Consciousness and Morality

This chapter considers three connections between consciousness and issues in ethics: first, the relevance of consciousness for questions surrounding an entity’s moral status; second, the relevance of consciousness for questions surrounding moral responsibility for action; and third, the relevance of consciousness for the acquisition of moral knowledge. This is a disparate set of connections, prompting a question: is there anything about consciousness these connections have in common? One might expect the answer to be no. But debate in each area has thus far failed to settle just what about consciousness is so intuitively important for moral status, moral responsibility, and moral knowledge. Given this fact, it remains possible that there is some common connection of these different issues in ethics to consciousness. The chapter takes up this possibility in its conclusion.


Introduction
It is well known that the nature of consciousness is elusive, and that attempts to understand it generate problems in metaphysics, philosophy of mind, psychology, and neuroscience.Less appreciated are the important -even if still elusive -connections between consciousness and issues in ethics.In this chapter we consider three such connections.First, we consider the relevance of consciousness for questions surrounding an entity's moral status.Second, we consider the relevance of consciousness for questions surrounding moral responsibility for action.Third, we consider the relevance of consciousness for the acquisition of moral knowledge.This is a disparate set of connections, prompting a question.Is there anything about consciousness these connections have in common?One might expect the answer to be no.After all, it is frequently noted that when used without qualification, the term 'consciousness' has multiple potential referents.Perhaps the central notion of consciousness is phenomenal consciousness -the kind of consciousness associated with 'what-it-is-like-ness.' Arguably separate is Ned Block's (1995) notion of access consciousness.As Block has it, access consciousness refers to an explicitly functional, dispositional feature of psychological entities -the availability of information to the psychological mechanisms and systems responsible for things like thinking, reasoning, planning, and action control.Building on access consciousness (and arguably on phenomenal consciousness), a third usage of consciousness refers to self-consciousness: that is, roughly, the capacity to think of oneself as oneself, and to think of features of oneself as features of oneself.Finally, a fourth usage of consciousness refers what David Rosenthal has called creature consciousness.As Rosenthal has it, creature consciousness "consists in a creature's not being unconscious -that is, roughly, its not being asleep or knocked out" (2002: 406).
The distinction between phenomenal, access, creature and self-consciousness is a start, but it does not exhaust the conceptual space.Various aspects of consciousness may be individually relevant to various moral issues.For example, phenomenal consciousness is a variegated phenomenon, encompassing modalities of perceptual experience, types of cognitive and agentive experience, emotional experience, the experience of pleasure and pain, and so on.Access and self-consciousness often combine to enable a range of sophisticated psychological capacities, including various forms of attention and memory.What is more, our higher psychological capacities depend on background systems that mediate degrees of functioning of these capacities, leading to talk of modes or levels of consciousness (Bayne et al. 2016).A full assessment of the relevance of consciousness to morality must take account not only of the kinds of consciousness that exist, but of various aspects of these kinds as well.
Given the complexity these various distinctions suggest, one might plausibly expect that the disparate nature of the connections we discuss below is due to the fact that different senses or aspects of 'consciousness' are at work in each area.As we will see, this is one possibility.But debate in each area has thus far failed to settle just what about consciousness is so intuitively important for moral status, moral responsibility, and moral knowledge.Given this fact, it remains possible that there is some common connection of these different issues in ethics to consciousness.We take up this possibility in this chapter's conclusion.

Moral Status
Moral status is a property (or set of properties) an entity has in virtue of which it deserves or has the right to certain kinds of treatment.The specific content of these rights may vary depending on one's more specific moral theory, and may vary as well depending on the kind of entity is in question.But at minimum moral status is thought to generate prima facie reasons against harming or killing an entity, as well as prima facie reasons against frustrating its interests.
Why think consciousness is relevant to moral status?Consider the fraught set of moral and legal issues concerning patients who apparently lack consciousness.Following traumatic brain injury, stroke, or disease, some patients enter into a coma before awakening into a state in which they are unresponsive to external stimuli.They fail to give any overt sign of responsiveness to their own name or to visual stimuli, despite a lack of impairment in their sensory systems.These patients are said to be in persistent vegetative states (PVS).Patients may remain in such a state for decades, dependent for their survival on the provision of tube feeding.
Almost everyone (see Gomes et al. 2016), including most medical professionals, hold that these patients lack consciousness (patients who exhibit some, often inconsistent, degree of responsiveness to stimuli are held to be minimally conscious, see Giacino et al. 2002).They are also often held to lack an interest in continuing to live.Family members have sometimes petitioned courts to allow the withdrawal of tube feeding, perhaps because they believe that sustaining them in this condition is incompatible with the dignity of a human being (where possessing dignity is an inalienable status, and arguably not one that depends on being conscious at a time -or, indeed, being alive).Courts have sometimes ordered hospitals to withdraw life support in these conditions, indicating that judges often accept that PVS entails a greatly reduced interest in continuing to live.
