Perception, Intuition, and Moral Judgment

 
 
 
 
 

Moral Perception 
 
 
 

Based on the Soochow Lectures, 2011 

Robert Audi

University of Notre Dame 
 
 
 

Draft of January 8, 2012 

Not for General Circulation 
 
 
 

 

Preliminary Contents 

(Pagination with 1.5 spacing; word count: roughly 47,000 words) 

Preface (3-5) 

Acknowledgments (6) 

Introduction (pp. 7-10) 

Chapter 1

Perception:  Sensory, Conceptual, and Cognitive Elements

(11-29) 

Chapter 2

Moral Perception: Causal, Phenomenological, and Epistemological Dimensions (30-45) 

Chapter 3

Perception and Moral Knowledge

(46-58) 

Chapter 4

Perceptual Grounds, Moral Intuitions, and Ethical judgment

(59-85) 

Chapter 5

Moral Perception, Aesthetic Perception, and Intuitive Judgment

(86-99) 

Chapter 6

Emotion and Intuition as Sources Moral Judgment

(100-116) 

Chapter 7

The Place of Emotion and Moral Intuition in Normative Ethics

(117-136) 

Conclusion 

Index (p. …) 
Preface 

      This book was prepared for the Soochow Lectures in Philosophy and, in an earlier, shorter version, presented at Soochow University in Taipei, in March 2011.   The Soochow Series is intended to stand as a source of continuing contributions to philosophy and a setting for establishing fruitful connections between philosophers in Taiwan and many other philosophers worldwide.  Even beyond those connections, the Soochow Lecture series aims at stimulating interchanges of ideas between philosophers working in Chinese and those working in other languages.  With this context in mind, I have selected a topic that spans three very broad areas and should engage the interests of philosophers and others working in any of these fields:  ethics, both theoretical and practical; epistemology, conceived as the theory of knowledge and justification; and moral psychology, conceived as inquiry in the areas of overlap between philosophy of mind and ethical theory.  The book is intended to engage the interests of philosophers and other thinkers working in any of these areas.

      Every chapter has been greatly expanded since its partial presentation in Taipei, but I have retained the concreteness and multiple examples needed for comprehension by an audience that did not have the text.  My aim continues to be a high level of readability so that students as well as professional readers can readily follow the argument.  Copious notes are included for those wanting certain qualifications or references to relevant literature, but the notes are not needed for comprehension of the main points.

      The main points in the book may well be comprehensible to readers who merely consult Chapter 1—which is perhaps the most complex—at certain points. Chapter 1 is required for any defense and full-scale understanding of Chapter 2 and later chapters, but they are sufficiently clarified by examples of their main points to be intelligible apart from it.  Readers interested in ethics but not epistemology, then, might try reading the introduction and proceeding directly to Chapter 2. To some extent, moreover, any chapter, but especially the last, may be understood by itself. It may certainly be taught on its own by an instructor with an understanding of the work as a whole.

        It is appropriate here to say something about the contribution of the book in each of the fields within its scope. Perception is a basic topic. My account of it draws on years of work on the subject but is presented here in a self-contained treatment and with footnotes indicating much relevant literature.   This book will show how the account can accommodate the idea that perception occurs in the moral domain.  Much will be said about perception in Chapter 1, with many examples drawn from the domain of vision.  On the basis of the account of perception outlined there, Chapter 2 presents a theory of moral perception and a case for countenancing it as an important element in moral life.  Perception is not only basic for success in both everyday life and scientific inquiry.  It is essential for moral knowledge and crucial for cross-cultural understanding.  It is the common root that nourishes our cognitive structure; it anchors that structure to the grounds from which truth emerges; and it sustains the vast and various superstructures we built from the foundational materials that perception provides.  If there is moral perception, and if some of our sound moral beliefs suitably rest on it, then there is moral knowledge.

      In stressing the importance of perception for understanding in ethics, I am not suggesting—and do not believe—that all moral knowledge is perceptual.  This point raises the question of how non-perceptual moral knowledge is related to perception.  I am especially interested in intuitive moral knowledge—whose existence I have argued for in The Good in the Right: A Theory of Intuition and Intrinsic Value (Princeton University Press, 2004) and in the ways in which intuitive moral knowledge is connected with perception and emotion. The role of moral imagination in relation to all three of these—perception, intuition, and emotion—is also a topic important in this inquiry.

      A major concern of the book is the issue of objectivity in ethics.  From that point of view, the influence of emotion on moral cognition is a major topic. A common assumption is that emotions bias us and undermine objectivity.  Doubtless they can do this, but I will show how they can also have evidential value and thereby contribute to our moral knowledge and, more broadly, to our ethical sensitivity and the refinement and justification of various important elements in our moral outlook.

      The book does not challenge moral skepticism directly, but if the account of perception and its relation to moral judgment is sound, the common-sense view that we have a good deal of knowledge in ethical matters gains support.  Even those who reject that view will be able to see how perception can at least provide some degree of rational support for moral judgment.  Intuition can be similarly seen to have evidential value, even if it is not taken to constitute knowledge or to provide evidence sufficient for knowledge.  And emotion can be seen—at least in people of maturity and discernment—not as a threat to objectivity but, under many conditions, as a source of evidence for moral judgment.

 

      

       

      Acknowledgments 

      To be expanded, but including Bagnoli, Dancy, Hagaman, Kennedy, McBrayer, Vyrynen, Witcomb, the Brackenridge Symposium, a seminar at Notre Dame, readers for the Press…

       With one exception, the relation of this book to my earlier work is indicated by various of the footnotes, but I should say that parts of Chapters 1 and 2 derive from my “Moral Perception and Moral Knowledge,” Proceedings of the Aristotelian Society Supplementary Volume 84 (2010),…. 
 
 
 
 
 

Introduction 
 
 

      A perennial quest of philosophy is to construct an adequate conception of the human person and to frame adequate standards for human conduct.  In the domain of ethics, standards of interpersonal conduct are central.  Ethical conduct is essential for human civilization, and in our globalized world, with its increasing international interdependence, nothing is more important than universal adherence to sound ethical standards.  Is there any moral knowledge that can serve as a basis for justifying such standards?  That is one of the broad questions motivating this book.

       With the successes and intellectual prominence of modern science, philosophers and many others who think about the status of ethics have been concerned with the apparent disparity between our ways of arriving at moral judgments and our ways of arriving at beliefs and judgments by using scientific methods.  A great many contemporary academics and others maintain—or simply presuppose—that if we have any moral knowledge, that knowledge must be broadly empirical and ultimately amenable to scientific confirmation.  This view is implicit in the most common kind of contemporary naturalism.  Much could be said about what counts as naturalism, and Chapter 2 will explore the extent to which my theory of moral perception may be considered naturalistic.  For our purposes, it is sufficient to bear in mind a wide conception.  In very broad terms, we might think of naturalism as the position that, first, nature—conceived as the physical universe—is all there is; second, the only basic truths are truths of nature; and, third, the only substantive knowledge is of natural facts.¹  Science, of course, is taken by naturalists to be the highest authority concerning what the truths of nature are.

      Naturalism as most commonly conceived contrasts not only with supernaturalistic theism but also with rationalism in epistemology.  In outline, rationalism is the view that the proper use of reason, independently of confirmation from sense experience, yields substantive knowledge (as opposed to knowledge of logical or analytic propositions). A robust rationalism extends to including certain sorts of moral knowledge as among the substantive kinds that may be described as a priori.²  Such knowledge, though not unscientific, is non-scientific. There is, however, a point of important agreement between rationalists and naturalists, even those naturalists who are empiricists.  It is that perception is a major source of possible knowledge of its objects and that if there is genuine knowledge of the physical universe, that knowledge depends on perception.

      My main project in this book is to show how perception figures in giving us moral knowledge and how moral perception is connected with intuition and emotion.  In showing this, I will combat stereotypes regarding both intuition and emotion, especially the view that they are either outside the rational order or tainted by irrationality.³  In doing this, I will at many points criticize one or another form of intellectualism, which is roughly the tendency to treat perception, cognition (especially belief formation), and rationality itself as dependent on intellectual operations such as inference, reasoning processes, and analysis.  Rationality is not intellectuality, and intellectual activity is not entailed by rationality in belief, action, judgment, or other elements that may be appraised in the dimensions or truth or rationality.

      More broadly still, I hope to lay out both major elements of a moral philosophy that only an epistemologist can develop and points of epistemology that only a moral philosopher would make.  I try to do this from the point of view of a philosophy of mind that makes it possible to understand human agency and cognition with minimal posits:  roughly, without burdening the mental life of rational persons—and doubtless our brains—any more than necessary for understanding the data.  Here I join forces with many colleagues in neuroscience and with many philosophers with views naturalistic than mine. In this spirit, and from the point of view of both epistemology and philosophy of mind, I aim at clarifying both the nature of intuition and emotion and their evidential role in yielding justified moral judgments and moral knowledge.  In doing this, especially in Chapters 4-6, which concern the ethically important interconnections among perception, intuition, and emotion, I will clarify the normative ethics that seems most plausible to me.

      If this overall project succeeds in the way I intend, it provides a foundation for affirming the possibility of moral knowledge that is, on the one hand, based on perception and hence empirical and, on the other hand, belongs to a framework of a priori moral principles that are not empirical.  Moral philosophy spans both the empirical and the a priori domains, and I shall argue that it does so in a way that makes possible both objective moral judgments and cross-cultural communication in ethics.

 

            

Chapter 1 

Perception:

Sensory, Conceptual, and Cognitive Dimensions 
 

      Perception is central in epistemology, and the concept of perception is among the most important in philosophy.  No one doubts that perception is essential for human knowledge, and we trust its deliverances.  If there is dispute about whether someone pointed a laser beam at an airplane in flight, honest testimony that one saw the act weighs heavily and normally settles the dispute.  It is even common for people to go so far as to say that seeing is believing. The prominence of this adage indicates the high importance that visual perception is taken to have for grounding belief and knowledge.  The sense of touch is also highly trusted.  If I feel my wallet in my pocket as I move through a crowd, I am confident that it is in fact there.  Indeed, tactile perception may have even greater psychological authority than any of the other senses.  If, looking at my wallet in my hand, I suddenly ceased to see it but could feel it in my grip, I would likely think that my vision is at fault rather than my sense of touch.  Whatever we might conclude about the relative power of different modes of perception over cognition, the clear and steadfast deliverances of visual perception—which will be the main kind of perception considered here—are not easily overridden.

      If the psychological authority of perception—its power to compel belief under varying conditions—is not in general contested, its epistemic authority—its power to yield knowledge and justified belief—is often taken to be limited to certain realms and to hold mainly for descriptive rather than normative propositions.  Paradigms of the former are propositions ascribing observable properties such as color and shape, to macroscopic objects. Related to these propositions are those ascribing to objects of scientific concern properties of the kinds that essentially figure in explanations in the natural sciences.  Paradigms of the normative propositions are those ascribing obligations to persons, wrongness to actions, or intrinsic goodness or badness to states of affairs.

      There are many people, in and outside philosophy, who, taking descriptive propositions to exhaust what is perceptually knowable, think that perception does not yield moral knowledge.  Commonly, such moral skeptics think that perception bears on settling moral disagreements only when they turn on differences over “facts” such as those observable in the scientific study of behavior.  Assessment of this skeptical view about the status of ethics requires both an account of perception and an understanding of the nature and basis of moral judgments.  The former topics will be central in this chapter, the latter in the next. 

 

I. The Major Types of Perception

      The term ‘perception’ is quite abstract and has long presented a challenge to philosophical analysis.  But there is no controversy about whether paradigms of perception include certain experiences in the five ordinary sensory modes:  seeing, hearing, touching, tasting, and smelling.  We should not consider this list of perceptual modes exhaustive.  For one thing, it does not include proprioception, for instance certain kinds of direct awareness of bodily processes, which, like an awareness of tensing muscles, might be considered a kind of inner perception.  There are also considerable differences among the kinds of perception the list represents; but at least in philosophy, seeing is the favorite paradigm of perception.  My aim will be to make points about perception that have wide application, but it will simplify discussion to use mainly visual examples and, of course, auditory cases involving reception of speech acts.

   We should note immediately that there is seeing in the intellectual, apprehensional sense, as well as perceptual seeing.  Seeing that realism in ethics is controversial is an important element in understanding current discussions in moral philosophy, but it is not perceptual seeing.  As this illustrates, one liability of focusing on seeing is conflating perceptual with intellective seeing.  The distinction may not be sharp, but it is clear enough to enable us to distinguish intellectual moral seeing (such as seeing that we ought not to cheat others) from the apparent moral perceptions that concern me.  Another case of seeing is “seeing in the mind’s eye,” which might be considered imaginational seeing.  That is best treated as an instance of visual imagination, which is morally significant, in ways I will illustrate in Chapters 6 and 7.  It is possible even given blindness—and would be possible even for a Cartesian non-embodied mind. It is not perceptual.

         It will help to begin with three main cases of perceptual seeing:  (1) seeing an object, such as a log in a fireplace; (2) seeing an object to have a property, say seeing a log to be smoldering; and (3) seeing that some “observational” proposition holds, for instance that a face is tanned.  I call these simple perception, attributive perception, and propositional perception.⁴  All three cases manifest the veridicality—the “factivity”—of perception.

   Let me illustrate.  First, if I see a face, there is in fact a face that I see; this is referential factivity.  The point is not that perception itself makes a reference, though the rationale for saying so is plain in such cases; but in specifying what someone perceived we must make a reference, and it will be to something real.  Second—to proceed to a different way of ascribing a perception—if I see a face to be bearded, there is a face that I in fact see and it is bearded. This illustrates both referential and attributive factivity—predicative factivity, in a terminology I take to be roughly equivalent. There must be something I perceive and, if I see it to be bearded (and so I attribute—or at least my perceptual system attributes—beardedness to it), then it is bearded.  Thirdly, suppose I see that that face is bearded. Then that proposition is true; this case illustrates propositional factivity: in effect, I see the truth of a proposition (loosely, I see that a proposition is true).  Normally, I also know the proposition in question to be true.

   These points about perception are apparently conceptual.  Reflection will show that someone’s using perceptual language in a way that flies in their face or presupposes their falsehood would be strong prima facie evidence of misunderstanding perceptual concepts.  Granted, someone may be said to “see ghosts,” but such special uses of ‘see’ normally indicate ascriptions of visual experience without an external object and may be set aside here.

   Very commonly, we not only see things but see them as something definite.  One usually sees a ship not as a mere hulk—as one might if one were next to it in a rowboat looking at just one side—but as a many-faceted conveyance with a base and superstructure.  Conceptually, seeing as is two-dimensional:  there is the thing seen and the set of properties it is seen as having.  From an epistemological point of view, seeing as is a hybrid.  This is because, at least in perceptual cases (the kind that concern us), seeing as is veridical and referential as to the object that is seen—here it is like simple seeing—but it is neither necessarily veridical nor referentially transparent as to what is, as it were, visually predicated.  In the first, referential dimension, then, it is like simple seeing; in the second, attributive dimension, it differs both from simple seeing and from attributive seeing, which is veridical in the predicative dimension. To see this difference, consider a child’s seeing a stuffed hound as a lion.  This entails that there is a stuffed hound seen, but not that a lion is seen.  The as-clause may be followed by a term for something merely imagined.

   Moreover, seeing as is also not transparent; i.e., it may fail to apply even if we substitute a true description for one indicating the aspect of the object the perceiver focuses on.  Someone who sees a table top as circular need not see it as having the shape of a figure whose circumference is pi times its diameter, even though these terms necessarily designate the same things.⁵  The position of the expression following ‘as’, then, is neither necessarily factive (as shown by the stuffed hound case) nor referentially transparent (as indicated by the table case).⁶

   Much more could be said about simple, attributive, and propositional perception, but the points made so far will enable us to proceed to other aspects of perception and then, in Chapter 2, to the nature and status of moral perception.

    

II. The Phenomenology and Ontology of Perception

      For both epistemological and metaphysical reasons, it is important to examine the phenomenology of perception.  We may plausibly assume that perception is experiential.  To see (or have some other perception) is to experience something, and the experience is distinctively qualitative.  To illustrate the phenomenal sense, there is something it is like to see a yellow grapefruit and something it is like to feel its surface. Here ‘like’ has its phenomenal sense, not its comparative sense.  A child can know what it is like to feel the surface of a grapefruit even if, before this occasion, the child has never felt anything else with a similar surface.⁷  Knowing what something is like, in the phenomenal sense, is a matter of the content of consciousness understood as object(s) of awareness; it does not require knowing what something else is like and knowing the former to share some property with the latter.

 

The representational element in perception

      Perception is not only experiential but also in some sense “representational.”  If it is indeed factive in the way illustrated—implying certain truths—the natural assumption would be that it represents its object:  the thing that is (e.g.) seen and (normally) seen to have some property. The perceiver may also see that the object has a property.  But seeing that something is so entails believing that it is so, and simple perception, at least, apparently does not entail having any belief about the object perceived.

      The representationality of perception is confirmed by the functional dependence—a kind of discriminative dependence—of the phenomenal element in perception upon the object perceived.  My perceptual experience depends on the perceived object and varies systematically with certain changes in that object so long as I continue to perceive it. Normally, if the tree that I see is windblown, my visual impression varies with the waving of its branches; if I am perceptually conscious of my chair and it vibrates as a passing train rumbles by, I have a tactile sense of vibrations; in thickening smoke, my olfactory sensations intensify; and so forth. This does not imply that we see observable properties of objects by seeing corresponding phenomenal properties.  The relation is not instrumental, and the relevant implication is that seeing the physical properties entails instantiating, not seeing, certain of the phenomenal ones.

         Given the representationality of perception and the typically rich information it provides to the perceiver, it has become common to speak of perceptual content.  This terminology needs clarification.  If you see a deer, is the perceived object, the animal itself the main content of your experience, or is that content simply a representation of the animal?  Are the properties you see the animal to have the content of your experience?  Does the proposition that there is a deer in the field (if you see that there is one there) belong to the content?  Do all three of these elements taken together constitute the content?  Perhaps we might say that all the properties phenomenally represented in a perceptual experience constitute its content, given the intuitive sense in which the content of something is what is in it, and given that what is most clearly “in” an experience is the properties, such as color, shape, and sound, that one is (phenomenally, as opposed to intellectually) aware of.

      To achieve clarity here, we need names.  Loosely speaking, we might call the perceived object the objectual content of the experience. This would be a kind of external content; but, it is “in” the experience, conceived as partly a relation to the external world.  Thus, a prancing deer before me that fills my visual consciousness might be considered a kind of content of my perceptual experience.  The deer might be conceived as the object that fills my consciousness.  Similarly, we might call the phenomenally represented properties—those properties one is in some way conscious of in the perceptual experience—the property content of the experience; and we might call the property-ascriptive propositions the perceiver can perceptually know on the basis of the perception the total propositional content.⁸

   The propositional content could be divided into at least two kinds:  call any propositions that are perceptually believed on the basis of the experience the doxastic  (propositional) content of the experience and those propositions that the perceiver is disposed to believe (but does not in fact believe) on the basis of the phenomenal perceptual elements of the experience the implicit (propositional) content. These two categories together may include all the propositions whose truth would make the experience, in respect of its phenomenal content, veridical, i.e., an objectively correct representation of some part of the world.⁹  If the experience represents two birds perched on a telephone wire before me, its propositional content includes the proposition that there are two birds on a telephone wire before me; if it represents a yellow dot in a blue patch, in includes the proposition is that there is a yellow dot in a blue path before me; and so forth.¹⁰  These three broad kinds of content, property content and the two propositional kinds, are not the only elements one might take as contents of perception, but they are major elements meriting that name. It will be the property content of moral perception that most concerns us. 

“Seeing is believing”

      This is a good place to examine the incalculably influential but rarely examined adage, “Seeing is believing.” Realizing what is and is not implied by a good understanding of this adage is perhaps even more important for moral perception than for perception in general. Consider the paragraph you are reading.  Do you believe that it is not in longhand or that has more than four words?  Well, you could see that these things are true before I asked.  Must you not “believe your eyes?” Now, for every property clearly present in your visual field, you might be thought to believe that the object you see bears that property:  say, being white, printed (rather than handwritten), appearing in a rectangular block, and having white margins.  This idea regarding belief-formation might underlie, and might seem to support, the view that the content of a visual experience is manifested in beliefs of all the propositions ascribing the visually represented properties (at least all such propositions one can understand).

      This idea inflates the doxastic content of the mind—what one believes—far beyond plausibility and is surely mistaken.  You need not believe—before I mention the point—that this paragraph has more than four words, even if, in merely looking at it, you see many more than four.  By virtue of seeing the full paragraph, you are, however, disposed to believe this if the matter arises, as where someone has said that she thought the page in question had, as its paragraphs, only one-line short aphorisms.  I contend that it is only under certain conditions that seeing entails (propositional) believing.  Apparently, it is mainly where what is seen, or an aspect of it, has some significance for the perceiver that seeing a property of the thing often produces the corresponding property-ascriptive belief.  Seeing may not produce belief even when one has the relevant thing in view.¹¹ One may even focally perceive something, as I do now the letters I am typing, without believing, for each, that it is a t, that it is a y, that it is a p, etc.  One may also merely peripherally perceive something, as where, concentrating on the birds, I see but pay no attention to a helicopter flying past them in the distance, and believe that it has some property, such as rising.  Focal perception may tend to yield belief; but need not, and merely peripheral perception may tend not to yield it, but may.  What is uncontroversial is that perception commonly disposes those with the appropriate concepts to form some beliefs and can enable the formation of indefinitely many that constitute knowledge.

         Even when perception produces belief, that belief need not be propositional, as opposed to predicative.  Consider seeing a normal, five-fingered hand in a play in which some characters are punished by amputation of a finger.  I could see, on stage, a normal human hand, and believe it to be five-fingered but not believe a proposition to the effect that the human hand before me is five-fingered. For although I am seeing a hand, I think that it is a plaster imitation.  It also seems—and this will later be seen to be important for moral perception—that we can respond to a property of something, such as a high obstacle in our path which we automatically step over—and we can take a thing to have a property—without believing it to have that property:  discriminating the property may be enough to guide our steps so that we do not trip. I doubt, then, that perception can guide behavior only through producing guiding beliefs.¹²

      These points may be generalized to other perceptual modes.  First, we cannot properly understand perception if we overintellectualize it in the way that, perhaps partly because of misunderstanding of the idea that seeing is believing, is natural for many philosophers.¹³  Second, it is crucial for understanding both belief—including moral belief—and its justification that we understand how experience, especially perceptual experience, can justify belief or render it knowledge. Believing that this paper is white is justified by my visual experience of white (or, arguably, by my actually seeing its whiteness). Saying that I see that this paper is white, as some would in justifying this belief to a skeptic, self-ascribes this visual experience and thereby suffices to express my justification; but, in specifying what justifies this belief, the self-ascription of seeing its whiteness, or of seeing that it is white, misleadingly suggests that I must believe that I see its whiteness in order to be justified in believing simply that it is white. In the order of justification, seeing is prior to its ascription, and perception can provide justification and produce knowledge whether or not we believe it to occur. 

The causal element in perception

  My stress on the discriminative dependence of phenomenal perceptual representation on the object perceived should indicate that I regard perception as (or at least as embodying) a causal relation.  I am suggesting a causal (and adverbial) theory of perception.  On this theory, to perceive something is, in outline, for it to produce or sustain, in the right way, an appropriate phenomenal representation of it. I refer to non-hallucinatory experience; but even hallucinations may have representational content.  In my terminology, even if a hallucination of an object is so vivid as to be phenomenally indistinguishable from—and even qualitatively identical with—perceiving it, it is only a sensory experience, not a perceptual one:  there is nothing seen, heard, etc.¹⁴ 

III. The Basis of Veridical Perception

      

Seeing three-dimensional objects is a good example for a reason not yet apparent.  Let me, then, indicate some core elements of seeing objects.

      

First, we never (directly) see all of an object such as a tree, animal, or building; we see only the part facing us.  On the most intuitively plausible account, we see objects in virtue of seeing certain of their properties, such as (even if distortedly) their shape.¹⁵

  Second, no inference is required:  the attributive seeing is constituted by seeing the appropriate properties; it is not a case of, or dependent on, inferring, from propositions about the object, that it is (say) flat. 

Phenomenal vs. physical properties

      

Perceptual experience has long seemed to some philosophers to entail acquaintance with phenomenal properties as well as with certain of the physical properties of the perceived object.  Consider viewing the rim of a drinking glass from an angle.  The rim will commonly appear—in terms of what we might call “uncorrected” visual experience, to be elliptical.  But, despite the look of the glass from this perspective, ellipticality is not a property of an ordinary drinking glass.  Thus, talk of seeing the properties of objects must sometimes be understood in terms of properties that, in some appropriate way, represent the properties of the object seen but do not belong to it.  There is of course a physical property of ellipticality, and we can understand how a phenomenally visual awareness of this property can, under the right conditions, indicate to the viewer the actual, incompatible shape of a round glass.  What, then, is the phenomenal property that we instantiate when we see the glass as having an elliptical rim?  There is no elliptical object present, nor, on my view, any special object, such as a sense-datum, that can bear a phenomenal property of ellipticality or indeed any properties at all.

      

The positive view I propose is that the phenomenal property in question is what we might call visually sensing elliptically or being visually appeared to elliptically (visually sensing in an “elliptical manner,” one might say, since we are speaking in terms of an adverb of manner).  This is an adverbially described common experience that we normally have in viewing a glass from a sharp angle.  I take having that experience—experiencing in the relevant phenomenal way—to be a psychological property of the perceiver.  It is plainly not a property of circular rims, nor does it entail that the viewer is acquainted with a sense-datum, an object “given” to sense, that in fact is elliptical. The perceiver need not even be aware of the phenomenal property itself; instantiating that property makes possible an awareness of it, but does not require that awareness.

      

By contrast, on the sense-datum view, there are phenomenal properties of these special, presumably mind-dependent, objects—sense-data—objects that have their own properties, such as phenomenal ellipticality; and these properties in turn represent physical objects to perceivers.  On my view, there are no such objects and we need not posit acquaintance with a phenomenal property of ellipticality in order to explain the experience of seeming to see an elliptical rim.  The phenomenal property in question—visually sensing elliptically—is a higher-order property of persons that they have by virtue of being sensorily aware of the property of ellipticality.¹⁶

There are persons; there are physical properties; and there are sensory experiences in which we are aware of those properties.  When, in perceiving an object, we are visually appeared to in a given way, we instantiate a phenomenal property.¹⁷

This experience may be just like one of seeing an elliptical rim; but that is not because we are acquainted with something that is elliptical.  It is because something round appears to us in a way that evokes the visual sense of ellipticality. If we may be said to be aware of the property of ellipticality, it is not because we perceive or are acquainted with something—whether physical or a sense-datum—that has this property.

      

It seems to be an empirical question whether one might have the awareness of ellipticality that goes with sensing elliptically if one has never seen an elliptical object. Perhaps the phenomenal (sensory) property might never occur in one’s experience if one has never had a perceptual experience causing one to instantiate it.  Nonetheless, the possibility of hallucination apparently shows that sensory awareness of ellipticality—a phenomenal as distinct from perceptual awareness of it—does not entail simultaneous awareness of any object possessing it.

      

The kind of awareness of properties so far considered—perceptual as opposed to intellective awareness—is phenomenal.  Perceptual awareness of properties, moreover, may require that instances of them play a causal role in producing or sustaining that awareness; but if so, what of hallucinations?  Since hallucinations can occur when the sense in question receives no external stimuli, as where someone who has lost eyesight visually hallucinates a loved one, an adequate account of sensory experience should avoid taking it to contain an external object that is in some attenuated or abnormal way perceived.  Here it is important to realize that what we are aware of can be abstract.

