SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

Revelation is the thesis that having an experience that instantiates some phenomenal property puts us in a position to know the nature or essence of that property. It is widely held that although Revelation is prima facie plausible, it is inconsistent with physicalism, and, in particular, with the claim that phenomenal properties are physical properties. I outline the standard argument for the incompatibility of Revelation and physicalism and compare it with the Knowledge Argument. By doing so, I hope to show that on various plausible interpretations of Revelation it is in fact consistent with physicalism. Moreover, there is a robust reading of Revelation that a posteriori physicalists can, and should, accept.


1. Thomas, Son of Mary

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

There's not much we don't know about poor Mary who, from birth, was held captive by evil professors and denied the opportunity of seeing colour while also, ironically, being forced to become the world's greatest colour scientist (Jackson 1982; 1986). But one fact that has long been kept quiet is that, shortly after gaining her freedom, in a moment of passion fuelled in equal measure by post-release euphoria and the Stockholm syndrome, she conceived a son. Nine months later, Thomas was born into ‘one great blooming, buzzing confusion’ of colours, sounds and tastes. Despite her own deprived upbringing, Mary was a fine mother. Yet, in about his fourth year, Mary noticed that something was missing from Thomas's life. For no reason in particular, Thomas had never eaten a peach. When she first noticed this, a line by Pablo Neruda came to her mind: “a man who has never tasted peaches . . . would quietly become sadder, noticeably paler, and probably, little by little, he would lose his hair.” Having now learnt everything there is to know about colours, Mary decided to find out if Neruda was right. From that day on she did everything she could to keep Thomas from tasting peaches.

Mary's experiment was, of course, impractical to the point of farce. And so despite her best efforts, a rebellious Thomas eventually managed to get his hands on a peach at school, and, as was inevitable, tasted the forbidden fruit for the first time. The taste was a revelation. He did not quite see the face of God, but what was revealed to Thomas that lunchtime, behind the gymnasium, was remarkable nevertheless. What was revealed was the full nature of the taste of peaches. Focusing on the experience that had been denied to him all his life, the essence of the phenomenal property, i.e. the taste of peaches, was laid open to view.

When Thomas put peaches in his mouth for the first time he also set off a cascade of chemical reactions in his mouth. His taste buds sampled some of the resulting chemicals, and various nerves became activated and transferred complex electrical signals to his brain. In his brain even more complicated neuronal activity took place and all sorts of chemical and physical properties came to be instantiated. One of these instantiated physical properties was also a phenomenal property: it was none other than the taste of peaches.

Most philosophers who have thought about these things think that the story I've just told is incoherent.1 Call the thesis that the nature of phenomenal qualities are revealed to us when we have experiences that possess them Revelation. The dominant view in the philosophy of mind is that Revelation is inconsistent with physicalism and obviously so. Some have even gone so far as to say that the only view consistent with Revelation is a thoroughgoing primitivism about phenomenal qualities (Stoljar 2009, 124).

I find this consensus odd. For one thing, it is quite unclear how Revelation should be understood: in fact, there are several plausible interpretations of it. For another, an argument is needed to get from Revelation to either primitivism or the denial of physicalism, and the required argument must have something obvious in common with highly controversial anti-physicalist arguments: it must argue from an epistemic gap to an ontological one. One would expect, then, that an argument from Revelation should also be controversial, even for those sympathetic to Revelation. Finally, Revelation is widely regarded, at least prima facie, as an attractive thesis. David Lewis thought that Revelation was an important part of our folk-psychological conception of phenomenal properties.2 And for the many who have felt the temptation towards a revelation thesis about colour, the temptation towards a revelation thesis about colour experience is surely at least as strong.3 Thus one would think that Revelation should be given up only with some reluctance and only after careful consideration of the arguments implying it conflicts with our more cherished commitments. But in fact discussions of these arguments are rare.

In what follows I consider the arguments. To do so, I distinguish several readings of Revelation and, to help map the options available to physicalists, I use the more familiar Knowledge Argument as my guide. As you'd expect, it turns out that whether Revelation is inconsistent with physicalism depends on how we interpret the crucial notion of ‘knowing the nature of a property’. And while determining which interpretations are amenable to physicalists is important, there are general lessons too. For example, on most readings of Revelation I discuss, familiar objections to the Knowledge Argument can be adapted to block the resulting anti-physicalist arguments. And there is at least one robust reading of Revelation that a posteriori physicalists can, and so should, accept.

2. The thesis of Revelation

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

Revelation is the thesis that phenomenal properties are fully revealed to us when we have experiences that instantiate them: experience gives us access to the essence of phenomenal properties. But what is it to have access to essence, or to know the full nature of a property? While there isn't a simple answer to that question, we can give a few glosses of the idea that will help to show why Revelation is tempting. For example, it is sometimes said that to know the nature of a property is, in some sense, to have complete knowledge of the property itself, or to know what makes the property the property that it is, what it takes to be that property. Or, it is to know what the property is in “an uncommonly demanding and literal sense of ‘knowing what’ ” (Lewis 1995, 327). Applied to Revelation, we get the idea that we can know what it takes to be the taste of peaches simply by experiencing that taste. Focusing on the taste, this has the ring of plausibility: my understanding of that taste seems deeper than my understanding of almost all other (nonphenomenal) properties in just the way described.

Still, these glosses of what it is to know the nature of a property leave us plenty of interpretative options. One option is so implausible that we should discount it immediately. We shouldn't read Revelation as saying that we are all like Thomas in that the nature of our conscious experiences are in fact known to us. After all, we might have tasted peaches only while being so distracted that we paid the experience little attention. Instead, we should formulate Revelation as follows:

Revelation: When we have an experience with phenomenal property Q, we are in a position to know the full nature of Q.4

While the concept of ‘being in a position to know’ is relatively intuitive, it could do with some filling out. I suggest we follow Williamson.

To be in a position to know p it is neither necessary to know p, nor sufficient to be physically or psychologically capable of knowing p. No obstacle must block one's path to knowing p. If one is in a position to know p, and one has done what one is in a position to do to decide whether p is true, then one does know p. The fact is open to one's view, unhidden, even if one does not yet see it. (Williamson 2000, 95)

In our case, what is open to view is the nature of the taste of peaches and all that must be done to know that nature is for us to reflect on our experiences.

3. The argument from Revelation and the Knowledge Argument

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

The dominant view is that Revelation is incompatible with physicalism. For ease of discussion, for the most part I'll treat physicalism as equivalent to a version of the identity theory.

Identity: For every actually instantiated phenomenal property Q, there is a physical property P such that P = Q.5

While developed arguments for the incompatibility of Revelation and Identity are rare, Lewis's discussion provides a good starting point. In what follows, Lewis speaks of experiences ‘identifying’ qualia in a demanding way. But it is clear that to ‘identify’ qualia in this way – to know exactly what qualia are – is to have propositional knowledge of their nature, just, as he explicitly says, knowing exactly what potassium is requires knowing its atomic number.

