2. The basic argument of Utilitarianism IV

I begin with the argument for hedonism from Utilitarianism IV.Footnote ¹⁰ As Mill himself notes, the first sequence of this argument shows only that pleasure is an intrinsic good, and not that it is the only intrinsic good – a more ambitious claim that I address in Section 8 below. In this section, I first offer a simplified reconstruction of the basic argument, before offering important supplementations to show how the argument can be more suitably – and substantively – formulated. Here is the basic argument:

Let us begin with (P1), which is offered as a definition of intrinsic value. It is crucial to notice that Mill is understanding inquiry about the good in a classic, agential way, as oriented around action and motivation.Footnote ¹⁴ That is, the sense of intrinsic value he is interested in is the value something has as an end of action, where the intended contrast is with what is valuable as a means to some further end.Footnote ¹⁵ Mill makes use of the notion of desirability in (P1) precisely to mark his interest in this end-sense of intrinsic value – in what is ultimately worth pursuing as an end in our actions. “Desirable” is a good synonym here for “worthy of pursuit.”

The controversy concerning Mill's argument typically stems from his explanation for (P2), that pleasure is desirable for its own sake. As I noted in the introduction, in this explanation he infamously suggests that just as the only proof that an object is visible is that people see it, the only proof that anything is desirable is that people actually desire it. And of course, as Mill points out, people do in fact desire pleasure for its own sake. Thus, pleasure is desirable in itself.Footnote ¹⁶ G. E. Moore excoriated Mill for making what he took to be an obvious and elementary mistake; something's being visible is about its being capable of being seen, while something's being desirable is about whether or not it is worthy of being desired.Footnote ¹⁷ However, I think there is a more charitable – and textually accurate – way to understand Mill's point, which requires us to attend to his epistemological remarks here: not only should we investigate the good (i.e. what is worth pursuing) by considering what people want and pursue, but this is the only way to investigate the nature of the good. So, we must ask, why does Mill take pursuit as a guide to value?

3. Mill's empiricist epistemology: “proof” of an axiom of reason

Prior to making the case for (P2) in IV.3, Mill makes several crucial remarks about the nature of the inquiry being undertaken. In IV.1, just a couple of paragraphs earlier, he repeats a point he stressed in I.5, namely that “questions of ultimate ends do not admit of proof, in the ordinary acceptation of the term. To be incapable of proof by reasoning is common to all first principles; to the first premises of our knowledge, as well as those of our conduct.”Footnote ¹⁸ That is, one cannot provide a logical proof to support an axiom of our reasoning, since the nature of such proofs is to use these axioms or first principles to derive new pieces of knowledge.Footnote ¹⁹ He most clearly and explicitly lays out this foundationalist position in A System of Logic, a text which is rarely read by contemporary ethicists (despite Mill viewing it as his magnum opus). Consider his comments there:

For Mill, this is true for both theoretical and practical axioms, and since ultimate ends – i.e. intrinsic goods – are axioms of practical reason, proof by reason is not a possible avenue to show that something is intrinsically good.Footnote ²¹

So, outside of this usage of the term proof (i.e. proof by reason), what could constitute proof of an axiom? Mill answers that first principles “may be the subject of a direct appeal to [two] faculties…namely, our senses, and our internal consciousness.”Footnote ²² Here we see Mill's empiricist commitments come to the fore: axioms are grounded in experience, which for Mill includes the senses and something he calls “internal consciousness.” While the language he uses is slightly different, Mill is here making a similar distinction to one found in Epicurean empiricism between sense-perception and sense-feeling. For the Epicureans, the objects of sense-perception consist of what they call impressions of external states of affairs, like the heat of a fire, the sound of a dog's bark, and so on. By contrast, sense-feelings are impressions of internal states of affairs; specifically, the feelings of pleasure and pain.Footnote ²³ Mill makes much the same move here, though he ultimately allows for more variety in terms of what counts as an internal state of affairs.Footnote ²⁴ Consider what he identifies in A System of Logic as paradigmatic instances of foundational knowledge, i.e. axioms or first principles of reason:

Though he does not explicitly say that “internal consciousness” refers to mental feelings in Utilitarianism, he does do so elsewhere:

