Faceless harms

Introduction

Zemiology often speaks of the harms ‘faced’ by different individuals and communities (see Canning and Tombs, 2021; Davis and White, 2023; Wright, 2025). But can we always be said to ‘face’ harms? Are harms always experienced as harms? In answering these questions, I’m going to claim, perhaps somewhat controversially, that zemiology has something of an experiential bias. Or if we wanted to get a little speculative realist, we might say that many (though, not all) zemiological accounts of harm are characterised by correlationism: the ‘idea according to which we only ever have access to the correlation between thinking and being, and never to either term considered apart from the other’ (Meillassoux, 2008: 5). In accounts characterised by correlationism, for something to represent a harm, we must become aware of it – our intentionality must be directed towards the negatively felt impact on our wellbeing. We must perceive a negative impact to our wellbeing. With the exception of events – such as workplace accidents (see Tombs and Whyte, 2007) – that cause immediate death, we can only parse harms that appear to us. Harms are only legible insofar as they are experienced, perceived, narrated, or made meaningful by those whose lives they negatively impact. This focus on recognised harms is completely understandable – the lion’s share of social harms that befall us are consciously experienced.

Yet many harms operate precisely by never surfacing phenomenologically qua harms. They make lives worse without generating an identifiable moment of injury. They represent harms without suffering ¹ : My life is made worse than it otherwise would have been, but I experience no pain, anguish or loss because the alternative never appears. Just as we can distinguish between phenomenological and ontological atmospheres (Lundberg, 2022), so too might we distinguish between phenomenological and ontological harm. Phenomenological harm captures our experience of harms insofar as they appear as harmful phenomena to us. Of concern in phenomenological harm is not simply ‘what’ is suffered, but rather cases where this suffering appears to the senses. Ontological harm, by contrast, structures a life’s possibilities, such that our lives do not go as well as they could have. Phenomenological harms are, therefore, indexed to subjective wellbeing, ontological harms to objective wellbeing. Put differently, phenomenological harms negatively impact our subjective wellbeing: our personal evaluations on how well our lives are going (see Billingham and Irwin-Rogers, 2022).

Phenomenological harm is always grounded in ontological harm, but not all ontological harm is experienced directly as phenomenological harm. ² All ontological harms are, in a sense, experienced phenomenologically in the form of lives not as good as they might have been. No ontological harm is an island, cut off from the remainder of our life’s trajectory. Their ripple effects often lower our subjective wellbeing, but in many cases, we cannot trace these negative affective states to these anonymous turning points in our trajectories. Often, these unrecognised pivots in the causal fabric of our wellbeing take the form of preventive harm: instances in which we are actively prevented from receiving a benefit (Feit, 2019). Take Bradley’s (2009: 71) following example:

Unbeknownst to me, Ned has left me tickets to a baseball game, and Hud has subsequently stolen them from my mailbox. I would have been much happier going to the game than staying at home, but I never find out about the tickets. . .

My life does not go as well as it could not just because I experience events that negatively impact my subjective wellbeing, but also because I am affected by a range of causal reverberations whose negative repercussions on my wellbeing I never see, hear, feel, smell or taste. I’m not haunted by these absences, for unlike the kinds of present absences that oogity boogity their way through hauntological accounts of life’s metaphorical ghosts (Fiddler et al., 2022; Fisher, 2014), I never catch a glimpse of them curtailing who I might otherwise be. Guillotining potentialities. Guillotining potential mes.

I term such cases faceless harms – faceless because we never recognise or ‘face’ them, but they nonetheless make our lives worse than they otherwise would have been. We do not confront faceless harms as injurious phenomena. ³ Doors close for me unbeknownst to me. I want to argue that faceless harms represent an important though under-acknowledged category of social injury for criminologists and zemiologists. I make this case through, firstly, distinguishing faceless harms from two other more commonly examined forms of social harm: latent harms, and epistemic harms. ⁴ I then show the currency of faceless harms to criminologists and zemiologists through detailing how they emerge in two case studies: doxxing and image-based sexual abuse, and allocative harms emerging from algorithmic social sorting systems. I conclude by detailing how the existence of such harms without suffering presupposes, at the least, an attitude-independent dimension of wellbeing, noting that several accounts of social harm already meet this condition (namely, Billingham and Irwin-Rogers, 2022; Canning and Tombs, 2021; Pemberton, 2016; Raymen, 2022; Wood, 2025; Yar, 2012).

