You are on page 1of 28

Making Contingency Safe for Liberalism: The Pragmatics of Epistemology in Rorty and Luhmann Author(s): Cary Wolfe Reviewed

work(s): Source: New German Critique, No. 61, Special Issue on Niklas Luhmann (Winter, 1994), pp. 101-127 Published by: New German Critique Stable URL: . Accessed: 07/11/2011 06:53
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact

New German Critique and Duke University Press are collaborating with JSTOR to digitize, preserve and extend access to New German Critique.

Making ContingencySafefor Liberalism: ThePragmatics of Epistemology in Rorty and Luhmann

What must immediately surpriseany readernew to the discourses of systems theory or what is sometimes called "second-ordercybernetics" is the rathersystematic reliance of this new theoreticalparadigmon the figure of vision and, more specifically, observation.That surprisemight turn into discomfort if not alarm for readersin the humanities who cut their teeth on the critical genealogy of vision and the look which runs, in its modernist incarnation,from Freud's discourse on vision in Civilization and Its Discontents through Sartre'sBeing and Nothingness to Lacan's seminars and finally to recent influentialwork in psychoanalysis and feminist film theory.' With the possible exception of Michel Foucault, no recent intellectual has done more to call into question the trope of vision than America's foremostpragmatistphilosopher,Richard early work Philosophy and the Mirror Rorty. From his ground-breaking Nature onward, Rorty has argued that the figure of vision in the of philosophical and critical traditionis indissolubly linked with represenassumes that "maktationalism and realism, where representationalism true' and 'representing' are reciprocal relations: the nonlinguistic ing item which makes S true is the one representedby S," and realism maintains the "idea that inquiryis a matterof finding out the nature of
1. See, for example, Slavoj Zizek's post-Lacanian analysis of the look in a few different texts, most importantlyTheSublimeObjectofIdeology (London:New Left, 1989) and, within feminism, the wealth of work by critics such as Mary Anne Doane, Laura Mulvey, Kaja Silverman,and many others.



Pragmatics ofEpistemology

something which lies outside the web of beliefs and desires," in which "the object of inquiry- what lies outside the organism- has a context of its own, a context which is privileged by virtue of being the object's ratherthan the inquirer's."2 Instead,Rorty argues,we should reduce this desire for objectivity to a search for "solidarity"and embrace a philosophical holism of the sort found in Dewey, Wittgenstein,and Heidegger, which holds that "words take their meanings from other words ratherthan by virtue of their representative character" "transparency and to the real."3Hence, Rorty rejects the representationalist position and its and argues instead that "Our only usable notion of privileged figure, 'objectivity' is 'agreement'ratherthanmirroring.'"4 Rorty's Deweyan reductionof objectivityto solidarityneatly disposes of all sorts of traditionalphilosophicalproblems, most significantly the problem of relativism, which may now be seen as a "pseudo-problem" for pragmatistphilosophy because, as Rortyputs it, "the pragmatistdoes not have a theory of truth,much less a relativisticone. As a partisanof solidarity, his account of the value of cooperative human inquiry has only an ethical base, not an epistemologicalor metaphysicalone" (ORT 24). On this point, the usefulness of the pragmatistreductionis particularly winning in Rorty's hands.In responseto the chargethatphilosophy in its postmodern and/or pragmatistincarnationautomaticallybecomes relativistonce it has ceased to be foundationalist, Rortyresponds: The view that every tradition as rational as moral as every is or othercouldbe heldonly by a god, someone who hadno needto use the or becauseshe (but only to mention) terms"rational" "moral," had no need to inquireor deliberate. Such a being would have into and escapedfromhistoryandconversation contemplation metTo anarrative. accusepostmodernism relativism to try to put a of is in metanarrative thepostmodernist's mouth (ORT 202). Thus, the Deweyan reductionof objectivity to solidarity provides the ethical basis for the pragmatist'sWittgensteinianepistemology, which insists that "it is contexts all the way down"and that 'graspingthe thing itself' is not something that precedes contextualization,but is at best a
2. RichardRorty, Objectivity, Relativism,and Truth,PhilosophicalPapers, vol. 1 citedparenthetically ORTin the text. as UP, 1991)4, 96. Hereafter Cambridge (Cambridge: 3. RichardRorty,Philosophy and the Mirrorof Nature (Princeton:PrincetonUP, 1979) 368. 4. Rorty,Philosophy and the Mirrorof Nature 191.

Cary Wolfe 103 focus imaginarius"(ORT 100). It would appear,then, that for Rorty the "outside"of belief or description- what used to be called the "referent" - is always already inside, insofar as meaning, to borrow Walter Benn Michaels's formulation,is not filtered through what we believe, but is ratherconstitutedby what we believe. The problem with this position - a problemwhich Rorty recognizes - is that it immediately raises the suspicion that "antirepresentationalism is simply transcendental idealism in linguistic disguise ... one more version of the Kantian attemptto derive the object's determinacy and structurefrom that of the subject"(ORT4). Critics of antirepresentationalism imagine "some mighty immaterial force called 'mind' or 'language' or 'social practice' . . . which shapes facts out of indeterminate Rorty continues, goo," and soothe, problemfor antirepresentationalists," theirpointwhichcarries such suggesno is to find a way of putting need tion. Antirepresentationalists to insistthat"determinacy" not is what is in question- that. . . it is no truerthat"atomsare what they arebecausewe use 'atom'as we do"thanthat"we use 'atom' atomsareas theyare" as we do because (ORT 5). But even if we agree with Rorty that "determinacy" not exactly the is issue here, the question of how the outside can be accounted for at all certainly is. What exactly is the philosophical status of those atoms whirling beyond the deterministicken of our descriptions?Discussing the work of Sellars and Davidson, Rorty writes: "what shows us that life is not just a dream,that our beliefs are in touch with reality, is the links between us and the causal, non-intentional,non-representational rest of the universe" (ORT 159). Pragmatistsdo indeed accept "the brute, inhuman, causal stubbornnessof the gold or the text. But they think this should not be confused with, so to speak, an intentional stubbornness, an insistence on being described in a certain way, its own way" (ORT 83). Were Rorty's account to end here, we would indeed be left with the outside as a black box, as that which must somehow be acknowledged and posited but is not permitted to do any representational work. In fact, though, there is a furtherwrinkle in Rorty'streatmentof this problem, and it is crucial to his avoidance of the double bind which plagues some of the more well-known neopragmatistaccounts of belief. Rorty imagines the recalcitrantrealist responding that the pragmatist,


