A Naïve Adverbialism

Chapter Five

A Naïve Adverbialism

The naïve realist needs to be able to explain the possibility of illusory experience, that is, experiences in which things in the world appear to be other than they are. I have looked at and rejected one strategy for the naïve realist to take. This strategy was to explain the illusions appealed to by the argument from illusion as perceptual experiences in which something phenomenally appears just as it is, but phenomenologically appears other than it is due to how the object of experience is taken. This approach failed because of the implausibility of supposing, in the case of illusions such as the white chalk seen in red light, that the chalk phenomenally appears just as it is. Furthermore, while the naïve realist could still try to accommodate this illusion in terms of variation in illumination conditions bringing about a change in surface properties of the chalk, there is a class of colour illusions for which this is not a viable strategy. These are simultaneous colour contrast effects, in which something appears a certain colour because of the surrounding colour context.

This whole approach to dealing with illusion accepted a claim about the phenomenal character of perceptual experience that is pivotal in the argument from illusion against naïve realism, namely, that it is constituted by the properties of the object of experience.

In this chapter I will argue that it is through a rejection of this central principle of the argument from illusion that the naïve realist can accommodate the possibility of perceptual error. By appealing to a form of adverbialism, the naïve realist can offer a picture of the determination of phenomenal character that allows them to maintain their commitment to the presentational nature of perceptual experience, and the psychological immediacy of perception.

Adverbialism is an account of the metaphysical nature of experience according to which experience is a mode of consciousness of a subject. It is motivated by a desire to undermine the necessity of explaining experience in act/object terms. The act/object account combines two thoughts. First, that experience is presentational, in the sense that it is an essentially relational affair consisting in a subject sensing some object. Second, that the core phenomenal character of experience, understood as the phenomenal properties of one’s experiential episode, are determined by the objects to which one is related in that experience, in the sense that these properties simply are the properties of said object. The act/object account is, then, a combination view of a presentational understanding of the nature of perceptual experience together with a certain conception of what determines the phenomenal character of such experience. It is this second aspect of the act/object account that is so deadly for the naïve realist. For it leads to acceptance of the phenomenal principle, according to which, if one’s experiences has a certain phenomenal quality, then there must be something that has that quality, to which one stands in the relation of sensing. And if we then recognise the possibility of illusion it seems we are committed to the existence of non-normal objects of sense in at least some perceptual cases. Which, together with the generalising claim, means that we are always aware of such objects whenever we perceive, and hence that naïve realism cannot be the case. According to adverbialism, experience is not an essentially relational state of affairs, but a mode or property of the experiencing subject, and the core phenomenal character of experience is not to be explained by the properties of some object to which one is related, but by the manner in which one senses. And so the phenomenal principle can be rejected and non-normal objects of sense avoided.

It is the aim of this chapter is to argue that there is space for a modified form of adverbialism according to which phenomenal character is constituted, not just by the way in which one senses, but by the way in which one senses an object. The presence of a particular phenomenal quality in experience is explained in terms of the manner in which some object is made present to the mind. So this adverbialism accepts the presentational claim of the act/object account, but rejects its phenomenal claim. Furthermore, this modified adverbialism can be seen as a version of naïve realism that is capable of defending itself against the argument from illusion.

1. The act/object account

I take the act/object account of the nature of experience to be a combination of two thoughts:

(1) Perceptual experience has the following nature: a subject stands in the relation of sensing to some object (the presentational thesis).[1]­­­[2]

(2) This object of experience determines the core phenomenal character of experience, in that the sensory qualities of the experience simply are the qualities of this object (the phenomenal thesis).

Acceptance of the phenomenal thesis entails acceptance of the phenomenal principle:

(3) ‘If there sensibly appears to a subject to be something which possesses a particular sensible quality then there is something of which the subject is aware which does possess that sensible quality.’ (Robinson 1994: 32)

According to the presentational thesis there is some object that is a part of perceptual experience and that is presented to the subject. What is it for experience to be presentational in this way? When enjoying a visual experience an object is before the mind in such a way that it is directly available for demonstrative reference. So the presentational thesis maintains that when we enjoy an experience there is something right there, available for demonstrative reference. Furthermore, this object is directly available for demonstrative reference. That is, it is not the case that this object is only available for demonstrative reference in virtue of the fact that something else is available for demonstrative reference.[3] This is the object of experience. But this does not exhaust the content of the position. The presentational thesis needs to be distinguished from a weaker relational claim that is consistent with there being an object of experience as so far described. This is the claim that the object that one perceives in virtue of having an experience is a necessary condition on the obtaining of that experience. While this is something that the presentational theorist should accept, this is best seen as a consequence of his fundamental view of the nature of experience, which is that there is an object of experience that is itself a constituent of the experience, to which the subject stands in the relation of sensing.