We might hold that the dramatic reduction in an interest in living is due not to the (putative) lack of consciousness, but instead to the patients' lack of a capacity to engage in worthwhile activities (cf.Sinnott-Armstrong and Miller 2013).However, the excitement produced by recent work indicating that some (almost certainly very small) proportion of patients who are entirely unresponsive to external stimuli are in fact conscious suggests that the possession of consciousness plays a significant role in explaining our intuitions here.
In an important initial study (Owen et al. 2006), fMRI was used to compare the neural correlates of instruction following in (conscious) controls to the brain scans of PVS patients given the same instructions (the instructions given were to imagine playing tennis and to imagine visiting every room in one's house).The brain scans of one patient were indistinguishable from those of healthy controls.Owen et al. interpret this as evidence of consciousness, on the (largely implicit) grounds that consciousness is required for instruction following (see Bayne 2013 for a defence of this claim).Later work by the same team provided stronger evidence for the possession of consciousness by a different PVS patient: this patient used the instruction following paradigm to answer yes/no questions (Monti et al. 2010). 1he general reaction to these studies clearly indicated that very many people understood evidence of preserved consciousness in PVS patients as evidence that they possess a higher moral status than we had previously thought.If this intuition is justified -if the possession of consciousness makes a very large difference to the moral status of entities -important conclusions follow, not only with regard to how we ought to treat people who have suffered brain injuries but also for other questions.For instance, the moral status of non-human animals, or of artificial intelligences, might depend on whether they possess consciousness (and perhaps on the degree of consciousness they possess, if consciousness is graded).But is the intuition justified?
This depends in part on the kind or aspect of consciousness at issue.Though there has been little systematic investigation of the question, many people seem to hold that it is the possession of phenomenal consciousness, or perhaps of a capacity for phenomenal consciousness, that underwrites a serious sort of moral status (for some discussion, see Kahane and Savulescu 2009, Seager 2001, Shepherd 2016).Jeff Sebo (2015) offers an articulation of this often-implicit view that he calls sentientism about moral status: (a) If you are sentient, then you have interests (where we can understand interests in terms of present subjective motivational states), (b) if you have interests, then you are capable of being harmed (where we can understand harm in terms of interest-frustration), (c) if you are capable of being harmed, then moral agents have at least a prima facie moral duty not to harm you, and (d) if moral agents have at least a prima facie moral duty not to harm you, then you have at least a prima facie moral right, against these moral agents, not to be harmed.(2015: 9) This articulation is useful, but it raises questions.For example, why think sentience is necessary for the possession of present subjective motivational states?And why think possession of such states is sufficient for the capacity to be harmed as well as the generation of prima facie rights against such harm?One attractive line of thought appeals to the intrinsic value of at least some phenomenal mental states.Although the notion of intrinsic value remains somewhat controversial, a number of philosophers have argued that at least pleasure is intrinsically valuable, and at least pain intrinsically disvaluable (e.g., Feldman 1997, Goldstein 1989, Kahane 2009, Rachels 2000).For expository purposes, it will be useful to focus on one version of such a view.According to Guy Kahane the kind of value we ought to associate with experiences of suffering (which he takes to be a composite of pain and a kind of experienced attitude of dislike), is objective in nature.2Its badness is not dependent on any contingent attitudes towards it a subject may have.Rather, the existence of a state of suffering gives one a prima facie moral reason to end that state, whether the sufferer is oneself or someone else.
One of Kahane's arguments for this view involves an appeal to the inconceivability of a kind of hedonic inversion.Kahane has us consider the claim that, on his view of suffering's badness, "suffering's badness supervenes on the character of an agent's total experiential state" (2009: 333).To deny this would be to allow hedonic inversion: "that I could be in the same total experiential state I am in when suffering from excruciating pain, yet that this state may not be bad at all, or may even be intensely enjoyable and thus good" (334).But, as Kahane notes, this "is not a suggestion we can make sense of " (334).As Kahane is aware, the truth of his supervenience claim does not establish the further claim that suffering is intrinsically bad because of its phenomenal character.But Kahane notes, not unfairly, that this further claim looks to be the best explanation we've got for the truth of the supervenience claim.