Pragmatics of Epistemology

given her account, cannot "find out anythingabout objects at all," since "You never get outside your own head." Rorty replies, in one of the more disarming moments in the book, "What I have been saying amounts to accepting this gambit."But, he hastens to add, one of the most central beliefs held by the pragmatistis that "lots of objects she does not control are continuallycausing her to have new and surprising beliefs." Hence, "She is no more free from pressure from the outside, no more temptedto be 'arbitrary,' thananyoneelse" (ORT 101). In contrast to what we might call the "hard"version of belief propounded by neopragmatists such as Steven Knapp, Walter Benn Michaels, and Stanley Fish - which holds that once you have a belief with "no distance"5- Rorty you will inhabit it "withoutreservation," here provides a soft account of belief, one in which beliefs are always held "with reservations"because they are held in a world in which (to use William James's picturesquephrase) things are constantly "boiling over" our beliefs about them and impelling us to revise those beliefs.6 In the parlanceof cyberneticsand systems theorywhich we will explore below, for any given code (or belief), the only source of new information is the as yet uncoded, the randomand unpatterned,the "noise" of the outside. Rorty's point is thatyou may believe whateveryou like, but that belief itself - and here is the pragmatistimperative- will have consequences because it is subject to "pressurefrom the outside." This is the sense, I think, of Davidson's assertion, which Rorty quotes approvingly,that "most of our beliefs are true"- because we are still aroundto talk aboutthem! (ORT9-10). From this vantage point, the imperative to theory, reflection on belief, derives not from an essentialist "appetiteof the mind," to use William James's phrase,7 nor a desire for transcendencein either its realist or idealist incarnation(as Knapp and Michaels would have it in their Against Theory polemic), but ratherfrom the strategic, adaptive, pragmatic value of theory which any act of intellection will ignore only at its own peril. One might insist, with William James, that the desire to
5. See the essays by Fish, Knapp,and Michaels,Against Theory:LiteraryStudies and the New Pragmatism,ed. W.J.T.Mitchell(Chicago: of ChicagoP, 1985)25, 113, 116. U 6. See William James, "Pragmatism's Conceptionof Truth," Pragmatismand The Mass.: Harvard 1978) 106-07. UP, Meaning of Truth,intro.A. J. Ayer (Cambridge, 7. See Frank Lentricchia's discussion of this moment in James - as a pointed contrast to the Against Theory polemic - in his Ariel and the Police: WilliamJames, Michel Foucault, WallaceStevens(Madison:U of WisconsinP, 1988) 124-33.

Cary Wolfe


theorize is "characteristicallyhuman," but "this would be like saying that the desire to use an opposable thumb remains characteristically human. We have little choice," Rorty continues, "but to use that thumb, and little choice but to employ our ability to recontextualize" (ORT 110). Thus, the pragmatist "takes off from Darwin rather than from Descartes, from beliefs as adaptationsto the environment rather than as quasi-pictures"(ORT 10); he thinks "of linguistic behavior as tool-using, of language as a way of grabbinghold of causal forces and making them do what we want, altering ourselves and our environment to suit our aspirations" (ORT 81). In this way, pragmatism "switches attention from 'the demands of the object' to the demands of the purpose which a particularinquiry is supposed to serve." "The effect," Rorty concludes, debatefroma methodologico-ontologiis to modulate philosophical what cal key into an ethico-political For now one is debating key. to purposesare worthbothering fulfill,whichare moreworthwhile the or thanwhichpurposes nature humanity of of thanothers,rather all us to have.Forantiessentialists, possiblepurposes realityobliges competewith one anotheron equal terms, since none are more than human" others 110). (ORT "essentially Yet, at precisely this juncture, the radically pluralist imperative of Rorty's pragmatistcommitmentto contingencybegins to break down or more specifically, begins to be recontainedby a more familiar,more complacentand uncriticalsort of pluralism.For it may be true, as Rorty puts it, that "holism takes the curse off naturalism"(ORT 109), but no sooner does it resituate the philosophical problems of naturalismin an key" than it createsenormous political problemsby rein"ethico-political within the horizonof a debilitatingethnocentrism scribingRorty'sproject and, beyond that, liberal humanism.Rorty's descriptionand defense of ethnocentrismin "Solidarityor Objectivity?" begins by sounding commonsensical enough: "For now to say that we must work by our own lights, that we must be ethnocentric,is merely to say that beliefs suggested by anotherculturemust be testedby tryingto weave them together with beliefs we already have" (ORT 26). But the issue which remains submergedhere, and which remainedsubmergedin the lengthy passage we quoted above, is this: just who is this generic "we" in Rorty's discourse?Thatproblemworks its way to the surfacelaterin the same essay, where Rortywrites,againin a seemingly commonsensicalmoment:


Pragmatics ofEpistemology The pragmatists' of free inquiry,and the justification toleration, for undistorted communication only take the form of a can quest comparisonbetween societies which exemplifythese habits and those which do not, leadingup to the suggestion nobodywho that has experienced bothwouldpreferthe latter... Suchjustification is not by reference a criterion, by reference variousdetailed to but to 29). (ORT advantages practical

Even if we leave aside the gesture toward "undistorted communication" (a gesture which Rorty himself has critiquedin Habermas8),and even if we subscribe to the bourgeois liberal values which Rorty inventories, the question that never gets asked is whether all members of Rorty's society experience these "detailedpractical advantages"in the way that Rorty imagines. Such liberal values and freedoms may extend to all in the abstract- that is, in theory- but do they in fact, in practice? Clearly the answer is no. This need not lead us to reject out of hand the liberal values Rorty regularlyinvokes; it is simply to point out that when Rorty claims that "we" should encourage the "end of ideology" (ORT 184), that "anti-ideologicalliberalism is, in my view, the most valuable traditionof Americanintellectuallife" (ORT 64), Rorty is staging a claim that is itself ideological throughand through.It is ideological, moreover,in termseasily legibleby the light of the most basic sort of Marxist critique of ideology familiar since The German Ideology: Rorty representsthe interestsand "detailedpractical advantages"of his own class, "postmodernbourgeois intellectuals,"as the interests of the entire society, ratherthan acknowledging,as his critics of various stripe have urged him to do, that those freedoms and advantages enjoyed by him and his class are not only not sharedby those beneath him on the social and economic ladder,but are in fact purchasedat their expense, built upon the very prosperity extracted from their exploitation and alienation and distributedwith gross and systematic inequity in the economic and social sphere. In other words, what Rorty does not recognize is that there is a fundamentalcontradictionbetween his putative desire to extend liberal advantagesto an ever larger community, and the fact that those advantages are possible for some only because they are purchased at the expense of others. This is simply to say that liberal free8. See RichardRorty, "Habermasand Lyotardon Postmodernity," Habermas in and Modernity,ed. RichardBernstein(Cambridge,Mass.: MIT, 1985) 161-75. See also note 29 below.

Cary Wolfe


doms in the context of global capitalism rest in large part upon economic power and freedom, and that both are operative within a context of scarcity. Rorty may bridle at FrankLentricchia'sclaim in Criticismand Social a Change that "our society is mainly unreasonable," charge to which "unreasonableby comparisonto what other society?" Rorty responds, (ORT 15). But what Rorty does not see - and this is the force of Lentricchia'scharge - is that Rorty's society, while perhaps not unreasonable for those in Rorty's position, is indeed unreasonable for many millions less fortunatethan he. As Nancy Fraser puts it, the problem with "the communitarian comfortof a single 'we'" is that that social space,assuming tendentiously there Rortyhomogenizes of are no deep social cleavagescapable generating conflictingsolidaritiesandopposing"we's."It follows fromthis assumedabsence that of fundamental socialantagonisms politicsis a matter everyof one pulling togetherto solve a commonset of problems.Thus, can Disconnected social engineering replacepoliticalstruggle. tinkwith a successionof allegedlydiscretesocial problemscan erings of structure.9 transformationthebasicinstitutional replace In this light, it is deeply symptomaticthat Rorty relies upon the lanwhose homogenizing connotaguage of "democracy"and "community," tions mask and submerge the "unevenness"in the social and economic sphere which a very differentlanguage- the language of "capital"and "class" or, as Luhmann will claim, of functional differentiation would force to light. To Rorty's credit, he tries in his most recent work to confront the If problems raised by his ethnocentrism. "we heirs of the Enlightenment think of enemies of liberal democracy like Nietzsche or Loyola as, to use Rawls's word, 'mad"' (ORT 187), then, he acknowledges, "sud9. Nancy Fraser,UnrulyPractices:Power,Discourse and Genderin Contemporary Social Theory (Minneapolis:U of Minnesota P, 1989) 104. This is not to agree with call Fraser'sHabermasian for a renewedattentionto the normative.Fraser'spoint is strikingly borne out, however, in Rorty'sresponseto CliffordGeertz'scritiqueof his ethnocentrism, where Rorty declares that the "whole apparatusof the liberal democratic state" functions just fine when it ensures that the drunkenIndian needing dialysis in Geertz's example is "going to have more years in which to drink than he would otherwise have had," and that the point for doctors and lawyers in such cases is - and here the quintessence of the sort of technocraticfunctionalismwhich Fraserdescribes- "to get theirjob Relativismand Truth204-05. done, and to do it right."Rorty,Objectivity,


