The 1990s saw an increasing acceptance that fair political representation was about more than the “politics of ideas” and had to incorporate a “politics of presence,” which would see legislatures become more descriptively representative of the populations from which they were drawn (Phillips Reference Phillips1995). A large part of the case in favor of greater diversity in the composition of political assemblies leaned on notions of legitimacy and democratic equality, arguing that “[d]escriptive representation matters because of what it symbolizes to us in terms of citizenship and inclusion—what it conveys to us about who does and does not count as a full member of society” (Phillips Reference Phillips2012, 517). Simultaneously, arguments were made that increasing the descriptive representation of traditionally underrepresented groups (i.e., mere presence) would likely result in the improved substantive representation of those same groups in terms of legislative outcomes. This relationship has been the focus of much research, with most empirical work reaching something of a consensus that the two are linked in a probabilistic but not a deterministic way (Celis et al. Reference Celis, Childs, Kantola and Krook2008). As Jane Mansbridge (Reference Mansbridge2015, 261) writes, “the evidence now seems clear that, once you look beyond roll call votes, Black legislators do a better job than Whites of representing Black constituents, women legislators do a better job than men of representing female constituents, and legislators from working-class backgrounds do a better job than others of representing working-class constituents.” Given the fact that such a link seems to map onto even the coarsely grained characteristic variables used to identify politically relevant groups in much empirical analysis, it appears hard to get away from the fact that something is shared, working, or transmitted at a greater rate within what are designated as in-group relationships than outside of them, and that this thing seems to underpin any link between descriptive and substantive representation.
What might this “something” be? A complex debate has developed over the past three decades that has tended to cluster in particular around the notion of shared experience, which serves as the bedrock of the various possible explanations. Yet, despite appearing to suggest that the link between descriptive and substantive representation is underpinned by something like shared experience, many writing on the subject stop short of explicitly stating that shared experience is the foundation of that apparent link. Indeed, there is almost a squeamishness about doing so. Anne Phillips (Reference Phillips1995, 93) exemplifies this view, writing that “[w]e do not see political views as following in some automatic way from the bare facts of experience, and, apart from anything else, we would question which particular experience was supposed to be shared.” This is a valid argument—theoretically, we want to avoid deterministic thinking, and, strategically, we want to avoid constraining representatives from previously underrepresented groups from acting counter to any expected norm (such as those based on gender or racial stereotypes).
In this article, though, I make the case that scholars of representation should be less reluctant to embrace the potential conceptual and empirical benefits of framing representation through the lens of experience. Having outlined the empirical evidence on the link between descriptive and substantive representation and sketched the debate on what might underpin it, I flesh out my claim in four steps.
First, I argue that, rather than thinking of “experience” in an unspecified way, we should be more concrete and in this context think of “experience” in terms of the epistemic content of, and epistemic capacities gained through, certain kinds of subjective experience—experience with a first-person character. Second, I draw on extensive philosophical debate to show how subjective experience can be seen to bestow individuals with knowledge and capacities that both permit and prompt them to think about the world in new ways, and that these benefits are accessible only through subjective experience. That is, there is something distinctive about learning something from the first-person point of view.
Third, I demonstrate how we can reframe the idea that experiences might be shared within groups, and how we can clarify what we mean by “group,” in such a way as to ameliorate concerns that that the notion is inherently essentialist or theoretically incoherent. Largely, this reframing consists in avoiding a characteristic-focused ontology of groups and instead thinking in terms of experiential proximity of various kinds, giving way to arguments around the sharing of experience that are probabilistic, not deterministic, and that can account for the intersection of different relevant experiences.
Fourth, I summarize the argument and outline my overall claim in the context of the foundational literature on representation, drawing out its especially political relevance and implications. Roughly, the argument is that when we talk about diversifying legislatures by increasing the descriptive representation of historically marginalized groups (or indeed any group), a key part of what we are talking about is bringing into political institutions the subjectively derived knowledge of experiences disproportionately undergone by those groups. In other words, one reason to care about the political presence of members of those groups is that experiences attach to persons, and certain kinds of experiences will either predominantly or only attach to certain kinds of persons. Consequently, I contend that this argument has the strategic implication that scholars of representation should be unafraid to argue in favor of political presence on the basis of experience and the empirical implication that future research on political representation should attend more closely to the role of subjective experience when seeking to understand why descriptive and substantive representation appear to be linked.
THE LINK BETWEEN DESCRIPTIVE AND SUBSTANTIVE REPRESENTATION
The debate regarding what link, if any, exists between descriptive and substantive representation largely stems from Hanna Pitkin's (Reference Pitkin1967) influential account, in which she sees substantive representation (“acting for”) as the most authentic kind of representation and the kind most worthy of the label, contrasting it with descriptive representation (“standing for”). Descriptive representation has traditionally been seen as a “correspondence between the characteristics of the representative and the represented” (Celis et al. Reference Celis, Childs, Kantola and Krook2008, 100). Suzanne Dovi (Reference Dovi2007, 27) summarizes: “Descriptive representatives are representatives who look like, or at least have experiences and interests similar to, the people they represent.” There are two broad questions that comprise the debate about the descriptive-substantive link. First, and perhaps more prominent, owing to its ability to be tested empirically, is the question of when this link seems to come to fruition? The second is what is the basis of this link?