We take it, then, that a Kahane-style view -that is, a view that ascribes objective value to at least some phenomenally conscious states -is at least defensible.In order to get from objectivism about the value of (at least some) phenomenal states to a view about moral status, one needs a further step.One needs to claim that the bearers of objective value merit certain types of treatment in virtue of the value they bear.Fortunately, this is a plausible claim.If a psychological subject bears objective value in possessing the capacity to enter into valuable mental states, then it seems that this subject merits certain types of treatment in virtue of this capacity.Further, it seems that this treatment ought to be sensitive in particular to the properties of the subject that have to do with its capacity to enter into pleasurable states.Prima facie, we ought not frustrate the subject's attempts to enter into such states, and prima facie we ought not cause the subject to move out of these states into non-or disvaluable states.
On this view, then, an entity's moral status is further explicated in terms of that entity's meriting certain kinds of treatment.And this treatment is further explicated in terms of sensitivity to the objective value that the entity bears.It emerges that an entity's moral status is closely related to, and mediated by, the kind and amount of objective value it bears.This raises questions that go beyond the scope of the present chapter.For example, what aspects of phenomenally conscious states constitute objective value?Kahane's view focuses on the disvalue of suffering, which is a composite of pain and dislike directed at the pain.Is it possible to generalize Kahane's arguments beyond suffering?Is there anything more to be said about the phenomenal state of dislike?In virtue of what does it count as dislike?Is it best thought of as a conscious desire, or emotion, or something else? Going beyond Kahane's specific view, how are we to think about the relationship between an entity's conscious mental life and the degree of moral status we attribute to it?Answering such questions seems an important task for objectivists about the value of phenomenal consciousness.Even without such answers, however, one might find objectivism an attractive way to make good on the intuition that phenomenal consciousness in some way undergirds moral status.
Attractive, but certainly not irresistible.Indeed, regarding the relationship between phenomenal consciousness and moral status, we are of two minds.One of us has suggested that the intuition that consciousness is necessary for a very serious interest in life is no mistake, but it is access consciousness and not phenomenal consciousness that plays this role (Levy & Savulescu 2009;Levy 2014b).In this view, the capacity to have future-oriented desires and plans and to care about how one's life goes is required for the kind of moral status that adult human beings typically possess and many non-humans lack, and this capacity depends on access consciousness.Information must be sufficiently available for rational thought and deliberation in order for a being to be able to have future-oriented desires or to conceive of itself as persisting in time, and this depends on the possession of access consciousness.Further, the states about which we care and constitute one's life going well are themselves (at least) largely informational states, not phenomenal states.Indeed, many of the roles that we might think can be played by phenomenal states alone (responding to the aesthetic qualities of art or of the world; having a sense of satisfaction that one's desires are satisfied; even having experiences of satisfaction) may be occupied by informational states rather than phenomenal.On this line of thought, phenomenal consciousness may underwrite an interest in having pleasant experiences and avoiding unpleasant, but it will fail to contribute much in the way of moral status.For it will not underwrite an interest in having a life.
Moving beyond discussion of access and phenomenal consciousness, a common assumption in the practical ethics literature is that an interest in having a life, or even significant interests in the future, rest on the possession of self-consciousness.According to Peter Singer, for example: Beings that are conscious, but not self-conscious, on the other hand, more nearly approximate the image of receptacles for experiences of pleasure and pain, because their preferences will be of a more immediate sort . . .They will not have desires that project their images of their own existence into the future.Their conscious states are not internally linked over time.(2011: 112) If this is right, then one might point to self-consciousness as the kind of consciousness that underwrites full moral status.Doing so would have important ramifications for our treatment of many non-human animals, and perhaps for those diagnosed as minimally conscious.For it is not obvious that the minimally conscious retain self-consciousness.One of us has argued, however, that this view of self-consciousness is misguided (Shepherd forthcoming).Arguably, self-consciousness is not on its own important for moral significance.Instead, its significance emerges along with a suite of psychological capacities that enable high-level cognitive sophistication: features like cognitive and attentional control and the coordination of perception, imagination, memory, and so on.If this is right, a satisfying account of moral status and of the psychological features that mediate its degrees might need to consider the significance of degrees of cognitive sophistication, and perhaps the significance of a closely related notion, namely, that of the levels or modes of consciousness available to a being (for related discussions, see Rowlands 2012, Varner 2012).

Moral Responsibility
As with moral status, many participants in debates about moral responsibility seem to have implicitly assumed that consciousness is necessary for an agent to be praise-or blameworthy for their actions (Sher 2009).And like the assumption that consciousness is necessary for a serious sort of moral status, this implicit assumption did not receive very much attention until relatively recently.Explicit attention has been motivated by two very different sets of considerations.On the one hand, work in neuroscience and in psychology has been widely interpreted as suggesting that if consciousness is necessary for moral responsibility, agents are never morally responsible for their actions (see Shepherd 2015 for a review of this work).Independently of this work, motivated more by thought experiments and by cases they find in literature, some philosophers have advanced positions according to which consciousness is at most a proxy for the kinds of conditions which are genuinely necessary for moral responsibility.
