HIGHER PLEASURES

The claim that some pleasures are more desirable than others seems to be an obvious one. Everyday practice clearly shows us that individuals often choose one pleasure over another pleasure. To suggest that some of these choices could be accounted for by a difference in quality does not seem particularly outrageous. And indeed, in order to prove his case, Mill sets up clear examples. At first glance, there seems to be no basis for questioning the evidence they provide. The straightforward nature of his examples further contributes to the impression that these are self-evident differences: pig versus person, Socrates versus fool.

In analysing these, it is important to realize that Mill is showing us what we expect to find. When we take a closer look, we notice that the examples to which he appeals are extreme. Who would choose to be a pig instead of a human being, a fool instead of an intelligent person? Since few philosophers choose to align themselves so completely with Jeremy Bentham, our responses to these examples appear to add credence to Mill's position. Yet one's choice could be based on many other factors. And, if it is driven by factors other than the quality of the respective pleasures, then the type of evidence that Mill provides us with falls short of proving that people actually prefer pleasures resulting from higher faculties to those pleasures resulting from lower faculties. A more careful investigation may show us that our preferences are less conclusive than Mill believes. To do this, we can begin with Mill's definition of higher pleasures, followed by a close analysis of the empirical evidence that he suggests is so unquestionable.

The adequacy of Mill's empirical evidence is something that has been ignored in the literature analysing Mill's distinction between higher and lower pleasures. Yet this is an important factor to consider in determining the success of Mill's own theory. Furthermore, since Mill's theory provides a model for other utilitarians who draw a distinction between higher and lower pleasures, the adequacy of his evidence is important. If Mill is unsuccessful, then other utilitarian theories cannot simply assume that reasons for subscribing to the distinction exist and have already been established. Although Mill purports to have shown decisively that people prefer pleasures resulting from their higher faculties, he is unable to do so with the evidence he uses. In fact, as will be shown below, the type of evidence Mill provides us with is neither necessary nor sufficient. This leaves utilitarian theories attempting to employ a qualitative distinction with further work to do.

MILL'S ARGUMENT

Mill's argument for the accuracy of the higher/lower distinction begins with a clarification of what he means by a ‘difference of quality in pleasures’.Footnote ⁵ He explains that in addition to varying in quantity, the value of pleasures can also vary in quality. Mill defines higher pleasures as follows: ‘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.’Footnote ⁶ I will refer to this statement as the weaker conditional. Mill uses this statement to explain what he means by ‘difference of quality’ and follows this with a further clarification which serves to emphasize the importance of qualitative differences. Because Mill does not stipulate that the initial definition and the follow-up clarification should be taken together as one definition, I will not assume that these form a single definition. Instead, I will proceed with caution and read the weaker conditional as a basic test that can be used to distinguish higher from lower pleasures. In line with this approach, I will take the stronger conditional (outlined below) as a test that shows the possible extent to which significant differences of quality can, and should, impact preferences. Here Mill suggests that for cases in which people are familiar with both pleasures but would refuse to give up one for any amount of the other, the quality of that pleasure is so great as to make the quantity offered almost or entirely irrelevant. Mill states:

Taken together, the weaker and stronger conditional are meant to establish a way of identifying higher pleasures and determining which pleasures are so significantly higher in quality that considerations of quality ought to outweigh considerations of quantity. For Mill, the next step is to demonstrate that there are cases in which considerations of quality justifiably outrank considerations of quantity by providing instances of actual cases. Unfortunately, Mill's proposed evidence and test are not as straightforward as he thought. As these are key components of Mill's proof of the distinction between higher and lower pleasures, they deserve further scrutiny. A closer examination of Mill's examples will show that although they appear convincing, there is an inherent flaw in his proposed empirical test.

Mill suggests that we need only look to everyday life to find proof of his distinction. Although most people lead lives with a moderate amount of pleasure, they would be unwilling to trade their lives (and their moderate quantity of pleasure) for a life as one of the ‘lower animals’ even if they were offered ‘the fullest allowance of a beast's pleasures’.Footnote ⁸ Mill views this as clear proof that some pleasures are clearly higher in quality than other pleasures.

