Article #328
Subject: The Lure of the Peak, and Mountaintop knowledge from the top
Author: Andrew W. Harrell
Posted: 8/8/2017 03:53:03 PM

Book Review of Derek Parfitt's "On What Matters: Volume I and II
PHILIP KITCHER

Part 1 of 2


I.The idea that ethics is the province of religion lingers even in relatively secular societies. On a recent
Saturday morning, the
principal news radio station in Berlin reported a dilemma facing German politicians as they attempt to
craft educational policy: children must be required to take classes in religion, or their ethical



education will inevitably be neglected. Yet the connection presup- posed by the politicians has often
been questioned. From Plato on, most philosophers have denied the possibility that the will of a dei- ty
could have anything to do with what is required of us. Although philosophy has shaped the ethical
teachings of the main Western religions, many of the most influential ethical thinkers have been
dedicated to explaining and defending principles in ways that are entirely independent of religious
doctrine. If the puzzled politicians had been aware of their own rich intellectual tradition, they would
have found easy ways of resolving their dilemma.

For over two millennia, philosophical ventures in moral theory have left their mark on everyday thought
and on concepts taken for granted in social and political life. Plato and Aristotle, Hume and Adam Smith,
Kant and Hegel, Bentham and Mill have shaped the lives of people who would not recognize their
names, people who make claims about virtue and vice, duty and obligation, the proper form of a market
economy, and the appropriate sphere of legal pro- tection. Nor does the influence end in the late
nineteenth century.
G.E. Moore’s assertion that human relationships and beautiful things are the sources of intrinsic value
inspired members of the Bloomsbury circle as they attempted to free themselves from the
claustrophobia induced by Victorian morality. More recently, John Rawls and Amartya Sen have
explored new directions for social and political theory, and Peter Singer has raised serious questions
about our treatment of other animals and about the responsibilities of the affluent few toward the many
people who live in acute pov- erty. Even in an age in which the relevance of philosophy to any aspect of
culture beyond its own arcane discussions is frequently (and justifiably) questioned, ethical theory
would seem to be one area in which philosophers still have important things to contribute.

At the heart of ethical theory are issues that seem inescapable, and that no other field of inquiry
promises to answer. How ought we to act? What kinds of things are worth wanting? What type of person
should you aspire to be? Many religious people suppose that there is an authoritative source of answers
to these questions. They are not moved by Plato’s cogent proof that the will of a being, howev- er
powerful, could not ground any moral duty. Nor are they per- suaded by Kant’s further development of
the same theme: moral reliance on the commands of another presupposes that the com- mander is not
only powerful but also good; hence those who obey must already have standards of goodness and be
able to apply them to the commander. After the twentieth century, and the many who attempted to
absolve themselves on the grounds that they were on- ly following the Leader’s orders, this Kantian
point would seem



especially forceful. Still, even those who continue to rely on reli- gious authority should at the very least
concede that their preferred manual of ethical instruction is profoundly incomplete. They, too, need
ways of guiding their conduct when their decisions lie beyond the scope of the commandments. They
might benefit, as genera- tions of religious scholars and teachers before them have done, from the
insights of philosophy.

Besides the straightforward questions about the actions, wants, and ideals of personhood, ethical
theory often ventures into higher- order issues, seeking to understand the character of ethics itself.
Rather than being a desertion of the more pressing problems in fa- vor of academic theorizing, this
trend could reasonably be viewed as part of a strategy for discharging ethical theory’s central task.
Asking how we might make sense of ethical truth and ethical knowledge, or answering the nihilist who
denies that anything mat- ters, can be a valuable initial step toward discovering what ought to be done
or what is worth cherishing. Still, as in any field of inquiry, aspiring theorists should beware lest they
lose all contact with the questions that provoked the line of investigation they are supposed to be
continuing. Thoughtful people who turn to the “literature” in recent ethical theory may well be puzzled
by the lack of connec- tion to the practical decisions and difficulties of contemporary life, and they may
harbor a suspicion (perhaps more than a suspicion) that ethical theory has become an academic game
of dubious rele- vance.

