Evocative Representation

Constructivist Normativity

Just as the norms of describing are internal to describing itselfare part of the language games of descriptionsso are the norms of representing. Inasmuch as we think of political representation as an action and process (Saward, 2006), it follows from the previous line of reasoning that it is guided by internal norms of correct functioning. Despite the general tendency of the literature to look upon representation as a process, what has been termed the constructivist turn in political representation has had a hard time also identifying norms internal to the process of representation. In other words, the constructed nature of the representative process has left little room for grounding what might count as good representation.

To illustrate the difficulty that the literature has so far encountered in defining representation's norms, I will discuss several ways of characterizing representation, each with its own view of what is normative for this activity. I will begin with Saward's representative claims and its elaboration in the 'shape-shifting' representative. I will then look at the idea of responsiveness as the yardstick of normativity and at Lisa Disch's mobilization conception of political representation. Lastly, I will discuss Disch's (2015) further contribution in what she calls the 'Constructivist Turn'. There, she asks whether epistemological commitments to constructivism are incompatible with political commitments to democracy, a question that can be interpreted in light of this discussion as a search for the (in)compatibility between epistemological constructivism and normative foundationalism.

After discussing these contributions in turn, I will argue that there is another way to see the norm of political representation, namely through when and why we tend to take representative claims as good, and when and why we tend to take them as wrong. We can then advance a foundational norm, discovered as it were in the operations of representation itself, what I call evocative representation.

The claim-making framework in representation theory, championed by Michael Saward (Saward, 2006), has successfully realigned theoretical interest in representation with linguistic and processual considerations. Instead of focusing on the various forms of political representationsomething that theorists had been doing for a long time, and in a sense had taken for granted as a dominant mode of theoretical inquiry -Saward proposed to think of the process of representation as akin to the process of presenting a claim. In subsequent elaborations of this basic idea, Saward (for example, 2014) argued that it is not the forms of representation that define its activity, but rather the general outlines of claim presentation. In fact, representatives are shape-shifters, meaning that they are free to change representative forms in order to fit the various circumstances of representing. This theatrical focus counters the tendency of "contemporary normative frames [to] stress singular and consistent roles of political actors or leaders" (2014, p.724). Said differently, the repertoire of representation is wider than "normative prejudgment" (p.725) would allow, and therefore stretches through and beyond the classical roles of trustee, delegate, and so on. In the terminology advanced through the discussion of Cavell, saying that representatives ought to act as trustees or delegates is prescriptive, not normative. It would be normative to say that when representatives do act as representatives, they do in fact act as trustees, delegates, and so on. Saward's insight is that when representatives act as representatives, they do in fact act through many different roles. In other words, what it is to represent, and what it is to represent well, cannot be given in one form over all others. Their simultaneous availability takes us some way toward what is normative for representation by recognizing its inherent multiplicity.

Saward's account of representation as centered around the presentation and contestation of claims unsettles two important aspects of how political representation had previously been theorized: it devalues the primacy of formal representative types, while questioning what the point of representation is. A representative under Saward's view is free to shape-shift as she sees fit, or according to the relevant situation, and in doing so she creates a persona for herself that is able to carry the representative work without identifying with a particular role. In other wordsand this is the constructivist foundation of his accountthe representative claim is first and foremost a contingent relation between parties, and its primary function is to bring these parties into being (the represented as much as the representative), and not just to convey interests and preferences from one being to another. The representative relation is therefore engaged in the creation of subjectpositions (Saward 2014; also see Tanasescu 2019 for a further elaboration of this concept).

This ideathat constituency interests are secondary to subject creationis further developed by Lisa Disch (2012b) in terms of "constitutive effects", and elsewhere in terms of a "mobilization conception of representation" (2011). She argues that representation "neither simply reflects nor transmits demands; it creates them as it actively recruits constituencies". This makes it clear that 'preferences' and 'interests' are products of the representative activity, or ways of speaking about representation that are in a sense derivative. Disch's (2011) mobilization conception of representation works with the same constructivist assumptions as Saward, and presents a picture of representative acts as being primarily about the solidification of identities around the making and unmaking of claims. This way of thinking about representation offers no bedrock that we can use to judge whether representation is done well or not. For example, the idea of weighing representation in terms of the responsiveness of the representative to the represented (for example Celis, 2012) cannot be held consistently with the idea that the subjects of representation and their interests are constituted through the representative claim. Indeed, Disch rejects the intelligibility of responsiveness and instead proposes reflexivity as "the normative standard for evaluating political representation" (Disch 2011, p.111). She characterizes this as a systemic quality that ensures that representation "can be judged as more or less democratic insofar as it does more or less to mobilize both express and implicit objections from the represented". This would mean that, insofar as a system is open enough to allow for contestation of, and debate around, representative claims, then we can, on light normative grounds, call it, at least, democratic. This kind of conclusion is forced upon constructivist accounts of representation because of their inherent commitment to foundationless norms, but it does not quite suffice. After all, contemporary politics is increasingly populated with regimes that feign all of the formal signs of representation, including reflexivity and open contestation, without them also being, on account of that, either democratic, or representative.