Just as the Epicureans hold that we can be certain we are seeing what we are seeing (even if it is a mirage or illusion), for Mill the fact that we feel what we feel is “sufficiently proved by consciousness itself” – i.e. by the experience of those feelings.Footnote ^27, Footnote ²⁸

With this context in mind, we can better understand the role that (P1) is playing in Mill's argument. As we have seen, Mill's empiricist contention – long underappreciated by ethicists – is that we can only prove that something is intrinsically good via an appeal to experience. Crucially for Mill, desire is “a state of passive sensibility” – that is, it is part of our experience (in this case our internal consciousness).Footnote ²⁹ Through (P1), which holds that what is desirable for its own sake is good in itself, Mill provides principled grounds for appealing to experience via desire to prove that something is intrinsically good. That is, Mill is appealing to our experiences of what we desire for its own sake as “proof” for what is desirable for its own sake, and thereby for what is intrinsically good.

At this juncture, one may rightly object that Mill is still susceptible to Moore's criticism. Just because we know that we desire pleasure for its own sake via our experience of desire does not mean that we should desire it for its own sake – that is, it does not mean that it is desirable. However, such an objection does not consider what, for Mill, constitutes an appeal to experience such that the appeal could ground an axiom of reason. On his view, the question of whether or not pleasure is desirable for its own sake is indeed “a question of fact and experience,” but it is “dependent, like all similar questions, upon evidence.” The answer, Mill goes on, “can only be determined by practiced self-consciousness and self-observation assisted by observation of others.”Footnote ³⁰ Whatever is meant here by “practiced self-consciousness and self-observation assisted by observation of others,” this is clearly what Mill requires of an appeal to experience for that appeal to ground an axiom of reason.Footnote ³¹

Fortunately, we do have the resources to understand what Mill meant by these processes of self-reflection: namely, his long treatise on empiricist epistemology, A System of Logic. There, contra Kant, he rejects the possibility of a priori knowledge (and thus the possibility of grounding axioms of reason via reason itself), and instead understands reasoning as a natural process:

It is Mill's foundational commitment to empiricism that lies behind his assertion that we can only figure out what is desirable in itself by considering what is actually desired. In John Skorupski's helpful formulation, as Mill sees it, the way we figure out what we should believe, what is good, how to act, and so on, is “by careful scrutiny of how we actually reason” and by “reflective analysis of which principles in this practice of reasoning turn out to be treated by us as normatively basic: ‘seeing the agent at work’.”Footnote ³³ For Mill, this is the only kind of evidence that can be offered for the normative claims that we hold.

4. Thinking from within

Although Mill never gave a name to this method of scrutiny and reflective analysis, I will follow Skorpuski in calling it “thinking from within.”Footnote ³⁴ This method of reflection is important for Mill because it investigates whether or not we can reduce some normative dispositions (feeling, thinking, believing, etc.) to others. At the same time, it can raise questions about whether our normative dispositions are justified in light of considerations about how they were formed, and if they are found to be unjustified, this subverts their normative authority. For example, let us say I think that Justin Bieber's singing abilities are mediocre at best. If through a process of self-reflection it turns out that I can show this thought to be solely a result of jealousy (I can reduce the thought to a feeling), the thought will be subverted. Even though it might still be true that Bieber's singing is mediocre, what I have discovered is that my reasons for thinking so were not good ones. That is, this exercise in examining alternative explanations does not show that a normative view, belief, or opinion is false but that it is not justified.

It is crucial to understand that thinking from within, far from looking to subvert our normative dispositions, actually aims to vindicate those thoughts, feelings, and judgments that survive the relevant form of scrutiny.Footnote ³⁵ It is the method through which we seek to establish the truth of basic normative claims – i.e. the claims we take to be first principles. If the claim is not subvertible by this process of thinking from within, we can consider this resilience a kind of “proof” of the claim, though obviously not proof in the traditional sense. This makes sense of Mill's comments about widening the scope of what we consider proof in Utilitarianism I.5, often ignored by his critics:

If upon serious scrutiny and reflection upon our experience (i.e. on sensation and/or internal consciousness), we cannot find an alternate explanation for a first practical principle that would subvert it in such a way that we withhold assent to it, then we offer our (albeit potentially tentative) assent. For Mill, this is functionally equivalent to the standard sense of proof, even if it is not the same. It is worth pointing out that this is a common view about the nature of evidence and proof for fallibilists about knowledge – who we can now see include Mill among their ranks.Footnote ³⁷ For Mill, any of the things we take ourselves to know could turn out to be false as we continue what he hopes is a life-long process of self-reflection and thinking from within. But in the absence of defeaters, Mill is perfectly happy to assent to normative claims that are resilient in the face of this reflective process.

