Minnesota Interdisciplinary Workshop on Well-Being, 24-26 October 2003
Hedonism Reconsidered
Since the 1960s, and especially over the last twenty-five or so years, a great deal of
significant, interesting, and suggestive work has been carried out in psychology and
related empirical disciplines on the question of well-being or happiness and its
During the same period, there has been a good deal of progress made on
the same sort of topics in philosophy. But there has been less interaction between
practitioners in the two debates than one might have expected. The bibliographies of
central works in philosophy, such as Derek Parfit’s Reasons and Persons (1984) or James
Griffin’s Well-Being: Its Meaning, Measurement, and Moral Importance (1986), for
example, make no reference to the debate in psychology; while a recent, important
collection of articles on the psychology of happiness – Well-Being: The Foundations of
Hedonic Psychology (1999) – includes no contribution by a philosopher. Of course there
has been interaction – a good deal of it by those participating in this workshop – but I
think my generalization still holds. I suspect this is largely a result of specialization.
People have enough difficulty keeping up with work in their own disciplines, and as other
disciplines specialize they become harder to enter. There has also been a readiness in
philosophy quickly to dismiss discussions of happiness in particular as outdated and
irrelevant. Anyway, whatever the reasons, the failure of communication is a pity, I think,
as there’s a good chance that each side could contribute to the other.
Both, after all, are
engaged in essentially the same task – working out what well-being consists in, what its
overall significance is, and how we might best make judgements about it. Certainly, the
idea of this workshop seems to me an excellent one.
As yet, I have not been able to dig deep into the empirical literature. But as far as
I can tell, psychological research centres around a notion usually called ‘subjective well-
See Diener 1984; Diener, Suh, Lucas, and Smith 1999; Kahneman, Diener, and Schwartz (ed.) 1999;
Ryan and Deci 2001; Layard 2003.
Nor should past philosophers be forgotten. Bentham is occasionally mentioned in empirical work, but
little use is made of his writings. And I would urge all empirical researchers to study with care the chapters
on ‘empirical hedonism’ in Sidgwick 1907 (bk. 2, chs. 2-3).

Page 2
being’ (SWB). SWB is often claimed to involve three aspects: positive affect, absence of
negative affect, and life-satisfaction.
Mere absence of negative affect, however, does not
seem to be a positive good (a stone has it). And it’s hard to see how mere satisfaction
with one’s life could itself be a good, unless we count that satisfaction as itself a positive
affect. So we’re talking about positive affect, and if our topic is a person’s overall well-
being, negative affect as well.
Then the question arises how we should understand such affects. Should it be as
pleasure and pain, or pleasures and pains? Emotions? Moods? Feelings in general? My
paper concerns this question. In a nutshell, my answer to the question is that the view
philosophers have called hedonism has been too readily dismissed by philosophers in
recent years. Whether hedonism, as I characterize it, is correct should be a matter of
significance to psychologists of happiness interested in clarity about what they are
measuring – and the implications of their results. To decide how much weight to give
SWB in policy-making requires a prior decision as to whether anything matters to human
well-being overall other than SWB. According to hedonism, only SWB matters. So the
implications of research on the psychology of happiness would be especially and
immediately important. (I should confess at this point that my paper as it stands is quite
programmatic. All I’m doing here is trying out a few lines of argument – so comment and
criticism would be very much appreciated.)
How, according to hedonism, should well-being be measured? I suspect that the
philosophical hedonist would approve several of the methods and methodological
research programmes presently underway in psychology, economics, and other areas.
But since, according to hedonism, what matters to well-being is how enjoyable an
experience is at the time it is experienced, and how much such experience the subject has,
the method involving multiple self-reporting might be thought the most likely to bear
fruit, especially given the doubts over the reliability of life-satisfaction reports.
But here
See e.g. Myers and Diener 1995: 11; Ryan and Deci 2001: 144; Frey and Stutzer 2002: 24.
For a brief survey, see Frey and Stetzer 2002: ch. 2.
See e.g. Diener 1984: 551; Kahneman, Diener, and Schwartz (ed.) 1999: ix-x; Tiberius 2003: 1-2 and refs.
therein. Preference hedonism, as I characterize it in this paper, is committed to the view that the
contribution of an experience to a subject’s well-being depends on the subject’s desiring that experience,
and that the stronger the desire, the more valuable the experience. One way to get at whether someone is
happy, therefore, is to ask them about their preferences. I suggest that we interpret the statement ‘I desire
experience e’, in this context, as implying that I prefer that experience to nothing at all. As soon as this is

