jiirgen Habermas, The Structural Transformation of the Public Sphere: An Inquiry into a
Category of Bourgeois Society
Jiirgen Habermas, editor, Observations on "The Spiritual Situation of the Age "
Axel Honneth, The Critique of Power: Reflective Stages in a Critical Social Theory
Axel Honneth, The Struggle for Recognition: The Moral Grammar of Social Conflicts
Axel Ho nn et h a nd Han s Joas, editors, Communicative Action: Essays on Jiirgen
Habermas's The Theory of Communicative Action
Axel Honneth, Thomas McCarthy, Claus Offe, and Albrecht Wellmer, editors, Cul-
tural-Political Interventions in the Unfinished Project of Enlightenment
Axel Honneth, Thomas McCarthy, Claud Offe, and Albrecht Wellmer, editors, Philo
sophical Interventions in the Unfinished Project of EnlightenmentHans Joas, G. H. Mead: A Contemporary Re-examination of His Thought
Cristina Lafont, The Linguistic Turn in Hermeneutic Philosophy
Maurizio Passerin d'Entreves and Seyla Benhabib, editors, Habermas and theUnfinished Project of Modernity: Critical Essays on Th e Philosophical Discourse ofModernity
Kirk Pillow, Sublime Understanding: Aesthetic Reflection in Kant and Hegel
Albrecht Wellmer, Endgames: The Irreconcilable Nature of Modernity. Essays and Lectures
Albrecht Wellmer, The Persistence of Modernity: Essays on Aesthetics, Ethics and
Postmodernism
Rolf Wiggershaus, The Frankfurt School: Its History, Theories, and Political Significance
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
3/216
On the Pragmatics of Social Interaction
Jiirgen Habermas
translated by Barbara Fultner
The MIT Press Cambridge, Massachusetts
Preliminary Studies in the Theory of Communicative
Action
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
4/216
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
5/216
Contents
Translator 's In tr od uc tio n vii
Reflections on the Linguistic Foundation of Sociology:
The Christian Gauss Lecture (Princeton University,
February-March 1971) 1
I Objectivist and Subjectivist Approaches to Theory
Fo rm at io n in th e Social Scien ces 3
II Th e Phe no men olo gic al Const itut ive Th eo ry of
Society: The Fundamental Role of Claims to Validity andthe Mon adol ogic al Fo un da tio ns of Intersubjectivity 23
III Fr om a Const itutiv e T he o r y to a Co mm un ic ati ve
Theory of Society (Sellars and Wittgenstein):
Com mun icat ive an d Cognit ive Uses of Lan gu age 45
IV Un iversal Prag matic s: Reflections on a T h e o r y of
Commu nicat i ve Co mp et en ce 67
V Tr ut h an d Society: T h e Discursive Re de mp ti on of
Fa ctua l Cla ims to Validity 85
Intentions, Conventions, and Linguistic Interactions
(1976) 105
Reflections on Communicative Pathology (1974) 129
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
6/216
I
vi
Notes
Index
171
183
Contents
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
7/216
Translator's Introduction
Philosophy in the twentieth century, in both its analytic and conti
ne nt al tradition s, ha s be en sh ap ed by wha t has co me to be kn ow n as
"the linguistic turn." Be it in metaphysics, epistemology, or value the
ory, philosophy of language has become a keystone of conceptual
analysis. Most pr of ou nd ly pe rh ap s, th e linguistic tu rn has affected
th e co nc ep tio n and un de rs ta nd in g of reas on. It is n o long er possible
tod ay to de fe nd th e univ ersal validity of a tr an sc en de nt , objective
reason, nor can language be regarded any longer as a neutral tool at
reason's disposal. The role of this movement in critical theory is due
in large me as u re to the work of Ji ir ge n Ha b er ma s . An d yet, in an increasingly po st mo de rn era, Ha be rm as has re ma in ed a de fe nd er of
moder nity. Whil e th e reason of th e en li gh te nm en t has come u n d e r
general at tack, he continues to endorse i ts emancipatory potential ,
albeit in the altered form of a "postmetaphysical" reason that is al
ways situated in contexts of interaction. Habermas locates the roots
of rationality in the structures of everyday communication such that
the critical po wer of re as on is i m m a n e n t in or di nar y lan gu ag e fro m
th e start. T h e ai m of his intersubjectivist ac co un t of "co mmu nic ati ve
reason" is to displace both subjectivist accounts that cling to Carte
sian co nc ep ti on s of mo no lo gi cal selfhood an d objectivist acco un ts
that ignore the agent's perspective entirely.
The essays and lectures collected in this volume explain why
Habermas considers a linguistic turn to be necessary, how he thinks
it is to be worked out, and what he takes its implications to be. They
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
8/216
viii
Translator's Introduction
address questions concerning the nature of social interaction and i ts
connection to communication, and they trace the implications for
de ve lo pi ng an ad eq ua te social theory. Th ey will be of int ere st n ot
only to readers who have followed Habermas's intel lectual develop
ment but also to those looking for an introduction to his theory of
communicative action. More generally, philosophers of language will
find in these essays a hos t of or ig inal idea s on the re la ti on sh ip be
tween language and society.
Since i ts pub licat ion in 1981 , Ha be rm as 's The Theory of Communica
tive Action1 has been widely acclaimed for its co nt ri bu ti on to phi loso
ph y a n d social theory. However, its two vo lum es are da un ti ng , no t
only for reasons of length, but owing to the breadth of its subject
matter and the denseness of i ts argumentation. The essays and lec
tures in the present volume provide a less arduous route to under
standing the theory behind that larger work. They are a part ial
translation of Vorstudien und Erginzungen zur Theorie des kommuni
kativen Handelns, which was pu bl is he d in 1984 an d con tai ns not on ly
preliminary studies leading up to the fully developed theory of com
municative action, but also several essays that complement The Theory
of Communicative Action in important respects. The first part of this
volume, "Reflections on the Linguistic Foundation of Sociology,"
formed the text of the Gauss Lectures that Habermas delivered at
Princeton in 1971.2
The se lectu res are drive n by th e sam e the or eti cala im that underl ies The Theory of Communicative Action, namely, the at
tempt to ground a theory of society on the foundations of communi
cative rationa lity. Th ey pr ov id e a larg ely self -con tained ac co un t of
the philosophical motivations behind the theory of communicative
action as well as an elucidation of its theoretical grounding in what
Habermas called first universal an d later formal pragmatics. Th e pre
se nt ati on of th e issues he re is m or e accessible th an in The Theory of
Communicative Action precisely becau se Ha be rm as is occ up ied withlaying out the general parameters of his project and situating it rela
tive to other theories of society. Moreover, he discusses certain issues
here, such as "systematically distorted communication," that he has
n o t ad dr es sed in th e sam e detail again. In shor t, the se "p re pa ra to ry
studies" offer imp or ta nt aids to un de rs ta nd in g his ma tu re philoso
phy and social theory.
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
9/216
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
10/216
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
11/216
x i _ _ _ _Translator's Introduction
reject "mea ni ng " as a basic, ir re du cib le co nce pt. In Hab er ma s' s view,
to un de rs ta nd the na tu re of intersubjectivity, we n e e d to u nd er st an d
how agents interacting with one another arrive at the same interpre
tations of their situation; in this respect, intersubjectivity is grounded
in sameness of meaning. Habermas regards communicat ion in lan
gu ag e as th e pa ra di gm case of achie ving such "identity of m e an in g "
and thus holds that l inguist ic normativity cannot be reduced to merebehavior in the sense of the objectivist paradigm. To distance himself
from the subjectivist paradigm, Habermas goes on to argue that the
normativity of me a n in g mu st be bas ed on th e intersu bjective (r ath er
than merely subjective) validity of a rule. Thus, intersubjectivity is to
be explained on the model of how two different individuals are able
to use a term with one and the same meaning.
That interlocutors succeed in assigning the same meanings to
their actions and circumstances attests to their mastery of what
Ha be rm as calls "co mmu nic ati ve c om pe te nc e. " O n this view, if we can
delin eate the struc tur e of comm uni cati ve c om pe te nc e, we will also
have captured the structure of communicative rat ionali ty. Haber
mas's formal pragmatics aims to provide a rational reconstruction of
this competence, that is, to transform an implicit knowledge, a
know-how, into a "second-level know-that."5
This is not to say, of
course, that a speaker actually has representations of the recon
str uct ed kno wled ge "in the he ad. " He r know-how is pr eth eor eti c: a
skill or mastery of a practice in the Wittgensteinian sense. Subjects
capable of speech and action have acquired a tacit mastery of
rule-governed practices that enable them to reach a mutual under
sta nd ing with on e an ot he r ab ou t the world. Successful co mmu ni ca
t ion requires, then, that the rules consti tuting such communicative
co mp et en ce be valid intersubjectively.