Recalling the earlier discussion of normativity, I want to suggest that reflexivity is not particularly normative. In other words, when someone asks whether a claim is representative, pointing to the openness of the system that produced the claim cannot provide an answer. It can provide some measure of legitimation, but there is no assurance that a representative claim will be a good one simply because it is more or less democratic. The fact of participation as such does not have any kind of necessary relation to the quality of representative claims. So although Disch offers a very helpful conceptualization of what the activity of representation consists of, that conceptualization is not enough for telling us what it is to represent well. Or rather, recalling the relation between norms and descriptions, what it is to represent well is buried in her account and not given its proper due.

Saward's account seems to remain agnostic as to representation's norm. The conceptualization that he offers of the activity of representation forces him to say that a claim is representative if it is accepted as such by a relevant audience. This is, in one sense, rightacceptance by the represented surely must weigh heavily on whether a claim is in fact representative. What is left out, and what points toward the need for further elaboration in this area, is the difference between being and feeling represented. Cavell pointed out that what is characteristic of actions is the possibility of their going astray. In the action of representing, what would it mean to both have audience acceptance and go astray? In other words, what would it be to feel represented, without being so?

The constructivist turn in representation theory has been very successful in showing that both the represented and the representative are constituted through the process of representation.

Commenting on Derrida, Disch (2011) shows that representation had previously fallen prey to the myth of primordial presence, namely the idea that the subjectivities involved in a representative process are ontologically primary. The subject-positions that arise from the representative process are not pure ontological givens. However, they are also not free to be whatever they wish. In other words, though representation, in advancing claims, creates constituencies and representatives, it cannot do so contingently; the representative cannot say anything and have it count as representative, or have it engender constituencies. What can be missed by focusing on the constructed nature of the representative relation is that the judgments that we do make about representative claims rely on foundational norms embedded in the ways in which we speak when representing. The constitution of subject-positions is bound by the grammar of the language in which they are articulated.

The importance of analyzing the act of speaking in order to analyze the process of representation has been amply recognized. In a further contribution, Disch (2012a) discusses the importance of speaking and naming in political representation by building on a productive tension between Pitkin and Laclau. She argues (2011, 2012a) that Laclau offers a concept of representation that is itself ontologically primary, and therefore develops representation genuinely free from the dominance of interests, whereas Pitkin remains wedded to a supposed ontological primacy of interests themselves.

For Laclau, representation is done in the act of naming a people. Under this account, the norm of representation-as-naming is what the act makes possible. The act is well done when it allows for the constitution of a people that does not kill difference in the name of its own unity, but rather incorporates it. So representation is well done when a people is created through the process of naming that is at the same time unitary and heterogenous. But this norm is not internal to the act of speaking itself: there is nothing that the representative can say that will ensure the survival of heterogeneity. In fact, the norm here is heterogeneity despite unity, which (like reflexivity before) seems to be a democratic prescription (and therefore an external principle) rather than a description of what we do when we speak well as representatives.

To sum up: thinking of representation as a process through which subject-positions are created and interact reveals the primacy of speaking in understanding what goes on when some are represented by others. But by the same token we are no longer able to specify what is normative of representation and are forced to, instead, prescribe democratic rules of functioningv. The constructivist turn in representation theory has argued convincingly that political representation is primarily about relations, and that the relations between subjectivities that it inaugurates are at the same time the inauguration of subjectivities themselves. It is salutary to be freed from the primacy of interests and the myth of presence, but there is also something intrinsic to being a subject that is not adequately taken into account. We do not speak of subjects and objects in the same wayobjectifying is, after all, a grave insult to the subjectand this is so because in subjectifying we are also engaging an intrinsic norm that separates subjects from everything else. Indeed, subjectifying previously objectified entities is a staple of progressive representation. Constructivism is right to insist that many previously discarded beings can be subjects (for example Tanasescu 2015Tanasescu , 2019, but in speaking about something as a subject and being recognized as speaking well, we are bound to discover what is normative of representative claims. I proposed therefore to think about what is normative of representation in foundationalist terms, because we can then both uphold the epistemological primacy of representative claims and say something about its internal, ontologically grounded, norms. I now wish to briefly turn to a way of discussing the act of speaking that can give us useful conceptual tools for identifying the internal norms of representative speech.

