代表概念的反思(Baker)

Parliamentary Affairs Vol. 59 No. 1 © The Author [2005]. Published by Oxford University Press on behalf of the Hansard Society for Parliamentary Government; all rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org doi: 10.1093/pa/gsj002 Parliamentary Affairs Vol. 59 No. 1, 2006, 155–172 Advance Access Publication 25 October 2005 Revisiting the Concept of Representation1 BY GIDEON BAKER IN an article on dealing with difference, Anne Phillips observes that ‘lacking any half-credible basis for seeing citizens as united in their goals, theorists of liberal democracy took issue with the homogenizing presumptions of a common good or common purpose, and made diversity a central organizing theme’.2 What would follow, however, if this prob- lem attached itself not just to the idea of the body of citizens, but to the body of the citizen herself? For, along with a growing awareness of dif- ference comes the possibility that, just as we can no longer talk of the polity as a singular subject with a common purpose, we might equally lack any reasonable grounds for seeing citizens themselves as homoge- nous, unified subjects capable of autonomy. If such is the case, how- ever, then there appear to be serious implications for our understanding of political representation, informed as this must be by notions of the subject that is supposed to be representable in some way. Before we can even consider such key questions as whether representation is norma- tively defensible, how to effect representation, or the representativeness of our representatives, we need first to look, then, at whether subjects can be considered capable of representation at all. In the light of developments in theories of the subject over recent decades, this question is no longer—if indeed it ever was—a simple one. The various attempts to deconstruct, displace or decentre the subject appear at first sight to radically undermine the coherent, stable agency necessary to both seekers and providers of political representation, who, on most liberal democratic readings, must claim certain fixed interests or identities, or to be able to represent these, over space and time. For some, indeed, the implications of these developments in theory are nothing short of terminal for representation: Along with the modern individual subject, what vanishes into the discursive condition [of postmodernity] is the entire humanist apparatus ... of common denominators upon which so much has depended, including representational politics.3 In short, there now appear to be all sorts of questions surrounding the very possibility of political representation. This article seeks to illuminate these questions further—to get the measure of the challenge that they pose—first, by summarising developments in theories of the subject and by speculating on the likely implications of these for conventional 156 Parliamentary Affairs accounts of representation. Second, it interrogates recent responses in democratic theory to these developments, questioning whether they are adequate to the task of reconciling new conceptions of subjectivity with representational politics. Third, and going into more detail on new the- ories of the subject, it explores issues surrounding the construction of subjects and the difficulties that these pose for conceiving of representa- tive politics in the conventional manner. Finally, it looks at ways in which this new subject itself might actually be resistant to representa- tional politics. Throughout the discussion Michel Foucault’s writings provide the main point of departure given their extraordinary influence on recent attempts at rethinking the subject. Developments in theories of the subject Nietzsche, speculating on the future of lawgiving, echoes Rousseau in looking to submission ‘only to the law which I myself have given, in great things and small’.4 For Nietzsche, however, solutions—such as Rousseau’s General Will—to the problem that this then creates for con- ceiving of a link between individual and collective autonomy can no longer stand in the face of the radically individualistic autonomy which he diagnoses as the only available to modernity after the death of God: What will not any more henceforth, and cannot be built any more, is—a society in the old sense of that word; to build that everything is lacking, above all the material. All of us are no longer material for society; this is a truth for which the time has come.5 The implication of this truth, for Nietzsche, is that the only escape from nihilism involves the individual becoming the authority for himself, which is, at once, to become responsible for himself.6 Prima facie, the possibility of political representation being compatible with autonomy would be excluded here. More recently, Michel Foucault, following Nietzsche, has famously subverted humanist ethics with the anti-humanist claim that there is no essential truth of what we are. Through genealogical ‘analysis which can account for the constitution of the subject within a historical framework’,7 Foucault too sought to disrupt the idea of the ‘constituent subject’: We believe that feelings are immutable, but every sentiment, particularly the noblest and most disinterested, has a history. We believe in the dull constancy of instinctual life and imagine that it continues to exert its force indiscrimi- nately in the present as it did in the past. But a knowledge of history easily dis- integrates this unity, depict[ing] its wavering course ... We believe, in any event, that the body obeys the exclusive laws of physiology and that it escapes the influence of history, but this too is false. The body is moulded by a great many distinct regimes ... Nothing in man—not even his body—is sufficiently stable to serve as the basis for self-recognition or for understanding other men.8 To radically historicise the subject thus, which is effectively the work of genealogy from Nietzsche to Foucault, is at once to severely limit its Revisiting the Concept of Representation 157 capacity for the sort of ‘objective’ knowledge of self and others that appears to be central to representation. For Ermarth, it was modernity’s ‘immense, historically developed capacity to keep the world in mind’ that was most essential to humanism, and this is a capacity that ‘is certainly the one essential to representational conventions from art to pol- itics’ also.9 So what happens to representation if this transcendental capacity is now lost to us? Ermarth’s critique here is a typically post- modern refutation of a wider ‘Culture of Representation’, but political