5. Elaborating the argument from Utilitarianism IV

Let us return to Mill's argument for the intrinsic goodness of pleasure, which can now be elaborated so as to include the significant epistemological points we have just discussed. First, we can attribute to Mill the following three premises:

As we have seen, (P1) comes from Mill's empiricism and his rejection of the possibility of grounding first principles of reason a priori via reason itself (his rejection of Kant).Footnote ⁴⁰ The claim made in (P2) is simply assumed by Mill. There is a long history in philosophy of taking the central question of ethics to be “What is the ultimate end at which all actions aim?,” going back to Greek philosophy. Mill places himself within this tradition, which is why he assumes (P2) and does not motivate it. (P3) simply combines (P1) and (P2); it replaces “a first principle or axiom” with “something's intrinsic goodness” in (P1).

We can now attribute to Mill the following four premises, beginning with the premise I originally cast as the first step of his argument:

To my mind, the move in (P4) to understand the good in an agential, conative way is plausible on its own terms. After all, we presumably think that intrinsically good ends are ends that we ought to pursue for themselves – i.e. ends that are desirable for their own sake. Moreover, (P4) serves Mill as a way to locate goodness in experience via (P5), which establishes desire as part of our experience. In service of treating goodness as a feature of experience, (P6) refines the inquiry at hand from goodness to desirability by combining (P3) and (P4), and (P7) continues this refinement by specifying which aspect of experience – desire, from (P5) – can be appealed to as the “proof” of desirability brought up in (P6).

At this juncture, we can see that the supposedly controversial idea in (P7), that we must investigate desire to determine what is desirable just as we must investigate what is seen to determine what is visible, is not so bizarre. As Henry West puts it, the analogy is that “as judgments of matters of facts such as visibility are based on the evidence of the senses and corrected by further evidence of the senses” accompanied by the relevant processes of psychological introspection, “so judgments of what is desirable are based on what is desired and corrected by further evidence of what is desired” – evidence that is produced by the process of thinking from within.Footnote ⁴⁴ Following this, we must note one more epistemological remark and make one last refinement:

Here, (P8) simply serves to explain what sort of “appeal to experience” Mill thinks can ground a practical axiom: his method of “thinking from within.” The following premise, (P9), simply replaces “appeal to the experience of desire itself” from (P7) with the method of “thinking from within” outlined in (P8). Essentially, (P9) is asserting that if, in thinking from within about what we desire, “we find that we desire this (whatever it may be) for itself, and acknowledge it as indeed desirable [in itself] – rather than subverting that judgment and trying to change our desires” – this constitutes evidence and “proof” of a fallibilist sort for the intrinsic desirability of that thing.Footnote ⁴⁶

Now we can finally take the last steps of the argument, and combine the method of thinking from within with what has been said about desire and goodness:

It is in (P10) that Mill makes his most important, and misunderstood assertion, which we can now properly understand as saying the following. In thinking from within about our experience of desire, we find that we desire pleasure for itself, and in further scrutinizing the nature of this desire we acknowledge that pleasure is indeed desirable in itself; we find nothing that could subvert the claim and change our desire for it.Footnote ⁴⁸ Although thinking from within cannot furnish “proof” in the traditional sense, for Mill it can functionally play the same role, which means we are now in a position to assert (P11) from (P9) and (P10): pleasure is desirable in itself. And finally, we may at last conclude from this and from (P4) – which holds that what is desirable in itself is good in itself – that pleasure is intrinsically good.

