How Do We Differ When We Differ In Taste?

1. Introduction

Adrie likes to eat olives. Beiro, on the other hand, does not. Unlike Adrie, Beiro does not like the experiences he gets from eating olives. Cases like this are differences in taste, or taste differences. But what sorts of differences do such cases actually involve? In what way, or ways, do Adrie’s olive-experiences differ from Beiro’s?

I will defend the claim that Adrie and Beiro’s olive-experiences differ phenomenologically. In other words, “what it is like” for Adrie to eat olives is not “what it is like” for Beiro to eat olives. There is some sort of difference in how their overall experiences feel for them “from the inside.”¹ Call this the phenomenal thesis. My sense is that most people are inclined to reject the phenomenal thesis. Most people, I think, would claim that olives taste the same to both Adrie and Beiro. As we will see, this is not quite the same as claiming that Adrie and Beiro get the same kind of experience from eating olives. But it is easy to conflate the two claims. So I think that most people would reject the idea that Adrie and Beiro get different kinds of experience from eating olives.

However, this is not why I think that the phenomenal thesis is interesting, and it is not what motivates me to defend it. I defend it for two reasons. First, differences in taste are interesting in their own right, and the phenomenal thesis is interesting if true. Second, some philosophers assume that the phenomenal thesis is false, and advance important arguments which turn on this assumption. If the phenomenal thesis is true, then those arguments are inconclusive. This is why the phenomenal thesis has philosophical significance, whether or not most people reject it.

The paper proceeds as follows. In §2, I formulate the phenomenal thesis in more specific terms. I then highlight in §3 how the resulting thesis has been implicated in previous philosophical disputes. I focus in particular on a pair of arguments by Chris Heathwood and David Sobel, each of which turns on the falsity of the phenomenal thesis. In §4 I argue for two key claims: first, that there is a link between our attitudes and our pleasures; and second, that there is a link between our pleasures and our phenomenology. Together, my conclusions link our attitudes and phenomenology in a way that vindicates the phenomenal thesis. In §5 I explore a few ways in which Heathwood and Sobel might argue for a more modest claim about taste differences: the claim that there are possible taste differences which do not involve phenomenological differences. I argue that even these more modest arguments are inconclusive. I close in §6 by taking stock of my conclusions.

2. The Phenomenal Thesis

The phenomenal thesis is a theory about taste differences. “Taste difference,” as I use the term, covers all cases with the following structure:

• Taste Difference: A case in which, for some subjects $s_{\mathit{1}}$ and $s_{\mathit{2}}$, and some activity $A\text{:}$

  • $A$ causes $s_{\mathit{1}}$ to have a certain type of overall experience $E_{\mathit{1}}$;

  • $A$ causes $s_{\mathit{2}}$ to have a certain type of overall experience $E_{\mathit{2}}$;

  • $s_{\mathit{1}}$ is robustly disposed to intrinsically like experiences of type $E_{\mathit{1}}$;

  • $s_{\mathit{2}}$ is robustly disposed to intrinsically dislike experiences of type $E_{\mathit{2}}$.²

This definition captures the sorts of taste differences I have in mind: cases in which subjects’ attitudes towards their experiences play a prominent role in explaining their attitudes towards the activities which cause those experiences. The case of Adrie and Beiro is a case of this kind. Adrie intrinsically likes the experiences she gets from eating olives; Beiro intrinsically dislikes the experiences he gets from eating olives. This is why Adrie likes eating olives and Beiro dislikes eating olives. There are other cases which are sometimes called “differences in taste,” in which experiences do not play a prominent explanatory role. For example, we say that some people have a taste for collecting old photographs. Plausibly, this “taste” is not explained by their attitudes towards their experiences of photographs. I set such cases aside in what follows.

A few terms in my definition require clarification. First, my definition appeals to attitudes of intrinsic liking and disliking. These attitudes are not to be characterized decision-theoretically; they are not reducible to actual and counterfactual choices. Rather, they are psychological phenomena which underlie and explain (some of) our choices. Furthermore, our intrinsic likes and dislikes should be distinguished from instrumental or otherwise extrinsic forms of approval and disapproval. To see the difference, suppose I take some cold medicine because I have a sore throat. I like the medicine extrinsically—I like that it will alleviate my throat pain. But I do not intrinsically like taking medicine; I do not approve of it for its own sake. In contrast, Adrie intrinsically likes the experience she gets from eating olives; she approves of it for its own sake. Going forward, I will drop the “intrinsically” qualification. But whenever I refer to a subject’s likes and dislikes, I mean her intrinsic likes and dislikes.