As an example of someone who adopts the weak relational claim without the presentational claim, we can consider an intentionalist for whom the content of perceptual experience is object-dependent. The intentionalist conceives of experience in the following way: it is a representational state that possesses a certain content. This content specifies the way the world must be in order for the experience to be veridical. According to some, this content, or at least some of it, is object-dependent. On this picture perceptual experience of, say, a green cup as a green cup, is an intentional state with the content ‘that is a green cup.’ It then follows that one could not have the very same experience in the absence of that particular green cup, because the content of the experiential state is that that very cup is green.[4]

How, then, does this intentional account of experience differ from the presentational thesis? On the intentional account, what is available through experience for demonstrative reference are things in the world, which are, from the subjective perspective, right there. When we reflect upon experience we find nothing between us and the world and are in a position to pick out and focus upon the things we perceive directly. So perceptual experience, on this intentional view, would have objects of experience in the sense described, and the things that we perceive would be the objects of experience. Furthermore, such experience would be relational, in the sense that it could not obtain in the absence of the things that we perceive. But, crucially, what is demonstrably available to us, through this experience, is not present in experience, but represented in experience. There is nothing in experience for us to demonstratively refer to.

Another way of saying this is that, on the presentational view, the object of experience is a constituent of the experience, whose presence in experience is to be explained by its being literally a part of the experience. On the relational intentional view, on the other hand, the object is not a constituent of the experience in this sense. The content of the experience involves the object in the sense that it is that very object that is represented.

We can now understand what the presentational theorist means by ‘sensing.’ It is simply the name for that relation that holds between subject and some object when that object is available to the subject in the way described above, and this availability is explained by the object being a constituent of the experience.

So far I have explained what a perceptual experience consists in, qua experience, on the act/object account. But what makes it a perceptual experience? Quite simply, the object of experience must be appropriately related to some external (mind-independent) object, such that, it is precisely in virtue of being so related that this external object is seen. Following H. H. Price, we can refer to this relation as ‘belonging to’ (Price 1950). An experience is a perceptual experience of some external object if and only if there is an object of that experience which belongs to that external object.

When a subject sees some external object then that object appears some way to the subject.[5] I use the expression ‘the phenomenal character of experience’ to refer to the way in which a subject is appeared to when perceiving something. So we might say that something appears blue to some subject at some time, and the subject being appeared to in that way (we can call it the ‘blue’ way, or we can say that ‘blueness’ is a phenomenal property of their experience) is a feature of the phenomenal character of their experience at that time. The phenomenal thesis, then, is a claim about why some subject is appeared to in a certain way. The claim is that a subject is appeared to in a certain way because there is an object of experience which constitutes the way that that the subject is appeared to. What is meant by ‘constitute’ here? Well, typically this is taken to mean that the object of experience has, or instantiates, the properties that characterise the way in which the subject is appeared to by some external object.

It is important to note that the phenomenal thesis only applies to what we can think of as the core phenomenal character of experience. Within the range of phenomenal appearances we can discern those appearances that are of what we can refer to as sensory qualities. There are some qualities the appearance of which, if veridical, entails that that which appears to possess such qualities must possess them. So, if something appears red, and this appearance is veridical, in that we are not misperceiving that thing, then that thing must actually be red. On the other hand, if someone appears sad, then it does not mean that they must actually be sad if we are not misperceiving them. Their appearing sad may properly feature in the phenomenal character of an experience, but not in the core of such experience. Sadness, unlike redness, cannot be a sensory quality. The phenomenal thesis is taken to apply only to this kind of phenomenal property.

Let’s put all this together in a concrete example that can serve to illustrate the act/object position. John consciously sees a table, and the table appears brown to him. John enjoys a visual experience of the table, which is to say that John senses some object. What this means is that John is related to some object in such a way that this object is available to him for demonstrative reference and is a constituent of John’s experience. This object is related in some appropriate way to the table, such that in virtue of being related in this way to the table, the experience counts as a perceptual experience of the table (the object of experience belongs to the table). Furthermore, the table appears brown to John. We are to explain this fact in terms of the object of experience actually instantiating this quality.

The phenomenal principle follows straightforwardly from the phenomenal thesis. If a subject S is appeared to in the F way, then there must be something that is F to which S stands in the sensing relation, because to be appeared to in the F way is to sense something that instantiates F-ness. And it is the phenomenal principle, together with the possibility of illusion and/or hallucination, that encourages us to think that we are related to objects in experience that are not the normal objects of perception.

I have presented the act/object view as a combination of two thoughts, the presentational thesis and the phenomenal thesis. The former is a thesis about the metaphysical structure of experience, while the latter is a thesis about the phenomenal character of experience. A crucial question is as to whether these are independent. Can one hold one and reject the other? They are quite frequently run together, and it is often assumed that to adopt a presentational conception of experience is tantamount to viewing the determination of phenomenal character in this way.[6] Before assessing whether this is the case, I shall now present the traditional adverbial response to the act/object view.

2. Adverbialism

If one accepts the phenomenal principle, then one is going to accept, at the very least, that illusions and hallucinations are experiences whose nature is to be explained in terms of a relation between subject and non-physical object. And, if one is persuaded by the argument from illusion, one will think that all experiences are like this. A popular way of avoiding this conclusion is to give an account of the nature of experience that does not adopt an act/object structure.

Intentionalism is one version of this strategy, according to which an experience is an intentional state, and hence an intentional relation to some object, rather than a genuine relation. In a genuine or non-intentional relation, the relata need to exist. It is a feature of intentional states, however, such as beliefs, that their objects need not exist. Adverbialism, like intentionalism, rejects a genuinely relational conception of experience. It does this by providing an account of the determination of phenomenal character that does not explain it in terms of qualities that objects of sense possess.