Page 3
I would especially welcome the views of experts in the psychological field as to what
they think philosophical hedonists should say. On the face of it, it appears that some of
the difficulties affecting life-satisfaction reports might affect reports of ‘instant utility’.
Cultural biases in favour of or against being happy, for example, might reappear at
instants; if the subject is engaged in an activity (‘flow’),
there could be similar problems
with memory of the experience of the activity; and if the subject makes her instantaneous
judgement by comparing that instant to others, then memory problems may occur here
also. And may there not be factors which distort instant utility reports in particular?
Irritation at repetitive sampling, for example, or laziness leading to unreflective reporting;
problems arising from the interruption of activities in which the subject would otherwise
be enjoyably engaged; or biases of mood which might disappear in a retrospective life-
satisfaction report.
Defining Hedonism
As Wayne Sumner says, ‘Time and philosophical fashion have not been kind to
My aim in this paper is to consider whether that unkindness is justified, by
attempting first to describe a version of hedonism which is at once in the spirit of the
hedonist tradition and not obviously implausible, and then to defend that view against
two of the most weighty objections that have led to hedonism’s current malaise. I shall be
considering hedonism only as a theory of welfare or well-being, not as a theory of
the case, then we have well-being. How much well-being? How are we to understand strength of desire? A
desire has various aspects. One is a disposition to go for its object. So one way in which a desire’s strength
might be increased would be for the subject’s disposition to become stronger. For example, if I was
previously disposed to go for p in the presence of q but not in the presence of r, but am now disposed to go
for p in the presence of q and r, other things being equal, my desire for p has strengthened. Desire also has
its phenomenology. In normal circumstances, if my felt satisfaction at possessing p increases, or if my
disappointment on failing to obtain p increases, then my desire for p has increased in strength. Self-report
tests do seem to be a good way to capture this phenomenological aspect of happiness, as long as
respondents are told that the primary benchmark for their judgements should be the zero level. What would
be needed for a unit of measurement would be both a temporal limit, and a way of measuring strength of
desire. Thus a unit might be stipulated to be a period of experience of some kind or another of one minute,
preferred to nothing with a certain strength to which some number is allocated. This would allow for both
intrapersonal and interpersonal comparisons of value. I take it that assumptions something like these lie
behind a good deal of empirical research in this area.
To use Daniel Kahneman’s term; see Kahneman 1999: 4.
Csikszentmihalyi 1990.
Kahneman (1999: 21) notes, for example, that bad weather affects self-reports.
Sumner 1996: 83.

Page 4
motivation (psychological hedonism) or as a theory of the good. Hedonism as an account
of what is good for any individual is consistent both with the view that agents are
motivated by goals other than pleasure, and the view that there are non-hedonist non-
welfarist values, such as, say, aesthetic or moral values.
It is often said that hedonism is the view that pleasure is the only good for an
individual, and pain the only bad. This is correct as far as it goes, but it glosses over an
important distinction between two types of theory of well-being – substantive and
A substantive theory consists in a mere list of the item or items that is or are
non-instrumentally good for people. Such items might include pleasure or pleasures,
accomplishment, knowledge, understood in independence from the views or desires of
the subject, or they may be more closely tailored to the subject, listing, for example, the
objects that will satisfy the subject’s informed desires. A formal theory constitutes an
account of what it is about the items on some list that make them good or valuable for the
individual concerned. For example, someone who has listed accomplishment might claim
that what makes accomplishment, or any other good, valuable for the individual is its
perfecting that individual’s human nature. That view might be called perfectionist. An
alternative would be to say that what makes accomplishment valuable is its satisfying a
desire for accomplishment, a desire, it might be suggested, which is universal and hence
an aspect of human nature. This view might be called a desire theory. Or one might claim
that something’s being an accomplishment is itself what makes it good – ‘being an
accomplishment’ is itself a ‘good-for making’ property. This position might be called an
objective list theory.
It might be objected that the distinction between these two levels of theory is
unnecessary. A perfectionist, for example, needs merely to say that what is good for an
individual is perfection of her nature. But this is both to ignore the difference between the
Often I shall speak merely of what is good for an individual, meaning ‘what is good (overall) for’ and
therefore including also what is bad for an individual. Thomas Hurka (1987) suggests banishing the phrase
‘good for’ from philosophical ethics, because of its ambiguity. But the distinction between what is good
‘impersonally’, in the sense of making a world or universe good, and what is good for an individual, in the
sense of making her life better for her than it would otherwise have been, seems to me fundamental. The
closest Hurka comes to describing this sense of ‘good for’ is as ‘good from the point of view of’, but this
notion itself seems to cut across the distinction between ‘good’ and ‘good for’. It may be that, from my
point of view, Moore’s beautiful world, never seen by anyone, is good, and that, from my point of view, my
headache’s coming to an end is good for me. See Moore 1903: 83-4.
See Sumner 1993: 101-2; Moore and Crisp 1996: 599.