Addressing the first question, there is increasing acceptance, based on a growing weight of empirical evidence, that it does matter who our representatives are: that representatives drawn from a given group appear to be “better” in some sense at representing the interests of that group than those drawn from outside of it. Most prominently, research has established this pattern in the cases of women and ethnic minorities, both historically excluded from political institutions in a variety of contexts, making any subsequent shifts especially clear. As Maria Sobolewska and colleagues (Reference Sobolewska, McKee and Campbell2018, 1) note, “The claim that, in certain circumstances, descriptive representation is linked to the substantive representation of historically excluded groups has been subject to empirical testing and is generally supported.” This empirical evidence is generated from contexts as varied as the United States (Dittmar, Sanbonmatsu, and Carroll Reference Dittmar, Sanbonmatsu and Carroll2018; Swers Reference Swers2002), United Kingdom (Catalano Reference Catalano2009; Childs Reference Childs2004; Gains and Lowndes Reference Gains and Lowndes2018), Hong Kong (Tam Reference Tam2017), and Sweden (Wängnerud Reference Wängnerud2000). Cross-national research suggests a relationship between women's presence and the representation of women-friendly policy preferences (Espírito-Santo, Freire, and Serra-Silva Reference Espírito-Santo, Freire and Serra-Silva2020). Outside of observational evidence, David Broockman (Reference Broockman2013) uses a field experiment to find that black politicians in the United States are more “intrinsically motivated” to represent the interests of black voters than white politicians, even when political incentives to do so are reduced or absent. The descriptive-substantive link also seems to hold for characteristics such as educational background (Bovens and Wille Reference Bovens and Wille2017), social class (Carnes Reference Carnes2013), and even occupational experience (O'Grady Reference O'Grady2019). Based on a growing body of evidence, there is reason to think that a link between descriptive and substantive representation exists across a wide range of politically relevant groups.Footnote 1
Turning to the second question—what is the basis of this link?—there is less consensus. Or, at least, less explicit consensus. The general thrust of the predominant response to this question of what links descriptive and substantive representation is as follows: by sharing some descriptive characteristics with our representative(s), they might be better placed to put across something akin to our experience by drawing on their own as, by virtue of our shared characteristics, we also share some experiences. In this vein, Mansbridge (Reference Mansbridge1999, 629) discusses shared experience as follows:
Few commentators have noticed that the word “descriptive,” modifying representation, can denote not only visible characteristics, such as color of skin or gender, but also shared experiences, so that a representative with a background in farming is to that degree a descriptive representative of his or her farmer constituents. This criterion of shared experience, which one might reasonably expect to promote a representative's accurate representation of and commitment to constituent interests, has a long history in folkways and even in law.
Here we see Mansbridge highlight the benefits of shared experience in terms of the presumed identification and understanding of constituent interests—knowing that these exist and what they are—and the accuracy with which a representative is able to convey them. This speaks to the thought that shared experience allows a representative to transmit lived experience in a committed and accurate way, the insinuation being that this accuracy and commitment would be less present (or absent) in a context in which a representative and constituent lacked the underpinning shared experience. Iris Marion Young (Reference Young2000, 41) gestures toward a similar idea—that experience differs across the population in politically pertinent ways—writing, “Every political unit has gender differences, moreover, that are sources of different social experience and often different interests. Differences of class and/or occupation importantly separate experience and culture in most societies.”Footnote 2
Perhaps the closest neighbor to the position I outline in this article is the empirical work of Barry Burden, who writes of “the personal roots of representation.” Burden (Reference Burden2007, 5) explicitly argues that “the information gleaned, and the interests and values formed, from life experiences shape [legislator's] behaviour on roll call votes, and more importantly and frequently, their proactive leadership on a smaller set of issues.” It is notable that he uses the term “experience” extensively—for example, “Seldom do congressional scholars acknowledge what experiences members of Congress bring to the table” (Burden Reference Burden2007, 14)—but “experience” does not feature in the index of his book, nor does he at any point offer a definition of precisely what experience is taken to mean. This is not to fault his analysis; the everyday usage functions perfectly well in this context. It is nonetheless noteworthy that, in Burden's account, experience is everywhere but still somewhat hidden in plain sight.
Many, however, have had reservations about the notion of shared experience, particularly the thought of the reasonable expectation of accuracy and commitment noted by Mansbridge that I referred to earlier. Some have argued convincingly that even within relatively clearly defined social groups, experiences vary wildly, perhaps sufficiently so as to render meaningless the idea that they are shared in any way. Dovi (Reference Dovi2007, 34–35) identifies this as a “tension”Footnote 3:
On the one hand, theorists of descriptive representation have argued that certain patterns of inequalities justify having an institutionalized voice of historically disadvantaged groups (the justice argument). Such arguments emphasize the shared obstacles facing members of those groups, obstacles that prevent members of these groups being present in the political arena in equitable numbers. On the other hand, these theorists increasingly acknowledge the diversity within historically marginalized groups. This diversity can seemingly undermine the presumption that historically disadvantaged groups will be served by increasing the number of descriptive representatives from those groups, for it suggests that some such representatives lack the desire or the experiences necessary for satisfying the reasons why descriptive representation of those groups is necessary for good democratic representation.
Some root these concerns in a consideration of intersectionality, the idea that inequalities intersect and accumulate in heterogeneous ways—“thinking about intersectionality reminds us that not all women are the same, and that they do not always share the same interests”—especially when reflecting on the challenges facing those scholars who wish to study empirically when substantive representation occurs (Joshi and Och Reference Joshi and Och2014, 170). Considering this tension in terms of the likely political outcomes we might expect from an increased diversity of descriptive representation, Phillips (Reference Phillips1995, 53) writes,
Most people will accept that experience has a formative influence on political beliefs . . . some might go one step further and say that past experience sets a definite limit to the shape of future beliefs. But the notion that shared experience guarantees shared beliefs or goals has neither theoretical nor empirical plausibility. It does scant justice to what is a multiplicity of identities and experiences, and it seriously underplays the capacity for reflection and transformation.
I agree with Phillips that shared experience will fail any test based on a guarantee of substantive representation, but a guarantee is arguably the wrong standard to consider, one that sets far too high a bar for shared experience to clear given what we know about the tricky and contingent workings of political institutions. Importantly, Mansbridge, Phillips, and others (see, e.g., Childs Reference Childs2004, 22–24) make clear that the guarantee of substantive representation on the basis of shared experience among descriptive representatives should not be used as the cornerstone of a “case for women's political presence.” Instead, such a case should focus on “women's former exclusion from politics” (Childs Reference Childs2004, 23). This point is, of course, more of a comment on political strategy—regarding the optimal argument in favor of women's representation—than a purely analytical one.