My definition also appeals to types of overall experience. A subject’s overall experience is the totality of that subject’s experiences at a time. Types or kinds of experience are individuated by their phenomenologies; that is, they are individuated by “what it’s like” to experience them. To illustrate, consider Adrie and Beiro. Let $E_A$ be the kind of overall experience Adrie has when she eats an olive, and let $E_B$ be the kind of overall experience Beiro has when he eats an olive. $E_A$ and $E_B$ are different kinds of experience just in case there is a difference between “what it is like” to be Adrie as she eats her olive, and “what it is like” to be Beiro as he eats his olive.

This definition individuates kinds of experience in a highly fine-grained way. Any phenomenological differences make for a difference in kind. Supposing that Adrie’s vision is slightly blurrier than Beiro’s, it follows that $E_A$ is a different kind of experience than $E_B$. Part of “what it is like” to instantiate $E_A$ is to have a blurry visual field; but this is not part of “what it is like” to instantiate $E_B$. Intuitively, however, this difference is irrelevant to Adrie and Beiro’s difference in tastes. Adrie’s blurry vision has nothing in particular to do with the fact that she likes the overall experience she gets from eating olives. So I will make a simplifying assumption: I will assume that the case of Adrie and Beiro does not involve any such irrelevant differences. I assume that Adrie and Beiro are very similar people, eating very similar olives in very similar circumstances. Neither of them has blurry vision, or ringing ears, or a pain in their foot, or any other feature which would make for an irrelevant difference in their overall phenomenology. Nothing of philosophical significance turns on this assumption, but it does make the discussion go more smoothly.

Simplifying assumptions notwithstanding, the case of Adrie and Beiro is an ordinary sort of case. It is an ordinary difference in taste, involving ordinary people. Adrie is robustly disposed to like the kind of overall experience she gets from eating olives. Barring extraordinary circumstances, she likes experiences of that kind whenever she has them. Beiro is robustly disposed to dislike the kind of overall experience he gets from eating olives. Barring extraordinary circumstances, he dislikes experiences of that kind whenever he has them. Similarly, some ordinary people are robustly disposed to like the experiences they get from eating durians, taking very hot showers, and getting deep Swedish massages. Other ordinary people are robustly disposed to dislike the experiences they get from those activities. The phenomenal thesis is a claim about these and all other ordinary differences in taste:

• Phenomenal Thesis: All ordinary differences in taste involve phenomenological differences.

The phenomenal thesis says that, for example, $E_A$ and $E_B$ are different kinds of experience—Adrie’s overall experience differs phenomenologically from Beiro’s overall experience. Even if we abstract away from irrelevant differences having to do with blurry vision, ringing ears, and so on, we ought to believe that their experiences feel different.

Of course, we could formulate a stronger version of the phenomenal thesis, according to which all taste differences involve phenomenal differences. This strong phenomenal thesis would extend even to extraordinary cases, involving extraordinary subjects and extraordinary circumstances. For example, it would cover cases involving aliens, as well as humans subjected to neuroscientific tampering. But I will not defend the strong phenomenal thesis, for two reasons. First: nothing about these science fiction-style cases is intuitive, so it is hard to see how we could appeal to them to support a general claim about taste differences. Second: the weaker phenomenal thesis is philosophically significant in its own right. As I will show in the next section, nothing stronger is required for the philosophical significance of my arguments.

Before we move on, however, there is one more clarificatory point to be made. It is crucial that we distinguish between the following two claims:

• Different Phenomenology: Adrie and Beiro get experiences of different phenomenal kinds from eating olives.

• Tastes Different: Olives taste different to Adrie than they do to Beiro.

If Tastes Different is synonymous with Different Phenomenology, then the phenomenal theory entails that olives taste different to Adrie than they do to Beiro. In fact, however, the two claims are not synonymous, and the phenomenal theory does not entail that olives taste different to each of them.