If, for instance, Q is essentially the physical property of being an event of C-firing, and if I identify the qualia of my experience in the appropriate ‘demanding and literal’ sense, I come to know that what is going on in me is an event of C-firing. Contrapositively: if I identify the quale of my experience in the appropriate sense, and yet know nothing of the firing of my neurons, then the quale of my experience cannot have been essentially the property of being an event of C-firing . . . If qualia are physical properties of experiences, and experiences in turn are physical events, then it is certain that we seldom, if ever, [know the nature of] the qualia of our experiences. Making discoveries in neurophysiology is not so easy! (Lewis 1995, 328)

Lewis's idea appears to be that if we did know the nature of qualia, and qualia are physical properties, then we would know that qualia were identical to those physical properties.6

A little later, in a footnote, Lewis offers a further remark:

If we know exactly what the qualia of our experiences are, they can have no essential hidden structure – no ‘grain’– of which we remain ignorant. (If we didn't know whether their ‘grain’ ran this way or that, we wouldn't know exactly what they were. Whatever we might know about them, we would not fully know their essence.) (Lewis 1995, 329, n.4)7

Here the idea seems to be that if Identity and Revelation were both true, we should be able to ‘see through’ the qualia to their physical essence: their physical nature should be ‘laid bare’. That is, merely by having the experience of the taste of peaches, Thomas should be in a position to know the full, complicated story about the nature of the neurophysiological property with which it is identical.

Lewis's discussion points to two different arguments, both of which can be found in various forms in the literature: the first is that if Revelation and Identity were both true, then, contra the facts, we would know some statement identifying the taste of peaches with some physical property; the second is that if Revelation and Identity were both true, then, contra the facts, we would know the intrinsic nature of some physical property merely by having the experience of the taste of peaches. At first glance, both these arguments can seem compelling. If we really knew the nature of some quale and that quale has a physical nature, then we should know that nature.

In what follows, it will be helpful to spell out the form that these two styles of argument have in common. Adapting Daniel Stoljar's name, I'll call it the Argument from Revelation.

1. If Identity is true and Thomas is in a position to know the full nature of the taste of peaches, then Thomas is in a position to know that p.

2. Thomas is in a position to know the full nature of the taste of peaches. [Revelation]

3. Thomas is not in a position to know that p.

Therefore

4. Identity is false.

This is really an argument schema, of course. To obtain an argument, we need to put some sentence in place of ‘p’. Where the two versions of the Argument from Revelation differ is simply in what sort of claim they substitute in: the first substitutes a phenomeno-physical identity statement and the second substitutes some truth about the physical nature of the taste of peaches. Having seen the general form of the argument, though, it is clear that there might be other versions of it that substitute other statements for ‘p’. It is also clear that a similar style of argument could be used against any identity thesis, whether the purported identity is between phenomenal properties and physical properties or phenomenal properties and spiritual properties. This is why Stoljar concludes that Revelation implies an “uncompromising version of primitivism about experience according to which [qualia] are primitive items in the world, wholly distinct from everything else” (Stoljar 2009, 124; compare Byrne and Hilbert 2007).

The Argument from Revelation is remarkably similar to the controversial Knowledge Argument, laid out below. (Mary, recall, is the colour scientist who has never seen colour but knows all physical facts.)

5. If physicalism is true and Mary knows all the physical facts, then Mary is in a position to know that p.

6. Mary knows all the physical facts. [By hypothesis]

7. Mary is not in a position to know that p.

Therefore

8. Physicalism is false.8

As with the Argument from Revelation, the cogency of the Knowledge Argument depends on what we substitute for ‘p’. The idea, I take it, is that in this case ‘p’ is some sort of phenomenal truth. For example, it may be the truth that ripe tomatoes cause experiences (in normal perceivers) that instantiate phenomenal red, or that phenomenal red = P for some physical property P. Also like the Argument from Revelation, this argument at least appears to work as well against alleged identities between phenomenal and spiritual properties as between phenomenal and physical (see Lewis 1988, 588–589).

I hope the similarity between the two arguments is striking. Where Thomas peers across an epistemic gap from the phenomenal side, Mary peers back across it from the physical side. And both arguments attempt to parlay this gap into an ontological one. The difference between the two arguments, of course, is that only the Argument from Revelation appeals to the (dark) idea that Thomas knows the full nature of a phenomenal property. What I hope to show is that this difference is less telling than it seems. There are various ways of interpreting Revelation, and many of them allow us to adapt familiar physicalist responses to the Knowledge Argument to block the Argument from Revelation.9

4. Knowledge by acquaintance

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

Consider (7) again.

7. Mary is not in a position to know that p.

I mentioned above some possible candidates with which we might replace p. Some, however, think that there are no plausible candidates: although Mary learns something, there is no new proposition that Mary learns. One way to develop this idea is to say that what Mary acquires when she comes out of the room is not new propositional knowledge, but instead new knowledge by acquaintance, in the Russellian sense of that term. In other words, Mary learns no new proposition, but merely becomes acquainted with a phenomenal property for the first time (Bigelow and Pargetter 1990).

Now, the same Russell that introduced the idea of knowledge by acquaintance is also the main source for contemporary discussions of the doctrine of Revelation, though his discussion of it was focused on colour properties. Here is what he said:

The particular shade of colour that I am seeing . . . may have many things to be said about it . . . But such statements, though they make me know more truths about the colour, do not make me know the colour itself any better than I did before: so far as concerns knowledge of the colour itself, as opposed to knowledge of truths about it, I know the colour perfectly and completely when I see it, and no further knowledge is even theoretically possible. (Russell 1912, 47)

Russell's assertion that seeing a colour brings complete knowledge of the colour itself is often read as an endorsement of the idea that seeing the colour brings us complete knowledge of the nature, or essence, of the colour. He is thus taken to be endorsing Revelation, only Revelation as applied to colour and not phenomenal properties.10

However, what is not clear from the passage just quoted is whether Russell thinks that seeing a colour gives us complete knowledge of a set of nature-revealing propositions about it. In fact, what follows the colon in the passage above seems to imply that seeing the colour does not provide us with complete propositional knowledge of it. Instead, as John Campbell has recently argued, it is plausible that Russell is making the point that our knowledge by acquaintance is complete when we see a colour.

There are not, on Russell's view, different ways of being acquainted with one and the same colour. The point is . . . that the knowledge of the thing is complete; there is no further non-propositional knowledge of the thing to be had, once you have encountered it in experience. (Campbell 2009, 661, my emphasis)

So, on this interpretation, Russell argued not that experience delivers all nature-revealing propositions about a colour, but only that it delivers all possible acquaintance knowledge of the colour.

Campbell's interpretation of Russell brings in to view a possible reading of Revelation. That is, it brings in to view one thing we might mean by saying that experience puts us in a position to know everything there is to know about a phenomenal property.

Revelation1: When we have an experience with phenomenal property Q, we are in a position to have complete knowledge by acquaintance of Q (to have complete knowledge of Q itself).

