Moral Ecologies and the Harms of Sexual Violation

University of Arkansas Press Moral Ecologies and the Harms of Sexual Violation Author(s): Quill R Kukla and Cassie Herbert Source: Philosophical Topics , FALL 2018, Vol. 46, No. 2, Gendered Oppression and its Intersections (FALL 2018), pp. 247-268 Published by: University of Arkansas Press Stable URL: https://www.jstor.org/stable/10.2307/26927958 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms University of Arkansas Press is collaborating with JSTOR to digitize, preserve and extend access to Philosophical Topics This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms philosophical topics vol. 46, no. 2, Fall 2018 Moral Ecologies and the Harms of Sexual Violation Quill R Kukla Georgetown University and Leibniz University Hannover Cassie Herbert1 Illinois State University ABSTRACT: Traditional moral explorations of sexual violation are dyadic: they focus on the relationship between the perpetrator and the victim, considered in relative isolation. We argue that the moral texture of sexual violation and its fallout only shows up once we see acts of sexual violation as acts that occur within an ecosystem. An ecosystem is made up of dwellers and an environment embedded in a broad, thick, interdependent, and relatively stable web of norms, practices, environments, material and institutional structures. We argue that many of the important and interesting harms wrought by sexual violation can only be understood as ecological harms. To illustrate this, we focus on sexual violations that occur within a specific type of ecosystem, namely an academic department with a graduate program. We examine the possible damaging effects of sexual violation on the ecology of a department. We also consider what makes 1. Both authors contributed equally to the conception and writing of this essay, and both were involved in every part. We alternate authorship on our coauthored papers and nothing else should be read into the order of authors. 247 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms an ecosystem resilient and relatively able to self-repair, and how sexual violation within an ecosystem may weaken its self-repairing resources. We show that looking at sexual violation through this ecological lens lets us identify harms that are otherwise obscured or difficult to locate. Discussions of sexual harassment, rape, and assault, whether scholarly or in the culture at large, typically frame these events as fundamentally dyadic encounters between a perpetrator and a victim.2 The pervasive assumption is that the best way to analyze the moral shape of sexual assault, rape, or sexual harassment (we hereafter use “sexual violation” to cover all three) is to figure out the harm suffered by the victim and the harm inflicted by the perpetrator. The core relevant moral statuses include being blameworthy, for the perpetrator, and being wronged and harmed, for the victim. Dyadic encounters of this sort also produce various relational moral statuses: the perpetrator may be obligated to apologize to the victim, for instance, and the victim may be owed restitution by the perpetrator. These are all moral statuses that attach to individuals or define a relationship between them. On this standard picture, there might also be some secondary moral statuses that accrue to people not directly involved in the violation: A bystander may have an obligation to help a victim, or to be a witness; a boss may have a duty to fire a perpetrator, and so forth. But these are collateral moral effects and they still attach to individuals. That is, on the standard dyadic picture, in essence, one person performs an action that impacts another. As a result, they have a new moral relationship (of beholdenness, obligation, etc.) and each acquires new moral statuses individually (having a duty, being aggrieved, etc.). If you add all these statuses and relationships together, you get a map of the moral terrain. In this essay we propose a substantial conceptual overhaul when it comes to framing the kind of harm that sexual violations inflict and the kind of moral repair they may call for. We argue that we need to see acts of sexual violation as acts that occur within an ecosystem, whose character is defined by a broad, thick, interdependent, and relatively stable web of norms, practices, environments, material and institutional structures, and which is inhabited by many dwellers. We can fully understand the kind of act that sexual violation is and the kind of moral impact it has only by understanding it as placed within such an ecosystem. We will argue that many of the important and interesting harms wrought by sexual violation are disruptions of the ecology of the ecosystem, and to its ability to provide resources for moral repair.3 2. Or sometimes multiple perpetrators, or multiple victims, but for our purposes this doesn’t change the traditional picture. 3. The idea of ecological analysis in a moral context has an important precedent in Code (2006), in which she argues for ecological thinking in epistemology. Her account is a friendly precursor to ours, although our primary focus is not on epistemic ecologies here, and although we are more interested than she is in the ground-level material character of ecosystems. 248 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms Thus the moral character of sexual violation cannot be understood by looking at the victim-perpetrator dyad taken in isolation. Nor, importantly, can we understand that moral character by adding individual and dyadic moral statuses together, for example by also considering the duties of bystanders, the harms to partners of victims, and so forth. Rather, we need to begin by understanding how it is placed within an ecosystem and what sort of damage it does to the ecology of that ecosystem. We can then work our way back in, from the ecological picture to the individual and relational statuses such as harm, accountability, and obligation. For example, on the dyadic model, we would view a date rape on campus as first and foremost an action done by one person to another, with some collateral damage along the way perhaps. On the ecological model, we view a date rape as first and foremost an act enabled by and embedded within campus rape culture, whose moral and pragmatic shape we can only understand by looking at practices of open and closed doors, the ecology of parties, how people travel home at night, social dynamics among students, and so forth. We can only fully grasp the specific harms that the sexual violation does to the victim, as well as to others, by looking at this total context. This in no way alleviates the rapist’s personal responsibility for the rape; our point is not about the systemic or ecological causes of the act. This is a separate and important debate. Our focus is not on causation, but rather on how the moral character of the act itself is integrally shaped by its place within its ecosystem.4 Likewise, we are not downplaying the harms done to the victim, because our point is also not about the magnitude of various harms and wrongs; instead, we argue that locating the assault within its ecosystem better enables us to capture the texture and significance of the violation and its effects on the victim. 1. ECOSYSTEMS AND MORAL ECOLOGIES In this section we explain what we mean by ecosystems, ecologies, and moral ecologies. In subsequent sections, we look at various ways in which sexual violation can disrupt the moral ecology of ecosystems. We believe that this lens allows us to identify harms from sexual violation that are otherwise difficult to locate. An ecosystem is a network in which an environment and its inhabitants are in thick-bandwidth interactions with one another. Human ecosystems are constituted by people, the environment, interactions between people, and interactions between people and the environment. The environment is material, social, normative, and 4. Thinking ecologically may change our priorities in responding to rape as well. On the dyadic model, bringing the perpetrator to justice and changing his behavior are the first concerns, whereas on an ecological model, our focus might be on changing the social and material practices on campus that facilitate rape, and on providing community-based support to those who have been harmed. 