Page 5
question, ‘What is good for individuals?’, and the question, ‘What makes what is good
for invididuals good for them?’, and to leave no room for specification – in this case, of
what perfection of human nature consists in.
The content of any substantive list will in most cases be shaped to some extent by
the formal theory, but it is quite likely that one will arrive at a substantive theory first and
then seek to explain its contents with a formal theory. Someone who begins with the view
that knowledge is a substantive good, and then reflects upon what they value in
knowledge and offers a formal desire theory, may reconsider their list in the light of the
fact that some people appear to possess a good deal of knowledge that does not fulfil any
particular desire . Knowledge might therefore be dropped, to be replaced by the objects of
desire (including knowledge, of course, if desired).
If a hedonist lists pleasure alone in her substantive theory, are we to take that
answer to be equivalent to ‘pleasures’? That will depend partly on our view of pleasure,
and partly on how we understand pleasures. If we take pleasures to refer to activities
which are enjoyed by the person engaged in them, as in ‘Playing golf is one of my
greatest pleasures’, and pleasure to refer to some kind of mental state or experience, then
a clear distinction between the two opens up. If, however, one follows Ryle and takes an
adverbial view of pleasure, then the gap begins to close.
It is certainly true that we
speak of taking pleasure in (and sometimes enjoying) many different kinds of object:
activities, other people, facts that, and so on. But it is most in the spirit of hedonism, I
suggest, to take both ‘pleasure’ and ‘pleasures’, listed substantively, to be equivalent to
‘pleasurable experiences’. What about hedonism at the formal level? What should a
hedonist say it is about pleasurable experiences that makes them good for individuals?
Here the answer closest to the spirit of hedonism, surely, is the view that it is their being
pleasurable (and not, say, their satisfying desires).
Hedonism about well-being, then, is the view that pleasure (i.e. pleasurable
experience) is the only thing that is good for anyone, and pain (i.e. painful experience)
the only bad, and that what makes these things good and bad respectively is their
pleasurableness and painfulness. Our next step, then, must be to consider the nature of
pleasure itself.
Ryle 1954.

Page 6
The Affective Model
Pleasure has been understood by some philosophers to be a special kind of sensation, that
is, something we sense, like sweetness.
A now-standard objection to any such view is
that the various experiences we find pleasurable – eating good food, taking a brisk
country walk, listening to music – seem to have no sensation or whatever in common.
This objection seems powerful, and an immediate response might be to look
beyond feelings to some ‘external’ criterion for an experience’s being a pleasure, such as
its being desired. Shelly Kagan has suggested an alternative position, however, according
to which pleasantness or pleasurableness is not a single common ‘component’ of pleasant
experiences, but a single ‘dimension’ along which experiences can vary.
Kagan uses an
analogy with the volume of sounds. Volume, he suggests, is not a ‘component’ of
auditory experiences, but ‘an aspect of sounds, with regard to which they can be ranked’.
If pleasantness is like volume, then arguing that pleasantness is not a single property
common to pleasurable experiences, because of the qualitative differences between them,
would be like arguing that, because sounds are so different from one another, there is no
single quality of volume.
How is the distinction between components of experiences and dimensions of
variation meant to work? Take the sound of a tinkling bell, and the sound of a honking
horn. The components of each are, respectively, tinkling and honking. Volume, Kagan
suggests, is not a ‘kind’ of sound. So a loud tinkling is the same sound as a soft tinkling,
whereas a loud honk is a different sound from a loud tinkling.
It is questionable, however, whether this distinction captures anything of great
metaphysical significance. We would indeed be inclined to say that the soft tinkling is the
same sound as the loud tinkling. But that is because we usually focus on aspects of how
things sound other than how loud they are. In fact loud sounds do form a kind. I might
ask you to group sounds together according to their volume, and you would then
categorize the loud tinkling with the loud honk, and the soft tinkling with the soft honk.
As Kagan himself goes on to say, ‘it seems … that there is a sense in which a specific
See e.g. Mill [1869]: 214; cf. Broad 1930: 229-30. For an excellent discussion, see Sumner 1996: 83-92.
See e.g. Griffin 1986: 7-8; Sprigge 1988: 127-31.
Kagan 1992: 172-3. Kagan himself refers to unpublished work by Leonard Katz.

Page 7
volume is indeed an ingredient of a given sound’. If we abandon any attempt to draw
distinctions between components, dimensions of variation, and ingredients, and speak
only of introspectible properties of experience, we are left with the original problem. In
the case of sounds, we can distinguish between them according to an introspectible
common property: volume; in the case of pleasant experiences, there is no introspectible
common property.
The Pro-attitude Model and Preference Hedonism
If there is no introspectible felt quality common to pleasurable experiences, an obvious
move to explain what pleasurable experiences do have in common
is to offer an account
according to which a pleasure is an experience towards which the subject has what may
be called a pro-attitude.
The notion of a pro-attitude is a broad one, so this model covers
a wide range of positions. One common view is preference hedonism, which might be
stated as the view that a pleasant experience is one that the subject desires to continue, for
its own sake.
Sumner objects to this view as follows: ‘Whatever its object, a desire can only
represent (or result from) an ex ante expectation that the continuation of some state or
activity will be experienced as gratifying; the satisfaction of the desire cannot guarantee
the ex post gratification’.
Sumner’s objection here seems to be the following. Take
some experience e at time t, which I desire to continue. According to the preference
hedonist, what would make e pleasurable would be the satisfaction, at t
, of my desire.
But, Sumner objects, e might continue, thus satisfying my desire, and turn out not to be
Sumner goes on to suggest that one can think of many valuable experiences –
such as the birth of a baby or a romantic moment -- which are not improved by
prolongation: ‘where many pleasures are concerned, more is not necessarily better’.
Justin Gosling had earlier provided the examples of a person ‘enjoying a subtle whiff of
scent, where the pleasure is in the ephemeral quality of the experience, and the person
I am assuming that one should try as far as is reasonable to avoid a Wittgensteinian ‘family resemblance’
See Sumner 1996: 90-1. Sumner calls such views ‘externalist’.
See e.g. Brandt 1966: 268-9.
Sumner 1993: 111.