The most famous experimental results that have been seen as casting doubt on the capacity of actual human agents to satisfy a consciousness condition on moral responsibility stems from the groundbreaking work of Benjamin Libet.In a classic experiment, he had subjects monitor the position of a dot that travelled very rapidly around the circumference of an oscilloscope, and note its position when they felt the 'urge' to move their hand.At the same time, he measured electrical potentials in participants' brains; in particular, the ramping up of the so-called 'readiness potential' , which is known to proceed voluntary action.This paradigm allowed him to generate relatively precise timings for the occurrence of the conscious urge to move and the neural activity reliably correlated with actual movement.Libet found that the readiness potential preceded participants' reported awareness of the intention to move by around 350 ms.These results have widely, though controversially, been interpreted as demonstrating that the conscious intention to move is epiphenomenal in action initiation.If it is epiphenomenal, it seems unlikely to be able to ground moral responsibility.
Libet's experiments have been widely and devastatingly criticized on multiple grounds (see, especially, Mele 2009).Its methodological flaws aside, it is difficult to see how they might constitute a serious challenge to the claim that consciousness is a necessary condition of moral responsibility.Libet's evidence concerns (at best) consciousness of the proximal intention which actually initiates an action.Morally responsible actions at least typically take time: even spontaneous actions like rushing to push a child out of the way of an oncoming car typically take much longer than 350 ms.They are often the upshot of conscious deliberation, in which the agent is responsive to reasons.Our capacity to weigh reasons for and against our actions may ground our moral responsibility, and this capacity may require consciousness.It is hard to see how a failure to be conscious of one's reasons precisely when one forms an intention to act could make a difference to one's responsibility.
If the gap between brain events that cause our actions and our consciousness of our having committed ourselves to action were much bigger, then our moral responsibility would indeed be threatened.It would be threatened because such a gap would entail that we have greatly impaired reasons responsiveness: we could not respond to a reason that might present itself after the relevant brain event.Soon et al. (2008) produced evidence that has been interpreted as demonstrating such a gap between intention generation and conscious awareness.In this experiment, subjects chose which of two buttons to press.Patterns of activation in parietal and prefrontal cortex predicted the choice, with around 60% accuracy, an average of 7 seconds prior to the action.The researchers took this to be evidence that the "subjective experience of freedom is no more than an illusion" (Soon et al.: 543).A little reflection should be sufficient to show that this conclusion rests on a claim that is obviously false: that the agent is unresponsive to reasons that arise once the relevant neural activity has occurred.Were it true that human agents could not respond to reasons which occur less than 7 seconds prior to action, everyday activities like driving a car would be completely impossible.The neural activity in parietal and prefrontal cortex cannot be the correlates of an irrevocable decision.More likely, they are correlates of an inclination (cf.Shepherd 2015).
More persuasive reasons for rejecting the claim that consciousness is required for moral responsibility stem from within moral philosophy.Philosophers like Arpaly (2002) and Smith (2005) have presented thought experiments which they argue show that we do not need to be conscious either of the facts that make our actions morally significant or the moral significance of these facts in order to be morally responsible for our actions.Consider the familiar case of Huck Finn.As the case is commonly understood, Huck helps Jim, a fleeing slave, evade pursuit despite believing that morally he ought to turn Jim in.According to Arpaly, Huck is morally praiseworthy for his action despite failing to be conscious (under an appropriate description) of the facts that make his actions right or (a fortiori) of the fact that they are right.If consciousness of facts about our actions were to be necessary for our being praiseworthy, the only plausible candidates are surely consciousness of the facts that make the action right or consciousness of the fact that it is right; in the absence of consciousness of these facts, we should regard our disposition to attribute praise as a strong reason to think that consciousness of any particular facts is not necessary for moral responsibility, she suggests.
Smith presents a different kind of case, in which an agent is blameworthy despite her failing to be conscious of any relevant facts.In her case, an agent omits to ring a friend to wish her happy birthday because she simply forgets.Clearly, she is not conscious of the fact that it is her friend's birthday, or of the fact that there is some action she ought to perform: she has forgotten the first, and only if she recalls the first can she be aware of the second.Again, there are no mental states to which the agent has (occurent) access which could plausibly ground the agent's moral responsibility.If she is indeed blameworthy, as Smith suggests (directly blameworthy, that is, and not in virtue of her failure to take steps to ensure that she did not forget her obligation), it appears that moral responsibility does not require consciousness of any particular facts.Like the Huck Finn case, Smith's example is designed to show that access to certain kinds of information is not required for moral responsibility.