Speaking and speaking well

In Tanasescu (2019), I developed an argument, based on J. L. Austin's theory of language and Judith Butler's use of it in social theory, to demonstrate that a representative claim is best thought of as an illocutionary speech act disguised as a constative. In short, the argument is that representative speech is most productively understood as inaugurating subjects (therefore as illocutionary, bringing something into being), though its very form of presentation is constative, that is to say, it claims to have discovered that which it creates. "I now pronounce you husband and wife" is an illocutionary, because it is the utterance itself that brings the state of affairsmarriageinto being. On the other hand, "the desk is wet" is a constative, because whether this utterance is happy or unhappy, to use Austin's terms, is not dependent on what it brings about, but is amenable to factual checking: the utterance is happy if indeed the desk is wet.

Because the illocutionary is not tied to truth claims in the same way as the constative, its happiness depends on a whole array of contextual factors that must work together. As Judith Butler (1997) demonstrates, social realities and relations are constantly negotiated through the performative use of language, which is instrumental to their genesis. A whole cultural apparatus is involved in the maintenance of particular kinds of social relations, and this can work in all directions. In other words, there is a whole culture of hate that gives hate speech its capacity to literally hurt its victims, as much as there is a culture of empathy that inaugurates expansive subjects.

Understanding representative claims as illocutionaries disguised as constatives allows for a more fine-grained interrogation of the operations of subject formation involved in representation and their relation to what is normative of this activity. Claiming to represent, whether human or non-human subjects, always involves the inauguration of subjectivities that have the general outlines specified by the claim itself. For example, if animal representatives speak of golden jackals as providing ecosystem services in the form of carrion disposal, this inaugurates the subject-position "golden jackal" (as much as the subject position "jackal representative") such that it appears that carrion disposal is part and parcel of being a jackal. The illocutionary claim, in other words, appears to be a discovery of a latent feature of jackal sensibility, and therefore the constructivist operations discussed earlier are successfully effaced.

The effacement involved in representative speech works both to mystify the process, and to give force to the illocutionary. The weight of empirical discovery is transmitted to what is essentially a process of invention. But, as with representative claims earlier, every claim is not equally amenable to credible presentation. One could also claim that the jackal is a thief, and therefore inaugurate a particular set of subject-positions implying particular kinds of relations, but one could not as credibly claim the jackal as a subject concerned with global climate change. The success of the transition from illocutionary to constative is in part given by the cultural resources available to give life to the created/discovered subject. The ideas of ecosystem services and the cultural imagination of jackal mischievousness give vital resources to achieve the happiness of illocutionary summoning into being. Thinking the jackal as concerned with global climate change is not as credible because we currently lack the adjacent cultural resources that would give it specific weight. This, however, is always amenable to change.

There is an internal relation between the illocutionary and the constative as used here that can be described by the play between the cultural and natural affordances surrounding a representative claim. Representative speech can only be thought of as sliding between illocutionary creation and constative discovery inasmuch as there are a number of symbols, practices, cultural truths, that conspire to make subjects seem given. Similarly, a particular subjective configuration is intrinsically related to the natural affordances (Descola 2014) that limit, to some extent, what can be said. One cannot claim anything at all, but one can always claim more than currently possible. Another way of expressing this askance relation is to say that representative claims function in the ways described here inasmuch as they rely on a common capacity among the subjects actively participating in representative processes to fashion subjectivities. This process of creation is intrinsically related to what is called the moral, or sympathetic, imagination, to which I will shortly turn.

To sum up: the representative claim can be understood as an illocutionary disguised as a constative.

Representative claims are illocutionary claims that are happy or unhappy depending on whether or not they achieve their desired effectbringing into existence certain subject-positions. Within this operation, there are norms, understood in the Cavellian way already discussed, that make a claim count as a good one. A good representative claim would then be one that successfully brings into existence the intended subject-positions. However, bringing subjects into beingwhat I call evoking themis an operation that, in virtue of the working of language and the concept of subject itself, relies on the moral imagination of both representative and represented. It is through the moral imagination that the natural affordances of the world come into play with cultural affordances such that new subjects can evolve. I now turn to completing the argument by spelling out the ways in which evocative representation is tied to the moral imagination, and what this means for representing well.