Pragmatics ofEpistemology

denly we liberal democratsare faced with a dilemma," for, "To refuse to argue about what humanbeings should be like seems to show a contempt for the spirit of accommodationand tolerance, which is essential to democracy." But Rorty quickly dispenses with this dilemma by insisting, "Accommodationand tolerance must stop short of a willingness to work within any vocabulary that one's interlocutorwishes to use" (ORT 190). Again, Rorty's position here may seem reasonable enough, but the problem is that those who are declaredbeyond the pale of reason during the course of Rorty's recent work include not only Nietzsche and Loyola, but also Gilles Deleuze, Jean-FrangoisLyotard, Michel Foucault, and all those whom Rorty calls, in a recent New York Times Op-Ed piece, the "unpatrioticleft" of the American academy, which "refuses to rejoice in the countryit inhabits"and "repudiatesthe idea of a national identity, and the emotion of national pride"10 that all those who have experienced "an apparentloss of faith in liberal is, democracy"(ORT220). Rorty constantly invokes the liberal intellectual's dedication to expanding the range of democratic privileges, freedoms, and values, but what becomes clear in his recent work is that such an expansion can take place only after the democratic ethnos has been purified of the sort of dissent it needs to encourage. Rorty wants a pluralism that is not too plural, a democracy that is not too democratic to call into question the values and benefits enjoyed by those of Rorty's class and cultural status. Time and again, rather than enlisting those with very of basic, fundamentaldisagreementsinto the "conversation" liberaljustice and democracy,he declares them out of the loop - declares them non-subjects, "fanatics," or "fantastics"- to begin with. And what kind of "conversation"is that? Rorty is quite clear on this point: to "accept the fact that we have to start from where we are," he writes, means that "there are lots of views which we simply cannot take seriously. . . [W]e can understand the revolutionary's suggestion that a

sailable boat can't be made out of the planks which make up ours, and that we must simply abandonship. But we cannot take this suggestion seriously" (ORT 29). For Rorty,to take this as a "live option" is not to be a rational partner in the democratic conversation, but - and here the red herring is hard to miss - to be one of those "people who have
10. RichardRorty,"The Unpatriotic Academy,"TheNew YorkTimes(February13, 1994) section E, 15.

Cary Wolfe 109 always hoped to become a New Being, who have hoped to be converted ratherthanpersuaded" (ORT29). Thus, the openness of liberal democraticprocess is purchased at the expense of the closedness of the liberal democraticcommunity,and the "conversation"of liberaljustice excludes from the very beginning those with whom substantive differences might be discussed. From this vantage point, it is clear, as Cornel West puts it, that Rortyan pragmatism "only kicks the philosophic props from underliberal bourgeois capitalist societies; it requires no change in our culturaland political practices." Rorty's project, "thoughpregnantwith possibilities ... refuses to give birth to the offspring it conceives. Rorty leads philosophy to the complex world of politics and culture,but confines his engagementto transformationin the academy and to apologetics for the modem West."ll In the end, then, Rorty's philosophicalcommitmentto contingency and the radical pluralism it promises is recontainedby his liberal humanism. Hence, we are forced to say that in Rorty's pragmatism,representationalism is indeed undone on the philosophicallevel only to re-emergein a more powerful and insidious form on the plane of the political.12 As we have seen, Rortyanpragmatismmoves to front and center the revisable, self-critical and reflexive nature of all beliefs and descriptions, only to recontain that commitmentto contingency within an ideology of liberalismwhich declares out of the picture from the outset (to turn the ocular metaphorback upon Rorty) those social others whose very otherness or difference might lead to a critical assessment of one's own belief. Hence, the Rortyan view gives us no way to theorize the between antagonisticbeliefs in the productive and necessary relationship
11. CornelWest, TheAmericanEvasion ofPhilosophy:A GenealogyofPragmatism (Minneapolis:U of MinnesotaP, 1989) 206-07. 12. In this light, Rortyanpluralism seems subject to Gilles Deleuze's critique of "state philosophy" as describedby Brian Massumi: "More insidious than its institutionbased propagationis the State-form'sability to propagateitself without centrallydirected inculcation(liberalismand good citizenship).Still more insidious is the process presiding over our presentplight, in which the moral and philosophicalfoundationsof nationaland personal identity have crumbled,making a mockery of the State-form- but the world keeps right on going as if they hadn't."See Massumi'sA User's Guide to Capitalismand Schizophrenia:Deviationsfrom Deleuze and Guattari(Cambridge,Mass.: MIT, 1992) 5. as In this connection, also see the discussion of the double sense of "representationalism" of in "representative" democracyas well as in the presumedtransparency the sign to thing signified in the conversationbetween Foucault and Deleuze entitled "Intellectualsand Practice (Ithaca: Cornell UP, Power," Michel Foucault, Language, Counter-Memory, 1977) 206.


Pragmatics ofEpistemology

social sphere. It is on the terrainof this last problem that Niklas Luhmann's systems theory takes a decisive step beyond what Cornel West has called Americanpragmatism's "evasionof epistemology-centered philosophy,"and in the process makes clearthata philosophicalcommitment to theorizing the pragmaticsof contingency needs more epistemologycenteredphilosophy,not less. For my purposes here, Luhmann'skey innovation in this connection is his theorization of "the observation of observation," which attempts to make use of the ocular metaphorby divorcing it from the sorts of unifying and representationalistdesigns which are critiqued by Rortyan philosophy only to reappearin Rortyan politics. For Luhmann, all observations are constructed atop a constitutive paradox or tautology which those systems cannot acknowledge and at the same time engage in self-reproduction.All systems, in other words, are constituted by a necessary "blind spot" which only other observing systems can see and disclose, and the process of social reproduction depends upon the undoing, distribution,and circulation of these constitutive paradoxes and tautologies (which would otherwise block systemic self-reproduction) by a plurality of observing systems. Both Luhmann and Rorty begin from the Wittgensteinianposition that "a system," as Luhmannputs it, "can see only what it can see. It cannot But Luhmann,unlike Rorty, derives from this see what it cannot."l13 formulation not the irrelevance of other observing systems (or beliefs), not their exclusion from the conversation of social reproduction, but rathertheir very necessity. Luhmann's theorization of the concept of observation and its relation to contingency is heavily indebted to the pioneering work in biology and epistemology of Humberto Maturana and Francisco Varela, and specifically to their theorization of what they term "autopoiesis." The cornerstone distinction for the theory of autopoiesis is between "organization" and "structure,"or what Luhmann will call "system" and "element," both of which roughly (but only roughly) correspond to Rorty's closed circuit of "belief' on the one hand and the open, "causal," and "non-intentional"set of relations between belief and the world of "facts" on the other. As Maturanaand Varela explain it, "Organization denotes those relations that must exist
13. Niklas Luhmann,Ecological Communication, trans.JohnBednarz (Chicago: U of Chicago P, 1989) 23.