H a b e r m a s uses Huss er l, Sellars, and Wit tgenste in as foils for hisown account. The subjectivist and objectivist paradigms are repre
sente d by E d m u n d H uss erl an d Wilfrid Sellars respectively. In The
Cartesian Meditations, Huss er l explicitly sets him self t h e task of re co n
structing intersubjectivity from a subjectivist starting point. (The tra
di ti on of in ter pr et ive sociology ini tia ted by Alfred Schiitz is r oo t e d
in this Husser l ian enterpris e.) H ab er ma s argu es that Hu sserl ' s
phenomenology of consciousness fails in the end to establish the
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
12/216
xii
Translator's Introduction
intersubjectivity of a community of transcendental egos, and so he
tu rn s to Sellars. Sellars's ap p r oac h is initially pr om is in g bec ause he
wants to model intentionality and the structure of thought on the
structure of language: He takes a l inguist ic turn within the philoso
phy of mind. However, his understanding of language, according to
Habermas, is essentially objectivist. That is, speakers come to mean
the same things by the same words because they respond similarly totheir environment and mutually observe each other 's responses
from a mon olo gic al , thir d-pe rson perspective ra th er than from a
dialogical second-person perspective. They are "monological lan
gu ag e us er s" with a full, in te nt io na l (i nn er ) life of beliefs a n d de sires ,
but lacking any interpersonal re la t ionships. But monological lan
guage that cannot be used for purposes of communicating with oth
ers, H ab er ma s ar gu es, is no t really lan gu ag e at all. Th us , in diffe rent
ways, Husser l a n d Sellars bo th pre su pp os e ra th er than acc ou nt forthe existence of intersubjectivity.
It is no accident that Habermas's argument against Sellars is remi
nis cent of Wittg enst ein's private-language arg um en t. Samenes s of
meaning is grounded in the validity of rules, and Habermas, follow
ing Wittgenstein, argues that a subject cannot follow rules in isola
tion . If s o m e o n e is follow ing a ru le , it mu st be at least in pr in ci pl e
possible for so me on e else to check wh et he r she is following th at ru le
correctiy; o n e per so n' s rule-following beh avi or is, in ot he r word s,
subject to eval uati on an d criticism by an ot he r. Thi s pr ec lu de s any
mo no lo gi ca l ac co un t of rule-following, for it pr es up po se s that differ
ent people have the same competence and are mutually capable of
assessing each other 's performance. ( i Wittgenstein emphasized that
me an in g is a ma tt er of use an d th at word s an d sen ten ces are used in
interaction with others; his "use theory of meaning" was in this sense
inh ere ntl y pra gm ati c an d intersubjective. As action a n d lan gu age areintimately interwoven, to understand an utterance is to grasp i ts role
in a language game, that is, to understand it as a move in a rule-
governed, interpersonal activity. Thus being able to engage in a
lan gu age g am e pr es up po se s sharin g a fo rm of life with on e' s int er
locutors. The rules constitutive of such language games are not stipu
lated arbitrarily, but have the status of conventions, a topic to which I
shall return below. Habermas elaborates on Wittgenstein's account
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
13/216
xiii
Translator's Introduction
in two ways. First, h e aim s to dev elo p a th eo ry of the st ru ct ur es of
intersubjectivity: Though Wittgenstein's language games clearly sup
pose dialogical relat ionships among part icipants in interaction, he
does no t analy ze these relatio nsh ips as such . Sec on d, H a b e r m a s
wants to do m o r e jus tice t h a n Witt gen stei n d id to th e fact th at lan
guage refers to the world.
Th e early Wit tge nst ein a t t em pt ed to ela bor ate a pu re ly cognit ivelan gua ge, the p ri me fun cti on of whic h was to re pr es en t the totality of
facts that make up the world. The later Wittgenstein aborted that at
tempt because, in Habermas's terms, he discovered communicative
language use(s). In other words, he came to realize that language
can be used for all sorts of purposes other than cognitive ones and,
according to Habermas, henceforth mistakenly downplayed the im
po rta nc e of the cognitive use altog ether . By cont rast, H a b e r m a s
maintains that reaching mutual understanding requires a speakera nd hea re r to oper at e at two levels: th e level of intersubjectiv ity on
which they speak with one another, and the level of objects or states
of affairs about which they communicate. His discussion here is argu
ably the best, most extensive elucidation of his conception of the
"double structure of speech." He makes it clear that the two uses of
language are interdependent. "A co mmu ni cat iv e theo ry of society mu st
do justice to the do ub le cognit ive- communicati ve st ruc tur e of
spe ech " (p. 64) . This dual str uct ure und erl ies the reflexive cha ract er
of language: Natural languages can function as their own meta-lan
guages, as Donald Davidson, for example, has also pointed out. Ac
cording to Habermas, we cannot communicate about things or states
of affairs in th e wor ld wi th ou t also "m et a- co mm un ica ti ng " ab ou t
what we are doing or how we are using the content of what we are
saying. It is here that speech act theory enters the picture. Every
speech act takes the form Mp, whe re M expresses the i l locutionary
force of the ut ter an ce (th e comm uni cati ve dimen sion ) a nd p ex
presses i ts pre pos it i ona l co nt en t (the cognit ive dim ens io n) ab ou t
which mutual understanding is to be reached. In this sense, al l
sp eec h acts ha ve a cognitive an d a co mmu ni cat iv e di me ns io n.
Ha be rm as 's key mo ve in l inking com mun icat ive rat ionali ty with
a theory of meaning is to connect the theory of meaning with a the
ory of ar g um en ta ti o n an d justification. T h e r e is, as h e put s it, a
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
14/216
xiv
Translator's Introduction
"validity basis" to speech; all sp eec h acts carry an implici t co mmit
m e n t to justification, to giving reason s that back on e' s claims. Wh e n
we use sp ee ch acts to communica te wi th o ne another , we move , as
Robe r t Br a ndom has recent ly re min de d us, in "the space of giving
and asking for reasons."7
Or, as Hab e rm as pu t s it , every speech act
raises certain claims to validity th at ar e o pe n to bein g chal lenged a nd
defended with reasons. T h e i l locut ionary com po ne nt of an ut terance expresses validity claims a speaker raises in per for min g speech
acts. Habermas init ial ly identifies four such claims: intelligibility,
t ru th , normat ive Tightness, a n d sincerity or t ruthfulness . That is, in
m a ki ng an u t t e rance , a speaker simultaneously raises the claims that
what she says is intelligible, that the propos i t ional c ont ent of what
she says is t rue, that she is m a k i ng the u t te rance in the appropr ia te
social context, a n d tha t she is spe aki ng truthfully. Fol lowing o n this,
Habermas classifies speech acts into four types, each of which corre sponds to one of the fou r validity claims: co mmu nic ati ves (e.g.,
speaking, asking, replying), constatives (e.g., reporting, asserting,
claiming), regulatives (e.g. , Ordering, requesting, demanding, re
m i n d i n g ) , a n d expressives (e.g., kno wing , thinki ng, fearing, ho pi ng ,
wishing). Communicative speech acts a r e used to m a k e explici t the
na t u r e of an u t te rance itself. In constative speech acts, speakers rep
resent states of affairs in the objective world a n d refer to something
in that world. In pe rf or mi ng regulative spe ech acts, speakers estab
lish intersubjective relationships with interlocutors and thus relate to
a social world. In expressive speech acts, speakers refer to things in
their subjective world by mak in g public inte ntio ns, desires, or other
private states or occurrences . In The Theory of Communicative Action,
the number of validity claims is r e duc e d to three ; intelligibility drops
out , leaving t r u t h , normative Tightness, and sincerity.8
Communicative action takes place against a back gro und consensus that it renews a n d develops. W h e n commu nicat iv e interact i on is
pr oc ee di ng smoothly, inte rlo cuto rs m a k e wha t the y ar e saying intell i
gible to one anoth er , g ran t what they are saying to be tr ue (i.e., the y
assume the refer ential expres sions they are using pick ou t objects to
which the at tributes they predicate of th em actually appl y), r eco g
nize the Tightness of the n or m tha t the speech act claims to fulfill,
and don't doubt each other 's sinceri ty. In shor t, they mut ual ly acc ep t
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
15/216
xv____ •Translator's Introduction
the validity of the claims being raised. In this "normal" case, a
speaker uses expressions such that the hearer understands the
speaker as the speaker wants to be understood, she formulates prop-
ositional contents such that they represent experiences or facts, she
expresses her intentions (sincerely), and she performs speech acts
such that they conform to recognized norms of accepted self-images.