Page 8
would be nauseated at the thought of lingering over it’, and of someone who is enjoying
breaking some good news to someone else but who must realize that they cannot go on
doing so.
This seems a good line of objection to preference hedonism so understood. To
avoid Sumner’s point about ex post gratification, the preference hedonist should insist
that the desire and its satisfaction are contemporaneous. Imagine that I am enjoying the
experience of teeing off in a game of golf. According to preference hedonism, my
enjoyment consists in my having that experience, and my desiring to have it (the
satisfaction of my desire, of course, follows from this combination). The same position
could be stated in terms of other pro-attitudes, such as liking or favouring. A pleasure,
then, is an experience that the subjects desires, wants, likes, favours, or whatever. And a
pain is an experience that subject desires or wants not to have, disfavours, dislikes, and so
on. For ease of exposition, I shall continue to speak mainly of desiring.
A version of preference hedonism which makes the desire in question
contemporaneous with its satisfaction seems able to make sense of the alleged problem
cases. Take the whiff of perfume. Gosling’s objection is that the pleasure of smelling it
cannot consist in the subject’s desire that the experience continue, both because part of
the enjoyment lies in the ephemerality, and because the subject would find the prospect
of its continuation nauseating. But the object of the subject’s desire is best not seen as the
continuation of the experience. That introduces the gap between desire and satisfaction
that led to Sumner’s first problem. Rather, the experience I desire when I am enjoying the
whiff of perfume is that very experience. I may be quite aware that its continuing would
make me sick, and hence not desire that (though I may well desire its continuing in the
absence of nausea). And there seems no difficulty in accounting for the pleasure I take in
ephemerality as an experience that I desire as it is, rather than an experience that I desire
to continue.
Pleasure cannot merely be an experience that is desired by its subject, however. It
must be desired in the right way, and for the right reasons. Imagine that I have never
experienced serious pain. I might, during my first experience of it, welcome it, at least for
a short time. Or a creative artist, who finds creativity acutely stressful, might desire that
Gosling 1969: 65.

Page 9
experience for its own sake, perhaps because she believes it to be valuable in itself.
Rather, one’s desire must be a response to how the experience feels. This is not to
reintroduce the idea that there is a single feeling of pleasure. Experiences feel very
different from one another – but they may all be enjoyable if we desire them.
Preference hedonism, then, suitably finessed, seems to provide an adequate
account of pleasure. I would add one further caveat. ‘Pleasure’ is quite commonly
understood to refer to bodily pleasures in particular.
For that reason, it is worth stating
hedonism in terms of enjoyments as well as as, or even rather than, pleasures. Pleasure
consists in any experience the subject desires – that is, any experience the subject enjoys.
And what makes such experiences good for their subjects is their being enjoyed.
Two Problems
One objection to any pro-attitude model of pleasure is that, in so far as it makes the value
of pleasure depend on something external or extrinsic – the desires, liking, or whatever of
the subject, it rules out the view that pleasure is intrinsically valuable.
For something is
intrinsically valuable in so far as its value depends on properties intrinsic to it.
Deciding which properties of something are intrinsic and which are not is
notoriously difficult. But it is clear that G.E. Moore, who is the prime source for such a
view of intrinsic value, would have allowed pleasure, as understood by the preference
hedonist, to be intrinsically valuable. Moore allows that an ‘organic whole’, comprising
the proper appreciation of beauty and the existence of the beauty itself, has intrinsic
So there seems no reason why pleasure should not be interpreted in an analogous
way as intrinsically valuable: the combination of an experience and its subject’s desire
for it has intrinsic value.
Another potential problem with preference hedonism emerges if we return to the
distinction between substantive and formal theories of well-being. According to
preference hedonism, what makes an experience pleasurable is its being desired. And,
My account here follows Kagan 1992: 173-4. As Kagan himself says, there is more (much more) to be
said here about what is meant by a feeling, and the kind of responses that count as pleasure-responses. I
hope to be able to say a little more about this in future.
See Sidgwick 1907: 402.
Feldman 1997: 135-6.
Moore 1903: 190.

Page 10
according to hedonism as a formal theory, an experience’s being pleasurable is what
makes it valuable. Thus it might seem that the ultimate good-making property here is
‘being desired’, and herein lies a problem: we desire many things other than experiences,
including those items often found on the lists of the constituents of well-being offered by
non-hedonists – friendship, knowledge, reputation, and so on. If a person’s wanting her
experiences can make those experiences good for her, why cannot her wanting her
friendships make those constituents of her well-being also?
The hedonist can avoid the extension problem here by suggesting that it is not
merely the desiring or liking of the experiences by the subject that makes them valuable,
but the experiences being what they are – that is, experiences. In other words, the ‘good-
making property’ here is disjunctive: ‘being an experience and being liked’. Nor does this
restriction seem ad hoc, since hedonism itself seems to begin from the strong intuition
felt by many that well-being must be understood in terms of conscious experience
The Philosophy of Swine?
A long-standing objection to hedonism is that it cannot distinguish between kinds of
pleasures, so that, in principle, the value of some apparently especially valuable pleasure
could be outweighed by a greater amount of some much more basic pleasure.
I myself
provided an example to illustrate this problem:
Haydn and the Oyster. You are a soul in heaven waiting to be allocated a life on
Earth. It is late Friday afternoon, and you watch anxiously as the supply of
available lives dwindles. When your turn comes, the angel in charge offers you a
choice between two lives, that of the composer Joseph Haydn and that of an
oyster. Besides composing some wonderful music and influencing the evolution
of the symphony, Haydn will meet with success and honour in his own lifetime,
be cheerful and popular, travel and gain much enjoyment from field sports. The
oyster's life is far less exciting. Though this is rather a sophisticated oyster, its life
Cf. Kagan 1992: 177-8.
See Sidgwick 1907: 398.
See Mill [1863]: ch. 2, para. 3.