Derek Parfit is rightly admired for the acuteness of his philosophi- cal intelligence and his dedication to
a thorough exploration of the questions he takes up. His first book, Reasons and Persons, which
appeared in 1984, was widely viewed as an outstanding contribu- tion to a cluster of questions in
metaphysics and ethics, although its crowning achievement was, I think, its fourth part, in which he of-
fered a penetrating series of arguments about the aggregation of value: how should we compare a state
in which some number of people enjoy lives of high quality with a state in which considera- bly more
people live at a slightly less exalted level? Questions such as this one bear on topics in welfare
economics and social choice theory (although Parfit does not make the connection ex- plicitly). Since
he has done so much to shape the character of con- temporary ethical theory—particularly by
introducing concepts and methods now central to academic philosophical ethics—Parfit’s new book has
been eagerly anticipated. Moreover, its sheer size— the two volumes of On What Matters comprise
1,365 pages— invites the thought that this is a magnum opus, a book that might do in our times what
Moore accomplished a century ago, or what



Kant achieved in the German Enlightenment—though both of them with many fewer pages.

One prediction is almost undeniable. On What Matters will be the subject of innumerable graduate
seminars, a book to be pored over for weeks and months by apprentice philosophers and their men-
tors, a source for journal articles that will refine a principle here or challenge an argument there. It will
be a paradigm in the original, uncorrupted sense of the word, one that will give rise to a profes- sional
practice of philosophizing. But will it—or should it—have an impact on broader cultural discussions,
shaping future thoughts about what we ought to do or want or aspire to become?

II.

It is a virtue of Parfit’s book that it aims to cover the traditional domain of ethical theory. Its discussions
of questions about the character and status of ethics are integrated with substantive con- clusions
about which actions are right or wrong, and which desires are worth having. Parfit opposes nihilists
(who think that nothing matters), social relativists (who suppose that what matters depends on society),
and subjectivists (who claim that what matters is a function of what people want). In the terms that he
favors, there are objective standards for “what we have most reason” to want and to do: the awfulness
of pain, for example, gives us an objective rea- son to avoid being in agony. It is not at all evident,
however, that posing the issues in terms of “reasons” is particularly helpful, or that the question of
“what we have most reason to do” is an im- provement on asking, “what, all things considered, ought
we to do,” because the notion of “something’s being a reason for someone” remains elusive and
obscure, despite Parfit’s many attempts to get it clear and right.

In any case, Parfit’s central task, undertaken in the first volume, is to think systematically about these
objective reasons and to formu- late a “supreme principle of morality.” This is an enterprise in which he
takes himself to be continuing the philosophical tradition, particularly as exemplified by his “heroes”
Kant and Sidgwick. (The latter’s Methods of Ethics—a “great, drab book,” in Parfit’s words—developed
the utilitarianism of Bentham and Mill in ways that absorb ideas from rival traditions.) On the face of
it, the thought that morality should have a “supreme principle” is puz- zling, for it is not obvious that our
ethical life can be subsumed un- der any single formula. Some people worry about the idea that
physical theory can be distilled into some mighty equation, taken to be the core of a “theory of
everything,” but the field of ethics



appears even less susceptible to such spectacular unification. If ten commandments are unable to
suffice, how can we hope to manage with one?

Many of Parfit’s remarks indicate that he thinks of ethical theoriz- ing as analogous to what occurs in
theoretical science, but his con- ception of a “supreme principle” is more subtle and more promis- ing.
As we learn when he begins to entertain serious candidates for the role, the fundamental law of ethical
theory is to be framed in ways that facilitate its application to human decisions: “The Kanti- an
Contractualist Formula: Everyone ought to follow the princi- ples whose universal acceptance everyone
could rationally will. This formula might be what Kant was trying to find: the supreme principle of
morality.” Like several of the ideas proposed in Kant’s seminal Foundations of the Metaphysics of
Morality, Parfit’s for- mula attains great generality by offering a test someone might ap- ply in making a
decision of any sort.