6. Ancient and early modern empiricist hedonism

With the above richer understanding of Mill's argument for the intrinsic value of pleasure in mind, we can proceed to discussing the more ambitious claim of classic hedonism, that pleasure is the only thing that is intrinsically good. Before discussing how Mill arrives at this position, I want to first briefly discuss how his hedonist predecessors made this argument; in particular, the ancient Epicurean school and the early-modern British empiricists. These philosophers were all empiricists, and so took all our concepts and knowledge to ultimately be grounded in experience, including practical knowledge and concepts. Because of this, their practical philosophy centered around identifying the practical component of experience, out of which we build our ethical concepts and knowledge. As we will see, Mill's position on pleasure as the ultimate end is in many ways a nuanced evolution from the more or less uniform position of his predecessors.

Let us begin by considering the hedonism of Epicurus. His argument for the intrinsic goodness of pleasure was staunchly empiricist, and like Mill he denied that any traditional kind of argument was necessary – or could even be given – to show why pleasure is to be pursued for its own sake. Instead, an appeal to experience is required – in this case, a direct appeal to experience without Mill's process of thinking from within to examine it. As Cicero put it, Epicurus “holds that we perceive [the intrinsic goodness of pleasure and badness of pain], as we perceive that fire is hot, snow white, honey sweet; it is unnecessary to prove any of these things with sophisticated reasoning; it is enough just to point them out.”Footnote ⁴⁹ On the Epicurean view, pleasure and pain are the practical side of experience – they are the evaluative elements of experience that inform action and can get us to move – and we know them to be good immediately from experiencing them. It is crucial to note that for the Epicureans, pleasure and pain are the only candidates found in experience upon which we can base all our other normative concepts: “Pleasure is our first and kindred good. It is the starting-point of every choice and of every aversion, and to it we come back, inasmuch as we make feeling the rule by which to judge of every good thing.”Footnote ⁵⁰ If the Epicureans are right that perception through the five senses and feeling through the two modes of pleasure and pain are the building blocks of all experience, then given their empiricism it would make sense that pleasure is the only possible candidate for intrinsic goodness (and pain for intrinsic badness); all other goodness would have to be derivative in some way from it.

The early-modern British empiricists, whose tradition Mill came from, shared this same position – though unlike the Epicureans, they rarely provided motivation for the view.Footnote ⁵¹ As Richard Fumerton has noted, generally speaking the early-modern empiricists “usually talked relatively little about the nature of our knowledge of our own mental states, nor of the need to build all other empirical knowledge on inference from that secure foundation. It's almost as if they took the framework within which they were working to be simply obvious.”Footnote ⁵² Despite this lack of explicit motivation, we can see that they were thinking along similar lines to the Epicureans, at least as far as arguing for the intrinsic goodness of pleasure is concerned. Hobbes, for example, noted that each man calls his own pleasure good and his own pain bad.Footnote ⁵³ His explanation for this is that we get the conceptions of pleasure and pain immediately via the senses; that is, our conception of them as good and bad respectively comes to us immediately from experience.Footnote ⁵⁴ He does not mention any other candidates for the practical component of experience, so it is likely that much like the Epicureans, he did not take there to be others. Similarly, Locke held that we come to know pleasure and pain “only by experience.”Footnote ⁵⁵ We know that they are good and bad respectively by reflecting upon that experience – in doing so, we will know that they are good and bad from “what we feel in ourselves” in experience.Footnote ⁵⁶,Footnote ⁵⁷ Only pleasure and pain are discussed as the practical elements of experience. This goes for Hume as well, who argued that like all other ideas, practical ones come from some impression or sentiment in experience, and that the relevant impressions by which good and bad are known “are nothing but particular pains or pleasures” – that is, particular experiences of pain or pleasure (not the abstract idea of pain or pleasure in general).Footnote ⁵⁸

As we can see, the classic hedonists that came before Mill similarly grounded their arguments in empiricism; they were not so much focused on arguing in a traditional way for the claim that pleasure is the only intrinsically good thing, but rather sought to explain its apparent axiomaticity by appealing to something that seems to be a secure source of knowledge: our own experience. As we saw explicitly in Mill's argument, no standard kind of proof can be given for axioms, since by their very nature they cannot be deduced from other propositions. In light of this, Mill's intellectual predecessors sought to provide an understanding of how to assess or ground an axiomatic claim, and their explanation in this case came by direct appeal to experience.