Indeed, Mill's evidence appears hard to refute. As he suggests, what ‘intelligent human being’ would agree to become a fool or a pig? We have little difficulty accepting his claim that few people would agree to be transformed into the ‘lower animals’, and that those who would agree are an exception to the rule – either incredibly unhappy with their present circumstances or clearly mistaken about what they really want.

This is such a poignant and seductive example that we forget to ask ourselves whether it really proves what Mill thinks it proves. And, as long as we are thinking in terms such as human being versus pig, intelligent being versus fool, we are unlikely to see how flawed Mill's evidence really is. In order to determine the success or failure of this evidence more clearly, we need to analyse the relation between the evidence and what Mill is trying to prove.

MILL'S TEST FOR HIGHER QUALITY

According to Mill, whether one pleasure is higher in quality than another can essentially be determined by looking at whether most people prefer it to the other, at least provided that the people in question are familiar with both pleasures, are making educated decisions and have no ulterior motive for the preference, i.e. the preference in question is not merely greater in quantity or preferred out of ‘a feeling of moral obligation’ to do so.Footnote ⁹ This gives us a way to differentiate between higher and lower pleasures. As Elizabeth Anderson points out, it seems to be ‘a straightforward empirical test’.Footnote ¹⁰ Anderson's investigation provides a starting point from which to question the supposed straightforwardness of Mill's test. Anderson's comparison of the pleasures of ‘reading books’ to the pleasures of ‘eating desserts’ exposes a significant problem. Although unwilling to ‘give up reading books for any amount of the pleasures of eating dessert’, she is also unwilling to ‘give up desserts for unlimited opportunities to read books’.Footnote ¹¹

In an attempt to rescue Mill, Anderson entertains the possibility that this problem could be limited to particular cases in which a person is equally unwilling to give up either pleasure and eliminated for cases in which a person is willing to give up one but unwilling to give up the other. If this were the case, then Mill could possibly show the legitimacy of distinguishing between higher and lower pleasures.Footnote ¹² But Anderson is right in rejecting this proposal. As she argues, it ‘cannot be a necessary condition for a difference in rank among goods’, because ‘it is not unreasonable to prefer a life which balances the pursuit of higher and lower pleasures to one singlemindedly devoted to the higher ones alone’.Footnote ¹³ Instead, Anderson attempts to solve Mill's difficulty by revising the ‘preference criterion’ so that it is ‘not whether experienced people would be willing to give up A for B but whether they would judge it worthwhile to do so’.Footnote ¹⁴ In doing so, she appeals to ‘nonhedonic value judgments’.Footnote ¹⁵ While this approach is worth considering, it saddles Mill with non-hedonistic values – something which seems to be a significant departure from Mill's intended theory. Given the problematic nature of this move and the fact that it would still fall short of resolving the problem at hand, it is not an approach that I will follow.

While Anderson's examples of reading books and eating desserts begin to reveal the problematic nature of Mill's preference criteria, there is more to the difficulty than her analysis suggests. Mill claims that if a competent person refuses to give up a higher pleasure for ‘any quantity of the other pleasure . . ., we are justified in ascribing to the preferred enjoyment a superiority in quality so far outweighing quantity’.Footnote ¹⁶ However, his earlier, weaker conditional does not require a proof that they ‘would not resign it for any quantity of the other pleasure’. Instead, this conditional defines difference in the quality of pleasures as that for which competent people have a ‘decided preference’. This is then what he needs to show in order to establish the existence of higher pleasures. And it may be all he intends to show when he goes on to suggest that ‘it is an unquestionable fact’ that competent people ‘give a most marked preference to the manner of existence that employs their higher faculties’, since if people show a ‘most marked preference’ then they can also automatically be said to have shown a ‘marked preference’.Footnote ¹⁷

Mill's own use of the weaker conditional shows that it might not be necessary for him to prove that people would be willing to give up desserts completely for a life of reading books, something that even philosophers might be hesitant to do. If the weaker conditional is sufficient, it might not be necessary for Mill to find people who are willing to give up all desserts for a future with reading in order to show a difference in quality. It may only be necessary to show that competent people have a ‘decided preference’ for reading books over eating desserts. Of course, this will not be sufficient to show that quality of pleasure always trumps quantity, but as Schmidt-Petri points out, Mill need not be read as requiring that ‘considerations about the quality of a pleasure automatically outweigh considerations about the quantity in which it is available’.Footnote ¹⁸