On reflection, the literature seems somewhat torn between arguing two positions simultaneously: first, that it is theoretically and empirically untenable to assume that shared experience underpins the link between descriptive and substantive representation, and second, to nonetheless maintain a belief, additionally supported by the available empirical evidence, that shared experience nonetheless must have some role to play in this link. We are presented, then, with both a challenge to offer an account that does not set these two positions against one another and an opportunity to sharpen our thinking about shared experience. I argue that the philosophical literature on the epistemic effects of subjective experience gives us greater purchase on the idea that experiences might meaningfully be shared among groups, and that this idea can in turn clarify our thinking about how the descriptive-substantive link can stem from such shared experience.
EXPERIENCE AND WHAT IT TEACHES
In the context of political representation and the descriptive-substantive link, what do we mean by “experience”? Perhaps more pertinently, what do we want “experience” to mean in this debate? A unifying feature of the philosophical discussion is the assertion that experience generally refers to an individual (or groups of individuals) going through some event and that “experience” refers to this occurring from their individual subjective viewpoint (Smith Reference Smith and Zalta2013). To put it another way, experience has a first-person quality that means it is like something to undergo it—for example, it is like something for me to go swimming, to have been a fan of the U.S. Women's National Team during the 2019 World Cup, and to have been alive during Barack Obama's presidency. Although we can describe these things from a range of points of view, for me to have experience of them means that I have a subjective perspective on them—if I Φ, then I know what it is like for me to have Φ-ed.Footnote 4 Over time, these experiences can accumulate, and taken together, this accumulation is often referred to as “experience,” with certain kinds of experiences being grouped as alike in some way (e.g., experience of a given sport, of parenthood, or of a certain occupational field). My major claim is that this meaning of “experience”—focused on subjective experience and associated knowledge—is what we should be most interested in when discussing political representation.
Quite a lot, although not everything, hinges on what is distinctive or even unique about the knowledge we gain from experience. Returning to Phillips's criticisms of the notion of shared experience that I noted earlier, she is correct that experience is not deterministic, but equally, it can bestow lessons that are hard to come by from other sources and, perhaps as a consequence of this, have more impact as a result. Phillips (Reference Phillips1995, 69) seems to acknowledge this point elsewhere in her work, writing that although women might witness men working to further what are ostensibly “women's interests,” “what they cannot really expect is the degree of vigorous advocacy that people bring to their own concerns.” In cases in which we are trying to give an account of some experience of ours to somebody else, it can often be hard to describe those instances when an experience just “felt like” something ineffable, but was nonetheless meaningful to us. We all know the feeling of telling someone about an experience we have undergone and feeling the disappointment of watching their face as our recounting of it does not land in the way we had hoped.
This discussion draws out and foregrounds the question of what subjective experience of Y teaches us above and beyond possession of all possible physical, scientific, and theoretical (i.e., nonsubjective) information regarding Y. For decades, philosophers have been engaged in a metaphysical debate on the question of whether “conscious experience involves non-physical properties” (Nida-Rümelin Reference Nida-Rümelin and Zalta2009). A famous thought experiment from Frank Jackson (Reference Jackson1982, Reference Jackson1986) initially sought to establish what is referred to as the “knowledge argument,” positing that even if an individual knew every single piece of physical information about a process undergone by a conscious entity, they would still discover something new if they underwent it themselves. He posited the case of Mary, a brilliant scientist who has lived in a black-and-white room for her entire life. While in the room, she has access to all available scientific knowledge regarding the phenomenon of color and becomes expert:
She specialises in the neurophysiology of vision and acquires, let us suppose, all the physical information there is to obtain about what goes on when we see ripe tomatoes, or the sky, and use terms like “red,” “blue,” and so on. She discovers, for example, just which wave-length combinations from the sky stimulate the retina, and exactly how this produces via the central nervous system the contraction of the vocal chords and expulsion of air from the lungs that results in the uttering of the sentence “The sky is blue.” (Jackson Reference Jackson1982, 130)
Jackson asks us to consider what happens upon Mary's release from the room; the first time she sees red, or blue, or green, does she learn anything? And if so, what? This example pushes against the intuition that if the world is a physical entity, there is no more to know of it than the physical information. And yet, it remains prima facie hard to deny that when she first sees the color red, Mary will learn, or gain, some knowledge or understanding of color distinct from that which is available through physical information. It appears that whatever is happening here cannot be cashed out solely in nonsubjective terms.
The upshot would seem to be that we gain something from undergoing an experience that we could not otherwise obtain (Paul Reference Paul2014). The precise metaphysical elements of this debate are interesting but not strictly relevant to the puzzle I posed at the outset of the article—because for the most part, as we go about our lives, we do not think or behave as metaphysicians, and for our purposes, the more naïve interpretation of the general argument outlined earlier will suffice. I think that we should accept the basic premise of the knowledge argument; it seems obvious that on a mundane level, first-person experience does teach us things that we cannot feasibly learn (in the same way, sense, or form) from third-person sources.
How might we formalize our thinking about what we learn from experience? David Lewis (Reference Lewis and Lewis1988, 20) claims that experience provides us with “phenomenal information,” adding that “when we learn what an experience is like by having it, we gain new epistemic capacities to remember, imagine, and recognize.” Along similar lines, Brian Loar (Reference Loar1990) distinguishes phenomenal from theoretical and scientific concepts, arguing that when we undergo some experience, we gain access to the phenomenal concept even if we already knew all there was to know in scientific and theoretical terms. As L. A. Paul (Reference Paul2017, 10) puts it, “there exists a distinctive way of grasping facts or understanding propositions under the subjective guise that gives us new abilities to imagine and simulate, and this way of presenting and understanding arises via experience.”