To see why the claims are not synonymous, consider the following case:

• Tinge of Disgust: Adrie eats one hundred olives in a row. Although Adrie likes olives quite a bit, even she has her limits—eventually, as she works her way through the one hundred olives, she becomes a bit disgusted with the taste. Let E_A-1 be Adrie’s first olive-experience, and let $E_{A\text{-}\mathit{100}}$ be her hundredth olive-experience. E_A-100 is tinged with disgust; $E_{A\text{-}\mathit{1}}$ is not.

$E_{A\text{-}\mathit{1}}$ differs phenomenologically from $E_{A\text{-}\mathit{100}}$, but nothing forces us to say that the hundredth olive literally tastes different to Adrie than the first olive. Perhaps the only difference between $E_{A\text{-}\mathit{1}}$¸and $E_{A\text{-}\mathit{100}}$ is that $E_{A\text{-}\mathit{100}}$ involves a disgust reaction, and perhaps this difference is not properly described as a way in which the olives taste to Adrie. Nothing forces us to accept this interpretation; but nothing forces us to deny it, either. Certainly it will sometimes be the most natural way to describe what is going on. To illustrate: imagine that Adrie is attending a dinner party, and the host is being somewhat pushy with the hors d’oeuvres. He offers Adrie plate after plate of olives, and eventually he notices her grimace slightly as she eats. “What’s wrong?” he asks. “Is there something wrong with these olives? Do they taste different than the last batch?” Adrie might respond: “No, there’s nothing wrong with them! They taste just the same. The trouble is I’ve had too many.” This suggests that there can be phenomenological changes in the experiences that Adrie gets from eating olives, without changes in how olives taste to her.

Applying this general lesson to the case of Adrie and Beiro, we can accept that Adrie and Beiro get different kinds of experiences from eating olives, while denying that olives taste different to each of them.³ Tastes Different is not synonymous with Different Phenomenology.

A better interpretation of Tastes Different appeals to the contents of Adrie and Beiro’s experiences:

• Content Interpretation: Tastes Different means: Adrie and Beiro’s experiences differ with respect to the qualities their experiences represent olives as having.⁴

On the Content Interpretation, Tastes Different says that Adrie and Beiro detect different qualities of olives, and do so by tasting them. In other words, it says that they are tasting different tastes. To see why this is a better interpretation of Tastes Different, consider another case:

• Salty Taste: Beiro eats one hundred olives in a row. As Beiro works his way through the plate of one hundred olives, he contracts a rare disease which renders him incapable of detecting salt by taste. Let $E_{B\text{-}\mathit{1}}$ be Beiro’s first olive-experience, and let $E_{B\text{-}\mathit{100}}$ be his hundredth olive-experience. In experiencing $E_{B\text{-}\mathit{1}}$, Beiro detects-by-taste the saltiness of an olive. In experiencing $E_{B\text{-}\mathit{100}}$, Beiro does not detect-by-taste the saltiness of an olive.

Most of us would certainly say that Beiro’s first olive tastes different to him than his hundredth olive. The first olive tastes salty to Beiro; the hundredth olive does not. We can also put this point in terms of experiential contents: Beiro’s experiences change with respect to the qualities they represent olives as having—$E_{B\text{-}\mathit{1}}$ presents him with saltiness; $E_{B\text{-}\mathit{100}}$ does not. The experiences differ in qualitative content.⁵

The Content Interpretation is fine as a first pass interpretation of Tastes Different. Talk of “the way olives taste” has something importantly to do with the qualities that olives are represented as having. However, I suspect that the best interpretation of Tastes Different will appeal to both phenomenology and qualitative contents. For example:

• Phenomenology-in-Virtue-of-Content Interpretation: Tastes Different means: Adrie and Beiro’s overall experiences differ phenomenologically in virtue of differing with respect to the qualities those experiences represent olives as having.

This, I think, is a promising interpretation. It seems to capture the sense in which phenomenology is relevant to our evaluations of “how things taste,” while making room for the important role of contents. The upshot is that the best interpretations of Tastes Different will appeal to the contents of experiences, and not merely their phenomenology.

Of course, the phenomenal thesis says nothing about the contents of experiences. It says that $E_A$ and $E_B$ differ phenomenologically, but it does not say whether or not they differ in qualitative content. For all the phenomenal thesis says, Adrie and Beiro’s experiences may share some or all of their qualitative contents.⁶ So there is a very natural sense in which olives may taste the same to both Adrie and Beiro, even if the phenomenal thesis is true. It could be that Adrie and Beiro taste (in the sense of detecting-by-taste) the same tastes (construed as objective qualities of olives). As we will see, this is an important point. It is easy to conflate the topics of phenomenology and qualitative content, but this tends to obscure what is going on in discussions of differences in taste. We will need to keep the distinction close at hand.