When x appears blue to S, this need not be explained in terms of S’s standing in the relation of sensing to some object. S’s being appeared to in the ‘blue’ way is not to be understood as awareness of the instantiation of blue, but as a way in which S is conscious. In so doing, support for the phenomenal principle is cut away, and the conclusion of the argument from illusion/hallucination avoided.[7] An analogy is usually drawn here between the adverbial understanding of visual experience and how we understand certain activities such as smiling, or dancing. We may say that someone has a broad smile, or is dancing a waltz, but, fundamentally, to have a broad smile is just to smile in a particular way, and to dance a waltz is just to dance a certain way. Here is Ducasse describing the position:

To sense blue is then to sense bluely, just as to dance the waltz is to dance ‘waltzily’ (i.e., in the manner called ‘to waltz’) to jump a leap is to jump ‘leapily’ (i.e., in the manner called ‘to leap’) etc. Sensing blue, that is to say, is I hold a species of sensing - a specific variety of the sort of activity generically called ‘sensing.’ (Ducasse 1942: 232-233)

As further refinement to the position, one could hold that the difference between, say, sensing redly and sensing bluely, is to be explained in terms of different modes, or determinations, of consciousness that are reliably caused, in normal observers, by distinct properties of external objects.[8]­­­ Perceptual experience, on this traditional adverbialist conception, is a modification of the subject brought about, in an appropriate way, by an external object.

3. Criticisms of adverbialism

3.1 The complexity of experience

The adverbialist contends that experiential statements such as ‘it seems to me as if I see something red’ do not express a relation between the subject and some entity, but are in fact reports of the manner in which the subject senses. The phenomenal character of experience is to be explained by this manner. There are two main objections to this picture, ‘the many property problem’ and ‘the complement objection,’ both of which utilise the complexity of experience to criticise the adverbialist.

The many property problem runs as follows. Firstly, appearances are frequently to be characterised in terms of multiple qualities. That is, appearances are rarely, if ever, is to be characterised by only one phenomenal property. So, if something appears red to some subject it will not just be the case that that thing appears red, rather it will appear both red and square, for example. The objection is that adverbial theories cannot respect this simple fact. Secondly, it is quite common for subjects to be appeared to by several things at any particular time, and these simultaneous appearances may be characterised by incompatible phenomenal properties. Something can appear round to a subject at the same time that something else appears square. Like the issue of multiple properties, it is argued that the adverbialist cannot accommodate this feature of experience either.

If it is true to say that it (phenomenally) appears to Jill as if there is something red and round, then it follows from this that it appears to Jill as if there is something red. The act/object theorist can explain this entailment straightforwardly because they take the state of affairs of it appearing to Jill as if there is something red and round to consist in Jill sensing some object that instantiates redness and roundness. Things are not as simple for the adverbialist, however. They have two options at this point. They can say that the phenomenal character of Jill’s being appeared to is determined by Jill sensing redly and roundly. In which case it follows straightforwardly that Jill senses redly, and thus that it appears to her as if something is red. Alternatively, they can introduce a new adverb ‘red-roundly’ and say that the character of the experience is determined by Jill sensing red-roundly. It must then be taken to be part of the meaning of ‘red-roundly’ that if someone senses red roundly then they sense redly and they sense roundly. But it seems that neither account will do. The adverbialist cannot say that its appearing to Jill as if there is something red and round consists simply in Jill’s sensing redly and roundly. This would leave them without the resources to distinguish between the following two states of affairs: it appearing to Jill as if there is something red and round, and it appearing to Jill as if there is something round and some other thing which is red. In both cases the adverbialist would have to say that Jill senses redly and roundly, collapsing the evident distinction between them. It would seem as if an object is required, in order to accommodate facts of this kind.[9]

In response to this the adverbialist could claim that in the case of it appearing to Jill as if there is something that is both red and round, we should understand Jill’s consciousness as being modified in a way that is distinct from saying that is modified redly and roundly. We should say instead that she senses red-roundly.

I take the central objection to this account that it seems to challenge the intuitive thought that, when it seems to Jill as if there is something both red and round, the redness and the roundness are distinct aspects of her experiential state. But if the character of her experience is to be understood as sensing red-roundly, the apparent distinctness of these elements are lost.[10]

The complement objection is based upon the following principle:

Just as it is not possible for something to be F and non-F at the same time, it is not possible for a person at a given time to V both F-ly and non F-ly. (Jackson 1977: 69)

As it is evidently possible that it can appear to someone as if there is something that is F and as if there is something that is not F, it is not possible to understand this adverbially as it would require us to say that that person senses both F-ly and non F-ly at the same time, violating the above principle. It is far from clear, however, that the above principle is a good one. In particular, it will not be true of those activities for which it is possible to engage in that activity more than once at the same time. Writing is one such activity. If one is ambidextrous it is possible to engage in two acts of writing at the same time.[11] In which case one could write clearly and illegibly at the same time, to the contrary of the above principle.