Yuri Cath (Reference Cath2019) elaborates this point in ways that are useful for the present discussion and, later, for thinking further about the explicitly political implications of my argument. Cath initially highlights the puzzle that, although it seems intuitively true that one cannot gain knowledge of what it is like to Φ without having Φ-ed, it seems equally true that we can learn something about what it is like to Φ, albeit in a nonphenomenal way, by consulting various kinds of testimony offered by someone who has Φ-ed. Cath (Reference Cath2019, 113) is optimistic about this apparent division and argues that “stories, testimony, and theories can help us to gain forms of knowledge falling between these two end points” when they alert us to a similarity between some experience we have had and the one we are hearing about. So, he contends, there are different ways of coming to understand “what it is like to Φ,” offering a gradation of what he refers to as “Knowledge of Experience” (KoE) (Cath Reference Cath2019, 113–14; n.b., the following quotation is adapted to replace the example in the original with the notation of Φ used here):
Gold-standard KoE. There is some way such that Mary knows that this way is a way that it feels to Φ, and Mary knows this proposition in a phenomenal way, in the sense that her concept of that way originated in acts of directly attending to the phenomenal properties of her own experiences of Φ.
Silver-standard KoE. There is some way such that Mary knows that this way is a way that it feels to Φ, and Mary knows this proposition in a phenomenal way, in the sense that her concept of that way originated in acts of directly attending to the phenomenal properties of her own experiences distinct from, but relevantly similar to, the experience of Φ (which she has not had).
Bronze-standard KoE. There is some way such that Mary knows that this way is a way that it feels to Φ, and Mary knows this proposition in some nonphenomenal way.
Let's flesh this out with an example. If I am a pedestrian going about my business in an urban area and I am struck by a car driven by a motorist, I obtain gold-standard KoE of what it is like to be hit by a car.Footnote 5 If, assuming I avoid this scenario, someone tells me about the time they were struck by a car driven by a motorist in an urban area and I am able to “form a concept of the way that it feels” that is grounded in my “direct acquaintance with experiences that are distinct from, but that share relevant phenomenal properties with, the experience” (Cath Reference Cath2019, 114), I can gain silver-standard KoE. So, in this case, it might be that I had been struck by a bike, or had been underneath a ceiling that gave way and had rubble from the collapse strike me at speed, and so on. Indeed, an interlocutor who was familiar with my personal history of such incidents might invoke them—“It felt a bit like how it must have felt when that bike struck you.” Finally, if I lack direct or relevant phenomenal experience of being struck by a car or by things that approximate what that feels like, I gain bronze-standard KoE by hearing or reading about what it is like to be hit by a car.
In these cases, I have some sense of what it is like to be hit by a car—I might become familiar with the kinds of injuries it imparts, or the physics of what happens when a car strikes a human body—but this is a nonphenomenal understanding. Putting these claims into more approachable language: For gold-standard KoE, Mary has done X; for silver, Mary has not done X, but she knows what it is like to X because she has done things that are like doing X; and for bronze, Mary has not done X, but she knows what it is like to X because she has encountered testimony of what it is like to X.
Thomas Nagel hits a similar note to Cath on the possibility of knowing what it is like to Φ without having Φ-ed (Reference Nagel and Nagel1979, 172; emphasis added):
There is a sense in which phenomenological facts are perfectly objective: one person can know or say of another what the quality of the other's experience is. They are subjective, however, in the sense that even this objective ascription of experience is possible only for someone sufficiently similar to the object of ascription to be able to adopt his point of view—to understand the ascription in the first person as well as the third, so to speak. The more different from oneself the other experiencer is, the less success one can expect with this enterprise.
I take Nagel's notion of “understand[ing] the ascription” as offering support for the idea that we do have some capacity to understand another's subjective experience by forming a phenomenal equivalency rather than through direct exposure—Cath's concept of silver-standard KoE. But his words also suggest caution against the thought that we may be able (or desire) to rely solely on silver-standard KoE in situations of deliberation or advocacy—that if there is sufficient apparent understanding of some experience by another, the presence of the original experiencer is unnecessary—as this understanding will be conditional on the degree of relevant similarity between them.Footnote 6 Nagel hints at possible extensions of the principle of understanding the ascription to other kinds of ways of encountering the world. These might include physical features such as height, weight, disability, or so on, as well as broader physical and nonphysical characteristics presenting in terms of sex, gender, sexuality, ethnicity, and race. Indeed, I think it is equally reasonable to consider things like having been poor and rich, or from certain parts of the planet or a country, in the same vein. Put simply, the level of “success” that one person may have in the “enterprise” of “understand[ing]” the quality of the subjective experience of another (Nagel Reference Nagel and Nagel1979, 172) will be conditioned by the degree of commonality between them in terms of their possession of certain kinds of gold-standard KoE.
A final point: I have focused mostly on the epistemic benefits of subjective experience that accrue in a first-person phenomenal way—what it is like to Φ. But it is also the case that experience can deliver objective knowledge: aspects of some experience that would appear the same, in the relevant sense, to any individual regardless of their phenomenal experiencing of them. For example, the gathering of wood for fuel in South Asian countries is an activity largely undertaken by women (Agarwal Reference Agarwal2009, 2297). The rules governing the “extraction and distribution of forest products” are controlled at the local level by community forestry institutions (Agarwal Reference Agarwal2009, 2297) and can be more or less strict (in terms of both their permissiveness and the consequences of breaking them) in different places. Bina Agarwal (Reference Agarwal2009, 2297) notes that “Strict rules (a complete ban on extraction, for instance) affect the poorest households and women in general more adversely, given their substantial dependence on forests for subsistence.” This fact is not dependent on appreciation of the phenomenal elements of the experiences of the women involved in wood gathering (that it is tiring or strenuous or tedious, for example) and can be communicated—as Agarwal communicates it to her readers—without reducing its utility, which in this case would be its utility as part of a critique of how these policies are developed and of their consequences.Footnote 7
The relevant communicable aspects of the experience are not dependent on an appreciation of what it is like, subjectively, to gather wood as a woman in a poor community in Gujarat or Nepal (as in Agarwal's study), but it is nonetheless true that exposure to the relevant objective fact generally is: the objective fact regarding the effects of the strictness of wood gathering policies would be the same if it were Bill Gates doing this work, but he does not—nor, generally, do other men in the communities in question. The people who uncover this objective fact will probably be women. Here, as will also be true of equivalent cases, members of the relevant group are more likely to discover the relevant objective fact and, in turn, one is more likely to hit on the relevant objective fact by speaking to a member of the relevant group or by encountering the written or other testimony of somebody who has.Footnote 8
To recap before proceeding, my main claim is that in the case of political representation, the presence or absence of various kinds of KoE in an assembly will condition the “success” that the assembly has in delivering certain kinds of substantive representation. Given that certain configurations of both the content and type of KoE are disproportionately likely to be found among certain kinds of people, when descriptive representation is improved, we should view the similar general improvement in substantive representation in this light. This series of claims, however, relies fairly heavily on the presumed existence of groups and the suggestion that certain kinds of subjective experiences are concentrated within them. The next task is to consider in greater detail the relationship between subjective experience and the kinds of groups that are a mainstay of the political representation literature.