This talk of access to information might be taken to suggest that the kind of consciousness at issue in these examples is access consciousness rather than phenomenal consciousness.Certainly in the moral responsibility literature this has often been (implicitly) assumed.But one should be careful here.According to many philosophers, phenomenal consciousness plays cognitively significant roles for psychological subjects, including the role of providing access to information.We discuss ways it has been thought to do so below.
Other kinds of evidence might be seen to support the contention that consciousness -whether access or phenomenal -is not required for moral responsibility.Agents sometimes perform morally significant actions (up to and including killing others) in states in which they appear to lack consciousness (such as somnambulism).Their capacity to perform complex actions in these states, including driving a car over long distances or playing a musical instrument, might suggest that their lack of consciousness does not excuse them: if consciousness is not necessary for the kind of reasons responsiveness exemplified in successful driving of a car, why should it be necessary for moral responsibility?These cases seem to indicate that not even creature consciousness is needed for moral responsibility (Suhler and Churchland 2009).
However, we shall suggest that the conclusion that neither source of evidence -neither the thought experiments cited by philosophers like Arpaly and Smith, nor the evidence showing that agents can exhibit impressive reasons responsiveness in the absence of creature consciousness -constitutes a persuasive case that consciousness is not needed for moral responsibility.The intuitions generated by both sources of evidence depend for their power on folk conceptions of what consciousness is and what it does.If consciousness plays different roles instead of or as well as the roles ascribed to it folk psychologically, then the intuitions generated by thought experiments and the contemplation of actual cases may be off track.We shall suggest that this is indeed the case: a better understanding of the role consciousness and conscious states play makes it very plausible that it is required for moral responsibility.Levy (2014a) argues that the impressive degree of reasons responsiveness exhibited by agents who lack creature consciousness is dependent on the prior acquisition of motor and perceptual routines.The agent has acquired overlearned scripts for responding to the kinds of reasons that typically confront the driver, or the piano player.The acquisition of these scripts themselves seem to require consciousness.There is evidence that acquiring new skills involves large areas of the cortex that are plausibly correlates of consciousness, whereas once these skills are the regions activated by performance shrink significantly (Haier et al. 1992;Raichle et al. 1994).Even implicit learning, such as the unconscious extraction of patterns from data, seems to require consciousness; the subject must be conscious of the stimuli, even though she need not be conscious of the pattern she implicitly learns (Baars 2002).If it is the case that the acquisition of such scripts requires consciousness, it may be that agents are indirectly responsible for actions they cause: responsible in virtue of having (consciously) acquired them.
Consciousness, Levy argues, constitutes or enables a global (neuronal) workspace (Baars 1997;Dehaene and Naccache 2001;Dehaene, Changeux and Naccache 2011), which is required for the agent to assess actions and their predicted consequences against their own values, plans and projects.Building on Fischer and Ravizza (1998), a strong case can be mounted for the claim that the lack of a capacity to respond to these reasons appropriately entails a lack of the kind of control required for moral responsibility.Further, because the resulting behaviour is not assessed against the agents' own values and projects, it does not seem to be expressive of their quality of will.It seems to follow that given the truth of either control-based accounts or of quality of will-based accounts of moral responsibility (by the far most popular accounts of moral responsibility in the literature), agents are not responsible for the novel actions they perform in the absence of creature consciousness.
Similar considerations seem to entail that agents are not responsible for their actions in the kinds of cases that feature in the work of Arpaly, Smith, and those who follow or parallel them.Though these agents are conscious, an absence of consciousness of particular facts entails an absence of the capacity to assess the moral significance of the actions against their own values and projects.This, in turn, entails a dramatic reduction of control and of the degree to which the action is expressive of the agent's quality of will, and this in turns seems to constitute strong grounds for excusing them (but cf.Sripada 2015).On Levy's view, then, it is a necessary condition on (direct) moral responsibility that an agent have controlenabling access to the facts that give her action its moral significance.This view has been influential in the moral responsibility literature.But there may be reasons to explore conceptual space beyond that covered by Levy's view.Levy's consciousness thesis depends on a particular view of consciousness (the global workspace view) that Levy ties explicitly to access consciousness.But, first, a number of other views of consciousness are on offer, generating curiosity regarding the relevance of these views to moral responsibility.Second, most proposed views of consciousness, including the global workspace view, aim to explain -or at least contribute to an explanation of -not just access consciousness, but phenomenal consciousness as well.Third, Levy's consciousness thesis states only a necessary connection.It would thus be consistent with Levy's view if another aspect or kind of consciousness turned out to be important in some way for moral responsibility.Restricting one's attention to access consciousness is fair play, but it seems there is ample motivation for considering the relationship between phenomenal consciousness and moral responsibility.