By distinguishing between acts of sensing, and allowing that a subject can engage in many acts of sensing at the same time, the adverbialist can straightforwardly avoid the complement objection by basically denying that the principle upon which it depends applies to sensings. The failure of this principle allows us to see how the adverbialist can respond, not only to the complement objection, but also to the many property problem. If it appears to Jill as if there is something both red and round, then this is to be understood as Jill sensing redly and roundly, where it is a single act of sensing that is modified by the adverbs ‘redly’ and ‘roundly.’ If, on the other hand, it appears to Jill as if there is something red, and it appears to her as if there is something else that is round, this is to be understood as Jill engaging in two different acts of sensing, one which is modified by the adverb ‘redly’ the other by the adverb ‘roundly.’[12]

It would seem, then, that the adverbialist does have the resources to deal with the complexity of experience, as long as he introduces the notion of acts of sensing that can occur at the same time. We can, however, extract something useful from this debate, in the form of (yet another) constraint which any account of sensory experience must satisfy. What motivates this whole line of objection to the adverbial theory is the thought that experience is complex, in that typically when someone enjoys visual experience they are appeared to by multiple things and in multiple, sometimes incompatible, ways.

This complexity of experience has to be explainable by any adequate account of sensory experience. Whether we take the core of experience to be a sensing, or the presentation of an object, or a representation of the world, the fact that experience is typically complex needs to be explicable. And as we have seen, the adverbial theorist is able to satisfy this constraint if they take sensings to be acts or events that can occur simultaneously.

3.2 The phenomenological problem

The second kinds of criticism are phenomenological in nature. When we reflect upon our experience, it seems to us as if we are reflecting upon objects present in our experience:

We are always aware of this core as the sensing, or sensory awareness, of some item […] and are aware of the sensible qualities involved as aspects or elements of this item. (Foster 2000: 179)

It does not seem as if, in so doing, we are aware of properties or modes of consciousness. If the adverbialist objects to this that it is possible that reflection upon our experience is misleading in this regard, this is possibly too much to accept, given that we never arrive at this through reflection on experience:

To suggest that, in the case of sensory experience, we simply have no procedure for getting a correct introspective view of the basics of the situation, even though we focus on the nature of our experience in the relevantly detached way, know exactly what we are looking for, and are fully aware of all that might mislead – this seems to have no rationale other than that of preserving, at all costs, the adverbialist’s position. (Foster 2000: 181)

In a related objection Mike Martin (1998) argues that our knowledge of the character of experience requires experience to have a subject matter that we can attend to. That is, we can only gain knowledge of what our experience is like ‘through directing one’s attention over the actual or putative objects of awareness’ (Martin 1998: 172) It is only through attending to what appears to us that we can come to know what our experiences are like.

When we reflect upon experience we are confronted with an object, with something that appears to us, and this is essential to our experience having the phenomenal character that it has. The phenomenal properties of experience are not independent of this ‘object of experience’ or ‘subject matter,’ but the adverbialist is committed to claiming that they are. Phenomenal properties, for the adverbialist, are objectless ‘ways of sensing.’

On the adverbial account, experience does not intrinsically have a subject matter, as it is a mode of the subject. It is at odds, then, with the observation that we come to know what experience is like by attending to some such subject matter. The transparency of experience is a point against the adverbial conception of experience.

3.2 The perceptual problem

The final kind of objection is that the adverbial account of the nature of perceptual experience fails to allow for a plausible account of how the world can genuinely appear to us. In perceptual experience things appear to a subject as being certain ways. Any adequate account of perceptual experience needs to explain what it is for something to phenomenally appear a certain way to a subject. If the apple appears red and round to John, then we need to explain what its appearing red and appearing round consists in.

For the adverbialist, what explains the way in which a subject is appeared to is a mode of sensing, a property of the consciousness of the experiencing subject. So the apples appearing red to John, as opposed to blue, consists in differences between the way in which John is sensing. But how can the adverbialist explain thingsappearing to the subject in experience? How can the adverbialist explain how the apple appears to John? The act/object theorist can explain this in terms of the subject taking, or interpreting, what is present in experience as belonging to things in the world.

4. The independence of the presentational thesis and the phenomenal thesis

In many discussions of the nature of experience there is no significant distinction made between the presentational thesis and the phenomenal thesis, and they are often described together. Here, for example, is John Campbell talking about the relational view of experience:

On a Relational View, the qualitative character of the experience is constituted by the qualitative character of the scene perceived. I will argue that only this view, on which experience of an object is a simple relation holding between perceiver and object, can characterise the kind of acquaintance with objects that provides knowledge of reference. (Campbell 2002: 114-115)

Here we have two claims (we can ignore, for our purposes, Campbell’s claims about the necessity of this view of perceptual experience in explaining our epistemic capacities.) One is about the phenomenal character of experience, to the effect that this is ‘constituted’ by the character of the objects perceived. This is what I have been referring to as the phenomenal thesis.[13] The other claim is a metaphysical one, that experience is a ‘simple relation holding between perceiver and object.’ From the rest of what Campbell has to say it would seem that this is the presentational thesis (or something very close to it).[14] The two claims are closely entwined in Campbell’s thinking.

