How should we explain ‘what it is like’ to perceive colour? One of the reasons why naïve realist theories of colour are interesting is that they promise to contribute towards a solution to the problem of consciousness.
There is something puzzling about the way that problems about consciousness and conscious experience are normally presented. Take Jackson’s Knowledge Argument as an example. In Jackson’s example, Mary is a scientist who knows all the physical facts about colour and colour perception, but she has never herself seen a chromatically coloured object – she has spent all her life in a black and white room. When she sees, say, a red apple for the first time, Jackson argues that she will learn something new, namely ‘what it is like’ to perceive colour. He concludes that there are facts that are not physical facts.
There is a lot that has and can be said about the Knowledge Argument, but here I want to focus on the question of what is the Knowledge Argument an argument for? It is widely assumed – both by those who accept the Knowledge Argument and those who don’t – that if Mary learns something when she sees a red apple for the first time, then she learns something about herself. Specifically, the argument is standardly taken to be an argument for the existence of qualia: intrinsic, qualitiative properties of experience.
But why, when Mary first sees an object of a certain kind, such as a red apple, should she be thought to learn something about herself? Why doesn’t she learn something about the object that she sees—namely, what redness, a property of the object, is like? (Or at any rate, if she does learn something about herself, why isn’t what she learns about herself parasitic on what she learns about the object?)
Here is one explanation of why the Knowledge Argument is normally assumed to be an argument for the existence of qualia, not colours: it is assumed in advance that there are no properties of physical objects that could explain ‘what it is like’ to perceive colour. And although there are different reasons for believing this, a common reason is that physics ‘tells us’ that colours as we perceive them do not exist. (This, at any rate, is something that Jackson himself argued for in earlier work defending a sense-datum theory of perception.) Assuming that Mary cannot learn anything new about physical objects when she first sees a red apple, then she must—if she learns anything—learn something about herself.
If this diagnosis is correct, then a central argument against physicalism itself relies on physicalist assumptions. This is not strictly inconsistent: it is possible to think that physics fully describes material objects but not conscious subjects. But there is at least a tension here. If science ‘tells us’ that there are no colours, then why doesn’t it ‘tell us’ that there are no qualia, either? After all, qualia play no role in scientific explanations. And if qualia can play no role in scientific explanations and still exist, then why should it follow from the fact that scientists do not appeal to colours to explain colour experiences that colours do not exist?
It is not just that there are arguments for the existence of colours that parallel standard arguments in the philosophy of mind for mind-brain distinctness—such as the Knowledge Argument, but also Kripke’s Modal Argument, the explanatory gap, and the hard problem of consciousness. Once these problem are internalised, and made into problems about conscious subjects, they can come to seem increasingly intractable. As it is normally understood, the Knowledge Argument is supposed to establish the existence of intrinsic qualitative properties of experience. But if we reflect on our experience, we seemingly aren’t aware of any properties of this kind. Experience is ‘transparent’: we ‘see through’ the experience to the objects in the mind-independent world that the experience is an experience of. So it is not just that there is no scientific evidence for the existence of qualia; there is seemingly no introspective evidence, either.
The naïve realist theory of colour provides an alternative way approaching problems associated with consciousness and conscious experience. According to the naïve realist, colours are mind-independent properties of objects that are distinct from their physical properties. The naïve realist therefore rejects one of the assumptions that is liable to make consciousness seem particularly problematic: that colours as we perceive them do not exist. If the naïve realist theory of colour is, in turn, combined with a naïve realist theory of perception, according to which ‘what it is like’ to have an experience is determined by the mind-independent objects and properties that we are related to in perception, then this provides an explanation of the qualitative character of colour experience. ‘What it is like’ to perceive colour is explained by the nature of the colours that we perceive.