Interestingly, work on an ancillary issue may here be relevant.Within moral epistemology a number of philosophers have advanced the view that phenomenal consciousness is important for the acquisition of moral knowledge -including, in some cases, knowledge of moral reasons for action.Although these philosophers have not to our knowledge explicitly considered the relevance of their views to moral responsibility, the connection is apparent.With this in mind, we turn to this literature next.Our initial aim will be to understand the proposed relevance of phenomenal consciousness to moral knowledge (which is an interesting issue in its own right).With such an understanding in hand, we return to the issue of moral responsibility.

Moral Knowledge
To see why phenomenal consciousness is thought to be relevant to moral knowledge, let us consider a recent proposal regarding the way conscious experiences lend epistemic force to the beliefs formed on their basis.According to Susanna Siegel, perceptual experiences are epistemically charged, where an experience's charge is a property "in virtue of which a subject's having the experience by itself redounds on the subject's rational standing" (2015: 291).For Siegel, the perceptual experiences an agent has make a pro tanto contribution to the rational standing of the agent.And the perceptual experiences an agent has make a pro tanto contribution (usually positive, sometimes negative) to beliefs formed on the basis of these experiences.
What is important about phenomenal consciousness here?Siegel gets at this question by way of another -what she calls the grounding question.In virtue of what, Siegel asks, do experiences have any epistemic charge at all ( 294)?Her answer involves appeal, initially, to the idea that experiences no less than beliefs can form a part of an epistemic subject's outlook or perspective on the world.
What might ground the rational standing of belief, and could that factor also ground the rational standing of experience?A natural idea is that what grounds the rational standing of both states is their role in the mind.Perhaps there is no further feature of belief, or routes to belief, that explains why beliefs can be evaluated as epistemically better or worse.Instead, it is their role as states that contribute to our outlook on the world.(286) Siegel does not wish to claim that every type of experience contributes to one's outlook.Rather, for Siegel it is perceptual experiences that paradigmatically play this role.And they do so in virtue of a feature of their phenomenal character.Siegel claims that perceptual experiences have a presentational phenomenal character.This is a character that 'purports to characterize how things in the external world are' (295).The thought appears to be that in possessing this presentational character, one's perceptual experiences strongly suggest that their contents are true.When one perceives a blue sky with a single cloud, one's experience presents the sky as actually blue, and as actually containing one cloud.Confronted with such an experience, one finds it very difficult to avoid believing that the sky is blue, and that there is one cloud in it.Nor, in the usual case, should one make any effort to avoid believing those things -in the normal case we have 'substantial epistemic support' for 'believing our eyes' (298).3Going beyond Siegel, then, but in our view consistent with what she says, we might say that in being presentational, perceptual experience commits the perceiver to the existence of what is presented.
So one aspect of some phenomenally conscious experiences -their presentational character -is argued to play a critical epistemic role.How does this relate to moral knowledge?If one can add to this view a further view about the contents of perceptual experience, one might think that perceptual experience plays critical roles for moral epistemology as well.We have in mind here recent proposals to the effect that "at least some moral properties can be part of the contents of perceptual experience" (Werner 2016: 298).According to proponents of moral perception, perceptual experience presents us not only with events, objects, and their properties situated in time and space in certain ways.Perceptual experience presents these events and objects as morally good or bad, right or wrong, and it does so, arguably, in a way that underwrites moral judgment, affords non-inferential moral knowledge, and allows for moral improvement (McGrath 2004, Audi 2013, Werner 2016, McBrayer 2010, Cullison 2010).Preston Werner's argument for moral perception is explicitly phenomenological, involving appeal to contrast cases.Consider Norma, a normal adult, and Pathos, who suffers from an empathic dysfunction.Werner has Norma and Pathos see some young people pouring gasoline on a cat.According to Werner, due to Pathos's empathic dysfunction, his experience will lack a certain strong phenomenological component tracked by features such as skin conductance response, elevated heart rate, and more.Werner argues that there is a phenomenal contrast between Norma's and Pathos's experience, and that "the best explanation is a difference in the perception of moral properties -Norma's experience represents the cat's burning as bad, whereas Pathos's does not" (Werner 2016: 304).
The thesis of moral perception is not without problems.First, as proponents of moral perception recognize, they are committed to a controversial view about the contents of perceptual experience.This is the view that perceptual experience contains high-level content.As Tim Bayne explicates the view, We see objects as belonging various high-level kinds, and this . . . is part and parcel of perception's phenomenal content.On this view, what it's like to see a tomato, taste a strawberry or hear a trumpet is not limited to the representation of various sensory qualities but also the representation of various 'highlevel' properties -being a tomato, a strawberry or a trumpet.(2009: 385) However, many philosophers argue that perceptual experience presents us only with low-level properties such as color, shape, movement, pitch, volume, and so on (Tye 1995, Lyons 2005).If all perceptual experience presents are low-level properties, then moving beyond these to mental states with contents as of tomatoes et al. will in some sense represent a cognitive, rather than a perceptual, achievement.