Later, in the second volume, Parfit makes it clear that his unified theory of ethics is not supposed to
decide every ethical issue. He does not believe that all ethical questions have determinate an- swers. In
this regard he cites issues about “the ethics of population or the morality of war.” For this reason,
whatever tests are en- joined by the “supreme principle” may not always be applicable: the task,
presumably, is to supply criteria adequate for resolving only the ethical questions that are in principle
answerable.

The central claim of On What Matters is Parfit’s proposal of the “Triple Theory.” It attempts to
characterize the wrongness of acts.

An act is wrong if and only if, or just when, such acts are disal- lowed by some principle that is

(1) one of the principles whose being universal laws would make things go best,

(2) one of the only principles whose being universal laws eve- ryone could rationally will....

(3) a principle that no one could reasonably reject.

Plainly, these formulations involve concepts about which many readers (and many philosophers) may
doubt their own fluency; and Parfit devotes pages, sections, and entire chapters to attempts to explain
some of these concepts. Even in advance of working through his discussions, however, his statement of
the Triple Theo-



ry brings out the structure of his ideas. Although the characteriza- tion of wrongness appears to involve
three different tests, he claims that the three criteria give the same results—acts debarred by one
would be ruled out by the other two. The point of exploring and offering all three is to reveal an
unexpected convergence among three major traditions in moral thought. In an image he em- ploys
repeatedly, the history of ethical theory is described as a se- ries of attempts to climb a mountain from
different sides—and now that we are closer to the peak, it is possible to recognize that these are
attempts on the same mountain and that the routes are coming together.

The three traditions of thought whose convergence Parfit wishes to affirm are consequentialism,
Kantianism, and contractualism. Con- sequentialists begin from the idea that some states of affairs are
better or worse than others, and construe right actions as those tending to produce outcomes that are
as good as possible. The ver- sion of consequentialism favored by Parfit is one that focuses on general
principles rather than specific actions: instead of declaring that an action is right if it generates the best
results, he understands right actions as those commended by rules, or “universal laws,” that “make
things go best.” Kantians suppose that the principles of morality are those that could rationally be willed
to hold as univer- sal laws—or, to come closer to one of Kant’s original formulations, those laws that
would be agreed upon for their self-governance by a community of rational beings. And contractualists
suppose that the principles of morality are those that would be agreed on in a discussion under some
type of ideal conditions.

Much of the work in the first volume of On What Matters consists in efforts to make some of the central
notions of these three tradi- tions clear and precise, and, on that basis, to formulate the most
defensible version of each. And after Parfit has argued that the principles are equivalent, we are
supposed to conclude that this unexpected convergence of approaches often viewed as starkly in-
compatible redounds to the credit of the Triple Theory, and that Parfit’s—and moral philosophy’s—work
is done.


WHY, THEN, is there a second volume, still longer than the first? Partly because of an accidental
feature: an earlier and shorter ver- sion of Parfit’s ideas was presented as a set of Tanner Lectures at
Berkeley, and, by the conditions of those lectures, publication of a developed version should be
accompanied by the discussions of distinguished commentators and the replies of the lecturer. Thus
the second volume opens with a different set of voices, three be-



longing to the original discussants (Susan Wolf, Allen Wood, and
T.M. Scanlon) and one subsequently recruited (Barbara Herman). But after 250 pages of such
exchanges, the bulk of the volume is then devoted to Parfit’s attempt to explain the status of ethics. He
hopes to demolish a threatening view—he calls it Naturalism— according to which the only facts are
natural facts: the world con- tains only those things that would be described by improved ver- sions of
the full range of human, natural, and social sciences, and philosophers should dream of no more things
than there are in heaven and Earth. For Naturalists, ethics either turns out to be con- cerned with a
certain type of natural fact, or it fails as a correct de- scription of anything. Parfit devotes great energy
to distinguishing many different forms of the Naturalist idea, arguing that all are in- correct, and then
offers his own preferred account of what ethics is about (it delivers a species of “non-ontological
truth”) and how we can know ethical truths (we have a special ability, about which lit- tle can be said,
that acquaints us with these truths). All this takes time.