As a second example here is Bill Brewer’s ‘object view’:

The basic idea is that the core subjective character of perceptual experience is given simply by citing the physical object which is its mind-independent direct object. This is what I earlier called the early modern empiricist insight, that perceptual experience should be conceived as a relation between a perceiving subject and the object presented. (Brewer 2008: 171)

The two views that Brewer puts together here are the presentational thesis together with a particular version of the phenomenal thesis which takes a stance on the relation between presented object and object perceived, namely, that they are identical.[15] And again we see the two views being regarded as more or less equivalent.

The thought that I take to be implicitly expressed in both of these passages is that a phenomenal property, such as ‘blueness,’ needs to be explained as the instantiation of this property by some object, if we understand experience to be a presentational state of affairs. But why should acceptance of the presentational thesis and acceptance of the phenomenal thesis go together?

One way of arriving at this position is that presented by Moore (1903). Moore presents the act/object view in this way: we can distinguish the awareness, or sensing of, say, red, and the instance of red that is so sensed. The difference between sensing red and sensing blue is in the nature of the thing that is sensed. For sensing is in common between the two. Given that when something appears blue to one, one’s sensing something cannot feature in an explanation of why that thing appears ‘blue’ as opposed to ‘red,’ the only resources we are left with is to explain this fact in terms of a difference between what one senses. The sensing component of sensing red and sensing blue are the same, and so, given the difference in phenomenal character between the two experiences, this difference must be explained in terms of the object that is present in experience in the two cases. One object instantiates redness, and the other blueness. But why on earth should the ‘sensing’ component (that which is in common between the two cases) be exactly the same in the two cases? What Moore is outlining is, really, a particular understanding of the presentational thesis, and not the only one available. The thought that the contribution of ‘sensing’ to the phenomenal character of experience is the same in all cases, is just an assumption.

A second reason for putting the presentational thesis together with the phenomenal thesis might be that we are first committed to the phenomenal thesis as a result of phenomenological observation, and then committed to the relational nature of experience as a consequence of this. It is clear that the phenomenal thesis entails the presentational thesis. If the phenomenal character of experience is constituted by some object, then experience is presentational, for there must always be some object that is a part of experience and which constitutes its character. On such an approach the presentational thesis would not require the phenomenal thesis, but the evident truth of the phenomenal thesis requires us to accept the presentational thesis, and this explains why they are often run together. The following passage from Price represents the claim that reflection on experience reveals the presence of instances of certain qualities:

When I say ‘This table appears brown to me’ it is quite plain that I am acquainted with an actual instance of brownness […] This cannot indeed be proved, but it is absolutely evident and indubitable. (Price 1950: 63)

But one could object that this is a quite fundamental mistake, and that reflection on experience does not straightforwardly reveal this at all. When we try to refer to the character of our experiences, we can be referring to at least two very different kinds of things. We might be referring to that, which we are aware of, the object of experience, or, we might be referring to our awareness of that object. We can be referring to the sensing, or we can be referring to that which is sensed. Now, according to the phenomenal thesis the sensing does not contribute to the phenomenal character of experience, so when I reflect on an aspect of the character of my experience I must be reflecting on the object of my experience. In which case, when I think about the character of my experience it will just seem obvious to me that I am acquainted with things that have the properties that characterise what my experience is like. But, if we are neutral as to the truth of the phenomenal thesis, it is a possibility that when I focus on this instance of red, I am focusing on the sensing of something, rather than the thing sensed. The line of thinking, then, assumes the truth of the phenomenal thesis and so cannot be used as a way of supporting it.

The final reason for combining the two theses together is that the phenomenal thesis provides the best explanation for the core phenomenal character of experience, given the truth of the presentational thesis. This seems to be the position that Broad, for example, adopts:

When I look at a penny from the side I am certainly aware of something; and it is certainly plausible to hold that this something is elliptical in the same plain sense in which a suitably bent piece of wire, looked at from straight above, is elliptical. If, in fact, nothing elliptical is before my mind, it is very hard to understand why the penny should seem elliptical rather than of any other shape. (Broad 1965b: 89-90)

Here we see that the actual instantiation of a particular shape property is appealed to as the best explanation for why the experience has the character that it does. In which case there is space for alternative explanations of phenomenal character that reject the phenomenal thesis, for the presentational thesis and the phenomenal thesis are independent.

4.1 Filling the space (1): the theory of appearing

Broad himself acknowledges a competing explanation for the character of experience that denies the phenomenal thesis but upholds the presentational thesis. It has been variously referred to as the multiple relation theory (Broad 1965b), the multiple relation theory of appearing (Jackson 1977) and the relational theory (Robinson 1994). Its most common name is, perhaps, the theory of appearing, and I will from now on refer to it as such (see Price 1950; Alston 1999; Langsam 1997). On this account, phenomenal properties of experience are not instantiated in some object, but in a relation between subject and object. This is in sharp contrast to the act/object views of the sense-datum theorists and modern day naïve realists for whom phenomenal properties are intrinsic properties of the objects to which we are related in experience.

Here are some recent articulations of the position:

The theory of appearing […] takes perceptual consciousness to consist, most basically, in the fact that one or more objects appear to the subject as so-and-so, as round, bulgy, blue, jagged etc. (Alston 1999: 182)

And according to Langsam (1997) experiences are ‘relations between material objects and minds’ where this is understood as the claim that ‘phenomenal features [phenomenal properties] are relations between material objects and minds’ (Langsam 1997: 35).