A second worry is perhaps more pressing.In an illuminating discussion, Jack McBreyer calls this worry The Looks Objection.Here is one way (among others) that McBreyer expresses it.
What bothers some about the possibility of moral perception seems to be the following: acts that are morally wrong often look just like acts that are morally permissible, where by 'look' we mean something like 'have the very same phenomenology.' When I see one person shoot another in an alley, the act could have been an act of senseless violence (and thus morally wrong) or an act of self-defense (and thus morally permissible).In either case, it would have looked the same to me.(McBryer 2010: 318) In response, McBreyer invokes analogies with other high-level facts.Hot things, for example, do not generate the very same phenomenology across cases.What is needed, McBreyer claims, are looks that normally correlate with the relevant instantiated facts: 'in order to have a perceptual experience as if X is F, things that look like this (where the 'this' picks out a certain phenomenology) are normally (but not always) F' .
It is not clear, however, whether this response adequately addresses the looks objection.It might be true that we perceive some high-level facts in the way McBreyer suggests.But what we want to know is whether we do so with respect to moral facts.Here, the proponent of moral perception must say something about the aspect of perceptual phenomenology that plays the role of McBreyer's 'this.' What kind of phenomenology is at issue, and how ought we to understand it? 4ne option is to leave moral perception at just this point in favor of a broader thesis.Consider the possibility that, even if perceptual experience alone cannot present the agent with moral properties or afford prima facie justification for beliefs with moral content, non-perceptual forms of conscious experience may do so.In particular, recent writers have emphasized the importance of intuitions -conceived as a type of phenomenally conscious experience5 -to moral judgment and moral knowledge.
In a recent paper, Elijah Chudnoff (2016) defends a view of conscious moral intuitions that does not rely on a thesis about high-level content in perception.In developing his view, Chudnoff appeals to a distinction between the content of perceptual experience and the object of perceptual awareness.As illustration, Chudnoff asks us to consider hearing a chime indicating that the oven is pre-heated to 450F.The content of this perceptual experience is as of a chime, and perhaps that the oven is ready.The object of one's perceptual awareness is the chime, but not (directly) the interior of the oven.Now, Chudnoff argues that an experience that p can be the whole basis for knowing that p "only if it both has p as part of its content and makes one aware of a truth-maker for p" (2016: 212).And he argues further that in cases of purported moral perception that one should ϕ, one's perceptual experience "can be at most a partial basis for knowing that one should ϕ in that situation" (212).This is because the truth-maker for p will include objects of which one is not perceptually aware, "such as that there is a prima facie duty of beneficence that isn't defeated by other features of the situation" (212).
Motivated by this argument, Chudnoff proposes that conscious experience makes us aware of the truth-makers for moral claims via a kind of collaboration between perception and conscious intuition.Consider a case in which an agent is confronted with a stranger in need of help, and suppose it is true that in this situation, the agent ought to help the stranger.In what Chudnoff calls low-level intuitions, an agent can have a perceptual-plusintuitional experience that she is in a position to help, and that there is a prima facie duty of benevolence.The agent can become "perceptually aware of a truth-maker for the proposition" that she is in a position to help, "intuitively aware of a truth-maker for the proposition that there is a prima facie duty of benevolence, " and "perceptually aware of a truth-maker for the proposition that there are no, or at least no apparent, defeaters of the prima facie duty of benevolence" (216).This is because in low-level intuitions, "you learn about concrete particulars by subsuming them under general truths" (216).
Importantly for Chudnoff, one need not possess background knowledge of or prior belief concerning the relevant general truths.The general truths themselves are presented as a part of the content of the intuition.And this feature of moral experience sometimes puts an agent in a position to know previously unknown moral truths, thereby enabling moral improvement.