249 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms institutional.5 An ecosystem is distinguished from a random set of such things by having a reasonably stable character and boundaries, supporting reasonably stable and robust activities and possibilities. These boundaries are usually semipermeable, allowing people and material objects to transition into or out of the ecosystem without destroying the boundaries themselves (although entering and exiting are not friction-free or cost-free). In a functional ecosystem, the activities and the environment are mutually constitutive, knit together and holding one another in place. In an ecosystem, the character of the parts is dependent from how they fit into the whole, and conversely the character of the whole emerges from the interrelated functioning of its parts. An ecosystem is a material entity—an environment with a certain kind of relational, normative structure. Ecosystems are not abstract. They are made up of material environments and practices, and their individual character and possibilities are dependent on that materiality. This structure is its ecology, or its internal logic that lets it hang together and have integrity. In other words, the ecology of an ecosystem is this pattern of activities, norms, and possibilities that it supports and that individuate it. Crucially, ecosystems with functional ecologies are reasonably elastic and self-repairing; they can reestablish equilibrium when something throws them off balance or disrupts them. Disruptions throw the normal patterns of the ecosystem into disarray; they risk destabilizing the ecology. A few quick examples will make the notion more concrete. Keep in mind that ecosystems are by nature complex, so these examples are by necessity simplified snapshots. We will dive into the ecology of academic departments, taken as ecosystems, in the subsequent sections. • A rabbit burrow is an ecosystem. It is created by its bunny inhabitants, but conversely the movements and activities of the bunnies are shaped by the burrow. One cannot understand a part of the system—a tunnel, for example—without understanding how it functions within the whole. A burrow with a healthy ecology is self-repairing. If a passage caves in because of some trauma above ground, it can be rebuilt or rerouted. This depends on the skills of the rabbits, the materiality of the soil, and the balanced structure and design of the material environment. Before the repair, the cave-in destabilizes the ecosystem; the bunnies can’t move as they are used to, and they don’t have easy ways of getting where they need to go and engaging in their normal bunny activities. • A boxing gym is an ecosystem. It is built and used by people, and the material and social design of the gym determines what can and will happen where. The bags are for punching, the ring is for sparring, the benches along the side are for socializing, and so forth. The timing bell breaks the movements in the gym into three- and one-minute cycles of action. One cannot understand the purpose 5. For our purposes in this paper, only human inhabitants of ecosystems matter, but for other purposes the inhabitants include other kinds of life forms. There are many detailed discussions of the nature of ecosystems, across many disciplines. Although they differ in the details, our description here identifies a common core. For a good, rigorous philosophical examination of ecosystems and their nature and importance, see Rouse (2016). 250 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms of the bits of space, time, and equipment that make up the gym without understanding their role in the larger system. This is also true for the gym dwellers. One cannot understand what the coach with the mitts is there for or up to without understanding the whole network of activities and roles that make up boxing. Less formally, the guy in the corner who never gets in the ring but dispenses unwanted advice to the women who are training only makes sense as a character within the ecology of the gym. A resilient gym can recover from a disruption. If the roof leaks and destroys the floor, a gym with enough money and material integrity can fix itself pretty quickly, but a broke gym in a dilapidated garage may not be able to. If one of the best fighters gets injured right before an exhibition night, a gym with a resilient ecosystem has other fighters to put in her place. A gym that has invested all its resources in one “star” won’t be able to do this. If two members have a conflict, a resilient gym will have the social resources to navigate and repair the situation, whereas in one riddled with backstabbing and hostility, the rift will likely metastasize rather than de-escalate. • A city is an ecosystem. What one can do and how one moves in a city depends on its material, institutional, and social resources and structure: what kind of natural topography it has; what transportation systems it has; what attractions it offers; what its municipal policies are; and what the local norms for talking, making eye contact, moving, queuing, dating, tipping, and so forth are. People act and move differently in different cities. On the one hand, this is because the cities afford different possibilities. A city with a subway system obviously enables different possibilities for mobility and interaction than one without. On the other hand, these differences in norms and behavior will influence the environment: businesses will survive or die depending on the tastes and practices of the locals, for instance, and roads and subway stops will be built to accommodate commuting patterns. The resilience of a city in the face of damage or disruption—from a hurricane, or a white supremacist rally in its streets, or a sting that leaves its mayor behind bars for corruption— depends on its physical, economic, and social resources. A strong ecosystem is elastic and resilient. It has built-in ways to repair itself after a shock. Conversely, a weak ecosystem is easier to disrupt and has less capacity for self-repair. Consider the carceral justice system, for example. Arguably, our system of incarceration is the opposite of a self-repairing ecosystem. Rather than offering resources for repair, it rips the wrongdoer out of the system abruptly, doing nothing to repair the families, neighborhoods, and micro-communities left behind. Sometimes, it later reinserts the person back into the system, but not only does it fail to provide adequate resources for reintegration, it actively prevents smooth integration by stigmatizing them for purposes of getting a job, renting an apartment, and so forth. Thus in the current system, the event of someone being punished for a crime (regardless of whether or not they have committed it) is especially disruptive to the ecology of the community, while also being especially hard to repair. The ability of an ecosystem to self-repair is also not static. Sometimes, an action or event can disrupt not only the ecology of an ecosystem, but also the 251 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms ecosystem’s ability to self-repair from the disruption. Consider a highway built through a tight-knit neighborhood. Most obviously, the highway disrupts the ecology of the neighborhood, changing how and whether people can move from one part to another, access neighborhood services and businesses, and gather together. But the highway also harms the neighborhood’s ability to self-repair, because it makes the community less cohesive (often to the point of dividing it into two separate communities forever) and harms its social capital, while also providing fewer resources to people on each side of the divide. We argue in subsequent sections that sexual violation in an academic context often has this second, meta-level effect: it both disrupts the ecology of a department and also harms the department’s self-repairing resources. Kevin Timpe (2018), like us, uses the language of moral ecologies. His core point is that human agency is inherently scaffolded by its environment; there is no determinate answer to what someone’s capacities and possibilities for action are independent of the social and material space they are in. How mobile someone is, for example, depends not just on their anatomy and fitness but on the built and natural terrain they need to navigate, including its transportation options and the like. As bioethicists have discussed at length, whether someone has the capacity to make an informed autonomous decision depends not just on the dispositional properties of their brain but also on how familiar their environment is, who is with them, how they are spoken to, and more. Thus, Timpe argues, if our goal is to enhance people’s ability to act well, including their ability to act morally, we need to work on their environmental scaffolding and not simply take them as individuals. We agree with Timpe’s account, which helpfully complements ours. But we do not need to draw upon his strong constitutive claim about agency for our purposes. We also mean something more specific than he does when we refer to moral ecologies. Moral ecologies, as we use the term, do not attach to special, distinct kinds of ecosystems. Rather, most sufficiently complex human ecosystems already have moral ecologies as well. That is to say, they constitute a structured moral environment. The moral shape of a particular action or situation—what