The theory of appearing is, then, an example of a position that endorses the presentational thesis but denies the phenomenal thesis.

The theory of appearing could be used by the naïve realist to explain illusion. The nature of perceptual experience can be thought of as presentational, but the naïve realist can deny that the phenomenal character of experience simply consists in the properties that this object has. Instead, the character of such experience consists in the relation that the object of experience and the subject stand in. It is this relation of ‘appearing’ that gives experience its phenomenal character, and so there is no problem with supposing that the object of experience appears to have properties that it lacks.

The fundamental objection to the theory of appearing, in my mind, is that it leaves phenomenal character unexplained. In virtue of what, exactly, is it the case that the tomato looks red to John? To say that there is no more fundamental explanation than just ‘the tomato stands in the “looks red” relation to John’ is deeply unsatisfying, and indeed, there is a feeling here that nothing has been explained. I do not, now, on being given this explanation, have an understanding of what the difference is between looking red and looking blue, or why there should be such a difference. We are brought back to the point that all parties are going to want to say that when x looks F to S, then S is related to x, and that to say that that is all we can say is not to explain what looking F is.

The theory of appearing is unsatisfactory, then, because it fails to provide us with an explanation of phenomenal character – it does not tell us what makes an experience the experience that it is. We might say, to say that something appears yellow to someone tells us something about the kind of experience that they are enjoying, but as yet says nothing about what it is that makes the experience of that kind.

The theory of appearing is a failure, then, but not because it rejects the phenomenal thesis. It is a failure because the appeal to a relation alone cannot explain, to any degree of satisfaction, the phenomenal character of our experiences.

4.2 Filling the space (2): a modified adverbialism

Adverbialism rejects the relational conception of experience, in order to avoid a conception of experience according to which it is a relation to a non-physical object. But the presentational thesis is only one part of the act/object account. The other is that the core phenomenal character of experience is determined by qualities of the object to which one is related in that experience, in that these qualities constitute that character.

Adverbialism is often presented in response to sense-datum theories, and as opposed to the presentational thesis. But perhaps it is better thought of, first and foremost, as a position that seeks to avoid the phenomenal principle, and the objectionable objects of sense that it brings in its wake as a consequence of the possibility of illusion. And as the phenomenal principle is a consequence of the phenomenal thesis, which is independent of the presentational thesis, there is no need for the adverbialist to deny the latter. What options are there available for the adverbialist who wants to retain the presentational thesis?

An object’s appearing red is to be understood in the following way: as sensing that object redly. But the object need not feature as a contingent element in this sensing, as it does on traditional construals of adverbialism, but could be thought of as a constituent of the sensing. The model to be adopted is not that of ways of dancing, or smiling, which are non-relational activities, but of ways of grasping, pushing or pulling, which are relational and hence require an object. A grasping, pushing or pulling is always a grasping, pushing or pulling of something. Sensing, then, is conceived of on this picture as a relational activity.

I shall use gripping as an example. There is an optimal grip for the human hand depending on the shape of the object to be gripped. Whether or not a grip should be considered optimal or not will be relative to some goal. So for example, to securely grip something in order to minimise the chance of its falling from your grasp, there is an optimal way in which one’s hand must be formed.[16] For different shapes there will be different optimal grips. Let us call the optimal grip for picking up a sphere the ‘spherical’ grip and the optimal grip for picking up a cube the ‘cuboid’ grip. The optimal grip for a sphere and the optimal grip for a cube will be at once similar and different. They are the same in that they are both grips, and, what is more, they are both optimal grips. But the grips have different characters, in that they involve different shapes of the hand around the object gripped.

Now let’s look at particular acts of gripping some object. A subject, Sam, grips a sphere in the optimal way, so Sam grips the sphere in the spherical way, which is to say, the grip Sam has of the sphere is the spherical grip. Now, what is the source of the character of this grip? Is it determined solely by the object gripped? No. The character of the gripping, the way in which Sam grips the spherical object, is clearly not determined solely by the nature of the object gripped, because clearly it is possible for Sam to grip the sphere in the spherical way on one occasion, and in a non-spherical way on another. The way in which Sam goes to grip the sphere, the way in which Sam shapes his hand in going to grip the sphere, contributes to the character of the grip Sam actually takes of that thing. But is the character of Sam’s grip on the sphere independent of the character of the object gripped? No. The character of the grip is clearly shaped and moulded by the actual shape of the thing gripped. The spherical shape of the object partly determines the kind of grip Sam takes of it. The shape of Sam’s hand in moving out to grip the sphere does not survive contact with the sphere, but accommodates, and settles into, its spherical shape.

It is important to note that the object is an essential constituent of the grip. One cannot grip nothing. A failed grip is like a failed push, or a failed touching, in that just as a failed push is not a kind of push, and a failed touching is not a kind of touching, so a failed grip is not any kind of grip. For something to be a gripping, there must be something gripped.