      

There appear, then, to be at least two kinds of awareness of properties, intellective and phenomenal. Suppose that, normally, the former requires—if only earlier in life—a route through the latter and in some cases may have a phenomenal element.  It may still be true that non-physical awareness of a property, as in hallucination, can produce a visual or other sensory experience that seems to the subject just like seeing.  The phenomenal awareness here can be phenomenally rich even if its object—a property—is abstract.  Intellective awareness can also be rich but is not my concern here.  It is, however, important: on the epistemology I find most plausible, it is the conceptual kind of awareness that apparently underlies a priori knowledge.¹⁸

      

Suppose, however, that we regard the objects of the most basic kind of perception as the sorts of physical properties by which we see spatiotemporal things.  Those things themselves are plausibly taken to be seen by virtue of seeing their properties:  the color and shape of a tree, the round, dark, treaded look of tires, the rising silvery shape of a plane taking off. How, then, can we maintain that perception is a causal relation to the object perceived?  This object, after all, is what causes the perceptual experience; the cause is not something abstract, such as a property. The problem here is one for perception in general and must be solved if we are to understand moral perception. 

The causation constitutive of the perceptual relation

      

For our purposes, it is best to approach the difficult question of how to characterize the terms of the relevant causal relations by first considering how to understand attributions of causation to substances.  Consider saying that what caused a dent in the door of a car was a stone, a paradigm of a substance.  Surely we mean something like this:  the event of the stone's hitting the door dented it, i.e., caused the event of its becoming dented (a distorting of its surface).  The problem here for a causal theory of perception is to explain what causes perceptual experiences.  Many perceptual experiences are not events, but ongoing impressions; nor need objects of perception, even in moral cases, be events.  One could see an injustice by viewing the distribution of someone’s grades for a set of students in a seminar one has fully attended (say, where the weakest two have very high grades while the strongest has a very low one).

      

A natural account of how substances figure in causing perceptual experiences arises from analysis of event-causation and sustaining causation.  My now seeing a hand is causally sustained by the hand's now having the observable property of being five-fingered, upraised, and so on for each of its properties that I see. With momentary perceptions, such as hearing a shot, the perceptual object's occurrence, a bang in this case (or a change, as where a rifle yields a change from silence to a banging) causes an event of, say, a heard banging sound.  Similarly, if I raise my hand before my eyes, that event causes my seeing its rising, which is a partly phenomenal event.

   

The metaphysical problem here is to explicate what kinds of objects are designated by the noun phrases naturally used in the relevant cases, for instance ‘my hand's turning’ and ‘the flagpole’s standing still’. We immediately encounter ambiguity:  such phrases can designate types (properties) or tokens (concrete instantiations of properties).  If I raise my hand twice, there are two tokens of the type, raising my hand.  If the flagpole stands still after each gust of wind, there are as many tokens of that state property, the flagpole’s standing still at the relevant time, as there are gusts. The type is abstract and hence the wrong kind of thing to be a cause.  I call the token a dependent particular.  It is a particular because it is a unique thing in time and, at least for physical events, space; it is dependent because there must be an object that instantiates the property of turning (and, for at least event properties, tokens it).  Ontologically, the hand is more basic than its turning:  the hand can exist without turning, but not conversely.

   

Connecting this ontological point about events with the causal theory of perception yields the view that what causes our perceiving a substance is its instantiating (tokening) some suitable set of properties, commonly including at least one observable property, such as its color or shape, or a sound it makes.¹⁹

  Unless the perception is strictly momentary, there will be both event causation and sustaining causation:  even for my seeing my hand steadily, in which case my perceptual experience is sustained by the hand, there is an event of my starting to see it, followed by my continuing to see it.  That state is sustained by the temporally extended instantiations of the relevant properties of the hand, normally observable properties such as a color or shape property. How this view of perception bears on moral cases will be shown in Chapter 2.  

      

***

      

Let me draw together some main points about perception that have emerged, again speaking in terms of seeing as a paradigm.  To see an object is to see some suitable subset of its properties.²⁰

Seeing these is a matter of an appropriate causal relation between its instantiating such properties and our phenomenal awareness of them:  i.e., an instantiation of certain phenomenal properties—a kind of sensory representation—of the visible properties in question.  This theory of perception is realist, mentalist, and (in the indicated sense) representationalist.  There must be a real object perceived; perceiving it must be experiential in a sense entailing mentality; and the perceptual experience must in some way represent the object.

      

The realism may seem indirect, but it is as direct as a plausible realism can be.  There is no more direct way to see a physical object than by seeing its properties.  We could say that perception is “mediated” by awareness of properties, but this is misleading:  the properties are not intermediary concrete objects, like sense-data, nor is one’s perceptual sense of the object distinct from one’s perceptual sense of its properties.  The only concrete objects posited—as opposed to properties, which are abstract entities—are the perceiver and the thing perceived.  The property instantiations are constituents in the perceptual relation itself and, ontologically, they are built from universals and the perceptible particulars that instantiate those universals.

      

How well can the theory of perception just sketched accommodate moral perception, such as seeing injustice? This is the main question addressed in the next chapter.  Of particular importance is to explain how the causal element in perception can be accommodated in the moral case, since moral properties are commonly conceived as normative and often taken to be outside the causal order.  We must, then, consider not only how causation figures in moral perception but some central points about the nature of moral properties themselves. 

Chapter 2 

Moral Perception:

Causal, Phenomenal, and Epistemological Elements 

      

  We should begin with clarification of the problem to be addressed.  What kind of experience might be thought to constitute moral perception, and why should we think that there is any genuinely moral perception at all? 

I. The Perception of Right and Wrong

      

Many philosophers think that moral knowledge is never perceptual and that perception is relevant to ethics only by representing certain non-moral facts.  These are the kind that, like the fact that one person clubbed another, can be ascertained without applying or even having moral concepts.  Responding to this skeptical view requires both an account of perception and an understanding of the basis on which singular moral judgments are made.  Singular moral judgments are the common kind in which a particular person or group of persons is said either (prospectively) to be obligated to do something (say, by a promise) or (retrospectively) to have done the right thing (for example by willing money to charities) or to have acted immorally (say, by taking a bribe).

      

It is essential to distinguish between moral perception and mere perception of a moral phenomenon, for instance perception which is simply of a deed that has moral properties—something possible for a person who has no moral sensitivity nor even any moral concepts.  Seeing a deed that has a moral property—for instance the property of being wrong—does not entail seeing its wrongness.  We can see a theft, as where we see someone break a store window and seize jewelry behind it.  We can hear a lie, as where people tell us they telephoned us at a time we know we were listening in vain for a phone call.  And of course we can feel a stabbing. These three cases, even conceived minimally in terms of what is sensorily represented in experience, are perceptions of non-moral phenomena.  They vary in what might be called directness, but they are all cases of perception.  Can we, however, also perceive the moral wrongs that these acts commonly entail? As to the case of seeing the rightness of something—a property that is positive from the moral point of view—we may, in everyday language, say such things as that we are relieved to see justice done or gratified to see a long-evaded moral obligation at last fulfilled.  But is justice itself, as manifested in, say, equal treatment of citizens, genuinely perceptible?

      

This book argues for a positive answer.  Further examples provide some support for that answer.  We sometimes speak as if we have actually seen a moral property (or an instance of one).  In answer to 'Did you ever see him wrong her?' one could properly say, 'Yes, I saw him spill hot tea on her hand when, at dinner, she was describing an incident that embarrassed him'.  And could one not see terrible injustice by viewing soldiers shooting citizens who are peaceably criticizing their government?

      

If we are guided by how ‘perceive’ is used in English, we might go further still and say that there can be perception of certain traits of moral character.  Some people seem visibly good.  Someone might say, ‘Just take a problem to her and in minutes you will see her goodness—it positively shines’. Doubtless the seeing of goodness here is by way of seeing manifestations of it.  That is not how the moral rights and wrongs so far illustrated are seen, but perhaps there are two (or more) significantly different ways to see a moral property.  Seeing her goodness is a kind of indirect seeing. Still, if the manifestations bespeak goodness—in reflecting it clearly—there appears to be a sense in which that goodness is indeed seen.  We can see a plane in the distance by seeing a speck and hearing a distant roar.  Might goodness not be sometimes manifested by comparably visible signs? To be sure, whereas we see, for instance, the wrong in the theft by seeing its ground—the seizure of someone else’s property—we see a person’s goodness by seeing its manifestations.  But this difference indicates two bases (and kinds) of moral perception; the difference is not between an instance of seeing as opposed to one of inferring.

      

    There is then, some reason to take literally discourse that represents moral properties—or apparent moral properties—as perceptible. One objection to taking such discourse literally is that we do not see or in any sense perceive moral properties, but only non-moral properties or events that evidence their presence.  Skeptics and noncognitivists may go further:  they might say that we simply perceive natural properties that cause us to tend to ascribe moral properties (or apply moral predicates) to their possessors or to express moral judgments about those acts, persons, or other objects of moral appraisal. Let us explore these objections.

   Suppose this causal hypothesis is true and that we do indeed tend to ascribe—and to say we see—moral properties because we perceive certain non-moral ones.  We must still ask what relations hold between the two sorts of properties²¹

and between non-moral judgments and moral judgments.  Second, we must ask whether these relations differ importantly from relations common outside the moral realm.  Third, if they do, does the difference show that we do not acquire moral knowledge or moral justification through perceptions of the kinds I have illustrated?  Answering these questions requires establishing some basic points about moral perception.

      

To begin with, we should not expect moral perception to be exactly like physical perception, at least exactly like perceiving everyday visible objects seen in normal light.  First, moral properties are not easily conceived as observable, in what seems the most elementary way:  no sensory phenomenal representation is possible for them, as opposed to intellective representations, though these may be integrated with phenomenal elements.²²

  Second, even the perceptible properties on which the possession of moral properties is often based may not be strictly speaking observable, at least in this elementary way.  You can see Wang do an injustice to Xiaohui by, for example, witnessing his stealing jewelry from her handbag or his falsely accusing her of theft.  But arguably, what is observable is Wang’s removal of (say) a bracelet or his audible action of accusing her of theft.  Still, you might count Wang’s false accusation among the “observable facts”—because you see and hear the accusatory speech act and you know that the accusation is false.  We must grant, however, that even though you can visually observe the basis of the falsity, such as, rolled into an accuser’s hand, the money the accused is said to have stolen, your seeing the injustice depends on your understanding the significance of the discrepancy between this visible fact and the content of the accusation.

      

Here we should distinguish between two kinds of properties that, though their presence can be known by inference, are each capable of being objects of non-inferential acquaintance:  the perceptual and the perceptible.  The former are sensory properties and include colors and shapes, textures and resonances, and other properties familiar from sense experience; and there is no ordinary perception without them.  The latter are not sensory and include certain moral properties: being wrong, being unjust, and being obligatory, among others.  A number of our examples indicate how perception reveals such properties. We can see this more clearly by considering the question whether the kind of perceptibility in question is a matter of being observable.    

      

Is injustice observable, in the most basic sense, which apparently goes with perceptual properties?  Is seeing injustice, for example, in any sense observational in the corresponding sense?  Or is such moral perception seeing—in a way that is at least not narrowly observational—a set of “base properties” for injustice, i.e., seeing properties on which injustice is consequential in a way that makes it obvious that an injustice is done?  In asking this I assume something widely held:  that actions and other bearers of moral properties do not have those properties brutely, but on the basis of (consequentially on) having “descriptive” properties. An act is not simply wrong, in the way in which an act can be simply a moving of one’s hand.²³

  It is essential to the wrongness of an act that is wrong that it be wrong on the basis of being a lie, or because it is a promise-breaking, and so forth.  Similarly, person is not simply good, but good because of having good governing motives together with beliefs appropriate to guide one toward constructive ends.²⁴

      

The question, then, is whether one ever really sees a moral phenomenon, such as an injustice.  Recall the distinction between seeing an action that is wrong and seeing its wrongness.  Clearly, one can see a deed that is wrong (unjust, a violation of a moral right, and so forth); this requires simply seeing the deed and its in fact being wrong.  One can also see the deed as unjust; this requires seeing it and in some way viewing it as an injustice.  Thirdly, one can see that injustice is done; one way this is possible is by seeing that a deed is done and, partly on that basis, realizing, from seeing properties one knows it to have, that it is unjust.  But, in any of these three cases, does one literally see injustice?  Consider the example of a colleague’s writing a B on a student’s grade report rather than an A, which, from our own careful reading, we know is deserved. Do we, in virtue of observing the falsification, see an injustice?  We may speak of moral perception here, but I grant that it is not the elementary kind of perception illustrated by seeing the shape of a drinking glass.

      

Does moral perception, however, differ in kind from every sort of non-moral perception?  One might think that the phenomenal elements in perception properly so called must be sensory in the representational way that goes with paradigms of seeing and with some of the other four ordinary senses.  But why should we expect perception of injustice, which is a normative, non-sensory phenomenon, to be just like perception of color, shape, flavor, or sound, which are physical or in any case sensory and non-normative?  Why should there not be, for instance, a phenomenal sense of injustice that is not “pictorial” in the way exemplified by the visual impression of a tree or a painting?  Must perceptual experience even be cartographic, embodying a “mapping” from phenomenal properties, such as a visual impression of four squares in a painting, to properties of the object perceived, say division into four squares? I think not. What we see must be perceptible; but even if perceptible properties, such as being wrong or unjust, must be seen by seeing perceptual properties, the latter do not exhaust the former.  The senses can yield the base from which we see certain perceptible properties without their being on the same level as the perceptual properties pictured or mapped by the senses.

      

If seeing physical objects dominates one’s thinking about perception, pictorial and cartographic representation may seem to exhaust perception.  The influence of seeing perceptual properties—the narrowly observational ones—on the philosophical accounts of perception is far-reaching.  Hume, for example, conceived perception in terms of impressions. If one imagines faithfully impressing what one sees on wax—which can record visualized color and shape quite faithfully—one can see why perceptual representation might be conceived cartographically.²⁵

  The colored wax impression is like a topographical map of what is seen.  But perception is not mental map-making, and perceptual representation need not be cartographic.  Even olfactory, gustatory, and perhaps auditory, perception indicate that point.  Many such experiences are one-dimensional in a way that common visual experience is not—though seeing nothing but, for instance, the black wall of a ship from a rowboat by its side could be one-dimensional.  Where there is no variation in a sensory field, there is normally nothing it seems to map.  Given these points, we can see that although, for certain cases of injustice, we can perhaps find, for the concrete deed seen to be unjust, a kind of mapping from the perceptual sense of injustice to certain of the base properties of injustice, there is no phenomenal property that has a cartographic representational relation to the property of injustice itself. 

II. The Representational Character of Moral Perception

      

In the light of what has emerged so far, we may distinguish between a phenomenal—and especially, a cartographic—representation of injustice and a phenomenal representation constituted by a (richer) perceptual response to injustice.  The sense of injustice, then—a kind of impression of it, one might say—as based on, and as phenomenally integrated with, a suitable ordinary perception of the properties on which injustice is consequential, might serve as the experiential element in moral perception.²⁶

      

An important constituent in this phenomenal integration is a felt sense of connection between, on the one hand, the sense of, say, injustice or (on the positive side) of beneficence and, on the other hand, the properties that ground them.  This felt sense of connection normally produces a non-inferential disposition to attribute the moral property of the action (or other phenomenon in question) on the basis the property or set of properties (of that action) on which the moral property is grounded.  If, for instance, I see injustice in a distribution, my sense of injustice normally yields a disposition to believe that distribution to be wrong because it is, say, giving more to one person than to another who is in the same needy position. My attribution of injustice need not be tied to the term ‘injustice’ or any synonym.  Any of a range of terms may be appropriate, and we may indeed leave open the extent to which the property attribution depends of the perceiver’s use of language at all.

      

I take moral perception to carry a phenomenal sense—which may, but need not, be emotional—of the moral character of the act. This sense may, for instance, be felt disapproval, or even a kind of revulsion, as where we see a man deliberately spill hot tea on his wife’s hand.  It need not be highly specific; it may, for instance, be a felt unfittingness between the deed and the context, as where we see a male and female treated unequally in a distribution of bonuses for doing the same work.  Positively, a felt fittingness may play a similar phenomenal role in moral perception.  Think of the satisfying sense of moral restoration when one sees a person one cares about apologize to someone else one cares about after an offensive remark; the apology befits the offense.  In each case, the moral sense of wrongness, injustice or, in the positive case, of welcome reparation is essentially connected to perception of non-moral properties on which the moral properties are based. In cases like these, we might be said to sense morally. This is not because moral properties are sensory—they are not—but because there is a kind of sensory experience suitably integrated with perceiving the properties that ground the moral property that we sense. Perceptibility is not exhausted by perceptuality. 

Moral properties and the perception of emotion

      

Here we can learn much by comparing moral perception with seeing an angry outburst that might evoke the comment ‘I’ve never seen such anger!’ Shall we say that the anger is not perceived because it is seen through seeing what might be called constitutive manifestations of it, such as redness of countenance, screaming, and heavy breathing?  Granted, these can be mimicked by a good actor; but the façade of a house may similarly mimic a complete structure, concealing empty space behind what looks like the front wall of a large residence. We should not conclude that houses are never seen. Why, then, may some injustices not be as perceptible as anger?

      

It is true that whereas anger is seen by its manifestations, moral wrongs (and other phenomena) are seen by their grounds.  But why should perception be conceived as limited to responses to effects rather than causes—or, more broadly, to other reliable indicators or determinants of the perceived phenomenon?  Suppose we think of perception as—in part—a kind of processing and reception of information by a causal path from an information source to the mind, where the processing need not imply events in consciousness.²⁷

This is certainly a conception that comports well with the role perception plays in providing empirical knowledge. On this conception, it should not matter whether the information impinging on the senses is determined by what is perceived or determines that.  We can know a thing either by its effects that mark it or by its causes that guarantee it.

      

Consider the latter kind of case—perception by grounds—through an example of perception that yields perceptual knowledge by seeing a causal ground.  May I not report that I saw a killing if what I saw is one person put what I know is a fatal dose of a poison for which there is no antidote into another’s whiskey which the latter then fully drinks? I see an act that is a killing and, given my background knowledge, see it to be one.  If the passage of time makes it seem that the action of killing is not complete until the death, note that the relationship between the causal information and the deed is the same as where one sees a powerful bomb explode next to a soldier but, owing to instantaneous smoke, does not see the inevitable resulting death. Has one not seen the killing?

      

Even if one insists that the killing cannot be properly said to have been seen until death occurs, the relation between that by which it is seen and the seeing of it is the same:  the seen determinants are still that by which the action, and the wrong it constitutes, are seen. The use of language I illustrate here is at least as old as Shakespeare.  In Romeo and Juliet, Mercutio, fatally stabbed, says, “I am sped” (Act III, scene 2); and in Hamlet, Laertes, after being fatally wounded by his own poisoned sword, confesses, “I am justly killed with mine own treachery” and continues with an affirmation that Hamlet too has been killed: 

      “. . . Hamlet, thou art slain;

      No medicine in the world can do thee good.

                              (Act V, scene 2) 


      To be sure, the relation between the grounds of wrongdoing and the wrongdoing seen by them is not causal.  The fatal poisoning does not cause the wrongdoing; it entails, in a way that implies partial constitution, the wrong-doing.  But, as Laertes’s wrong-doing illustrates, the relation can be highly reliable.  It can even be a necessary relation and so, from the point of view of perception as a kind of receiving of information, perfectly suitable to ground perception and indeed knowledge.

      It must also be granted that whereas someone perceptually normal cannot ordinarily view a tree in daylight without the usual arboreal representational experience, many people can view a perceptible injustice without perceiving injustice or having any moral phenomenal response, such as an intuitive sense of wrong-doing.  But the proportion of people who are morally percipient, or indeed even morally normal at all, is smaller than the proportion who are perceptually normal.  Moreover, injustice can be subtle, and the difficulty of perceiving it may, for some morally normal people, trace to the need for greater moral sensitivity than goes with mere moral normality.  Compare seeing a painting in which a small figure of a person is sketched visibly but is not sharply distinct from background shrubbery.  Someone who is perceptually normal but not an experienced viewer of paintings might not, without careful scrutiny or guidance, have any phenomenal response representing that figure.  This does not imply that the figure is visually imperceptible.  Similarly, a person’s lacking a sense of injustice upon witnessing an unjust deed does not imply that its injustice is morally imperceptible—certainly not to other observers.  Insensitivity to a property does not imply its imperceptibility.²⁸

      It should be apparent, then, that the theory of perception I have outlined can accommodate moral perception by incorporating a distinction between perceptual representations of an ordinary sensory kind and perceptual representations of a richer kind that have a moral element.  Can this broad theory of perception, however, explain how moral perception can have a causal character?  It can.  It does not do so by treating (moral) perceptual property instances like seeing injustice as causally produced or sustained by instances of moral properties.  The theory is neutral regarding the possibility that moral properties are causal.  It does, however, construe seeing certain subsets of base properties for injustice as—at least given appropriate understanding of their connection to moral properties—a kind of perception of a moral property; and this kind includes, as elements, such ordinary perceptions as seeing a violent seizure of an old man’s wallet and hearing an abusive vulgarity screamed at a conference speaker.  Depending on our psychological constitution, we may be unable to witness these things without a phenomenal sense of wrongdoing integrated with our perceptual representation of the wrong-making facts.²⁹ In many people, moral perceptions of certain salient moral wrongs are virtually irresistible. 


Is moral perception necessarily conceptual?

      We have seen the difference between a moral perception of wrong-doing and a perception that is merely of an act that is wrong. We have also seen that moral perception does not entail the formation of moral belief or judgment.  Still, although moral perception is not belief-entailing, it remains true that given how we see certain base properties that are sufficient for injustice, we sometimes perceptually know, and are perceptually warranted in believing, that, for instance, one person is doing an injustice to another.  We are thus warranted in seeing the deed as an injustice.  When we have such perceptual knowledge or warrant, we are often properly describable as seeing that the first is doing an injustice to the second and, indeed, as knowing this.

      This point does not imply that seeing an injustice is intrinsically conceptual.  But seeing that an injustice is done is conceptual.  A child who has yet to develop the concept of injustice can see an act that constitutes an injustice. A decade later, of course, the same perception might immediately yield a moral conceptualization of the act or indeed moral knowledge thereof.  Between these two points, the child may be disturbed at seeing the kind of act in question, say giving a visibly much nicer pair of shoes to a twin sibling of the same sex taken to the same store just before the school year.  Prior to conceptualization, the child may have a sense of unfittingness in such action:  the disparity in treatment disturbs the child who sees the sibling treated better.  That perception of disparity, together with the sense of its unfittingness, reflects a discriminative sensitivity to differential treatment and puts the child in a good position to develop the concept of injustice.

      My view does not imply that all non-inferential moral knowledge of perceptible moral facts is itself perceptual.  One might know such facts from memory or testimony.  Those are non-inferential ways of knowing.³⁰  Conceivably, there could even be a subliminal detection capacity by which someone non-inferentially knows some such facts more directly.  This kind of knowledge would be analogous to the sort possible through a kind of “blindsight.” An experience constituting a moral perception, by contrast, must have a phenomenal element.  Many elements of this kind might be called perceptual moral seemings.  These are not equivalent to intuitive moral seemings, though they may produce those.  Nor need the proposition known be itself intuitive, in roughly the sense that we find it plausible upon considering it even in isolation from perceptual or other evidence for it.  It may well be, however, that some perceptual moral seemings are related to intuitive seemings based on a hypothetical version of the same case, much as perceptual physical seemings, as with ordinary vision, are related to an imaginational seeing of the same object in the mind’s eye. Recollective imagination can replay perception; prospective imagination can preplay perception; and creative imagination can reach beyond perception both in the variety of properties imagined and in their combinations.  Responsiveness to property instantiations is crucial in both cases.

      Nothing said here implies that what perceptually seems to have a property actually has it, nor need every perceptual or intuitive seeming regarding a proposition—a (conscious) perceptual or intuitive impression that it is true—yield belief of the proposition it supports.  A perceptually knowable proposition may be only a potential object of an intuitive seeming, as where someone sees a wrong and considers the nature of the deed but does not initially have a sense that it is wrong or, especially, see that it is wrong:  here seeing a wrong may not even be a moral perception and certainly need not yield a propositional perception that the deed is wrong.  We might see one man we view as domineering shake the hand of another, smaller man of lower rank before a meeting and notice a hard squeeze, with the result of redness in the other’s hand.  It might not seem to us until later that we have witnessed an intimidation, though we could have been more alert and seen at the time that the former was wrongfully intimidating the latter.  Moral perceptual seemings, moreover, may or may not be partly emotional, as where indignation figures in them.³¹

      It should be evident that the theory of moral perception developed here is reliabilist in taking moral properties to be genuine properties of actions and persons.  But the theory also stresses phenomenal, attitudinal, and even emotional aspects of moral perception.  With these kinds of psychological, elements in mind, some philosophers have developed an extensive analogy between moral properties and secondary qualities or indeed treated the former as a subcase of the latter. This issue is too complicated to pursue here in depth.³² It is appropriate, however, to indicate why my position is not fruitfully considered a secondary quality view. First, it is characteristic of such a view to take secondary qualities, for instance colors, to be not intrinsic properties of their possessors—not “in” the object that has them, as Locke put it—but dispositions on the part of the object to produce certain phenomenal properties in the observer.  On my view, moral properties do belong to, and are in that sense “in,” their objects. They do entail that competent observers tend to have a certain kind of experience, but they are not constituted by this tendency. Second, whereas the secondary quality theorists tend to think that the nature of moral properties depends on human responses in the way color properties apparently do, I deny this.  Certainly a morally sound observer of, say, brutal injustice will tend to have a disapproving experience; but surely there could be such moral properties (and acts exhibiting them) even if there were no morally sound observers but only barbarous offenders.  


      ***

      On the view of perception presented in Chapter 1, perception is a kind of experience causally based on an information-conveying relation to the object perceived.  I have not offered a full analysis of this relation but have said enough about it to indicate how, even if moral properties are not themselves causal, they can be perceptible.  Perceiving them, as where we see an injustice, commonly evokes belief, but need not.  When it does, it may do so in a way that grounds that belief in perception of the properties of the unjust act in virtue of which it is unjust.  This kind of grounding explains how a moral belief arising in perception can constitute perceptual knowledge and can do so on grounds that are publicly accessible and, though not a guarantee of it, a basis for ethical agreement.  There is more to be said about how this is possible and about how perceptual moral knowledge is connected not only with other moral knowledge but also with intuition and emotion, which, as later chapters will show, are significantly related to perception and perceptual knowledge.

 


Chapter 3 


Perception and Moral Knowledge 


      Simple perception, whether moral or not, does not entail, though in normal case it amply justifies, belief formation.  Its non-doxastic character does not in the least preclude its presenting the perceiver with much information. That it does this explains in good part why it can both justify beliefs appropriately connected with its content and ground knowledge about the perceived object.  But if, in perceiving an object, we in some way processes information—as is widely held among psychologists as well as philosophers of perception—one may wonder whether perception is in some way inferential.  Understanding perception requires pursuing this question, and that in turn requires clarifying what constitutes inference.  This chapter will explore how perception and inference differ and how each may yield moral knowledge. 