Page 11
will consist only of mild sensual pleasure, rather like that experienced by humans
when floating very drunk in a warm bath. When you request the life of Haydn, the
angel sighs, ‘I'll never get rid of this oyster life. It's been hanging around for ages.
Look, I'll offer you a special deal. Haydn will die at the age of seventy-seven. But
I'll make the oyster life as long as you like’.
If all that matters to my well-being is enjoyable experience, will there not come a
point at which the value of the oyster life must outweigh that of the life of Haydn? And if
so, is that not a strong objection to the reductionism of hedonism?
J.S. Mill, of course, developing some lines of argument from Plato’s Republic,
sought to answer this kind of objection by distinguishing between higher and lower
pleasures, on the basis of a distinction between quantity of pleasure (understood in terms
of intensity and duration) and quality:
It is quite compatible with the principle of utility to recognise the fact, that some
kinds of pleasure are more desirable and more valuable than others. It would be
absurd that while, in estimating all other things, quality is considered as well as
quantity, the estimation of pleasures should be supposed to depend on quantity
If I am asked, what I mean by difference of quality in pleasures, or what
makes one pleasure more valuable than another, merely as a pleasure, except its
being greater in amount, there is but one possible answer. Of two pleasures, if
there be one to which all or almost all who have experience of both give a decided
preference, irrespective of any feeling of moral obligation to prefer it, that is the
more desirable pleasure. If one of the two is, by those who are competently
acquainted with both, placed so far above the other that they prefer it, even though
knowing it to be attended with a greater amount of discontent, and would not
resign it for any quantity of the other pleasure which their nature is capable of, we
Crisp 1997: 24.

Page 12
are justified in ascribing to the preferred enjoyment a superiority in quality, so far
outweighing quantity as to render it, in comparison, of small account.
Mill’s argument has come under a great deal of scrutiny, and it is commonly
thought that he faces a dilemma.
Either the higher pleasures are higher because they are
more pleasurable, in which case no special distinction between higher and lower
pleasures can be drawn on the basis of anything except intensity and duration; or they are
higher for some other reason, such as their being more ‘noble’, in which case Mill has
abandoned hedonism by allowing in non-hedonistic goods.
In a previous discussion of this passage, I sketched a way in which Mill might
avoid the dilemma.
Logically he is not prevented from claiming that properties such as
nobility do in fact increase the pleasantness of experiences, thus adding a dimension
along which value can increase in addition to intensity and duration. But I suggested that
this solution is not entirely satisfactory, since it is not clear why, if nobility can increase
enjoyableness and hence value, it cannot be a good-making property in its own right.
Further, it is not clear why we should not allow that an experience could be noble, and
not pleasurable.
I now think, however, that this version of the ‘extension problem’ can be resolved
in a more satisfactory way by closer attention to what is involved in the distinction
between quantity and quality, and reference to the idea of competent judges. Consider a
simple bodily pleasure, such as that involved in sipping a small glass of moderately
chilled lemonade on a hot day. If we ask the drinker what it is about that experience that
makes it pleasurable, she will refer to qualities of the drink: its being cool and sweet,
perhaps; and we may restate that claim in terms of her experience of coolness and
sweetness. Now we might ask her to compare that experience with two others: that of
drinking a small glass of tepid water, and that of drinking a large glass of ice-cold
lemonade. As a competent judge, she may prefer the experience of drinking the small
glass of lemonade to that of drinking the tepid water, on grounds of quality: the water is
neither cool nor sweet. And she may prefer the large glass of lemonade to the smaller on
Mill [1863]: 2.4-5.
See Crisp 1997: 32 and refs.
Ibid.: 33-5.