Because it leaves open the possibility that $E_A$ and $E_B$ share qualitative contents, the phenomenal thesis is more plausible than it might appear at first glance. Indeed, one might worry that it’s so plausible as to be uninteresting. In my experience, people tend to react to the phenomenal thesis in one of two ways: either they think that it is false (because it says that differences in taste involve differences in phenomenology), or they think that it is uninteresting (because it does not say that different foods taste different to different people). I think that, on the contrary, the thesis is both interesting and true. To show that it is interesting, I will rehearse a pair of arguments which appeal to differences in taste. Both arguments turn on the rejection of the phenomenal thesis as I have formulated it. Then in §4 I will make the case for the phenomenal thesis.

3. Two Arguments from Taste Differences

Both Chris Heathwood and David Sobel advance arguments which appeal to differences in taste. They both assume that the phenomenal thesis is false—indeed, that it is obviously false. They assume that there are many cases in which subjects differ in their likes and dislikes towards a particular (phenomenal) kind of experience. This assumption plays an important role in each of Heathwood and Sobel’s arguments. However, the arguments differ importantly in subject matter: whereas Heathwood argues for a thesis in the philosophy of mind, Sobel argues for a thesis in value theory. I will consider Heathwood’s argument first.

Heathwood argues for externalism about pleasure: the thesis that pleasant experiences are pleasant in virtue of their extrinsic properties.⁷ In contrast, internalism about pleasure is the thesis that pleasant experiences are pleasant in virtue of their intrinsic properties—and usually the relevant intrinsic properties are assumed to be phenomenal properties.⁸ Heathwood’s preferred theory of pleasure is a paradigm form of externalism. His theory tells us that “a sensation S, occurring at time t, is a sensory pleasure at $\text{t}$ iff the subject of S desires, intrinsically and de re, at t, of S, that it be occurring at t” (2007b: 32). The upshot is that when I have a pleasurable experience of sipping coffee—for example—the experience is pleasurable for me in virtue of a certain extrinsic property—namely, its being desired in a certain way—rather than its intrinsic phenomenal character.

Heathwood defends externalism about pleasure, and argues against internalism about pleasure, by appealing to differences in taste. He tells us that:

The cases that most clearly support externalism involve sensations that some people like and others don’t (especially gustatory sensations), or sensations that bother some people but not others. The sound of fingernails scratching on a chalkboard is extremely unpleasant to many people, but not at all unpleasant to others. If unpleasantness is intrinsic to sensations, then one of these groups has to be mistaken. If this sound really is intrinsically unpleasant, then those whom it doesn’t bother and who therefore judge it to be not at all unpleasant, are wrong. That is hard to swallow. (2007a)

Internalists about pleasure have an obvious response to Heathwood’s worry. They can claim (and have claimed, see Smuts 2011) that those who differ in tastes have different types of experiences. Suppose that Beiro, but not Adrie, is bothered by the sound of fingernails scratching on a chalkboard. The internalist may claim that Adrie’s experiences differ phenomenologically from Beiro’s experiences, and that this intrinsic difference explains the difference in their experiences’ unpleasantness. Thus, both Adrie and Beiro are correct—Beiro’s experience really is unpleasant; Adrie’s experience really isn’t unpleasant. Neither Adrie nor Beiro is mistaken.

Heathwood addresses this line of thought indirectly. He considers changes in taste, which he takes to also support externalism. He tells us:

Flowers and perfume initially smell nice, but can begin to nauseate after a while. One way this may happen is that the sensation itself somehow transforms after prolonged exposure—you start getting a different smell. But surely another way this happens is that the smell stays the same while our feelings about it change. What we once liked, we now dislike. […] Internalists could respond by insisting that such cases always involve intrinsic changes in the sensation. But this seems like a desperate move, akin to the desperate strategies used to defend views like psychological egoism. (2007a)

Importantly, Heathwood is not merely claiming that some changes in taste do not involve phenomenological changes.⁹ Rather, he claims this is so obvious that it would be a “desperate move” to deny it. This gives some context to Heathwood’s earlier remarks regarding differences in taste. Clearly, he would have the same reaction to the internalist’s suggestion that all differences in taste involve differences in phenomenology. He would regard this, too, as a “desperate move.”