I. Perception and Inference

      Some philosophers might contend that although we can see that (say) someone is writing ‘He did only two hours of work’, we cannot see that an injustice was thereby done even when we know he worked much longer.  Rather, they might claim, from what we visually know and background propositions we already believe, we infer that an injustice was done.  I grant that making such inferences is possible and also that the phrase ‘see that’ can designate inferential cognitions, as where ‘see’ means ‘realize’ and what is realized may come via premises. Such cases often represent an inferential kind of knowledge that rests on a premise yielded by perception.  My point is that for some moral knowledge, we need not posit an inference, as opposed to belief-formation that is a direct response to a recognized pattern.  An inference, as a tokening of an argument, is a mental event or process that requires a set of premises and a conclusion.  Inference is not needed for responses to patterns, nor even for certain kinds of interpretation of patterns or other complex phenomena.  Think of the kind of momentary outburst characteristic of breaking a fingernail when one is trying to open a battery compartment.  Someone observing the incident may instantly interpret the outburst as frustration rather than anger.³³

      Examples alone will not settle this matter.  The relation between inference and moral perception deserves further comment.  Suppose we do posit an inference underlying the kind of moral knowledge that is, on my view, perceptual.  We must then treat as inferential all our perceptual beliefs except the most elementary.  We could not be properly said to see that someone is angry or even that water is coming from the tap.  The properties that are observable in the narrow sense that goes with sensing these phenomena do not discriminate between anger and a theatrical imitation of it or between water and ethyl alcohol.  This is important if we need to show that we are seeing anger or water; but we must not impose the requirements for showing something on simply knowing it.³⁴ Moreover, positing inferences is not needed to account for how perceiving a pattern can mediate between perceptions of the elements in it and a belief the pattern produces.  Compare facial recognition.  We believe that someone approaching is (say) Karl because of the facial pattern we see, but seeing that is a matter of seeing many features of his face, not of drawing on myriad tacit premises attributing them.

      Granted, facial recognition depends on seeing the features of the face, as is evidenced by the impossibility of recognition where a number of features are blocked, say the eyebrows.  But such recognition is not dependent on inference from the relevant features, as is evidenced by the possibility of recognition even where the perceiver has no belief corresponding to those features.  I need not believe Karl’s brows have the look they do until I focus on the matter. Indeed, the look they have that is important for my recognition may be so distinctive—or so subtly related to other features, such as the nose and hairline—that it would be difficult, or perhaps impossible, to capture in the content of beliefs—or, correspondingly, in a set of premises for inference—at all.

      Even if propositional moral perceptions were in some sense inferential, attributive perceptions need not be.  Just as we can see a spruce to be a conifer without inferring that it is a conifer from facts about it, we can see an act to be a wrong from properties of it, without inferring that it is wrong.  Believing a thing to have a property does not even require believing that it has that property and thereby conceptualizing it as, say, a spruce.  Moral perceptions are similar in this respect.  In seeing one person wrongfully intimidate another we may have a phenomenal sense of the first wronging the second but only on reflection form a belief in which the concept of wrongness (or any moral concept) figures.

      Similarly, imagine hearing a judge issuing a sentence. We may have a sense of its unfittingness to the crime even prior to our forming—or without our forming—a belief, on that basis, that the sentence is unjust. To be sure, the phenomenal representation of voicing and diction may be psychologically so much more prominent than the moral sense of injustice that the latter is difficult to isolate and easy to miss, especially where it is not heightened by emotion.  But the visual sensations representing a sad face may be similarly subtle.  They may still be a basis for a perceptual belief about the person’s mood, whether or not emotion figures in the perceiver’s phenomenal response.  Our responses to persons and their deeds, like our responses to paintings and sonatas, may be very finely adjusted to myriad perceptible properties without our drawing a single inference.  Those responses may be moral, aesthetic, recognitional, or of some other character; and the complexity of their basis yields grounds for much genuine knowledge. 


II. Can Moral Perception be Naturalized?

      One might think that accounting for moral perception requires naturalizing moral properties so that they can figure in the causal order as do the observable properties familiar in the natural sciences.  I am not seeking to naturalize moral properties, nor does explaining the data we have considered in describing moral perception require such naturalization.³⁵ At least three points are needed here.  First, I have already noted that the experiential responses to moral properties that entitle us to speak of moral perception are causally explainable in terms of their having a basis in the natural properties on which moral properties are consequential.  Second, these responses can have this basis whether or not moral properties are themselves causal.  My account thus allows, though it does not require, naturalizing moral properties by providing a causal account of their constitution.  Third, the question whether moral perception—or any other kind—can be somehow inferential is orthogonal to the question whether only natural properties are perceptible:  whether perception is inferential or not does not depend on whether its only direct objects are natural phenomena.

      In one way, however, I am (non-reductively) naturalizing moral perception.  For I not only take moral perception to be a causal relation but grant that the base properties for moral properties are natural properties and have causal power if any properties do.  Moral perception is a phenomenon that occurs in the natural, causal order.  Moreover, the non-causal element important for understanding moral perception and knowledge acquired through moral perception does not require positing any supernatural being or even a Cartesian conception of the human person.  That non-causal element is crucial for conceptual capacities that go with an adequate understanding of moral concepts and with the a priori character of the relation between moral properties and the non-moral, natural ones on which moral properties are consequential.  The experience in virtue of which a moral perception counts as a perception is causally grounded in perception of natural properties, and the experience may be considered a causally grounded response to a moral property that is itself consequential on those natural properties.  This holds even if the phenomenal element in that response is not narrowly representational.³⁶

      My conclusion at this point, then, is this.  Although moral properties are apparently not natural properties, they are constitutively anchored in natural properties, in an intimate way such that seeing or otherwise perceiving the natural properties or relations that are their base suffices, given an appropriate phenomenal response, to make it reasonable to describe certain experiences as perceptions of such moral properties as injustice and, more generally, wrong-doing.  When moral perceptions like these occur, whether they are simple or attributive perceptions, the perceiver is in a position to see that something, such as an action or person, has the property in question.  Such propositional perception embodies a kind of moral knowledge. 


III. Moral Perception as a Basis of Moral Knowledge

      Some of my examples bring out that moral perception always comes by way of non-moral perception.  The relation here is not causal or instrumental, but rather constitutive.  Moral perception is partly constituted by a certain kind of response to perceiving, for instance to seeing, a moral phenomenon, such as an act constituting wrong-doing. Moral perception is not, properly speaking, caused by that seeing or by some other perception of a moral phenomenon; it is perceiving that phenomenon, in the moral way I have described.  Compare seeing a smile by seeing the distinctively happy countenance that we call a smile.  Where this is a hedonic perception and not a quasi-photographic representation of the face such as a dog might have, it is a way of seeing the facial expression in question.  Moreover, just as it is by being a stabbing of a helpless old man that the deed seen constitutes wrong-doing, when we smile by exhibiting the relevant facial expression, the ‘by’ here indicates a constitution relation:  in the context, producing the happiness of countenance is smiling.  It is not an ordinary means to smiling.³⁷

 

Basic and non-basic perception 

      If we reflect on cases like facial recognition, in which we acquire perceptual knowledge by perceiving other properties (though non-inferentially), we can distinguish between basic and non-basic perception.  Perception of the shape and color properties of William’s face is basic relative to perception of the property of being—as we might put it—“William-faced,” but we do not normally infer that the face is William’s from ascriptions of the more basic properties.  With moral perception, the relation of the base properties—those on which moral properties are consequential—to the moral properties grounded on them is at least as intimate as that of facially constitutive properties to that of having the face in question.  It is more intimate than the relation of anger-expressive properties to being angry.  I see no good reason not to speak of moral perception if we can speak of facial perception and perception of anger.  Perception is multi-levelled. Some kinds are more basic than others; and even if there happens to be a level that is ultimately basic for us as we are now constituted, the concept of perception does not dictate any final level, such as that of perceiving colors and shapes.³⁸

      Regarding the epistemology of moral perception generally, I have argued that there is a kind of experience properly called moral perception and that it can ground a certain kind of moral knowledge.  Even apart from skepticism (which I here assume is avoidable), we should ask whether the grounding of the moral-perceptual beliefs in question is sufficiently reliable to qualify them as knowledge.

      It must be granted that if we do not have good grounds for believing that the base properties are present (those on which a moral property is consequential), then we do not have good grounds for ascribing the moral property in question.  This kind of dependency, however, is not peculiar to moral beliefs.  If we lack grounds for believing that Wang is (say) red-faced and screaming as he hears of his son’s wrecking the family car, we also lack good grounds for believing, on that basis, that he is angry.  But notice this:  although his having these properties is excellent evidence that he is angry, it does not entail this (nor do any perceptible evidences); whereas A’s knowingly hiding the money that A is falsely accusing B of stealing does (non-formally) entail that A is (prima facie) wronging B.  In the first case, the grounding relation is empirical and contingent; in the second, it is a priori and necessary.³⁹  Moreover, although in the first case we perceive a fact by what it determines and, in the second, by what determines it, in both cases the perceptual knowledge is reliably grounded and is so in part by virtue of a causal relation.

      Consider a different example of evidential dependence. If we lack good grounds for believing that the lights have come back on, we also lack good grounds to believe on the basis of this fact that electricity is flowing.  But notice this:  whereas the lights being on is excellent evidence that the electricity is flowing, it does not entail that proposition, yet if one person really is knowingly hiding, behind his back, the money he is falsely accusing someone else of stealing, it does follow that he is wronging the other. In the one case the grounding relation is empirical and contingent, in the other it is a priori and necessary. 


Epistemic vs. inferential dependence

      My position, then, is that moral cognitions, such as moral judgments, can constitute perceptual knowledge but depend epistemically, though not inferentially, on non-moral elements.  Take inferential dependence first.  Suppose we know or justifiedly believe that a student plagiarized.  This is because we know or justifiedly believe that, for instance, the student’s paper is copied from the internet (this would illustrate both epistemic and inferential dependence).  Our knowledge is premise-based.  Suppose, however, that we see a man slap his wife’s face upon her asking him not to have another whiskey before driving home.  If, as would be normal, we know perceptually that he wronged her, we know this non-inferentially, on the basis of our adequate perceptual grounds.  Our need for this perceptual basis manifests an epistemic dependence, but not an inferential, premise-dependence.  Our grounds are perceptual, not propositional.

      Granted, inferential processing comes in many forms.  Two people could play a game in which they slap each other after various insults meant to be simply witty.  If I know this game is being played at a cocktail party, my seeing a husband’s slapping his wife might then lead me to believe he has wrong her only after I form the belief—perhaps without articulating it to myself—that the wife’s remark about driving after drinking was not part of the game.  It should also be granted that it may be difficult to tell whether a belief is or is not inferential.  But the difficulty of drawing a distinction in some cases does not imply that it is not quite clear in many others.

      Our justification in my slapping example, like our knowledge in that same case, is also perceptual:  we see the man slap his wife and, as with facial recognition, believe, on that visual basis and non-inferentially, that he wronged her.  Our justification for this perceptual belief is as strong as the justification we would have for believing this non-perceptually, on the basis of premises ascribing to him the properties on which his wronging her is consequential, properties whose instantiation we have perceived.  The strength of our justification is a matter of how well-grounded it is, not of the form in which the grounds are possessed:  specifically, whether they are possessed in the form of perceptually received information or in the form of beliefs whose propositional objects express the same information.  The belief that the tipsy husband wronged his wife can also count as perceptual knowledge because of the way it is based on a phenomenal responsiveness to the moral property. That responsiveness, in turn, is causally grounded in perception of certain of the natural properties on which the moral property is consequential. 


Phenomenological reliabilism

      The position on perceptual moral knowledge defended here might be called a phenomenological reliabilism.  It enables us to ground the possibility of a major kind of ethical objectivity.  It accounts for the availability of intersubjectively accessible grounds for a wide range of moral judgments.⁴⁰ It also explains how some moral knowledge can meet strong reliabilist constraints, that is, how perceptual beliefs constituting moral knowledge are based on perceptions that, in the circumstances, guarantee the truth of the beliefs or makes them very highly probable.

      Explaining how certain moral beliefs and the corresponding moral judgments can meet strong reliabilist constraints is important.  It paves the way for countenancing moral perception and the perceptual knowledge it can yield. The position does not imply, however, the even stronger conclusion that all non-perceptual moral knowledge rests on a foundation of perceptual moral knowledge, or even that we could not have the former without the latter.  The account is consistent with the plausible view that, without our having some perceptual moral knowledge, we would have no moral knowledge; but it leaves open the possibility that even apart from moral perception, we could have both inferential and non-inferential moral knowledge and justification for moral judgments.  The account does not foreclose the possibility that moral concepts are acquired through a combination of non-moral perception and moral concept-formation.⁴¹

      Developmentally, these two cognitive elements in our lives—non-moral perception and moral concept-formation—yield an understanding of the notions of right and wrong, of the obligatory and the permissible, and other normative concepts.  I have stressed that moral perception naturally occurs in this developmental process, but I have left room for other developmental views.  A certain kind of Platonism might posit development of moral concepts by rational insight into the abstract and explain singular moral knowledge by appeal to subsumption of particulars under moral concepts.  This subsumptivist view is not required by a Platonistic theory of general moral concepts, but it might be worked out compatibly with that view as naturally as in the case of the contrasting position I have presented.  On my view, even if we understand moral concepts as abstract entities of some kind, we can take moral properties to be anchored in the natural world and perceptible by their naturalistic grounds.   


***

      The theory I have advanced here suffices to explain how ethical objectivity is possible.  It provides for the possibility of our finding intersubjectively accessible grounds for at least a wide range of moral judgments—a matter to be explored in more detail in the next two chapters.  It also explains how moral knowledge is possible even under strong reliabilist constraints on what constitutes knowledge. The explanation takes account of the distinction between pattern-dependence, as is exhibited by much perceptual knowledge, and premise-dependence, as applies to inferential knowledge.  The possibility of moral knowledge by way of perception also does not depend on whether inference is in some way “implicit.” The reliability can be accounted for on either interpretation of perception, but it is best accounted for, and explainable in the psychologically simplest way, without taking perception to be implicitly inferential.

      If, however, ethical objectivity is possible and if, in addition, there is a kind of moral perception that has high reliability, why is there so much apparently rational moral disagreement?  One would think that if there is the kind of intuitive a priori connection that, on my view, holds between perceptible properties and moral properties consequential on them, there would be less disagreement in moral matters or, where there is such disagreement among rational persons, resolution would be less difficult.

      Here I can make just three points that are implicit in what has so far emerged.  First, just as quite rational persons differ in aesthetic and even perceptual sensitivity, they differ in moral sensitivity and may disagree as a result, even where they witness the same morally right or wrong actions.  Second, much moral disagreement centers on propositions that the disputants believe inferentially, and the parties may differ in their standards for sound inference, as indeed scientists may in theirs, or one or another party may draw an invalid inference.  Third, even apart from these points, rarely do parties to a moral disagreement respond to identical evidences, for instance exactly the same non-moral facts about well-being or, where disagreement concerns a deed witnessed by both, the same perceptions.  Moral disagreement among rational persons, even where each is in some way responding perceptually to the same phenomena, does not show that there is no moral perception or that moral perception cannot often be a basis of knowledge.  If it can be a basis of moral knowledge, then at least some moral judgments may be both objectively grounded and, as may be increasingly important in our globalized world, a basis for cross-cultural agreement.  How moral disagreement is to be understood on the theory I am developing and how it is connected with intuition and emotion will be among the topics of Chapter 4. 


Chapter 4 

Perceptual Grounds, Moral Intuitions, and Ethical Disagreement 

      

The previous chapters presented an outline of a theory of perception and explained how that theory enables us to explicate moral perception.  I have resisted the naturalistic temptation to subsume moral perception entirely under the ordinary perception of observable natural properties, but I have brought out important respects in which it is similar to that.  Moral perception has, for instance, a causal element. It also exhibits both the discriminative dependence, and some of the phenomenal elements, that go with its causal perceptual structure.  These characteristics are manifested in different ways in different people and may also be colored by elements of culture and upbringing.  One person's moral perception may be another's observation of mere behavior.  Our normative standards may also influence what we perceive and certainly what we perceptually believe.  The bereaved mother and the terrorist will see very different things in a car bombing; the priest and the pimp will see very different things in the servitude of a teenage girl.  They may, to be sure, have the same basic perceptions, say of color and shape, voice and movement.  But the mother and the priest have moral perceptions that the terrorist and the pimp lack.  This makes a great difference in what, overall, they see.

      

If, however, there really is moral perception, and if it rests on non-moral perception in the way I have indicated, why is there apparently more disparity in judgment in the moral realm than for ordinary non-moral perception? Chapter 3 briefly described how rational disagreement is possible in ethics and elsewhere, but it left much to be determined in explaining how rational disagreement is possible even where moral perception is brought to bear in settling it.  I pointed out that not all moral knowledge is perceptual and that not all moral perception produces moral knowledge—it may not even produce moral belief.  But there remains much to say about how rational disagreement is possible in ethics.  In addressing this matter, I will consider not only disagreements regarding perceptible moral properties, but disagreements involving moral intuitions and other kinds of moral cognitions.  This inquiry is particularly appropriate given that one of my overall aims in this book is to explain how objectivity in ethics is possible.   

I. Does Moral Disagreement Undermine Justification in Ethics? 

      

There are many disagreements on moral questions, and in some cases all of the disputants are rational regarding the subject under discussion.  Does this show that we are rarely if ever justified in holding a moral judgment, at least when we realize that some comparably informed rational person rejects that judgment?  It is essential in understanding disagreements of any kind to ascertain whether the disputants differ regarding the same proposition.  Sometimes a person rejects what another says without seeing just what that is, perhaps because the language used seems threatening or promises to put the person at a disadvantage, as where one can tell that one is being accused of something but does not see exactly what it is.  We could call this kind of disagreement illocutionary, since the disagreement is focused on the speech act (the “illocution”) conceived in some generic way and not specifically on its content.⁴²

  Illocutionary disagreement may seem to be only a pragmatic phenomenon rather than a substantive difference on the truth-value of some proposition.  But this is not quite correct. We should distinguish several kinds of disagreement.

 

Three kinds of disagreement

      

Suppose someone says something indefinite, for instance that my students are not happy with their assignments.  I may reject this not because I disbelieve the vague claim but because I can think of several propositions that might explain what the speaker has in mind, and I disbelieve each of those.  One interpretation might be that the assignments are too hard, another that they are too frequent, still another that they are uninteresting.  These need not even come before my mind to figure in my rejection of the indefinite claim.  It may help to call this a case of indefinite disagreement (the first kind of disagreement that concerns me). It is indefinite because the speaker is not clearly committed to any of the specific propositions I disbelieve, nor need I take the speaker to be asserting one of those.  We disagree on the vague claim, and I take it that there is something more specific on which we differ or would differ if we considered it; but I do not ascribe to my colleague belief of any specific claims.  My disagreement with the person is thus substantive (and not merely verbal or even a matter of contrasting attitudes toward the same proposition), though it is not specific.  I reject the person’s assertive speech act—which is why the disagreement may be called illocutionary—but not some particular proposition asserted.  It is as if I said, if only to myself:  I don’t know exactly what you have in mind regarding my students, but the things that come to mind are all propositions I reject.

      

This example brings us to our main concern here: content-specific disagreement. This takes two main forms and contrasts with the illocutionary kind, which is indefinite and is a matter of rejecting what is said rather than of believing any specific proposition that is a contrary of what is said.  Ccontent-specific disagreement is focal and occurs when one person affirms something and the other, realizing the incompatibility with something the other believes, rejects it or at least suspends judgment on it. This need not, however, be propositional disagreement, which is one kind of content-specific that concerns me and probably the most important kind. This kind occurs where the disagreement is over a proposition, which (in cases of such disagreement) I take to be a truth-valued element accessible to more than one mind.  But, as is evident from our discussion of the modes of perception and its different kinds of content, content-specific disagreement may be attributional—a matter of attributive beliefs and their differing predications—rather than propositional.  Let me illustrate this third kind of disagreement.

      

Suppose we each see a moving shape in the woods ahead of us and you say, ‘It’s dangerous—let's not go in there’, whereas I say, ‘It's not dangerous’.  It may be that neither of us has done any more than predicate different properties of the thing we both see but do not recognize, thereby differing in attributive belief. You might believe it to be large and lumbering toward us, while I believe it to be stationary or even retreating.  We might also believe different propositions that are tied to our different points of view and so can be identified only by their believer.  You might believe that the shape is like that of a bear you saw in a zoo (one I have never even visited), whereas I may believe that the shape has stubs like those of a tree-trunk I have seen that was shorn of its branches.  These propositions are not expressed by either of us and are not what we disagree on, even if they are part of the source of our disagreement about danger.  Still, we may understand each other to ascribe different properties to whatever it is, and that is sufficient to put us in content-specific disagreement.  For a moral example, recall the observation of an interview.  One of us, struck by diction and body language, may believe the interviewer to be unethical; the other, struck by the animation of the discussion, may believe the interviewer to be simply vigorous and might then be indisposed to make any negative moral judgment on it.

      

To be sure, even if we disagree regarding an object of discussion only in our ascriptions of properties to it, we are in a position to arrive at propositional disagreement as well.  We may nonetheless agree that we differ regarding the properties of what we are both willing to describe as the tall shape we see over there. Moreover, it may be that we can rarely settle an attributive disagreement without formulating propositions and seeking agreement on them; but here I am explaining what constitutes disagreement. How disagreement arises and how it is settled are different matters, though closely related to the content of the disagreement.⁴³

 

      

Now consider the moral case of the priest and the amoral case of the pimp.  The priest sees the girl who, in desperation, turns to prostitution as being used merely as a means or indeed as violated.  The pimp, by contrast, may see her amorally, as making the best living she can.  The cognitive difference here is due to perspectival disparity.  Each observer has a different perspective from which an indefinite range of propositions will seem true. Each will be disposed to dispute some of these, but the perspectival disparity is largely a matter of the categories (in some sense) in which they see the girl.  It will tend to produce definite contentual disagreement of both kinds.

      

The pimp may, to be sure, have certain moral concepts and a good sense of the base properties for them, but may also be amoral in one sense:  he may have no moral commitments regarding the girl or anyone else and no motivation to act on any moral propositions he may happen to believe about her or anyone else.  Even if he is not amoral in this sense, he may lack the morally important notions of violation of a person and of treating a person merely as a means.  Moreover, supposing he does have these notions, he may not apply them by all the same criteria as the priest, or may simply be insensitive to the evidences that indicate their application.  This is in part a matter of moral education.  Similarly, just as some people may not have the concept of electricity or, more commonly, do not know, or are not sensitive to, all the main evidences of its presence, the same may hold for wrong-doing. 

Epistemic peers as idealized disputants

      

One might accept all this and still note that rational persons can disagree in moral matters even when they do have essentially the same relevant concepts and are focusing on the same proposition.  Executives may disagree on what bonus is merited by a good employee; legislators may disagree over whether polygamy should be legal; parents may disagree on what punishment is deserved by a wayward adolescent.  These truths are important for understanding ethics, but again we must disambiguate:  there is disagreement over prima facie moral appraisal and disagreement over final (on balance) appraisal.  We find differences of judgment among rational persons over whether, say, assisted suicide is wrong on balance; but few if any rational persons disagree over whether killing people is prima facie wrong, i.e. (roughly), on balance wrong unless there is an opposing consideration, such as self-defense, of at least equal moral weight.  There is a great deal to say about the nature and resolution of moral disagreement, and I have elsewhere examined it.⁴⁴

  Here, then, I will simply summarize some of the ways in which, without skepticism, we can explain rational moral disagreement among what are called epistemic peers.

   Consider a possible rational disagreement between two people—epistemic peers—who occupy, on the matter at issue (say the claim that an interviewer was unethical), positions of epistemic parity.  Roughly, this is to say that on this matter they are (a) equally rational and equally thoughtful (in the relevant matter) and (b) have considered the same relevant evidence equally conscientiously.⁴⁵

 

By contrast with most descriptions of epistemic parity with respect to a proposition, this one explicitly requires that the relevant parties consider the proposition and do so equally conscientiously.  If we require only sharing the same relevant evidence and having the same epistemic virtues (or being equally rational in the matter, which is a similar condition), nothing follows about how fully these virtues are expressed, and it thus possible that, for instance, despite equal epistemic ability and equal possession of evidence, the parties have devoted very different amounts of time or effort or both to appraising the proposition.⁴⁶

  In that case—exhibiting a kind of epistemic asymmetry—disagreement may be readily resolved by an equally conscientious consideration of the relevant evidence.  Particularly where we are concerned with the whole range of significant cognitive disparities, it is important that parity be understood to have a non-dispositional element—for our purposes, actually considering relevant evidence. This element provides a way to account for important disparities that might not be evident in a peer disagreement in which no consideration, or only differentially conscientious consideration, of the relevant evidence occurs.

  Let me illustrate.  If I believe that a colleague with whom I disagree satisfies (a) and (b), this may prevent me from concluding that I am clearly right and my colleague clearly wrong.  It should prevent dogmatism about my own correctness.  The possibility that (a) and (b) are jointly realizable in rational moral disagreement should be granted, but it should be stressed that—as skeptics might be the first to emphasize—it is very hard to be justified in believing that someone else satisfies (a) or (b) or, especially both. The breadth, complexity, and quantity of evidence needed about the other person are great, and error in assessing that evidence is difficult to avoid.⁴⁷

It may also be difficult to determine precisely what factors, among the many that may figure in a disagreement, are relevant to the matter in question.  Evidence is constituted by facts and other elements that are relevant to indefinitely many propositions about which people may disagree, and the scope of its relevance is not written on its face. In practice, we may be at most rarely justified in believing anyone to be an epistemic peer on a given point at issue.  This applies especially to parity regarding complex moral issues.

      

Moreover, there is at least one other relevant set of variables.  One variable is potentially influential background theories a disputant may hold, such as skeptical ones or ethical theories with myriad implications for the kinds of cases in dispute.  Another variable is background beliefs, including prejudicial ones that may not surface in the discussion or in the reflections of either party, say that highly educated people are more reliable in moral matters.  Beliefs need not even be conscious to influence other cognitions.  Still another variable is difference between disputants in conceptions, especially normative ones, that may affect assessment of an issue without even coming to consciousness.  Consider the effects of reading philosophical ethics, say in Aristotle or Confucius in the virtue ethics tradition, Kant or Ross in the deontological tradition, or Bentham and Mill in the utilitarian tradition.  We are also influenced by ethically significant novels, such as those of Jane Austen, Charles Dickens, and Fyodor Dostoyevski.  Two people can be equally rational and consider the same evidence for a proposition, p, but differ in the background cognitions and conceptions they bring to the assessment of that evidence.  Such background elements may also include religious convictions or theoretical commitments, say to Calvinism, skepticism, or epistemological coherentism.  Sometimes such background factors are included in “total evidence,” but I use ‘evidence’ more specifically (their influence may also be causal rather than justificatory, in which case they are not naturally included under the term ‘evidence’).

      

Still other factors must be taken into account if we are to achieve a good understanding of rational disagreement.  In addition to the difficulty—especially in complex cases—of acquiring justification for believing someone else (a) to have considered the same relevant evidence, (b) to have done so equally conscientiously, (c) to be equally rational in the matter, and (d) to be free of background cognitions that reduce the person’s overall justification regarding p, there are at least three further factors.  One is that someone else’s disbelieving p is itself a reason, for a person who rationally believes p, to doubt that the other is correct in denying p or is a full-scale epistemic peer in the matter.⁴⁸

The second is that we are better positioned to make a critical appraisal of our own evidence and of our responses to it than of anyone else’s evidence or responses to that evidence.⁴⁹

  Other things equal, then, we are better justified in our assessment of our own basis for believing p and of our response to that basis than in our assessment of the basis of anyone else’s believing it or of anyone else’s response to that basis.

      

The third factor is that, as we check and re-check our own grounds for a justified belief that p and our responses to them, we tend to increase our justification for believing p, at least where we retain that belief (but possibly even if, say from a skeptical disposition, we do not retain it).  Insofar as we are self-critical and have justified self-trust, as some of us do, our retention of a belief after such scrutiny tends to be confirmatory. The belief survives a kind of test. Thus, the very exercise of critically seeking to establish the epistemic parity of a disputant may give one a justificatory advantage in the dispute.  Perhaps we may conclude that other things equal, making a rational conscientious attempt to establish the epistemic parity of a disputant tends to favor the conscientious inquirer, at least where one retains a disputed belief. 