the morally possible responses are, what sort of harms it inflicts, what possibilities for generosity or building relationship it opens and closes, and so forth—is deeply shaped by the ecosystem in which it occurs. Particular human ecosystems have particular sets of local moral norms built into them. Moreover, features of the ecosystem that have no overtly moral content—such as who controls a certain resource or has a certain kind of institutional power, or whether there are gender-neutral bathrooms, or whether there are easy transportation options for leaving—may shape the moral significance of actions and the moral texture of situations within the system. The moral ecology of an ecosystem is made up of these patterns of moral possibility and significance. Consider a family. Imagine that a brother skips his sister’s college graduation, and the sister is angry. Whether this was a moral violation in the first place, and how serious it was, depends on expectations and traditions within the family; the particular history of the relationship between the siblings; how hard it was for the 252 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms sister to make it to this point given the family’s resources; and much more. As every good advice columnist knows, there is no way to assess the moral texture of this situation or the meaning and gravity of the brother’s act without knowing many such details about the ecology, including the moral ecology, of this family. Most or perhaps all families carve out and sediment roles for its members—troublemakers, relationship-repairers, the ‘smart one’, the ‘weird one’, whatever—and these are part of what gives meaning to the act. What kinds of responses are possible will also depend on this family’s ecology. Can the sister just talk to her brother, or will an aunt need to intervene on her behalf? Can the brother just take the sister out for a celebratory dinner some other time to make it up to her, or is this family deeply entrenched in and attached to traditions and rituals such as graduations? Does their fighting put a younger sibling in the middle, so that sibling cannot act at all without being interpreted as choosing one side or another? All these details directly impact the moral statuses of the players involved: the brother’s obligations, the harm to the sister, her right to feel aggrieved, the moral pressure on the aunt to intervene. Moreover, how easy it is to repair the family ecology after this disruption will depend on all sorts of subtleties. Is dad the type to take sides and escalate the situation? Is the family prone to grudges? Does it have an internal culture of brief explosions of anger that burn out quickly? And so forth. This example lets us start to see what’s wrong with the approach to moral analysis that starts with dyadic relationships and individual moral statuses, adding them together to get a total moral picture. The point is really twofold. First, as we just emphasized, the harm that the brother does to his sister (for instance) only has the particular character and gravity it has because of the larger context of this family’s ecological dynamics. In some families, skipping the graduation would be a deeply hurtful act; in others it’s a minor brush-off, easily fixed with a nice dinner the following week. But also, and crucially for our purposes, some of the moral harms at play here are harms to the ecology itself, and so they cannot show up at all on a dyadic picture. If the brother’s act shows disrespect for a longstanding family tradition of getting together at graduations and taking them seriously, then it can destabilize this norm. If it causes family members to take sides, it can cause a fracture in a network of relationships. If it is read as diminishing his sister’s accomplishment specifically, it might introduce a sexist dynamic that fits poorly with the family’s established values. Further, the effect that this has on other family members such as the younger sibling isn’t well captured by merely analyzing the harms or obligations that accrue. Instead we need to look at how the framework that gives meaning to their actions is itself reshaped or disrupted. Now, showing up to a standing Wednesday night dinner with the brother may count as siding with him, while canceling dinner becomes siding with the sister. The brother’s action thus disrupts the moral ecology, in a way that changed the possibilities for action that other family members now have available to them. These are ecological disruptions, and accordingly, if they are to be repaired, the repair has to happen at the level of the ecosystem. Until the family repairs, its normal, sedimented patterns of interaction and activity will be destabilized. 253 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms Importantly, notice that a strong and resilient ecosystem may not have a good moral ecology; the norms and values and practices that it holds in place with resilient stability may be morally crappy, and not conducive to the flourishing of all its dwellers. In some families, the kind of conflict we just described is the norm and part of the ecosystem. There are families in which hurting a particular member or escalating conflict so that everyone is involved is the local, sedimented tradition. A community with very traditional sexist and homophobic norms in place, which prevents its girls from doing anything other than growing up to be mothers subservient to a man, may be very stable, and have a well-entrenched set of values. It may have norms in place that make it almost impossible for anyone to effectively challenge or question the system, and it may self-repair quickly by simply demonizing and disowning norm-violating girls. Thus it would be wrong to say that it is an overriding moral goal to make ecosystems resilient, or to avoid their disruption. Sometimes they need to be disrupted, or even destroyed if they are sufficiently harmful to their members. In the academic context, clearly, many departments have stable, sedimented norms that enforce unjust hierarchies, disrespect members that do not conform to a narrow intellectual profile, maintain white supremacy, and so forth.6 These departments need to change. However, ecological disruption is itself typically harmful and disorienting to the members of the ecosystem, in its own right. It makes action more difficult, depletes resources, fractures social relationships, and undermines bodily and emotional comfort. Thus ecological disruption, and struggles over self-repair, are generally morally bad things in and of themselves, even if they may sometimes serve a greater good in the long run. People ousted from their ecosystems are left without the ecologically based roles around which they’ve developed their skills, sense of self, and values. Even in the case of a deeply morally distorted community, such as an evangelical cult perhaps, ecological disruption can leave members alienated, traumatized, and seriously lacking in social and material resources for rebuilding. Someone who has developed skills for being a good community member within the evangelical cult, say, by organizing Bible camp for the neighborhood kids and clothing drives for church members, is left without a framework for how to be a good neighbor once they’re removed from the cult. Disruption is therefore worth studying as a moral phenomenon on its own terms, even if in some cases we believe that it is instrumentally worthwhile. 2. SEXUAL VIOLATION AND ECOLOGICAL HARMS From hereon we focus on a specific type of ecosystem, namely an academic department with a graduate program. We also focus on a specific type of disruptive, morally harmful act, namely a sexual violation. We want to see what happens when we 6. Many thanks to Susan Brison for pushing us to develop this point. 254 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms try to capture the moral contours of sexual violation by thinking ecologically rather than dyadically. We examine the possible effects of sexual violation on the moral ecology of the department, and how it might destabilize or disrupt the ecosystem itself. We also explore what ecological features of the department might make selfrepair after this kind of disruption possible or impossible, difficult or easy. We will try to show that looking at sexual violation through this lens lets us identify harms that are otherwise difficult to locate. As we go, we will also briefly consider how our current institutional responses to sexual violation help or hinder repair. The kinds of effects we are interested in here are maximally contextual and local, and are dependent upon the institutional, material, and social complexity and specificity of an academic department. If we wished to argue that any of the effects we are talking about are necessary in a given situation, we would need to write an entire novel developing a compelling example and giving it rich moral texture. Thus the best we can do is point at possible effects, and count on our readers to use their imaginations to fill in messy details that make our examples plausible. We hope and expect that anyone who has been in an academic department that has lived through sexual violation allegations will recognize many of the effects and phenomena we describe below. We also want to make clear that while our discussion draws heavily on our own experiences with multiple departments in which a sexual violation has occurred, and on discussions with many friends and colleagues who inhabit many different roles in these departments, we are not targeting or focusing upon any one department or any one actual set of events in this essay. 