RETHINKING HOW EXPERIENCE MIGHT BE SHARED WITHIN GROUPS
I previously noted a squeamishness about thinking of experience as being “shared” among individuals, usually within various characteristic and societal groups. This reticence stems partly from the assumption of a certain kind of ontological underpinning for groups, at least the kinds of groups most frequently discussed in the literature—those based on characteristics of identity that are rooted, in some sense, in bodily attributes (e.g., sex or ethnicity).
A less problematic framing is not to appeal to a single essential element of shared experience that individuals can move closer to and thereby become more authentically part of some underlying group, but rather instead to think about shared experience as proximity within the possible universe of lived experience—as a question of proximate resemblance. For example, people living in the Middle Ages share something that I do not share. At the same time, it is also fair to say that I (an average academic living and working in an advanced Western democracy) probably share more aspects of my daily lived experience with Queen Elizabeth II than I do with an individual living as part of an uncontacted community in the Amazon rainforest.
None of this framing need rely on essentialism—it simply leans on intuitions about how we all experience the world in ways that are more similar to the experiences of some people than to others. So, for example, two different groups who have both suffered historical discrimination are likely to have experiences closer to one another in the grand scheme of things than a group who has never suffered discrimination in such a systematic and sustained way—there will be some shared silver-standard KoE, to put it another way. Crucially, this framing focuses on the content and nature of experience, not the extent to which any individual might measure up against some immovable criteria of identity based on psychological or somatic measures, among others. So, for example, in the case of individuals for whom the relevant identity is regularly contested—such as being a member of a social class—we need not suggest that there is some singular or immutable core of “working class-ness” that must be converged upon before any acknowledgment that they will experience the world in similar and relatable ways can begin. Moreover, when we are concerned with shared experience, we are often thinking within relatively small bounds—those of certain times or places, and so on—bounds within which it is more likely that we will find overlap at the level of silver-standard KoE, if not gold. So although it might be untenable to assert that women or workers, thinking globally and over time, possess shared experience, it is perhaps more plausible to contend that women and workers in given time periods and specific places (depending on the geographic scale in question) might have something nearer to this. Young (Reference Young2000, 82) is close to this position when she writes that “we should conceptualize social groups according to a relational rather than a substantialist logic. Secondly, we should affirm that groups do not have identities as such, but rather that individuals construct their own identities on the basis of social group positioning.”
Alternatively, we could think of experiences as being related hierarchically. By this I mean that some fine-grained experiences can be related to others but at a higher level (Salakhutdinov, Tenenbaum, and Torralba Reference Salakhutdinov, Tenenbaum and Torralba2012). For example, it is obviously different in one sense to fly in a helicopter than to fly in a plane. However, at a higher level, both involve flying (seeing the world from above, having to breathe internally circulated air, and so on); gold-standard KoE of one produces a silver-standard KoE of the other. Similarly, although the experiences of a wealthy white American woman born in New York City will greatly differ from those of an African American woman born in rural Louisiana, it nonetheless remains plausible to consider some of the experiences of gendered discrimination that they may encounter in their lives as related at a higher level. Of course, the lessons of intersectionality should caution us here—experiences of inequality intersect in varied and changing ways (Crenshaw Reference Crenshaw1990). That said, even if there is something irreducible about the nature of some experiences that we might consider to be examples of intersectionality in action, it remains unlikely that there is not some commonality or ability to draw parallels between them. Strategically speaking, points of difference can still be highlighted within political appeals for recognition. In this light, Miranda Fricker (Reference Fricker1999, 201) helpfully distinguishes between experiences being “the same” and having “similarities,” arguing that “to differentiate between sameness and similarity in this way is no verbal trickery, for differentiating them brings out a genuine distinction. Whereas the idea of sameness of experience across difference may be a romanticism from identity politics, similarities of experience across difference are a linguistic necessity.”
Importantly, either formulation—proximal or hierarchical—allows for the experiences that are shared to differ over time and space for members of the notional group, reflecting the fact that the way they experience the world will change depending on the shape it takes around them. This shape, in turn, may change the configuration of related experiences that are shared at any moment or, indeed, prompt a shift in the emphasis that members of the group place on certain experiences relative to others. Taking into account Phillips's criticism noted earlier, either the proximate or hierarchical formulation allow for agency, transformation, and change: although experience might initially set one down a certain path, it offers no guarantee of walking it in full. This said, change and transformation still entail change and transformation from some point or arrangement of points and, I think, we can consider this point or points to have been shaped by experience.Footnote 9
Building on the definition of experience offered earlier, I think we can additionally use the ideas of proximity and hierarchy that I have outlined to give an account of group identification, formation, and sustainability. Kwame Anthony Appiah (Reference Appiah2005, 86) describes how collective identities generally have three features: there is some “social conception” of that group that allows for the recognition of people as members of it, some people identify as part of the group, and people are often treated as if they are part of it. Thinking experientially, association with a social conception, identification-as, and treatment-as will all shape the experience of life of members of a group in similar ways. The process of collective intentionality involved in the development of social conceptions and treatment-as may also serve to rule certain experiences as essential or not to being counted within the membership of a group or even to adjudicate on the authenticity of that membership.