Cary Wolfe 111 among the components of a system for it to be a member of a specific class"; structure, on the other hand, "denotes the components and relations that actually constitute a particular unity and make its For Maturanaand Varela, what characterizes all organization real."l14 is that the relationship between structure and organizaliving things tion is one of "autopoietic organization," that is, "they are continually self-producing"15 according to their own internal rules and requirements. In more general terms, what this means is that all autopoietic entities are closed or, to employ Luhmann's preferred term, "self-referential" on the level of organization, but open on the level of structure. This is most clear, perhaps, in Maturanaand Varela's theorization of what they call "operational closure."16 "It is interesting to note," they write, that the operational closureof the nervoussystem tells us that it to does not operate it according eitherof thetwo extremes: is neither nor representational solipsistic. It is not solipsistic, becauseas partof the nervoussystem'sorganin of ism, it participates the interactions the nervoussystemwith its Theseinteractions environment. continuously triggerin it the strucits of that tural changes modulate dynamics states.... for it Nor is it representational, in each interaction is the nervous statethatspecifieswhatperturbations possiare system'sstructural ble andwhatchanges them.17 trigger The theorization of the operational closure of autopoietic entities allows Maturana and Varela to break with the last vestiges of the representationalist view and to "walk on the razor's edge, eschewing the extremes of representationalism(objectivism) and solipsism (idealism)."18 "These two extremes," Varela, Evan Thompson and Eleanor Rosch contend in The Embodied Mind, "both take representation as their central notion: in the first case representation is used to recover what is outer; in the second case it is used to project what
14. HumbertoR. Maturanaand FranciscoJ. Varela, The Tree of Knowledge: The revised edition, trans.Robert Paolucci, foreBiological Roots of Human Understanding, word by J. Z. Young (Boston: Shambhala,1992) 46, 47. and 15. Maturana Varela43. and 16. See Maturana Varela163-67 for a more detaileddefinitionand discussion of operationalclosure. and 17. Maturana Varela169. and 18. Maturana Varela241.


Pragmatics ofEpistemology

is inner."19As Maturanaand Varela frame the problem in The Treeof Knowledge, "If we deny the objectivityof a knowable world,"they ask, because everythingis pos"arewe not in the chaos of total arbitrariness Gordianknot," as they put it, is to sible?" The way to "cut this apparent change the nature of the question, to realize that the first principle of any sort of knowledge whatsoever is that "everythingsaid is said by someone"20- that is, to foregroundthe contingency of what Luhmann will call observation. Luhmann'srefinementof the concept of observationis a key component of his extension of the concept of autopoiesisfrom the realm of living systems (the focus of Maturanaand Varela) to social systems. "If we abstractfrom life and define autopoiesisas a general form of system building using self-referential closure," Luhmann writes, "we would have to admit that there are nonliving autopoietic systems."21For Luhmann as for Maturanaand Varela,the attractionof the concept of autopoiesis - or what Luhmannwill more often treat under the term "selfreference"- is not least of all that the theorizationof systems as both (operationally) closed and (structurally)open accounts for both high degrees of systemic autonomyand the problem of how systems change and "adapt"to their environments(or achieve "resonance"with them, as Luhmannexpresses it in Ecological Communication).22 But Luhmannextends and refines the work of Maturanaand Varela in the particulartheoreticalpressurehe applies to the problem of observation. It will come as no surprisethat Luhmannagrees with Maturana and Varela that "Autopoieticsystems . . . are sovereign with respect to the constitution of identities and differences. They, of course, do not create a material world of their own. They presuppose other levels of reality.... But whatever they use as identities and as differences is of their own making"(ESR 3). Yet in his essay "Complexityand Meaning,"
19. FranciscoJ. Varela,Evan Thompson,and EleanorRosch, The EmbodiedMind: Mass.: MIT, 1993) 172. CognitiveScience and HumanExperience(Cambridge, and 20. Maturana Varela135. "TheAutopoiesisof SocialSystems,"Essays on Self-Reference 21. Niklas Luhmann, to references this volumeof essays will be pro(New York:Columbia 1990)2. Subsequent UP, vided parentheticallyin the text as ESR. See also the translator'sintroductionto Luhmann'sEcological Communication. 22. In Luhmann's of work,this is partof the moregeneraltheorization what he refers to as "functionallydifferentiated" modem society (as opposed to hierarchical premodern see ones). For a short summary, Luhmann'sessay "TheSelf-Descriptionof Society: Crisis Fashion and Sociological Theory," International Journalof Comparative Sociology XXV, 1-2 (1984): 59-72.

Cary Wolfe


Luhmann pushes beyond Maturana and Varela and distinguishes between a system's operation and its observation. "By operation," he writes, "I mean the actual processing of the reproduction of the system." "By observation, on the other hand," he continues, "I mean the act of distinguishing for the creation of information" (ESR 83). The distinction between operation and observation, Luhmann writes elsewhere, "occupies the place that had been taken up to this point by the unity-seeking logic of reflection. (This means, therefore, a substitution of differencefor unity)"23 about which we will say much more in a moment. Luhmann distinguishes a third term here as well: self-observation. "Self-referential systems are able to observe themselves," he writes. "By using a fundamental distinction schema to delineate their selfidentities, they can direct their own operationstoward their self-identities" (ESR 123). If they cannot distinguish what is systemic and internal from what is environmentaland external,then they cease to exist as autopoietic, self-producing systems. This is why Luhmann writes that the distinction between "internal"and "external"observation "is not needed," and that "the concept of observation includes 'self-observation'" (ESR 82). In other words, to observe at all requires an autopoietic system, and an autopoietic system capable of observation cannot exist without the capacity for self-observation - that is, without the capacity "to handle distinctions and process information."24Hence, observation and, within that, self-observation,are necessary operations of autopoietic systems. All of this leads us to the centralpoint we need to understandabout Luhmann's concept of observation and its relationship to the epistemological problem of constructivism. Luhmann'sposition is clearest, perhaps, in his explanation of the observation of observation in his importantessay "The Cognitive Programof Constructivismand a Reality that Remains Unknown," where he writes, "An operation that uses
23. Niklas Luhmann,"The Cognitive Programof Constructivismand a Reality that Portrait of a Scientific Revolution,eds. Wolfgang Remains Unknown,"Selforganization: Krohn, Gunter Kuppers, and Helga Nowotny (Dordrecht:Kluwer, 1990) 68, emphasis mine. 24. Luhmannqualifies this somewhat in "Complexityand Meaning":"it has to be decided,"he writes, "whetherself-observation(or the capacityto handle distinctions and process information)is a prerequisiteof autopoieticsystems," Luhmann,Essays on SelfReference 82. It seems, though, that the position outlined earlier in the essay - that the concept of observationautomaticallyincludes that of self-observation- would seem to requireself-observationas such a prerequisite.