If Heathwood is right that differences in taste do not involve differences in phenomenology, then it would indeed appear that internalism about pleasure has a worrisome result—viz., that some people are thoroughly mistaken about the pleasantness of their experiences. Take Adrie and Beiro, for example. Adrie and Beiro differ in tastes. Thus, by my definition, Adrie is robustly disposed to like $E_A$, and Beiro is robustly disposed to dislike $E_B$. Presumably, then, Adrie will judge that $E_A$ is pleasant, and Beiro will judge that $E_B$ is unpleasant. If Heathwood is right about differences in taste, then $E_A$ and $E_B$ are the same kind of experience. So, if internalism about pleasure is true, then Adrie and Beiro’s experiences do not differ with respect to pleasantness. Either both experiences are pleasant—in which case Beiro is mistaken—or they are both unpleasant—in which case Adrie is mistaken.

Suppose that the experiences are pleasant, and Beiro is mistaken. It’s not that he is confused, distracted, or somehow “out of touch” with his own experiences. He is fully aware of what those experiences are like, and he says with great confidence that they are unpleasant. Nevertheless, he is mistaken—he is wrong about the pleasantness of his experiences. Heathwood regards this as a deeply implausible result, and I tend to agree.¹⁰

We can reconstruct Heathwood’s argument as follows:

• Heathwood’s Argument from Taste Differences:

  • (P1) There is a difference in taste which does not involve phenomenal differences. (The Existence Claim)

  • (P2) If P1, then if internalism is true, there is a difference in taste which does not involve a difference is pleasurableness.

  • (P3) If internalism is true, and if there is a difference in taste which does not involve a difference is pleasurableness, then at least one person is thoroughly mistaken in their judgments about the pleasurableness of their experiences.

• Conclusion: If internalism is true, then at least one person is thoroughly mistaken in their judgments about the pleasurableness of their experiences.

P1 says that there is at least one difference in taste which do not involve phenomenological differences. That is, there is at least one case with the following structure: for some kind of experience $E$, and some subjects $s_{\mathit{1}}$ and $s_{\mathit{2}}$, $s_{\mathit{1}}$ is robustly disposed to like her experiences of $E$, and $s_{\mathit{2}}$ is robustly disposed to dislike his experiences of $E$. Going forward, I will call this the Existence Claim. It seems clear that Heathwood believes that there are lots of cases with this structure. However, he only needs the weaker Existence Claim for this particular argument to go through.

Still, it might be tempting to interpret Heathwood as adopting a premise that is weaker than the Existence Claim. Returning to our distinction between phenomenology and content, we might be tempted to interpret Heathwood as making a point about the qualitative contents of our experiences. We might interpret him as saying that prolonged exposure to perfume does not cause us to detect different qualities of the perfume. This interpretation might be suggested by Heathwood’s remark that, on the view he rejects, “you start getting a different smell” (emphasis mine). But this cannot be the right way to interpret Heathwood. To make a case against internalism, Heathwood cannot be satisfied with claiming that there is no change in content. He needs to claim that there is no change in phenomenology.

To see this, suppose we endorse a rather naive theory of pleasure: we think that pleasure is a simple, contentless tingling sensation. With this theory of pleasure at hand, we give a simple treatment of Heathwood’s perfume case. We say that as one smells perfume continuously, one feels less and less of the pleasant tingles. This does not amount to a change in the content of one’s experiences, since the tingles have no content. If Heathwood were merely arguing that the perfume case does not involve a change in contents, then he would have to allow this treatment of the case. But he would not allow this treatment of the case. The naive theory of pleasure is exactly the sort of internalist theory which Heathwood is arguing against. So Heathwood cannot be satisfied with denying that cases like the perfume case involve differences in content. He has to deny that such cases involve phenomenal differences. He has to accept the Existence Claim.