A bare survey of the principal contours of On What Matters fosters the impression that it is a leisurely
book, one that did not need to swell to its exceptional length. Such a judgment would be unwar- ranted.
Parfit’s book is noteworthy for the density of its argumen- tation. At almost every paragraph, readers
who want to achieve a firm grip on the thoughts that he is developing will have to pause and to ponder.
This is partly a matter of the difficulty of the no- tions that Parfit is trying to make precise, and partly a
tribute to his honesty in considering alternatives and pursuing them doggedly. Yet it is also the result of
the method that Parfit favors, one now prominent in ethical theory as a result of his influence.

In hundreds of staccato paragraphs, we start with some candidate principle, confront it with a story
about some artificially simplified situation, announce a judgment about the way to appraise that situ-
ation, and render a verdict on the status of the principle. Many of the stories involve schematically
described predicaments in which people are in danger of bodily damage or of death: trapped on rocks
or by an earthquake, or lying tied to a railroad track in a tun- nel (with possibilities for someone to divert
an oncoming train, thereby sending it on a track through a different tunnel where some lesser number
of people are bound—or even for someone to have the option of pitching a bystander from a bridge
above the track, so that the train will be stopped by his body). Under the names as- signed to them,
these exercises in moral fiction recur throughout the web of argument. Here is an example of the
dominant argu- mentative style, which I have selected at random:




As in Tunnel, however, this nondeontic reason [the awfulness of acting to bring about someone’s death]
could not decisively outweigh your reason to do what would save several people’s lives. If Bridge is
significantly different from Tunnel, as many people would believe, this difference could not, I believe, be
that, since you would be killing me as a means, you would have a decisive non-deontic reason not to act
in this way. This feature of this act might give you a decisive reason not to act in this way. But it could
do that, I believe, only by making this act wrong. This decisive reason would have to be deontic. If that
is true, the objection we are now considering fails.

Since passages such as this invite scrutiny of Parfit’s distinctions, concepts, and principles, and since
they supply ample fodder for varying his preferred stories and challenging or confirming his judgments
about them, the prediction I made earlier appears com- pletely safe: On What Matters is a treasure
trove for ethical theo- rists, looking to make their mark in the professional journals. Whether it can
contribute to broader ethical discussions about the pressing problems that occupy people and their
societies is, of course, a different matter.