2.1 NETWORKS OF TRUST Any functioning ecosystem relies on networks of trust, in the sense that dwellers count on one another to follow certain norms, behave in predictable ways, and have particular competencies. They also trust their environment to scaffold their habitual activities in dependable ways.7 In an academic department, students trust faculty to know the material in their areas of expertise, and to advise them appropriately. Faculty trust one another to carry out their assigned duties, and they trust students not to plagiarize their work, and so forth. Trust is especially important in a context such as an academic department, in which there are marked and necessary power hierarchies, with some members having a great deal of control over others’ careers and well-being. Sexual violation in an institutional context such as an academic department obviously violates the victim’s trust and renders the perpetrator untrustworthy. But more on point for our purposes, it can disrupt networks of trust. After an incident of sexual violation, often, no one knows exactly who knows what happened, or who is (perhaps of institutional necessity) protecting whom, or which side people are on. Students may not know who they can safely 7. In a suitably thin sense of trust, this applies to nonhuman ecosystems and nonhuman members of ecosystems that include humans as well. 255 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms turn to for support or with questions. The event, typically shrouded in all sorts of formal and informal confidentiality protections, can fragment everyone’s ability to interpret others’ words and interactions. Sally Haslanger (2012) has analyzed the role of presumed common ground in making conversations and interactions comprehensible. Sexual violation in a departmental context is particularly effective at undermining common ground. If one person knows what happened and the other does not, and neither knows what the other knows, conversations quickly go off the rails.8 The disorienting erosion of hermeneutic trust based on misinformation, partial information, and not knowing where people stand can happen at every level—between faculty, between students and faculty, and between students. For those in the know, everyday interactions become fraught with potential hidden codes and meanings. For those not in the know, various events and speech acts can become altogether illegible. If people do end up openly taking sides, this can introduce a different kind of serious moral distrust. In turn, this erosion of trust destabilizes the ability of the ecosystem to function as usual. Moreover, it directly damages the system’s ability to repair itself, since moral repair of this sort is a social and collaborative activity. The legal and informal culture of confidentiality that surrounds these cases directly contributes to the ecological damage to trust networks. Much, though not all, of the erosion of trust comes from people knowing different fragments of the story, not knowing who knows what, and not being able to ask or tell. In this climate, rumors and half-truths and incoherent pieces of stories circulate, adding to the general problems of hermeneutic breakdown. Furthermore, neither the victim nor the perpetrator can play an active role in moral repair if they are bound to silence. Crucially, nothing in our analysis here is about individuals having an untrustworthy character or making untrustworthy choices. The perpetrator presumably made untrustworthy choices, of course, but this isn’t our focus. Rather, the disruption of networks of trust is a systemic ecological event. It cannot be reduced to the moral or other agential features of the players involved. 2.2 NETWORKS OF FRIENDSHIP Friendships in academic departments increase intellectual community and learning. Intradepartmental friendships can occur between faculty, between students, between faculty and graduate students, and so forth. They lead to formal and informal collaborations and mentorships. They provide much-needed social and emotional support. These things are critical to the ecological well-being of a department and 8. In one case we know of, about half the faculty knew that a case of sexual harassment by a faculty member of a graduate student had been reported and the other half did not. One faculty member assigned the perpetrator an official role that put him in a position requiring intimate and trustworthy interactions with graduate students. Another faculty member assumed that this was because the first did not take the harassment seriously. The first had no idea it had happened and was confused by the second’s objections to the appointment and anger. Hijinks ensued. 256 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms they shape its moral texture. Friendships are important resources for repair if the ecosystem is disrupted, as they provide a banked supply of presumed good will and good intentions, as well as an entrenched basis for collaboration. Moreover, they can provide a ready-made network for distributing and coordinating the labor of repair. Sexual violation harms friendship networks directly, in straightforward ways that presumably don’t need spelling out, but also indirectly. It may well result in the tightening of rules for faculty-student socializing, for instance. Even if there are no new formal rules, there may be significantly more discomfort around facultystudent socializing on everyone’s part. Furthermore, already-established friendships may be reinterpreted with suspicion, by third parties or even by the friends themselves. They may become sexually suspect, or they may be seen as possible alliances in a new culture of distrust and choosing sides. One of the most complex forms of ecological damage to friendship networks surrounds friendships with the perpetrator. Sexual violation rarely results in complete ostracization and denial of all friendships with the perpetrator.9 And indeed, there can be well-founded reasons for continuing those friendships, such as complex webs of emotional entanglement, layers of merited gratitude and obligation, and a longstanding history that is not easily severed.10 This is not to say that friends should ignore the violation or fail to hold their friend accountable for it; indeed, friends are the ones who may be best positioned to intervene and help change the perpetrator’s behavior. But being friends with a perpetrator can put one in an untenable social position. Refraining from publicly denouncing one’s friend can easily be perceived as an act of complicity or tacit support. In fact, we have seen friends of the perpetrator held accountable for that friendship more publicly and to a greater degree than the perpetrator themself was held to account for the sexual violation. Being seen, within the department or the discipline at large, as the friend of a known perpetrator can itself contribute to a culture that condones sexual violation.11 But denouncing them will surely undermine the friendship and may set back efforts at ecological repair. Thus friends of the perpetrator may find themselves stuck in genuine moral dilemmas. Moreover, being friends with the friend of a perpetrator can also be dangerous. Not only can it be emotionally scary, but it is hard to feel confident that your friend isn’t passing on information to the perpetrator, innocently or maliciously. 9. Social ostracization from doesn’t prevent the perpetrator from sexually violating people in other communities and, as we’ll discuss, rarely helps with genuine ecological moral repair. 10. We recognize that a genuine friendship typically involves a substantial degree of mutual trust, support, and esteem for one another, and these things may or may not be possible after a friend has committed a sexual violation. At the same time, erstwhile friends of the perpetrator may choose to continue to uphold other dimensions of the friendship, especially when they work together in the same department. Whether the relationship remains a genuine friendship or becomes some other kind of collegial relationship depends on the case. 11. See BoJack Horseman, Season 5, Episode 10, “Head in the Clouds,” for an interesting discussion of the dilemmas of being publicly friends with an abuser. 