Is rational disagreement a crippling problem for intuitionism?

      

These points about rational disagreement are quite general, but they have a particular relevance for ethical intuitionism, which, in positing moral knowledge not based on premises—a central element in the view—may appear to free at least many intuitive judgments of the need for evidential grounds or even to shield them from criticism.  But this is not so, nor do similar concerns undermine the case for countenancing moral perception and non-inferential knowledge yielded by it.  Intuitive moral judgments may have just the kinds of evidential grounds we have seen in the case of moral perception.  Moreover, even though intuitions are non-inferential, they may—again like moral perceptual beliefs—be defended by inferences where a need for scrutiny or justification arises.  Being held on a non-inferential basis does not preclude being defensible on an inferential basis.

      

Even if (as I doubt) a justified intuitive moral belief might not be defensible by finding plausible premises for it, inferences might be enlisted in its defense by way of clarifying it and rebutting criticism.  I have also provided (in other work⁵⁰

) a conception of self-evidence on which the claim that at least some basic moral principles have this status entails neither that comprehendingly believing them implies having indefeasible justification for them, nor that they cannot be justified inferentially, on the basis of other propositions.  This section provides a case for allowing intuitionists, like others, to retain their convictions in the face of rational disagreement provided certain critical standards are met.  These standards seem satisfiable, under some conditions, for at least a good many reflective moral agents.

      

The points just made about rational disagreement are meant to show that moral knowledge is possible despite the skeptical concerns arising from the possibility of rational disagreement between epistemic peers.  If my case is sound, it certainly supports the possibility of moral knowledge—whether of prima facie moral obligations or of overall moral obligations.  I am not ignoring the common kinds of disagreements in ethics, but these are not between people who are epistemic peers in the relevant matter.  That point eliminates some of the difficulties that attempted resolution may face. Neither this point nor any made above imply that the possibility of moral perception makes resolving moral disagreements easy or that, by educating one’s perceptual sensibilities in ethics one can become a moral expert.⁵¹

But even the existence of moral and other disagreements that are difficult to resolve does not support skepticism, nor does an objectivistic realism in ethics imply that there must be moral experts in ethics. 

      

My points about rational disagreement apply not only to empirical moral judgments, say singular judgments appraising a specific person or act, but also to self-evident moral principles, such as the kind expressing general prima facie obligation to avoid killing, cheating, lying, and promise-breaking.⁵²

Many of the moral judgments whose status is in question are based on what I have called moral perception.  These include many that are intuitive, but intuitive moral judgments also include many that are not perceptual in kind.  It is time to consider the intuitive category and its relation to moral perception. 

II. The Concept of an Intuition

      

Disagreement occurs at all levels, including that of intuition, though of course one may hope that at least where intuitions are difficult to support by argument, our disagreements in intuitions are either not common or not major for either practical or theoretical matters.  How reasonable is such a hope?

      

This question cannot be answered unless we transcend a common stereotype of intuition which represents it as a “gut response,” a kind of automatic cognitive (and often affective) reaction, not mediated by reflection or assisted by inference.  If we detach from our idea of intuition the suggestion of mere emotionality or mere personal preference, then this description may hold for some intuitions.  Suppose someone said that believing is an action. My intuitive response to this claim has always been instantly negative.  This does not make it a gut response, though perhaps it once was.  In any case, just as, in a short moment, one can take in myriad details of a painting, one can quickly appreciate a conceptual classification (such as treating beliefs as actions) and place it in a wide context.  Thus, even if gut responses typically are among those that come quickly, the rapidity of a response does not imply that it is a gut response—and certainly not a response that does not arise from a background of considerable understanding or is unjustified.

      

Rapidity of response is, in any case, by no means characteristic of all intuitions.  Some intuitions emerge only upon reflection, and even some that do arise immediately on considering something are quickly replaced by other intuitions as consideration continues or passes into critical reflection.  Think of a case in which you have to decide whether to cease life support for your terminally ill father, who has never addressed the matter.  You will reflect on his values, his reactions to relevantly similar cases among relatives, and so on.  You may need quite some time, even more than one sitting.  When you feel you have a sufficiently informed picture of his values and of what is best for him—a picture sometimes by imagining a kind of narrative of his life—it may seem to you that you should not allow further life support.  If this impression is an intuition, it is non-inferential; but as the example shows, that does not entail its being a gut response.  Nor does its being non-inferential entail that it arises independently of inferences in the background, as where one infers facts about medical consequences from what one reads about the available treatments and only then intuitively sees what should be done.  We may need ladders to ascend to a plateau, but from there we may be able to ascend to the summit by a direct path. The path may be long and laborious but may show us a wide view.  Understanding a problem, reading a passage, and taking in a painting may be similar.  Intuition, like understanding, may also come slowly and only when we have taken a long and wide view.⁵³ 

Five notions important for understanding intuition

      

An account of intuitions should clarify at least five related notions. Let me characterize each briefly and then proceed to indicate how some of the notions figure in ethical intuitionism as I conceive it.

      

Cognitive intuition. This is the most common kind of intuition considered in ethical literature:  it is intuition that p (a proposition), say that a friend deserves an apology for an offensive remark or, to take a different kind of example, that one should not accept a major gift from a former student whom one must again teach in the future.  In the latter case, at least, the intuitive proposition could be subsumed under a moral rule, say one to the effect that we should not do things that produce risky conflicts of interest.  But subsumability under a rule need not prevent a proposition from being the object of an intuition, any more than there need be one route to a given destination.  Moreover, some who would have the intuition might not accept any such rule—at least antecedently to reflecting on such cases and generalizing from them.⁵⁴

      

Objectual intuition. This is roughly direct apprehension of either (a) a concept, such as that of obligation, or (b) a property or relation, such as the property of being a promise, the property of being unjust, or the relation of entailment.  As a kind of objectual knowledge, such intuitions are in a sense epistemic. They may also be considered cognitive, though not propositional, but I prefer to reserve the term ‘cognitive intuition’ for intuitions with propositional objects. In a sense to be explored below, objectual intuitions constitute intellectual perceptions.

      

Intuitiveness. This is a property primarily of propositions, but it is also predicated of concepts, arguments, and other intellectual phenomena.  Applied to a proposition, it is equivalent to the proposition’s being intuitive—having the property of evoking (under certain conditions) what might be called the sense of non-inferential credibility (this notion may be relativized, for instance to persons of a certain description or to a certain level of understanding needed for the intuitive sense to be manifested). That sense is roughly equivalent to the sense of a proposition’s (non-inferentially) seeming true—seeming true where this impression of truth does not derive from any sense of the proposition’s being supported by a premise.  This sense of its seeming true normally produces an inclination to believe it.⁵⁵

      

Propositional intuition. This is intuition conceived as a proposition taken (by the person using the term) to be intuitively known, say that capital punishment is wrong, or that what is colored is extended.  We can imagine someone saying to colleagues on a committee writing an ethics code for a corporation, ‘Let’s jot down the various intuitions that have been expressed on what counts as harassment and see if we can frame a definition worth putting into our code of ethics’. Here, intuitions are constituted by propositions, though in these contexts the term may be loosely used in a way that is not factive and implies only a strong presumption of truth.  Hence, unlike cognitive intuitions, propositional intuitions are abstract, non-psychological elements and (on most but not all uses) include only truths.⁵⁶

      

Apprehensional capacity. This is intuition as a rational capacity—facultative intuition, for short—a kind needed for philosophical reflection and manifested in relation to each of the other four cases. It is roughly a non-inferential capacity by whose exercise what is intuitively believed or known is believed or known.  Ethical intuition, like logical intuition, is a special case of the output of this capacity.

      

Cognitive and objectual intuition will be my main concern. But the other notions should be kept in mind; we gain clarity by distinguishing them from the former two concepts of intuition and connecting them with those when necessary.⁵⁷ 

The analogy between intuition and perception

      

There is much to be learned from conceiving facultative intuition as analogous to the “faculty” of perception and, correspondingly, intuitions as analogous to perceptions.  Take seeing as a paradigm of perception.  Intuitive apprehension is analogous to seeing. First, it has objects, typically, or at least most importantly for our purposes, concepts and properties.  (We need not here discuss what kinds of objects these are, but I assume they are both abstract and in some way connected with linguistic usage, and they are surely accessible through understanding that usage.) Second, in addition to simple apprehension—apprehension of—there is also apprehension to be (attributive apprehension), and apprehension as: aspectual apprehension. The former, apprehension simpliciter, is de re (roughly, of the thing that is its object) and apparently does not entail cognition, as where a small child apprehends the wrongness of smashing a toy but does not have a propositional belief that this is wrong or even see it as wrong.  Attributive and aspectual apprehension entail at least some degree of understanding of the property the object is apprehended to have, or as having, say the wrongness of smashing the toy.

      

The third point here concerns the connection between an apprehension and an intuitive seeming.  Given an apprehension of a type of act, as where we are considering what to do and envisage various possible acts, it may also intuitively seem to us that (for instance) the kind of act in question is wrong in general.  This is a propositional seeming.  Fourth, we may, on the basis of this seeming, believe that such acts are wrong.  Fifth, if this belief is true, then we intuitively see (apprehend) that the act is wrong.  Such true doxastic intuition, a case of intuitive belief, would at least normally constitute intuitive knowledge. The same points hold for an apprehension of a concrete act, though apprehension implies not mere observation but having at least a sense of certain properties of the act, say its being a promise-breaking or a truth-telling in the face of temptation to lie.

      

We thus have, with intellectual perception as with sensory perception, simple, attributive, aspectual, and propositional perception: (1) apprehensions of the relevant orject or  kind of object, such as the act-type, lying; (2) apprehending the object to have a property, say seeing an interview to be unfair, (3) apprehensions of the object as something, for instance of an act as wrong; and (4) apprehensions that it is something, say, is unjust. Simple and attributive apprehensions have the same kind of veridicality as perceptions of objects; aspectual and propositional apprehensions may or may not embody beliefs; but, when they do, they commonly represent knowledge.

      

Are there, as the perceptual analogy suggests, apprehensional counterparts of illusion and hallucination?  Might someone mistakenly (illusorily) apprehend the property of redness as coming in precisely discrete shades in concert with its discrete wavelengths?  This would be a misapprehension but could still be of red, somewhat as visually misperceiving a glass’s round rim as elliptical is still seeing that rim. In neither case, moreover, need the illusion produce false belief. An analogue of hallucination is more difficult to delineate.  Mere possibilia, such as mythological beasts like chimeras, are apprehensible; thus, if we assume that hallucinatory perceptions have no object at all, we must focus on something like thinking of a round square.⁵⁸

How could one apprehend “round-squarely,” even granting that no false belief is implied? Perhaps someone could apprehend each of the elements with a false sense of their unity and perhaps also a sense of properties consequential on or otherwise closely connected with each.  One could perhaps speak of a hypothetical object here, analogous to an object that might be posited for hallucinations, but this is not the place to carry the idea further.  It is enough that the analogy between intuition and perception is extensive.  Let us now explore its application to justification.

      

Where something’s seeming to have a property is a kind of intellectual impression and not a belief, we may speak of an intuitive seeming.  An intuitive seeming, like a sensory seeming of the kind that occurs when one views the rim of a glass from an angle, may often generate a belief with the same content; but where there is no need to form a belief or to make a judgment, and especially where one’s attention shifts after the seeming arises, a seeming may well remain just that.  But commonly you respond to such reflections aimed at arriving at an answer to a question by forming a belief that is intuitive—this is a doxastic intuition.⁵⁹

  The propositional content of a doxastic intuition will normally seem true to its possessor, but no such seeming is a necessary precondition for forming a doxastic intuition.

      

An intuitive seeming can evidentially support, and can causally produce or sustain, a doxastic intuition, but not every intuitive seeming does this.  A counterexample to one’s view may be like this.  It may be felt as seeming to indicate a truth, but not evoke one’s conviction, at least not immediately. The more we have invested, the more we tend to resist devaluation.  Moreover, not every doxastic intuition arises from or is based on an intuitive seeming, though it would be unusual (if it is possible at all) to have an intuitive belief, as opposed to, say, an inferential belief, which is, for the person in question, perceptibly inconsistent with an intuitive seeming.  An intuitive seeming implies a felt inclination to believe; the inclination is characteristically based on a phenomenal sense of something's having a property.  It is analogous to a perceptual inclination to believe, as where, seeing a property inclines to believe something to the effect that the object in question has it. Beliefs, including doxastic intuitions, that are perceptibly inconsistent with the truth of an intuitive seeming would tend to create a sense of tension. That tension is a kind that a rational person tends to avoid and, if it should arise, to try to resolve.  Resolution often occurs through the person’s giving up one or the other intuitive element, but such tensions may also lead to giving up a standing belief. (I will say more about this relationship in discussing moral intuitions.)

      

As already suggested, on any plausible conception of intuitions, they are non-inferential.  This might appear to be obviously so for intuitive seemings, since they may appear to be the wrong kinds of elements to be premise-based.  But perhaps sense can be made of one seeming's being based on another in at least a quasi-inferential way, as where a person seems anxious on the basis of seeming to be preoccupied with possible failures in a plan you are jointly making. Should we say, then, that a non-doxastic seeming could be inferential at least where the premise is another seeming?  This would be at best misleading.  Where one such seeming is based on another of the same kind, we have a close analogy, not to one belief’s being inferentially based on another (a premise-belief), but rather to the relation that non-basic perception bears to basic perception.⁶⁰

  If someone’s seeming to me upset is based on the person’s seeming to me fidgety, I need not have an inferential belief or inferential seeming as a result and certainly need not have actually drawn an inference (an event in consciousness). Rather, the seeming fidgetiness is part of a pattern to which this seeming to be upset is a response. The most important point about intuitions in relation to inference is that they are direct responses to something the person sees or otherwise senses or considers, not indirect intellectual responses to a premise that, for the person in question, is in some sense prior to the intuited proposition. The complexity of the object, pattern, or even narrative, to which intuitions (including seemings) respond does not imply that they are inferential.  Complexity in what one perceives or considers does not imply forming premise-beliers regarding it.

   This ascription of non-inferential directness to intuitions goes well with the paradigmatic status of intuitions whose objects are luminously self-evident axioms, though there are also clear cases of intuitions whose contents are not self-evident.  Among those who use the notion of intuition discriminatingly, it is widely agreed that it is by intuition that we see the truth of such self-evident propositions as that if all human being are mortal, and we are all human beings, then we are all mortal. 

The self-evident and the obvious

      

In part because the paradigms of self-evident propositions are both intuitive and obviously true, it is important to see that neither obviousness nor even intuitiveness is essential to them.  Consider the proposition that the mother-in-law of a spouse of a person's youngest sibling is that person's own mother.  This takes most people some time to see to be true, but it is evident in itself, without the need for a premise. It is thus not obvious (even for the same people who can see its truth and come to find it self-evident). That point is confirmed by how intuitive it seems to us once we see how it is true.  A self-evident proposition, then, can be far from obvious and yet seem obvious to us once we see its truth on the basis of an adequate understanding of it.  The concept of self-evidence in question—and most appropriate in ethical theory—is this: 

Self-evident propositions are truths such that (a) in virtue of adequately understanding them one has justification for believing them (which does not entail that all who adequately understand them do believe them), and (b) believing them on the basis of adequately understanding them entails knowing them.⁶¹

  

      

This account makes it easy to see not only why a self-evident proposition need not be obvious but also why it may be withheld or even disbelieved by at least some people.⁶²

  For even when a proposition that is not obvious is considered by someone, it may not be psychologically compelling for that person, as are, for most normal adults, such simple analytic truths as that if x and y are people of the same height and x is five feet tall, then y is also five feet tall.  A psychologically compelling proposition is one such that adequately understanding it and simultaneously considering it entails believing it.  The obvious self-evident truths are compelling.  Such examples as my in-law case make it clear, however, that we can comprehend a self-evident proposition and still need time to reflect on its content, say on the relationships it expresses, before we believe it.  Clearly, then, not all self-evident truths are initially intuitive, that is, intuitive on first comprehending consideration.  I doubt that all of them must be capable of being intuitive at all, but if that is so it is not inconsistent with anything I maintain.

      

Return now to the case of the intuition, formed after much thought, that, we should discontinue life support for the terminally ill patient.  Not only is the object of this intuition not self-evident (which it could not be given that the self-evident is a priori);⁶³

we must also grant that the belief in question represents the person's conclusion in the matter.  There are, however, at least two kinds of conclusions:  conclusions of inference, which are premise-based, and conclusions of reflection, which are properly so called because they conclude or wrap up a matter on which one has reflected but, in my terminology, are non-inferential.  The latter may be as direct as a master conductor’s concluding verdict, concerning a violinist playing in a mediocre way at an audition, that the violinist should not be employed.  The conductor may have to listen for several minutes, but the intuitive conclusion may be based on an overall response, not on such premises as that the violinist rushed through the delicate passage in the middle—there may indeed be no such premises in the judge's mind. A piece played without “mistakes” is not thereby played well. A performance having only parts that are beautifully played, like a painting composed only of beautiful parts, may fail to be beautiful. 

III. Intuitions as Apprehensions

      

Some of my examples have indicated that moral intuitions can have general propositions as their objects, for instance the proposition that there is a prima facie obligation not to kill people (which is non-existential in that its truth does not strictly require that people exist).  There are also philosophical intuitions, say that believing is not a process.  It is plausible to take such general and philosophical propositions to have as constituents (even if not their only constituents) the concepts they are in some sense about, at least in the sense that they reflect facts concerning those concepts.  One kind of intuition is what I have called simple:  apprehension of concepts or of properties or relations, where these are viewed as abstract entities.  We may, for instance, say, of poor logic students that they have trouble apprehending the concept of validity or have too few or misguided intuitions about specific entailments (an analogue of perceptual illusions, wherein there is contact with the object perceived but a false impression about it); and we may say, of good students in ethics, that they readily apprehend fittingness relations.  Here the analogy is to simple perception, but in the case of intuition the objects—entailment and fittingness relations—are abstract.

      

If the analogy between intuitive apprehension and moral perception is as far-reaching as I am supposing, then just as one can see the injustice of a deed or the goodness in a person, one may apprehend the concept of injustice or that of personal goodness.  Indeed, might the perceptions of the corresponding properties be a normal route—perhaps the only ordinary route—to apprehension of the corresponding concepts?  Here is how that might be so.  Apparently, from a conceptual and epistemological point of view, there is a progression—one that seems to correspond to normal human development—in which the first stage is sensory acquaintance (normally of a perceptual kind) with properties and an ability to discriminate some from others; the second is conceptualization of the properties in question, a stage that makes it possible to apprehend them in a conceptual way; and the third is framing propositions in which the concepts figure—exercising the concepts, we might say—as where one forms a belief that, for instance, hitting other people is wrong.⁶⁴

The process may be arrested at any stage, but it is commonly completed.

      

Where moral perceptions and, especially, moral intuitions are in question, it may strike some people that my account is too heavily cognitive and that motivation should be given a larger role in understanding moral intuition.  We normal moral agents are, after all, motivated to avoid what we intuitively believe to be injustice and we are motivated to seek, maintain, support, honor (or the like) what we intuitively believe to be good.  I see no need to build motivation into the account of either moral perception or moral intuition, but certainly I have made room for a strong if contingent connection between moral intuition and motivation to act accordingly.  There may be a non-contingent connection between these in rational persons, or at least rational persons in whom there is a high degree of integration between cognition and motivation.⁶⁵

Some moral intuitions and many moral perceptions—which have the phenomenal vividness that goes with perception and that need not belong to intuition—are no doubt motivating.  This may have theological or evolutionary significance.  By the grace of God or evolution, or both,⁶⁶

we may be sensitive to at least some kinds of (intrinsically) good and (intrinsically) bad things, and naturally attracted to the former and averse to the latter. But that aspect of our psychological and epistemic constitution is another topic.

  If moral intuitions differ from moral perceptions in their phenomenal character and vividness, they may in many cases be analogous to perceptual beliefs in their “naturalness,” especially where they respond to paradigms of the kind important in moral education.  Indeed, it may be that just as one cannot help forming certain beliefs upon having certain perceptions, say believing that a growling dog is running toward one if one hears and sees it doing this, so one cannot help having the intuition that someone is doing a wrong if the person is one sees the person place greased banana peels at the top of a stone staircase in a public square.

      

If, moreover, we reflect on some of the major stages in moral education, we encounter something that bears on the contrast between nominalistic empiricism and Platonistic rationalism: the ability to generalize, often acquired from observing a single paradigm.  Consider a child who takes a model airplane from a friend without permission and is seen by its parents playing with it at home.  Scolded with a stern utterance of ‘This is stealing! How would you like it if he took one of your cars?’ and ordered to return it to the friend, the child may, with little or no explanation of what stealing is, realize that one must not just take things from people and that returning the plane is required.  The child may even see, or be disposed to realize upon thinking about the matter, that one cannot, for instance, return the plane and take a model truck in its place, or return it and then, by way of replacement, take a toy from a different friend.

      

The generality of what is learned from experiencing a paradigmatic instance of a concept is commonly almost unlimited.  The same child might, perhaps a year or two later, recognize, as stealing, one student's copying a line of poetry composed by another and handing it in as the student’s own.  This would illustrate cross-categorial generalization from the material to the intellectual.  Whatever the explanation of our capacity for generalization, it is a capacity that partly underlies the importance of both perception and intuition in recognizing instances of moral properties. 

      

***

      

If there is, as I have contended, both perceptual and intuitive knowledge in moral matters, and if the two kinds of knowledge are related as I have maintained they are, then not all our singular moral knowledge, or even all moral cognition regarding specific acts or particular persons, is an application of principles. Indeed, not all moral knowledge even depends on believing principles. Given that so many moral properties are commonly instantiated in virtue of perceptible base properties, and given our intuitive responsiveness to these base properties and to moral properties themselves, moral knowledge need not wait for deduction from premises.  It is often prior to deduction. Much moral knowledge rests on perception, on intuition regarding ethical questions about actual or projected action, or on intuitions concerning hypothetical cases that help us decide what is just or unjust, right or wrong, permissible or obligatory.  Indeed, I doubt that the categorical imperative or the principle of utility would seem plausible apart from confirmation by intuitive results of their application in concrete cases.  Moral knowledge is commonly not derived from application of such master principles.  In this respect, it seems to be like aesthetic knowledge.

      

The comparison between moral and aesthetic knowledge and other aspects of the similarity between moral and aesthetic cognition will a main concern of the next chapter. It should be clear at this point, however, that in some important if elementary cases, moral knowledge is perceptual.  It may certainly arise from perceptually grounded knowledge by a bridge from the descriptive to the normative—especially from a discernment of the descriptive properties that ground moral properties—to the moral. That bridge from the descriptive to the moral, and often from is to ought,⁶⁷

  has both the strength and the accessibility of the a priori, and, under different conditions, it can sustain both perceptual and intuitive moral knowledge.

 

      

      

Chapter 5 
 

Moral Perception, Aesthetic Perception, and Intuitive Judgment 
 

  The theory of value properly includes aesthetics as well as ethics, but too few contemporary philosophers have adequately explored the bearing of aesthetics on ethics.  The connections between the two are especially important for understanding moral perception and moral intuition.  There is aesthetic perception, as opposed to mere perception of an aesthetic object, just as there is moral perception, as opposed to mere perception of a morally significant phenomenon; there is aesthetic intuition, just as there is moral intuition; and, as in ethics, we find aesthetic disagreements that, even more than with moral disagreements, challenge the view that there are objective standards. Much can be learned about both ethical and aesthetic judgment from the comparison between the two domains.  

1. The Role of Intuition in Aesthetic Experience

   

Many aesthetic experiences are perceptual:  hearing a sonata, seeing a painting, feeling a sculpture.  But not all aesthetic experiences are perceptual. Reading a poem is different.  Aesthetic experiences in reading poetry depend on visual perception, but are not themselves perceptual, much less visual—unless it is because the poem evokes visual images of aesthetic significance for the reader.  The experience may indeed be much like the experience one has in rewardingly “hearing” the poem in silent soliloquy.  To be sure, we might speak of inner perception of the mentally uttered lines; but this is not perception in the ordinary sense.  It is, however, experiential, and the aesthetic experience is a response to what is experienced.  We can best understand the role of intuition in aesthetic experience if we focus on the broad notion of an aesthetic response rather than on aesthetic perception, which—even if it is a central case of aesthetic experience—represents a narrower category.

      

In many aesthetic responses, intuition figures importantly.  Some aesthetic responses are temporally immediate; but many, like that of the orchestra conductor listening to the competent but uninspiring violinist during an audition, take time. As suggested earlier, it is essential to see that temporal immediacy is not required for intuition. This holds even for unobvious self-evident propositions that, once they are seen to be true, are highly intuitive.  We needn’t “just see” what is, at some point in our thinking about it, highly intuitive for us.  Gradual discernment and even reflection may be required to see an intuitive truth.  In cases like this there may or may not be, at the end of one’s consideration of the proposition, a kind of naturalness in the cognition.  Conductors at an audition may mistrust their intuitions and, for that reason, allow the candidate to play longer than normal, or they may consult a colleague before making a judgment.  Intuitions differ not only in the time or reflection needed for their formation, but also in other dimensions, including, where they are propositional, the confidence levels they embody toward their propositional objects.

      

Even where intuitions arise naturally from aesthetic experience and exhibit high confidence, the irresistibility of belief formation common with certain perceptions is very often absent.  The counterpart point holds for moral intuitions formed from moral experience, but aesthetic intuitions seem, on the whole, to require more experience and education than do moral intuitions. My contrast between the way perceptual experience, whether moral or aesthetic, gives rise to intuitions and the way it yields perceptual beliefs is not meant to imply that perception must produce belief (which I think is false). The point is that, given the perceptual presence of certain non-aesthetic, non-moral properties of an object—most notably observable properties—if something or someone raises the question whether it has that property, the perceiver cannot in general resist forming the belief that it does.  Seeing a round cloud in the sky above may evoke no belief about it when, though it is in my visual field, I am concentrating on an intellectual problem; but if I am at that moment asked whether there is a round cloud above, it is at best difficult to resist forming the belief that there is.  Here to see almost certainly is to believe.

      

This point about irresistibility of belief-formation in certain perceptual cases is doubtless partly a matter of empirical phenomenology and may apply both to some cases of moral perception and to some cases of aesthetic perception.  Suppose Evelyn is both morally and aesthetically sensitive.  Hearing one person speak offensively to another might, for her, make a perception of wrong-doing irresistible; seeing a dancer wobble during a waltz might inevitably evoke an aesthetic perception of clumsiness.  These perceptions are non-propositional, but they might evoke, virtually irresistibly, intuitions that the first person is wronging the second and that the dancer is waltzing clumsily.

      

One possibility in the moral case is that the intuition is quite confident; it might also yield, perhaps after some reflection, a moral judgment with the same propositional content. That judgment in turn might express moral knowledge. Another possibility is that the perception quickly yields a moral judgment, without first producing an intuition that leads to making the judgment.  Whether, if so, the judgment is silent or expressed in a disapproving utterance is a quite different matter—heavily dependent on such pragmatic variables as the context and the relation between the observer and the offender.  Moreover, in a moral case like this, or in its aesthetic counterpart, the content of the judgment may be intuitive for the person who makes it even it does not arise from an intuition but is an instance of simply seeing something obvious without the phenomenal element that goes with intuition. 