This seems to rescue Mill from Anderson's criticism to some extent, because it allows for situations in which people are reluctant to give up either reading books or consuming chocolate. However, this still does not put Mill in the clear. Although Mill need only show that people have a ‘decided preference’ for one pleasure over another in order to establish the existence of higher pleasures, his only attempt to do so involves showing that people have a ‘most marked preference’ for pleasures resulting from their higher faculties even if the alternative were to consist of ‘the fullest allowance’ of lower pleasures. Thus, his attempt focuses on proving the stronger conditional. (This is not particularly surprising since Mill considers his existing evidence ‘unquestionable’ and successfully proving the stronger claim would also provide evidence for the weaker claim.) Mill gives us the following evidence:

So Mill's ‘unquestionable fact’ is his rather reasonable belief that most people would refuse to give up their lives as human beings for lives as ‘lower animals’. But, even if Mill's reasonable belief turns out to be true, this particular example does not prove his stronger conditional. And if the only evidence he provides for the weaker conditional is the evidence he uses in attempting to establish the stronger conditional, then the weaker conditional is also unsupported.

As suggested earlier, it seems likely that most of us would agree that we are unwilling to give up our lives as humans for lives as pigs. So what is wrong with the seemingly reasonable evidence Mill provides? The problem is that this fact falls short of proving that the pleasures in question are actually higher pleasures as well as failing to prove that the reason for our decision depends on quality.

Looking at other examples may help to clarify this point. Given the choice between a life with some apples and oranges or a life full of as many oranges as I can possibly enjoy, I am likely to prefer the life with some apples and oranges. But this does not allow me to conclude that the pleasure I gain from apples is of a higher quality than the pleasure I gain from oranges. Similarly, people may prefer to live life with both their father and their mother, rather than a completely pleasurable life with just their father. Again, this does not seem to warrant the conclusion that life with one's mother is preferable to life with one's father. Although this is one possible conclusion, there are other possible conclusions. It may only show that life with both parents is preferable to life with just one parent. Or it may show that when we are given a choice, we prefer a variety of pleasures. If these examples fall short of proving that the preference in question is based on qualitative superiority, then Mill's evidence is insufficient.

A possible objection could be raised at this point. The examples outlined above do not involve pleasures resulting from different faculties. Mill's example, on the other hand, clearly involves a comparison of pleasures attainable by different faculties – in particular a beast's pleasures versus a human's pleasures. This can be easily remedied. By turning again to Elizabeth Anderson's example of reading books and eating desserts, we can see that the same problems result. True, we may choose a life with moderate pleasure gained from both reading books and eating desserts over a life with the ‘fullest allowance’ of pleasure but limited to dessert-eating. So does this prove that reading books is preferable to eating desserts? As philosophers, we may well be predisposed to agree with the conclusion. However, before we do so, we need to remind ourselves that the question is whether the fact that we are unwilling to give up a life with both reading and eating for a life with just eating is sufficient proof that reading is preferable.Footnote ²⁰

There is a danger here that the subject matter of this example could obscure the real issue at stake – the relation of Mill's evidence to what he is trying to prove. I suggest we look to another more neutral example, one in which we have not already accepted the higher/lower distinction that Mill is trying to prove. Imagine that you are given the choice between living a somewhat satisfactory or pleasurable life with five senses or a life with the ‘fullest allowance’ of pleasure possible from four senses. Although we may again choose the life with five senses, I would again hesitate to claim that this proves that the fifth sense, e.g. ‘hearing’, is better than any or all of the other senses. At best, this seems to show our preference for living with five senses over living with four senses. Although this example involves different faculties and it could even be argued that it provides evidence that there are different types of pleasures, it doesn't seem to prove that Mill is justified in concluding that a difference in faculties (coupled with a refusal to resign two types for one) automatically shows that the resulting pleasures vary in quality (higher or lower).Footnote ²¹

Thus, passing Mill's empirical ‘test’ is not an adequate proof of the distinction between higher and lower pleasures. As such, his example falls short of proving a ‘marked preference’ of one pleasure over another. But this is not the only problem. Not only does Mill's example fail to provide sufficient proof of his higher/lower distinction, but it is also not clear that it even provides a necessary condition for proving a higher/lower distinction. In order to supply a necessary condition, we must read Mill as suggesting that quality considerations must trump quantitative considerations all of the time. But is this what Mill had in mind? Schmidt-Petri suggests it is not, while Riley argues that it is. To resolve this question we need to go back to the text once again.