257 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms The erosion of friendship networks and of trust networks are closely linked. Disrupted friendship networks tend to create hurt feelings, perceptions of favoritism, secret-keeping, and gossip, all of which erode trust. Conversely, eroded trust is not conducive to friendship. 2.3 MATERIAL CULTURE Ecosystems are materially and spatially embodied systems. Sexual violation can disrupt the material culture and practices of a department; it can alter how the spatial environment functions to support the activities that happen within it. For example, the department lounge may go from a safe common space for socializing to one that excludes certain people. The victim, and perhaps other potential victims, may not feel able to use the lounge if the perpetrator may walk in at any moment. Some people’s discomfort and disrupted spatial patterns may ramify out, and shift spatial patterns more dramatically. Closed office doors may take on new or unsettled meanings; it may not be clear who is hiding, being aloof, engaging in untoward activity, or just writing. Conversely, work patterns may be disrupted if people feel the need to keep their doors open to avoid such interpretive ambiguity. Spatial patterns of working and socializing may be disrupted: asking to meet a student at a coffee shop to collaborate may become open to sinister interpretation.12 The department may need to shift its formal and informal rules for when there can be alcohol at events, and where and with whom professional socializing can happen. Sudden, abrupt shifts in material culture—in how we can use space—are disorienting and can be exhausting. They require us to explicitly think through where and how to do things, and they make our embodied negotiation of space cumbersome and anxiety-producing. The cognitive and emotional load involved in navigating a space that is not working as it was designed to work, and that suddenly thwarts your habits, is intense. Such shifts place burdens on everyone in the system. Everyone now needs to attend to, decipher, and physically renegotiate shifting material practices. Ex hypothesi, we are considering a department whose sedimented ecology enabled sexual violation to happen. This brings us back to our earlier point that some ecosystems support a flawed moral ecology and really need to be rebuilt (or in extreme cases, destroyed). Some of the material shifts we have discussed may well be for the good in the end. While it’s almost certainly bad if some people don’t feel comfortable using the lounge, it may well be good to get rid of alcohol at departmental events, or to have tighter constraints on off-site collaboration and advising. But importantly, material disruptions to an ecosystem are burdensome even when they lead to something better in the end. Disrupting even a morally 12. In some departments, it is already part of the ecology that such off-site meetings don’t happen and would seem sketchy if suggested. This might be part of a stable ecosystem. We are specifically considering a case in which this is a sudden change in an ecosystem. 258 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms flawed ecosystem comes with burdens and harms that might be quite serious. Abrupt damage to an ecosystem from a moral violation is surely almost never the best way of bringing about needed ecological change. At a minimum, it has collateral moral costs for everyone in the ecosystem, and not just for the perpetrator and the victim. A sudden shift in material patterns in response to a crisis is draining and destabilizing, even if the emerging new patterns might be part of a better moral ecology in the long run. 2.4 INSTITUTIONAL SCAFFOLDING In order to function, academic departments rely on a variety of institutional mechanism and systems. Members of the departments need to count on these mechanisms and systems working as they were designed to work. Some are formal and some are informal. Faculty members count on there being a procedure for the fair and sensible distribution of teaching and service assignments. Graduate students count on procedures for choosing their committees, getting letters of recommendation, being assigned suitable teaching assistantships, receiving support on the job market based on their work, and so forth. When someone in the department sexually violates someone else, it can throw any or all of these systems off. Part of the perpetrator’s sanction may be that they cannot teach or do certain service jobs, leaving their colleagues to fill roles for which they are not really qualified or prepared, and perhaps leaving students without certain courses. If the perpetrator continues to teach, students may not feel comfortable taking their classes or serving as their teaching assistants, and may be left to navigate the process of working around them without any procedures in place. A sexual violation may change who can get letters of recommendation from whom, and how certain letters will be read by outsiders. A student who was counting on a letter from the perpetrator may now be caught between being a letter short, or having a letter that risks being seen by outsiders as a sign of an unsavory alliance or even a sexual tryst. Committees are sometimes reformulated in ways that do not track or maximize helpful research expertise. The presence or lack of presence of the perpetrator on a committee may be interpreted by others in all sorts of strange and perhaps inscrutable ways, which vary depending on how complete and accurate outsiders’ information is. This kind of departmental disorder is stressful and disorienting for everyone involved. It creates a great deal of extra work. It also means that who does what job, takes which courses, writes which letters, and so forth less accurately tracks everyone’s actual scholarly work and expertise. In this sense the disruption undermines the telos of an academic department. We want to emphasize that these are not theoretical risks; we have both witnessed this kind of disarray in action, in several different departments, in the wake of sexual violations. We hope it is clear by now that these are harms to the ecology of the department; they are specifically systemic harms that undercut people’s ability to act smoothly and well within the environment, and which distort the results of various actions. Moreover, putting aside the original act of sexual violation, they are harms 259 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04an 1976 12:34:56 UTC All use subject to https://about.jstor.org/terms that are not caused by individual bad moral actors or individual incompetence. Everyone involved may be doing their best to restabilize the ecosystem and compensate for the damage, and yet the system itself may not have the scaffolding and resources to cope smoothly. 2.5 DISTRIBUTED NETWORKS OF LABOR Overlapping with but conceptually distinguishable from the destabilization of institutional scaffolding is the disruption of the ecological distribution of labor. The work of running a department is distributed and coordinated. When one person can’t teach a core class, or supervise grad students, say, because of having sexually violated someone in the department, then others have to take up the slack, and they end up with an unfair and perhaps seriously burdensome workload. Colleagues may have to abruptly take over tasks or courses that they don’t really have the capacities or knowledge to take over efficiently. Moreover, sexual violation is the kind of event that not only redistributes existing labor, but generates new forms of labor. As fallout from the ecological damage that it inflicts, it often creates the need for extra emotional labor: calming fears about sexual violation itself; managing anxiety and concerns about the impact of the disruption on job searches; fielding sensitive questions; mediating between parties now suddenly in conflict; connecting people with support resources. It also creates new social and administrative labor: meeting with Title IX officers; giving witness statements; organizing meetings to respond to concerns or come up with new policies, etc. Indeed, this added labor can be immense. Furthermore, all this extra labor is not evenly or justly distributed, even bracketing the frustrating fact that the perpetrator generally gets a reduction rather than an increase in workload. For instance, typically, the people called upon to take up the slack in advising are also the same people to whom students are already more likely to turn for emotional labor. As ‘safe’ members of the department, these people are also more likely to be charged with negotiating and mediating sensitive conflicts, planning responses, and the like. We recognize that sexual abusers come in every demographic, as the recent Avita Ronell case made vivid. Still, surely there is no denying that professionally secure men are the most common sexual violators in academic departments. Moreover—regardless of the identity of the perpetrator—women faculty, faculty of color, disabled faculty, and queer faculty are often uniquely positioned to give both emotional and practical support to students and junior faculty who share their demographic. They are also often the ones seen as ‘safe’ more generally, and they are the ones from whom departments are most likely to demand extra service in a time of need. Hence sexual violation not only throws off the ecology of labor; it is likely to exacerbate existing unequal burdens and to harm those already more vulnerable within the department. When members of the department are overworked, and especially when those who are disproportionately responsible for emotional and social labor are particularly overworked, the capacity of the ecosystem to self-repair is also damaged. 