Page 13
grounds of both duration (the experience lasts longer) and intensity (it is more chilled). In
other words, judgements concerning the value even of simple bodily pleasures depend on
quality as well as quantity.
Now consider a typical ‘higher’ pleasure. Imagine that our lemonade drinker has
recently completed her first reading of Jane Austen’s Pride and Prejudice, and we ask
her to rank the experience of drinking the ice-cold lemonade against that of reading the
novel. She ranks the novel higher than the lemonade. Why? There is much more to this
judgement than mere intensity and duration. The drink is perhaps of greater intensity, but,
our judge may suggest, it would not matter how long that experience could be prolonged:
she would never enjoy it as much as she enjoyed the novel. For what she enjoyed in the
novel was its wit, its beautiful syntax, and its exquisite delineation of character. The loss
of such enjoyments (that is, enjoyable experiences) – in the context of her own life --
could never be compensated for by any amount of lemonade pleasure. But this does not
mean that quantity is irrelevant to valuing mental pleasures. Just as quality matters in
valuing bodily pleasures, so quantity matters in valuing mental pleasures. Austen’s
juvenile work Love and Friendship is worth reading, but Pride and Prejudice is far
better. And two of the reasons for preferring Pride and Prejudice, other things being
equal, are that the enjoyment in reading it is more intense, and it is a significantly longer
work. But the point remains that a hedonist, if she takes sufficient note of the fact that we
often refer to the qualities of our enjoyment or what we enjoy in explaining why we value
that enjoyment, has the resources to explain the vastly greater value we put on certain
pleasurable experiences without introducing non-hedonist elements into the account of
well-being. It is true that at this point a hedonist about well-being would be well advised
to admit the existence of certain non-hedonistic aesthetic values, the appreciation of
which can be enjoyed to such an extent that such enjoyments become discontinuously
more valuable than certain bodily pleasures. But it is still enjoyment alone that matters to
well-being, so the extension problem can be avoided.
If we are to ascribe such value to the enjoyment of appreciating the beauty of Jane
Austen’s syntax, however, should we not admit that such appreciation on its own, without
enjoyment, can increase a person’s well-being? Or at the very least that what is adding
value in such cases is an ‘organic whole’ composed of appreciation (which may well be

Page 14
valueless without enjoyment) and enjoyment? Here the hedonist must first insist that, on
reflection, we should conclude that pleasureless appreciation is without value. (This
objection – that things other than enjoyment matter to well-being – has recently been
stated in a particularly devastating form by Robert Nozick, and I shall discuss his so-
called ‘experience machine’ in the following section.) What about organic wholes? The
hedonist will have no objection to allowing in reference to, say, aesthetic appreciation at
the level of substantive theory. That is, enjoyable appreciation of aesthetic value may
feature on the list of welfarist goods. But at the level of formal theory, she will insist that
what makes such appreciation good for the subject is its being enjoyed, and that alone.
Reference may be made to aesthetic value, as we have seen, in explaining what makes the
experience enjoyable and what is being enjoyed. But allowing any contribution to welfare
in such cases to come from the appreciation itself leaves unanswered the extension
problem – if appreciation can contribute to welfare alongside enjoyment, why can it not
contribute on its own?
The hedonist, then, appears to have a response to the philosophy of swine
objection, as stated in terms of my case of Haydn and the Oyster. But now consider a
new version of that problem, in which the angel in charge offers to manipulate my desires
in the case of the oyster, so that even were I fully acquainted with the kind of pleasures in
each life, I would now desire the oyster life much more strongly, and would, during my
life as an oyster, have very strong desires for the experience I was having. (If I express
doubt concering whether an oyster could have very strong desires for anything, the angel
will respond by saying that this particular oyster will really be just like a human being
who happens to have very strong desires for oyster pleasures.) Is the hedonist not
committed to the view that the oyster life will be better for me?
Here I think the hedonist must bite the bullet. She is not committed to the view
that the life of an oyster could be better for us as we are than a life such as Haydn’s. But
if well-being is increased by enjoyment, and enjoyment is increased by strength of desire,
then increasing strength of desire will increase well-being. Why does this conclusion
strike us as so counter-intuitive? There are at least a couple of things a hedonist might
say here. One is that our judgements are based on our own strong preference for a life
such as Haydn’s. Another is that we see certain things – such as creativity or aesthetic

Page 15
appreciation – as valuable in themselves, independent of enjoyment. This leads us into
the second major objection to hedonism that I propose to discuss.
The Experience Machine
Hedonism is a form of mental state theory, according to which what matters to well-being
is experiences alone. That leaves it open to the following charge:
The Experience Machine. ‘Suppose there were an experience machine that would
give you any experience you desired. Superduper neuropsychologists could
stimulate your brain so that you would think and feel you were writing a great
novel, or making a friend, or reading an interesting book. All the time you would
be floating in a tank, with electrodes attached to your brain … Would you plug
in? What else can matter to us, other than how our lives feel from the inside?
Nozick believes that the experience machine example shows that various things
do matter to us in addition to our experiences: (1) we want to do certain things; (2) we
want to be a certain kind of person; (3) we want to be able to make contact with a reality
deeper than one that is entirely man-made. We might call these the values of
accomplishment, personhood, and authentic understanding.
Let me avoid the question whether we as individuals would plug in to such a
machine, since it raises a variety of unnecessary technical and empirical issues, and rather
consider the well-being inherent in various parallel lives. First consider P. P writes a great
novel, is courageous, kind, intelligent, witty, and loving, and also makes significant
scientific discoveries. In other words, her life includes all three of the things Nozick
suggests we value in addition to mere experience. Let us add, in the light of some further
doubts Nozick has about such machines in general, that P makes her major life choices
quite autonomously.
And let us add also that P enjoys all these aspects of her life.
Now consider Q. Q is connected to an experience machine from birth, and has
experiences which are introspectively indiscernible from P’s (imagine that the superduper
Nozick 1974: 42-3.
Ibid.: 44-5.