Schmidt-Petri is certainly correct in pointing out that Mill's conditional is not written as a two-way conditional, and as such, it should not be assumed to read, ‘If some pleasure is of higher quality, then it will be or ought to be chosen over the pleasure of lower quality regardless of their respective quantities.’Footnote ²² However, Schmidt-Petri tries to take this fact too far. The problem arises when Schmidt-Petri moves from the fact that we cannot just assume the biconditional to the claim that there is no reason to think that Mill believes that considerations of quality, strictly speaking, ever really trump considerations of quantity. Here Schmidt-Petri believes that considerations of quality are relevant only because they speak to the intensity and amount of pleasure that one can obtain.

The conclusion Schmidt-Petri reaches here is unwarranted. First of all, it is not necessitated by the passage in question. Second, it directly conflicts with other passages where Mill clearly states that differences of quality are not simply to be understood as mere differences in intensity. One example of this is Mill's suggestion:

This passage provides clear evidence that there are at least some higher pleasures which cannot be differentiated from one another simply on the basis of their intensity. Here it is likely that one of the problems with Schmidt-Petri's response can be found in his use of examples. Whereas Mill's view of the difference in quality of two pleasures clearly involves a difference in the type of pleasure, Schmidt-Petri uses examples which involve the same kind of pleasure, e.g. two bottles of wine or two cars.Footnote ²⁴ This is misleading since both examples, while involving a clear difference in quality, still involve the same type of pleasure.

Jonathan Riley, on the other hand, argues that Mill subscribed to the stronger view, suggesting that Mill did in fact hold that considerations of quality always ought to override considerations of quantity.Footnote ²⁵ Riley's claim is a reasonable one. It is quite plausible that Mill personally believed that qualitative superiority ought to be preferred to quantity. However, while passages in Mill's Utilitarianism clearly entail that qualitative superiority may outweigh and, in some cases at least, ought to outweigh considerations of quantity, these sections do not entail that considerations of quality must always override considerations of quantity. While Mill's empirical test allows us to conclude that qualitative superiority can be or should be an overriding consideration in at least some cases, it falls short of proving that Mill was committed to the claim that qualitative superiority is an overriding factor in all cases involving higher and lower pleasures. Instead, we are left with a test which only shows us that Mill believes that there are at least some pleasures for which qualitative considerations do in fact outweigh quantity.Footnote ²⁶

To resolve this dispute, we need to be careful as to the type of pleasures we are comparing. It is clear that, in order to reflect accurately what Mill is trying to do, we need to choose examples in which the pleasures compared are pleasures which differ in kind. Here we are better off using examples such as those chosen by Anderson. Her comparison of the pleasure we get from books versus the pleasure we get from chocolate clearly involves a difference ‘in kind’ and, as such, is a more helpful way of thinking about the problem. Other examples, such as Riley's discussion of ‘justice’ and the pleasure that arises from it, can further help to expand our understanding of Mill's theory and the type and degree of qualitative difference which Mill is trying to allow for within his theory.Footnote ²⁷ Riley's example of ‘justice’ can also help us to understand the kinds of pleasures Mill may have in mind when he suggests that there are certain pleasures for which considerations of quantity can never override one's choice.Footnote ²⁸ Recent empirical research may provide further support for this by furnishing examples in which the presence or absence of certain beliefs about the world (e.g. beliefs about justice or inequality) may influence – significantly – an individual's own degree of happiness.Footnote ²⁹ However, even these possibilities do not demonstrate that differences in quality are always such that they must trump quantity or that all pleasures which are ‘different in kind’ are pleasures which are so superior that considerations of quantity will always be secondary.