For example, the scientific community expects at some point to be confronted with the question of whether artificial intelligence is conscious or human in some sense (Tegmark Reference Tegmark2017). One's response to this question will rely heavily on one's sense not only of how we should weigh various factors in our thinking—for instance, the importance of certain biological or physical characteristics to the category of “human”—but also one's sense of the purpose of the category of “human.” These are questions about fluid categorizations that are bound up in deeper conflicts of power, resources, and aims (Young Reference Young2000, 71). In other words, they are political questions.
On this point of purpose, we can think of “groups” as a social kind, constructed to a greater or lesser extent to do a certain kind of work in the world. It is fruitful to ask what purpose we want these groupings to serve. As Sally Haslanger (Reference Haslanger2000, 35) puts it, what is it that “we want the concept in question for”? In academic debates on the descriptive-substantive link, we often invoke the idea of groups as part of a wider category of traditionally underrepresented groups (Caul Reference Caul1999). This intervention highlights the fact that when we are interested in charting the descriptive-substantive link, this tends to be because we have a political concern in play, often relating to a collective experience of historic exclusion from political power that we hope is in the process of being rectified (Haslanger Reference Haslanger2000). In many cases, our interest is in groups of individuals deemed to have been wronged or unjustly treated in the past (Mansbridge Reference Mansbridge1999); arguably, this experience of unjust treatment is the thing shared by members of what we identify as more or less unified groups deserving of representation rather than any essential characteristic in and of itself. Indeed, the shared experience of exclusion often brings about the formation of groups, or coalitions of groups, in the first place (Young Reference Young2000, 6). That is, the identity itself is formed from the oppositional statement that “you do not represent me.”Footnote 10
There is more to be said on the ontology of the kinds of groups we discuss when thinking about political representation, specifically regarding their fixity and stability over time. Yet, there is insufficient space to do that topic justice here.
RELOCATING SUBJECTIVE EXPERIENCE IN POLITICAL REPRESENTATION
Let us return to the issue of the descriptive-substantive link in light of the preceding discussion. The philosophical debate highlights the relationship between experience and knowledge that, when combined with a reframing of the idea of groups and how experience might be shared among members of those groups, allows us to locate subjective experience at the heart of the process of political representation. Given that many experiences are disproportionately undergone by certain groups within society (e.g., the experience of sexism or racism), acceptance of the argument from experience underlines existing claims around the importance of having individuals from these groups present when legislating on issues that affect those groups. On this view, we can express the primary argument of much of the existing literature—that good political representation requires the presence of different kinds of people—in a slightly different way that emphasizes how good representation requires the presence of a diversity of KoE, attached to different kinds of people. Although some might see this as a semantic or nitpicky point, it clarifies the claim made here linking descriptive and substantive representation. It is beyond the scope of this paper to mount a new empirical investigation of the specific effects of various forms of KoE on the work of political institutions as legislators interact with one another. However, I want to briefly revisit parts of the literature outlined at the outset of the article to draw out more explicitly the role of experience in light of the theoretical discussion offered in the previous sections.
Experience will provide representatives with certain things, often things that they can gain only from experience and, in the cases that often are of most interest to scholars of representation, from undergoing experiences that are disproportionately concentrated among certain societal groups. In instances in which members of those groups have previously been excluded from political institutions, this analysis sees them as bringing either relatively or completely new gold-standard KoE to the legislature. Because these representatives have a somewhat different epistemic position to what might be considered “normal” for the institution, they may see anew what might be familiar themes and practices in legislative politics. I outline three examples of this—targeted curiosity, testimonial credibility, and epistemic openness—before discussing when and how experience might matter most.
Targeted Curiosity
Representatives who have had certain experiences directly may develop a kind of targeted curiosity, whereby possession of certain kinds of first-person experiential knowledge leads an individual to seek out other kinds of knowledge about the world. Possessing a certain configuration of knowledge has the effect of wanting to have more of other, related kinds. This effect bears some similarity to Young's (Reference Young2000, 137) reflections on how one's “social perspective” can consist “in a set of questions . . . and assumptions” that one carries into the process of reasoning on an issue. There are a few ways in which targeted curiosity could work: in one sense, it could relate to a simple intensification of interest in some aspect of a relevant field, akin to the impulse to follow an investigatory lead.
Alternatively, targeted curiosity might be politically instrumental in cases in which some experience has shown me that a certain piece of knowledge missing from a debate will make the acquisition of my preferred political outcome more likely. This other knowledge need not be experiential but could instead include certain kinds of factual knowledge. In principle, this knowledge is accessible to all, yet possession of certain gold-standard KoE will result in a greater awareness of it among members of these groups than is the case for people in general. Lawmakers’ curiosity is akin to a finite resource that possession of certain kinds of experience is more likely to direct in certain directions than others, for example “greater scrutiny of the dominant perspective and its justificatory resources” (Ashton and McKenna Reference Ashton and McKenna2020, 9), perspectives and resources that, given the historical development and continued male dominance of political institutions, are likely to be white and male (Dahlerup and Leyenaar Reference Dahlerup and Leyenaar2013).
Perhaps most recognizably, this targeted curiosity will take the form of asking questions that might otherwise not be asked. Many examples in the literature reveal lawmakers invoking experience or perspective as the reason that some legislative outcome came to be (Sawer Reference Sawer2012). Kelly Dittmar and colleagues (Reference Dittmar, Sanbonmatsu and Carroll2018, 2), in their account of how women elected to the U.S. Congress interpret their progress in the institution, frequently encountered invocations of similar ideas:
these congresswomen assert that their experiences as women provide them with perspectives different to those of their male colleagues. They describe the ways in which those experiences both fuel their passion to address issues affecting women and provide a gendered lens that they apply to their work on a variety of other public policy concerns.
Often these experiences relate to a focus on a specific policy area. For example, Senator Heidi Heitkamp noted, “So many women, at least at my age, are dealing with elderly parents and see the challenges. And these are experiences we have that are different from the male members” (Dittmar et al. Reference Dittmar, Sanbonmatsu and Carroll2018, 47).