representation is also brought under this pejorative rubric: We have been rethinking the culture of representation, its obsession with power and knowledge, its constraint of language to primarily symbolic rather than poetic function, its categorical and dualistic modes of definition, its belief in the quantitative and objective, neutral time and objectified world, its irreducible subject ... [This] raises some discomfiting but unavoidable questions: not just about the ‘individual’ subject but about the future viability of democratic insti- tutions.10 The idea of a loss of transcendence relates closely to Foucault’s claim that the subject is ‘moulded by a great many distinct regimes’. We are reminded here that, as a general theory, he views the subject not at all as a product if its own doing, but rather as an effect of discourses of power. A number of troubling implications for representative practices follow from this, however, implications that will merely be flagged up here, for further exploration later. First, on the basis of an analysis of the ubiquitous web of power relations that always precede, and consti- tute, the subject, Foucault explicitly rejects the classic, juridical account of social contract theory within which liberal democratic representation is traditionally understood. Here, power is taken to be a right, which one is able to possess like a commodity, and which one can in consequence transfer or alienate, either wholly or partially, through a legal act or through some act that establishes a right, such as takes place through cession or contract. Power is that concrete power which every individual holds, and whose partial or total cession enables political power or sovereignty to be established.11 But of course, in undermining the humanist discourse of authenticity in relation to the subject upon which this juridical model of power depends, Foucault also undermines the idea of our freely consenting to surrender power to those who would represent us. In its place he substi- tutes a politics of humanism, understood as containing a ‘logic of legis- lation’—a discourse of truth that legitimates the authority of the truth teller, or expert, ‘to legislate norms for us’.12 For what else, other than a struggle of power, could be at work when representatives, along with others who would legislate on our behalf, are unable to rely on some essential truth about the human condition. As with Nietzsche, truth and power become intertwined in such a way as to steal authority away from ‘truth tellers’ generally. Thus just as Nietzsche disavows any 158 Parliamentary Affairs relation between individual and collective autonomy, so too for Foucault this connection is now highly problematic: [T]he idea of a consensual politics may indeed at a given moment serve either as a regulatory principle, or better yet as a critical principle with respect to other political forms; but I do not believe that that liquidates the problem of the power relation ... perhaps one must not be for consensuality, but one must be against nonconsensuality.13 Second, as Judith Butler points out, if a subject position is no more than a site of converging power relations, then no subject is ‘self-identical’ (a point also made by Foucault14), which is to say that no subject ‘houses or bears these relations’ since ‘this converging and interarticulation is the contemporary fate of the subject’.15 In Foucault’s words, this rather weak identity, which we attempt to support and to unify under a mask, is in itself only a parody: it is plural; countless spirits dispute its posses- sion; numerous systems intersect and compete. The study of history makes one ‘happy, unlike the metaphysicians, to possess in oneself not an immortal soul but many mortal ones’. And in each of these souls, history will not discover a forgotten identity, eager to be reborn, but a complex system of distinct and multiple elements, unable to be mastered by the powers of synthesis ...16 From a Foucauldian perspective, then, to attempt representation under these conditions could be to threaten violence against subjects by freezing their identities and foreclosing possible change. Only one element of that subject’s otherwise pluralistic identity ends up being represented at any one time, an artificially homogenous representation of a heteroge- neous subjectivity. The other is effectively being characterised as simply too ‘other’ to be representable. Recall the latter part of the statement: ‘Nothing in man—not even his body—is sufficiently stable to serve as the basis for self-recognition or for understanding other men’. Before we can adequately unpack a Foucauldian view of specifically political representation, however, we need to consider Foucault’s deeper analysis of the act of representation itself, as undertaken in The Order of Things. In this work, Foucault suggests that the possibility or represen- tation rendering beings ‘as visible in their truth’ is lost to us along with the ‘Classical’ age (from roughly the mid sixteenth to late eighteenth centuries) within which this system of knowledge made sense. In Foucault’s Classical era we find a language of knowledge that represents in the sense of naming, patterning, combining, connecting and discon- necting things as it gives them form in words. Language works as a ‘translucent’ medium through which representation must pass in order to render beings visible ‘in their truth’. ‘The possibility of knowing things and their order passes ... through the sovereignty of words’.17 Foucault goes on to argue that as long as this language was spoken in Western culture, it was not possible for human existence to be called into question on its own account, since it was precisely the human subject Revisiting the Concept of Representation 159 that contained the nexus of representation and being. The discourse that provided the connection between the ‘I think’ and ‘I am’ of the being undertaking it constituted the fundamental nature of Classical language, ‘for what was being linked together in it was representation and being’.18 The classical subject therefore lacks the perspective neces- sary to see that ‘he who ties together all the interlacing threads of the representation in the form of a picture or table ... he is never to be found in that table himself’.19 When the Classical coherence between the theory of representation and the theories of language gave way in the late eighteenth century, therefore, it had the most profound consequences: the theory of representation disappears as the universal foundation of all pos- sible orders; language as