“Why representationalists can’t be desire theorists (and vice versa)" - Matthew Kinakin (University of Edinburgh)
Synthese, 2026
Let’s say that you, like me, enjoy ice cream. Let’s say, further, that on some particular occasion, you are having a bowl of vanilla ice cream, sprinkled with pistachios and a dollop of raspberry jam, topped off with a bit of honey, sea salt and olive oil. Let’s assume that you, as I do, find the dessert delicious. Note some things about this case. When you take bites of the dessert, you surely undergo a variety of sensory experiences: the sweetness of the jam and honey, the crunch of the pistachio, the aroma of vanilla, and the silky texture of the olive oil. A true whirlwind of sensory experience. If we’re realists about phenomenal character—i.e. we take there to be “something-it-is-like” to undergo those sensory experiences—then if anything has phenomenal character, it’s surely the many sensory experiences we undergo when we devour that dessert.
But notice, also, that not only do these sensory experiences have some corresponding phenomenal character—they have a qualitative feel to them—but that you and I find these sensory experiences pleasant. That is, on top of the sensory phenomenology of our experiences, there is an affective dimension as well, i.e. a quality of the experience we find pleasant. Moreover, not only are these sensory experiences pleasant, but they are also presumably motivational and normative: once that first bite hits your taste buds, good luck stopping; and why stop when, presumably, you’re getting something that is, in some sense, good-for-you, namely: pleasure!
All these features call out for explanation. We have
(i) Sensory phenomenology (the sweetness of the honey, the texture of the oil)
(ii) Affect (the positive or negative feel of the sensory experience)
(iii) Affective motivation (the motivation toward or away from the affect itself)
(iv) Affective value (the prudential (dis)value of the affective quality itself)
In philosophy, there are two highly popular, naturalistic theories used to explain a variety of these features. But here’s the problem: they are incompatible. For instance, if you buy into the leading theory of phenomenal consciousness, namely representationalism, then you cannot say that affect, its motivation, and value consists in a desire targeting the sensory phenomenology. Similarly, if you think that affect itself is a phenomenal matter, then you won’t be able to say that affective motivation and affective value consists in a desire targeting the affective phenomenology.
To see all this, consider the following.
A very popular view about the nature of phenomenology claims that phenomenology (whether sensory or affectual) consists entirely in what an experience is about, i.e. in what the experience’s representational content is. Representationalism about phenomenal character is a reductive, physicalist view. Hence, its appeal to those with naturalistic commitments.
Setting aside sensory phenomenology, another popular view sets out to capture affective motivation and affective value (and, in some contexts, even affect itself). That popular view goes by the name of desire-satisfactionism or desire-fulfilment theory since it claims, in its most popular form, that for a sensory experience to be pleasant, motivational, and good-for-you, the sensory experience itself—i.e. its phenomenal properties—must be the object of a satisfied or fulfilled desire. Arguably, the most popular view of the affective qualities of sensory experience, a version of which is propounded by Chris Heathwood (2007), says just that: your sensory experience feels good (and you want more of it, and it is good-for-you) because you are getting something you want. Moreover, some (Cutter and Tye 2014; Bain 2017) take affect itself to be a phenomenal matter, and although they eschew desires when it comes to affect, they do invoke them when it comes to the motivational and normative properties of affect. And note, importantly, for these sorts of desire theorists, the type of motivation we are explaining isn’t any sort of world-directed motivation, e.g. motivation directed towards one’s body, but is rather experience-directed. Now, there are important details we can set aside for the time being, but we can note the two general kinds of desire theorist that are within the scope of the argument: (i) sensation-directed desire theorists, and (ii) affect-directed desire theorists.
Here’s the point I argue for in Why representationalists can’t be desire theorists (and vice versa).
Representationalism is a very popular view in philosophy of mind of experiential phenomenology. Desire-based theories of affect, its motivation, or normativity are very popular views in value theory and metaethics. Interestingly, both are naturalistically respectable views. In fact, both attempt to provide naturalistic reductions of two things that have resisted such reductions, namely phenomenal consciousness and value. But the two views are incompatible. One cannot be a representationalist about phenomenal character while being a desire theorist about affect, its motivationality, or normativity (and vice versa). Moreover, as stated above, some will even be inclined to say that affect itself is a phenomenal matter, and hence, will be representationalists about affective phenomenology too. Even so, those representationalists (of which there are a few) still can’t be desire theorists about the motivationality or normativity of affect. That’s not only surprising and interesting, but troubling for those theorists who have endorsed both. So, how does the argument go?