2. Aesthetic and Moral Properties:  A Comparison and Contrast

      

The significant analogy between ethical and aesthetic perceptions and intuitions has other aspects we should note.  One concerns the relation of aesthetic properties, for instance beauty, delicacy, balance, unity, and enchantment, to their base properties—their grounds.  That aesthetic properties are consequential seems virtually as clear as the counterpart point for ethics.  Just as an action cannot be brutely wrong but must be wrong in virtue of being, for example, a killing or a lie, a painting cannot be just beautiful or a sonatina just delicate.  The painting must be beautiful in virtue of composition and coloring, say, and the sonatina must be delicate in virtue of a combination of, for instance, melody, harmony, and rhythm.  Moreover, it appears that aesthetic properties are like moral ones in being normative and in being consequential (at least ultimately) on descriptive properties.

      

To be sure, one can describe a sonatina as delicate, but the sense of ‘describe’ here is generic.  It means something like ‘characterize’ or even simply ‘apply an adjective to’.  In that wide sense one can also describe a person as scrupulously moral, but this moral property is not what philosophers have considered a descriptive one.  There is no easy way to specify just what constitutes a descriptive, as opposed to normative, property; but perhaps we can at least say that descriptive properties, when possessed by concrete objects, are causal.  This does not seem to me to apply to moral properties, and I doubt that it applies to any normative ones. That an action was obligatory, for instance, does not imply anything about its effects or causes, nor is it an ascription of a causal power to it; and the same applies to attributions of intrinsic (aesthetic) goodness to a poem or a painting.

      

Another way to see the similarity between moral and aesthetic properties is to note that there is a quite strong supervenience in each case:  no two actions or persons can be alike in all their non-moral properties and differ in their moral ones, and no two paintings or other artworks can be alike in all their non-aesthetic properties and differ in their aesthetic ones.⁶⁸

  Consequentiality (described in Chapter 2) is a stronger relation than supervenience in being a determination relation.  It is instantiated in the aesthetic domain (and others) as well as in the moral  It appears that just as an act’s obligatoriness is determined by, say, being promised, a painting’s beauty is determined, in some very complicated way, by such non-aesthetic properties as its colors, textures, lines, and shapes.⁶⁹

      

These metaphysical truths—that both moral and aesthetic properties are supervenient and consequential—seem to me a priori and necessary truths about the moral and aesthetic realms; but the analogies we have noted should not be allowed to obscure some important differences.  One is a difference in the kinds of consequentiality relations that occur in the aesthetic and ethical domains:  in the aesthetic case, as opposed to the ethical case, the relation of consequential properties to their base properties is not (at least in general) a priori, and not in all cases necessary.  We may be able to explain, by pointing to its qualities, such as fineness of description and development of a single theme, why a poem by Shakespeare is delicate or unified, but we will not in general be able to frame a priori generalizations to the effect that poems with those qualities (described in general terms) tend to be unified or, to put much the same point differently, to the effect that those qualities are delicacy-making or unity-making in the way lying is wrong-making.  One hypothesis to explain this is that aesthetic properties are, like final as opposed to prima facie obligation, organic—applicable on the basis of some non-additive combination of aesthetically contributory properties—but this is a possibility that cannot be pursued here.⁷⁰ 

3. The Rule-governed Element in Ethics and Aesthetics

      

If, in aesthetics, there are no close analogues of a priori moral principles, it does not follow that there are no aesthetic principles or rules at all.  An important question that arises here is whether there can be aesthetic rules or, if there can be, whether such rules must simply be more vague than moral ones, partly because they play a different role in our lives.  The role of artworks and even of natural beauty in human life is crucial for its quality but not for our survival in social groups.  By contrast, without people's conforming, by and large, to ethical standards at least within their own group, human life in society would be unlikely to persist. This does not hold for aesthetic properties; and, in many cases, they (as opposed to the natural properties they are consequential on) are also not as readily accessible to perception as are some important moral properties in the kinds of observable cases I have described.  Aesthetic properties are, however, similar to overall obligation in being consequential in a way that very often implies the need for training in order to perceive their presence. Consider balance, unity, engagingness, and insight.  Here is the first part of Shakespeare’s Sonnet 116: 

      

Let me not to the marriage of true minds

      

Admit impediments. Love is not love

      

Which alters when it alteration finds

      

Or bends with the remover to remove. 

      

Oh no! It is an ever fix’ed mark

      

That looks on tempests and is never shaken.

      

It is the star to every wand’ring bark

      

Whose worth’s unknown although its height be taken. 

This is fine poetry.  Compare it with this parody: 

      

Let me not to the junction of two lands

      

Omit a border wall.  Neighbors cannot each other like

      

When mixing’s indiscriminate.  Oh no! Firm hands

      

Must we have, to keep us free from strife. 

      

Walls are the mark that guides us in the night,

      

The line we trust to keep our fields in sight….   

We need not draw inferences from rules to see that this versification is bad poetry.  We can articulate defects of it, but their relation to its aesthetic deficiency is not as close as that of, say unequal sentencing of equal partners in crime is to injustice.

      

A related point about the aesthetic in contrast with the ethical is that in the former case education and experience may be even more important for the development of intuitive capacity.  But does the fact that cognitions come to people only after training entail that their justification for those cognitions is relative to that training in a sense that makes their intuitions provincial and undermines any claim those intuitions may have to cross-cultural validity?  I think not, and this can be confirmed even in mathematics, a field that is paradigmatically cross-cultural.  Mathematical education is required even to have certain intuitions; those intuitions are not thereby rendered provincial or limited in evidential significance.  In the literary realm, Shakespeare is fully accessible only to those who can comprehend his rich and subtle English; but the aesthetic value of his poetry and drama is not thereby rendered provincial.  It can be appreciated by those to whom English is not native, but simply well learned.  Indeed, substantial elements in the value of any great poetry can survive translation.

      

Much as the aesthetic value of a poem can survive translation, the moral value of certain kinds of acts can survive a behavioral analogue of translation.  Consider the (prima facie) obligation to express gratitude when someone does something good for us that is not obligatory and not easy.  In one culture a bow might be the normal way to thank someone on an appropriate occasion; in another some form of words, such as ‘thank you’ might be the norm and bowing might be considered sarcastic.  Thus, to translate ‘thank you’ into the language of the culture in which bowing is the norm for expressing gratitude, and to expect the resulting words to constitute a conventional expression of gratitude, would be in effect be ethically erroneous.  The ineffectiveness of the resulting words for expressing gratitude would not, however, imply substantive moral disagreement between the cultures.  The concept of the obligation to express gratitude is instantiable in multifarious cultural forms.  There can be good and bad ethical translations among these forms. We need not suppose that the difficulty or impossibility of translations in some cases implies that there is no cross-culturally valid standard—one expressing genuine moral value—any more than we need conclude that there is no real poem—or, more important here, no aesthetic value in the poem—as distinct from the plethora of linguistic presentations it may have.⁷¹

  

IV. The Reliability of Intuition

      

Perception is widely considered to be (as I shall assume it is) a broadly reliable “faculty,” in roughly this sense (using seeing as our paradigm):  if we see a thing, then if it visually seems to us to have a property, it usually does.⁷²

  Perception is also highly informative, in the sense that if (for instance) we see an object, there are certain things, often a great many, that we can visually know about it.  Commonly, we can discern its coloration, something about its shape and texture, its location in relation to us, and more.  Much could be said here, but my main concern is simply to explore whether ethical intuition, though not itself a kind of propositional perception, is grounded in a similar way and may, in a significant number of cases, be comparably reliable.⁷³

      

We may immediately set aside the cases of plainly factive intuition in which one both sees that some moral proposition holds and thereby has an intuitive moral belief, say that one person is being unfair to another in giving a negative evaluation of the latter's work.  For (at least normally) if one sees that something is so, one knows it is, and it is uncontroversial that cognitions constituting knowledge are a reliable kind.  This is a point that even skeptics will grant, though they may hasten to claim that we either never in fact see that something is so or, at least, never know or justifiedly believe that we do. The cases of special interest here are those in which someone considers an act or person, whether actual or hypothetical, and forms a moral intuition about the person or act.  Given the reliability and informativeness of perception, we have two questions:  How reliable is intuition?  And how informative is the basis on which it rests? 

Intuitions as responsive to grounds

      

We cannot determine, in an a priori and general way, just how reliable moral intuitions are, but once we see the extent to which discrimination and even reflection can underlie them, and once we view moral properties as such that their presence is readily discerned given perception or apprehension of their base properties, there is no reason to be skeptical about the epistemic status of all intuitions, and there is some reason to consider them often sound.  Intuitions, though not based on premises, do normally arise from supporting grounds, as opposed, say, to mere cultural conditioning, applications of stereotypes, or wishful thinking.  The main kinds of grounds on which ethical intuitions rest—in rational persons with a mastery of moral concepts—are a priori indications of moral properties, for instance properties that are a priori wrong-making, obligation-making, or good-making.  That is, the presence of these properties entails, in an a priori way and often by a self-evident connection, that the act in question is prima facie obligatory, or that the thing with a good-making element (say, pleasure in an experience) is prima facie good: roughly, good overall if there is no opposing element of at least normatively equal weight.

      

Aesthetic intuitions are not similarly grounded; there are few if any a priori connections between non-aesthetic and aesthetic properties.  This is one reason the reliability of moral intuitions should not be considered wholly on a par with that of aesthetic intuitions.  For both kinds of intuitions, however, our reliability can be enhanced by education and experience.

      

In holding this view I am maintaining an intuitionist position regarding basic common-sense moral principles, such as the prohibitions of killing, beating, and lying.  These posit a wrong-making relation between such non-moral act-properties as killing and the moral property of being wrong. The relation of such non-moral act-properties to the moral property of wrongness self-evidently obtains.  The same holds for the relation of such right-making properties as being a promise-keeping and being obligatory.  Given my account of the self-evident, it is not difficult to see how some people can be superior to others in intuitional reliability:  roughly, in the proportion they exhibit (in a broadly normal range of circumstances) of true to false intuitions. People differ markedly in the relevant kinds of discernment, understanding, and sensibility. But even apart from our success in arriving at true beliefs, at least our ethical intuitions grounded on a perception or apprehension of the base properties of the moral properties that the intuitions ascribe tend to be true.  Recall the cases of the subtly intimidating handshake and the covertly hostile interview.  The intuitions in cases like these are commonly also cases of knowledge; and when they are embedded in propositional perception, they are also instances of propositional moral knowledge. 

Intuitions regarding overall vs. prima facie obligation

      

To be sure, if an intuition is that an action is obligatory overall, or wrong on balance, as opposed to prima facie obligatory or prima facie wrong, then its reliability depends on its taking appropriate account of a complex pattern of factors.  But we often have an awareness of a pattern whose presence implies, for instance, the overall obligatoriness of keeping a promise or the overall wrongness of lying.  Think of cases in which a close friend is plainly counting on one’s veracity and will be harmed if one lies.  A ready awareness of this kind of pattern is possible in part because of what we have retained from moral experience, but the most important point here is that the number of factors relevant to determining overall obligation (and perhaps overall goodness) is often quite small.⁷⁴

      

If overall obligation is often a matter of an organic combination of grounding elements—say, a promise to do something for a friend, together with the promised deed’s offending a second friend, burdening a third, and causing one to miss a rare opportunity for self-development—then it is like aesthetic value in the complexity of the combinatory normative weight of relevant elements.  But, in addition to the differences already noted between the moral and the aesthetic in the kinds of connections that link descriptive grounds of normative properties to the possession of those properties, there is also a difference in complexity.  Whereas it is common for a moral property like obligatoriness or wrongness to have a single main ground such as a promise or the need to render aid, it is at best rare that any significant aesthetic property may be reliably ascribed on the basis of a single ground. Granted, a painting of a single shade of grey might thereby be visually dull, but this is a rare kind of painting and visual dullness is a special (and simple) kind of aesthetic property. Most aesthetic properties—especially the positive ones such as beauty, elegance, and unity—are consequential on a complex of base properties, often too numerous to list and not readily describable.  Art can be required to describe art.

      Aesthetic intuition, then, even when refined and based on extensive experience, is not exactly analogous to moral intuition. But both are responses to discernible patterns, and both can often be justified by appeal to elements in those patterns.  The similarities support the idea that intuitive responses are not results of unconscious inferences; the differences, though significant, do not entail either that justificatory inferences can never be drawn or that, even when they cannot be, the intuitions in question are merely subjective responses lacking any rational basis.

      

If Ross was correct in The Right and the Good, the essential factors determining moral obligation include considerations of non-injury, veracity, beneficence, and gratitude.  (These and other factors will be considered in some detail in the next chapter.)  I grant to utilitarianism that we may always be properly concerned with whether, in a world with as much suffering as ours has, we are doing enough for others to fulfill our obligation of beneficence. But the complexity that these variable factors introduce does not prevent our often having adequate grounds for believing that we ought overall to avoid certain harms to others, abstain from lying, and make charitable donations.

      

If these points are correct, we may also consider the capacity for ethical intuition to be an informative faculty, in this sense:  given a consideration of a morally significant act or situation, a person who understands moral concepts and is aware of the base properties for the relevant moral properties exhibited by the act or situation is in a position to have intuitive knowledge of, or at least to have justification for, propositions attributing those moral properties to the act or situation.  Roughly, in people with an adequate understanding of moral concepts, the discernment of the non-moral properties on which moral properties are consequential tends to be highly informative and is a basis for intuitive knowledge and justification. Having the concept of beneficence, if I observe someone slip on an icy sidewalk and bleed from the forehead, I will perceive an obligation to render aid and will tend to judge that I should help.  My obligation to help is consequential on the combination of relievable distress and my own ability to relieve it.  My obligation may be overridden or even canceled by the pre-emptive action of someone else; but it is initially both present and perceptible.  It may also be overridden if an even weightier obligation conflicts with it, and the overridingness may be something not perceived but intuitively seen only after reflection.  Cancelation of the obligation may result from the person’s asking me not to help, and the obligation may be eliminated (but not canceled) by someone else’s quickly taking over the responsibility to give aid.

      

Discernment of the base properties of moral properties is not the only ground of moral knowledge and moral justification.  For instance, nothing prevents us from bringing a sound moral principle (such as, for many moral philosophers, some version of the categorical imperative or of the principle of utility) to bear on a situation and arriving at moral knowledge by subsuming an act under such a principle.  But such subsumptive moral knowledge is not the basic kind, even of singular propositions.  This is not to assume without argument that conformity to the categorical imperative, for instance by acting from a maxim that conforms to it, or acting on knowledge that one is maximizing utility, are not bases of moral knowledge or of moral justification at all.  But moral knowledge by subsumption is inferential knowledge based in part on the relevant generalization; it is not a direct response to the base properties and consequentiality relations that ground the truth of that generalization. 

      

***

      

The previous chapter brought out important ways in which perceptual and intuitive knowledge in moral matters are related. This chapter showed how singular moral knowledge is like singular aesthetic knowledge in resting on a response to properties which ground the truth of the propositions that are the object of the knowledge.  This similarity transcends an epistemological  difference in the realm of general knowledge:  the aesthetic domain is unlike the moral domain in not being governed by comparable a priori principles, if by any such principles at all.  This difference does not undermine the crucial point of comparison:  in both the moral and the aesthetic cases our knowledge has an experiential basis, and in neither case need it depend upon deduction from principles. In both realms, moreover, the basis of singular judgments can be intersubjectively sharable and a ground for objective justification.

 

      

Chapter 6 

Emotion and Intuition as Sources Moral Judgment 


      Moral perception is possible for virtually any normal person with an elementary mastery of moral concepts.  It is common among people with highly discriminative moral sensibilities.  It characteristically makes many moral propositions intuitive, including many whose content matches or is supported by that of perceptually known moral propositions.  It is a major route to moral intuition.  It often yields moral knowledge.  But moral perception is by no means the only route to moral intuition or moral knowledge.  Reflection is another route.  Its subject-matter may be concrete, as where a practical decision must be made regarding life support for an accident victim; it may be abstract, as where the topic is philosophical, say a question of whether promising-breaking is always prima facie wrong; and in either case reflection may be what yields moral intuition.  Moral intuitions may also arise in a quite different way:  from emotion.  How this is so and what evidential value emotions may have in ethical matters, are central concerns of this chapter.

 

I. Emotion and Intuition: Interaction and Integration

  Intuition and emotion are sometimes associated.⁷⁵  Some emotions, such as certain misgivings, and perhaps a sense of foreboding, are at least next door to intuitions.  Intuitions, especially those constituting insights into others' consciousness, can be vivid and tied to keenly felt emotions, such as compassion.  Correspondingly, intuitive people are commonly viewed as people of feeling.  But intuitions need not be connected with emotion, as in the case of purely intellectual intuitions; and emotions, such as rage that arises from someone’s carelessly crashing into one’s car, need not be connected with intuition.  Emotion may also differ from intuition in having no definite object, if any at all, as seems to be the case with free-floating anxiety.  In the realms of contemplation and conduct, however, there is a close connection between emotion and intuition. 


The structure of emotion

      Consider moral emotions, for instance indignation, some kinds of disapproval, and certain types of resentment.  These are not uncommon in everyday life.  Are there intuitions that are equivalent to these or other emotions?  One obstacle to any such identification is that emotions are strictly speaking not truth-valued (or at best not happily considered truth-valued), whereas at least cognitive intuitions, the most common kind of intuition, are.  To be sure, we might speak of false fears where a person fears that an accident will occur but is mistaken.  It seems preferable, however, to call such a fear ill-grounded or to call the person (or the belief in question) mistaken.  Note too that ‘fear’ has a non-emotional use:  we might also speak, for instance, of fearing that there will be no fresh fruit on the menu, when no emotion is involved.  Fear is not the only emotion that can have a propositional object, as with fearing that a low-flying plane will hit one.  We can also be angry that the plane violated the law.  But there is no temptation to speak of true or false anger, except perhaps where these terms mean something like ‘genuine’ and ‘merely ostensible’).

      Granted, there are intuitions, as there are emotions, with non-propositional objects.  Take intuitions of dishonesty in an examinee or a salesperson.  Nonetheless, attributive intuitions are typically apprehensional, not emotional (though they may lead to emotion); and propositional emotions, among other emotions—say, indignation that someone did not keep a promise—are not themselves properly considered true or false, as opposed to well-founded or ill-conceived.  Even if propositional emotions might in special cases be coherently said to be true or false, the many emotions lacking propositional objects, such as rage and grief, may not be coherently described as true or false.⁷⁶

      A second obstacle to identifying emotions with intuitions is that intuitions, unlike emotions, are not intrinsically motivational. To be sure, in integrated persons, many kinds of intuitions do produce or are characteristically accompanied by motivation to act accordingly; but an intuition about, say, the validity of a logic-book argument with trivial content, need not be motivating.  In any case, intuitions admit of rationality and may be well-grounded or ill-grounded.  When Shakespeare’s Othello described himself as having “loved not wisely but too well” (Act V, scene ii), he was not just providing a poetic figure.  Loving wisely is not simply a combination of love with wisdom, and Othello is admitting a defect in his emotion, of a kind connected with its apparently narrow basis.  (Early in the play he says of Desdemona’s fascination with his account of his harrowing exploits, “I loved her that she did pity them.”) Intuitions are essential for developing an adequate moral epistemology and important for understanding emotion, moral judgment, moral perception, and other phenomena of concern in this book.

      Although I offer no analysis of the concept of emotion, I will work with a conception of emotions (including moral emotions) as multifaceted intentional elements that are typically responses to experience—whether of the outer or inner world—or to the real or merely imagined. Three kinds of constituents characterize at least most emotions; these constituents are commonly described as cognitive, motivational, and affective.  They are understood differently by different writers on emotion⁷⁷ and need brief explanation.

      The cognitive element of emotion need not be propositional, but even with objectual emotions, such as fearing a growling dog, there will normally be something believed about it, say that it will bite one.  Why only normally? We should allow that a tiny child who is not yet psychologically developed enough to form even attributive beliefs could be frightened by the growling dog.  This fright might not be full-blooded fear, but fear it could be.⁷⁸  If being startled is an emotion, and not just a kind of shock that can be akin to fear, it too has no cognitive component beyond what is required by the causative perception.  In any case, cognitive content, for emotions having it, need not be carried by propositional beliefs, such as the belief that the dog may bite me. A disposition to form such beliefs might have to be present, but that is a different psychological property. 

      There is less controversy concerning whether emotions must have a motivational component.  In experiencing an emotion, such as fear, anger, or excitement, there will be something or other one wants, say to avoid the feared thing, to strike out at the object of anger, and to experience the exciting thing.   Apathy, a sort of settled indifference, may seem to be an exception.  But is apathy an emotion?  If it is an emotion, and not the absence of emotion where one might expect it, apathy at least has a motivational dimension:  that is, we expect no motivation where we realize we would otherwise see it.  The person need not be indifferent to, say, what is genuinely frightening, but is not as easily excited and does not spontaneously feel such emotions as grief or nostalgia when, for normal persons, they are natural, even difficult to resist.

      The affective dimension of emotion is roughly its feeling dimension—a phenomenal element that contrasts with the dispositional character of belief and motivation.  Anger has a certain experiential feel; felt affection a quite different one; fear another still.  Even apathy (which I include for comparison and not because I take it to be an emotion) might have a phenomenal element—a sense of impassivity or even of the unimportance of much in life.  But whatever we say about the status of apathy, we can say that a person who cannot easily feel anything of the kind that goes with fear, anger, indignation, excitement, felt affection, disappointment, distress, and anxiety is unemotional; someone who easily feels these (or a similar range of emotions) might be considered emotional.

      Emotions are often regarded as intentional—to have intentionality as a kind of directedness toward their objects.  This is chiefly because their objects are taken to be represented under some conceptualization.  Consider first a propositional emotion: someone’s being angry that Benito fired his secretary.  This does not imply being angry that Benito fired the man who stole the petty cash, even if these are one and the same man.  The angry person might not realize that the second, incriminating description applies to the secretary.  Still—to move to a non-propositional emotion—fearing the growling dog is fearing the attacking pit bull terrier if they are one and the same.  Objectual emotions—those with a “direct object” such as a dog—like simple perceptions, can be truly ascribed using any correct description of their object. The identity of that object, as opposed to the attributive content of the emotion (or perception) is not affected by the subject’s intentionality.

      We might also speak of attributive emotions, on analogy with attributive perceptions, where the emotion is toward a thing as taken to have a property.  These emotions constitute an intermediate case, lying between simple emotions like fearing the pitbull and propositional ones like fearing that an accident will occur.  I might see an approaching bear at nightfall as I take out recycling on a farm.  Fearing such an approaching object as threatening is possible without conceptualizing the object, as is required by, say, fearing that the bear will attack, but the property of being threatening still figures in the intentional content of my emotion. Attributive emotions, then, are intentional, but not in the way propositional ones are.  This point is easily missed if one thinks that emotions are judgments and one takes judgments to have propositional objects, in which case one should also be able to predicate truth or falsity of emotions.⁷⁹

      I have already indicated that, although emotions have an essential cognitive element, they are not in general properly considered true or false.  But propositional objects are not the only elements that suffice for the intentionality of a psychological phenomenon. We may take emotions to be intentional because, at least typically, they embody beliefs, even if only attributive beliefs.  Even attributive beliefs exhibit intentionality.  Believing a distant object, such as a bear, to be dangerous, for instance, does not entail believing it to have a significant likelihood of harming someone or doing damage, even if that is necessarily equivalent to being dangerous.  The predicative contents of attributive beliefs are intentional objects in roughly the same way as the propositional objects of propositional beliefs.

      As our examples suggest, beliefs are paradigms of the cognitive element in emotions, even if that element might, in the convictional dimension, sometimes be weaker than belief, say a strong supposition (it might also be an intuitive seeming).  With some emotions, such as fearing that the tornado will hit me, the belief or other cognitive element, such as an anticipation too weak to be belief, is propositional; but it is, in many emotions, attributive.  An emotion may also embody beliefs of both kinds, such as fearing that the tornado will hit me and being terrified by the growing roar I believe to indicate its approaching.⁸⁰

      Desires (in the widest sense) are paradigms of the motivational element in emotion. Those elements may in any case be described as some kind of wanting.  Love embodies a desire for the good of the person loved; envy typically embodies a desire for something (or for a kind of thing) believed to belong to someone else (the object of envy); fear implies (typically) a desire that a felt danger be averted (or simply wanting the thing in question to change or go away).  What of free-floating anxiety?  This can unsettle a person, but its motivational power concerns more what the subject regards as relieving it than an object of worry—if it had a definite object, such as one’s vulnerable child, it would not be free floating.  It should be stressed, moreover, that there is no sharp distinction between certain aroused emotions and certain moods, such as sadness and agitation. Some cases of anxiety may be better classified as distressing moods rather than as emotions, and there are certainly borderline cases.

      We have already seen cases which suggest that the affective element in emotion is more difficult to characterize than the cognitive and motivational elements.  Perhaps we can say that it is a matter of feeling; but different emotions are associated with different kinds of feeling, and of course non-occurrent emotions, such as anger with someone who is far from one’s mind during one’s tennis game, embody only dispositions to have feelings and do not entail presently experiencing any feeling. Consider the very different feelings that go with love vs. hate, or with fear vs. delight, or with joy vs. grief.

      All emotions may exist in an occurrent form, since all can be elements present in—even dominant in—consciousness.  But emotions may also exist dispositionally, as where they are possessed in the way love for one's family is when one is wholly occupied with some professional project and does not have them in mind in any sense.  If there are exceptions, they are such cases as elation:  whereas we can be angry with someone when the emotion is in no way present in consciousness, it is at least not obvious that, when our minds are wholly elsewhere, we can be elated about, say, a result of a project or revolted by someone’s treatment of farm animals. Perhaps, however, one can be; what is perhaps not possible is being elated or revolted simpliciter in a purely dispositional way.  Note too that we can call someone an angry person and leave it at that, implying that the person is is either dispositionally or occurrently angry; but there is no comparable category of elated or delighted persons, and to call a person elated or revolted and leave it at that, with no qualifier, implies being occurrently so at the time of attribution.

      In any case, to understand emotions, we must understand dispositional psychological properties. Possession of dispositional properties is more than a capacity to exhibit—given eliciting conditions—the kinds of events that count as manifestations of those properties.  But possessing dispositional properties does not entail either a strong tendency to manifest them given those conditions, or even a definite probability of doing so given the conditions.⁸¹ Consider resentment toward a coworker and an eliciting condition for that emotion, for instance someone's asking, privately, what one thinks of the person.  There will be some tendency to say something negative; but it may easily be inhibited by, for instance, discretion.

      The relation between emotion and behavior manifesting it is typically indirect, and assigning a definite probability to any particular manifestation, such as damning with faint praise, is unlikely to be plausible.  To be sure, we can usually count on certain eliciting conditions to cause an emotion possessed dispositionally to become occurrent, as where a close friend’s question 'What do you think of him?' evokes felt resentment.  But the emotion’s becoming occurrent is not the only constitutive manifestation of the disposition, and it may be better called a realization of it. 


Intuition as a cognitive basis of emotion

   Can the cognitive component of an emotion be constituted by an intuition?  Consider indignation again.  If it seems to me, intuitively, that one person interviewing another is being unfairly inquisitive, that intuition may be the central cognition in both eliciting and sustaining my indignation toward the interviewer.  I would be indignant toward the interviewer and would naturally explain this by saying that the questioning seems, say, intrusive.  An intuitive seeming might suffice; I need not actually believe that the interviewer is being intrusive. If this just intuitively seems so, that may suffice to arouse emotion.  But certainly a doxastic intuition can also serve.  This is not to imply that only intuitive cognitions can serve to evoke or sustain emotion:  the unfairness may be so plain that one readily perceives, believes, and indeed knows that it is occurring. Here one may have not a moral intuition but simply a moral perception and perceptual moral knowledge.  But intuition may guide our thought and emotion where the evidence does not yield knowledge or, sometimes, before it yields knowledge or even belief.