Pragmatics ofEpistemology

distinctions in order to designate something we will call 'observation.' We are caught once again, therefore, in a circle: the distinction between operation and observation appears itself as an element of observation."25 Most readers would probably agree with Luhmann (and beyond that, with the work of George Spencer Brown, which Luhmann draws upon) that the most elementary intellectual and psychical act is to draw a distinction, to distinguish figure from ground, "x" from "not-x." The point Luhmann wishes to underscore, however, has been a familiar one ever since the "liar's paradox" of antiquity,26 or more recently the theory of "logical types" of Russell and Whitehead which tried to solve such antinomies: that drawing such a distinction, which is the elementary constitutive act of observation, is always either paradoxical or tautological, and that this is both necessary and unavoidable.27 "Tautologies are distinctions" Luhmann writes, thatdo not distinguish. Theyexplicitly negatethatwhatthey distinmakesa difference. thus guishreally Tautologies blockobservations. schema: is They are alwaysbasedon a dualobservation something whatit is. This statement, however, negatesthe positeddualityand assertsan identity. thus Tautologies negatewhatmakesthempossithe ble in the first place, and, therefore, negationitself becomes (ESR136). meaningless To many readers,this descriptionwill evoke nothing so much as the famous Hegelian postulate of "the identity of identity and non-identity." What Luhmannwishes to stress, however, is not the identity of identity and non-identity but rather the non-identity(or difference) of identity and non-identity.As he puts it in Ecological Communication, that the unity(of self-reference) wouldbe unacceptable the form in of a tautology(e.g., legal is legal)or a paradox (one does not have the legal rightto maintain one's legalright)is replaced a differby of ence (e.g., the difference legal andillegal).Thenthe systemcan to oscillatewithinit and develop proceedaccording this difference, to regulatethe ascription the operations the code's of of programs
25. Luhmann,"CognitiveProgram" 68-69. xiv. 26. See Luhmann,Ecological Communication, 27. Luhmann addresses the so-called "theory of logical types" of Russell and Whitehead in many places; see, for example, "Tautology and Paradox," Luhmann, Essays on Self-Reference 127; and, for a more extensive refutation,Luhmann,Ecological Communication23-24.

Cary Wolfe 115 withoutraisingthe questionof the positionsand counter-positions code'sunity.28

Two points need to be stressedhere. First,what enablesthis crucial of and emphasison the difference identityand non-identity separates fromthe Kantianism whichhe bearsmorethana passing Luhmann with is resemblance Luhmann's strident of rejection anypossibilityof a tranFor all are scendentalsubject-observer. Luhmann, observations proobserver who could always,in theory,describe duced by a contingent and Hence, all observations all systemsdescribed by thingsotherwise.
elementof complexity. WilliamRaschputs As them contain an irreducible it, for Luhmann- contra Hegel and Kant - "complexitycan never be fully reducedto an underlyingsimplicity since simplicity, like complexity, is a construct of observationthat could always be other than it is. Contingency,the ability to alterperspectives,acts as a reservoirof complexity within all simplicity."29 The second point that needs to be underscoredin Luhmann'sposition on tautology which we quoted above - and it is one whose pragmatic impulse will distinguish Luhmann'sposition from that of Derrida and deconstruction,at least in Luhmann'seyes30- is the insistence that the nature of all observation tautological (or, more strictly, paradoxical31) constitutes a real, pragmaticproblem for all social self-descriptions.
28. Luhmann,Ecological Communication, see also 37. xiv; 29. William Rasch, "Theoriesof Complexity,Complexities of Theory: Habermas, Luhmann,and the Study of Social Systems," GermanStudies Review 14 (1991): 70. As Rasch points out, this is precisely the point that is missed by Habermas'sprojectof a universal pragmatics. "The whole movement of Habermas'sthought,"he writes, "tends to some final restingplace, prescriptivelyin the form of consensus as the legitimate basis for social order,and methodologicallyin the formof a normativeunderlyingsimple structure which is said to dictatethe propershapeof surfacecomplexity."Rasch, 78. 30. In a recent essay, Luhmannwrites of the lineage that runs from Nietzsche is throughHeidegger to Derridathat in theirwork, "Paradoxicality not avoided or evaded but, rather,openly exhibitedand devotedlycelebrated..... At present,it is not easy to form a judgment of this. Initially,one is impressedby the radicalitywith which the traditional Europeanmodes of thoughtare discarded.... [But it] has so far not producedsignificant of results. The paradoxicalization civilization has not led to the civilization of paradoxito cality. One also starts to wonder whether it is appropriate describe today's extremely dynamic society in terms of a semantic that amounts to a mixture of arbitrarinessand trans. Bernd Widdig, StanfordLiterature paralysis."Niklas Luhmann,"Sthenography," Review 7 (1990): 134. 31. As Luhmannpoints out, tautologies are actually "special cases of paradoxes" such that "tautologiesturn out to be paradoxes,while the reverse is not true."Luhmann, Essays on Self-Reference136.


Pragmatics ofEpistemology

This is so, Luhmannargues, because "An observer can realize that the self-referential systems are constituted in a paradoxical way. This insight itself, however, makes observationimpossible, since it postulates an autopoieticsystem whose autopoiesisis blocked"(ESR 139). The solution to these obstacles or blockages is that self-referential paradoxes must be, in Luhmann'ssomewhat frustratingnomenclature, "undone" by the system [Entparadoxierung].We have already mentioned two ways in which such undoing might take place: the theory of or logical types, which "interrupts" undoes the vicious circle of paradoxical self-reference "by an arbitraryfiat: the instruction to ignore and the operations that disobey the command to avoid paradoxes";32 reliance upon binary coding, which enables the system to undo its paradoxes by orienting the operationsof the system to the difference of x and not-x (legal and non-legal, for example) without raising the question of theirparadoxicalidentity. But if Luhmann'sconcernwith the pragmaticsof tautology and paradox for social reproductionseparateshim from Derridaand deconstruction, his position on how the practical-politicalundoing of tautology and paradox ought to be handledseparateshim from consensus-seeking liberals such as Rorty or Habermas.For if the processes of undoing paradox require that a system's constitutiveparadoxremain invisible to it, then the only way that this fact can be known as such is by the observation of another observing system. As Luhmannputs it, "Only an [other] observer is able to realize what systems themselves are unable to realize" (ESR 127). What is decisive about Luhmann'sinterventionhere is his insistence on the constitutiveblindness of all observations,a blindness which does not separate or alienate us from the world but, paradoxically, guaranteesour connectionwith it. As Luhmann explains it, in a remarkablepassage worth quoting at length: The sourceof a distinction's guaranteeing realitylies in its own operativeunity[as in, forexample, distinction the It legalvs. not-legal]. is, as cannot observed be however, precisely thisunitythatthe distinction - except by meansof another distinction which then assumesthe of of function a guarantor reality. Another of expressing is to this way with the worldwhich as a the operation say emergessimultaneously resultremains to unapproachabletheoperation. cognitively
32. 24. Luhmann,Ecological Communication

Cary Wolfe 117 The conclusion be drawn with to fromthisis thatthe connection worldis established the blindspotof the the realityof the external by Realityis whatone does notperceivewhenone cognitiveoperation.
perceives it.

Perception and cognition of reality, in other words, are made possible by the deployment of a paradoxical distinction to which the observation utilizing that distinction must remain "blind" if it is to perceive and know at all. Here, Luhmannneatly traverseswhat has traditionally seemed an insoluble epistemological problem:how to avoid the untenable reliance upon the distinction between science and ideology which has traditionally buttressed ideology critique and the sociology of knowledge, and at the same time avoid lapsing into epistemological solipsism. Luhmann's negotiation of this problem is possible only on the strength of systems theory's articulation of the observation of observation, which enables us to view the "blind spot" or "latency" of the observations of others not merely as ideological bias or the distortion of a pregiven reality knowable by "science," but rather as the unavoidably partial and paradoxicalpreconditionof knowing as such.34 This, Luhmannwrites, is "the systematic keystone of epistemology taking the place of its a priori foundation."3 "In a somewhat Wittgensteinian formulation,"he writes, one couldsay thata systemcansee onlywhatit cansee. It cannot see it see see whatit cannot. Moreover, cannot thatit cannot this.... Nevertheless,a system that observes other systems has other of possibilities.... [T]heobservation a systemby anothersystem - following Humberto Maturana will call this "second-order we forced on the observation" can also observe the restrictions observedsystemby its own modeof operation. It can observe .... the horizons of the observedsystem so that what they exclude becomesevident.36
33. Luhmann,"CognitiveProgram" emphasismine. 76, "The assumption- to be found above 34. As he relates it in "CognitiveProgram": all in the classical sociology of knowledge- thatlatentstructures, functions, and interests lead to distortionsof knowledge, if not to blatanterrors,can and must be abandoned.The impossibility of distinguishingthe distinctionthat one distinguishes with is an unavoidable preconditionof cognition. The questionof whethera given choice of distinctionsuits one's latent interests only arises on the level of second order observation [that is, on the 73. level of the observationof observation]." Luhmann, "CognitiveProgram" 75. 35. Luhmann,"CognitiveProgram" 23. 36. Luhmann,Ecological Communication