Knowing the nature of a property in this sense does not imply any propositional knowledge of physical nature at all. The argument from Revelation fails, therefore, because it incorrectly supposes that Thomas' complete knowledge of the taste of peaches implies that he knows certain truths about the nature of peaches.

So, one way to obtain the result that Revelation and Identity are compatible is to read Revelation as a claim about our knowledge-by-acquaintance of phenomenal properties. While this may not give everyone what they wanted from Revelation, it is at least a plausible reading of that passage from Russell which is a main source for the doctrine of Revelation. And it is also a natural interpretation of that doctrine for those who endorse the acquaintance reply to the Knowledge Argument.

5. Knowledge of essential truths

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

Nevertheless, just as there are those who insist that Mary comes to learn some new proposition, so will there be those who insist that Revelation should be read as implying that Thomas has complete propositional knowledge of the taste of peaches. For those who think this, the natural way to read Revelation is as follows:

Revelation2: When we have an experience with phenomenal property Q, we are in a position to know all essential truths about Q.

On this reading, Revelation seems to straightforwardly imply that Identity is false. For if Thomas knows all essential truths about the taste of peaches, and Identity is true, then he should at least know that the taste of peaches = P, for some physical property P.11

But while Revelation2 allows us to derive the desired conclusion, there are problems with using it as a reading of Revelation. One problem is that, unlike the intuitive doctrine it is meant to capture, Revelation2 may imply that Thomas should know all sorts of esoteric necessary truths about the property of being the taste of peaches. For example, it may imply he needs to know that the property of being the taste of peaches belongs to the singleton set containing that property or, supposing the property is an Armstrongian universal, that it is multiply instantiable. Even if, following Fine (1994), we recognize that some necessary truths about a property are not essential truths, there may still be a difficulty in excluding these and similar esoteric metaphysical truths from the truths that Thomas must know in order to know the essence of a phenomenal property. While this may be a problem for Revelation2 (and other readings of Revelation discussed below), I'll ignore it in what follows. There are other problems that remain even if this one can be solved.

To see why, let's return to the Knowledge Argument. Although Mary knows all physical facts, it may not follow that Mary knows, or is in a position to know, all truths, even granted that there are no non-physical properties. For if there can be two concepts for the same property and truths are individuated by the concepts they involve while facts are individuated by the properties they contain, then there will be more truths than facts. Moreover, Mary may simply lack the concepts required to understand certain truths and not be in a position to acquire those concepts. In particular, it is may be both that one can't acquire phenomenal concepts without having an experience which instantiates the relevant phenomenal property and that the only truths Mary is yet to learn involve phenomenal concepts for those phenomenal properties she is being prevented from experiencing. So, Mary may know all the facts, and yet not know those truths that involve phenomenal concepts she is yet to acquire. When she does acquire them she will learn new truths – containing concepts for properties for which she already has physical concepts – but no new facts. This is the ‘missing concept’ reply to the Knowledge Argument.12

This reply can also be adapted to the Argument from Revelation. For what it reminds us is that it is implausible to read Revelation as entailing knowledge of all essential truths about a phenomenal property: on that exacting standard no-one could ever know the nature of anything without possessing all possible concepts for that thing. It is possible, therefore, that Thomas knows the full nature of the taste of peaches, and yet does not know certain essential truths about it that involve physical concepts, simply because he does not possess those concepts.

In light of this issue Revelation2 needs to be reformulated so that it only requires that knowledge of essence requires knowledge of all essential truths that we understand.

Revelation3: When we have an experience with phenomenal property Q, we are in a position to know all essential truths about Q that we understand.

Revelation3 is also open to objection, however. To see this, notice that the ‘missing concept’ reply to the Knowledge Argument arguably doesn't go far enough. Consider Experienced Mary. Experienced Mary once escaped her captors and briefly experienced the colours (the reds! the blues!) before being recaptured. She has thus acquired the phenomenal colour concepts. But it is arguable she might retain those concepts and yet forget which one applied to her experiences of the sky. It seems, therefore, that Experienced Mary has all the concepts she needs, but still doesn't know the relevant truths.13 Thus physicalism remains under threat.

The a posteriori (or Type-B) physicalist, however, has a response to this problem. Like the missing concept strategist the a posteriori physicalist individuates concepts more finely than properties and agrees that Mary learns a new truth, but no new fact. What gives this position its title of a posteriori physicalism is the further claim that Mary can't reason a priori from the physical truths she knows to the phenomenal truths she doesn't, even though she has complete knowledge of the physical facts, and even if she already possesses the relevant phenomenal concepts (Horgan 1984; Loar 1997; Tye 1999; Papineau 2002; 2007).

Now, on the face of it, a similar reply can be applied to the Argument from Revelation which uses Revelation3. For it seems possible to know the full nature of a thing for which one has a concept, grasp another, co-extensional concept, and yet not be able to work out, a priori, that the two concepts are co-extensional. Here's an example. I may know the full nature of Mohamed Ali. That is, I may come to know everything there is to know about his essence – including that he is human, perhaps who his parents are if biological origins are essential, and so on. I may then acquire the concept cassius clay14, which rigidly designates the selfsame person. Nevertheless, it is easy to imagine that I acquire the latter concept in a way that I cannot work out, a priori, that Mohamed Ali is Cassius Clay. At least, it is hard to see why those who think that that identity statement is a posteriori should think that it must be a priori derivable for those who know the nature of Mohamed Ali. Here's another example. Suppose I know the full nature of water. Then I notice that a child has her own concept for a liquid she loves, namely retaw, and I then acquire that concept. Still, it's not obvious that I should be able to discover, a priori, that retaw = water. At any rate, it seems no more obvious than that water = H2O should be knowable a priori. Moreover, both of these examples involve, or could be elaborated so that they involve, identities between two concepts that both rigidly designate.15

Thus we still do not have a persuasive argument from Revelation to the denial of physicalism. The current formulation of that argument relies on reading Revelation as Revelation3. But this reading begs the question against an a posteriori physicalist, who holds that some phenomeno-physical identity statements may be necessary a posteriori truths which may not be derivable a priori. As the above examples prima facie suggest, an a posteriori physicalist should also hold that these statements are not derivable a priori even by someone who knows the nature of the relevant property. That is, an a posteriori physicalist has independent reasons for rejecting Revelation3 on the grounds that it is an inappropriate, question-begging way of interpreting Revelation.16

6. The Enhanced Knowledge Argument

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

The argument of the previous section drew an analogy between the a posteriori physicalist's response to the knowledge argument and a response to the Argument from Revelation which relies on Revelation3. However, there is also an important point of difference between these two responses which might be thought to undermine the analogy. In this section I explain the difference but argue that it is unproblematic: in fact, I hope to show that reflecting on the difference shows why a posteriori physicalists are committed to allowing that one can know the nature of a property without knowing all essential truths about it. In the following section I resume the search for a more appropriate way to unpack Revelation.