260 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms People simply have less time and emotional bandwidth to work on repair, and again, those most likely to be in repairing roles are the same people who will be especially over-tapped. 2.6 EPISTEMIC ECOLOGIES, EPISTEMIC HYGIENE, AND EPISTEMIC BURDENS A species of labor whose ecology can be disrupted after a sexual violation is epistemic labor. Sexual violations can create extra epistemic labor that isn’t neatly distributed in an ecosystem. They generate traditional epistemic labor, such as figuring out university and Title IX codes and procedures, preparing witness statements, and the like. They are also prone to generating gossip, misinformation, and legitimate questions and concerns without easy answers, and informal but often intense epistemic labor is required to manage this extra burden. All of this extra labor can constitute a form of ecological destabilization and disruption, as already-busy department members—often the same ones now performing extra institutional labor and emotional labor, as we discussed in the last section—need to make room for it, and there are no preestablished, sedimented norms in place for performing it. Much of the epistemic labor that comes along with gossip, rumors, and partial information—especially when the topic is legitimately stressful and high stakes for many of the people involved—is what, building on Karen Frost-Arnold’s work, we call epistemic hygiene. Frost-Arnold (2018) points out that “all epistemic communities produce discarded and unwanted by-products of epistemic practices of knowledge, understanding, and communication. Unjustified claims, false beliefs, misunderstandings, and miscommunications are all general types of what we might call epistemic trash” (our emphasis). Epistemic trash is the byproduct of messy epistemic practices. This kind of epistemic detritus, produced by rumors, incomplete stories, and the like, has to be cleaned up in order for an ecosystem to have a functional epistemic ecology. This process of cleaning up the trash— clearing up misinformation, dispelling rumors, filling in stories—is what we are calling epistemic hygiene, and someone has got to do it, even though it can be unpleasant, stressful, and recalcitrantly difficult. Note that epistemic hygiene is required for all messy epistemic practices, which is to say all interesting and rich epistemic practices, whether or not they are unfortunate. When we perform clinical trials, we need to clean up data by throwing out outlying data points and data points generated outside of proper protocol. When we moderate a talk or teach a class, we need to shut down lines of questioning that are drifting way off-topic. When we learn a new sport we often pick up extra habits and motions that seem to be part of the skill, but are actually just inefficient or counterproductive rituals that we need to identify and clean up. But when an ecosystem is disrupted by the kind of event that naturally generates anxiety, gossip, and misinformation, and with respect to which the information flows have to be tightly managed for legal and privacy reasons, the labor of epistemic hygiene can be quite intense. 261 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms Moreover, Frost-Arnold suggests, quite plausibly, that burdensome and stressful epistemic hygiene “is disproportionately performed by some (typically traditionally oppressed) groups while other (typically traditionally dominant) groups benefit from this labor while usually being oblivious to its performance.” For instance, Frost-Arnold points out that when hate speech circulates around a campus (via Jordan Peterson or ‘trans-exclusionary radical feminists’ or whoever the source of the moment is), there is often a great deal of epistemic clean-up to do, and it is the people who are already vulnerable who tend to see that it needs to be done and take responsibility for doing it. Likewise, often, after a sexual violation, senior men in the department will barely be aware of the ecological disruptions around them, not to mention the anxiety, pressing concerns, and rumors circulating among the students and other faculty. Complex social dynamics pass under their radar. Meanwhile, faculty members or student leaders who are women, or junior, or otherwise especially trusted or relatable for people feeling under threat or disoriented will be the ones bearing the brunt of cleaning up the epistemic trash that has been dumped on them. This may require them to acquire new knowledge, quickly learning how best to support survivors and others, and familiarizing themselves with university procedures and best practices around sexual violations. The epistemic work created by a sexual violation within a department can be ongoing for an astonishingly long time. Knowing what work needs to be done involves tracking who is struggling; deciphering who knows what and who needs to know what; and updating information about what sorts of harm are playing out and how information is spreading and circulating. Doing the work requires repeated conversations, meetings, e-mails, and so forth. This work is often stressful and emotionally draining, and may require emotional labor in addition to the epistemic labor. Keeping the epistemic situation tidy and cleaning up debris is difficult and can be exhausting, and it can be made all the more so because it is work done to clean up someone else’s mess. One of the people in the ecosystem who is virtually certain to be saddled with new epistemic responsibilities is the victim, who typically ends up in a protracted negotiation over who to tell, when and how much to tell, and when and how to correct misinformation. A victim who accuses a fellow department member of a sexual violation will not have the luxury of avoiding choices about how to manage the flow of information. Either way, they need to figure out how to respond when they are asked directly about the climate in their department, or about whether a rumor about a sexual violation is true, or when they hear a distorted version of their own story.13 They have to decide whether or not to warn others who are considering coming to the department or working with their assailant. Likely they will repeatedly face direct pressure to speak simply in virtue of the ecological structure itself. 13. We know of one case in which a victim’s dissertation director asked her repeatedly if she knew who the victim was and whether the event had actually happened. 