Pragmatics ofEpistemology

Here we need to sharpen our sense of the pragmatic implications of Luhmann's epistemology and how it differs from Rortyan pragmatism. The assertion we quoted a moment ago that "the constructed reality is not ... the reality referred to" must surely remind us of Rorty's attempt to situate descriptions within a "non-intentional" and "causal" world without having that world do representational work. But what follows - that "the connection with the reality of the external world is established by the blind spot of the cognitive operation," and that "Reality is what one does not perceive when one perceives it" - separates Luhmann's crucial reformulation from Rorty. For Luhmann stresses the contingency and paradoxicality of that very observation itself and, contra Rorty, derives from that contingency the necessity of the observations of others: it is only in the mutual observations by different observers that a critical view of any observed system can be formulated. If we are stuck with distinctions which are paradoxical and must live with blind spots at the heart of our observations, Luhmann writes, "Perhaps, then, the problem can be distributedamong a plurality of interlinked observers" who are of necessity joined to the world and to each other by their constitutive but different blind spots. The work of social theory would then consist in developing "thoughtful procedures for observing observation, with a special emphasis on that which, for the other, is a paradox and, therefore, cannot be observed by him."37 And while this reformulation is neither, strictly speaking, a politics nor an ethics, it does provide a rigorous and persuasive theorization of the compelling necessity of sociality as such - that is, of necessary, reciprocal, and yet asymmetrical relations between self and other, observer and observed, relations which can no longer be characterized in terms of an identity principle (be it of class, race, or what have you) which would reduce the full complexity and contingency of the observer's position in the social space. In these terms, Luhmann's insistence on the blind spot of observation and, therefore, the essential aporia of any authority which derives from it (the authority, say, of the system which enforces the distinction legal/illegal) bears more than a passing resemblance to the proposition of a fundamental "antagonism"at the core of social
37. 137. Luhmann,"Stenography"

Cary Wolfe 119 relations as recently theorized by Ernesto Laclau, Chantal Mouffe, and Slavoj Zizek. As Zizek articulates the concept, "far from reducing all reality to a kind of language-game, the socio-symbolic field is conceived as structured around a certain traumatic impossibility, around a certain fissure which cannot be symbolized."38 Or, to remind ourselves of Luhmann'sformulation, "the connection with the reality of the external world is established by the blind spot of the cognitive operation. Reality is what one does not perceive when one perceives it." For Zizek as for Luhmann, "every identity is already in itself blocked, marked by an impossibility,"39and thus "the stake of the entire process of subjectivation, of assuming different subjectpositions" - or in Luhmann's system, of a plurality of interlinked observers whereby paradox and tautology can be distributed in the social field - "is ultimately to enable us to avoid this traumatic experience"40 of the fact, as Luhmann articulates it, that it is our blind spot that assures our connection with the real, that "The constructed reality is ... not the reality referredto."41 For Zizek, the concept of social antagonism, which countenances "an ethics of confrontation with an impossible, traumatic kernel not covered by any ideal," constitutes "the only real answer to Habermas, to the project based on the ethics of the ideal of communication without constraint,"because it unmasks the constitutive disavowal at work in Habermas's model: "I know very well that communication is broken and perverted, but still . . . (I believe and act as if the ideal speech situation is already realized).'"42For Habermas, we will remember, complexity and contingency always contain the threat of relativism and even nihilism, and thus the proliferation of different systems of knowledge and value must be grounded in some sort of
38. Slavoj Zizek, "BeyondDiscourse-Analysis,"afterwordto ErnestoLaclau,New Reflections on the Revolutionof Our Time(London:New Left, 1990) 249. 39. Zizek, "BeyondDiscourse-Analysis" 252. 40. Zizek, "BeyondDiscourse-Analysis"253. 41. This is to leave aside, of course, the pronounceddifferencesbetween Zizek and Luhmann:Zizek's conjugationof these issues in a psychoanalyticregister in which the concepts of traumaand affect are crucial;and the fact that Zizek's project remains very much within the terms of Cartesianor Kantianidealism, even if it inverts those terms so that the symbolic, signifier,or idea now appearsas the failed "gentrification" the Real, of "the thing," the body - of all that Kant in the Critiqueof Practical Reason calls "the pathological." 259. 42. Zizek, "BeyondDiscourse-Analysis"


Pragmatics ofEpistemology

underlying simplicity. For Habermas - but not, significantly, for Rorty43- that simplicity is harboredin the very nature of language itself and its fundamentalpresuppositionof an ideal speech act, of an undistorted communicationthrough which the claims of different systems of thought and value can be adjudicatedin a process of rational dialogue that arrives at common norms and values.44 But Zizek, like Luhmann,does not disavow the "brokenand perverted"(i.e., paradoxical and tautological) nature of communication,but ratherderives from that brokenness the necessity of sociality as such. He holds that "what this fetishistic logic of the ideal is masking is, of course, the limitation proper to the symbolic field as such: the fact that the signifying field is always structuredaround a certain fundamentaldeadlock"45or what Luhmanncalls the "blockage"of paradoxicalself-reference. Like the theorists of social antagonism- and like them, against Habermas and against Rortyanethnocentrism- Luhmanninsists that the distributionof the problem of paradoxicalityand the circulation of latent possibilities can take place only if we do not opt for the quintessentially modernist and Enlightenmentstrategy of the hoped-for reduction of complexity via social consensus. If all observation is made possible by a paradoxicaldistinctionto which it must remainblind, or Thisis why all projection, the setting a goal,everyformation of of necessitates recursiveobservation why, furthermore, and episodes observation makespossiblenot so muchthe elimination recursive of as and onto paradoxes theirtemporal socialdistribution different operaof tions.A consensual of is, integration systems communication given thatshouldsoonerbe fearedthansought such conditions, something can for. For such integration only resultin the paradoxes becoming to that invisible allandremaining wayforanindefinite future.46 For Luhmann,the Habermasian strategy(or, for that matter,the Rortyan
43. For a sketch of their differences,see Rorty's essay "Habermasand Lyotardon ed. Postmodernity," RichardBernstein,Habermasand Modernity 161-75. As Rorty puts of it, "the troublewith Habermasis not so much that he providesa metanarrative emancipation as that he feels the need to legitimize, that he is not content to let the narratives which hold our culturetogetherdo theirstuff. He is scratching where it does not itch" 164. 44. See, for example, the balancedaccount of Habermas'sproject in Steven Best and Douglas Kellner, Postmodern Theory:Critical Interrogations(New York:Guilford, 1991) 233-55. See also Rasch, 70-72. 259. 45. Zizek, "BeyondDiscourse-Analysis" 75. 46. Luhmann,"CognitiveProgram"