At the same t ime, part icipants in communicative action are assumed
to be pr ep ar ed to reac h mu tua l un de rs ta nd in g— th at is, their a tt i
tud e is co mmu ni cat iv e ra th er th an strategic (o ri en te d towar d realiz
ing one's own ends). As such, they are assumed to be accountable,
th at is, ca pa bl e of justifying th eir actio ns a n d expr essio ns. A cco un t
ability thus refers to a general presumption of rationality, cashed out
in te rm s of on e' s re ad in es s to justify the cla ims o n e raises. Becau se,
normally, in raising validity claims, a speaker takes on the warrant to
make good on them, formal pragmatics as a theory of "communicative rationality" can serve as a foundation for a critical theory. As a
speak er can be call ed up o n to justify th e claims ra is ed in her utt er
ances, the bu r d en of jus tifi catio n a n d th e possibility of cri tiq ue ar e
buil t into the very structure of language and communication.
Wh en the consensus unde rlyi ng smoothly funct i oning co mm un i
cative interaction breaks down and the flow of the language game is
in te rr up te d, p art icu lar claims to validity may be them ati ze d. To re
deem problematic claims to truth or to normative Tightness, we must
resort to a level of argumentation that Habermas calls discourse,
through which we seek to attain a rational consensus on these claims.
But how are speakers able to distinguish a true (or rational) from a
false (or merely co nt in ge nt ) consen sus? No te tha t we rou tin ely as
sume, as a matter of fact, that we are able to do so, and that, in this
sense, speec h is fun dame ntal ly rat ional . To mo de l the ass ump tio ns
buil t into the ideal of rat ional discourse, Habermas introduces the
notion of the ideal speech situation. T h e ideal sp eec h situat ion is sub
j ec t only to the "unforced force of t he be t te r a rgument " ; it is devoid
of all other constraints. All interlocutors are equally entitled to make
assertions, raise questions and objections, or provide justifications
for their positions. And all express their true intentions. It is crucial
to remember that discourses as a matter offact usually do not manifest
the conditions of the ideal speech situation, but the model can serve
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
16/216
xvi
Translator's Introduction
as a standard in identifying deviations from the ideal of rational
consensus.
T h e que sti on of wh en ce th e justifications for th em at iz ed validity
claims are drawn brings us to the notion of the lifeworld, which is
complementary to that of communicative action. The l ifeworld pro
vides a context of relevance within which communicative actions
(and actors) are "always already" situated. As such, it always remains
in the ba ck gr ou nd , st an di ng "at the backs" of part ic ipant s in co mm u
nic ati on, as it wer e. It ca nn ot be tr an sc en de d: Speak ers and actor s
cannot act by placing themselves outside of it. It has, on the contrary,
a tr an sc en de nt al ch ar act er insofar as it fun ctio ns itself as a co nd it io n
of possibility for communicative action. In this sense, it functions as a
ba ck gr ou nd of mu tu al intelligibility. I nte r subjectively sha red , it
makes possib le th e smo oth fu nct ion ing of everyday co mm un ic ati ve
action. In general, speakers do not have explicit but only tacit knowledg e of it; no ne th el es s, th e lifeworld provides com mu ni cat iv e actor s
with a shared stock of taken-for-granted interpretations on which
they can draw in trying to understand others. In discourse, elements
of this imp lic it kn ow led ge can be r en d er ed explicit in or d er to re
deem validity claims that have been challenged. This connection be
tween univ ersal prag mat ics a n d th e lifeworld is discussed in th e
fou rth Gau ss Lect ure , wh er e Ha b er ma s def ends the l inguistic t u r n in
phenomenology and suggests that universal pragmatics aims to elucidate basic structures of the lifeworld. Thus we can see that he early
on conceived communicative action and l ifeworld as complemen
tary, a connection he later strengthened and elaborated in The The
ory of Communicative Action.
Of special no te is Ha be rm as ' s discussion of tr ut h in th e Gauss Lec
tur es, for t r u th clai ms enjoy par ad ig ma ti c status as validity claims
(p. 86). When we raise a truth claim, we use language cognitively.And Habermas's discussion of cognitive language use in the Gauss
Lectures focuses on questions of reference and perception—ele
ments that are not emphasized in his subsequent art iculations of for
mal pragmatics. When attributing a property to an object, he claims,
a speaker presupposes that the object exists and that the proposit ion
she asserts is true. That is, she assumes that the subject expression
has a referent and that the predicate can be correctly applied to it .
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
17/216
xvii
Translator's Introduction
Ha be rm as he re end ors es a desc ripti on theo ry of refe renc e. Interest
ingly, he also states that our experience is in the first instance sensory
a n d onl y in th e se co nd in st an ce co mm un ica ti ve (p. 79). In lig ht of
recent criticisms to the effect that he needs a theory of reference to
avoid some form of linguistic idealism, the Gauss discussion is there
fore important . 9
It is also important because it contains an early treatment of theso-called consensus theory of truth, which emerges from Habermas's ac
count of the discursive redemption or vindication of validity claims.
As we have seen, a claim is discursively vindicated if rational consen
sus is re ac he d co n ce rn in g its validity, an d th e m e a n i n g of tr u th , ac
cor din g to Ha be rm as , is exp lica ted by specifying th e con dit ion s
under which validity claims can (or could) be vindicated. All of this
suggests an episte mic co nc ep ti on of tr ut h as wha t is rationa lly ag re ed
upon under ideal condit ions. The interest of the "consensus theoryof truth," however, lies not so much in what it says about the nature
of tr ut h, as in wh at it says ab out ho w we rea ch a g r e e m e n t on claims
to truth. Thus it is not so much a theory of truth as a theory of
justification. An d in fact, Ha be rm as has since ab an do n ed an epis
temic conception of truth and has developed this conception of ra
tional consensus primarily in the context of his theory of discourse
ethics, whi ch he de ve lo pe d after the co mp le ti on of The Theory of Com
municative Action.
While the Gauss Lectures focus on truth as a dimension of validity,
"Intentions, Conventions, and Linguist ic Interactions" (1976), an es
say m o r e explicitly loc ate d within t he phi lo so ph y of action , focuses
on the validity of social norms and examines the conceptual inter
connections between rules, conventions, norm-governed action, and
intentionality. Rather than establishing the need for a theory of communicative action on the grounds that other theories have failed to
provide adequate accounts of intersubjectivity, Habermas is here
concerned to demonstrate the need for a theory of action that is
intersubjective. Th e con cep t of com mun icat ive action is to acc ou nt
for intent ion al action, tha t is, actio n caused by in te rn al int ent ion al
states (in Bre nt an o' s sens e) of th e agen t, as well as fo r n o r m -
co nf or mi ng acti on or beh avi or in the sense of act ion in a cc or da nc e
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
18/216
xvm
with external rules. The essay aims at developing a concept of com
mun icat ive action (or, as h e pu ts it he re , int era cti on me di at ed
through interpreta t ion) that incorporates both intent ional and
norm-governed act ion. Habermas concludes that the two models of
intentional and norm-governed action that he discusses are comple
mentary, and that l inguist ic communication can be seen as consti tu
tive for both. But he does not advocate assimilating or reducing
social to linguistic theory. Indeed, he argues against taking language
as a paradigm for rule-following, or assimilating semantic and social
co nv ent io ns an d takin g th e fo rm er as pa ra di gm at ic of the latter,
since this wou ld obscu re th e crucial dist inction betw een co mmu ni ca
tive and strategic action. Rather, he conceives conventions "in the
sense of valid—that is, intersubjectively recognized-norms" as a sub
set of rules of action in general. The latter includes rules of instru
mental action and strategic rules as well.The duali ty of cognit ive and noncognit ive orientations continues
to play a role in this essay as well. Habermas draws an analytic distinc
tion between two types of intentionality, one referring to a cognitive
re lat io n to a wo rl d of objects, the ot her ref er ri ng to the sta nce a sub
j ec t adopts toward the p roposi t ional con ten t she is expressing. In ten
tional action can be understood on the model of teleological action,
in that the agent has a goal that she intends to accomplish and which
thus functions as a cause of her actions. When we examine inten
ti onal act ion wi th a view to th e ag en t' s cognit ive rela tio n to t he
world, it is po ssi ble —up to a po in t— to un d er st an d this relati on
mono log icall y. Th a t is, we can c onsid er h er as an indi vid ual in isola
tion from others and independently of the culture in which she lives.
Bu t as soo n as we try to give a n ac co un t of h o w t h e a g e n t come s to
have the goals she has, this model begins to break down. For her
goals d ep en d o n he r desires a n d o th er intent ion al states, whic h i n
tu rn result from what Ha be r ma s calls he r "need int erp reta tio ns. "
Th es e in tu rn a re a fu nct ion of the ag ent 's cultur al values a nd
no rm s, an d this me an s that int ent ion al action ca nn ot be ac co un te d
for monologically. Rather, our account of need interpretations re
quires looking at how subjects interact in accordance with mutually
recognized norms and values, and this establishes a nexus between
intersubjective cultural tradit ions and individual needs.