Some instances of targeted curiosity could stem from shared experiences within the legislature itself (Childs and Krook Reference Childs and Krook2009, 131). That is, shared experiences following their election might prompt some coalitions (however loose) of legislators to focus attention on some aspect of the design or procedures of the institution that had previously been considered benign, such as the encouraged or expected behavioral norms in the main legislative chamber (Lovenduski Reference Lovenduski2012). These experiences could also include issues such as a pervasive culture of sexual harassment that was previously not publicly remarked upon, or how the apparently neutral working patterns of the legislature are, in fact, gendered or racialized, excluding or impinging upon the ability of representatives from traditionally underrepresented groups to do their work (Puwar Reference Puwar2004). As noted by Josefina Erikson and Tània Verge (Reference Erikson and Verge2020, 5), the presence of previously absent groups in political institutions highlights “the fallacy of the gender-less representative who has no care responsibilities underpins formal rules about the sitting time of committee and plenary sessions, which is also reflected on the limited work-family policies parliaments have adopted (e.g., parental leave options, proxy voting, regulation of working hours or child-care facilities).”
Testimonial Credibility
Representatives who are descriptive of a previously excluded group will also bring with them, and be the impetus for the development in others of, a range of other related kinds of silver- and bronze-standard KoE as a result of the Lewisian abilities to “remember, imagine, and recognize” (Lewis Reference Lewis and Lewis1988, 20). Legislative success in terms of political outcomes will likely necessitate the co-optation of these other colleagues who lack the relevant gold-standard KoE themselves. How can legislators who are keen to push for the promotion or adoption of certain policy positions create “critical actors” or advocates who were not initially convinced by the proposed policy (Childs and Krook Reference Childs and Krook2009)? Such critical actors, lacking the relevant gold-standard KoE, do not materialize from nothing and are not convinced to take the positions they ultimately do solely on the basis of internal reflection. Rather, thinking epistemically, the claim is they are exposed up close to first-person testimony of an experience undergone by a fellow representative and that witnessing this testimony might bring about the targeted epistemic curiosity mentioned above. As Mansbridge (Reference Mansbridge1999, 644) writes, representatives with experiential knowledge can “speak on those issues with a voice carrying the authority of experience.” She considers the utility (and meaning) of such experiential knowledge in terms of speaking to constituents in a “horizontal” way, but this kind of testimony will also have an effect on fellow legislators.Footnote 11 To put it another way, testimony drawing on gold-standard KoE will carry authority with other legislators that might prompt them to view an issue through the lens of their own similar experiences (silver-standard KoE), or simply to seek to learn about it in a way they previously had not (bronze-standard KoE).Footnote 12
For example, in June 2018, member of Parliament (MP) Danielle Rowley told the U.K. House of Commons chamber, “I'm on my period”—a first for the institution (Marsh Reference Marsh2018). This took place during a debate on “period poverty,” the term used to describe the fact that many women are unable to afford adequate sanitary products. A few months earlier, MP Heidi Allen discussed her abortion during a debate on the subject, another first for the Commons. While speaking, she broke down in tears and was visibly emotional, telling her story in the first person (BBC News Reference News2018). Similarly, Marian Sawer (Reference Sawer2012, 330), discussing the notable 2006 debate that took place in Australia regarding women's access to the abortion drug RU486, highlights moving examples of “the introduction of deeply personal experiences into parliamentary debate.” It is hard to measure precisely the effect of this kind of testimony on those present who have never had (or never can have) an abortion or a period, but one would be hard-pressed to deny that the imparting of a personal experience is likely to have multiple qualities missing from an account that lacks such a grounding.
Epistemic Openness
When representatives from previously marginalized groups enter political institutions, it is not just themselves and their experiences that they bring into the legislature—in the course of their work, they also have the power to bring in the experiences, expertise, and testimony of others who might also have been previously excluded in some way. This raises the question of whom they choose to bring into the legislature, both figuratively in the form of the kinds of documentary knowledge and information they source and where they source it from, and literally in terms of bringing expert witnesses to committees and other venues where they can have input into the legislative process (Geddes Reference Geddes2018). Drawing on Dovi's (Reference Dovi2007) account of how good representatives conduct themselves, we could see experience working throughout this process in terms of her virtues of “critical trust building” and “good gatekeeping.”
Thinking about the link between descriptive and substantive representation, it may be that the presence of a diversity of gold-standard KoE instills in representatives a desire to undertake this trust-building and gatekeeping role in an epistemically just way. That is, they will seek to avoid behaviors or practices that might cause or result in epistemic injustice against marginalized groups, which Fricker (Reference Fricker2007, 1) defines as “consisting, most fundamentally, in a wrong done to someone specifically in their capacity as a knower.” She identifies two specific forms that epistemic injustice can take—testimonial injustice and hermeneutic injustice. Testimonial injustice reflects prevalent existing inequalities in social relations, such as sexism (where women's word is not taken as seriously as men's) or racism (where the epistemic contributions of individuals from ethnic minorities are devalued). Fricker (Reference Fricker2007, 4) refers to such cases as those of “identity prejudice,” which result from an “identity-prejudicial credibility deficit.” Hermeneutical injustices occur when groups are marginalized from the process of creating shared hermeneutical resources—those “collective social understandings” that allow us to make sense of the world around us (Fricker Reference Fricker2007, 147). On this definition, an individual can be said to be hermeneutically marginalized when “they belong to a group which does not have access to equal participation in the generation of social meanings” (Fricker Reference Fricker2013, 1319). This prior pattern of marginalization becomes visible when someone seeks to express themselves, to describe their lived experience to another, and comes up short “in a certain kind of failed or semi-failed attempt to render an experience intelligible, either to oneself or communicatively to another” (Fricker Reference Fricker2013, 1319).