A posteriori physicalism remains controversial because although most agree that there are necessary a posteriori truths – such as water = H2O – there is less agreement about whether the identity ‘P = phenomenal red’ is a posteriori (or, more accurately, whether that identity is a priori derivable from all the physical facts). This claim is controversial because phenomenal concepts seem importantly different from concepts like water which, although rigid designators, appear to have contingent reference fixers. To put the point another way, the extension of water in counterfactual scenarios apparently depends on how things are in the actual world, or is actuality-dependent. In contrast, taste of peaches seems to not only be a rigid designator, but also have a rigid reference fixer: neither fool's pain nor fool's taste of peaches seem possible. That is, taste of peaches appears to be actuality independent. This special feature of phenomenal concepts suggests to some that the relevant phenomeno-physical identities should be a priori. At least, they should be a priori if we also grant that the relevant physical concepts are also rigid designators with rigid reference fixers, or are also actuality independent.17 To overcome this challenge, a posteriori physicalists have typically tried to make it plausible that there can be two co-extensional, actuality independent concepts which can't be known to be co-extensional a priori. They have typically tried to show this by appealing to further postulated special features of phenomenal concepts (Loar 1997; Papineau 2002; 2007).

This extra wrinkle raises a problem for the a posteriori physicalist response to the Argument from Revelation discussed in the previous section. The Mohamed Ali = Cassius Clay and water = retaw examples rely on at least one of the concepts involved having a contingent reference fixer, or being actuality dependent. So if our phenomeno-physical identities lack this feature, then the line of thought just canvassed suggests that they should be a priori after all, especially if we also know the nature of the relevant property. And it does not seem that the attempt to appeal to any special features of phenomenal concepts will alleviate this worry. This is because those who appeal to such special features in the context of the Knowledge Argument often do so as part of a strategy to show that phenomenal concepts can be actuality-independent and yet those who possess such concepts, and have had the relevant experiences, might still not be in a position to know the nature of phenomenal properties (Papineau 2007, 131; Loar 1997, 603).18 Such a strategy obviously can't be applied to the Argument from Revelation.

To respond to this problem, allow me to introduce the Enhanced Knowledge Argument. In doing so I hope to show that the a posteriori physicalist faces certain explanatory burdens independently of the Argument from Revelation and that the strategies they must use to discharge those burdens can also be used to block the Argument from Revelation.

The Enhanced Knowledge Argument is ‘enhanced’ because it begins with the assumption that Mary knows the nature of all physical properties, rather than merely knowing all physical facts. Otherwise it has a similar form to the original argument. (To finesse worries about the missing concept reply, suppose also that Mary is Experienced Mary.)

Enhanced Knowledge Argument

 9. If physicalism is true and Mary knows the nature of all physical properties, then Mary is in a position to know that p.

10. Mary knows the nature of all physical properties. [By hypothesis]

11. Mary is not in a position to know that p.

Therefore

12. Physicalism is false.

The crucial difference between the Enhanced Knowledge Argument and the original is obviously (10). But while there is this difference, talk of Mary knowing the nature of the physical doesn't substantially alter the dialectic between physicalists and their opponents. For there seems to be only two ways to deny (10) given Mary knows all physical facts. First, it could be said that the nature of the physical somehow depends upon non-physical facts which Mary doesn't know. But even if such an idea is coherent, it is not open to the physicalist, who holds that there are only physical facts. Second, it could be argued that Mary doesn't know the nature of the physical, because the intrinsic nature of the physical cannot be known through experience and experiment, both of which reveal only the extrinsic, structural properties of the physical. This is Russellian Monism. Yet, as has often been pointed out, Russellian Monism can also be used to block the original Knowledge Argument. So, overall, there is no point of difference here between the original and enhanced versions of the Knowledge Argument: for physicalists who aren't Russellian Monists, both versions are equally pressing (and for those who are, there seems to be an easy way to avoid both).

If we grant (10), this leaves (9) and (11). (11) is just as plausible as it was in the original Knowledge Argument – or will be when we substitute for p the same proposition that was to be substituted for it in the original argument. So the only option left to the physicalist is to deny the relevant instances of (9).

One such instance is obtained by substituting ‘phenomenal red = P’ for ‘p’, where P is the relevant physical property. If the physicalist denies that instance of (9), as it seems she must, then she is denying that Mary must know the proposition ‘phenomenal red = P’. But this means she is in effect saying that identity statements between actuality-independent physical and phenomenal concepts are a posteriori even when one knows the nature of the property at issue. And this is just the claim that is needed to block the Argument from Revelation. So, when it comes to the instances of (9) that involve phenomeno-physical identity statements, the extra wrinkle does not show that the a posteriori physicalist will face any new difficulty in attempting to reconcile Revelation with physicalism.

However, it is not only identity statements that Thomas doesn't know. To use Lewis's term, Thomas also doesn't know whatever essential truths there are about the ‘grain’ of the property. More precisely, he doesn't know about the internal constitution of the taste of peaches: if the taste of peaches is a conjunctive or structural property, Thomas knows neither what it is compounded from or that it is a compound property.19 And yet since these seem essential truths about the taste of peaches Thomas should know them if Identity and Revelation were true.

In fact, it's not clear that he must. The physicalist might deny that Thomas should know about this structure. Again, drawing a parallel to Mary and the Enhanced Knowledge Argument helps to see this. This time, however, consider a Mary who doesn't know all physical facts, but only enough about the physical to know the nature of that physical property which is phenomenal red (assuming Identity). Now ask whether Mary would know the following essential truth about phenomenal red: phenomenal red appears to introspection to be a simple and intrinsic property of experiences.20 (Since this is part of what it's like to have an experience of phenomenal red I think it's undeniable that it is part of the essence of that phenomenal property.)

Whether Mary would know that truth depends on the sort of property that phenomenal red turns out to be. On the one hand, suppose it is the property of being an R-fiber firing, for example, where R-fibers are a type of neuron defined by their internal make-up (rather than their functional role). In this case, it is intuitive that Mary could know the nature of this property in physical terms – know its entire physical make-up – without knowing that it appears to introspection to be a simple and intrinsic property of experiences. Thus there is a burden on the physicalist to explain how one can know the nature of a property without knowing an essential truth about it, an essential truth which is not an identity statement. But this is the same burden facing the physicalist who endorses Revelation.

On the other hand, it may be that the relevant physical property is a functional role property, where the role is defined by its place in the entire information processing system of the brain, including a system responsible for monitoring its own internal states. It may even be that somewhere among the facts about this system are the facts that determine that phenomenal red appears to introspection to be a simple and intrinsic property of experiences. Nevertheless, it is still not clear that our current Mary should know this essential truth. That is, it is not clear that Mary should be able to ‘read-off’, a priori this essential truth from the facts she knows about the functional role property. At least, it seems natural for an a posteriori physicalist to maintain that the connection between that essential phenomenal truth and the facts concerning the functional role property can only be discovered a posteriori. And if it seems natural to say this about Mary there should be no further cause for concern in saying that Thomas knows the nature of the taste of peaches without knowing all the physical, essential truths about it. Once again Thomas and Mary appear to be on all fours.