Translator's Introduction
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
19/216
r
Translator's Introduction
One of the negative consequences of starting from the teleological
means-ends model is that values and motives of action are repre
sented as private needs and wants—the most serious flaw of an em
piricist ethics, in Habermas's view. But if a person's motives are to be
intelligible to others, need interpretations must be intersubjecdve,
although their intelligibility does not yet constitute a normatively
bi nd ing sta nd ar d. An intel ligib le mot ive is n o t yet a just ific ation; thelatter requires reasons that all can share: "To say that a norm is valid
is to say that it claims to express a universalizable interest and to de
serve th e co ns en t of all th os e affected " (p. 122 ). This fo rm ul at io n an
t icipates Habermas's subsequent formulation of the principle of
universalizability of discourse ethics. More importantly, however,
these relatively early writings show th e deep c on ne ct io n bet wee n the
universalizability of interests and their origin in intersubjectivity. In
sofar as o ur want s an d n ee ds always ap pe ar u n de r so me int erp ret at ion, they pr es up po se a co mm un it y that has a lan gua ge con tai nin g
evaluative expressions, which in turn are rooted in an inter-
subjectively sh ar ed tr adit ion of cu ltu ra l values. Th es e values b ec o m e
normatively bind in g w h e n th er e is a cons ensu s th at is re pr od uc ed in
language and sedimented in the form of conventions.
Conventions, of course, are commonly appealed to in order to ex
plain how we un de rs ta nd o ne ano ther . Ha be rm as does not pr esu ppose that there simply are such conven t ions tha t ma ke mut ual
understanding possible any more than he presupposes that there
simply are subjects wh o abi de by t h em (let al on e stipu late th e m ) . In
stead, relying on G. H. Mead's analyses, he offers a developmental ac
count of how such conventions are established as normative
expectat ions presupposed in speech acts .1 1
Once we accept that both
having intentions and acting in accordance with norms presuppose
linguist ic interaction, we can understand how subject formation isthe result of linguistic interaction, how we are socialized in and
through communicat ive interact ion.
Finally, the essay "Reflection on Communicative Pathology" (1974)
seeks to address the question of deviant processes of socialization—a
topic that any dev elo pme nta l accoun t of interactive co mp et en ce
must address—and contains an analysis of the formal conditions of
xix_
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
20/216
XX
Translator's Introduction
systematically distorted communication. Habermas's guiding as
sumpt ion here is tha t the deve lopment of interactive com pe te nc e is
connec ted to th e d e ve l opm e n t of in te rna l mechanisms for control
ling behavior, b u t th at these two deve lopments are disti nct (since
m o r a l j u d g m e n t s a n d actual beh avio r in conflict resolution d o n ot
coinc ide) . His analysis stresses t h e co nn ect io n be tw ee n l inguist ic
c om m uni c a t i on an d ego dev elop ment : "Commu nicat ive act ion is t hem e d i u m of socialization" (p. 131). Pickin g u p on a t heme men t io ned
above, this essay establishes the co nn ec ti on betw een subjectivity and
intersubjectivity by showing that a subject 's intent ions a r e socially,
th at is, intersubjectively, st ru ctu red .
Habe rmas want s to show tha t b o th social a n d individual p athol o
gies ca n be analyzed in t e rms of dis turbances in interactive co mp e
tence. Such an acc oun t, however, pre sup pos es a m ode l of undisturbed
or normal communica t ion and interaction in th e t e rms of his f or ma lpragmatics . O n this approach the no t ion of normalcy is not deter
m i ne d by any part i cular cul ture , n o r is it a stat ist ical no r m; it is r at he r
a culturally invariant normative notion.
H a b e r m a s has be en cri t icized for p re sen t ing too idealized an ac
c oun t of com mun ica t io n, part icularly owing to his no t ion of the
ideal speech situation. This last essay shows that he is very mu ch at
t u n e d to the empirical vagaries of communica t ion . In clai ming tha t
the validity basis of speech has t ra ns cen den tal status, Ha be rm as cer
tainly does not m e a n to impl y that we can no t deviate from the condi
tions of no rm al commun ica t io n; o therwise, we would n o t have to
explicate the normative basis of speech. The condi t ions of possible
communica t ion are t hus not t r anscendenta l in th e sa me sense as, say,
Kant 's t ranscendental intui t ions of space and time qua condi t ions of
possible percept ion. Nonetheless , the formal presu pposi t ions un de r
ly ing communica t ion a re , according to Habe rmas , unavoidable.Moreover, as su ch they fun cti on so mew hat like regulative ideals in the
Kantian sense. They are not inviolable, b ut in cases where the inter
nal organizat ion of speech is violated, the pa t te rns of c om m uni c a
t ion a r e pathologically d istort ed. We ha ve al re ad y seen th at
in te r locutors may cha l lenge the validity claims raised by o thers and
thereby prompt communication shifts from action to discourse . It is
also possible t ha t the claims to intelligibility, t r u t h , Tightness, or sin-
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
21/216
xxi_
Translator's Introduction
ceri ty are continually sus pen ded or flawed wit hou t pr om pt in g such a
shift. If this ha p p e n s , th e result is systematically di st or te d c om mu ni
cation. T h e ki nd of violation of the universal pre su pp os iti on s of
co mm un ic at io n t ha t leads to systematic dist ort ion is no t the resul t of
a lack of co mp et en ce in the la ng ua ge , a mi sc on ce pt io n of the level of
discussion, or a retreat from communicative to strategic action.
These all involve a cessation of communicative action, whereas in thecases that Ha be rm as has in min d, comm uni cati ve action con tin ues
in spite of th e viol ation of its fo rma l pr esu pp os iti on s. Th e str ong est
cases of systematic distortion are those in which the speaking sub
jects themselves are unaware of their violation of communica t ive pre
supposit ions, such as whe n a co mp et en t spea ker expresses he rself
unintelligibly without realizing it , when one spouse deceives herself
about her feelings for the other, or when a speaker thinks she is act
ing in accordance with social norms but is actually violating them.
Ideally, the rejection of a validity claim leads to discourse, in which
th e sp ea ke r seeks to justify the claims she is ma ki ng ; or th e sp ea ke r
shows by h e r actio ns tha t she is sinc ere. But this do es no t h a p p e n in
cases of disto rted co mm un ic at io n th at stem from conflicts th at can
not be quite suppressed yet must not become openly manifest—
because, for example, they threaten the identi ty or self-understand
ing of one or more interlocutors. This sort of situation results in a
kind of paradox of systematic distortion of communication, for the
very validity claims th at are be in g violated " serve to ke ep u p t he ap
pearance of consensual action" (p. 155).
This discussion makes clear that the idealizations required by this
model of communication may fail. However, the idealized model al
lows for a systematic un de rs ta n di ng of th e diff erent sorts of failure
and provides the norms or standards for criticizing them.
As the empirical l i terature upon which Habermas draws in this discussion indicates, systematic distortions connected with subject-
fo rm at io n occu r par tic ula rly of ten withi n families. No t only is this a
context in which people's identities are formed and confirmed, it is a
con tex t in which a part icularly hig h pre mi um is pla ced on co mm un i
cative rather than strategic action. On the one hand, families are ex
pe ct ed to fun ctio n as units ; on th e ot he r ha nd , the ne ed s an d wants
of individuals have to be met within the family structure. Thus there
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
22/216
xxii
Translator's Introduction
is a tensi on an d po ten tia l conflict be twe en th e or ien tat ion towar d
mutual understanding and the orientat ion toward individual need
satisfaction. Failure to resolve such conflicts explicidy can lead to sys
tematically dist orte d co mm un ic ati on , in which me mb er s empl oy dif
ferent strategies for maintaining or producing a "pseudo-consensus."
The y ma}' seek to safeguard an end an ge re d cons ensu s an d pr eve nt
cha lle nge s to it by, for inst ance, in te rr up ti ng or br ea ki ng off conversation, refor mul atin g a di sa gr eem ent as an agr eem en t, falsely re
ciprocating another 's action, or—in the most extreme case that
threatens the very intelligibility of their utterances and actions inas
much as incoherence violates the norms of rat ionali ty—behaving in
consistently. T h e no ti on of systematically dis tor ted co mm un ic at io n
evidently introduces a third option between the successful comple
tion of a speech act and what J. L. Austin terms a "misfire" in which
the speech act itself fails.