the spontaneous tabula, the primary grid of things, as a indispensable link between representation and things, is eclipsed in its turn; a profound historicity penetrates into the heart of things ... [A]bove all, language loses its privileged position and becomes, in its turn, a historical form coherent with the density of its own past ... But as things become increasingly reflexive, seeking the principle of their intelligibility only in their own development, and abandoning the space of representation, man enters in his turn, and, for the first time, the field of Western knowledge.20 The key thought here is that, in the modern era and consequent upon the dissolution of classical discourse and its replacement by pluralised language, what comes to light is human historicity—‘the fact that man as such is exposed to the event’. ‘Man ... was constituted only when language, having been situated within representation and, as it were, dissolved in it, freed itself from that situation at the cost of its own frag- mentation: man composed his own figure in the interstices of that frag- mented language’.21 But since the human being is now thoroughly historical himself, as subject to ‘the movement of History’ as anything else, he can no longer remain stable.22 Foucault’s point in application to our concern with acts of political representation is then that any act of representation is today as unstable as the subjects attempting it now realise themselves to be. Moderns can do no more than to interpret their experience as ‘an application of the forms of language to the human order’. The human subject who repre- sents is now visible to us as an interpreter rather than invisible to us as a knower. With the visibility of the subject comes the invisibility of (objective) representation. Symbolic representation At exactly the same time as Foucault was confining representation to the past as an artefact of a Classical form of knowledge, Hannah Pitkin, in The Concept of Representation, was arguing that, to the contrary, we can clarify what representation is as a timeless concept. Given the influence of Pitkin’s discussion on the study of political representation, 160 Parliamentary Affairs it is useful to juxtapose her analytical concept of representation with Foucault’s historicised practice of representation. What this juxtaposi- tion reveals is that it is only a historical account of representation that allows the subject of representation to come fully into view. Pitkin, for all the sophistication of her analysis, dehistoricises representation and so makes the mistake—from a Foucauldian perspective—of assuming a universal subject for political representation. Pitkin is well aware that the practice of representation involves making present what does not in fact exist. Indeed she illuminates, amongst other facets, a symbolic, or ‘standing for’, side to representation. This part of representation, she argues, means to make present something that is not in fact present; ‘symbols are not proxy for their objects, but are vehicles for the conception of what they symbolise’:23 Because of the element of arbitrariness in most if not all symbols, because their connection with what they represent is not based on resemblance ..., the only criterion of what constitutes a symbol is in people’s attitudes and beliefs. If we ask what it is that makes symbols out of anything ... the answer is clearly: the beliefs, attitudes and assumptions of people.24 Yet Pitkin is critical of models of representation that take this symbolic form as definitive of representation in toto, arguing that when this move is made, ‘the political representative will then seem to have only arbitrary, conventional, or hidden connections with those he represents. Representing people will seem no different from symbolizing an abstraction like the nation’.25 The idea that the human subject might come to be seen as an abstraction itself is not anticipated by Pitkin, and this is why her argument against seeing symbolic representation as the whole of political representation seems less compelling now than when she originally made it. Pitkin’s argument against a symbolic-only understanding of represen- tation is also weak in the face of the Foucauldian reading for being pri- marily normative—symbolic-only representation will reduce political representation from an activity which we might offer a critique of to a mere state of affairs to be described. For the sense of representation as acting for others implies that those others might be more or less objec- tively represented, while to simply ‘stand for’ them admits only of sub- jective analysis—‘so long as people accept or believe, the political leader represents them, by definition’.26 The problem with Pitkin’s argument here is twofold. First, in light of the decentred subject, we might now say that just because a representa- tive ‘stands for’ the represented, rather than ‘objectively’ representing them, does not imply that representation is ‘merely in the mind of the governed’.27 This way of making the objection implies that, while the represented can be manipulated into seeing representativness, those that represent are knowing subjects able to work ‘on the minds of those who Revisiting the Concept of Representation 161 are to be represented’.28 This division between knowing and ignorant subjects now appears unstable. Second, and more significantly, the primarily normative critique of symbolic representation marks a subtle but decisive shift away from Pitkin’s stated goal of providing a purely conceptual account of representation. To put the point bluntly: just because we might reject a symbolic under- standing of representation as politically undesirable does not mean that we can necessarily escape from it. Representation and differences between subjects Having demonstrated that even one of the most sophisticated accounts of political representation to date has been relatively blind to the subject of representation, let us turn next to the field of democratic theory gen- erally. We may find that democratic theory, taken more broadly, has already assimilated a more critical approach to the subject of represent- ative politics. The positing of irreducible differences between subjects is now a staple of democratic theory, though it has long been a central feature of liberal thought. The rise to prominence in recent years of deliberative models of democracy is pertinent here, where these models, working with a post- metaphysical concept of the subject and accepting the ‘fact of pluralism’, variously highlight the beneficial transforma