Page 16
neuropsychologists have somehow copied P’s experiences, which are then ‘replayed’ to
Q). According to hedonism, P and Q have exactly the same level of well-being. And that
is surely a claim from which most of us will recoil.
What can the hedonist reply? The first point is that there is an issue here about
philosophical methodology analogous to one which arises in the rather frequent alleged
counter-examples to consequentialism. Most now accept that at the very least those
counter-examples, appealing as they do in the first instance to our gut reactions, are
insufficient immediately to dislodge consequentialism. There are questions to be asked
first about the source of those reactions, and how – if they stand up to reflection – they
are to be taken into account in a coherent moral theory.
Nevertheless, many will say that their gut reaction to the experience machine case
does stand up to reflection, so the task of the hedonist must be somehow to fit that
reaction into the hedonist framework. She can achieve this most effectively, I suggest, by
following strategies similar to those developed by consequentialists to deal with alleged
(i) Inherent pleasure. First, we should note that the things on Nozick’s list of non-
experiential goods all involve experiences, and indeed experiences people tend to enjoy:
the writing of a great novel, and the consequences that flow from that, not least the
contemplation of one’s own achievement; the self-conscious exercise of the virtues;
exercise of human capacities for enquiry and the satisfaction of curosity; the satisfaction
that comes from contemplation of one’s own self-government.
(ii) The paradox of hedonism and secondary principles. But it is not enough for
the hedonist to claim that what we really value in the case of these goods is the enjoyable
experience alone, since we have already allowed that the gut reaction to the experience
machine stands up to reflection. The next stage in the defence of hedonism requires
reference to the paradox of hedonism. Pursuit of these non-experiential goods is likely to
be enjoyable in itself, and to have enjoyable consequences. So, just as ‘secondary’ moral
principles, such as those forbidding killing of the innocent or lying, can be claimed by
consequentialists to have value in so far as they promote the overall consequentialist goal,
See Sidgwick 1907: 401.
Mill [1863]: 2.6, ll. 7-10; Sidgwick 1907: 396.

Page 17
so ‘secondary’ prudential principles – such as those to which Nozick is referring – can be
said by the hedonist to be valuable as guides to the most effective way to promote
(iii) The evolution of value. The hedonist may also ask us to reflect upon the
origin of our values. Consider accomplishment. It is clear that we form our values in the
light of others’ attitudes from an early age. It is more than likely that, in the hunter-
gatherer phase of human development, those who achieved more in the field would have
been rewarded by their fellows, partly with a greater share of the available goods, but
also presumably with esteem and status. Since authentic understanding might also be
seen as a form of accomplishment, and accomplishment itself involves the assertion of
individuality, this evolutionary story might also throw some doubt on the claims to value
of Nozick’s other three goods. But what of pleasure? Again, there is an evolutionary story
to be told about its origins. But on reflection its status as a value seems less contingent
and dependent on the views of others. The values Nozick mentions are highly culturally
relative: some cultures seem to have emerged without the individuality they involve (and
presumably one might tell an evolutionary story about the development of non-
individualist, communal values). But all human beings have always, in different ways,
sought pleasure and avoided pain.
(iv) Perspectives. The question of which is the correct perspective to take on our
lives is difficult. It is certainly true that many people, if asked, would probably place a
significantly higher value on their accomplishments, character, understanding, and
autonomy than on their enjoyments. But from what perspective should we assess the
place of these values? If we attempt to take a more objective, ‘God’s eye’ view of our
own situation, comparing our own achievements and lives in general to that of God or
some being far greater than ourselves, we may return to the anthropocentric perspective
more sceptical of the great significance we attach to these values. What is writing War
and Peace compared to creating the universe? What is even Einstein’s understanding
compared to omniscience? Or imagine that we were all pretty much as clever as
Would doing physics at the level of Einstein then be thought so valuable? And
Sidgwick 1907: 403, 405-6.
Cf. Nozick 1974: 241, 245.

Page 18
if the idea is that value lies in achieving the potential we have, why do we think that
Mozart’s achievements were so much greater than those of a mouse who, by mouse
standards, excelled as much?
When it comes to enjoyment, the matter of perspective seems less significant. Is it
worth going to great lengths, perhaps even suffering significantly, to achieve some goal?
Well, yes, if you enjoy achieving it more than you would have enjoyed some other series
of experiences. There is no need to decide whether to take a God’s eye view or not.
(v) Agency. I have already mentioned some doubts about the alleged value of
individuality, resting on its historical contingency. Consequentialists, or those
sympathetic to consequentialism, might also be asked by the hedonist to note the
emphasis on agency in the values under discussion, especially accomplishment.
According to consequentialist moral theory, what matters is what happens, not who
brings it about; likewise, in her theory of well-being, a consequentialist should be wary of
attaching the kind of value to an agent’s bringing about an outcome, as opposed to that
outcome’s merely occurring, that is involved in the value of accomplishment.
(vi) Free will. Questions may be raised about autonomy, which also have
implications for the value of accomplishment and personhood. For autonomy to be
valuable – as opposed to merely its enjoyable aspects, and the lack of the unpleasant
aspects of its absence – seems to require that we have free will. If we are determined,
then it is not clear why Q should not be as autonomous, and as much an independent
person, as P. But there is some doubt as to whether libertarianism can be made coherent.
At the very least, the controversial nature of libertarianism should make the claims of
autonomy and personhood to value less secure.
(vii) The pleasureless life. There is one further mode of argument the hedonist
may carry across from the consequentialist debate. When non-consequentialist moral
goods are alleged – such as desert or equality – a standard consequentialist response
consists in isolating those alleged goods in cases where nothing good comes of them, in
the hope thereby of throwing them into doubt. Do we really think that a criminal should
be punished even if it does no one any good? Do we really think that equality is desirable,
even in cases in which its promotion harms all concerned?