This model can be extended to the sensing of the object of experience. When a subject senses red-ly, this is to be understood as sensing something in the red way. The determination of the character of experience is not to be thought of in terms of the object of experience alone, but in terms of the way in which one senses that object. But neither is the way in which one senses an object independent of that object. The object of experience plays a role in determining the character of the sensing of it. It shapes, or moulds, the act of sensing directed upon it. Like traditional adverbialism, the phenomenal character of experience is to be explained in terms of a mode of sensing, but unlike traditional adverbialism this mode of sensing is in part determined by an object that is sensed. The traditional adverbialist might respond to this by saying that on his view, the mode of sensing is determined by the object that is sensed. On their account, the external object that is perceived causally determines the mode of consciousness.

What is crucially different between the traditional adverbialist account of sensing and this modified adverbialism is that on this latter view the object of sensing is an essential part of the act of sensing, whereas on the former view it stands external to the act of sensing itself, as its cause.

This picture requires us to think of the relationship between the object of experience and the phenomenal properties of experience in a different way to how traditional relational conceptions of experiences conceive of it. The object of experience is a constituent of the experience that is available for demonstrative reference. The phenomenal character of the experience does not, however, consist simply in the presence in experience of the qualities that this object possesses. Rather, a particular phenomenal property is the sensing of this object in a particular way. On this picture, then, experience is presentational, but the phenomenal thesis is not true.

This modified adverbial conception of perceptual experience copes well with the criticisms that are directed at its traditional counterpart. There are three criticisms, each of which exploits the lack of an object in the adverbialist account of the determination of the phenomenal character of perceptual experience. According to the modified adverbialist, the way in which one is appeared to in perceptual experience is determined by the way in which one senses an object. So the issue of complexity is easily accommodated. Any adequate account of phenomenal character must be able to capture the difference between it appearing to someone as if there is something that is both F and G and it appearing to someone as if there is something F, and as if there is something else that is G. The modified adverbialist can respect this because in the former case there is something present in experience that is sensed F-ly and G-ly, while in the latter case there are two things present in experience, one of which is sensed F-ly and the other which is sensed G-ly. In the same way the theorist can accommodate the fact that it can appear to a subject as if something is F at the same time as it appears to that subject that there is something that is not F. In such a case there will be two objects present in the overall experience of the subject. One of these is sensed F-ly while the other is sensed non-F-ly.

The account also preserves the phenomenological data. It can fully allow that we come to know what experience is like only through reflecting upon the objects of experience, because there are such objects, presented to the conscious subject. To the charge that we never become aware of ‘modes of sensing’ when we reflect upon what our experience is like, the theorist can respond that while this was a problem for the traditional adverbialist who maintained that phenomenal character is determined by an objectless mode of consciousness, our being aware of objects of experience is our being aware of the way in which we sense these objects.

Finally, this modified adverbialism does not have the same problem of explaining how we can be perceptually conscious of an external world that faces the traditional adverbialist. There are two possible ways in which they can conceive of this coming about. The first would be to conceive of perceptual consciousness (its appearing to some subject as if there is something with such and such a quality somewhere) as having a sensory core, that determines the phenomenal character of the experience, and which is taken, or interpreted, by the subject. There is no more problem to understanding how this is taken to be of the world than there is for the act/object account, for there is something present in the experience which is, actually, a part of the world. The second would be to think of perceptual consciousness as not consisting of sensory core plus interpretation of this core. Rather, a subject is appeared to in perceptual experience simply because they are sensing some object in the world, where this sensing is to be understood as an engagement with that object, and so could already by thought of as ‘of’ that object. Something appears to the object because something is sensed.

5. The explanation of error

This modified version of adverbialism can be seen as a version of naïve realism that answers the problems posed to it by the argument from illusion, as it rejects the phenomenal thesis that such an argument relies on.

Recall that the presentational theorist is committed to ‘objects of experience’ that are present in experience in the sense that they are available for demonstrative reference and are constituents of experience. When a subject sees an object then that object appears to them, and it does so in virtue of the subject being aware of, or sensing, some object of experience that is appropriately related to the object seen.

Let us understand naïve realism in the following way – it is the claim that the object of experience that one is aware of, or senses, in virtue of which one sees some external object, is that external object. The awareness of the object of experience does not, therefore, fall short of awareness of the object seen. When a subject, Sally, sees an object, lets say an apple, then that apple appears to her. Let us say that one of the ways the apple appears to her is red. What is the explanation we should give of the phenomenal feature, redness, which characterises the way that the apple appears to Sally? The naïve realist claims that the state of affairs that is Sally being appeared to in the red way, is the same state of affairs that is Sally seeing the apple. There is a single state of affairs that we can think of as having two aspects. When we say ‘Sally sees the apple’ then we are reporting what we can call the perceptual aspect of this state of affairs, and when we say ‘Sally is appeared to in the red way,’ we are reporting what we can call the phenomenal aspect of this state of affairs.

The modified adverbialism described above is a version of naïve realism. The phenomenal character of the apple appearing to Sally is to be understood in terms of Sally sensing red-ly. But Sally’s sensing red-ly is the same thing as the apple’s appearing red to her – they are the mental and object centred counterparts of the same state of affairs. The object that Sally senses in the red manner, which is the object of experience, is the red apple. Like the adverbial theory, phenomenal character is determined by the manner of Sally’s sensing. Unlike the adverbial theory, Sally’s experience is presentational, because sensing F-Ly is always sensing some object F-ly.