How does epistemic injustice relate to our concern regarding the link between descriptive and substantive representation? Mansbridge (Reference Mansbridge1999, 641) writes,
Representatives and voters who share some version of a set of common experiences and the outward signs of having lived through those experiences can often read one another's signals relatively easily and engage in relatively accurate forms of shorthand communication. Representatives and voters who share membership in a subordinate group can also forge bonds of trust based specifically on the shared experience of subordination.
This appeals to the thought that being in possession of a given piece or type of knowledge is likely to result in a kind of epistemic openness as regards other knowledge that is thought to be related, specifically an openness as to both the conveyor of that knowledge and the mode of conveyance. As the number of individual representatives who share knowledge of some experience with a constituent (or constituency) increases, we might also expect this constituent's ability and willingness to communicate with the formal political system to similarly increase, likely as the result of a positive feedback loop over time (Broockman Reference Broockman2013).
Additionally, having representatives who possess experiential knowledge that was previously absent from the legislature may, over time, allow for the creation of a set of hermeneutical resources on which other (future) legislators and citizens can draw when discussing their own experiences (Fricker Reference Fricker2007). Phillips (Reference Phillips1995, 70), writing about early feminist consciousness-raising groups, notes how participants found that “the sharing of experience was part of a process in which women freed themselves from a cycle of passivity and self-denial, stretched their sense of what was possible and desirable, and reached different conclusions about what they might want,” shifting politically “from an objectively defined set of interests . . . to a more exploratory notion of possibilities so far silenced and ideas one had to struggle to express.” Consequently, the political field of discussion can expand in such a way that brings previously absent topics, and hitherto marginalized perspectives on already present issues into view.
When Will Experience Matter Most?
Echoing the contingency of Mansbridge's (Reference Mansbridge1999) original “yes” regarding the importance of descriptive representation, we should perhaps expect the extent to which experience matters, in both empirical and normative terms, to depend on certain functions in certain contexts. It seems likely that the role of subjective experience will be more visible and influential in “contexts of uncrystallized interests” in which “issues have not been on the political agenda long, candidates have not taken public positions on them, and political parties are not organized around them” (Mansbridge Reference Mansbridge1999, 643). Although the ability of individual legislators to articulate and persuade will depend on institutional design, contexts of uncrystallized interests are likely to constitute political moments in which varied and novel epistemic claims have greater opportunity to be acknowledged.
In terms of institutional setting and democratic design, it seems fairly straightforwardly clear that experience will be more visible in certain areas of legislative life than others (e.g., less visible in roll call vote patterns in situations of strong party discipline and more visible in lengthy speeches in committee chambers), and that the design of institutions will affect this. Generally speaking, experience will probably matter more, both normatively and empirically, in deliberative contexts than in nondeliberative contexts. As Mansbridge (Reference Mansbridge1999, 635) writes, “the open-ended quality of deliberation gives communicative and informational advantages to representatives who are existentially close to the issues.” When constrained by institutional rules and norms, any advantages that experience accrues may end up being less visible when we look for them and less important when accounting for a given political outcome.
A final area of interest would lie in the heterogeneity of any effects correlated with specific kinds of KoE shared across legislators. That is, in some instances, representatives have a shared experience but do not behave in the same way when legislative activity that we might see as directly related to it arises. This might be along the lines of political party. In his study of the U.S. Congress, Burden (Reference Burden2007, 9) demonstrates that “the effects of many of the member-specific variables are asymmetric across parties, depending on whether the party is a proponent or opponent of the policy status quo.” This important point underscores the fact that certain gold-standard KoE is no guarantee of any specific outcome—the KoE can appear to produce effects in different policy directions. The lessons that two individuals learn from the same experience may differ in consequential ways. What has been argued in this article thus does not contradict what has been referred to as the “simple premise” relationship between descriptive and substantive representation: “once present in politics women representatives will voice women's concerns and transform the political agenda” (Childs and Withey Reference Childs and Withey2006, 10). Rather, the argument is that, although this process of transformation is a complex one involving multiple essentially contested concepts, when seeking to study it the existing literature has often dispensed with some of these concepts too quickly and, in doing so, has prematurely closed off certain theoretical and empirical lines of inquiry.
CONCLUSION
In this article, I have located the idea of experience, specifically first-person subjective experience, at the heart of the process of political representation. Existing theoretical work on the probable basis of the link between descriptive and substantive representation (as established in empirical studies) often focuses on notions of “shared experience” but ultimately steps back from fully endorsing this owing in part to a fear of essentialism. I have argued that this fear derives from the vague way in which we talk about “experience.” Instead, my claim has been that we should think about the role of experience in political representation as experience qua the epistemic fruits of subjective experience; as a particular kind of knowledge that one gains from the subjective perspective—knowledge referred to as “gold-standard KoE” (Cath Reference Cath2019). Such knowledge can be considered to be shared either hierarchically, with various experiences sharing some higher-order qualities, or in terms of proximity, where some are closer in character to others. I have argued that although the presence of gold-standard KoE is in and of itself no guarantee of substantive representation, thinking through the process of representation in this way can help us both theorize and explain it with greater clarity.
Empirically, future research could focus on these areas as potential explanatory mechanisms for the descriptive-substantive link. For example, hypotheses might test how specific gold- or silver-standard KoE relates to legislative style, voting behavior, or contact with constituents, building on existing, similar, work that is less fine-grained (Burden Reference Burden2007). Indeed, some work has already begun to explore the role of gold- and silver-standard KoE in the substantive representation of women by men (Nugent Reference Nugent2019). Comparatively, examining the political potency of KoE across different cultural contexts may yield notable insights into representative style.
More than this, I hope that this account of shared experience will embolden those engaged in the pursuit of greater diversity in political institutions to foreground this kind of experiential knowledge in their arguments regarding democratic equality. Simply, if we feel that the experiences of our representatives matter, and that certain relevant experiences will largely accrue only to certain types of person, we want a diversity of persons and, by implication, a diversity of experience. In light of what has been argued in this article, we should be confident that such an argument is theoretically sound and be less concerned that arguments for a politics of presence are weakened when associated with arguments for a politics of ideas.