Where have we got to? We have seen that physicalists need to discharge various explanatory burdens in response to the Enhanced Knowledge Argument and that the strategies open for them to do so can also be applied to block the Argument from Revelation. More specifically, responding to the Enhanced Knowledge Argument requires denying that we must know all the essential truths about a property in order to know its nature. Thus, while Loar, Papineau and others have designed their accounts to have the result that phenomenal concepts are actuality independent without thereby making certain essential truths derivable a priori, this does not entail that they must deny phenomenal concepts are essence revealing. On their own a posteriori physicalist position, I've argued, phenomenal concepts might be essence revealing and yet possessor's of those concepts may not know all essential truths about the phenomenal property. A posteriori physicalists, having drawn a distinction between facts and truths, should hold that even if knowing the full nature of a property requires knowing all essential facts about that property – or all essential properties of that property – this does not imply knowing all essential truths about that property. I will further explore the idea that knowing the nature of a property requires knowing all essential properties of it in the following section.

Before doing so, however, there is a further potential disanalogy between the Enhanced Knowledge Argument and the Argument from Revelation to deal with. It might be thought that physicalism implies that the canonical description of the world is in physical terms, or that the essence of a property is best captured by a physical description of it. If so, then it is plausible to say that Mary already knows the full nature of phenomenal red, say, and no phenomenal truth can further reveal its essence. While, in contrast, if Revelation and Identity are true then Thomas would know the nature of the taste of peaches under the physical description since that best captures its essence. This would be a worrisome disanalogy if it were genuine. However, since physicalism is an ontological thesis we should resist the idea that it implies that any particular mode of description is the ultimate or best description of reality. Physicalism implies that all the facts are physical, but this does not imply that these facts can't be equally well captured using concepts other than those of basic physics.21

7. Knowledge of essential properties

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

The discussion so far suggests that the Argument from Revelation might be more profitably developed by interpreting Revelation in terms of our knowledge of essential facts or essential properties rather than truths. In other words, we might unpack Revelation as follows:

Revelation4: When we have an experience with phenomenal property Q, we are in a position to know for each essential property, F, of Q that Q is F.

This thesis requires only that Thomas know all the essential properties of the taste of peaches. It does not require that he possess any particular concepts for those properties. Instead, he must only have some concept for each essential property of Q, and under that concept he must know that Q has that property.

On this approach, Thomas's failure to know various identity statements will not be to the point in the Argument from Revelation. For if we suppose, say, that being a P-fiber firing is the same property as being the taste of peaches, then the property of being identical to a P-fiber firing is the same property as being identical to the taste of peaches. According to Revelation4, therefore, knowing the nature of the taste of peaches does not imply that Thomas must know that the taste of peaches is identical to P-fiber firings: it is sufficient that he knows the trivial truth that the taste of peaches is identical to the taste of peaches.

However, as I in effect pointed out in the previous section, there may be other essential properties of the taste of peaches that Thomas is unaware of, including those properties related to any ‘internal constitution’ it has. Whether there are such properties or not depends on two things. First, it obviously depends on whether it turns out that the physical property has essential properties related to its internal constitution. Second, it depends on whether Thomas is already in a position to know, without empirical investigation, that the taste of peaches has all those properties.

The answer to neither of these questions is obvious. Suppose for the sake of argument, though, that the first is answered affirmatively. To answer the second, we need to know which essential properties of the taste of peaches Thomas is in a position to know. He is certainly in a position to know that it is a taste experience; that it appears to introspection to be a simple property of experiences; that it appears to introspection to be an intrinsic property of experience; perhaps also that it stands in certain relations to other taste experiences.22 These properties are, if Identity is true, physical properties. Furthermore, they are essential physical properties of the taste of peaches and Thomas knows about them under phenomenal, or other, concepts. So it seems he knows at least some of the essential properties he needs to know about if Revelation and Identity are to be compatible, only he knows them under nonphysical concepts.

Does Thomas know enough to know of all the essential properties of the taste of peaches, though? Again, I think the answer is far from obvious. Moreover, at this point I think that the answer we give will depend on the type of physicalism at issue. If we continue to restrict our attention to Identity, then it may well be that the taste of peaches has an ‘internal constitution’ of some sort, and that there are essential properties of it that Thomas doesn't know about under any concept. But in the absence of further argument that there are such further, hidden, essential properties, whether there is conflict between Revelation4 and Identity remains unclear.

It is also important to remember that, for the sake of discussion, I have been ignoring the most plausible version of physicalism, namely supervenience physicalism. Yet if we allow that the taste of peaches is a multiply realizable property – i.e. that Identity is false – then the taste of peaches itself would seem not to have any particular ‘internal constitution’ essentially at all, even though particular instantiations of that property may be physical, structural properties.23 Thus it may be that on more plausible versions of physicalism Thomas does know all the essential properties of the taste of peaches.

The upshot, then, is that (1) it remains unclear whether very strong versions of physicalism are inconsistent with Revelation4, but we can say that (2) a more plausible supervenience physicalism does not seem to conflict with it.

Whether or not Revelation4 is consistent with physicalism, however, it is important to remember that it is but one of several plausible readings and it is a demanding one. For on this approach, to know the nature of a property which has an internal constitution, one must apparently know its nature ‘all the way down’. If we are to know the nature of water, for example, we need to know not just that water is composed of H2O molecules, but also that Hydrogen molecules contain a proton and an electron combined in a certain way and that protons are made from quarks and quarks are made from. . . . At any rate, we need to know all these things until we reach a point at which the relevant properties, being a quark say, have no essential internal constitution. And if we never reach such a point – if the properties each have an essential internal constitution and the universe is gunky24– then it is in principle impossible to know the nature of water. These consequences bring out just how demanding this conception of knowing the nature of a property is. For one might think that we already know the nature of water, or, less permissively, that whether or not the universe is gunky shouldn't determine whether or not we can know the nature of familiar properties like water. That the current attempt to clarify the concept of knowledge of essences conflicts with these thoughts is not sufficient to rule it out as a plausible clarification of that concept. But it does suggest that we also have a less demanding conception of what it is to know the nature of a property. In the next section I consider one way of developing such a conception.

8. Knowledge of counterfactual extension

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

A less demanding conception of knowledge of essence holds that to know the nature of a property is to know the counterfactual extension of some concept one has of it. We can apply this approach to Revelation as follows.