But now lets take a case of illusion. Sally sees a red apple, but this time the apple appears blue to her. The challenge for the naïve realist is to show how it is possible in such a situation for Sally’s seeing the apple and the apple’s appearing blue to her to be different aspects of the same state of affairs, when the apple is not blue. For a naïve realist wedded to the phenomenal thesis, this is a serious difficulty, because the phenomenal aspect of the state of affairs must be explained only by reference to the object of experience, which, according to the naïve realist, is the object that is seen. But the object that is seen in the illusory case does not have the property that the object of experience has. So how can it be identical with this object? It is plain that for the naïve realist position to be tenable it must be able to explain the source of perceptual error in such a way that it is not located in a discrepancy between properties the object of experience has, and properties the object perceived has.

The modified adverbialism described above can be used to show how this is possible. The source of perceptual error, on this account, is in the sensing of the object. And the sensing of the object is determined both by the object of experience and the way that object is sensed.

The blue manner of sensing is the sensing of something in the blue way. Just as the shape of an object determines the correct grip we must adopt relative to some goal (say in order to pick it up securely), so the colour of an object determines the correct sensing we must employ in order to have the colour of the object available for us in thought and action (i.e. in order to have thoughts about that colour instance and act towards it). Now, just as Sally might grip a round object in the square way, and so make a mistake in the gripping of the object, she can sense a red object in the blue way. The locus of perceptual error is, then, sensing in the wrong way, and not a discrepancy between the properties of the object of experience and the object of perception.

6. Conclusion

I have argued that acceptance of a presentational conception of experience does not bring with it acceptance of the view that the core phenomenal character of such experience is simply to be identified with certain aspects of the character of the object involved in the relation. Once we accept this, it is open to a theorist to reject the latter while maintaining the former. I presented a modified adverbialism that occupies this logical space and respected what I regard as the fundamental insight of adverbialism, namely that phenomenal character can be, at least in part, determined by the way in which one senses. According to this form of adverbialism, experience is the sensing of some object in a particular way, rather than just sensing in a particular way. This modified adverbialism can be regarded as a version of naïve realism. What is more, it is a version of naïve realism that is well placed to accommodate the possibility of perceptual error.

It does not, however, provide the naïve realist with an answer to the argument from hallucination. That will require a different strategy. It is my purpose in this chapter only to outline a defence of a particular conception of what the experience one enjoys when perceiving consists in. The consideration of non-perceptual experience and naïve realism will be the subject of the next two chapters.

[1] Recall that as I am using the term, ‘perceptual experience’ denotes only those experiences that we enjoy when perceiving objects. A hallucination, on this usage, would not be a perceptual experience, but a non-perceptual experience. [2] This should be distinguished from the presentational claim of the naïve realist as described in chapter one. Recall that this is the claim, not only that experience is presentational in nature, but that what is presented is the object of perception. The presentational thesis of the act/object account, as I have presented it, leaves open the nature of the object of experience. [3] For a fuller explanation of this notion of ‘non-dependent’ demonstrative reference see Snowdon (1992). [4] For this kind of view about the nature of experience, see McDowell (1998). [5] Except, perhaps, in cases of non-conscious perception such as blindsight. I am concerned here only with an account of conscious perception. [6] See, for example, Brewer (2008), Foster (2000) and Campbell (2002). [7] Ducasse (1942; 1951) first put forward this particular alternative to the act/object structure. [8] See, for example, Chisholm (1957). [9] See Jackson (1988: 120) and Foster (2000: 173-174). [10] See Jackson (1988: 122). [11] Jackson considers the possibility of someone writing with both his left and right hand at the same time but does not consider this a counterexample to the principle. He does not, however, seem to recognise that there are two different acts of writing going on the same time in such a case (see Jackson 1988: 125-126). [12] See Foster (2000: 175-178) for this solution on behalf of the adverbialist. [13] For example: ‘we have the ordinary notion of a “view,” as when you drag someone up a mountain trail, insisting that he will “enjoy the view.” In this sense, thousands of people might visit the very same spot and enjoy the very same view. You characterise the experience they are having by saying which view they are enjoying. On the relational picture, this is the same thing as describing the phenomenal character of their experiences’ (Campbell 2002: 116). I take this to mean that what one describes, when one describes the view, and what one describes, when one describes one’s experience of that view, are the very same thing, namely, the objects and properties of the scene before one’s eyes. [14] For example: ‘on the relational view […] In the case in which there is a dagger [to which you are consciously attending], the object itself is a constituent of your experience’ (Campbell 2002: 117). And later, he explains that ‘constituency’ must be something other than the object entering into the content of perceptual experience: ‘on the Relational View, experience of objects is a more primitive state than thoughts about objects, which nonetheless reaches all the way to the objects themselves’ (Campbell 2002: 122-123). [15]A further, minor difference is that Brewer takes perceptual experience to be a three place relation between subject, presented object and conditions of perception, whereas I have just been assuming a two place relation between subject and presented object. [16] See, for example, the experimental use of ‘Blake shapes’ in determining efficiency of grasp in test subjects (Goodale and Milner 2005: 24-27).