H a b er ma s links a family's po te nt ial for conflict to po wer relat ions,
cl ai mi ng t ha t a "family's ability to solve . . . p r ob lems st ands in an in
verse rela tion t o its in te rn al pot ent ial for conflict. Th e latter in t u r n
is a fun ctio n of th e di str ib ut ion of pow er" (p. 161) . H e do es not ,
however, sug ges t th at a heal thy family mu st succeed in tr an sc en di ng
power relations. Rather, he allows for a "healthy" distribution of
power, which, nevertheless, is connected to an "asymmetrical distri
bution of opportunities" for gratification. Once again, there is a
clear rec ogn it i on that empiric al circum stance s—even in com mu ni
catively st ru ct ur ed con tex ts— div erg e significantly from t he id eal
speech si tuation.
While these lectures and essays provide a good introduction to the
theory of communicative action, they are also transitional in nature:
The y fo rm a bri dg e be twe en Ha be r ma s ' s work of th e 1960s a n d tha tof the 1980s. His linguistic turn was initially motivated by the convic
tion that a critical social th eo ry re qu ir ed a sou nd me th od ol og ic al
an d epist emolog ical fo un dat ion : he nc e the projec t of pro vidi ng a
linguistic grounding for sociology. However, the project of develop
ing a co mp re he ns iv e th eo ry of rationality, whi ch is wh at the t heo ry of
communicative action in effect attempts, cannot be carried out
mer ely fro m the me th od ol og ic al persp ective of finding an al ter na-
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
23/216
x x.iii__
Translator's Introduction
tivc to objectivist an d subjectivist social th eo ri es . Th us H a b e r m a s
soon fo un d i t necess arv to devel op an acc ou nt of the pre supp osit ions
of act ion ori ent ed toward reac hin g un de rs ta nd in g ind ep end en t ly of
an account of the transcendental presuppositions of social-scientific
knowledge . 1 2 This helps explain why the distinction between cogni
tive an d com mu ni cat iv e la ng ua ge use, whi ch is so cen tra l in the
Gauss Lectures, becomes less prominent in his subsequent formulat ions of the theory of communicative action.
1 3
At the same time, while Habermas has been working out a com
plex theory of action, he has also elaborated his conception of
spe ech act the or y an d of fo rma l pra gm at ic s to serv e as th e basis fo r a
social theory of meaning. Since the writing of the Gauss Lectures, he
has developed a systematic classification scheme for theories of
meaning, in which he dist inguishes formal semantics, intentionalist
semantics, and use theories of meaning. In his view, each of these fo
cuses on but one of the three functions of language that an adequate
the ory of me an in g mus t in co rp or ate . As we saw, in co mmu ni cat in g,
we represent facts about the world, we express our subjective states,
an d we interact with others; an d these thr ee functi ons c or re sp on d to
the three validity claims of truth, sincerity, and Tightness th at for mal
pragmatics analyzes.1 4
In his recent work, in addition to speech act
theory, Habermas also draws on Michael Dummett 's assertibilist se
mantics, according to which the meaning of a sentence (or utter
an ce ) is given by th e co nd it io ns u nder whi ch it is ac ce pt ab le to
he ar er s. Dev elo pin g this asp ect of fo rm al pragma tics once agai n u n
derscores the aspects of rationality and intersubjectivity; for to say
that un der st and ing an ut terance is knowin g the condi t ions u nd er
which it is acceptable entails that a speaker-hearer does not fully un
der sta nd a given ut te ra nc e unless she knows wh at reas ons cou ld be
offered to back up the claims raised in the utterance. And construedin this way, acceptabil i ty con dit ion s can no t be de te rm in ed ind e
pe nd en tl y of an intersubjective prac tice of ar gu me nt at io n an d
justification.
Habermas has recent ly re turned to some of the themes adum
brated in the early 1970s. One of these is the question of the nature
of tru th , as I in di cat ed above . An ot h er is th e distinction be twe en
communica t ive and noncommunica t ive language use .1 5
Yet the views
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
24/216
xxiv
Translator's Introduction
art icu late d in thes e early work s are no t only relev ant to H ab er ma s' s
cu rr en t tho ugh t ; they bea r on con tem po ra ry phi losophi cal discus
sions more broadly. Within Anglo-American philosophy, there has
been a resurgence of interest in pragmatics and in social theories of
me an in g t ha t do justice to th e intersubjectivity of social int er act ion .
A pr ime ex amp le is Ro ber t Br an do m' s Making It Explicit, which is an
elaborate working out of a semantic theory based on social practicesan d, in particular , practice s of justification. O n B ra nd om ' s view, se
ma nt ic s is ba se d on t h e giving of a n d asking for re as on s: T o give th e
meaning of a sentence is to art iculate the condit ions under which i ts
ass ert ion is just ifie d, w hi ch is to artic ula te a we b of jus tif icat ory rela
t ions. However, unlike Habermas, Brandom does not dist inguish be
tween irreducibly distinct types of validity claims; the focus of his
analysis remains the assertion, which he continues to regard as basic.
This consti tutes a potential challenge to Habermas's system not onlywith respect to the irreducibility of the three validity claims to truth,
Tightness, a n d sincerity, b u t also wit h res pe ct to the sta tus of th e as
sert io n withi n his ow n framework. Given the pr of ou nd influ ence
Sellars has had on Brandom, Habermas's discussion of Sellars might
also be a potentially fruitful point of engagement, with regard to
both semant ics and percept ion.1 6
It wo ul d be a way of fle sh ing ou t
th e cognitive dim en si on of lan gu age an d clarifying its rel ati on shi p to
the communicative dimension, a relationship that lies at the heart of
Habermas 's project .
Acknowledgments
This project would not have been possible without the help and sup
p o r t of ot he rs . I am in d eb te d to Je re my Sh apir o for his ex cell ent
or ig in al tran slat ion of th e Gauss Lec tur es, wh ic h m a d e my ow n taskimmeasurably easier. My thanks to Harry Heft, Jonathan Maskit, Ste
ven Vogel, Christopher Zurn, and especially Thomas McCarthy who
all pro vid ed inva luab le feed back o n the translation an d in tro du c
tion. I wo ul d also like to t h a nk Pat Davis, wh o tr an sf er re d th e origi
na l type scri pt of th e lect ur es on to disk, as well as La rr y C o h e n an d
Judy Feldmann of MIT Press for their assistance. Finally, I am grate
ful to Jii rge n Ha be rm as for his gener osit y in re sp on di ng to my
queries.
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
25/216
Reflections on the Linguistic Foundation ofSociology
The Christian Gauss Lectures (Princeton University,
February-March 1971)
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
26/216
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
27/216
Objectivist and Subjectivist Approaches to
Theory Formation in the Social Sciences
There are competing theoretical approaches in the social sciences
that differ not only in the kinds of problems they address and the research strategies th ey apply, b ut in th ei r fu nd am en ta l pri nci ple s.
They diverge in their choice of categorial frameworks and in how
they con cep tua liz e their objec t do ma in —t ha t is, in ho w they defi ne
what it is they are actually studying. These differences of conceptual
strategy express more deeply rooted conflicts: conflicting views of sci
ence and cognitive interests. My aim here is not to investigate and
systematically ex p o u n d these theo retic al ap pr oa ch es . I in te nd ra th er
to deve lop a pa rt ic ul ar co nc ep tu al strategy for t he social sciences an dto establish its theoretical plausibility and potential. I should like to
begin with some comp arati ve met ho do log ica l con side rati ons that
lead to some preliminary classifications. These are to serve exclu
sively as a provi siona l de lim itat ion of a co m mu nic at i on th eo ry of so
ciety. This theory does not yet exist in a satisfactory form, and I can
only discuss a few issues that moti vat e me to con si de r suc h an ap
pr oa ch to be fruitful.
In te rms of co nc ep tu al strategy, th e first dec isi on th at is of fun da
mental significance for a theoretical program in the social sciences is
whether to admit or re ject "meaning" [Sinn] as a pri miti ve ter m. I
take the paradigm of "meaning" to be the meaning [Bedeutung] of a
word or a sentence.1
Thus I am assuming that there are no pure or
a priori speaker intentions; meaning always has or finds a symbolic
expression; to attain clarity, intentions must always be able to take on
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
28/216
4
Lecture I
symbolic fo rm an d to be expr esse d. This exp res sio n can be an ele
ment of a natural language or linguistically derivative (it may, for ex
am pl e, be lo ng to a system of signs with wh ich th e deaf -mu te or
drivers in traffic co mm u ni ca te ). T h e express ion may also be no n
verbal, that is, it may take the form of an action or of a bodily ex
pression (a grimace or gesture), or of an artistic or musical
re pr es en tat io n. I a m assu ming that a me an in g tha t is express ed
nonverbally can in principle be rendered, at least approximately, in
words: Whatever can be meant can be said. 2 T h e converse, however,
does not hold. Not everything that can be said is necessarily express
ible nonverbally.