   Similar points hold for anger, which is next door to indignation, as the same example shows.  What of resentment, which can be moral in content or morally based, as where it arises because of a moral judgment?  To be sure, resentment can persist after the offending party has made full reparations.  Surely intuition can serve as the cognitive component of emotion here too; but whereas the (correct) intuition that one person wronged another may justify the resentment initially, it may cease to justify it when, because the person should see that generous reparation has been made, the resentment remains.  This resentment outlives its justification.

      To be sure, if the reparation is not perceptible to the resentful person, who may reasonably think it has not in fact occurred, the person may remain justifiably resentful.  Suppose I knew that A owed reparation to B, whom A wronged.  Knowledge that such reparation is still owed would not be possible:  whether or not I can perceive reparations, if they have been made, it is false that they are still owed and so this cannot be known.  Justification, by contrast, is importantly different from knowledge, here as elsewhere.  My resentment can still be justified.  As with belief, a justified emotion may yet be objectively groundless.  This difference between justification and knowledge affects our assessment of emotion and not just our appraisal of beliefs. Some but not all emotions are, in their propositional forms, knowledge-entailing; many are factive; but few if any are constitutionally justified.   


Perceptually based intuitions

      So far, the intuitions mentioned have not been viewed as the cognitive content of moral perceptions.  But since intuitive beliefs may arise in moral perceptions, as well as with them and from them, we may say regarding all of these cases that the kind of intuitive knowledge that perception can yield may produce a cognition central in an emotion in the ways we have illustrated for non-perceptual intuitions that figure in emotions.  You might hear intimidation in the tone of what one person says to another, and, through that and your sense of the vocabulary used, you might sense that the first is wronging the second and become angry.  This sense, if properly grounded, can be a moral perception whose basis is perception of properties that ground the wrongness of the act and whose cognitive content is an intuition.  You may intuitively see that the one is wronging the other.  That same intuition, under different conditions or in a different person, can be the cognitive component of a different emotion, say resentment.  Seeing this wrong might then be the central basis of that moral emotion.

      This is not to imply that every moral propositional perception has an intuition as its content.  Seeing that someone is being unfair to a coworker by making a false allegation may be a clear case of knowledge based on conclusive evidence. That the first is wronging the second is not unintuitive, but this kind of cognition is also not properly called an intuition.  Even if it is non-inferential, there is no exercise of discernment.  Intuition is best understood as representing not a grasp of the obvious but rather a cognition of the kind that, often because a complex pattern is in view, can be the core of, or at least provide evidence for, judgment. With the perception of intimidation, for instance, there is a need for discernment, and the intuition that the interviewer is intimidating the interviewee arises in the perception of the intimidation.  With this perception one may also have the intuition that the interviewer should not be allowed to continue.  From it, a bit later, one might (perhaps retrospectively) have the intuition that the interviewee is hurt.  The same threefold distinction applies to emotion in relation to perception:  seeing one person cheat another can yield the cognitive core of anger; resentment can arise with this perception; and from resentment of a certain acute kind one can become revolted by the cheater.

      Intuitions may have causal as well as constitutive relations to emotions.  A cognition that is central in an emotion and partly constitutive of it will not be at the time in question a cause of the emotion; but the cognition (say an intuition) might have played a role in the formation of the emotion.  An instructor’s anger that a student cheated is partly constituted by believing that cheating is wrong and a betrayal of trust, even if that same belief did not cause (as opposed to being a necessary condition for) the anger of which it later became a part.  It seems clear that any of the cognitive elements just described can give rise to emotion.  They can cause it, and they can certainly sustain it.  There is simply a change of role when such elements as intuiting that an interviewer is being unfair passes from causing indignation to being an element therein.  A match that lights a fire can continue burning as part of the conflagration.  


II. The Evidential Role of Emotion in Moral Matters

      An emotion can arise in response to properties that evidence the belief, or some belief, that is essential in the emotion.  This has been illustrated with respect to indignation when it arises from intuiting unfairness in an interview. Where the unfairness is so blatant as to be seen, we have a case where a moral perception, which entails moral knowledge or the possibility of it, provides evidence for the judgment that the person in question is unfair.  The same evidence may support such moral emotions as indignation or resentment. 


The evidential potential of emotion 

      What evidential potentiality might an emotion, especially a moral one, have in its own right?  This question arises mainly when the emotion does not emerge from a cognition that either constitutes knowledge or is well evidenced and, in either of those cases, gives the emotion an element that provides evidence for a moral judgment, such as that one person is being unfair to another.

      To see how emotion can be evidential in its own right, apart from such a supportive cognition, consider a different case of indignation, one that arises in administering an oral examination.  Suppose I am judging a colleague who is conducting such an examination. I must rely on my memory of the student’s performance as well as on sensing the nuances of the appraisal the colleague is apparently making. Emotion, like perception and intuition, is often a response to a pattern.  Perhaps in such cases emotion often responds to the whole as more than the sum of the parts.  The content and style of an oral examination or an interview, for instance, may globally ill-befit the level of competence that one can expect in the kind of student or interviewee under examination. This unfittingness may be apparent during the oral examination, as one watches the individually difficult though acceptable challenges and sees the pained expressions of the candidate struggling to reply.  There may also be an unfittingness of the colleague's descriptions of what the candidate said; they may be too harsh, say by exaggerating or even ridiculing the candidate’s mistakes.  My indignation, mixed with a sense of an examiner’s demandingness ill-befitting the low level of the student, may be part of my basis for thinking that the examiner has been unfair to the student.

      It will help here to consider a quite different emotion, anxiety, which is not a moral emotion.  I was once temporarily hosted by someone acting in a disturbingly strange way.  We were sitting alone in a dining area where several kitchen knives lay on the table at which I was having breakfast.  He stared at them for a time and was silent while doing so, though the conversation resumed.  I found myself uncomfortable.  I had no belief that he might be dangerous or even that he was seriously disturbed, and I do not think that I drew any inference from anything I believed concerning his psychological make-up.  This is not to say that I could not have formed beliefs that would be a basis for having the emotion I had begun to feel.  But I later saw that my anxiety—which was likely a response to many more indications than his staring at the knives—was some evidence of his being seriously disturbed.  (Later incidents unmistakably confirmed that he was indeed disturbed that morning.)

      In this kind of case, although the emotion does not embody an intuition or a cognition that evidences a judgment, the emotion itself may play an evidential role in supporting such cognitions.  It may be a perfectly rational response to a pattern that may at least temporarily evade description.⁸²  In this and similar kinds of cases, moral emotion (among other emotions) may be quite analogous to certain kinds of aesthetic responses.  In part because it may be a rational response to what is perceived, a moral emotion may support a moral intuition, such as that one person is being unfair to another. It may also support a non-moral response, such as a belief that a person is dangerous or disturbed.  Those intuitions in turn may evidentially support moral judgments.  For some judgments, moral as well as non-moral, an emotion may be the primary support, at least initially; in other cases the emotion may be a response to an intuition or other cognition that is itself evidence, or intuition and emotion may be common responses to the same evidencing factors, as where a moral perception yields both. 


An evolutionary speculation

      One may naturally speculate here that there is an evolutionary explanation for the evidential value of emotion.  Emotions often motivate appropriate behavior, and their arising in situations of some danger might have survival value. Take fear.  Because it is highly motivating and may rapidly produce avoidance behavior, it can play a protective role that a belief that there is danger cannot as readily play.  The belief must evoke motivation; the fear has motivation as a constituent.  If fear is sufficiently often warranted by the sense of danger or threats that evoke that sense, its role in producing rapid defensive responses might be expected to give it fitness value.

      Liking—as felt attraction for another—provides a different example.  If courting behavior always had to wait upon the formation of approving beliefs about the object of pursuit, human life would be quite different.  This is not just because such beliefs—say, to the effect that the person has good character—typically do not motivate as well as does positive emotion; it is also because many relationships would not come into being at all if they were not fueled by an initial glow.  Delight may come faster than approval, approval more readily than sober positive judgment.  On the aversive side, the survival value of fear as part of a warning system is clear, and similar points hold for other “negative” emotions.  The survival value of the “positive” emotions is connected not just with their capacity to facilitate mating, but with their capacity to support cooperative behavior.⁸³

      If emotions might have a kind of fitness value in part because of their ability to evidence such phenomena as danger—whether from natural forces or from the machinations of other people—we might wonder whether they may have a kind of evidential autonomy.  In particular, are there kinds of facts which, apart from emotional evidence, one could not have evidence for or, especially, know?  I cannot see that in principle there are facts that can be evidenced or known only by emotions or emotional elements.  For one thing, emotions themselves are in large part responses to perceived phenomena, such as the behavior of others, or at least experienced elements such as angina or other ailments.  It would appear that in principle, at least, the evidential aspects of these phenomena, whether external or internal, can be captured by propositions that can be known or justifiedly believed, or by predications that can be known or justifiedly accepted, apart from emotion.  Suppose this is so.  Assume, for instance, that one person’s being unfair to another can be known through a cognitive grasp of the actions of the former toward the latter, their manner of performance, and their motivation.  It does not follow that, given our natural constitution, we in fact can know such unfairness to occur without emotional discernment. 


      ***

      The importance of emotional evidence in ethical matters is very great and is best appreciated when its relation to moral perception on one side and, on another side, moral intuition is taken into account.  Even if all the moral truths that we can know could be known apart from emotional evidence—and I do not claim this—we might well know many of them much less readily if we had to learn them through forming beliefs carrying sufficient evidential information that is independent of emotion.  The reasonable conclusion to draw here is that it is not only possible that, through the evidence of emotion, often where the emotion is connected with intuition, we sometimes know things we would not otherwise know; we also know some things more readily through that evidence, than we would have if we depended on non-emotional evidence.  This point may often hold even where, apart from emotional discernment, we would have discovered the same things.  Its importance in the moral domain will become still more apparent as we consider a wider range of moral judgments. 

 


Chapter 7 


The Place of Emotion and Moral Intuition in Normative Ethics 


      We now have before us the core of a theory of moral perception and its relation to both intuition and emotion.  Many examples have been provided to illustrate and support all three of these interrelated notions.  The theory can be further supported and clarified, however, by considering all three in relation to the kinds of moral judgments central for practical ethics.  This will require further discussion of both emotion and intuition, an illustration of how they arise in several moral domains, and a sketch of the place of moral imagination in bringing both to bear on the formation of moral judgments. 


I. Emotion and Moral Intuition

  Emotion may issue in and support intuition by responding to a pattern of factors that befit both.  This may occur in at least three ways.  Take indignation produced by witnessing subtly exploitive conduct.  First, witnessing that conduct can produce and evidence an intuition of injustice.  In a second kind of case, emotion may indirectly affect intuition, causally and evidentially.  It may, for instance, lead to reflection that supports intuition, as where one is surprised to find oneself annoyed with someone, reflects on why one is annoyed, and then has an intuition that the person's manner of speaking to one is slightly patronizing.  The intuition may arise here as a kind of conclusion of reflection (as opposed to a conclusion of inference).  It may be a phenomenal seeming and in some cases may pass into a confident intuitive belief.

      In a third kind of case, which may initially resemble the second kind, an emotion may yield premises from which we infer a proposition.  Here we arrive at a conclusion of inference.  This proposition may or may not be one that, in the circumstances, we might have inferred.  Recall my anxiety in the awkward kitchen scene.  By contrast with directly giving rise to an intuition that something is wrong with my companion, the anxiety could have led me to think that he looked at the knives too long, that he seemed nervous, and that he became strangely distant from me in the conversation.  These indications support the proposition that he is disturbed.  But they may also arouse an emotion directly, rather than through producing cognitions that elicit and propositionally justify it.

      It should be no surprise, then, that emotion can support intuition and its propositional content in the ways I have described.  Emotion is often a discriminative response to perceptible aspects of people or of things in our environment.  In this way it can magnify, unify, or extend the work of perception and thereby provide evidence concerning the person or situation perceived.  The evidence may derive its force largely from these perceptible elements that underlie emotion, but emotion may also provide a kind of evidence of its own. 


Can moral emotions discriminate among kinds of obligations?

      Moral emotion can be evidentially important even if its range in discriminatively responding to moral phenomena is narrower than that of intuition.  It is narrower, since intuitions are as diverse as the huge range of propositions constituting their contents, and there are surely more propositions having moral significance than there are distinct moral emotions.  But in absolute terms, the range of moral emotions is not narrow.  A good way to test this thesis is to consider W. D. Ross’s principles constitutive of basic moral obligation and, with those principles in view, to explore the kinds of emotions appropriate to observing their violation or fulfillment.

      Ross provided a list of apparently basic moral principles, which I here describe only briefly.⁸⁴ The principles posit eight quite comprehensive prima facie obligations, and all of them may be associated with moral emotions that, in widely varying circumstances, can evidence violation of the obligations that the principles designate or, by contrast, conformity with what the principles require. In this way, emotions may play a major role in morally important conduct and are an important topic in normative ethics.

      Let us consider the Rossian principles in what seems a natural order.  In illustrating the kinds of emotion that may be appropriate to them, we can also clarify the range of phenomenal elements that befit moral perceptions with similar content. Those phenomenal elements alone do not individuate the perceptions or fully account for their befitting the moral situations in question, but the phenomenal elements in moral perception and in associated emotions may play both evidential and motivational roles.

      (1) Justice: the obligation to treat people in accord with their merit (merit is the notion Ross stressed, though he also had in mind such notions as equal treatment and desert). Consider someone with a sense of justice who sees disproportionate distributions, say in grading papers or in apportioning food to the needy after a flood.  Such a person is likely to exhibit a negative feeling of unfittingness, or an unsettling sense of discord, or a feeling of imbalance. Here emotion evidences injustice.  On the positive side, justice may also be indicated by emotion and may also be perceived.  There is, for instance, a sense of satisfaction in giving grades that genuinely reflect actual examination scores and seminar performance marks.  Moreover, sometimes one does not settle on a difficult distribution until such an emotion arises.  It may play a confirmatory or disconfirmatory role—something later verified by the review and further information possible with the passage of time.  Felt unfittingness in a projected distribution, such as a set of grades, may produce anxiety.  The sense of justice may replace anxiety by relief.  In cases like these, emotion, when suitably responsive to morally relevant elements, can be evidence for justice as well as for injustice.

      (2) Non-injury. A second obligation Ross noted is non-injury: the obligation not to injure or harm people.  We may experience moral revulsion on seeing a man lash a child for spilling milk, or slap his wife for smiling at a friendly waiter who greets them as they reach their table.  Such violence, especially when self-defense is not in evidence, is repelling in a way that is moral and readily evokes disapproval.  There may be shock, but there is also the kind of negative response that makes criticism natural and energizes defensive or punitive treatment.  Injury and harm may also produce other negative emotions. Film footage of genocidal slaughter may evoke moral outrage.

        (3) Veracity: the obligation not to lie—to be truthful in speaking (as opposed to simply speaking truly⁸⁵). The emotion of indignation may arise on hearing a lying accusation about a coworker, told to the worker’s supervisor. Again, moral emotion may contribute to the phenomenal element in moral perception; and for moral emotion as for intuition, the sense of unfitingness can be a main basis.  The unfittingness of saying that p when p is perceptibly false is a clear case.  Thought and imagination may work much as perception does in such matters.  Envisaged action, as well as perceived action, may also evoke emotion; and here the power and vividness of a person’s moral imagination is especially important.  People who have internalized the principle of veracity may feel emotion prospectively, as where a perceived temptation to lie evokes revulsion. That revulsion may in turn produce a negative judgment opposing the lie.  Internalization of other moral standards may yield the same kind of sensitivity. On the positive side, the prospect of resisting a temptation to lie can give one a feeling of relief of tension or a sense of gratification.

       (4) Promissory fidelity:  the obligation to keep promises.  Seeing someone breaking a promise to help a friend’s child—say, selfishly spending the money given for this purpose—may arouse anger, disgust, or, where the offense is not grievous, moral disappointment, a disapproving sense of the unfittingness of the deed done to the promissory word. Experiencing promises broken toward oneself may evoke emotion even more readily. Indignation is another emotion appropriate here, particularly where there is evidently no mitigating factor. The emotion may be focused on the combination of the broken promise and the absence of any excuse; the sense of discord between deed and circumstance may here be part of the phenomenology.

      (5) Beneficence: the obligation to do good deeds, for instance in reducing the suffering, or enhancing the well-being, of others. Take the negative case first—failure to fulfill this obligation.  Imagine the anger and distress of witnessing, when one has no phone, someone’s refusal even to call for help when an injured and bleeding accident victim requests it and the only explanation is the hurry of the person who, with a cell phone in evidence, rushes away. Such evasive conduct ill-befits the victim’s salient need, and where the injury is not serious, it may evoke not anger but only an empathic sense of frustration. By contrast, imagine someone who risks danger to save a drowning person.  There is such a thing as moral admiration, a kind of pride in the good voluntarily done.  We might feel like applauding as the person pulls the victim to safety.

      (6) Self-improvement: the obligation to improve one’s character or knowledge. This is perhaps less likely to be evidenced by emotion, in part because it is self-directed and its fulfillment is often laborious, so that non-fulfillment is commonly mitigated by one’s legitimate reasons to expend energy elsewhere.  But we can be angry with ourselves for failures, and if they are moral we can feel a kind of distress; we can also feel gratified or even delighted if we sense improvement in developing some capacity we care about.  Here we might also consider a parent who discovers the children playing a video game instead of doing their homework or practicing their instruments. Even where the children have broken no promise, the parent may have a sense of their abandoning what should be their values, and a likely emotion is a keenly felt combination of disappointment and disapproval.

      (7) Reparation: roughly, the obligation to make amends for harms or injuries to others.  Imagine seeing someone who, while borrowing a computer to do an e-mail, loses an associate’s unsaved file filled with data and then does nothing—barely saying ‘sorry’ and then leaving the scene.  Here one might easily feel moral annoyance, even resentment.  The associate has been “used” in a way that approaches what Kant called treatment of a person “merely as a means.”

      (8) Gratitude: roughly, the obligation to reciprocate, at least by expressing appreciation, for good deeds done for one.  Consider someone who accepts help with a flat tire and, after the polite young man helping completes the hard work and soils his clothes, does not even thank him, saying only ‘I’m glad that is over with’.  This is an occasion for sharp disapproval, likely felt keenly.  Anger may also be natural here.  A mere acknowledgment of the successful work ill-befits the man’s generosity and sacrifice that call for gratitude.  The unfittingness here parallels that illustrated by failure of reparation.  Failure in reparation ill-befits one’s own bad-doing; failure in gratitude ill-befits someone else’s good-doing.

   In earlier work I have proposed adding to Ross’s list prima facie obligations of two other kinds:  obligations to preserve and enhance liberty (roughly, to preserve and advance liberty in human conduct) and those of respectfulness (obligations to do what one is obligated to do in an appropriate way, understood in terms of the manner of action as opposed to its type).⁸⁶  Here too moral emotion may be revealing and may produce intuition or judgment or both. Let us consider these obligations in turn.

      Regarding the obligation of liberty, consider the natural feeling of relief from moral distress, sometimes combined with a sense of moral satisfaction, when someone innocent of a crime is set free.  Similarly, we may feel threatened, or at least feel an empathic distress, when we see someone unduly restricted, as where authorities prevent a person with a minority view from speaking in a meeting.  To be sure, the sense of justice is also a possible source of emotion there, but justice need not be involved.  An unwarranted restraint, even where it is within the rights of the chair of a meeting, is enough to evoke a sense of violating an obligation to respect freedom. As this suggests, our sense of the value of liberty may be manifested even where no moral right is threatened.  Imagine someone who, out of irrational fear, cannot fly in a plane.  One ought to help here if one can.  We might feel both grateful and relieved at the good done as we watch someone patiently dispel the unfounded fears and feel that we have helped to liberate the person for flying.

      With obligations of manner, a multitude of emotions may occur.  This is because the adverbial properties in question (typically expressed by descriptive terms ending in ly) are higher-order, presupposing the act-properties they modify.  We may thus be indignant at the right deed cruelly done (done in a cruel manner); and we may be annoyed at a potentially fine gesture of reparation made clumsily.   Annoyance might overshadow the moral satisfaction we might otherwise feel, or the two emotions might mingle to yield something quite unusual.  Emotion is even more likely to be aroused when something hurtful to a person is done in a wrongful way.  Arresting offending teenagers may be necessary when they have damaged a library; but, when they are cooperating, they should not be arrested violently, say dragged into a van or spoken to in a racially or ethnically insulting way. The sight of either immoral way of making the arrest will tend to arouse negative emotions in sensitive impartial witnesses. 


Internal and external modes of emotional support

      Emotions can be evidential for intuitions or judgments in many ways.  Some are internal, others external.  Let me explain these in turn.

      When an intuition that an emotion supports is based on and colored by the experience that both generates the emotion and provides a rational basis for that emotion, the emotion supports the intuition in an internal way.  Imagine a physician who considers ordering force feeding for a dying patient in our care.  An emotional revulsion to the envisaged prospect, whether this revulsion arises before or simultaneously with an intuitive judgment that the act in question is wrong, may support that judgment.  The vivid thought of the manipulated helpless patient may produce both an emotional revulsion and an intuitive sense—or an intuitive belief or judgment—that the deed would be wrong.  The rational and sustaining basis of the emotion is an element that is central for the justification of the intuition (or other cognition).  The emotion and intuition are unified as responses to the same perceptions and thoughts. In a way, they may be mutually reinforcing.  Each is a response to the same pattern that justifies the negative judgment, and the emotion contributes to the phenomenal element in the intuition.

      The emotion may not only confirm the presence of properties that support the content of an intuition; it may also magnify evidence supporting that content. A third way emotions may support intuitions is this.  An aspect of a  emotion, such as the negative feeling that goes with moral revulsion, may also figure in the intuition itself, which may be to the effect that one person was unjust to another.  When there is internal evidential support of the intuition by the emotion, this is largely a matter of the emotion’s reflecting the property (or a property) that the intuition centrally represents, say being unjustly intrusive.  Revulsion to a wrong, for instance, may evidence the wrong-making element(s) that both evoke that emotion and support the negative intuition that the act is wrong or, like the envisaged forced feeding, would be wrong.

   An emotion can also externally support an intuition, belief, or judgment. This occurs when it operates as an element in a moral deliberative process where moral judgment and action are required.  It may also occur where emotion is a response to an abstract prospect, as where we consider thought experiments in a seminar. The same intuition of wrongness in ordering force feeding could be based on my reflection on the artificiality of sustaining life in that situation by those invasive means.  Now suppose I then note that my emotional reaction is also negative and, if I am not skeptical about the value of emotional responses, I may take this to be confirmatory.  One might think: why would I feel upset at the prospect of the forced feeding if it were really an acceptable option?  Here it is the fact that the emotion has evidential value that is crucial; in this case, my taking account of the emotion itself, and not its basis, is what, given my appropriately registering it, plays the confirmatory role.  That role may be non-inferential, as where I simply form an intuition in response to the pattern of which the emotion is a confirmatory part, or inferential, as where my possessing the intuition is a premise for a judgment that seems indicated by its presence or character.

      Broadly speaking, then, where an emotion externally supports an intuition, the support is by way of taking account of that emotion, with the emotion or some aspect of it operating as a datum that confirms the propositional content of the intuition and may be cited in support of that proposition, whereas emotions internally support intuitions when both are non-inferential responses to the same phenomena and the character of the intuition is affected by the emotion in a way that enhances its credibility in relation to the evidential facts that support both.  Where the external support provided by an emotion yields a belief or judgment, that support may be inferential, with a premise concerning the emotion (or some aspect of it) playing a major role.  An intuitive response with the same content, by contrast, is cognitive, but non-inferential; and in contrast to all of those (and other cognitive cases) the emotional response has affective and motivational elements as well, and their presence, as responding discriminatively to the same phenomena, may significantly support the intuition or other cognitive element.

      None of this implies that all emotions must confirm some intuition. Emotions may profoundly mislead, and some occur with no connection to the person’s intuitions or other cognitions, such as judgments.  But emotions are often rational and, contrary to one stereotype, they need not tend to falsify the beliefs they engender or help to sustain.

      An interesting and difficult question that arises here is whether, if one accurately observed a situation that would ordinarily arouse emotion—say the sight of one person’s patronizing another (for instance by “talking down” to the other)—one might then form beliefs which yield premises that, at least as well as the emotion of indignation, provide support for the intuition that (say) the patronizing person is being unfair to the other.  This is possible in principle.  It is the facts that ultimately evidence singular moral judgment; whether we respond to them emotionally or intuitively does not change this.  It might happen, however, that for some people the direct route from emotion to moral intuition or moral judgment is more natural than the inferential route from beliefs expressing the relevant perceived facts to the judgment; there may be other people for whom that direct route is rarely traversed or often closed.⁸⁷

      The capacities in question may vary independently: some people are more emotionally sensitive, or more intuitive than others, though a person might be emotionally sensitive yet not particularly intuitive or intuitive but not particularly emotional.  People also differ in the extent to which their moral beliefs and judgments arise from evidential beliefs as opposed to perceptual or emotional responses to the evidencing facts themselves.  It must also be kept in mind that overdetermination is possible:  an emotion can directly support an intuition at the same time one realizes facts—which may or may not be bases of the emotion—that also support it. 


II. Moral Imagination

      Most of our examples of emotions as responses to experience have been cases in which perception plays a crucial causative role.  This range of examples is appropriate given our concern with moral perception.  But the importance and pervasiveness of perceptual experiences that give rise to emotion must not be allowed to obscure the point that emotion and intuition can arise from imaginative experience and can indeed provide evidence in those cases as well as with actual perception.  Such imaginative experience can manifest what is sometimes called moral imagination. The term ‘moral imagination’ is not to be construed in the most natural way, on analogy with ‘moral person’, ‘moral virtue’, or ‘moral character’.  That construction yields a contrast with non-moral imagination, and this is not central for the notion of moral imagination in question.  That notion is not a concept of a kind of imagination, but rather of imagination in the moral sphere.

      Deliberation and moral judgment may be assisted by imaginative representation of the kinds of situation in question.  We can sometimes be as emotionally sensitive, or as intuitively insightful, when we imagine a situation calling for decision as when we actually see one person relating to another in an actual situation of just that kind.  Much as memory can preserve what we have perceived in situations of certain kinds, imagination can outline what we would perceive in situations of those kinds.

      By contrast with perception, imagination is creative.  Perception fails us if it is not faithful to its object; it is properly limited by facts.  Imagination fails us if it does not transcend its starting points; it is properly limited by (at most) possibilities.  Imagination apparently depends on the world, or at least on experience, for raw material; but it can build indefinitely many structures from that material, whereas perception is more a reproductive than a constructive faculty. Both are representational, but perception has a functional dependence on the object perceived, whereas imagination can manipulate the objects it represents—whether spontaneously or in accordance with our will.  Perception is tied to its objects; imagination, even if it requires raw material from perception, is limitlessly combinatory, often dynamic, and readily responsive to our desire for even minute alteration.

      In the moral realm, imagination can construct morally significant scenarios we have never experienced.  This is perhaps the most common kind of exercise of moral imagination, and if is a great aid to prospective deliberation about what to do.  But it may also bring to the evaluation of past deeds, whether by ourselves or by others, alternatives we have experienced in other situations.  We may wonder why, in a difficult matter, we did not do what succeeded elsewhere or why friends who failed in an attempt to do right did not do what they themselves should have observed their mentors doing.  Such reflection is capable of great subtlety if it is aided not just by memories of relevant experiences but also by imagining possible alternatives.