Revelation5: When we have an experience with phenomenal property Q, we are in a position to know the counterfactual extension of some phenomenal concept, C, of Q.25

This thesis implies neither that we are in a position to know all essential truths about, nor all essential properties of, phenomenal properties. It implies only that the relevant experiences put us in a position to know enough essential truths or properties of the taste of peaches so that we can determine the counterfactual extension of taste of peaches. While this means Revelation5 is a less demanding interpretation of Revelation than some of those considered above it is arguable that it still captures the core idea of knowing the nature of a property. In the actual world, the things which instantiate a given property may share certain features only contingently. By considering the counterfactual extension of a concept for that property we abstract away from such contingently shared features and consider which features a thing can change and yet still possess the relevant property. This is why metaphysical disputes over the nature of a property often turn on whether and how that property is instantiated in counterfactual worlds (Nida-Rümelin 2007, 311–312). Thus there is a robust sense in which knowing the counterfactual extension of a concept for a property entails knowing what makes the property the property it is, or being able to identify it in an uncommonly demanding way. Moreover, to know the counterfactual extension of taste of peaches it is not necessary to know truths such as that the property of being the taste of peaches belongs to the singleton set containing it or, if it is an Armstrongian universal, that it is multiply instantiable. Revelation5 thus also has the plausible consequence that Thomas might know the nature of the taste of peaches without knowing these other esoteric metaphysical truths.

In recent work, both Martine Nida-Rümelin (2007) and David Chalmers (2004; 2009) have given accounts of phenomenal concepts which develop this approach, though each do so in slightly different ways. Despite their differences, both are committed to the following general picture. For some concepts we are in a position to say, for each world considered counterfactually, what the extension of that concept is. In other words, for every counterfactual world there is some presentation of it such that we are able to determine the extension of the concept at that world. If we have this ability, then we can be said to know the nature of the property associated with the concept. Consider water, as our ever-reliable example. Water is not an actuality independent concept. Therefore, mere possession of that concept does not put us in a position to determine its counterfactual extension: or,as one might say in this context, it is not revelatory, since mere possession of water doesn't put us in a position to know the nature of water. But having ascertained through a posteriori investigation that water is H2O in the actual world, we are now able to determine for every counterfactual world presented in molecular terms where the water is at that world – we now know the nature of water. (Note that we shouldn't be said to know the nature of water if we could determine the counterfactual extension of water only when worlds were described in terms of the very concept at issue (i.e. water) or by referring back to the actual world (e.g., the stuff I actually drank a few minutes ago).)

Now, in contrast to their claims about water, both Chalmers and Nida-Rümelin argue that phenomenal concepts are actuality independent and so are revelatory in the appropriate sense26: we know the extension of phenomenal concepts in counterfactual scenarios independently of any knowledge about which world is actual (except for knowledge required to possess the phenomenal concept).27 Moreover, both Nida-Rümelin and Chalmers also use their claims about phenomenal concepts to mount anti-physicalist arguments. To simplify greatly, both arguments move from the idea that phenomenal concepts and certain physical concepts are revelatory, in the sense above, to the claim that the identity statements between phenomenal and physical concepts are, if true, a priori.28 Since no such statement is a priori, they argue, none is true.29 Or, to put the same argument in another way, they hold that a person in possession of two actuality independent concepts with the same counterfactual extensions should be able to determine the extension of one concept in a counterfactual world when that world is described using the other concept. Since we can't do this with phenomenal concepts when worlds are described with actuality independent physical concepts, physicalism is false.

However, as we saw above, a posteriori physicalists are already committed to denying that identities between the physical and the phenomenal are a priori, even though both concepts are actuality independent. They are, therefore, committed to denying the crucial premise required to derive the falsity of physicalism from Revelation5.30 That is, an a posteriori physicalist like Loar or Papineau will maintain that even though phenomenal concepts are actuality independent, a person in possession of a phenomenal concept may still not be able to determine the extension of that concept in a counterfactual world described in purely physical terms. Indeed, their accounts of phenomenal concepts are designed to make such a position plausible. Thus if Revelation involves only the commitment to knowledge of the counterfactual extension of phenomenal concepts, then a posteriori physicalists have nothing further to fear from accepting it. So we have arrived at a version of Revelation that a posteriori physicalists can and should accept.

In saying that a posteriori physicalists can and should accept Revelation5 I have not forgotten that they typically insist that by agreeing to the claim that phenomenal concepts are actuality independent they are not agreeing to the idea that such concepts reveal the essence of phenomenal properties. But according to Revelation5 such a position is plainly inconsistent, since on this understanding concepts that are actuality independent are revelatory by definition. It must therefore be an alternative understanding of Revelation that a posteriori physicalists have in mind when they reject it. If it is one of the versions of Revelation considered above they have in mind, though, we have seen that even on the most threatening of those, i.e. Revelation4, it is far from clear that it should be rejected on the grounds that it undermines Identity and it does not seem to undermine supervenience physicalism. It is hard to see what other grounds an a posteriori physicalist would have for rejecting Revelation.31

9. Russellian Monism

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

We've seen several ways that we might interpret Revelation so that it is compatible with physicalism. Before finishing I'd like to briefly consider one alternative way we might reconcile the two ideas, namely by adopting Russellian Monism. In this context, Russellian Monism is the position that phenomenal properties do not supervene on the structural, physical properties, but do supervene on those properties together with the intrinsic properties of the physical.

Now, if both Identity and Revelation are true, then when Thomas tastes peaches he is in a position to know the nature of a physical property. So suppose that, in keeping with one possible form of Russellian Monism, the intrinsic nature of the physical is in fact phenomenal. Then it could be argued that the phenomenal knowledge Thomas is in a position to gain when he tastes peaches is knowledge of the intrinsic nature of some physical property. And this might be held to be the essence of that physical property. Whatever the plausibility of this position, if this is how things are, there is no conflict between Identity and Revelation.

More viable versions of Russellian Monism, however, point up a difference between the Knowledge Argument and the Argument from Revelation. Suppose the taste of peaches is identical to P-fibers firing. The full nature of P-fiber firings presumably consists of many other physical properties distinct from the taste of peaches. A Russellian Monist might hold either that the intrinsic nature of these properties is also phenomenal (panpsychism) or that while not themselves phenomenal their intrinsic nature is such that when combined in certain ways they constitute phenomenal properties (panprotopsychism) (compare Chalmers 2004, 283). On either approach, more basic phenomenal or proto-phenomenal properties somehow sum to create the taste of peaches. Since Thomas presumably does not know of such properties, or how they combine to make phenomenal properties, the tension between Identity and Revelation re-emerges. Thus these positions do not, on their own, provide a way of blocking the Argument from Revelation (compare Goff 2006). On the other hand, they do provide a way of blocking both the original and enhanced versions of the Knowledge Argument. For according to Russellian Monism, Mary can't know the intrinsic nature of the physical on which the phenomenal depends while locked in her room: she has access only to the structural and relational physical properties which science and perception can uncover. Since she doesn't have access to part of the supervenience base for phenomenal truths, she shouldn't be expected to be able to infer those truths.