If we may def ine "m ea ni ng " from th e ou tse t as linguistic m e an i n g
[Sinn], th at is, wi th re fe re nc e to th e significati on [Bedeutung] of
words and sentences, then this first basic decision in conceptual strat
egy can be re fo rm ul at ed m o r e precisely. It is in fact a m et at he or et ic aldecision as to wh et he r linguistic co mm un ic at io n is to be reg ar de d as
a constitutive feature of the object domain of the social sciences. The
te r m "con stitutive" me an s th at the object d o m ai n stu die d by th e so
cial sciences is itself determined in terms of linguistic communica
tion. Sometimes we describe language by using categories of
observable behavior or transmissible information and explain l in
guistic processes in terms of learning theory. In these cases, mean
ingfully st ru ct ur ed forms are ta ke n as objects am o n g ot he r physicalobjects; the latter are described in a conceptual framework that is
not specific to any language and are studied by means of empirical
the ori es. In cont rast , la ng ua ge is constitutive of an object do ma in if
its cate gor ial fra mewo rk is suc h th at meani ngfu lly s tr uc tu re d fo rm s
(such as pe rs on s, expressi ons, a n d institutions) ca n a p p e ar in it as
p h e n o m e n a req uir ing expla nat ion. "Mea nin g" has th e s tatus of a
pri miti ve t er m in the social scie nces if we use it to cha rac ter ize th e
structure of the object domain itself, rath er t ha n ju st individual ele
ments within it. I should like to clarify this first metatheoretical deci
sion by pointing out three of its implications.
(a) Behavior versus action
Only if "meaning" is admitted as a primitive term in sociology can we
distinguish action from behavior. At present I should like to set aside
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
29/216
Objectivist and Subjectivist Approaches to Theory Formation
the prior pr ob le m of the de ma rca tio n betwee n those ob servab le
events tha t we in te rp re t as beh av io r an d th ose events that we ca nn ot
interpret as behavior. The interpretive scheme that allows us to un
derstand the motion of a body as the expression of an organism, in
other words, as the movement of a living body, has not yet been satis
factorily analyzed. 3 By des cri bin g an obs erv abl e mot io n as behavio r,
we ascribe it to an or ga ni sm th at re p r odu ce s its life by ad ap ti ng t o its
envir onmen t. We un de rs ta nd i t as a mo ve me nt b ro ug ht abo ut by an
organism. In so doing, we are assuming that there is an endty X that
in some broad sense is "responsible" for this movement. In this con
text, of course, the category of responsibility can be used only in
scare qu ote s, th at is, wit h cert ain res erva tion s. Fo r an an im al ca nn ot
be held responsible for its behavior in the same sense that a subject
capable of speech and cognition can be held responsible for its ac
tions. Nonetheless, we apparently derive the perspective from which
we interpret motions as modes of behavior from a private modificat
ion of the pre-understanding of our own social lifeworld. I call this
modification privative be ca us e we are capa ble of dis tin gui shi ng be
havioral resp onse s from ot he r events wit ho ut ha vin g to app eal to th e
cate gor y of me an in g. For that catego ry ma ke s it possible to differen
tiate between behavior that I can understand as intentional action
and behavior that cannot be comprehended under this descript ion.
I call behavior intentional if it is go ve rn ed by n o r m s or or ie nt ed torules. Rules or norms do not happen like events, but hold owing to
an intersubjectively recognized meaning [Bedeutung]. No rm s have se
mantic content: that is , a meaning [Sinn] that be co mes th e rea son o r
motive for behavior whenever they are obeyed by a subject to whom
things ar e mea nin gfu l. In this case we spe ak of an action. Th e in te n
tion of an acto r wh o or ien ts his or h er beha vio r to a ru le co rr es po nd s
to the meaning of that rule. Only this normatively guided behavior is
what we call action. It is only actions that we speak of as intentional.
Observable behavior fulfills a prevailing norm if and only if this be
havior can be understood as produced by an acting subject who has
gras ped the me an in g of the no r m an d obeye d i t intentionally. Behav
ior that we observe over a particular period of time can de facto ac
cord with a given norm without being norm-governed. That is why
we dist inguish regula r beh avi or fr om rul e-g ove rne d behavior, o r
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
30/216
6
Lecture I
action. We discover regularities through inductive generalizations;
eit he r the y exist or they do not. In contrast , we mu st und er st an d th e
meaning of rules; thev have normative validity. We can break rules;
but it is me an in gl es s to say t hat regular itie s ar e violate d. Rul es that
underlie a practice can be accepted or rejected, whereas regularities
in behavior can be affirmed or denied. Naturally we can assert the
existence of regularities in contexts of intentional action as much as
in a chain of behavioral responses. But in the former case, we can de
duce the assertion from the fact that norms are followed with spe
cifiable probability, whereas in the latter we must base our assertion
on an induc tive gene raliz ation from obs erve d behavior.
(b) Observation versus the understanding of meaning [Sinnverstehen]
The distinction we have drawn between behavior and action leads to
a fur the r dis t inction betwee n different mo de s of exp er ien ce inwhi ch be hav ior al respo nse s and acti ons are accessible to us. We ob
serve behavior and behavioral regulari t ies, whereas we understand
actions. O n c e ag ai n it is th e cat ego ry of me a n in g that differ entiates
the two mo de s of ex pe rie nce . I ca nn ot obs erve actions as me r e be
havior. For if a given behavior is to be described as an action, then I
mu st relate feat ur es of this be ha vi or to rules on whic h it is ba se d an d
un de rs ta nd th e me an in g of these rules . Of course the app reh ens ion
of s t ru ctur es of act ion th ro ug h the unde rst and ing of me an in g rests
on observations.
Let me co mp ar e two pe rce ptu al ju dg me nt s or "observation state
ments." "I see a fly bouncing against the window" is a sentence in
which I report the observation of a behavior. In contrast, "I see John
returning from work" is a sentence with which I describe an "ob
served" action. I us e th e exp re ss ion "to see" in the sam e way in bo t h
cases. For both sentences report events that the speaker claims to
perceive at the time. Nevertheless, in the former case "seeing" means
obs erv ing an even t th at can be co m p r eh en d e d as behavior, whe reas
in the lat ter i t means understanding an action. Of course this under
s tanding is based on the obse rvati on of an occur ren ce (th e do orb ell
rin gin g, a pe rs on ent er in g the roo m, etc.); b ut th e ob ser ved behav
ioral elements and events are interpreted with reference to a structure
of action . T h e latte r consists in no r ms , in this case, social no r ms th at
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
31/216
7____
Objectivist and Subjectivist Approaches to Theory Formation
regulate work hours and transportation to and from work. I have to
be acquainted with norms of this sort and the conditions of their ap
plication in order to know when a given occurrence can be inter
pr ete d as a case in wh ich th e no r m appl ies. "I see J o h n r et ur n in g
from work" me an s that I un de rs ta nd an obser ved occ urr enc e as the
fulfillment of a norm: that is, as a particular action—in this case, as
"returning from work." To see, observe, or perceive an action always
involves un de rs ta nd in g a n o r m (or the co rr es po nd in g int ent ion of
the actor) and interpreting movements (or states of affairs) in the
light of the n o r m (or int ent ion ) that is un de rs to od .
The decis ion whether intendonal act ion should be admit ted has
meth odo log ical implic ations precisely with rega rd to th e mo d e of ex
perience. This can be seen at the level of problems of measurement.4
Measurements serve to transform experience into data that meet the
de ma nd s of inters ubjecti ve reliability a n d on t he basis of whi ch th e
claim to empirical validity of theoretical statements can be verified.
Observations of events (and behavioral responses) can be linked
with th e la ng ua ge gam e of physica l me as ur em en t. Bu t th er e is no
co rr esp on din g system of rel iably incu lcate d basic me as ur em en t op
erations, such as the one available for moving bodies (or points of
mass) for objects such as actions, which are accessible only through
communica t ive exper ience [sinnverstehende Erfahrung]. In ot he r
words, observations that can be expressed in descriptive sentences ofa la ng ua ge for thin gs an d events can be verified th r o u gh re co gn ize d
procedures that are reducible to physical measurement . The inter
pretation of the meaning of symbolic forms such as actions, which
can be re pr ese nte d in descripti ve se nte nce s of a la ng ua ge fo r pe r
sons and expressions, cannot be reliably operationalized in analo
gous fashion. Unti l now, the measurement of symbolized meaning
has depended on ad hoc procedures that in the final analysis rest on
an understanding of language that remains prescientific, al though i tmay be shaped by the discipline of hermeneutics. In principle, any
one wh o maste rs a na tu ra l lang ua ge can, by virt ue of co mm un ica ti ve
competence, understand an infinite number of expressions, if they
are at all mea nin gfu l, a n d m ak e th e m intelligible to others . T h at
is, she can interpret them. Some are more practiced at this than
othe rs: He rm en eu ti cs is an art an d not a me th od .5
We make use of
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
32/216
8
Lecture I
hermeneut ics , the ar t of interpreta t ion, instead of a meas ur em en t
procedure; but i t is not such a procedure. The only thing that would
allow for the dev el op me nt of basic me as ur em en t oper ati on s for
me an in g woul d be a the ory of ordinary- language com mun ica t io n
that did not merely guide and discipline the natural faculty of com
muni cativ e co mp et en ce , as he rm en eu ti cs does, bu t tha t could also
explain it.