Volume II
III.
SOME OF THE chief anxieties about Parfit’s version of ethical theory are already voiced by the
commentators whose essays open his second volume. One prominent theme in the commentaries is
whether Parfit’s claim about the convergence of three independent traditions is surprising (and thus has
the probative force he attributes to it). The critics point out, with considerable justice, that Parfit has
carefully selected just those elements in the three ethical traditions that have some kinship with one
another, ignoring more central features that would be far harder to reconcile. The points they raise are
valuable for philosophers interested in the history of ethical theory, and particularly for those concerned
with Kant’s writings on morality. Parfit replies to the historical concerns at some length, but spends far
less time addressing two more fundamental objections raised by his commentators.
Yet a collapse of the “surprising convergence” would seem less problematic for the Triple Theory than
these two further objections. The first of them concerns the goal of the project. Owing to his fondness
for thinking of ethical theory as analogous to theories in areas of the sciences, Parfit often writes as if
the goal of the enterprise is to produce a collection of principles that could be more or less
mechanically applied to ethical decision-making. After the Triple Theory has been fully developed, when
the three traditions have shaken hands on the mountain peak, there will be a new way of guiding our
ethical life: sensitive judgment will give way to accurate calculation. But would that really be an
improved, ultra-efficient way of guiding our ethical life? Would it not be, rather, something entirely
different—something in which the value of sensitive judgment would have been lost? Ethical decisions
owe part of their value to the person’s own activity of thinking through the problem, which often
involves engaging with the situations and feelings of others. Judgment is not a matter of applying some
formula that has been delivered by a correct and complete theory.
An alternative to Parfit’s way of thinking about ethics and ethical theory would take the point of the
enterprise to be one of assembling ideas that can be used both in helping people to develop into more
sensitive judges and in serving as resources on occasions of judgment. Perhaps there is no single peak.
Perhaps, as Susan Wolf suggests in her commentary, there is only an indefinitely extending range. Parfit
is much too brief in responding to her cogent suggestion.
________________________________________
THE SECOND OBJECTION concerns the method employed throughout On What Matters. Short
schematic fictions—“puzzle cases”—are used as if they were analogues of experimental results that
could be used to test putative theoretical hypotheses. One deep difficulty with this method is that, for
all the words that Parfit expends on attempts to clarify his central concepts, particularly the notion of a
reason, the concepts finally remain imprecise, and readers must constantly struggle to decide whether
his assertions about the bearing of the evidence are justified. Even more importantly, the reactions he
intends us to share are strikingly different from the kinds of reports that play a valuable role in the
development of the sciences: whereas the standardization of observations and experimental findings is
crucial to scientific objectivity, when people offer their judgments about puzzle cases in ethics there are
absolutely no standards for when they are doing it well, no serious understanding of what they are
doing or how, no sense of how their judgments might be distorted by prior commitment to some ethical
principle—and thus no way of knowing whether their reports have the slightest evidential worth.
Consider the case that Parfit refers to as “Bridge,” a variant on a muchdiscussed scenario. In the
canonical version, five people are bound to a track and threatened by the approach of a train. On the
rail of the bridge over the track sits a fat man, whose heft would be sufficient to stop the train. Would it
be right to push him from his perch onto the track below, thus using him as a buffer to protect the five?
Of course, if you imagine yourself on the bridge faced with this choice, all sorts of awkward and
practical questions arise. Would you be able to dislodge the fat man? (For the puzzle case to work, you
have to be of lesser girth—otherwise you would have the option of sacrificing yourself.) If you pushed
him, would he fall in a way that would halt the train? Is there some other way to prevent the deaths—a
signal that can be given or a switch that can be thrown? Could you persuade the fat man to jump? Could
you say, “Fat man, let us leap together”?
To avoid some of these questions, Parfit’s variant of the story stipulates a remotecontrol device that you
can use to launch someone from the bridge on to the track. In this way he seeks to dodge or escape
certain questions—but his modification introduces many others. How can you tell what will happen if
you use whatever device you have? Could you stop the train simply by opening the trap, without
anything falling through? Can you signal to the potential victim and arrange for some appropriate