Similarly, Heathwood cannot be satisfied with claiming that some part of one’s phenomenology remains constant in the perfume case. This interpretation is suggested by Heathwood’s claims that, on the view he accepts, the smell of perfume “begin[s] to nauseate after a while,” and “our feelings about it change.” Here, Heathwood seems to suggest that our overall experiences change over time. So it’s natural to read him as claiming that the core olfactory experience remains constant, not that the overall experience remains constant. But again, this cannot be the right way to interpret Heathwood, because this interpretation is consistent with internalism. The internalist can claim that our core olfactory experience remains constant, but our overall experience becomes unpleasant because we feel less and less of the pleasant tingles (or more and more of the unpleasant nausea).¹¹ Again, this is exactly the kind of naive internalist theory that Heathwood rejects. So Heathwood must deny that the case involves phenomenal differences— any phenomenal differences. This interpretation is mandatory to establish the Existence Claim, and nothing weaker than the Existence Claim will suffice for Heathwood’s purposes.

Strictly speaking, the Existence Claim is consistent with the phenomenal thesis. Heathwood could grant that all ordinary cases involve phenomenal differences, while also insisting that some extraordinary, science fiction style cases do not involve phenomenal differences. In this way, Heathwood could accept his Existence Claim while also making room for the phenomenal thesis. But Heathwood is not in a position to make this concession. Heathwood does not argue for the Existence Claim; rather, he regards it as an obvious piece of common sense. He regards it as obvious that cases like the perfume case do not involve phenomenological differences. This attitude towards the Existence Claim is deeply in tension the phenomenal thesis. If we accept that all ordinary taste differences involve phenomenal differences, we will not assume that some extraordinary cases do not involve such differences. So, if Heathwood were to concede the phenomenal thesis, he would need to argue that there are extraordinary cases of the kind he describes. I will consider some arguments of this kind in §5; for now, it’s enough to note that Heathwood himself does not provide any such arguments.

So much for the argument for externalism about pleasure. David Sobel gives a similar argument with the same structure—and he, too, assumes the falsity of the phenomenal thesis. Sobel appeals to taste differences in an argument against strong objectivism. This is the thesis that none of our reasons (or almost none of our reasons) are grounded in our desires. To illustrate: suppose I have a reason to finish writing this paper, and a desire to finish it. The strong objectivist will deny, contra subjectivism, that my reason to finish my paper is grounded in my desire to finish my paper (or in any of my other desires). The strong objectivist will instead offer some other grounds for my reason. They might claim (very charitably) that finishing my paper is objectively worthwhile, and this is what grounds my reason to finish it.

Sobel introduces taste differences as posing a prima facie problem for strong objectivism. He says:

An immediate worry one might have about strong objectivism is how such a view can account for the irresistible thought that some people have more reason to taste this chocolate ice cream rather than that strawberry ice cream and other people have more reason to taste the strawberry where such reasons stem from the taste of the desert (rather than, say, health concerns) and one’s reaction to it. (2005: 438)

To explain this “irresistible thought,” an objectivist could adopt the same strategy which I offered on behalf of internalists about pleasure. They can claim (and have claimed, see Scanlon 1998: 42) that the “irresistible thought” is explained by phenomenological differences. Supposing that Adrie but not Beiro likes to eat strawberry ice cream, the objectivist may claim that Adrie’s overall experience differs phenomenologically from Beiro’s overall experience, and that furthermore this phenomenological difference grounds a difference in pleasurableness. On the resulting picture, Adrie has more reason than Beiro to eat strawberry ice cream.

Sobel anticipates this line of thought, and his response is familiar. He tells us that:

It must be metaphysically possible, on this [phenomenological] conception of pleasure, that someone not like it. We would perhaps be similarly surprised if we learned that someone did not like the taste of chocolate or did like the taste of dirt—we would in the first instance reach for explanations that do not entail that they really do not like what we find so easy to like or that they really do like what we find so disgusting. But eventually, surely, there could be evidence that these surprising tastes are really theirs. We could, of course, always plead inverse qualia in cases like this, but that will often seem an unwarrantedly drastic understanding of what is going on. […] So let it be that we finally find someone who really does not like the flavor of sensation of pleasure. Should we think that this person is necessarily making some sort of mistake? Well what mistake would it be? I myself do not understand what sort of mistake could be thought to be necessarily involved in a failure to like this or that phenomenological state. (2005: 444–45)

Sobel clearly holds that some taste differences do not involve phenomenal differences. Indeed, he thinks that denying this would amount to “plead[ing] inverse qualia” and would “often seem like an unwarrantedly drastic understanding of what is going on.” This is why Sobel thinks that strong objectivism implausibly entails that some people are mistaken about their experiences.¹² Take Beiro, for example. He would judge that he has little or no reason to eat olives, since he does not like the experiences he gets from eating olives. But if those experiences are objectively pleasurable, then in fact he is mistaken—he has a fairly weighty reason to eat olives.