(c) Conventionalism versus essentialism
No matter how the problem of measuring the meaning of symbolic
expressions is solved, the experiential basis of a theory of action re
mains distinct from that of a behaviorist theory in the strict sense.
For the adequacy of a description of a meaningfully structured con
struct, an ut te r an ce or an action, can be tested only by ref ere nc e to
the knowledge of the subject who produced the expression. In many
cases a subject capable of action may not be able to specify explicitly
the norms according to which it orients his behavior. Nevertheless,
insofar as it masters norms and can follow them, it has an implicit
knowledge of rules. On the basis of this know-how it can always de
cide wh et he r a given behav ioral respo nse c orr esp ond s to a kn own
rule a t all, th at is, wh et he r it can be un de rs to o d as actio n. Th e sub
jec t can dec ide whether , in a given case, such a behavioral r esponse
acc or ds wi th or deviates fro m a given n o r m , an d to wh at ext en t it deviates from a n un de rl yi ng n or m. Th e situation is simil ar wit h re ga rd
to linguistic utterances. Usually, competent speakers are able to ex
plicate the grammatical rules of the natural language in which they
fo rm and u nd er st an d sent enc es only incompletely, if at all. Non et he
less, every adequately socialized speaker has at his disposal a
know-how that enables him to dist inguish phonetic utterances from
mere sounds, to distinguish semantically meaningful and syntacti
cally well-fo rmed s ent enc es fro m th ose that are deviant, an d to o rd ersuch sentences according to the degree of their deviation. This intu
itively available knowledge of rules that competentiy speaking and
acting subjects have, which can also be discursively articulated at any
time, provides the required experiential basis for theories of action.
Strictly behav ioris t the orie s, on the ot he r ha nd , de pe n d exclusively
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
33/216
9
Objectivist and Subjectivist Approaches to Theory Formation
on obser vat ion al dat a. This fact gives rise to an i m p o r t a n t dif fer enc e
between the structures of the two types of theory and their relation
to their respecdve object domain.
Theories that are to explain the phenomena accessible through
the unders tanding of meaning [Sinnverstehen]—that is, t he ut te r
ances an d exp res sio ns of subjects capable of sp ee ch an d act io n—
m u s t tak e the fo rm of a systematic expli catio n of th e kn ow led ge of
rules base d on wh ich com p et en t speakers a n d actor s ge ne ra te their
expressions. Th e o r y fo rm at io n serves to re co ns tr uc t the systems of
rules acco rdi ng to whic h meaningfu lly s tru ctu red forma tion s, sen
tences an d action s, ar e pr od uc ed . Th es e genera tive rules nee d no t
be directly read off the surface structure of expressions. As with
grammar, there may be deep structures, which underlie the surface
structures that have been produced and yet are part of a competent
speaker 's implici t know-how; he nc e they ar e no net he les s known. T h e
goal of such a theory is the hypothetical reconstruction of rule sys
tems that disclose the internal logic of the rule-governed generation
of intelligible surface st ruc tur es. Let us no w assu me th at these intelli
gible surface stru ctur es cor re sp on d to th e emp iric al regulari t ies of
observable events (and behavioral responses). In this case we could
compare the reconstruction of the abstract systems of rules underly
ing surface structures with theories of the empirical sciences from
whic h we derive laws of n at u r e tha t so meh ow "un der lie" empir icalregularities. Bu t this co mp ar is on clearly reveals th e differe nce in
status of the two types of theories. The hypothetical reconstructions
advance an almost essentialist claim that is absent from the nomo-
logical theories in the empirical sciences. For, insofar as the latter ref
er to the object domain of physicallv measurable events, the
primitive terms of systems of nomological statements are primarily
introduced by convention. They provide the idiom for a theoretical
constru ct tha t can be cor ro bo rat ed indirect ly th ro ug h the derivat ionof lawlike hy po th es es th at ar e sub ject to con fi rma tio n. O n e mi gh t say
that nomological hypotheses, if they are true, correspond to struc
tures of a reality objectified in terms of physics or the behavioral sci
ences (or that they pick out invariant features of this objectified
reali ty). But they cannot be said merely to reconstruct an intuit ive
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
34/216
10
Lecture I
knowledge that competent observers of this reality always already
possess. Rather, the sort of knowledge thus produced is, as a rule,
quite counterintuit ive.
In contrast, the rational reconstructions of the knowledge of sub
jects capab le of speech and act ion do raise such an essentialist claim.
T h e primitive term s to be emp loye d in the recon str uct ion of struc
tures of operationally effective generative rules, therefore, are not in
troduced conventionally. Instead they are introduced in connectionwith categories that must be derivable from the self-understanding of
the very subjects who produce these structures. As I see it, the
essentialist moment consists in the fact that hypothetical reconstruc
tions, if tr ue , co rr es po n d no t to str uct ure s of an objectified reality
but to structures of the implicit know-how of competent subjects ca
pable of ju d gm en t. Wh at is to be explicated by these rec ons tru ctio ns
are the operationally effective rules themselves.
I ha ve discussed the meta the ore tica l decision wh et he r me an in g
should be admitted as a primitive term in the social sciences by look
ing at three methodological implications of great import . Having
do ne so, I ca n pro vid e a prov ision al de ma rc at io n be twe en objectivist
and subjectivist approaches to theory formation. I shall call a theo
ret ical program subjectivist if it conceiv es of society as a me an ingf ul ly
structured system of life [Lebenszusammenhang], and as a sys tem of
symbolic express ions and structu res that is cont inuo usly p ro d uc edac co rd in g to und erl yin g abstract rules. T h u s the or y is given the task
of rec on str uct ing a proc ess whereb y a meaningfu lly s tr uct ur ed social
reality is pr od uc ed . In con tras t, I shall call a the ore tic al p r o g r a m
objectivist if it conceives the life process of society not internally as a
proce ss of con str uct ion , that is, of the pro du ct io n of mean ing ful
structures, but externally as a natural process that, like other pro
cesses, can be observed in its empirical regularities and explained by
means of nomological hypotheses. In this sense, all strictly behaviorist theories, such as classical learning theory, are objectivist. At this
po in t I pre fer not to dec ide betwee n these two co mp et in g ap
proaches. I shall content myself with pointing out that the objectivist
theoretical program, which has been quite successful within its lim
its, has to deal with difficulties arising from its methodological ne
glect of the symbolic prestructuring of social reality. These
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
35/216
n___ _ _Objectivist and Subjectivist Approaches to Theory Formation
difficulties show up at the level of the measurement problems in
volved in atte mp tin g to re du ce action to behavior, wh ich have b ee n
discussed by Cicourel (1965) and others.6
T h e s e difficulties are
exemplified by wh at can no w be con si de re d th e failed at te mp t to de
velop a behaviorist theory of language.7
1 do not want to enter into
this discussion he r e. Inst ea d I will limit myself in th ese l ecture s to th e
theoretical program of subjectivism. The generative theories of soci
ety [Erzeugungstheorien], as I sha ll call t h e m fr om n o w on , hav e com
plementary difficulties to deal with. A theoretical program of this
kind must answer three questions. Assuming that society is conceived
as a process of generating a meaningfully structured reality:
(a) Who is the subject of this generative process, or is there no such
subject?
(b) How is the mode of this generative process to be conceptual
ized—as cognitive activity (Kant and Hegel), as linguistic expression
(H um bo ld t) , as la bo r (M ar x) , as artistic cre ati on (Schelling , Nie
tzsche), or as inst inct (F reu d) ?
(c) And, finally: Are the underlying systems of rules according to
which social reality is constructed invariant for all social systems, or
do even these abstract rule systems develop historically, and is there
possibly an inner logic of their development that can be recon
structed as well?
Before setting u p a typology of ho w th e most i m po r t a n t ge ne ra ti ve
theories of society have answered these questions, I want to discuss
very briefly two fu rth er fu nd ame nt al decisions co nc er ni ng co nce p
tual strategy that have great bearing for theory formation in the so
cial sciences.
T h e second metatheoretical decision is w he th er int en tio na l action is to
be conceptualized in the form of purposive-rational action or in theform of communicative action. Let me first characterize these two
types of action with reference to the status of the rules that govern
behavior in each case. By purposive-rational action! un de rs ta nd ei ther
instrumental action or rat ional choice or a combination of the two.