The argument is pretty straightforward, and there are details I will omit here (check out the paper for the nitty-gritty details). But here’s the gist of it. Representationalism about phenomenal character entails the following view about the nature of phenomenal character. When we try to attend to the phenomenal properties of experience, our attention finds nothing but what the experience represents, i.e. the experience’s representational content. In other words, our attention only gloms on to the putative represented objects and properties of experience. Try it. Focus on something within your perceptual reach, say, the sweetness of the honey in that last bite of dessert. Now try to attend to the experience. What do you notice? Surely, your attention only focuses harder on what your experience is about, in this case the sweetness of the honey in your mouth. If representationalism is true, then attending directly to the experience itself should be impossible. If it weren’t, then it would be possible to separate phenomenal properties from representational content, something that would count directly against the strong reduction of phenomenal character to representational content. So, representationalism entails that experiences are transparent, and the introspective data bear this out. Things seem, according to the representationalist, as they should be.
Another way of putting the point is as follows. Sensory phenomenology, or any phenomenology for that matter, cannot be directly attended to. That is, experiential phenomenology cannot be the direct object of inner attention. But, as I argue, desire theories of sensory affect, its motivationality, or its normativity entail, for their plausibility, that subjects be directly aware of their phenomenal episodes. The key point being: why your sensory experience feels good, motivates you, and is good-for-you is because you are experiencing (i.e. are directly aware of) the very thing you have an occurrent, de re desire for, namely: the sensory experience itself. But that intimate picture between second-order desire, awareness, and first-order experience is precisely what is ruled out by representationalism’s commitment to the transparency of experience. So, sensation-directed desire theorists obviously can’t think of sensory phenomenology in terms of representational content. And affect-directed desire theorists can’t think of affective phenomenology in terms of representational content either.
There are, of course, more details, and the picture of how I get to that conclusion is slightly more complicated than I’ve outlined here. For starters, there are other ways to be aware that one’s desire is satisfied, ways that don’t entail direct awareness. For instance, as I consider in the paper, you might believe that your desire is satisfied, and the desire might still plausibly be de re. So, on one plausible view, you can “get a thing into a thought”, so to speak, without being directly aware of it; and then you can become aware of your desire’s satisfaction in the same indirect way the object got into the thought. But saying that affect, its motivationality, and normativity consists in a subject indirectly believing as against directly experiencing a satisfied desire stretches credulity, or so I argue. But the representationalist, with their commitment to transparency, seems committed to this implausible picture where, for example, affect (or its motivationality or its normativity) is constituted by believing your desire is satisfied. That, I argue, is implausible. Hence, why representationalist can’t be desire theorists (and vice versa).
So, why care about this? For starters, both views are highly popular, and on some occasions, philosophers have explicitly endorsed both. Hence, those affect-directed desire theorists can’t think in terms of representational content. But they do! Hence, they need to jettison one of those views. Furthermore, I wager that some desire theorists of affect—i.e. some sensation-directed desire theorists—think in representationalist terms when it comes to the sensory phenomenology their desires target. If that’s true, they need to be aware of this incompatibility and look elsewhere when it comes to their preferred theory of phenomenal consciousness. And you might say the same thing about representationalists: they must look elsewhere if they want to explain something like the motivationality and normativity of affect. But eschewing desire-based explanations of those two features of affect spells trouble for representationalism. There’s really nowhere else plausible to go, or so I argue in a different paper… summarized below…
Paper #2: Why intentionalists can’t take painkillers
Another paper of mine, Why Intentionalists Can’t Take Painkillers (Philosophical Studies 2026), continues this line of thought and argues for why the above-discussed incompatibility is detrimental for the representationalist (i.e. intentionalist). Above, we ruled out that intentionalists can appeal to desires to explain something like why we are motivated to continue or stop an experience. Question: Can intentionalists appeal to something other than a desire to explain how subjects are motivated toward (or away from) their own experiences? Short answer: Not really… In the following paper, I use painfulness to argue for this conclusion. Hence, why, if you’re an intentionalist, then you can’t explain why we take painkillers.
Transparency. The trouble for intentionalism, as mentioned above for paper #1, begins and ends with their commitment to the transparency of experience, what I will call Transparency. In short, Transparency entails that subjects cannot directly attend to the experience itself and any of the properties thereof (phenomenal or otherwise). I basically run through a variety of motivational strategies and argue that they are either incompatible with Transparency, or they are highly implausible.