7. A note on the consequences of separating hedonism from empiricism

Before turning to the final part of Mill's argument for pleasure as the sole intrinsic good, I want to briefly comment on what is lost when the nuances of Mill's argument, as I have reconstructed it thus far, do not receive adequate attention from ethicists. Although hedonism has a long history in ethics, the only hedonist who is regularly taught to undergraduates is Mill.Footnote ⁵⁹ Offered as the representative of hedonism in most ethics classrooms, he is typically taught in a way that separates his ethical arguments in Utilitarianism from his empiricism. There are a number of good pedagogical reasons for this, including that the classic Moorean misreading of Mill's argument simplifies the argument to a level of abstraction that is appropriate for the introductory level, and enables instructors to introduce the concept of normativity to young philosophers.Footnote ⁶⁰

However, because this way of teaching Mill is so widespread, the classic hedonism known to philosophers at large in the present day is precisely this falsely simplified hedonism, which ignores classic hedonism's traditional roots in empiricist epistemology. The cost of this pedagogical approach to Mill is philosophical. In not understanding that classic hedonism and empiricism are traditionally a packaged deal, contemporary ethicists often fail to see the most important metaethical merits and nuances of classic hedonism – namely, that it is best understood as the result of endorsing two broad positions: (1) empiricism, considered a serious position in epistemology to this day, and (2) a view of pleasure on which it is the practical element of sense experience, also considered a serious position in the philosophy of pleasure today.Footnote ⁶¹

8. Mill's classic hedonism: pleasure as the “only” intrinsic good

As we saw above in Section 6, Mill's hedonist predecessors largely use the same method to arrive at classic hedonism: a direct appeal to experience, in which they combine their empiricist epistemology with the view that pleasure is the positive practical component of experience, and pain its negative analogue. Mill expands on this basic methodology by going beyond a direct appeal and asking us to reflect upon our experience of pleasure in a particular way to rationally evaluate what experience seems to tell us – namely, that pleasure is intrinsically desirable (and thus good). This gives us Mill's (P10) above, according to which anyone who thinks from within about their experiences of desire and pleasure will acknowledge the intrinsic desirability of pleasure.Footnote ⁶² As I will discuss below, Mill's full explanation for (P10) involves him positing an extremely intimate connection between desire itself and pleasure. Among Mill scholars, there is significant debate about how to interpret what he says about this connection in a way that coheres with other remarks he makes both in and out of Utilitarianism. For my purposes here, I am not interested in weighing in on these interpretative debates. Rather, I want to consider the two interpretive options endorsed by most scholars, show that each of them offers a philosophically plausible account of the relationship between desire and pleasure, and show how these plausible accounts lead to the thesis of classic hedonism: that pleasure is the only thing that is intrinsically good.

Before looking at any interpretations, let us first consider what Mill says about why thinking from within will not subvert the claim that pleasure is desirable in itself. He does not merely say that there is no evidence this process can unearth which would undermine pleasure's intrinsic desirability, but that the evidence it unearths makes it impossible to deny this claim. According to Mill, the process of thinking from within – which can tell us which normative dispositions can be reduced to others – reveals that “to think of an object as desirable (unless for the sake of its consequences), and to think of it as pleasant, are one and the same thing; and that to desire anything, except in proportion as the idea of it is pleasant, is a physical and metaphysical impossibility.”Footnote ⁶³ Upon engaging in this process himself, Mill came to think that “desirable” and “pleasant” are just “two different modes of naming the same psychological fact.”Footnote ⁶⁴ He takes this point – that pleasant and desirable are the same thing – to be self-evident: “[s]o obvious does this appear to me, that I expect it will hardly be disputed.”Footnote ⁶⁵

With this in mind, Mill's assertion of (P10) makes sense, for denying that pleasure is desirable is in effect to deny that pleasure is pleasant. That is, given his view that thinking from within about our experiences of desire and pleasure reveal that “desirable” and “pleasant” are the same thing, it would be impossible to reject the claim that pleasure is intrinsically desirable, since that would mean rejecting the claim that pleasure is pleasant. Presumably, Mill's thought here is that no one would deny that pleasure is pleasant, just as no one would deny that a circle is circular. Thus, we can see that his argument that pleasure is intrinsically good because it is desirable really boils down to this: pleasure is intrinsically good because pleasure is pleasant.