Thus, some versions of Russellian Monism highlight an important difference between the Knowledge Argument and the Argument from Revelation. Whether one takes the difference to imply that the physicalist should therefore be wary of adopting Revelation depends on whether one takes those versions of Russellian Monism as serious physicalist options.32

10. Conclusion

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References

The obscure thesis of Revelation has been much maligned by physicalists and much loved by anti-physicalists. Yet, as I hope to have shown, it is far from clear that Revelation and physicalism are in tension. Some readings of Revelation are consistent with physicalism and others build in conceptions of what it is to know the nature of a property that are either implausible or unacceptable to a posteriori physicalists. Even those readings of it that are otherwise acceptable aren't obviously inconsistent with physicalism. Finally, it is worth reiterating that a posteriori physicalists can accept a robust interpretation of Revelation (section 8). Given the attractiveness of the principle, this should be welcome news.*

References

  1. Top of page
  2. A
  3. 1. Thomas, Son of Mary
  4. 2. The thesis of Revelation
  5. 3. The argument from Revelation and the Knowledge Argument
  6. 4. Knowledge by acquaintance
  7. 5. Knowledge of essential truths
  8. 6. The Enhanced Knowledge Argument
  9. 7. Knowledge of essential properties
  10. 8. Knowledge of counterfactual extension
  11. 9. Russellian Monism
  12. 10. Conclusion
  13. References
  • Adams, R. 1987, “Flavors, Colors, and God”, in: The Virtue of Faith and Other Essays in Philosophical Theology, Oxford: Oxford University Press, pp. 243262.
  • Armstrong, D. 1978, Universals and Scientific Realism, v II, Cambridge: Cambridge University Press.
  • Ball, D. 2009, “There Are No Phenomenal Concepts”, Mind, 118, pp. 935962.
  • Bealer, G. 1994, “Mental Properties”, Journal of Philosophy, 20, pp. 185208.
  • Bigelow, J. and Pargetter, R. 1990, “Acquaintance with Qualia”, Theoria, 61, pp. 129147.
  • Byrne, A. and Hilbert, D. 2007, “Colour Primitivism”, Erkenntnis, 66, pp. 73105.
  • Campbell, J. 2009, “Consciousness and Reference”, in: B. McLaughlin and A. Beckermann, eds, Oxford Handbook of the Philosophy of Mind, Oxford: Oxford University Press, pp. 648662.
  • Chalmers, D. 1996, The Conscious Mind, Oxford: Oxford University Press.
  • Chalmers, D. 2004, “Phenomenal Concepts and The Knowledge Argument”, in: P. Ludlow, Y. Nagasawa and D. Stoljar, eds, There's Something About Mary, Cambridge, MA: MIT Press, pp. 269298.
  • Chalmers, D. 2009, “The Two-Dimensional Argument Against Materialism”, in: B. McLaughlin and A. Beckermann, eds, Oxford Handbook of the Philosophy of Mind, Oxford: Oxford University Press, pp. 313337.
  • Fine, K. 1994, “Essence and Modality”, Philosophical Perspectives, 8, Logic and Language, pp. 116.
  • Goff, P. 2006, “Experiences Don't Sum”, in: A. Freeman, ed., Consciousness and Its Place in Nature: Does Physicalism Entail Panpsychism? Exeter: Imprint Academic, pp. 5361.
  • Horgan, T. 1984, “Jackson on Physical Information and Qualia”, reprinted in: P. Ludlow, Y. Nagasawa and D. Stoljar, eds, There's Something About Mary, Cambridge, MA: MIT Press, pp. 301308.
  • Jackson, F. 1982, “Epiphenomenal Qualia”, Philosophical Quarterly, 32, pp. 127136.
  • Jackson, F. 1986, “What Mary Doesn't Know”, Journal of Philosophy, 83, pp. 291295.
  • Jackson, F. 1998, From Metaphysics to Ethics, Oxford: Oxford University Press.
  • Johnston, M. 1992, “How to Speak of the Colours”, Philosophical Studies, 68, pp. 221263.
  • Kripke, S. 1980, Naming and Necessity, Oxford: Blackwell.
  • Lewis, D. 1986, “Against Structural Universals”, reprinted in: Papers in Metaphysics and Epistemology, Cambridge: Cambridge University Press, pp. 78107.
  • Lewis, D. 1988, “What Experience Teaches”, reprinted in: N. Block, O. Flanagan and G. Guzeldere, eds, The Nature of Consciousness: Philosophical Debates, Cambridge, MA: MIT Press, 1997, pp. 579596.
  • Lewis, D. 1995, “Should a Materialist Believe in Qualia?”, in: Papers in Metaphysics and Epistemology, Cambridge: Cambridge University Press, pp. 325331.
  • Lewis, D. 1997, “Naming the Colours”, in: Papers in Metaphysics and Epistemology, Cambridge: Cambridge University Press, pp. 332358.
  • Loar, B. 1997, “Phenomenal States (Revised Version)”, reprinted in: P. Ludlow, Y. Nagasawa and D. Stoljar, eds, There's Something About Mary, Cambridge, MA: MIT Press, pp. 219240.
  • Mclaughlin, B. 2003, “The Place of Color in Nature”, in: R. Mausfeld and D. Heyer, eds, Colour Perception: Mind and the Physical World, Oxford: Oxford University Press, pp. 475505.
  • Nida-Rümelin, M. 1995, “What Mary Couldn't Know: Belief About Phenomena States”, in: T. Metzinger, ed., Conscious Experience, Exeter, UK: Imprint Academic, pp. 219241.
  • Nida-Rümelin, M. 2007, “Grasping Phenomenal Properties”, in: T. Alter and S. Walter, eds., Phenomenal Concepts and Phenomenal Knowledge, Oxford: Oxford University Press, pp. 307338.
  • Papineau, D. 2002, Thinking About Consciousness, Oxford: Oxford University Press.
  • Papineau, D. 2007, “Phenomenal and Perceptual Concepts”, in T. Alter and S. Walter, eds, Phenomenal Concepts and Phenomenal Knowledge, Oxford: Oxford University Press, pp. 111144.
  • Russell, B. 1912, The Problems of Philosophy, Oxford: Oxford University Press.
  • Stalnaker, R. 2008, Our Knowledge of the Internal World, Oxford: Oxford University Press.
  • Stoljar, D. 2005, “Physicalism and Phenomenal Concepts”, Mind and Language, 20, pp. 469494.
  • Stoljar, D. 2006, Ignorance and Imagination, Oxford: Oxford University Press.
  • Stoljar, D. 2009, “The Argument from Revelation”, in: D. Braddon-Mitchell and R. Nola, eds, Conceptual Analysis and Philosophical Naturalism, Cambridge, MA: MIT Press, pp. 113138.
  • Strawson, G. 1989, “Red and ‘Red’ ”, Synthese, 78, pp. 193232.
  • Tye, M. 1999, “Phenomenal Consciousness: The Explanatory Gap as a Cognitive Illusion”, Mind, 108, pp. 705725.
  • Tye, M. 2000, Consciousness, Color and Content, Cambridge, MA: MIT Press.
  • Tye, M. 2009, Consciousness Revisited: Materialism Without Phenomenal Concepts, Cambridge, MA: MIT Press.
  • Williamson, T. 2000, Knowledge and Its Limits, Oxford: Oxford University Press.
Footnotes
  • 1

    For example, Lewis 1995; 1997; McLaughlin 2003, 478; Stoljar 2006; 2009; Stalnaker 2008, 99. A related revelation thesis is also discussed in relation to colour, from which the ‘Revelation’ label has been taken. There are many more philosophers who hold that thesis to be incompatible with physicalism about the colours. See, for example, Byrne and Hilbert 2007; Johnston 1992; Jackson 1998, 102.