      If moral imagination can take us far beyond what we have perceived, it also tends (with the help of memory) to preserve that and to use it—or materials constituting elements in a perception—in facilitating judgment.  One case is this:  if one has seen a kind of wrong-doing, then imagining a case significantly similar to it may well tend to evoke moral disapproval, a sense of wrong-doing, or even an intuition that the imagined deed is wrong.  It may also evoke indignation, which may well befit the case.  The more closely the imagination “brings back” the perceived wrong-doing (or right-doing), the more reasonable it is to expect moral emotion or a moral cognition, whether intuitive or doxastic, with similar content.

      The path between moral perception and moral imagination is not, however, a one-way street.  We often imaginatively create a situation of action in order to decide what to do in a similar one.  We also imaginatively create a scenario to fill out moral narratives we read or hear, as where a friend describes ill-treatment.  Such exercises of imagination can heighten our perceptual sensitivity.  If we imagine how we would regard a kind of treatment by someone, whether we are hoping for it, fearing it, or simply envisaging it, then we may be more likely to perceive such treatment if it occurs.  Moreover, if moral intuition is evoked by what we imagine, might this not increase the likelihood of our having moral perceptions with a similar content?  These hypotheses are empirical and seem quite worth exploring scientifically.  I cannot claim to know that they hold.  What is evident even given the reflections before us here, however, is significant:  it is certainly possible that perception supplies both raw materials and cognitive inclinations to the imagination and that imaginative activity heightens our perceptual sensitivity to phenomena of similar content.

      It seems beyond doubt that imagination exercised in the moral sphere can yield intuition.  Intuition, in turn, can produce emotion, which, in ways we have seen, can support moral judgment.  Moral imagination can also yield emotion directly.  Whether or not moral imagination produces intuition, emotion, or moral judgment, and in what order, is variable; and how each of these three is related to the others in a given case is a contingent matter.  The point is that moral imagination, like moral perception, can yield and support intuition, emotion, and moral judgment. This is perhaps to be expected:  the exercise of moral imagination can, through vivid imaging of morally significant events, and through envisaging diverse possibilities, produce an experience significantly like a moral perception.

      Some people have more imagination, especially more in the way of moral imagination, than others.  This tends to make them more intuitive and more susceptible to moral emotion.  But the connection between imagination and, on the other hand, intuition and emotion is not tight.  One might be imaginative in moral matters, but not especially intuitive therein and not at all unusual in emotional sensitivity in the moral realm.  It also seems possible to have clear and reliable intuitions, in moral or other matters, even without being imaginative, or to be emotionally sensitive without being imaginative.  One would think that being highly discriminating perceptually, at least where this occurs in the moral realm, would tend to stimulate moral imagination; but there too we find only a contingent and variable connection.  Moral imagination, then, can provide evidence and can do so in a way that parallels the way moral perception provides it.  Whether, in a given person, either moral emotion or moral perception plays a large role in moral life depends on many variables.  The development of moral emotion and moral perception at a high level is surely a mark of good moral education and cultural maturity.

 

III. Intuition and Moral Judgment

      So far, I have talked of intuitions and other cognitions and of their relation to emotion, but I have not focused specifically on judgment.  The notion of judgment is among the central concepts in ethics.  I refer to judging and to judgment as the cognition one retains when one judges that p (some proposition) and then holds the judgment dispositionally.  This retention of judgment is common where we make a judgment on which no action is needed at the time, and we retains the judgment either spontaneously or with the idea of (for instance) following or expressing it later.

      One might wonder why judgment, as opposed to belief with the same content, is so important.  It is, after all, beliefs and thoughts that are the basic cognitions we act on (desire is crucial for action, too, but I leave it aside for now since desire is conative and not cognitive and raises question that cannot be pursued here⁸⁸). One reason for the importance of judgment is that judgment that p—in the sense of ‘judging’ that p—can invest the belief that p with both sharper focus and additional power to yield inference or action.  We must at least briefly entertain or focus on what we judge.  Moreover, the natural occasions for judgment commonly invite, and often demand, reflection; but we need not even think of, much less focus on, consider, or entertain, all the propositions we believe.  For many kinds of propositions we believe, and even for many we act on, we do not consider or even entertain them.

      Testimony and perception provide two major sources of examples that illustrate the contrast between judging that p and merely believing it.  When friends tell me about the sights and sounds on their recent trip, I form beliefs as naturally and quickly as they express them to me.  I do not normally consider the propositions they affirm; and I can believe what you tell me as quickly as I hear it and without focusing on it.  If I immediately accept p, q, and r as you affirm them in a narrative, I will have the passing thoughts, that p, q, and r; but this does not require entertaining or even focusing on those propositions.  Entertaining a proposition is focal and goes beyond simply having the thought of it.  A different kind of example concerns cases in which one forms beliefs upon making observations (and there are others).  When I check the tool closet for bolts, I form beliefs about their sizes and number as I observe the supply; I need do nothing that is plausibly viewed as thinking of or, especially, considering the propositions I come to know.

      This difference between judgment and belief is one factor that makes judgment more subject to critical reflection or filtering out than belief in general.  For considering or even just focusing on a proposition, especially if it complex, tends to give us a better grasp of it than simply coming to believe it as someone asserts it or as we observe that it holds.  Indeed, typically, we have no occasion to make judgments unless we see some need to consider some matter.  This typical background of the making of a judgment contributes to its special role in our understanding of cognition and action.  The judgments people make tell us something about them that we do not learn simply from knowing that they hold the beliefs those judgments express—which is not to say that we cannot learn much about people from knowing some of the beliefs they hold that were not formed through making judgments with the same propositional content.  As to action, what we do on the basis of judgment tends to be deliberate in a way not all intentional action is.  Indeed, deliberating about what to do is a major route to making judgments in the first place.

      It is not easy to say what constitutes a judgment, but we should consider this briefly to set the stage for some of the points to come.  Like the other intentional attitudes, judgment has (as already indicated) dispositional and occurrent forms. Judging that p is true, in the sense of making a judgment to this effect, is occurrent; more specifically it is a kind of doing.  Judging in the sense of appraising is also occurrent and is not only a kind of doing but also action:  this is the action or activity of observing or focusing on something with a view to making a judgment about it.  But one may also judge—in the sense of simply holding the judgment—that another person is highly competent.  Here we have judgment of a dispositional kind.  Dispositional judgment is like dispositional belief in many ways; and in the common sense of the term—‘holding the judgment’—it entails believing the proposition in question.  (The converse entailment of course fails.)

      Judging in the activity sense does not entail holding any belief of the kind broadly aimed at, say that Katherine's painting is the best (or the worst) in the gallery; but making the judgment that p apparently does entail believing p, even if the belief is immediately lost, as where one instantly realizes that p entails a false proposition or immediately sees decisive counterevidence to it.  What is not altogether clear is whether making a judgment manifests the formation of belief produced by other things, such as the evidences being considered, or whether the judging itself normally produces belief.  The former is the more plausible view, but the latter may hold in cases where the evidence seems to weigh about equally in favor of p and some contrary, and the judging is like a push that gets the mind off the fence.

   In moral matters, it is common to be, for some issues and for at least some time, on the fence: equally attracted to (or indifferent between) competing alternatives.  Notoriously, prima facie obligations, such as promissory obligations and obligations of beneficence, sometimes conflict.  This is emphasized by moral philosophers as different as W. D. Ross and John Stuart Mill.⁸⁹ Judgment is often needed to arrive at a position on what one's final (preponderant) obligation is.  What obligation is final is often unclear. We may need time to arrive at a belief, and often we make a judgment that something is so as a way of at once manifesting the formation of a belief and expressing that belief.  As this suggests, making a judgment can be concurrent with forming a belief.

      We can now see how intuitions and emotions can figure in yielding judgments of final obligation.  Sometimes one option is intuitively right, but we withhold judgment and reflect in any case, especially if the matter is important or we have reason to doubt whether we are in fact right.  Reflection can confirm or disconfirm our initial intuition.  It may do this by evoking supporting or opposing intuitions, by leading us to a theoretical analysis, by providing premises that confirm or disconfirm the initial intuition, or in other ways.  Reflection may also invoke emotion, though emotion may also be present initially.  Moreover, emotions, for instance resentment and empathy, can pull in different directions, much as intuitions can. Take cases in which we consider the consequences of favoring one option over another.  The thought of punishing a wrong-doer may yield a flush of gratification, but the thought of hurting the person might produce a guilty feeling that opposes the punitive inclination.

      In a sense, the kind of reflection I am describing, with its projection of consequences of what we do, is like using visual observation to get a better perspective on an object to be judged.  If we are sufficiently foresightful, we can see what we should do.  This is not moral perception; but it may be analogous, resting on grounds supplied by the imagination rather than by sense.  For cases in which we are actually observing a situation of moral conflict, say where two people disagree on who should have a bequest, we may be able to achieve moral knowledge, partly based on moral perception, of what ought to be done, in this instance moral knowledge of who ought to receive it.  We might perceive wrongs on both sides and, on seeing the favored division realized in the writing of checks to the deserving parties, see that justice is done.  In either case, the theory of moral perception developed here is applicable:  moral properties are consequential on the same kinds of properties whether the former are envisaged in projective imagination or discerned in an actual situation of human interaction.

      Here again the analogy between moral and aesthetic judgment is instructive.  We judge paintings and musical performances that are in our perceptual field, but we also judge imaginary plots not yet written down or melodies sung only in the mind's ear.  Intuition and emotion are relevant, as in the ethical case, though in the aesthetic realm emotion may be even more important, since a major function that is proper to at least some artworks is to please, to delight, or to evoke other feelings.  Aesthetic pleasure can, however, be taken in what does not merit it, as moral approval and disapproval can also be misplaced.  When are conditions optimal? David Hume's famous pronouncement on taste in aesthetic matters captures some of the major variables important in ethics as well: 


Strong sense, united to delicate sentiment, perfected by comparison, improved by practice, and cleared of all prejudice, can alone entitle critics to this valuable character…[and, he added] the joint verdict of such, wherever they are to be found, is the true standard of taste and beauty.⁹⁰

 

The strong sense Hume speaks of is roughly the discriminative capacity that facilitates the formation of intuitions; the delicacy of sentiment he refers to bespeaks emotional sensitivity; and the comparisons he notes include the kind relevant in ethics.

      As to the prejudice Hume warns of, I have not treated intuition, emotion, or moral judgment as immune to its influence.  But the point that they are fallible must be balanced by another:  in correcting them we must rely on elements of the same kind.  Sometimes the very content of an intuition is of a kind that can be judged only in an intuitive way.  Do we, for instance, have obligations of self-improvement, or are reasons for self-improvement simply a matter of prudence?  This cannot be answered apart from other intuitions—concerning, for instance, prudence, morality, what we “ought” to do whether we want to or not, and what is just a matter of self-interest rather than goodness or excellence.

      What we can also appeal to when conflicts arise for us among our intuitions, our emotions, or some combination of these, of course includes perception, our own and that of others.  To be sure, our sensory responses to the world, like our intuitions, are fallible; but apart from a far-reaching skepticism that seems implausibly pessimistic, we may reasonably take perception to have the kind of basic reliability implied by a good majority of the beliefs it yields being at least approximately true.⁹¹  Moral perceptions are among those we may rely on. They may evidence intuitions; they may justify emotions; and they often serve to anchor moral judgments in the objectively real world we share. 


      ***

      Emotion is not just a response to what we see, intuit, or judge. It may enable us to see more, it may evoke judgment, and it may provide evidence for judgment or intuition.  Once it is realized how discriminative and subtle emotions can be, it is easy to see how, like intuitions and judgments, emotions may, even on a single occasion, be mixed or even in conflict.  When conflicts arise for us among our intuitions, our emotions, or some combination of these, we can sometimes appeal to perception to gain further information or refine information we have.  Doing so is compatible with exercising moral imagination or building on what perception teaches us.  Our sensory responses to the world are, to be sure, fallible; but apart from a far-reaching skepticism that seems implausibly pessimistic, we may reasonably take perception to have the kind of basic reliability I have sketched.  Moral perceptions are among those we may rely on. They reflect the properties that determine moral truths.  They may both evoke and justify both intuitions and emotions; and they often provide a basis on which moral judgments can be refined and moral disagreements resolved.

 


Conclusion 


      It is time to recall some major points that have emerged in this book.  I have extended a detailed account of perception to the moral domain. Establishing that moral perception, like its non-moral counterpart, is genuinely a response to the world supports the objectivity of ethics.  Moreover, understanding moral perception can contribute to cross-cultural communication, which is now more important than ever.  My account of moral perception does not require naturalizing moral properties, though, in virtually every detail, it is consistent with doing that if it should be possible.  But the account does require that moral properties have a base in the natural world.  The bridge from that naturalistic base to moral judgment has the intelligibility of the self-evident and, very often, the reliability of necessary truth.

      Moral perception is not best considered to be a kind of simple perception in which moral properties are seen in the elementary way in which we see shapes and colors.  Simple sensory perception of natural properties is at the core of moral perception; but it is moral properties that are seen in moral (visual) perception, and it is moral propositions that are perceptually seen to be true in such cases.  Many moral properties, then, are, though not perceptual, nonetheless perceptible.  Where moral properties are perceptible, moral propositions may be seen to be true in much the way many non-moral propositions may be seen to be true:  on the basis of perceptually experiencing properties that ground their truth.  Our moral perceptions are, moreover, a major route, though by no means the only route, to intuitions.  Those perceptions may ground moral intuitions about what we perceive.  Indeed, seeing that an act or a person has a moral property may itself be a manifestation of an intuitive perceptual capacity that has considerable discriminative subtlety regarding descriptive natural properties.

      Intuitions are cognitive, but they may or may not be doxastic:  they may be phenomenal seemings rather than a kind of belief, though phenomenal seemings often yield belief and are not always clearly distinguishable from “occurrent” beliefs.  Many intuitions of either of these cognitive kinds, like the perceptions that may underlie them, are responses to patterns.  Intuitions may also be formed only upon reflection, particularly where they are responses to complex patterns.  The idea that intuitions are immediate reactions, or are more a product of feeling than of intellect, is a stereotype.  Some people are more intuitive than others, as some are more logical; but the capacity for intuition, like the capacity for logical thought, is to some extent present in all normal human beings.

      Moral intuitions may be of prima facie obligation or of overall obligation.  The same holds for permissibility and other deontic notions and for axiological notions such as that of the intrinsically good.  Whether a propositional object of a moral intuition has a prima facie or overall status, the intuition may represent either a particular action, or a kind of act, as fitting, whether to another act, a judgment, an emotion, a human relationship, or something else again.  An act, for instance, may be intuitively seen to be fitting either in relation to one element supporting it, as where it is (prima facie) obligatory only in virtue of being promised, or in relation to a pattern of elements, as where, on the basis of the direction of conduct indicated by many normative factors taken holistically in the situation, it is an overall obligation.

      Self-evident propositions may also be objects of intuition, but they are far from being the only objects of it and are not its primary objects.  Moreover, the view that we “just see” the truth of self-evident propositions—often mistakenly considered paradigms of the objects of intuition—is mistaken.  The self-evident need not be obvious or even readily understandable, much less initially intuitive—even by someone capable of seeing its truth on reflection.  The same holds for propositions that, like many singular moral propositions, are intuitive but not self-evident.  What is self-evident is knowable without dependence on premises; but its internal constitution may require reflection before its truth can be seen.⁹²

      Intuition may arise from emotion, but may also produce it.  Neither entails the other: an emotion may have nothing to do with intuition, and an intuition may be unconnected with one's emotions.  In moral matters, however, there can be a kind of wisdom of the emotions:  they can be rational responses to experience and can yield intuitive judgment.  Precisely because of the way in which emotions are often rational, discriminative responses to the properties on which moral properties are consequential, emotions may provide evidential support for intuitions.  Intuitions, especially moral ones, may also provide evidential support for emotions and may evoke them.  Without an emotional element, much intuition would not engage our motivational system as it does; and without emotion, our intuitions would be deprived of a major source of their data. Without moral perception, the objectivity of our intuitions and of the moral judgments they produce, the objectivity of ethics overall, and the scope of moral knowledge would be greatly reduced.  Moral perception not only yields moral knowledge; it also provides some of the raw material from which moral imagination constructs scenarios that help us both to arrive at moral judgments and to articulate ethical standards that can guide our judgments in the future.  Moral properties are in many cases perceptible elements in our experience, and the major task of normative ethics—a task that requires the efforts of us all—is to enhance the human capacity to discern those properties, to help us to make sound moral judgments, and to strengthen the motivation to abide by universally valid moral standards. 


Robert Audi

University of Notre Dame




¹ Detailed discussion of what constitutes naturalism and whether normative notions, such as obligation and intrinsic goodness, can be naturalized is provided in my “Can Normativity Be Naturalized?” forthcoming from Cambridge University Press in a volume edited by Susana Nuccetelli and Gary Shea.

² How rationalism may be conceived is discussed, and the position defended, in my “Skepticism About the A Priori:  Self-Evidence, Defeasibility, and Cogito Propositions,” in John Greco, ed., The Oxford Handbook of Skepticism (Oxford, Oxford University Press, 2008), 149-175; and how rationalism applies to moral knowledge is indicated in chs 1 and 2 of my The Good in the Right: A Theory of Intuition and Intrinsic Value (Princeton: Princeton University Press, 2004).

³ Here I overlap with the concerns of Jonathan Haidt in his much-discussed work on social intuitionism, “The emotional dog and its rational tail:  A social intuitionist view of moral judgment,” Psychological Review 108 (2001), 814-34, and related work by developmental psychologists who study ethics. For the latter see, e.g., Darcia Narvaez and Daniel K. Lapsley, eds., Personality, Identity, and Character:  Explorations in Moral Psychology (Cambridge:  Cambridge University Press, 2009).

⁴ Here and in discussing perception generally I draw on ch 1 of my Epistemology, 3^rd ed.

(New York and London:  Routledge, 2010). I have, however, substituted ‘attributive’ for ‘objectual’ with the idea that the former is more intuitive and can serve the same function.

⁵ These examples suggest that seeing as is conceptual. But that does not follow, and I leave open that in some cases it is not.  Non-transparency is necessary but not sufficient for the conceptual character of the position following ‘as’ in such locutions, and there may be ways of seeing x as F that simply require a determinate way of responding to something’s being F.

⁶ Seeing as may be cognitive rather than perspectival; but our subject is perceptual rather than intellective, seeing as, e.g. doxastic seeing as, in which the perceiver believes the object has the property it is seen as having. In part because seeing as does not meet the factivity standard applicable to objectual and propositional seeing, I will not discuss it in relation to moral perception.  The three factive cases are also more important for the epistemology of perception.

⁷ Doubtless there can be “blindsight,” understood as a “direct” cognitive response to visible properties unaccompanied by relevant visual experience.  But a person incapable of visual experience could have that; even a mechanical robot could, if it could have knowledge at all. We need not call either kind of knowledge seeing.  Not all knowledge of the visible is visual knowledge. For a discussion of blindsight that contrasts with mine, especially in giving a lesser role—if any essential role—to conscious elements as conditions for genuine seeing, see Tyler Burge, Origins of Objectivity (Oxford: Oxford University Press, 2010), esp. pp. 374-75 (his note 10 cites many recent scientific studies of blindsight).

⁸ We should also distinguish perceptual propositional content from sensory propositional content. Macbeth’s hallucination of a dagger has the latter but not, in my terminology, the former.  To be sure, he might have had a mixed experience:  hallucinating a dagger while actually seeing the wall before which it seems to hover.  Then his overall visual experience would have both kinds of propositional content.  Much discussion of the content of visual experience is provided by Susana Siegel in The Contents of Experience (Oxford: Oxford University Press, 2010).

⁹ Arguably, all such propositions would be included if the person had sufficient conceptual sophistication to believe or be disposed to believe every proposition attributing to the percepetual object some property of it that appropriately figures in the perceptual experience. 

¹⁰ Here one might think of the speckled hen problem. If I perceive such a creature, are all the speckles I in some sense see represented in my experience?  There is much to say here, but we need not solve the problem.  It is enough to note that anyone who posits propositional content of the kind described need not say either that the subject can know by introspection how many speckles there are or that the attempt to determine it may not change the number that are phenomenally “visible.”

¹¹ For a case supporting the distinction in question see my “Dispositional Beliefs and Dispositions to Believe,” Nous 28 (1994), 419-434.  Concerning the relation between seeing and believing, consider the locutions ‘I could not believe my eyes’ and ‘I could hardly believe my eyes’.  Both indicate how difficult it is to resist believing a proposition that comes to mind and seems to one (visually) to be true.  Shakespeare’s Othello demanded the “ocular proof” of Desdemona’s infidelity.  He surely must have thought that seeing is believing—or the surest kind of believing. (Unfortunately all he saw was an ostensibly incriminating handkerchief.)

   ¹² The points made here are supported in my “Justifying Grounds, Justified Beliefs, and Rational Acceptance,” in Mark Timmons, John Greco, and Alfred R. Mele, eds., Rationality and the Good:  Critical Essays on the Ethics and Epistemology of Robert Audi (Oxford: Oxford University Press, 2007), esp. 230-39, which replies to Timothy Williamson’s “On Being Justified in One’s Head,” 106-22. I might add here that how loose we can allow conceptualization to be depends on whether we are thinking of de dicto or only de re belief. My suggestion is that one need not even believe a rock to be slippery in order to respond to its slippery appearance by sidestepping it; but suppose explaining this response does require positing the agent’s believing it to be slippery.  That entails property attribution, but does it require the agent’s conceptualizing slipperiness?  I doubt that, but certainly the discriminative response in question is a likely stage in developing a concept of slipperiness, and some might consider the de re belief to require conceptualization of the property it ascribes.

¹³ In places Thomas Reid seems to take visual perception to entail belief regarding the object. He says, e.g., in his Inquiry into the Principles of Common Sense, uman Mind on the P“the perception of an object implies both a conception of its form and a belief of its present existence.” See Ronald E. Beanblossom and Keith Lehrer, eds., Thomas Reid: Inquiry and Essays (Indianapolis: Hackett, 1983), p. 84; and that “My belief is carried along by perception, as irresistibly as my body by the earth” (p. 85). Whether he was committed to taking propositional belief, as opposed to objectual belief, to be entailed by perception in such cases is not clear.   

¹⁴ For valuable discussion of perceptual representation different from, but in some ways supportive of, mine, see Justin P. McBrayer, “A Limited Defense of Moral Perception,” Philosophical Studies (2009). He plausibly replies to the charge that moral perception is not causal in “Moral Perception and the Causal Objection, Ratio 23 3 (2010), 201-307.

¹⁵ The most intuitive account of such seeing (an account which I take Bishop Berkeley, in his Principles of Human Knowledge, to have relied on for his phenomenalism) is that we see objects in virtue of seeing their properties.  We can grant the perspectival character of seeing, however, without taking the object seen to be a construct out of the perceptually accessible properties.

¹⁶ The position articulated here is a clarification of the one expressed in my “Moral Perception and Moral Knowledge,” Proceedings of the Aristotelian Society lxxxiv (2010), 79-97. An earlier version of the position, which contrasts it with the sense-datum view, is expressed and defended in more detail in my Epistemology, ch 2. I should add here that the property is higher-order not because it is a property of a property but because the property of sensing elliptically conceptually “contains” a reference to the less complex, first-order property of being elliptical.

¹⁷ Not all phenomenal properties are sensory, even where that term extends to inner sense, as where inner sense yields a dull pain above the eyes.  Imaging—seeing in the mind’s eye—inner speaking, and even inferring are also phenomenal.

¹⁸ A sketch of the associated account of the a priori and references to alternative views is provided in my “Skepticism About the A Priori:  Self-Evidence, Defeasibility, and Cogito Propositions,” in John Greco, ed., The Oxford Handbook of Skepticism (Oxford: Oxford University Press, 2008), 149-175. For a related exploration of the epistemic role of visual experience in yielding a priori knowledge, see Marcus Giaquinto, Visual Thinking in Mathematics (Oxford:  Oxford University Press, 2011).

¹⁹ One might wonder how the causative property could fail to be observable; but the causative token need not be observable, as opposed to intimately connected with an observable property as, e.g., tokening light-ray reflection is intimately connected (but not identical) with having a color.

²⁰ The property might be relational:  I can see a distant plane when I misperceive its color and shape but I see its approximate location and its relation to me produces a suitable phenomenal responsiveness to changes in it. I must here ignore these and other complications.

²¹ Here and elsewhere I omit discussion of noncognitivism, according to which there are no moral properties and we should speak instead of moral language. I reject that view but cannot argue the matter here.  If my overall view in this book is plausible, that will in itself constitute reason to favor cognitivism in ethics.

²² Compare, however, Mark Wynne’s characterization of “the perception of value.” See esp. ch 3 of Emotional Experience and Religious Understanding (Cambridge:  Cambridge University Press, 2005). He seems to agree with John McDowell that “we should think of values as ‘in the world’. . . on McDowell’s account, it is by way of our affective responses that we come to recognize these values” (p. 9). The reference is to McDowell’s “Non-Cognitivism and Rule-Following,” in S. Holtzmann and C. Leich, eds., Wittgenstein:  to Follow a Rule (London: Routledge and Kegan Paul, 1981).  Cf. the passages Wynne quotes from Quentin Smith and Friederich Schleiermacher (pp. 64 and 66).

²³ I do not assume that there are any properties not grounded in others, but only that the way moral properties are grounded in descriptive ones is different from that way the property of moving one’s hand is grounded in, say, intentional elements such as desires.  Even if the difference is not one of modality, conceptual or ontic, there is an epistemic difference (clarified in the text) and that is the more important one for this book.

²⁴ That moral properties are consequential is a view elaborated in G. E. Moore’s Principia Ethica (Cambridge:  Cambridge University Press, 1903) and Ross’s The Right and the Good (esp. ch 2). It is developed further in ch 2 of The Good in the Right.

²⁵ Cf. Locke’s idea that the mind at birth is a “tabula rasa”—a blank tablet waiting for the hands of experience, especially perceptual experience, to inscribe on it the information experience supplies.  Again, it is easy to tend to conceive perceptual representation cartographically.

²⁶ Compare Jonathan Dancy: “though we can discern reasons across the board, our ability to do it is not sensory; it is not sensibility that issues in the recognition of reasons (though sensibility may be required along the way); it is rather our capacity to judge. . . We might, I suppose, conceive judgement in general as a response to recognized reasons. . .” See Ethics Without Principles (Oxford: Clarendon Press, 2004), p. 144. This seems consistent with my view; and if recognizing reasons can be accomplished by discriminating base properties for a moral property central in a judgment, then if, as I hold, some cases of recognition are integrated with a certain phenomenology and an understanding of the relevant moral concepts, my view of moral perception (simple and propositional) accommodates the conception of certain (prima facie) moral judgments sketched here.

²⁷ For discussion of the sense in which perception is information processing, Fred Dretske’s Knowledge and the Flow of Information (Cambridge, MA: MIT Press, 1981) is a good source. Processing information is more than its mere reception; see Burge, op. cit., e.g. pp. 299-301, for discussion of the both notions and points concerning Dretske’s view.

²⁸ The view proposed is consistent with my ethical intuitionism developed in The Good in the Right: A Theory of Intuition and Intrinsic Value (Princeton: Princeton University Press, 2004) and elsewhere; and this paragraph indicates how to meet an objection by Sarah McGrath to the idea that there is intuitive knowledge (not all of which, to be sure, is perceptual) of particular moral facts. See her “Moral Knowledge by Perception,” Philosophical Perspectives 18 (2004), 209-229, esp. 223. 

²⁹ For related work developing a partial phenomenology of moral perception see Terry Horgan and Mark Timmons, “What Does Phenomenology Tell us about Moral Objectivity?”, Social Theory and Policy (2008), 267-300. They also explore phenomenological aspects of fittingness.