Page 19
Consider now the life of R. R’s life is as far as is possible like that of P, with all
the enjoyment – and, let us assume – all the suffering stripped out. R writes a great novel,
possesses the virtues, makes scientific advances, and lives autonomously, but takes no
pleasure in what she is doing, in what she is, or in what she finds out. She is motivated,
perhaps, by thoughts about duty, or about the objective good. Is it plausible to think that
that R’s pleasureless life is of any value for her? Certainly it is hard to see how R herself
could care about her own well-being, since caring about something involves feeling
appropriate pro- and con-attitudes, involving enjoyment and suffering, depending on the
fate of that thing.
It has become common in the philosophy of religion to attempt to demonstrate not that
belief in the existence of God is required, but only that it is reasonable. That hedonism is
a reasonable – that is, not clearly implausible – view is all I have sought to show in this
paper. Maybe the time will come for us to bury hedonism for good; but that time is not
Roger Crisp
St Anne’s College, Oxford
9 September 2003

Page 20
Brandt, R.B. 1966. ‘The Concept of Welfare’, in Krupp (ed.): 257-76.
Broad, C.D. 1930. Five Types of Ethical Theory, London: Kegan Paul, Trench, Trubner.
Crisp, R. 1997. Mill on Utilitarianism. London: Routledge.
Csikszenmihalyi, M. 1990. Flow: The Psychology of Optimal Experience, New York:
Harper Perennial.
Diener, E. 1984. ‘Subjective Well-Being’, Psychological Bulletin 95.3: 542-75.
Diener, E. 1999. ‘Subjective Well-Being: Three Decades of Progress’, Psychological
Bulletin 125.2: 276-302.
Feldman, F. 1997. Utilitarianism, Hedonism, and Desert: Essays in Moral Philosophy,
Cambridge: Cambridge University Press.
Frey, B. and Stutzer, A. 2002. Happiness and Economics, Princeton and Oxford:
Princeton University Press.
Gosling, J.C.B. 1969. Pleasure and Desire: The Case for Hedonism Reviewed, Oxford:
Clarendon Press.
Griffin, J. 1986. Well-Being: Its Meaning, Measurement, and Moral Importance, Oxford:
Clarendon Press.
Hooker, B. (ed.) 1993. Rationality, Rules, and Utility: New Essays on the Philosophy of
R.B. Brandt, Boulder, San Francisco, and Oxford: Westview.
Hurka, T. 1987. ‘“Good” and “Good For”’, Mind 96.381: 71-3.
Kagan, S. 1992. ‘The Limits of Well-Being’, in Paul, Miller, and Paul (ed.): 169-89.
Kahneman, D. 1999. ‘Objective Happiness’, in Kahneman, Diener, and Schwartz (ed.): 3-
Kahneman, D., Diener, E., and Schwartz, N. (ed.) 1999. Well-Being: The Foundations of
Hedonic Psychology, New York: Russell Sage Foundation.
Krupp, S.R. (ed.) 1966. The Structure of Economic Science, Englewood Cliffs, NJ:
Layard, R. 2003. Happiness – Has Social Science a Clue?, Lionel Robbins Memorial
Lectures 2002-3, London: Centre for Economic Performance.
Mill, J.S. [1863] 1998: Utilitarianism, ed. Crisp, R., Oxford: Oxford University Press.

Page 21
Mill, J.S. [1869] 1989. ‘Notes on James Mill’s Analysis of the Phenomena of the Human
Mind’, repr.Robson, J. (ed.), Miscellaneous Writings, Toronto and London:
Toronto University Press and Routledge.
Myers, D. and Diener, E. 1995. ‘Who is Happy?’, Psychological Science 6.1: 10-19.
Nozick, R. 1974. Anarchy, State, and Utopia, Oxford: Blackwell.
Parfit, D. 1984. Reasons and Persons, Oxford: Clarendon Press.
Paul, E.F., Miller, F.D., Paul, J. (ed.) 1992. The Good Life and the Human Good,
Cambridge: Cambridge University Press.
Ryan, R. and Deci, E. 2001: ‘On Happiness and Human Potentials: A Review of
Research on Hedonic and Eudaimonic Well-Being’, Annual Review of Psychology
52: 141-66.
Ryle, G. 1954. ‘Pleasure’, Proceedings of the Aristotelian Society, suppl. vol. 28: 135-46.
Sidgwick, H. 1907. The Methods of Ethics, 7
edn., London: Macmillan.
Sprigge, T.L.S. 1988. The Rational Foundations of Ethics, London: Routledge and Kegan
Sumner, L.W. 1993. ‘The Evolution of Utility’, in Hooker (ed.): 97-114.
Sumner, L.W. 1996. Welfare, Happiness, and Ethics, Oxford: Clarendon Press.
Tiberius, V. 2003. ‘“How’s It Going?” Judgments of Overall Life-satisfaction and
Philosophical Theories of Well-Being’, unpub. ts.