262 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms Extra epistemic burdens on victims are both ongoing and more or less inevitable. Even a victim who chooses to say nothing at all will need to make this choice over and over again, as new people come into the department, or as new students sign up to work with the perpetrator, or when their department wins praise for its social climate, or whatever it may be. Moreover, this is not just a passive choice to stay quiet, but one with ecological effects. Withholding information can chip away at the victim’s networks of trust and friendship and their smooth embeddedness in the department ecosystem. Their behavior may become ‘odd’ from the outside, if they skip departmental events or don’t sign up for classes to avoid their assailant. Staying quiet may isolate them over time.14 Hence the epistemic burden of managing information, even through silence, is intertwined with other kinds of ecological harms. Moreover, a victim’s choice to stay quiet may itself constitute an epistemic injustice, if her choice results from what Kristie Dotson (2011) calls “testimonial smothering,” which is a disempowered speaker’s choice not to speak because of their reasonable fear that their testimony will be received incompetently in ways that will harm them. Elizabeth Harman (2019) argues that a victim in fact has moral obligations to report their experiences in order to protect future victims, even when doing so is harmful to the victim, and that this is one of the second-order harms of sexual violation. We strongly resist this interpretation, as we think the victim never has a moral obligation to tell her story or to manage the harms wrought by the person who victimized them. Nor, conversely, does the victim ever have an obligation to keep the incident to themselves. The event is theirs to share when and how they do or don’t see fit. Rather than framing this situation in terms of moral obligations, our account lends itself to an ecological analysis: one of the ecological disruptions wrought by sexual violations in the context of an ecosystem such as a department is this type of epistemic disruption, which does practical harm to victims regardless of how they decide to navigate the extra epistemic labor they find themselves saddled with. We can deny that sexual violations impose new moral obligations on victims while acknowledging that they find themselves with new moral pressures, including moral epistemic pressures, to navigate whether they like it or not. Insofar as navigating these decisions is stressful, exhausting, and often retraumatizing, it constitutes a separate, essentially ecological layer of harm to the victim. In an ecological context bound up by institutional secrecy, these epistemic burdens will be all the more intense, because the victim may be the only person in a position to issue warnings and correct misinformation as needed. Thus survivors are often called on to perform an undue amount of the work of epistemic hygiene, along with other forms of burdensome epistemic labor. There are three points that come out of this subsection. The first point is that disruptive events within an ecosystem—especially those, paradigmatically including 14. In the final section, we discuss the case of a victim who stays quiet because she is already isolated to begin with. 263 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms sexual violations, that are shrouded in privacy, induce gossip, and raise the specter of real risks to others in the community—produce a great deal of chaotic and unexpected epistemic labor that many ecosystems are ill equipped to absorb. Moreover, like the administrative and emotional labor discussed in the last section, they deplete crucial resources. They drain the time and energy of department members and thereby further weaken the resilience of the system. But without this labor, the ecology further fragments, as trust, friendship, and the other systems we have discussed come increasingly under threat. The second point is that the need for epistemic hygiene after a disruption tends to generate a special form of epistemic injustice, as the burdens it imposes are not evenly distributed. More privileged members of the department may be oblivious to the epistemic labor going on around them, and less privileged members are stuck taking out the epistemic trash.15 The third point is that victims, because of their distinctive epistemic relationship to the original event and their distinctive social position, will more generally be particularly burdened with difficult epistemic labor, including but not restricted to the labor of epistemic hygiene, in the wake of their victimization, and this can constitute a further layer of harm they must endure. 3. ECOLOGICAL VS. DYADIC MORAL ANALYSIS Over the course of section 2, we tried to make the case that sexual violations within the ecosystem of an academic department are likely to cause a variety of ecological harms—harms to the ecosystem as a whole, and harms to people that can only be understood insofar as they are embedded in the ecosystem. Moreover, as we went, we argued that many of these harms are also what we might call metaharms, insofar as they damage the ecosystem’s resilience and reparative abilities. These acts not only cause harm directly, but they harm the capacity of the system to repair those direct harms and exhaust its resources for doing so. At this point, we want to return explicitly to the topic of moral ecologies. The harms we have discussed do not just impact the ability of the ecosystem to function; they more specifically change and distort the moral terrain of the department. They alter the moral valence and significance of various actions. They also change the possibilities that exist for acting well or poorly, and the opportunities for flourishing or suffering. Consider, for instance, the erosion of trust networks that we examined in section 2.1. Trust makes possible a variety of morally positive actions. In some departments, people are reluctant to share drafts of their work because they are concerned others will poach their ideas. A strong network of 15. Indeed, many academic philosophers unlikely to be personally harmed by epistemic trash currently seem to be actively in favor of what we might call epistemic littering: allowing anyone to make any public claim no matter how ill-informed and damaging and hateful, while leaving others to do the labor of cleaning up and taking out the epistemic trash. 264 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms trust enables people to give feedback on work as an act of moral generosity. Trust enables a faculty member to do right by a student by suggesting that the student replace them on a committee with someone more intellectually appropriate, without this being read as a betrayal, or as implying that the professor does not like the student’s work. Prior to any particular action, a strong trust network makes a certain space of moral activity possible, and a weak one undermines it. Moreover, moral ecologies shape the moral texture and status of particular actions once they happen. An offer to take a student out for a celebratory glass of wine after they get their first paper accepted for publication may be a socially fraught invitation that puts the student in an uncomfortable position with no good way forward, in the context of a department with very few opportunities for friendship and a history of untrustworthy behavior, or in one in which the material culture is strongly rooted on-site in the department. Within a different ecology, in which these two people are friends with a history of doing various things together in various places, this might be a completely normal, considerate, pleasant offer, which can be easily declined if the student is too busy or for whatever reason. Thus when we engage in moral analysis with respect to an ecosystem such as an academic department, a crucial set of questions concerns the moral quality of the ecosystem itself. Does it make generosity possible, and in what forms and from whom? What burdens does it attach to invitations, and are these burdens systematically unevenly distributed? Does it support and enable moral goods such as friendship, or does it impede and undermine them? Does it build in systematic epistemic injustice by creating hierarchies of testimonial status, or by unfairly distributing epistemic labor? And so forth. These are all moral questions that cannot be answered by looking at any set of particular individuals’ actions, relationships, and statuses, no matter how many we add in; rather they concern the framework that makes those possible and gives them their shape. Conversely, when we do morally assess a particular action in an ecosystem, we often need to place it within its ecology to get the moral assessment right. The elderly tenured man who invites a new, untenured young woman assistant professor over to his house to discuss her work over wine may insist that he is just issuing an invitation, which she is free to turn down. But he is quite likely being willfully obtuse concerning how the moral valence of the invitation is shaped by the power hierarchies, gender and age inequities, and trust networks in place in that department. Sexual violations reshape the possibilities for moral behavior within an ecosystem, and thus shift the moral ecology. For example, in a department where some people know of the violation and others don’t, a professor who knows of the violation may not be able to explain to an unknowing mentee why they can’t cochair the student’s committee with the perpetrator without violating the survivor’s trust. The problem is not simply that sexual violations create conflicting obligations (which is nothing new), but rather that sexual violations alter the structure of the ecology itself, such that people may not be able to ethically fulfill their roles within that ecology. 