³⁰Why these kinds of knowledge should be considered non-inferential is explained in some detail in chs 3 and 7 of my Epistemology.

³¹ Here and in many other places it will be evident that I am taking moral perceptions to arouse or bear some important connection with motivation and behavioral tendencies.  My view can take account of many points about perception made by Alva Noe in his wide-ranging Action in Perception (Cambridge, MA: MIT Press, 2004), but I do not accept all the elements in his “enactive” theory of perception, e.g. that “for perceptual sensation to constitute experience—that is, for it to have genuine representational content—the perceiver must possess and make use of sensorimotor knowledge” (p. 17). Perhaps, however, this requirement is less stringent than one might think: a page earlier he says, “The enactive view insists that mere feeling is not sufficient for perceptual experience (i.e., for experience with world-representing content)” (p. 16). I agree that it is insufficient; and for me the ‘i.e.’ suggests that in determining what counts as perception, kind of content may be more important than the sensorimotor condition and may at least qualify what that condition requires. 

³² For a detailed critical discussion of the secondary quality view, see Elizabeth Tropman, “Intuitionism and the Secondary Quality Analogy in Ethics,” Journal of Value Inquiry 44 (2010), 31-45. For a contrasting sympathetic presentation of the analogy see Justin D’Arms and Daniel Jacobson, “Sensibility Theory and Projectivism,” in David Copp, ed., The Oxford Handbook of Ethical Theory (Oxford: Oxford University Press, 2006), 186-218.

³³ The Good in the Right addresses the role of inference in moral epistemology; and in Practical Reasoning and Ethical Decision (London and New York:  Routledge, 2006), esp. chs 4-6, I provide an account of the nature of inference and its relation to belief formation.

³⁴ This is argued in detail in Epistemology, esp. chs 13-14.

³⁵ For critical discussion of Nicholas Sturgeon’s “Cornell Realist” attempt to naturalize moral properties, originally presented in “Moral Explanations,” in David Copp and David Zimmerman, eds., Morality, Reason, and Truth (Totowa, NJ: Rowman and Allanheld,1985), see my “Ethical Naturalism and the Explanatory Power of Moral Concepts,” in Steven Wagner and Richard Warner, eds., Naturalism: A Critical Appraisal (Notre Dame:  University of Notre Dame Press, 1993), 95-115. That paper argues that naturalizing moral explanations is possible without naturalizing moral properties and that some of the work those explanations do is non-causal.

³⁶ Space does not permit comparing this view with moral sense theories, but I take those to be best understood as naturalizing moral properties and making them response-dependent; I do neither. For a version of this view usefully contrasting with mine see Michael Smith, “Objectivity and Moral Realism:  On the Significance of the Phenomenology of Moral Experience,” in Michael Smith, ed., Ethics and the A Priori (Cambridge: Cambridge University Press, 2004). A related theory developed more closely in relation to empirical psychology is the “construct sentimentalism” of Jesse Prinz in The Emotional Construction of Morals (Oxford: Oxford University Press, 2007), which is examined “Prinz’s Subjectivist Moral Realism,” a detailed critique by David Copp (Nous 45, 3 (2011), 577-94). An informative discussion of sensibility accounts of moral judgments, with an examination of an anti-realist version, is provided by D’Arms and Jacobson, op. cit.

³⁷ . Note that when x is a constitutive as opposed to instrumental means to y, y is not possible without x.  Smiling is not possible without the facial movements that constitute it, but telling a joke may or may not be a means to producing smiling.  To be sure, the constitutive means can have more than one cause, as can smiling itself; an artificial smile, caused by stimulating the facial muscles through brain manipulation, is still a smile. What is not contingent is the relation between the constitutive means and what it constitutes.

 

³⁸ Note that I say, “if there happens to be a level that is basic for us”; I leave open that there is none, though in any given case of perception there cannot be an infinity of levels and so must be a level basic on that occasion. This indicates my response to Jonathan Dancy, who, reacting to my point that moral properties are not observable in the sense that they are phenomenally representable, expressed doubt that “there is something elementary in all perception, so that, for instance, the mechanic’s basic awareness cannot be of the defective functioning of the water pump.”  See “Moral Perception,” Proceedings of the Aristotelian Society Supplementary Volume 84 (2010), p. 111 (this is a commentary on my “Moral Perception and Moral Knowledge,” immediately preceding it).  I agree that a mechanic can simply hear the defectiveness of a water pump. But I doubt this is a basic perception.  The mechanic surely hears the defectiveness in the way one normally recognizes a face—not by drawing an inference from an ascription of some tell-tale property but by responding to, say, a scraping at a certain pitch and loudness, an origin near the radiator, a contrast with the normal sounds, and so forth.  Granted, we could evolve so that certain complex patterns are not in this way “decomposable”—perhaps some are not. This is a contingent matter on which my theory of perception is neutral.

³⁹ That the relation between the base properties and the moral ones consequential on them is necessary and a priori is argued in chs 1 and 2 of The Good in the Right and “Skepticism about the A Priori,” but the main points in this book do not depend on that strong view.

⁴⁰ My reliabilism here concerns conditions for knowledge, not perception:  it is not implied that we perceive something only given a “reliable” connection between its having, in the circumstances,  the relevant properties and our perceiving it (though this connection cannot be, in a certain way, accidental).

⁴¹ This paragraph indicates some of the reasons why my theory of moral perception should not be conceived as a “perceptual account of moral knowledge” in general, as seems to be suggested by Carla Bagnoli in “Moral Perception and Knowledge by Principles,” in Jill Graper Hernandez, ed., The New Intuitionism (London and New York:  Continuum, 2011), pp. 101-103.  Knowledge of moral principles and subsumptive knowledge of singular moral propositions are not perceptual, though at least the latter may epistemically depend on perception.

⁴² For the notion of an illocutionary act see J. L. Austin’s seminal How To Do things with Words (Oxford: Oxford University Press,1962) and (among the many discussions since then) William P. Alston, Illocutionary Acts and Sentence Meaning (Ithaca:  Cornell University Press, 2000).

⁴³ There are still other cases of disagreement, and indeed disagreement as considered here (and in the epistemological literature) is a special case of what I call cognitive disparity. This notion and related ones are discussed in detail in my “Cognitive Disparities: Dimensions of Intellectual Diversity and the Resolution of Disagreements” forthcoming in a collection edited by David Christensen and Jennifer Lackey. An example is difference in degree of conviction.  If one person barely believes p (or simply accepts it, as with a well-confirmed but still disputed scientific hypothesis, without actually believing it) and another says, with great conviction ‘Surely p’, this may occasion illocutionary disagreement and may of course lead to content-specific disagreement as the two, in discussing the matter, develop different higher-order beliefs, say through differing ascriptions of probabilities.

⁴⁴ In, e.g., “The Ethics of Belief and the Morality of Action: Intellectual Responsibility and Rational Disagreement,” Philosophy 86 (2011), 5-29.

⁴⁵ Roger Crisp, in “Intuitionism and Disagreement,” in Timmons, Greco, and Mele, op. cit., 31-39, forcefully raises this kind of problem, and I have responded in the same volume, pp. 204-208.  I have discussed disagreement further in “The Ethics of Belief and the Morality of Action: Intellectual Responsibility and Rational Disagreement,” Philosophy 86, 335 (2011), 3-29, and my discussion of the topic here reflects that paper.

⁴⁶ Consider, e.g., a not atypical characterization by Feldman and Warfield, op. cit., meant to capture (as it surely does) a notion common in the literature: “[P]eers literally share all evidence and are equal relative to their abilities and dispositions relevant to interpreting that evidence” (p. 2). Cf. Lackey’s characterization (op. cit., 274). She presupposes in the context (as do many studies of peer disagreement) that consideration of the proposition by both parties has occurred and, often, has occurred over time and in a way that requires some thought regarding the relevant evidence.

⁴⁷ Two points may help here. First, we may use (a) and (b) to characterize parity simpliciter: above all, we widen the scope to include all times and all issues both can consider. Second, though I am discussing only rational persons, the notion of parity would apply to people who are not rational—a complicated case I cannot consider here.

⁴⁸ To be sure, unless it is rational for us to take ourselves to be rational in believing p (which is not to imply that we must in fact have this higher-order belief), we may not rationally take ourselves to have a reason to consider the disputant incorrect in denying p.  But that point is consistent with our having such a reason.

⁴⁹ The reference to evidence here must be taken to designate mainly grounds of an internal kind, such as the “evidence of the senses.”  For evidence conceived as publicly accessible supporting fact, I am not suggesting that any one person is necessarily in a better position than another to appraise it, though we may still have a kind of intrinsic advantage in appraising our response to it.  But for assessing rationality the central concern is the person’s experience, memory impressions, reflections, and other internal elements.  

⁵⁰ In, e.g., “Self-Evidence,” Philosophical Perspectives 13 (1999), 205-228.

⁵¹ For discussion of what might constitute a moral expert, see Sarah McGrath, “Skepticism about Moral Expertise as a Puzzle for Moral Realism,” Journal of Philosophy 108, 3 (2011), 111-37. The kind of realism implicit in this book can account for the difficulties she poses for the view that there can be moral experts. It should be noted that expertise even in scientific matters is neither required for achieving scientific knowledge nor an attainment guaranteed simply by acquiring a scientific education.

⁵² As I have noted in many places, the self-evident may be withholdable, even disbelievable. This is important for explaining how competing theorists, such as moral “particularists,” can deny self-evident moral principles of the kind I have defended in The Good in the Right.  How this denial is consistent with rationality is explained in my “Intuition, Inference, and Rational Disagreement in Ethics,” Ethical Theory and Moral Practice 11 (2008), 475-492. 

      ⁵³ After writing this I read a paper by Patricia Greenspan which makes the important point that “[E]motions can serve as rational barriers to discounting reasons.” See “Emotions and Moral Reasons,” in Carla Bagnoli, ed., Morality and the Emotions (Oxford: Oxford University Press, 2011), p. 40.  This holds for intuitions too; e.g., an intuition about a sick father might lead one both to recognize and to give adequate weight to reasons implicit in his values, some of which one might have difficulty articulating at the time one’s moral intuition takes shape. My account of the evidential value of emotion in Chapter 6 will show how, in the same non-inferential way, emotions can respond similarly to reasons, often in concert with intuition. But whether or not one articulates reasons, intuition may reflect discernment of their weight, much as we may take account of multiple elements in a poem or a painting as we intuitively interpret and judge it.

      ⁵⁴ Cf. Ross’s remarks on intuitive induction in ch 2 of The Right and the Good. Despite the term ‘induction’ he seems to allow for non-inferential formation of the relevant general belief arising from apprehending instances of one or more crucial properties figuring in the proposition believed.

⁵⁵ One might think intuitions are constituted by “a subclass of inclinations to believe,” as argued by Joshua Erlenbaugh and Bernard Molyneux, “Intuitions as Inclinations to Believe,” Philosophical Studies 145 (2009), 89-109. It seems to me that an entailment here is the most that can be claimed, at least on the plausible assumption that we are speaking of (occurrent) phenomenal states and may take inclinations to believe as dispositional in nature.

⁵⁶ Moore said of propositions about the good, for instance, that “when I call such propositions `Intuitions' I mean merely to assert that they are incapable of proof.” See (1903), p. x). See also Wilfrid Sellars, Science and Metaphysics (London: Routledge & Kegan Paul, 1968) for his use of ‘intuiteds’ to designate intuited propositions (this category is wider than Moore’s in that he intended to use ‘intuition’ as he here characterizes it for propositions that are not only true but also self-evident).

⁵⁷ Most of these notions are discussed in The Good in the Right, e.g. pp. 32-39, 48, and 208, note 37.  But it contains little explicit treatment of intuitive seemings.

⁵⁸ For detailed discussion of the ontological and other aspects of hallucination (as well as an account of the a priori that supports intuitionism), see chs 1, 2, 5, and 6 of Epistemology.

⁵⁹ Cf. the view of Paul Thagard and Tracy Finn concerning what they consider representative of moral intuitions: “The intuitions that killing and eating people is wrong and that aiding needy people is right are judgments of which we usually have conscious awareness” (emphasis mine). See “Conscience:  What is Moral Intuition?”, in Bagnoli, Morality and the Emotions, p. 151.  They are conceiving intuition as equivalent to (or at least entailing) a kind of judgment. I find this conception too narrow, though I agree that one can have an intuition dispositionally, as where it is simply stored in memory—and one can even have it in focal consciousness without being aware that one has it or aware of it under that description.

⁶⁰ This is not to deny that one can draw an inference and arrive at a proposition that seems to one true, but (perhaps from a skeptical disposition) one does not believe.  Nor should it be denied that one can both inferentially believe that p and find p intuitive. Finding a proposition intuitive does not require any particular route either to considering it or to believing it if one does; and inferring a proposition does not entail believing it.  It is consistent with either disbelieving it (as where one is deducing a contradiction to refute an opposing view) or, by contrast, with its seeming true to one without one’s accepting or believing it.

⁶¹ My “Self-Evidence” contains a more detailed account of self-evidence than The Good in the Right.  I should add here that we might also speak of full understanding to avoid the suggestion that adequacy implies sufficiency only for some specific purpose.  Neither term is ideal, but “full” may suggest maximality, which is also inappropriate.

⁶² Obviousness is relative in a way self-evidence is not:  what is self-evident, as my account makes clear, is justifiably believable by anyone who adequately understands it; but what is properly called obvious is understood to be readily seen to be true by a kind of person whose comprehensional level is presupposed in the context in which the obviousness is ascribed to a proposition.  We may also speak of self-evidence for a person where this means not ‘believed by that person to be self-evident’ (as in some uses) but ‘self-evident and capable of being justifiedly believed by that person on the basis of adequately understanding it’.  The latter—appropriately—makes understanding self-evident propositions relative to persons but not self-evidence itself.

⁶³ That the self-evident is a priori should be obvious from my characterization of it as knowable on the basis of (adequately) understanding it. This holds of axioms as commonly understood, and that point helps to explain why it is plausible to take the self-evident as the base case of the a priori, underlying, e.g., provability.  For an explication of this view see chs 5 and 6 of Epistemology, 3^rd ed.

⁶⁴ I have explained this developmental view in more detail in “Intuition, Inference, and Rational Disagreement in Ethics,” cited earlier.

⁶⁵ In “Moral Judgment and Reasons for Action,” in my Moral Knowledge and Ethical Character (Oxford: Oxford University Press, 1997) I have defended this point is detail and noted much relevant literature.

⁶⁶ The biological evolution is compatible with theism perhaps obvious and an explanation of its compatibility with various kinds of theism is provided in ch 10 of my Rationality and Religious Commitment (Oxford: Clarendon Press, 2011).  For a brief treatment of how morality has apparently evolved in the history of our species, see Frans de Waal, Primates and Philosophers:  How Morality Evolved (Princeton:  Princeton University Press, 2006). His discussion of loyalty (pp. 163-66), which he calls “a moral duty” (p. 165) is representative.

⁶⁷ Does the point here violate “Hume’s law,” according to which one cannot deduce an ought from an is?  I am not claiming that there are formally valid inferences of the kind in question, nor even that there are informally (“synthetically”) valid ones from descriptive propositions to ascriptions of overall (final) obligation.  The a priori connection is between descriptive propositions and ascriptions of prima facie obligation.  But prima facie obligation is important; it is certainly a moral status, and it should lead us to act in accord with it if there is no overriding set of factors. The epistemological point in question is defended in, e.g., chs 1 and 2 of The Good in the Right.

⁶⁸ Granted, historical properties are relevant to the value of aesthetic objects. But the value of an aesthetic object need not be, and certainly need not be exhausted by, aesthetic value. The beauty of a sculpture is not a matter of who shaped and textured it but of the shape and texture it has.  A perfect replica would be equivalent in beauty, grace, elegance, and so forth for other aesthetic properties. 

⁶⁹ For a detailed account of the determination relation in question see Paul Audi, “Grounding:  A Theory of the In Virtue Of Relation,” forthcoming in the Journal of Philosophy.

⁷⁰. For an informative discussion of how aesthetic properties are related to other kinds and of the extent to which attribution of the former may be appropriately subsumptive, see Richard W. Miller, “Three versions of objectivity:  aesthetic, moral, and scientific,” in Jerrold Levinson, ed., Aesthetics and Ethics (Cambridge:  Cambridge University Press, 1998), 26-58. Peter Railton’s “Aesthetic value, moral value, and the ambitions of naturalism” (59-105) also bears on the basis of aesthetic properties. It is fruitfully compared with Sturgeon’s “Moral Explanations,” cited above.

⁷¹ Much more could be said about the ontology of artworks, but there is no reason to think we must settle for an ontology that undermines the possibility of aesthetic value that is analogous in the suggested way to moral value.  Regarding the example of gratitude, if one thinks it represents etiquette rather than ethics, I would suggest that in any case the most important standards of etiquette may overlap ethical standards and may in any case be sufficiently analogous to them for our purposes here.

⁷² Notice, however, that we cannot validate perceptual beliefs without relying on perception itself.  This circularity problem has been noted by a number of epistemologists, including William P. Alston.  See his “Epistemic Circularity,” Philosophy and Phenomenological Research 46 (1986), 263-290. 

⁷³ Critical discussion of my view is provided by Walter Sinnott-Armstrong in “Reflections on Reflection in Robert Audi’s Moral Intuitionism,” in Mark Timmons, John Greco, and Alfred R. Mele, eds., Rationality and the Good:  Critical Essays on the Ethics and Epistemology of Robert Audi (Oxford, Oxford University Press, 2007), 19-30 (with a response from me on 201-204). Other relevant papers are contained in his three-volume collection from MIT Press (200x). See also Michael R. DePaul, “Intuitions in Moral Inquiry” in Copp, op. cit., 595-623. Sinnott-Armstrong raises further problems for intuitionism, though in this case concerning justification more than reliability, in his “Framing Moral Intuitions,” in… and “An Empirical Challenge to Intuitionism” in Hernandez, op. cit., 11-28.

⁷⁴ I have proposed a theory of intrinsic value in “Intrinsic Value and Reasons for Action,” Southern Journal of Philosophy XLI Supplement (2003), 30-56.  On the prospects for conceiving intrinsic value hedonistically see Roger Crisp, Reason and the Good (Oxford: Oxford University Press, 2007) and Fred Feldman…Philosophy and Phenomenological Research) …

⁷⁵ Thagard and Finn, e.g., say, “Introspection supports the claim that moral intuitions are a kind of emotional consciousness” (op. cit., p 151), though they carefully avoid implying that either intuition or emotion is irrational:  “moral intuition, like emotional consciousness in general, is not just visceral, because our overall interpretation of a scene is colored by cognitive appraisal as well” (p. 152). Although Mark Wynn does not explicitly identify emotion with feeling in the sense in which he associates the latter with intuition, he sympathetically (perhaps with full agreement) cites Schliermacher’s view that “Intuition without feeling is nothing and have neither the proper origin nor the proper force; feeling without intuition is also nothing . . . they are originally one and unseparated” (quoted in Wynne, op. cit. p 66). Wynne’s quotation from Quentin Smith is a more recent indication of the association:  “In this rejoicing I am experiencing a captivated intuition of the determinately appearing importance of global fulfillment.”  See Smith’s The Felt Meaning of the World:  A Metaphysics of Feeling (West Lafayette, IN:  Purdue University Press, 1986), p. 64 in Wynne. For a wide-ranging treatment of emotion in relation to intuition see Sabine Roeser, Moral Emotions and Intuitions (Houndmills, Basingstoke, Palgrave Macmillan: 2010).

⁷⁶ This is not to deny that there can be a proposition that expresses a kind of cognitive norm important for both the constitution and the rationality of the emotion. Here I think of what Robert C. Roberts calls a “defining proposition” for an emotion; e.g., “Relief’s defining proposition is this:  “It is important that X be in condition Y, though X was not, or might not have been, or was not known to me to be, in condition Y.” See Emotions: An Essay in Aid of Moral Psychology (Cambridge: Cambridge University Press, 2003), p. 279.

⁷⁷ A great deal has been written on emotion, much of it confirming this tripartite conception (though our examples will show that there are cases in which at least the cognitive element may be, if not absent, then indefinite.  Among the valuable book-length treatments are Robert M. Gordon, The Structure of Emotions:  Investigations in Cognitive Philosophy (Cambridge: Cambridge University Press, 1987), Robert C. Solomon, The Passions: Emotions and the Meaning of Life (Indianapolis:  Hackett,1993), Martha Nussbaum, Upheavals of Thought: The Intelligence of Emotions (Cambridge: Cambridge University Press, 2001), Roberts, op. cit.; and Linda Zagzebski, Divine Motivation Theory (Cambridge:  Cambridge University Press, 200x). A recent short but wide-ranging treatment is Jon Elster’s Explaining Social Behavior:  More Nuts and Bolts for the Social Sciences (Cambridge: Cambridge University Press, 2007), esp. ch. 2, pp. 145-61. For him emotions “do not seem to form a natural kind” (146), which I find initially plausible. He goes on to list six features that are common and important, including triggering by beliefs (147).  For reasons that will soon be apparent, I take many factors that evoke emotion to be pre-doxastic.

⁷⁸ Cf. John Deigh’s view that any account of emotion must do justice to the fact that “emotions are common to both humans and beasts. . . [though] the set to which humans are liable is much greater than the set to which beasts are liable. . .” He then illustrates: “the terror of horses fleeing a burning stable, the rage of a bull after provocation by a tormentor, and the delight of a hound in finding and retrieving its quarry are all examples.” See his Emotions, Values, and the Law (Oxford: Oxford University Press, 2008), 18.  On my view, these are not full-bloodedly emotions unless they embody beliefs; but it is difficult to say whether animals might have beliefs, and certainly they may come close enough so that the kinds of emotions they may have are at least very much like the same ones in human beings.

⁷⁹ Here I depart from cognitive theories of emotion on which, as Roeser puts it, “emotions are judgments of value” (op. cit., p. 202). Her view is, in part, that “emotions are states that are cognitive and affective . . . non-inferential, normative judgments. We can understand emotions as fulfilling the role of non-inferential judgments or intuitions . . . They bridge the is-ought gap and they let us make context-sensitive, holistic moral judgments. Paradigmatically, particular moral intuitions are emotions” (p. 204).  The judgmental view also encourages thinking of emotions as more behavioral than they are, at least insofar as judgments are made.  But presumably a judgment can arise and be held without having been made, say in an episode of appraising someone.  The view in question, then, might best be understood as taking emotions to be held judgments, rather than judgings or products of judgings.  The truth-value problem and others remain, however. Further critical points bearing on the judgment view are made by Roberts, op. cit., e.g. pp. 87-89.

⁸⁰ Arguably, fearing that p is incompatible with believing that p, but the point might be pragmatic. If I believe p, it might be odd to say I fear that p, but does one stop fearing that the tornado will hit when one sees (and thereby believes) that it will?  In any case, anger will illustrate the relevant point in the text: that an emotion can embody both kinds of belief.  Anger with a man can be partly constituted by propositional belief that he did something nasty to his wife and believing him to be likely to continue the pattern.  

⁸¹ An account of the difference between dispositional beliefs and related phenomena, with much analysis concerning the constitution belief itself, is given in my “Dispositional Beliefs and Dispositions to Believe,” cited in note 11.

.

⁸² Compare this with the strong view that “since moral judgments depend, at bottom, on how we respond emotionally to the world around us, the idea of a purely rational approach to morality is an oxymoron.  To be sure, the idea of a purely emotional approach to morality would also be a contradiction in terms; to engage in moral deliberation is to step back from one’s immediate reaction and think critically about it.”  See Gregory E. Kaebnick, “Reasons of the Heart:  Emotion, Rationality, and the ‘Wisdom of Repugnance’,” The Hastings Center Report 38, 4 (2008), 36-45, p. 36.  My position does not require dependence of all moral judgments on emotion, nor would I contrast the emotional with the rational in the way suggested here, but Kaebnick’s paper does provide useful examples of the way emotional responses may have evidential weight.

⁸³ There is much literature concerning how, from a neurobiological point of view, emotions might conduce to survival; see e.g., Thagard and Finn, op. cit.; Jorge Mall, Ricardo De Oliveira-Souza, and Roland Zahn, “Neuroscience and Morality:  Moral Judgments, Sentiments, and Values,” in Darcia Narvaez and Daniel Lapsley, op. cit., 106-35; and, for a developmental perspective, Darcia Narvaez, “Triune Ethics Theory and Moral Personality,” in Narvaez and Lapsley, op. cit., 136-58.

⁸⁴ See ch 2 of The Right and the Good for Ross’s introduction and most influential discussion of these principles. In ch 5 of The Good in the Right I discuss them further and introduce refinements and some corrections of Ross’s formulations.

⁸⁵ Two clarifications will help. First, saying something true need not constitute being truthful; the statement may be intentionally misleading or even accidentally made. Second, veracity—roughly honesty—may be mistakenly conceived as the disposition to communicate truth when a circumstance or question—say, ‘Do you believe p?’—makes that relevant.  But withholding truth even when asked what one believes on a familiar topic need not count against veracity.  Some people should be put off or greeted with silence; nor does taciturnity imply dishonesty. What count against veracity is chiefly lying and certain kinds of deception. Some tendency to cooperate with certain inquirers may be a requirement of veracity as a virtue; but to explicate all of that would require an excursion into the theory of veracity.

⁸⁶ These two prima facie obligations are introduced and discussed in some detail in ch 5 of The Good in the Right.

⁸⁷ The distinction between internal and external roles of emotions in supporting moral judgment may be clarified by considering the roles that—despite common interpretations of Kant—emotions can play in Kantian ethics.  Here Carla Bagnoli’s “Emotions and the Categorical Authority of Moral Reasons” (in her Morality and the Emotions, 62-81), which explores Kant’s resources for taking account of emotions in ethical theory, is instructive.  She treats respect, e.g., as the “emotional aspect of practical reason.”  See esp. pp. 75-78.

⁸⁸ In chs 3-6 of The Architecture of Reason: The Structure and Substance of Rationality (Oxford:  Oxford University Press, 2001) I have extensively discussed what constitutes rational desire. Further points are supporting arguments are given in “Prospects for a Naturalization of Practical Reason:  Humean Instrumentalism and the Normative Authority of Desire,” International Journal of Philosophical Studies 10, 3 (2002), 235-63.

⁸⁹ See esp. ch 2 of John Stuart Mill, Utilitarianism, George Sher, ed. (Indianapolis:  Hackett, 1979), pp. 24-25, in which Mill says “There exists no moral system in which there do not arise unequivocal cases of conflicting obligations . . . and there is no case of obligation in which some secondary principle is not involved” (p. 25). What Mill calls secondary principles would be much like Ross’s principles of prima facie obligation. Mill would have denied their self-evidence but would have agreed with Ross in affirming (though on utilitarian grounds) the prima facie obligations in question.

⁹⁰ David Hume, “Of the Standard of Taste,” reprinted in Mark Schorer, Josephene Miles, and Gordon McKenzie, eds., Criticism: The Foundations of Modern Literary Judgment (New York: Harcourt, Brace, and World, 1958), p. 446.

⁹¹ My Epistemology (cited earlier) contains discussion of (and literature references to) reliability in relation to knowledge in general and perception as a source of knowledge.

⁹² This is not to deny that the reflection appropriate to acquiring justification for believing a self-evident may involve inference. Consider the proposition that if p entails q and q entails r, and not-r, then not-p. Someone unfamiliar with formal logic may have to realize that p entails r in order to see how the formal truth corresponding to modus tollens is here involved in the overall truth.  But that p entails r is entailed by the proposition in question; it is not a premise for that proposition, nor is there any need of a separate premise.  In “Self-Evidence” (cited earlier) I have called this an internal inference and explained how it differs from the external kind that indicates premise-dependent knowledge and justification.