Chudnoff comments:
[S]ometimes we see our immediate situation as both illustrating and illuminating moral reality itself.These are the moral perceptions that amount to low-level intuitions.And these are the moral perceptions that enable moral improvement.For when we confront moral reality itself, we can check our prior moral beliefs against how it is now presented to us.(217) Chudnoff's proposal offers us one way to develop a view of the epistemic significance of conscious experience that moves beyond perceptual experience alone. 6In so doing, it affords a potential way around the looks objection.A proponent of what we might call moral experience, the thesis that at least some moral properties can be part of the contents of phenomenally conscious experience, can accept that there is no stable way morally right or wrongs acts or events perceptually look.Even so, she can maintain that intuition plus perception offers the kind of phenomenological clarity and stability perception alone lacks.Intuition fills a gap by evaluatively presenting the objects often given by perception.On the view that emerges, it is not moral perception but evaluative phenomenology more generally that provides a route to moral knowledge. 7

Moral Responsibility Redux
Let us return briefly to the relationship between consciousness and moral responsibility.As we saw above, some philosophers have argued that consciousness is not required for responsible behavior, and Levy has argued against these philosophers that a kind of access consciousness is in fact necessary.But we raised the possibility that phenomenal consciousness could be relevant as well.In this connection, the discussion is section four reveals two ways one might develop a phenomenal consciousness/moral responsibility connection.
The first is fairly straightforward.If aspects of perceptual and intuitional experience put one in contact with one's reasons for action, then these experiences may play a role similar to the role Levy argues access consciousness plays.Namely, these states may enable reasons-responsive action.Whether this line is ultimately persuasive will depend on the reasons given for favoring phenomenal (as opposed to, or in addition to) access consciousness with respect to the function of putting agents in contact with their reasons.
While the first route will likely appeal to so-called reasons-responsive theorists, the second draws on the socalled quality of will views of moral responsibility.In an influential article, Angela Smith argues that "what makes an attitude "ours" in the sense relevant to questions of responsibility and moral assessment is not that we have voluntarily chosen it or that we have voluntary control over it, but that it reflects our own evaluative judgments or appraisals" (2005: 237).Smith herself does not emphasize phenomenal consciousness (indeed as Levy reads her she is an opponent of the importance of consciousness).Even so, implicit in Smith's case are certain patterns of conscious experiences a subject undergoes -what she notices (and fails to notice), the thoughts, evaluations, and emotions that occur to her as she interacts with others.One attracted to Smith's quality-of-will view might develop a connection with phenomenal consciousness as follows.Recall Siegel's argument that perceptual experiences can themselves redound to one's standing as a rational agent.Analogously, one might argue that an agent's moral experience -the moral properties presented in her perceptual plus intuitional states -redound to the agent's moral standing in such a way as to render her the apt target of morally reactive attitudes.Whether this line is ultimately persuasive will depend in part on the reasons given for taking the tokening of morally salient phenomenal states to redound in just this way.
Certainly the role of phenomenal consciousness in reasons-responsive accounts of responsibility, or in quality of will accounts requires further development to be deemed important to such accounts.8But given the evident connections between work on consciousness and moral knowledge and work on consciousness and moral responsibility, it seems developmental efforts in this direction will prove fruitful.

Conclusion
In conclusion, let us return to a question we broached in this chapter's introduction.We have discussed the potential importance of consciousness for moral status, moral knowledge, and moral responsibility.Is there anything about consciousness these connections share?
As most of the views we have discussed remain under dispute, we take it to be an open possibility that there is not.Nonetheless there is a thread running through this chapter that merits emphasis.In our discussion of moral status we appealed to phenomenal states that possessed intrinsic value, the most obvious of these being painful and pleasurable experiences.In our discussion of moral knowledge we appealed to perceptual/intuitional/ emotional phenomenal states that present aspects of the world as good or bad, or as obligating or permitting courses of action.And in our discussion of moral responsibility we (all too briefly) considered views that might appeal to the epistemic and moral qualities of such states to undergird more traditional views of moral responsibility.One thing all of the above discussions share, then, is an appeal to phenomenal states that are broadly evaluative in nature.And this leads to the suggestion that it is evaluative phenomenology that is, in the first instance, intimately connected to morality.This suggestion might afford insight into why -other than a perhaps unhealthy obsession with tidiness -one might care whether the connections between consciousness and morality share a common feature.If evaluative phenomenology is a key to understanding the relevance of consciousness to morality, then work on evaluative phenomenology emerges as important not only for the philosophy of mind, but for ethics as well.Ethicists may have reason to pay closer attention to phenomenology than they have hitherto done.
In this connection, however, readers will have noticed that an anti-phenomenalist thread runs through this chapter as well.In our discussion of moral status we considered the view that it is access consciousness rather than phenomenal consciousness that undergirds high degrees of moral status.In our discussion of moral responsibility we saw that the most influential view in the region is Levy's which explicitly downgrades the importance of phenomenal consciousness in favour of functional aspects of cognition.And in our discussion of moral knowledge we noted problems not only for the thesis of moral perception, but also for views that would give phenomenal consciousness more broadly a central epistemic role.
With respect to the connections between consciousness and morality, there is little philosophical consensus.Even so, relationships between aspects of consciousness and aspects of morality appear to constitute fruitful and important areas for further research.