Interesting thoughts. I’m sympathetic with the idea that a naive realist theory of color might go some way toward explaining the puzzling “qualitative character” of color experience. But I don’t know if it can take us very far all by itself. It seems to me that if there’s an explanatory gap between the physical facts and the facts about consciousness, then there would still be an explanatory gap between physical facts + irreducible-colors-in-the-world on the one hand, and the facts about consciousness on the other (even if we focus only on color experience). One reason for this is what Chalmers, in “Panpsychism and Panprotopsychism,” calls the “quality-awareness gap.” Let P* be a conjunction of all the physical facts + all the facts about the distribution of primitive colors in the world, including facts about the causal and other natural relations that may hold between those colors and states of my brain, etc.). Let Q be the fact that I am now phenomenally aware of redness. It seems like “P* & ~Q” is conceivable. For example, it seems conceivable that there could be a zombie who was causally related to instances of primitive redness in just the way that I am, or that someone could be causally related in this way to instances of primitive redness while having an experience phenomenally just like my experiences of green.
I am no qualia freak — but I suppose we’d want to distinguish the “What is it like?” question from “Why is it like anything at all?” Naive realism would help more directly with the former than the latter. Right?
Hm.. I’m not sure. As suggested above, it seems to me that P* (the conjunction of all the physical facts about me + the distribution of primitive colors in the world) is a priori consistent with both the zombie hypothesis and inverted-phenomenology hypothesis. Given that P* is a priori consistent with the inverted-phenomenology hypothesis, it seems like there’s going to be a problem explaining *what it’s like* even granting that there’s *something* it’s like. In other words, if I’m right, there would be an explanatory gap between “P* & there’s something it’s like for me” on the one hand, and the facts about *what* specifically it’s like for me, on the other.
That’s fine, but still it would take a further argument than the standard zombie one to challenge the explanatory role that Keith is proposing for NRism. And the inverted spectrum hypothesis is — on empirical grounds — pretty problematic, I believe.
Thanks Brian and John. I think the initial question puts very nicely the problem with appealing only to a naive realist theory of colour. I think this is why we also need to appeal to a naive realist theory of perception, according to which the phenomenal character of an experience is determined by what the experience is an experience of. The hope is that understanding experience relationally in this way is what is going to guarantee that fixing the colours fixes the characters of colour experiences.
Hi Keith, sounds like you have written an excellent book, I look forward to reading it! And I think you are right that regaining naive realism about color (and other ‘secondary properties’) is a crucial step towards solving – or rather dissolving – the mind/body-problem. But your claim that “‘What it is like’ to perceive colour is explained by the nature of the colours that we perceive” can still be interpreted in different ways, and I would like to ask you to elucidate it a bit more. On the most radically relationist and externalist construal, favored e.g. by John Campbell, the color properties literally constitute the phenomenal properties. Perhaps that is what you have in mind, but I hope not, because this seems to eliminate internal experiential states (or render them propertyless, which to my mind amounts to the same thing) just like the traditional outlook eliminates colors in the external world. I think that we need to acknowledge both, that we need to be naive realists about mind and world, about color and its experience, not just about one of them. So here’s a suggestion how we can. Let’s identify the phenomenal and the intentional properties of color experience. The what it is like of color experience is the what it is like of experientially representing colors. The second step is to say that experience cannot be understood independently of what it is experience of. (More could be said about that, but I’ll leave it at this for now.) So then the nature of color explains the nature of color experience without identifying the two.
Hi Michael, thanks for the comment. The ideas here owe a lot to John Campbell’s work, and I am tempted towards the more radically austere account you find there. For the purposes of presenting a solution to the problem of consciousness along roughly these lines, though, you are right that you don’t need to go this far — the key is to accept that perceptual experience is essentially relational, and you can do this within a representationalist framework. In the book I work within the framework of a naive realist theory of perception, but whether we should accept this (rather than a representationalist alternative of the kind you outline) is something that I’d like to consider in more detail in future. I like the way that you put the challenge for the autere relationalist.
Thanks very much Keith for your thoughtful posts and answers to the questions including my own. The series was timely for me as I have just been writing a page on the nature of colour for a website on colour for painters, though without reading your book I remain uncertain to what extent you feel naive realism/primitivism can be reconciled with current science and to what extent it is proudly independent, so I hope I’ve not been unfair to your position:
https://www.huevaluechroma.com/012.php