substitute object to fall through the trap? Are there other devices you should seek that would allow you
to communicate directly with the driver, or to stop the train in less messy ways? Your response to any
actual situation would depend on how you would answer or address questions such as these—on how
you would cast around in attempting to avoid any death or injury (just as, in the original story, you would
seek alternatives to the stark choice assigned to you). Parfit’s emendation of the canonical scenario is
guided by no standard of objectivity for evoking reliable responses, and thus it generates further
versions of the disease it is intended to cure.
You cannot respond to the imagined predicament without thinking hard, but hard thinking leads through
a cloud of questions to a state of confusion. A few conditions are simply declared: the outcomes are
known and the options limited. But since that sort of certainty and limitation is exceedingly remote from
the circumstances in which we make our practical decisions, our judgmental capacities cannot be put to
work in their normal ways. Readers are pitched into a fantasy world, remote from reality, in which our
natural reactions are sharply curtailed by authorial fiat. When we are called on to render a verdict, the
dominant feeling is a disruption of whatever skills we possess, and a corresponding distrust of anything
we might say-often publicly visible when lecturers ask their audiences to respond to some puzzle case:
only partisans of some particular theory answer confidently, while the rest sit in uncomfortable silence.
The reader may even be left with a deep sense of unease that matters of life and death are to be judged
on the basis of such cursory and rigged information. (Allen Wood makes similar points trenchantly in his
contribution to Parfit’s book. This part of Wood’s critique goes unaddressed by Parfit.)
Parfit’s only defense of his use of puzzle cases occurs in passing, in a much earlier passage in which he
proposes that thought-experiments are as valuable in ethics as they are in the sciences. The
comparison prompts an obvious response—that many scientists think of thought-experiments as
motivational rather than probative, as preludes to real experiments that will elicit genuine evidence. (It
should also be noted that the great thoughtexperiments in the history of science occur in domains in
which the questions can be precisely defined.) And matters are made even worse when the puzzle
cases are used to interrogate the reasons that an imagined agent within the story might have for acting
in a particular way. The vagueness of Parfit’s concept of a reason—a concept he takes to be indefinable
—vitiates any serious attempt to survey the range of reasons someone has at his disposal. Even after
the many claims about reasons Parfit makes, a sensitive reader should still wonder if those claims are
justified. Moreover, his extensive discussion of issues about ethical truth and ethical knowledge renders
our capacities for arriving at judgments, whether about individual situations or about general principles,
so mysterious that the reader’s sense of hopeless floundering is further compounded.
After his lengthy attempt to scotch all naturalistic approaches to ethics, Parfit addresses the worry that
admitting “non-natural facts” is to venture into an obscure and possibly incoherent metaphysics. His
solution is that some truths (including ethical truths) are true in some “non-ontological sense,” and that
“we form many true beliefs because these beliefs are intrinsically credible, or because we are aware of
facts that give us reasons to have them.” In short, we possess an ability—Parfit thinks of it as shaped by
natural selection, but it is probably better to view it as the product of natural selection and sociocultural
learning—that can be exercised, in ways we do not understand at all, to yield a special sort of truth that
we also do not understand at all.
It is hard to feel confident about the existence of such an ability, or that our judgment results from its
proper exercise, especially when we are putting it to work on an artificial case in which our natural
thoughts are constrained in many ways and when the verdict we are to provide concerns imprecisely
formulated principles. Parfit’s best-developed attempt to defend his postulated ability is to draw on an
analogy with mathematical knowledge. Yet many things mathematicians once took to be selfevident
were later rejected by their successors, and principles now judged basic emerged from a complicated
history of mathematical exploration. In mathematics, self-evidence is achieved, not given.
So we have not been given any good reason to think that the Triple Theory is the true ethical theory.
While the traditions from which Parfit draws have supplied useful resources for ethical judgment, his
versions of their central ideas are not especially precise, nor readily applicable, nor well-supported by
the evidence that he offers. On What Matters does not contribute—perhaps it does not intend to
contribute—to fostering ethical discussion and ethical practice in the broader world.
________________________________________
IV.
IS THERE A BETTER alternative? I believe there is, and it is a version of the Naturalism that Parfit so