  • 2

    “Why do I think [Revelation] must be part of the folk theory of qualia? Because so many philosophers find it so very obvious” (Lewis 1995, 328). Lewis then names Adams 1987 and Kripke 1980 as examples. Lewis rejected qualia “strictly so called” because he thought Revelation incompatible with physicalism.

  • 3

    McLaughlin 2003, 476–478 also makes this point. In fact, Russell 1912 and Strawson 1989, 213, two sources for the contemporary discussion of Revelation for colour, were focused on sense-data and qualia respectively.

  • 4

    Following Stoljar 2009, 119, here and throughout I assume that if it is in the nature of the taste of peaches that p then Revelation only requires that we know that p, and not that it is in the nature of the taste of peaches that p. This is in contrast to Byrne and Hilbert 2007.

  • 5

    Since it is harder to block the Argument from Revelation against the identity theory than to block it merely against a supervenience claim, this won't beg any important questions.

  • 6

    Stoljar 2006; 2009 focuses on this form of the argument.

  • 7

    Stalnaker 2008, 99 endorses this argument explicitly. Compare the reason Jackson 1998,103 gives for saying that a revelation thesis about heat (the property heat, not the phenomenal feel of heat) is false. He points out that “[w]e did not hesitate to identify heat with something [molecular kinetic energy] whose full nature is manifestly not given to us in the experience of heat.” Likewise, compare Johnston 1992, 225 explaining why revelation about colour is inconsistent with various physicalist theses about colour: “In any case, visual experience certainly leaves us with a lot more to know about the nature of both the categorical microphysical properties of surfaces and the reflectance properties of surfaces. So these properties do not satisfy Revelation.”

  • 8

    Although this formulation may appear simplistic given the contemporary debate, it is more or less the formulation of the argument in Jackson 1986, 293–294.

  • 9

    Not all responses to the Knowledge Argument can be applied to the Argument from Revelation. In particular there is no room to apply Lewis's 1988 response, namely the ability hypothesis. It is perhaps not surprising, therefore, that Lewis thought the Argument from Revelation, but not the Knowledge Argument, was successful.

  • 10

    Though, as mentioned in n.3, Russell held we are acquainted with sense-data, so it is not entirely clear whether he meant Revelation to apply to colours or phenomenal properties.

  • 11

    See Stoljar 2006; 2009. Lewis's 1995 discussion seems to imply a weaker version of this approach – he requires knowledge of a necessary and sufficient property of the phenomenal property, but knowledge de dicto not de re. In the literature on colour, see Byrne and Hilbert's 2007Self-Intimation, which they claim is implied by Revelation and implies primitivism, and Johnston 1992.

  • 12

    For an example see Tye 2000, 3–20, but for recent doubts see Tye 2009, ch3. See also Ball 2009.

  • 13

    The example of Experienced Mary is from Stoljar 2005, but a similar, and more developed, example can be found in Nida-Rümelin 1995.

  • 14

    Here, and in what follows, I use small caps to form the name of a concept.

  • 15

    I am ignoring complications about how to spell out rigid designation for concepts. There is another worry about the fact that Cassius Clay and retaw both have contingent reference fixers to which I'll return below.

  • 16

    The same response can be made to Lewis's weaker version of Revelation2, described in n.11.

  • 17

    For example, Chalmers 1996; 2009 and Bealer 1994. Jackson 1998 argues more generally against a posteriori physicalism.

  • 18

    Loar actually distinguishes two notions of ‘capturing the essence’ and tries to show that his account of phenomenal concepts don't capture the essence of phenomenal properties in the more demanding of his two senses.

  • 19

    For more on structural properties, see Armstrong 1978, vII, 69–71 and Lewis 1986.

  • 20

    Adams 1987, 253 and Lewis 1995, 329 both point out that we don't notice any structure when introspecting qualia.

  • 21

    Thanks to an anonymous referee for pressing the concern discussed in this paragraph.

  • 22

    Whether Thomas knows this last truth depends on whether he needs to have had other taste experiences to fully acquire taste of peaches.

  • 23

    Notice that although a supervenience thesis is consistent with primitivism, the point made in the text holds true even on non-primitivist versions of supervenience physicalism, so this is not a capitulation to the view that Revelation entails primitivism.

  • 24

    A gunky object is one whose parts all have proper parts.

  • 25

    Notice that since this says only that one is “in a position to know” it implies neither that experience is necessary to acquire the phenomenal concept, nor that experience will always yield full understanding of a phenomenal concept. Ball 2009 and Tye 2009, 63–73 argue that such consequences are implausible.

  • 26

    For recent criticism of this view, see Tye 2009, 70.

  • 27

    Nida-Rümelin 2007, 325 allows that for actuality independent concepts it is sufficient for knowing the nature of the associated property that one can determine the extension at counterfactual worlds described using that very concept. I think this threatens to trivialize Revelation5 and so would apply the limitations expressed in the previous paragraph here as well. While I can't argue for it here, I say we know the counterfactual extension of a phenomenal concept because when the phenomenal property is imaginatively presented as being instantiated at a certain region in a counterfactual world we can recognize it falls under the concept.

  • 28

    Nida-Rümelin calls her thesis “Cognitive Transparency”; the equivalent thesis in Chalmers is “CP”.

  • 29

    Although Chalmers runs his in terms of supervenience and conceivability, the issues are the same.

  • 30

    Chalmers has himself made a similar point informally: http://fragments.consc.net/djc/2006/12/nidarumelin_on_.html

  • 31

    Some may have thought that to call a concept ‘revelatory’ means, by definition, that an identity between it and another revelatory concept must be a priori if true. But if that idea takes us beyond the claim that phenomenal concepts are actuality independent it is obscure what the resulting notion of a revelatory concept is, or what reason we have to think that phenomenal concepts are revelatory in that sense.

  • 32

    Chalmers 2009, for example, is unsure enough to formulate the conclusion of his argument against physicalism as the disjunctive thesis that either Physicalism is false or Russellian Monism is true.

  • * 

    For helpful comments and discussions thanks to Miri Albahari, Stewart Candlish, Dave Chalmers, Benj Hellie, Daniel Stoljar, Barry Maund, Adam Pautz and the anonymous referees for dialectica. Too many philosophy departments have heard this paper and given helpful feedback to list them here, but special thanks to Raamy Majeed and the University of Sydney.