Instrumental action follows technical rules based on empirical
knowledge. These rules imply condi t ional predict ions regarding ob
servab le events, wh et he r physic al or social. Th u s suc h pr ed ict io ns
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
36/216
12
Lecture I
ca n pr ov e to be co rr ect or inc orr ect . Rati ona l cho ice is go ve rn ed by
strategies based on analytical knowledge. They imply derivations
from preference rules (value systems) and decision procedures.
These proposit ions are derived either correctly or incorrectly. Pur
posive- rational action attains definite goals u n d e r given con dit ion s.
But whe rea s th e mea ns organ ize d by in str ume nt al action are app ro
priate or inappropriate according to criteria of effective control of
reality, strategic action depends only on the correct assessment of
possible behavioral options, which results from derivation using val
ues and maxims alone.
By communicative action I un de rs ta nd symbolically me di at ed inter
action. It is go ve rn ed by bi nd in g no rm s that define rec ipro cal e xpec
tat ions ab ou t behav ior an d th at mu st be un de rs to od an d
acknowledged or recognized by at least two acting subjects. Social
no rm s are enf orce d th r ou g h sanctions. Th ei r me an in g is objectifiedin symbolic expressions and is accessible only through ordinary lan
guage communication. Whereas the effectiveness of technical rules
and strategies depends on the validity [ Gultigkeit] of empirica lly tru e
or analytically correct propositions, the validity [Geltung] of social
n o r ms is en su re d by an intersubjective r eco gn iti on t ha t is base d on a
consensus about values or on mutual understanding. Violating a rule
has different c on seq ue nc es in each case. In co mp et en t behavior,
which violates proven technical rules or correct strategies, is cond e m n e d p er se to failure t h r o u g h its lack of success. T h e "pun ish
m en t ," so to spe ak, is bu ilt in to its fo un de ri ng o n t he shoal s of reality.
Deviant behavior, on the other hand, which violates prevailing
norms, triggers sanctions that are connected with the rules only ex
ternally, by convention. Learned rules of purposive-rational action
pr ovi de us wit h a reg ime of skills; in contrast, int er na liz ed no rm s fur
nish us with a regime of personality structures. Skills enable us to
solve problems, whereas motivations allow us to conform to norms.
Table 1 sum mari zes thes e definit ions. Th ey requ ire a mo re precise
analysis, wh ic h I ca nn ot und er ta ke her e.
Theory formation in the social sciences has the option of defining
the object domain to comprise either actions of the strategic type
only or both strategic and communicative actions. Strategic action
can be considered as a limiting case of communicative action; it oc-
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
37/216
13
Qbjectivist and Subjectivist Approaches to Theory Formation
Table 1Rules of Action
Technical and StrategicSocial No rm s Rules
Linguistic Means of
Definition
Elements of Definition
Mechanisms ofAcquisition
Function of the Type ofAction
Sanctions for ViolatingRules
intersubjectively sharedordinary language
reciprocal normative ex
pectations of behavior
internalization of roles
maintenance of institutions (conformity to
norms based on reciprocal reinforcement)
punishment based onconventional sanctions;
failure against socialauthority
context-free language
conditional predictions;
conditional imperatives
learning of skills andqualifications
problem-solving (goal attainment, defined interms of means-endsrelations)
lack of success; failureagainst reality
curs whe n ord ina ry langu age com mun ica t io n between interlocut ors
breaks do wn as a me an s of ma in ta in in g consensus , an d each assu mes
an objectifying attitude toward the other. For strategic action is based
on rules for the purposive-rational choice of means; and, in princi
ple, each actor can make this choice by herself. Maxims of behaviorin strategic actio n are d e te r mi n e d by th e interests of max imi zin g
gains and minimizing losses in the context of competition. In this
case, my other is no longer an alter ego whose expectations I can
fulfill (or dis app oin t) ac co rd in g to intersubjectively re co gn ize d
no rm s. Rather, she is an o p p o n e n t who se decisi ons I seek to
influence indirectly by me an s of pu ni sh me nt s an d rewards. In stru
mental actions, on the other hand, are not social actions at all;
rather, they can appear as components of social actions (i.e., as ele
ments of role definitions). If only strategic actions are admitted, we
can develop rational choice theories such as theories of exchange. If
com mun ica tiv e action s are ad mi tt ed as well, we can deve lop conv en
tional theo rie s of acti on such as tho se of We be r or Par sons .
T h e third metatheoreticcd decision that I consider to be of major con
se qu en ce is wh et he r we sh ou ld ch oos e an atomisti c or a so-called
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
38/216
14
Lecture I
holistic approach. I cannot go into the details of the extensive litera
ture on this issue. 8 Both conceptual strategies seem to me to be feasi
ble, al though the different theories have different domains of
application and capacities. At the level of theories of action, the
atomistic approach takes the form of methodological individualism.
J. W. N. Watki ns (in ag r ee me n t with Po pp er ) has fo rm ul at ed two in
dependent postulates: (a) "the ultimate constituents of the social
world are individual people who act more or less appropriately in the
light of their dispositions and the understanding of their situation."9
Thus all social phenomena must be analyzable in the form of state
ments about the actions of individual subjects. Statements in a theo
retical idiom, which contains expressions for supraindividual social
entities, such as roles, institutions, value systems, and traditions, are
inadmissible unless they can be reduced to statements in another
theoretical idiom in which the only predicates are for acting subjects,
their utterances, and their motivations. The second postulate is that
(b) "no social tendency exists which could not be altered if the indi
viduals concerned both wanted to alter it and possessed the appro
pria te informat ion."1 0
This stronger claim has the status of a
philosophical assumption. It states that subjects capable of speech
a n d action ar e th e only age nts of c h a n g e in th e historical deve lop
ment of social systems. Social change can be explained with refer
ence to the properties of supraindividual units (such as systems,gr ou ps , or struct ures) if a n d only if these supraindivi dual pr op ert ies
are reducible to properties of individual subjects capable of speech
and action. The counterposition to an individualistic theory of ac
tio n is r ep r es en t ed tod ay by social systems th eo ry (such as th e wor k
of Deutsch, Parsons, and Luhmann). Systems theory takes into ac
count the fact that the structure of social norms transcends the sub
jectively i n t e n d e d m e a n i n g of individuals acting according to n o r m s .
Systems are introduced as units that can solve objectively given problems through learning processes that transcend individual subjects.
Th e th re e op t io ns th at I have me nt io ne d, from whi ch a con cep tual
strat egy fo r the social scie nces mus t be cho se n, offer co nv en ie nt cri
teria for classifying the most important theoretical approaches, as in
dicated in table 2.
I do not want to examine the relative fruitfulness of these various
theor etica l appr oac hes . This overview is in te nd ed ra th er for the
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
39/216
15Objectivist and Subjectivist Approaches to Theory Formation
Table 2Approaches to Social Theory
behavior
notadmissible
Meaning as primitive term
admissible
strategic action communicative(and strategic)
action
atomistic behaviorist
psychologyrational choicetheories (e.g.,pure economics)
"interpretive"sociology (e.g.,
ethnomethodology)
holistic biological systems theory
social cybernetics
(e.g., organizational sociology)
structuralist andfunctionalist systems theories; sym
bolic interactionism
purpose of classifying the generative theories of society that I am in
terested in. Obviously they do not belong among strictly behaviorist
theories; n o mor e, however, do they bel on g am o n g the orie s of strate
gic action. These theories incorporate assumptions of rat ionali ty that
obtain—approximately—but for limited segments of social reality.
Both rational choice theories and social cybernetic models have a
normative analytical status. They can be applied only on one of two
pr esu ppos iti ons. The first is that subjects act rat ionall y a n d tha t t he
assumed maxims of behavior are in fact the basis of their actions.
The second is that the self-regulating systems are stabilized in pre
cisely the state that is conventionally postulated to be optimal. Gen
erative theories of society cannot belong to this type of theory, since
they claim to c o m p r e h e n d th e life pr oce ss of society as a wh ol e an d,
in particular, as a concrete process of the generation of meaning
structures. They do not content themselves with segments of realitythat can be seen as approximating models of rationality. In other
words, they are not satisfied with a normative analytical status. There
fore they must be classified under the type of theory that appears in
the right hand column of the table.
To be able to use this fra mew ork to differ entiate be tw ee n differ ent
co nst ru cti on th eo ri es of society, we mu st re tu rn t o th e quest ion s tha t
8/15/2019 HABERMAS, Jürgen, On the Pragmatics of Social Interaction
40/216
16
Lecture I
already forced themselves upon us with regard to the as yet unclear
concept of the generation of meaningfully organized structures of
life. If I am right, we can try to clarify