vigorously opposes.
Parfit is gripped by a particular picture of ethical knowledge, one in which people can discover ordinary
kinds of facts by ordinary kinds of means, the sorts of methods used by ordinary observers and
investigators, as well as by the most insightful researchers, methods that deliver information about mice
and molecules, murder, monsoons, and mayonnaise. In his view, the task of understanding ethical
knowledge is either that of building a bridge from these facts to ethical judgments or of finding some
separate source for those judgments. After failing to find a bridge from facts to values, Parfit sets out
on his quest for his (nebulous) source of ethical knowledge. Yet for at least a century, philosophers have
known that this kind of picture of knowledge is mistaken. Human beings begin in the middle, with a mix
of beliefs, and the proper topic concerns the grounds for a change of belief. People acquire, early in
their lives, a complex collection of ideas about the natural world and about how they should think, feel,
and act. As they grow, they change their minds, sometimes producing large collective transformations
in the prevailing views about what is to be done.
According to a well-worn joke, an American traveling in Ireland spends a long time trying to find the
remote village he aims to visit. At last he stops to ask one of the local inhabitants how to get there. And
he is told: “Well, I shouldn’t start from here if I was you.” The philosophical predicament is often quite
similar. Ethics might be better understood, and ethical life might be improved, if we began with the right
questions.
One version of Naturalism starts by thinking of ethics not as the search for a single immutable all-
serving principle, but rather as an entirely human endeavor, a project begun by our remote ancestors
tens of thousands of years ago and continuing indefinitely into the future. There is no mountain to
climb, no final compendium of ethical truths, but only a central human predicament, from which we
escaped by learning—imperfectly—to regulate our own conduct. The philosophical study of this project
must absorb the insights of various natural and human sciences, bits of evolutionary biology and
primatology, of psychology and anthropology, of archaeology and history. (Naturalism should be
elaborated broadly, recognizing the potential contributions of all rigorous forms of inquiry across the
entire spectrum, from art history and anthropology to zoology; there is no need for Naturalists to lapse
into the scientism of taking some particular area of physical science as fundamental.) Sensible
conclusions cannot be reached by pitting imprecise principles against fanciful cases, but rather by
looking, as carefully and as comprehensively as we can, at the details of ethical practice and ethical
change.
Our ancestors once lived in small groups, mixed by age and sex, in the fashion of contemporary
chimpanzees. To participate in this type of social life, they required some capacity for identifying and
responding to the desires of the other members of the band, but the limitations of that capacity made
their lives together tense and fragile. Self-regulation began as a social technology, directed at
overcoming the limitations of our altruism. Rules for conduct, discussed within the group, came to
govern human lives. Through a long period of time, probably at least fifty thousand years, different
small societies engaged in social experiments. The ethical practices that exist today are the heirs of the
most successful of these experiments.
Along the way many things happened. The initial framework of rules and motivating devices was
probably very crude: the first commands were likely focused on the most prominent causes of social
tension, giving rise to prescriptions for sharing scarce resources and proscriptions against initiating
violence; band members were motivated to comply through their fear of punishment. Subsequent
generations added more subtle ways of inducing conformity, recruiting emotions of solidarity, shame,
and pride, feelings of respect and awe, often directed toward a powerful being viewed as the source of
the group’s way of life. Attempts to resolve the challenges imposed by scarcity fostered a division of
labor, out of which roles and institutions emerged. The extension of some protections to neighbors
paved the way for an expansion of group size, and increased cooperation generated higher forms of
altruism. The ethical framework familiar to us evolved gradually, through a series of small steps. The law
codes that are among the earliest written documents testify to hundreds of generations of prior
discussion. During recorded history ethical changes become perceptible. Although instances of ethical
progress may be rare, it is hard to resist the thought that some of these modifications are progressive.
________________________________________
YET HOW CAN a naturalist approach make sense of a concept of ethical progress? Not by conceiving of
it as the discovery of a prior and independent truth. Better to think of it as consisting in the solution of
problems—as progress from, not progress to. Ethics begins as a social technology, aimed originally at