There are at least two ways to interpret Mill's claim that “pleasant” and “desirable” are the same. The first, accepted by many Mill scholars, is to take what Mill says at face value and read this as an identity claim that is discovered empirically.Footnote ⁶⁶ Given this identity relation, if Mill is right about both his understanding of intrinsic goodness as what is desirable for its own sake and in his empiricist contention that the only evidence we have for what is intrinsically desirable is our experience of desiring things for their own sake, then pleasure is the only possible intrinsic good, because pleasant things are the only things we desire more of for their own sake. That is, on this reading pleasure and intrinsic goodness amount to the same thing; there is not one property called “intrinsically good” and another called “pleasant.” Rather, to borrow a phrase from Mill, these words are “two different modes of naming the same psychological fact.”Footnote ⁶⁷ In the contemporary literature on the nature of pleasure, this identity relation is considered a plausible view of the relationship between pleasure and desire; one of the main theories on the table as a live option is the motivational theory of pleasure, according to which pleasure reduces to desire – what it is for something to be pleasant just is for us to desire more of it for its own sake.Footnote ⁶⁸ Thus, on this interpretation of Mill, we have good reason to take his view seriously, or at least as seriously as we take contemporary motivational theorists – Mill's position is one worth engaging with.

A second way to understand what Mill means by saying “desirable” and “pleasant” are the same focuses on what it means to “think of an object” as desirable or pleasant. On this interpretation, thinking from within about desire and pleasure reveals that when we desire something for its own sake, we desire it precisely because we think of it as pleasant.Footnote ⁶⁹ As Skorupski notes, this idea, that “whatever we desire, we desire under the idea of it as pleasant,” is regarded by Mill “as a psychological law about all desires.”Footnote ⁷⁰ In other words, Mill can be understood as endorsing a kind of psychological hedonism; in order to be motivated to pursue something as an end at all, we must in some way conceive of that thing as pleasant.Footnote ⁷¹ Thus, on Mill's view, if I want chocolate cake I want it precisely because I think its taste will be pleasant – it will taste good. Similarly, if I want to attend a live performance of the Concierto de Aranjuez, I want to do this precisely because I think of watching and listening to such a performance as pleasant in some way.

If we understand Mill as a psychological hedonist, then this explains why he asserts both that pleasure is the sole intrinsic good and that things like health or listening to music “are desired and desirable in and for themselves.”Footnote ⁷² For Mill, things like health, virtue, or listening to music are desirable for their own sake not as a means to pleasure but only insofar as they are conceived of, in some way, as pleasant. This is what Mill means when he says that whatever “is desired otherwise than as a means to some end beyond itself, and ultimately to [pleasure], is desired as itself a part of [pleasure], and is not desired for itself until [it] has become so.”Footnote ⁷³

Given Mill's psychological hedonism, to say that something like virtue is desired in itself is to say that virtue is conceived of as pleasant in some way. For Mill, when this happens it is not that virtue is itself taken to be the ultimate good, but rather that in conceiving of it as desirable in itself (and thereby pleasant), it becomes part of one's conception of pleasure as a whole (or one's conception of the pleasant life as a whole). So, while we can in some respect refer to things like virtue and health as intrinsically desirable and thus intrinsically good, this desirability and goodness ultimately derives from their being thought of as pleasant. In this respect, pleasure should be understood as the only true intrinsic good, without which nothing else could be understood as desirable for its own sake. In the same way that Mill's classic hedonism – when he is read as an empiricist endorsing a motivational theory of pleasure – ought to be taken seriously by contemporary philosophers, so too should his hedonism be taken seriously when read as an empiricist psychological hedonist. While psychological hedonism has its detractors in contemporary philosophy of action, it also has its fair share of proponents.Footnote ^74, Footnote ⁷⁵ Mill's position is not any less viable than other contemporary psychological hedonists.