Latent versus faceless harms

Though they bear some passing resemblances, faceless harms should not be confused with latent harms. When there is a period of latency between the event that harms us and the manifestation of that harm, we have experienced a latent harm (Anderson et al., 2026; Wood, 2025). Exposure to environmental contaminants such as carcinogenic pesticides, represents perhaps the paradigmatic example of such latent harms (Revesz, 1999). The harms wrought from exposure to such chemicals may remain dormant for many years – the cancer that finally metastasises after a long period of latency. Harm and harming are separated by an often long temporal disjuncture; the individual suffers the harm well after the event that harmed them or events that collectively harmed them (see Feit, 2015; Parfit, 1987). To be more precise, victims of latent harms only experience a negative impact on their wellbeing after this long temporal disjuncture, for we could say that their wellbeing – all things considered – has been negatively impacted as soon as their exposure to carcinogenic chemicals generates virtual – but inexorable and undetected – potentialities that will lead to their future illness.

Faceless harms need not entail such a temporal disjuncture between harming and harm (though they often do). When, unbeknownst to me, an algorithmic system rules me out of receiving some benefit, my overall lifetime wellbeing is immediately lowered from this surreptitious allocative harm (Shelby et al., 2023). What distinguishes faceless harms from latent harms, then, is that we remain unaware of them. Latent harms eventually announce themselves to us; faceless harms do not. We experience latent harms as harms when they finally manifest; when the cells mutated by our exposure to asbestos finally metastasise. Faceless harms negatively impact our wellbeing while we remain entirely ignorant of them, and entirely ignorant to how our lives would go better had we not been unwitting victims to them.

Epistemic versus faceless harms

Does learning of these harms that occur behind my back, even when I can do nothing to thwart the gravity of their impact on my wellbeing, make me better off? Might even learning of these lost opportunities and truncated trajectories not harm me further? Have I, in short, suffered an epistemic harm, where I as ‘agent qua knower [am made] worse off’? (Dunne and Kotsonis, 2025: 426). Epistemic harms, as Wright (2025: 2) details, ‘can be understood as an outcome of epistemic injustice, which renders a person, or a group of people, disadvantaged in and through practices pertaining to knowing’. What is the relationship, then, between faceless harms and epistemic harms? And do all faceless harms make me worse off as a knower?

This, I argue, will be contingent on whether we are further disadvantaged by our lack of knowledge about the harm to one or more dimensions of our wellbeing. We are further disadvantaged when our knowledge of a harm such as, say, reputational damage, enables us to limit the gravity ⁵ of this damage, and live a better life than we would have had we remained in ignorance of it. I suffer an epistemic harm when my inability to recognise that I have suffered a harm thwarts my ability to address this harm. But there may also be instances where, given the ineliminability of the unrecognised harm I have suffered, learning of its existence only serves to further harm me. If there are cases where ‘ignorance is bliss’, where, ‘a reasonable person would pay not to have certain information’ (Kadane et al., 2008: 30), then surely at least some faceless harms fall into this category. Faceless harms are, therefore, not equivalent to epistemic harms, though epistemic harm can supervene on faceless harms (see Anderson and Wood, 2022).

But this all raises the question: am I harmed as a knower, or is it rather my lack of access to certain knowledge that is extrinsically harmful to my wellbeing? To say that we can be harmed as knowers would require that we take knowledge to be an intrinsic rather than instrumental good for us; knowledge is good in and of itself, rather than on account of what it can do for us. There are certainly accounts of wellbeing that take knowledge as an intrinsic good for humans, but I think in many instances, the harm that is alluded to in speaking of epistemic harms is not harm to the individual qua knower, but rather harm to other dimensions of their wellbeing on account of their ignorance about things undermining their interests. The issue is not epistemic harm to the individual qua knower, but rather, to use Gregson’s (2025) terminology, the issue of how ignorance can sustain harm.

In many instances, my ignorance about a faceless harm I have experienced will lead to its sustainment: it will increase the gravity and the duration of the negative impact it has on my wellbeing. But there are two other options. In some cases, my knowledge of a faceless harm won’t move the needle in one direction or the other – it will neither improve or further damage my net lifetime wellbeing. And in some other cases, my knowledge of a faceless harm will only serve to further damage my net lifetime wellbeing – I am not able to rectify the harm’s gravity, and learning of it only serves to add psychological distress to the wound – ignorance would have been bliss.