Cary Wolfe 121 one of liberal recontainmentof contingency via ethnocentrism) is a doomed and potentially dangerous project which might result in the blockage of communicationsand the "renderinginvisible," rather than the unfolding, distributionand undoing of paradox. Clearly, then, the Luhmannian concept of observation is not "intended to provide a groundingfor knowledge, but only to keep open the possibility of observation operations' being carriedout by very differentempirical systems - living systems, systems of consciousness, systems of communications."47 And just as clear, too, is Luhmann'sresolute posthumanism, which concludes that what Habermas characterizesas the project of Enlightenment and modernity has, and must, come to an end. "With this,"he writes, of the traditionalattribution cognitionto "man"has been done that is with. It is clearhere, if anywhere, "constructivism" a away new theory, knowledge,a posthumanistic of one. This completely but is not intended maliciously only to makeclearthatthe concept for as "man" the singular!), a designation the bearerand guar(in mustbe renounced. realityof The antorof the unity of knowledge, of cognitionis to be found in the currentoperations the various systems.48 autopoietic There is a pragmatic premium in this philosophical difference, for in Luhmann's view the movement to a posthumanist perspective has the practical benefit of enabling the "betterfunctional performance"49 of a highly differentiated society and its component systems. For example, in Ecological Communication,Luhmann argues that "a sensible handling of systems-theoreticalanalyses" will "lead more to the expansion of the perspectives of problems than to their suppression."50 Such analysis, he contends, can provide an importantcounterbalance to destructive social anxiety, which "is more likely to stop the effects of society on its environment, but . .. has to pay for this by risking unforeseeable internal reactions that again produce anxiety."51 Here, we might think of the so-called Spotted Owl controversy, where social anxiety about biodiversity and habitat destruction did indeed "stop the effects of society on its environment,"but at the expense of
47. 48. 49. 50. 51. 78. Luhmann,"CognitiveProgram" 78. Luhmann,"CognitiveProgram" 128. Luhmann,Ecological Communication 131. Luhmann,Ecological Communication 131. Luhmann,Ecological Communication


Pragmatics ofEpistemology

creating a severe generalized backlash of anxiety about environmental protectionat the expense of economic well-being, one which threatened, ironically enough, to have severe repercussionsfor the re-authorization by Congress in 1993-94 of the EndangeredSpecies Act, the very act which mandatedthe protectionof the SpottedOwl in the first place! It is importantto note, however, that Luhmannmakes it abundantly clear that the pragmatic value of his theorization of complexity and functional differentiationis to enable this world - and more specifically this liberal, western, capitalistworld - to engage in systemic selfreproduction without destructiveblockages of autopoiesis and thereby achieve maximumresonancebetween the system and its environment.52 Quite surprisingly, given his epistemological innovations, Luhmann takes wholly for granted the enclosure of thought - even putatively revolutionary thought - by the western, liberal, capitalist social system. As he expresses it in Political Theory in the WelfareState, the basic problem for any would-be criticalposition is that act, everyoperational everystructural process,everypartialsystem in participates the society, and is society, but in none of these instancesis it possibleto discernthe existenceof the whole sociof out ety. Even the criticisms societymustbe carried withinsociEven the planning of society must be carried out within ety. of out society.Even the description societymustbe carried within society.53 And while Luhmannhere would seem to registernothing more than an epistemological truism, in fact he goes a good bit farther- as Danilo Zolo has pointed out - in his tacit endorsement liberalcapitalistsociof ety and "neo-liberal"policies (a fact more than hinted at in Luhmann's political essays and in his systematicallyreductive glances at Marxist
52. This is to leave aside Danilo Zolo's critiqueof the pragmatic value of Luhmann's work. As Zolo points out, it is so relentlessly abstract circuitous it is hardto imagine and that how anyone otherthana quitesophisticated couldmakepracticaluse of systemsphilosopher Luhmann'saccount.And even if we are suspiciousof this charge,which might, it could be we argued,harbora latentanti-intellectualism, must agreewith Zolo that Luhmann paintsa that pictureof complexityso daunting it is hardto imaginehow one couldever,based on this the theory,anticipateand directdecisions regarding social system - as Luhmannhimself See Crifreely, sometimesalmostgleefully,often acknowledges. Danilo Zolo, "Autopoiesis: Telos86 (Winter1990-91):79. tique of a Postmodern Paradigm," 53. Niklas Luhmann,"The Representation Society Within Society," in Political of State, trans.JohnBednarz,Jr.(Berlin:Walterde Gruyter,1990) 17. Theoryin the Welfare

Cary Wolfe 123 theory).54As Zolo puts it, Luhmanninterprets the crisisof the welfarestatein termsof the loss of the law'sregulatas invades Accordingly, spheres well as ing ability. legislation private and other functionally differentiated autonomous sub-systems.In overloads law the doingso, the welfarestate'sinterventionist strategy its Thisoverload results to the pointof distorting regulatory function. in chaoticlegislation whichcomplicates legalsystemandprevents the and its rational this,Luhmann the reflexive Against self-reproduction. defendthe autopoietic law theorists of autonomy social sub-systems - particularly thoseconcerning and education, familylife. economy, Thus,theautopoietic supports paradigm deregulatory policies.55 To recall our discussion of Rorty, then, we may say that Luhmann, while he does evade pragmatism's "evasion of philosophy" and its reduction of complexity, does not evade a pervasive liberalism which, even more than in Rorty,takes the form of a technocraticfunctionalism that is content to operatewholly within the purview of what Lyotardhas called the "performativityprinciple" of "positivist pragmatism."56In these terms, John McGowan's recent critique of Rorty would surely apply to Luhmannas well. As McGowanputs it, "the importantthing to note is that the negative endorsementof change, of the ever continuing conversation"or, we should add, of the continualundoing of complexity and distributionof paradox in Luhmann'ssystem, "is dependent upon and presupposes a much more ositive version of the social world than the conversationalistsinhabit."7 It is here that Luhmann'scomplacent taking for grantedof western, capitalist,liberal society short-circuitsthe second political promise of his work: his rigorous theorization of the epistemological necessity of sociality as such, of the fact that the social is always virtual, partial, and perspectival, mutually constituted by observerswho can andmust expose the aporiasof each other'spositions.
54. See for example the ratherstunningandwillful naivet6of Luhmann'sdiscussion of the relationshipbetween politics and economics in his essay "The Representationof Society WithinSociety," 11-19; and see, for a typically reductiveglance at Marxisttheory, 17-18 of that same essay. 63. 55. Zolo, "Autopoiesis:Critiqueof a Postmodern Paradigm," 56. See Jean-Frangois Lyotard,"DispatchConcerningthe Confusion of Reasons," The Postmodern Explained: Correspondence1982-1985, trans. Don Barry et al., eds. JulianPefanis and MorganThomas,afterword Wlad Godzich (Minneapolis:U of Minby nesota P, 1992) 66. See also 123-24. For a particularly strikinginstanceof Rorty's technocratic functionalism,see his responseto CliffordGeertz'scritiqueof his ethnocentrismin 203-210 (see 9n above). Relativismand Truth, Objectivity, and 57. JohnMcGowan,Postmodernism Its Critics(Ithaca:Cornell UP, 1991) 198.