making up for the limits of human altruism. Some prominent episodes in the recorded history of ethical
practice take up versions of the original problem: when slavery is abolished, when women’s choices are
expanded, when prejudices against certain forms of sexual expression are overcome, a prior situation in
which there is a systematic failure to identify with the desires and the aspirations of other people is
changed—a class of failures of altruism is resolved.
The changes come about not through recognition of some special ethical fact, hitherto unappreciated,
but through the discovery of natural facts, about people, their capacities, sufferings, and aspirations,
on the basis of which there are new possibilities for mutual engagement. Reformers come to see that
desires that have been ignored or viewed as perverse are central to the lives of others, and through a
more informed, inclusive, and sympathetic conversation, they learn how those desires can be satisfied
without interfering with anyone’s fundamental aims.
But the identification and overcoming of our failures of altruism is not the only mode of ethical progress.
Like other forms of technology, the ethical project is not limited to the problem out of which it arose. It
generates new problems as it evolves. The “ethical truths” we arrive at are those principles introduced
in progressive—that is, problemsolving—transitions, and retained in subsequent progressive changes:
in William James’s happy phrase, “Truth happens to an idea.”
The great ethical theorists, on such an account, are those who supply resources for human decisions—
collective human decisions—directed at problem-solving. Many of these thinkers have both reflected
extensively on the practices of the societies in which they find themselves and have been deeply
immersed in the theoretical ideas of their predecessors. (In our own times, John Rawls is an outstanding
example.) Whatever their intentions, they offer no final theory, no “supreme principle of morality.”
Moreover, although they facilitate conversation, serving as philosophical midwives, they cannot claim
any special expertise in discovering ethical truth. Just as ethical practice began in negotiation within a
small group, so, at its best, it continues by involving many—ideally all—human perspectives, under
conditions in which each strives to accommodate the interests of all others.
________________________________________
WE SHOULD DISPENSE with the myth of the sage, the guru, the teacher, and the ethical expert, which
has often distorted the ethical project. The image of the moral philosopher as expert is a latecomer: in
human history, by far the most prominent claims to special ethical expertise have been those of
religious teachers, who have sometimes, although not always, used their alleged access to the
supernatural to inscribe their own prejudices into a group’s ethical code. But philosophers who reject
that version of the story fall into a rival fable, according to which they are the people who have insight
into pre-existent ethical truth, and they are the pioneers who climb the mountain. For all its intellectual
perseverance and academic penetration, On What Matters remains trapped in the acceptance of this
latter myth.
Whether the approach I have sketched is ultimately satisfactory (and it probably is not), I am convinced
that its kind of Naturalism, and the questions that it poses, offer an important corrective to
contemporary fashions in academic ethics. Although he does not construe Naturalism in terms of an
interdisciplinary perspective on the ethical project, Parfit regards the forms of Naturalism he considers
as a serious threat. On several occasions he remarks that, if Naturalism is true, then much of his life will
have been wasted. My last disagreement with him concerns this wrenching judgment.
If Naturalism is true, then many of Parfit’s claims are indeed wrong and his perspective is indeed askew.
Does it follow that his efforts (and consequently much of his life) have been wasted? I do not think so.
Almost all those who have engaged in any form of inquiry have been wrong and misguided. That is our
predicament: fallible investigators start from the conclusions of their fallible predecessors. Yet even the
dedicated mathematical astronomers of the late Middle Ages who explored the complicated details of
the equant point in Ptolemaic theory contributed to the advancement of the science, by supplying
standards by which more promising post-Copernican systems might be judged, and by introducing
possibilities and options into future debates. On What Matters belongs with such achievements. It
stands as a grand and dedicated attempt to elaborate a fundamentally misguided perspective. Its
diligence and its honesty command respect. Perhaps these real virtues will set standards for very
different ventures in academic ethics, Naturalist or otherwise—for a return to the tradition of attempts
to understand and improve everyday judgment, and to provide resources for people and policymakers
everywhere. In the end, that is what matters.
Philip Kitcher is John Dewey Professor of Philosophy at Columbia University and the author, most
recently, of The Ethical Project (Harvard University Press). This article appeared in the February 2,
2012, issue of the magazine.