Two examples

Two examples that have recently received attention from criminologists and zemiologists give a good sense of the nature and currency of these faceless harms: the rippling reputational harms associated with delegitimising doxxing and image-based abuse, and allocative harms stemming from opaque algorithmic systems (Elias, 2025). Both implicate digital technologies, but digital technologies should not be taken as a necessary ingredient of faceless harms.

The challenge of faceless harms

Another way to the understand faceless harms is through recourse to the distinction that philosophers of harm – and particularly proponents of causal accounts of harming – occasionally make between harm and harming (Gardner, 2021; Unruh, 2023). Harming entails ‘what it is for an action or event to inflict harm upon someone’ (or something; Hanser, 2019: 859). Harm, by contrast, entails negative impacts to someone or something’s wellbeing that are caused by harming (Johansson and Risberg, 2022).

In all of the examples I’ve marshalled so far, the victim of a harm does not experience the harming event whatsoever. But we can also identify a secondary more ubiquitous form of faceless harm. In the second form, I experience the harming event, but not as a harm – I do not recognise the event’s harmfulness to my wellbeing. If we can befall faceless harms through our own actions, then faceless harms are truly ubiquitous. We regularly act against our own interests, all the while remaining ignorant of how our actions have undermined our interests. We are the victims of our own counter-intentional harms (Wood et al., 2025). Indeed, the faceless harms we experience through these counter-intentional actions are one of the key ways that ideologies and Lukes’ (2005) third dimension of power negatively impact our wellbeing. One doesn’t have to subscribe to a Marxian ‘false consciousness’ view of ideology to agree that ideology can, and often does, make us act against our interests.

It is all too easy to proceed as if harms must appear – be felt, narrated, recognised by those whose lives are affected by them – to count as harms in any analytically meaningful sense. Yet many harms do their work without directly surfacing as such phenomenological harms. They worsen lives without producing an identifiable moment of injury, and without giving victims a coherent object to name, contest, or seek redress for. Their existence speaks to the need for accounts of harm that can capture injuries to subjective and/or objective wellbeing. Zemiology is far from devoid of such accounts. Though I’ve argued that zemiological research often centres phenomenological harm, accounts of social harm’s ontology developed by zemiologists and zemiology-adjacent scholars often do provide a theoretical foundation for capturing harms without suffering. Indeed, accounts of social harm’s ontology have tended to presuppose objective-list accounts of wellbeing – Yar (2012), Pemberton (2016), Billingham and Irwin-Rogers (2022), me (Wood, 2025) – or perfectionist/developmentalist accounts of wellbeing – Canning and Tombs (2021), Raymen (2022) – that do account for an attitude-independent dimension of wellbeing. These accounts do not necessarily pick out faceless harms from other forms of injury, however, their ontological purview is broad enough to capture these harms without suffering.

Epistemically, faceless harms are troubling because they evade problem recognition and therefore evade complaint, mobilisation, and narrativisation. As Presser (2013: 15) emphasises, ‘stories—or narratives—are a crucial type of rhetorical device for constructing who we are and what we intend to do’. Narratives are, at an individual level, critical to help-seeking behaviour, which commonly begins with the cognitive task of recognising and defining a problem (Liang et al., 2005). Faceless harms interrupt this sequence – no recognised problem, no help-seeking. They often easily fold into business as usual, particularly where they arise as emergent effects of systems rather than individual actions. They are unstoried causal reverberations that, remaining unnoticed, remain unsaid, unaddressed (see Presser, 2022).

Deleuze and Guattari’s (1987) account of faciality is instructive here. ⁶ Faciality is not ‘about faces’ in a merely anatomical or interpersonal sense; it names how power sorts, judges, and normalises by making bodies and environments readable. For Deleuze and Guattari (1987), the face operates as a regime that organises signification and subjectivity, producing recognisable subjects that institutions can read and govern (Deleuze and Guattari, 1987: 167–191). Harm can be sustained through an analogous demand: show us the moment, show us the victim, show us the harm someone faced. Often, faceless harms fail this test, and so they persist. Faciality therefore clarifies a structural paradox: institutions often require a stable ‘face’ of harm (legible victim, recordable event, pin-pointable cause) before they can act; yet contemporary harm-production frequently works by defacing – through removing the experiential legibility of events that leave us worse off. Just as Deleuze and Guattari (1987: 171) exhorted us to ‘dismantle the face’, we need to dismantle the harms we phenomenologically face into their often faceless progenitors.