Putting these points together, we can reconstruct a familiar argument:

Sobel’s Argument from Taste Differences:

• (P4) There is a difference in taste which does not involve phenomenal differences. (The Existence Claim)

• (P5) If P4, then if strong objectivism is true, some people are thoroughly mistaken in their judgments about their reasons for having certain experiences.

• Conclusion: If strong objectivism is true, some people are thoroughly mistaken in their judgments about their reasons for having certain experiences

To reiterate, the Existence Claim is the claim that there is at least one case with the following structure: for some kind of experience $E$, and some subjects $s_{\mathit{1}}$ and $s_{\mathit{2}}$, $s_{\mathit{1}}$ is robustly disposed to like her experiences of $E$, and $s_{\mathit{2}}$ is robustly disposed to dislike his experiences of $E$. With respect to this Existence Claim, Sobel is in essentially the same dialectical situation as Heathwood. Like Heathwood, he cannot settle for a weaker claim. Also like Heathwood, he does not argue for it—rather, he assumes that it is true. He cannot maintain this attitude while at the same time conceding the phenomenal thesis. For if Sobel were to grant the phenomenal thesis, he would need to argue for the Existence Claim. And he does not give any such arguments.¹³

In the remainder of this paper, I will make the case for the phenomenal thesis. If the phenomenal thesis is true, then that would constitute an interesting result in its own right. But it would also be dialectically significant, in that it would send the ball back into Sobel and Heathwood’s court. For if the phenomenal thesis is true, then the onus is on them to show that there are taste differences of the kind they describe.

4. From Attitudes to Phenomenology

According to my definition, differences in taste are differences in attitudes. Subject $s_{\mathit{1}}$ is robustly disposed to like the experiences she gets from a certain activity; subject $s_{\mathit{2}}$ is robustly disposed to dislike the experiences he gets from that same activity. According to the phenomenal thesis, ordinary differences in taste involve phenomenal differences—those who differ in taste are getting experiences of different (phenomenal) kinds. So, according to the phenomenal thesis, there is a link between our attitudes and our phenomenology. I will argue for this link in two stages. First, I will argue that there is a link between our attitudes and our pleasures. Second, I will argue that there is a link between our pleasures and our phenomenology. Together, these claims link our attitudes to our phenomenology in a way that vindicates the phenomenal thesis.

5. The Existence Claim Revisited

In this section, I will consider a few arguments for the Existence Claim. To reiterate, this is the claim that there exists at least one case with the following structure: for some (phenomenal) kind of experience $E$, and some subjects $s_{\mathit{1}}$ and $s_{\mathit{2}}$, $s_{\mathit{1}}$ is robustly disposed to like her experiences of $E$, and $s_{\mathit{2}}$ is robustly disposed to dislike his experiences of $E$. Strictly speaking, such arguments are irrelevant to the truth of the phenomenal thesis. We can accept the phenomenal thesis, while granting that some extraordinary taste differences do not involve phenomenal differences. But it is nevertheless instructive to consider the arguments. If it’s hard to find conclusive arguments for the Existence Claim, then this should make us all the more confident in the phenomenal thesis.

6. Conclusion

It is striking that, in recent years, much has been written about taste disputes—the sorts of things people say when they differ in taste. See for example Egan (2010), Sundell (2011), Palmira (2015), Capraru (2016), Ferrari (2016), Zeman (2016), Wyatt (2018), and Beddor (2019). But much less attention has been paid to taste differences themselves—that is, the differences that obtain between people who differ in tastes. This, I hope to have shown, is an important oversight. There is much to be learned from thinking about how we differ when we differ in tastes.

It can be natural to suppose that olives taste pretty much the same to everyone, and this can make it natural to think that we all get the same kind of experience from eating olives. But however natural this thought may be, it ought to be rejected. We ought to think that those who love olives have different experiences than those who hate olives—and we ought to reject arguments that assume otherwise.