265 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms Thus no story about obligations, permissions, beholdenness, blameworthiness, rights, or other individual or dyadic moral statuses can fill in the ecological story for us. An attempt to generate a full moral picture using this traditional toolkit of moral theory will necessarily fail. Correspondingly, if we want to understand the harm an action inflicts, we cannot just look at the totality of harms to individuals and relationships; we also need to understand any ecological harms it does to the system as a whole and to situate harms to individuals as occurring to people who are materially and socially located within that ecology. The character and texture of the harm done to a victim can only be fully grasped when situated within an ecosystem. Was the perpetrator a trusted mentor? How do other members of the department view the perpetrator? Is the perpetrator at the center of the department’s social life or on the fringes? What is the material layout of the department? Are there multiple exits, or might the survivor easily be trapped by their assailant’s presence? Are there windows from classrooms or offices into the hallways, or might the survivor have to be constantly on edge looking for the perpetrator? Are they badgered to speak up out of a supposed duty to improve the department, or are they made to feel that speaking up will ‘ruin’ the department, and so are pressured to stay silent? Can the victim easily leave the department without drastically reshaping their priorities, obligations, and sense of self? All of this makes a difference to the victim’s ongoing relationships, comfort in department spaces, and access to departmental events going forward. Just what was disrupted and what was taken away is contextual and granular. The harms of intradepartmental sexual violation to the victim are ecological even if the victim is someone who was already quite isolated from the rest of the ecosystem, and who does not tell anyone what happened. Isolation is, essentially, an ecological position; it is a (degenerate) way of being embedded in an ecosystem. Outside of an ecosystem, one may be alone but not isolated. An isolated victim will be more cut off from friend networks and networks of trust from the start, and less able and likely less willing to access support systems and to share information. Indeed, anecdotally, it certainly seems like many perpetrators particularly target isolated victims, who are more likely to stay quiet and less likely to have an integrated support system in the department. Isolated victims may have a harder time getting testimonial uptake and have fewer social resources for recovery. Furthermore, like the victim who starts off well integrated but makes an ongoing series of choices to stay quiet, whom we discussed above, the isolated victim who doesn’t feel able to share what happened for whatever reasons will need to confront and negotiate her own silence over and over again—as she sees new people enter the department, or sees the perpetrator win accolades or gain access to new potential victims, for instance. That some members of a department are isolated (as opposed to merely introverted) is already a kind of ecological failing of the space. Healthy, self-repairing ecosystems do not strand some of their members without support. A sexual violation of an isolated victim who does not tell anyone what happened may do less 266 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms new ecological damage to the department. But this is partly because its ecosystem was already weak in some ways. Moreover, in this situation, self-repair is often difficult, because the weakness that needs to be repaired can less easily be seen to start with. While it is important to us to explain how many of the distinctive harms victims may suffer are inherently ecological, we have shown repeatedly that ecological disruptions do not only harm victims. The harm to the ecosystem from a sexual violation can have weighty consequences for many members of the department. They may find themselves with fractured friendships, eroded trust, reduced opportunities for collaboration and exchange, exhausting disorientation in their space, and extra work of various sorts. They are also denied opportunities for moral actions, such as various forms of generosity. The point is not to compare the amount of harm the victim suffers with the amount that others suffer, but rather to emphasize that sexual violation is not just a dyadic harm against a victim, but an ecological harm against an ecosystem. This ecological disruption is among the harms for which perpetrators are accountable. They are responsible for these disruptions, and for the burdens and destabilization they impose on their colleagues, and for the ways in which their actions truncate the moral possibilities open to others. We do not wish to take away from the unique harms that victims may experience. But we simply will not understand the moral impact and texture of sexual violations unless we look at the ecological harm they do, which does not reduce to the sum of the individual harms they do. And indeed, these ecological harms may be serious even if a particular victim doesn’t happen to feel traumatized or to perform having been harmed in accordance with our cultural expectations. It is not the job of victims to prove the badness of their assailants’ actions by manifesting sufficient damage. Many of the ecological harms from sexual violations are not directly dependent on the victim’s specific experience or responses. Once a departmental ecology is damaged, it cannot be fixed merely by punishing the perpetrator, nor by somehow giving restitution to the victim. Indeed, as we saw, punishing a sexual violator often comes with other kinds of ecological disruptions such as giving other people more and unexpected work and straining friendships. This doesn’t mean that violators shouldn’t be punished, but it does mean that punishment may cause further ecological disruption, which also may need to be addressed. Sequestering the victim, curtailing their ability to seek support by binding them with confidentiality agreements or informal shame, or taking away their committee members are also morally and practically poor roads to repair. More generally, simply ripping one person or two people out of an ecosystem is a terrible approach to repairing it. To heal a department and to best support those who have been harmed, disruptions must be repaired at the ecological level. In short, we urge a shift in how we think about sexual violations within ecosystems such as academic departments (or campuses, or families, or prisons, or almost anywhere they occur). We should move from thinking of a sexual violation as primarily a dyadic event in which A harms B, to understanding it as an event 267 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms that can rupture an ecosystem and damage its ecology, including its moral ecology, often profoundly impacting those within it, and harming its ability to self-repair. We can acknowledge the importance of the individual and dyadic statuses and their robust reality while also understanding that they cannot tell the whole story. The ecological lens shows us a moral layer that is not reducible to sets of monadic and dyadic moral statuses, and we hold that it is essential to grasping the full moral texture and breadth of harm from sexual violations. ACKNOWLEDGMENTS We wish to thank the audience and fellow panelists at the 2019 Eastern APA symposium session on the #metoo movement and sexual harassment for helpful feedback on an earlier version of this paper. In addition, this paper benefited from our conversations with Susan Brison, Carolina Flores, Karen Frost-Arnold, James Mattingly, and Paddy McShane. This paper also owes its existence to all the sexual harassers and abusers across academic philosophy. REFERENCES Code, Lorraine. 2006. Ecological Thinking: The Politics of Epistemic Location. Oxford University Press: New York. Dotson, Kristie. 2011. “Tracking Epistemic Violence, Tracking Practices of Silencing.” Hypatia 26(2): 236–57. Frost-Arnold, Karen. 2018. “Epistemic Justice and the Challenges of Online Moderation.” Presentation at the A Mind of One’s Own Anniversary Conference, Cambridge, MA. Harman, Elizabeth. 2019. “#metoo and the Failure to Warn Others.” Presentation at the Eastern Division meeting of the American Philosophical Association, New York. Haslanger, Sally. 2012. “Ideology, Generics, and Common Ground.” Resisting Reality: Social Construction and Social Critique. Oxford University Press: New York, 446–78. Rouse, Joseph. 2016. “Toward a New Naturalism: Niche Construction, Conceptual Normativity and Scientific Practice.” In Normativity and Naturalism in the Philosophy of the Social Sciences, edited by M. Risjord. Routledge: New York. Timpe, Kevin. 2018. “Moral Ecology, Disabilities, and Human Agency.” Res Philosophica 96(1): 17–41. 268 This content downloaded from 141.161.91.14 on Sun, 27 Sep 2020 00:04:30 UTC All use subject to https://about.jstor.org/terms