Motivational view 1. So, why take painkillers?
First-order view 1: Attitude
Subjects are motivated to eliminate their painfulness in virtue of the internal attitude of the painfulness itself.
This idea is ruled out simply in virtue of the fact that it violates the reductive ambitions of the intentionalist program, namely to reduce all phenomenal character to intentional content. Appealing to the attitude or psychological mode of the mental state rather than just the content violates those reductive ambitions.
Motivational view 2. So, why take painkillers?
First-order view 2: Reflexive content
Subjects are motivated to eliminate their painfulness in virtue of the internal reflexive evaluative content of the painfulness itself (that is, content that evaluates itself as bad).
One might be inclined to think that it’s the painfulness itself that motivates its own elimination. And in line with intentionalism, the idea would be that the intentional content constitutive of painfulness somehow motivates the subject to eliminate the painfulness. The idea here appeals to reflexive contents, i.e. contents that are about themselves. Although some take that idea seriously, it won’t do for the intentionalist here since reflexive contents clearly violate Transparency. In order for a content to be about itself, it would first need to not be transparent. For if it were transparent, then the content couldn’t represent itself; rather, the content would see through itself (because it’s transparent). Hence, reflexive contents violate Transparency.
Motivational view 3. So, why take painkillers?
Second-order view 1: Second-order evaluative experience
Subjects are motivated to eliminate their painfulness in virtue of a second-order evaluative experience of the painfulness as bad.
This view is also clearly ruled out given it also violates Transparency. For this view says that a subject is motivated to eliminate their first-order experience because they have a second or higher-order experience directed at the first-order experience. But according to Transparency, first-order experiences cannot be the intentional objects of second-order experiences.
Motivational view 4. So, why take painkillers?
Second-order view 2: Second-order desire
Subjects are motivated to eliminate their painfulness in virtue of a subjectively frustrated external second-order desire for the painfulness not to occur.
This is perhaps the most plausible view, but we’ve already ruled this out above.
Motivational view 5. So, why take painkillers?
Second-order view 3: Second-order evaluative judgment
Subjects are motivated to eliminate their painfulness in virtue of a second-order evaluative judgment that the painfulness is bad.
This view notes that when subjects respond to their pains by taking a painkiller, they are often acting rationally. Hence, the suggestion that the very badness itself of painfulness plays a motivating role vis-à-vis painkiller-taking behaviour. Hence, why the judgment contains evaluative content. Note that since we’ve ruled out both a second-order experience and a second-order desire, the relevant attitude must be a belief, and the content of the judgment must be truth-apt. Hence, I argue, this entails truth-apt motivational internalism, the idea that evaluative judgments—all on their own—intrinsically motivate.
But independent of that, what I argue is that intentionalists can’t make epistemic sense of this judgment. For Transparency commits them to a view of introspection according to which subjects introspect their mental states (phenomenal character included) on the basis of what their experience represents, or what it is about. I won’t bog us down here with the details, but what I argue is that there are two general ways intentionalists could make epistemic sense out of the relevant motivating judgment above. The first is to say that there is some a posteriori justification for the belief. I argue that this won’t work. The second way is to argue that there is some a priori justification for the belief. I argue that this won’t work too for various reasons. I end the paper by considering some potential ways intentionalists could respond. One way is to invoke a kind of intuitionism about evaluative knowledge, something I suggest might be at odds with intentionalism’s naturalistic commitments. Another suggestion is for intentionalists to embrace an error theory about pain’s badness, namely that we are simply in widespread error when we evaluate our pains as bad. That also won’t work since it fails to explain why it is rational to take a painkiller, something intentionalists set out to do.
The upshot is that if we want to account for the intuitive and common-sense idea that subjects exhibit actions directed at their experiences, then we need to allow for the possibility that experiences (and any phenomenal properties thereof) can be the objects of direct, inner attention. Because intentionalism precludes this, it has a very difficult time accommodating that intuitive idea.
References
Bain, D. (2017). Why Take Pain Killers? NOUS, 53:2, 462–490.
Cutter, B. and M. Tye. (2014). Pains and Reasons: Why it Is Rational to Kill the Messenger.
Philosophical Quarterly 64(256): 423–433.
Heathwood, C. (2007). The reduction of sensory pleasure to desire. Philosophical Studies 133:
23–44