Pragmatics ofEpistemology

This shortcoming will be clearest, perhaps, if we compare Luhmann's work with Donna Haraway'sreinterpretationof the figure of "observation" and, more broadly, of vision in her recent work. Both Luhmann and Haraway attempt to retheorize the figure of vision by situating it, that is, by detranscendentalizingit and divorcing it from its representationalist associations. Luhmann would, I think, agree with Haraway's insistence on "the embodied nature of all vision" and her theoretical rejection of "a conquering gaze from nowhere" which claims "the power to see and not be seen, to representwhile escaping And similar to Luhmann's,Haraway's epistemologirepresentation."58 cal project is dedicated above all, to use her paraphraseof Althusser, to resisting "simplification in the last instance" (SK 196). But here, as Haraway's sense of "embodiment" the name for this theoretical fact needs to be distinguished from Luhmann'stheorization of the contingency of all observation. What Haraway wants is a concept of "situated knowledges" (SK 188) which emphasizes the physical and social positionality of the observer - not least of all, for Haraway, the observer's gender - in short, the specific conjuncture of qualities which mark the possibilities and limits of what the observer can see. In this sense, she writes, "objectivity turns out to be about particular and specific embodiment, and definitely not about the false vision promising transcendence of all limits and responsibility. The moral is simple: only partial perspective promises objective vision" (SK 190). In Haraway's articulation of observation and vision, "embodiment" names contingency, "objectivity"names political and ethical responsibility for one's observations, and both are "as hostile to various forms of relativism as to the most explicitly totalizing versions of claims to scientific authority"(SK 191). There can be little doubt that Harawaywould find in Luhmann'stheorization of observation, in his "unmarking"of it, we might say, through a relentless formalism and abstractionwhich socially and historically disembodies it, confirmationof her suspicions aboutrelativism. Luhmann would need to be told, as Harawayreminds us, that "social constructivismcannot be allowed to decay into the radiant emanations of cynicism" (SK 184). And indeed, Luhmannwould seem to invite this charge in many places in his work - both on the level of his theory
58. Donna Haraway,"Situated Knowledges:The Science Questionin Feminism and the Privilege of PartialPerspective,"Simians, Cyborgs,and Women:The Reinventionof Nature (New York:Routledge, 1991) 188. Hereafter cited parenthetically SK in the text. as

Cary Wolfe 125 and on the level of his rhetoricand tone. In Ecological Communication, for example, he writes: The problemseems to be that one has to recognizethe dominant seenas "capitalism" "functional or differsocialstructure whether a entiation" to assume position it. against ... A functional equivalent for the [nineteenth-century] theoretical construct"dialectics/ it is a revolution" not in sightandtherefore is notclearwhatfunction criticalself-observation societywithinsociety could fulfill. ... of Likethe "Reds" the "Greens" also lose coloras soonas they ... will officeandfindthemselves with assume confronted all theredtape.59 My guess is that Harawaywould justifiablydetect the levelling political at extrapolation the end of this passage from the epistemologicalclaims at its beginning as an instanceof that relativismwhich is, in her words, "a way of being nowhereand everywhereequally.The 'equality'of positionand ing is a denial of responsibility criticalenquiry.Relativismis the perfect mirrortwin of totalizationin the ideologies of objectivity;both deny the stakesin location,embodiment, partial and (SK perspective" 191). Luhmann's theory of observation doesn't sufficiently recognize the imperative of Haraway's"embodiedobjectivity":that "vision is always a question of the power to see" (SK 192). Again, a passage from Luhwill mann'sEcological Communication help make the point: that can of Investigations areinspired theoretically alwaysbe accused a lack of "practical reference." for Theydo not provide prescriptions to the that others use. .... Thisdoesnotexclude possibility serviceable in of resultscan be attained this way.But thenthe significance themethodof creating will alwaysremainthat a more controlled ory of ideas can increasethe probability more serviceableresults of above all, that it can reduce the probability creatinguseless excitement.60 The question that Haraway puts squarely on the table is never broached by Luhmann:"serviceable" whom? And in the absence of for addressingthat question - and of any detectable interest in addressing it - Luhmann's position seems ripe for interpellationinto Haraway's reading of systems theory in terms of the historically specific "management" strategies of post-WorldWarII liberal capitalist society, in which systems theory, like sociobiology, populationgenetics, ergonomics, and
59. 60. 126. Luhmann,Ecological Communication xviii. Luhmann, Ecological Communication


Pragmatics ofEpistemology

other field-models, is crucial to "the reproductionof capitalist social relations" in the specific era of "an engineering science of automated technological devices, in which the model of scientific intervention is
technical and 'systematic,' .

. [t]he nature of analysis is technological

functionalism, and ideological appeals are to alleviation of stress and other signs of humanobsolescence."61 We need not agree with Haraway,I think, that the systems theory paradigm always already carries with it a tacit endorsementof liberal capitalist society in its post-WorldWarII incarnation.Indeed, Maturana and Varela have drawn very different ethical and political implications from very similar epistemologicalpremises.62But it seems clear that in Luhmann'shands, the systems theory paradigmdoes in fact indulge the same sort of blithe liberal functionalism embraced by Rorty in its refusal to confront the uneven and asymmetricalrelations of power especially economic power - which undeniably constrain and indeed often renderutterly beside the point the undoing of complexity and the distribution of paradox which in Luhmann's thought remain too squarely within a political if not philosophical idealism. If Rorty sanitizes the social field by limiting conversationto the liberal ethnos, Luhmann levels it by refusing to complicate his epistemological pluralism - that we are all alike in the formalhomology of our observationaldifferences - with an account of how in the real social world where those observations take place some observers enjoy more resources of observation than others. The complexifying and open-ended imperative of Luhmann's theory is, following George Spencer Brown, "distinguish!"
61. Donna Haraway,"TheBiological Enterprise: Sex, Mind, andProfitfrom Human 44. Engineeringto Sociobiology,"Simians,Cyborgs,and Women 62. See the last chapterof Maturana Varela,The Treeof Knowledge, esp. 245and 46, for a discussion of the ethical andpolitical implicationsof theirepistemology,

Cary Wolfe 127 and "observe!"but we must still subject that imperativeto the critique levelled by Steven Best and Douglas Kellner at the metaphorof a cultural conversation of diversity and plurality as it is deployed by Rorty, "that some people and groups are in far better positions - politically, economically, and psychologically - to speak," or to observe, we might add, "than others. Such calls are vapid when the field of discourse is controlled and monopolized by the dominant economic and political powers."63 We might say, then, that Luhmann blind spot, his unobservablecon's stitutive distinction, is his unspoken distinction between "differentiation" and what historicist, materialist critique has theorized as a "contradiction," blind spot which manifestsitself in Luhmann'sinabilor unwillingness to adequatelytheorize the discrepancybetween the ity formal equivalence of observersin his epistemology and their real lack of equivalence on the material,economic plane. It seems that the category of contradiction,insofar as it names this difference, proves more difficult to dispose of than Luhmann'ssystems theory imagines.64 Or rather, it is disposed of by systems theory, but only "abstractly,"as Marxists theorists like to say, only in thought, but not in historical, material practice. What Luhmann'sepistemological idealism refuses to confront is that the differentiation,autonomy,and undoing of complexity it imagines remains muffled and masteredby the economic context of identity and the rule of exchange value within which systems theory historically arises. And in that refusal, in its pragmaticeffect of socialyreproducing the liberal status quo, it is clear that there are powerful ideological reasons, as well as epistemological ones, why one cannot see what one cannot see.

63. Best and Kellner,PostmodernTheory288. 64. As Eva Knodt has suggested in response to an earlier draft of this essay, this point in turn raises the most far-reachingquestion of all: whether or not postmodern society is adequately described as primarilyfunctionally differentiatedor stratified and hierarchical. In this connection, we might - to stay within the Marxist paradigm invoke Raymond Williams's famous revision of the base/superstructure model to say that the paradigmof functionaldifferentiationis emergent,even though it might be more pervasive socially, within a system in which dialectical contradictionremains dominant in the form of the asymmetrical importanceof the economic system. See Williams's "Base and Superstructurein Marxist Theory," Problems in Materialism and Culture (London: New Left, 1980).