by Steve Bayne (March 2005)

This paper is the first part of a larger work devoted to examining P. F. Strawson's contributions to philosophy and in particular his work relating to the notion of objectivity understood from a Kantian perspective. Objectivity is an especially fertile area of philosophical inquiry inasmuch as it appropriates traditional epistemological issues to a more analytical enterprise, where by 'analytical' is included an approach to philosophical problems governed in large measure by attention to linguistic nuance and logistical analysis. Presented here in its first rough form is a set of more or less informal remarks, which serve as a prologue to a more complete undertaking. That the remarks are roughly constructed and much in need of further elaboration will become evident to the persistent reader. My objective in this incomplete and unfinished commentary is to bring out the durability of Strawson's ideas in part by pursuing them in connection with more traditional issues such as phenomenalism and more recently the matter of the ontological status of dispositions and their relationship linguistically to the nature of counterfactual and other conditionals. The treatment of conditionals will be perceived as weak no doubt by many who are well apprised of the details of the discussion. This has not deterred me from invoking my own half baked thoughts on the matter, mainly because I will in the second part of this work reject as misleading and innappropriate the application of the concept of a disposition we in analytical philosophy have inherited it from Carnap and Ryle. The tradition is an old one going back as far as Locke, at least, but it is one that obscured the relationship of self ("Apperception") to the external world both in action and in cognition. A situation I hope to move in the "right" direction by these modest efforts. I begin with some exegetical remarks on Strawson's Individuals.
The Identification of Particulars

(1)   The possible topics of public discourse include among other things "objective particulars." The way we think about our world, which consists in all possible topics of discourse, constitutes our conceptual scheme. A conceptual scheme exists independently of any particular language. What a scheme "schematizes," to use a term Strawson does not use, is how we talk about the things we do talk about. We succeed in talking about a thing only if we can identify what it is we are talking about. How we identify a thing relies on certain features of our conceptual scheme. When a speaker makes use of an expression having as its "standard function" that of enabling a hearer to "identify" some particular that speaker has made "identifying reference to that particular.

Making an identifying reference to a particular is not sufficient to enable the hearer to identify that particular. When all goes well both the speaker and the hearer have identified the particular but in different ways. The conditions that allow for agreement between what the speaker makes identifying reference to and what the hearer identifies as the object of the speaker's reference determine the ontological commitments possible within our common conceptual scheme. A particular can be said to exist only if it is possible that it be identified either as what is thought about or as what, for the hearer, is the topic of conversation. There are certain dependency relations between acts of successful identifying reference. Thus I can identify for some hearer a house as one that Jack built only if the hearer can identify Jack.

A dependency of identification of one kind of thing on another is the more general case and tells us something about "the general structure of the conceptual scheme." It indicates an ontological priority of things of one kind on things of another, within a conceptual scheme. Ontology is made dependent on conceptual schemes - not the other way around. Some types of things are more independent than others. This lays the ground for the following question: Is there a kind of particular that is least dependent with respect to identifying reference, and therefore in some scene "basic"?

(2)   Of crucial significance for all of what Strawson has to say about reference is the distinction to be drawn between "story relative" identification and other forms of identifying reference. Consider Strawson's own example: two people are talking and one says to the other "A man and a boy were standing by a fountain. The man had a drink." In this example use of the definite description, "the man," makes reference to the man referred to, if any, by "a man" in the preceding sentence. The hearer can go no further than this in identifying the man, as long as these two sentences are all that have been uttered and there are no gestures etc. that might locate the man in question. In this and all other such cases what we have is "story relative" or, simply, "relative" identification. One consequence is that the hearer can make reference to the man only by placing within his own frame of reference someone else's frame of reference. This means that the hearer cannot in fact identify a particular man. The speaker cannot even be said to have made an identifying reference since no particular individual can be selected on the basis of what he has said. In case the story is true, if identification is to be successful it must be possible at some point to exit what is merely story relative identification.

To be sure, in cases of "demonstrative identification," cases where the object referred to by the speaker is in immediate purview, story relative identification is not an issue. Serious questions arise, however, in cases where the object is not in purview. It might be thought that the use of proper names will enable successful speaker identification, but names require the "backing" of descriptions and so we must be able to make use of descriptions in referring to objects outside our sensory or visual field if we are going to succeed at anything more than story relative identification. But relying on descriptions is problematic because of what is called the "problem of duplication." This is the problem presented by the possibility of there being more than one particular that meets the conditions expressed by the description. For any description of a contingently existing object it is at least possible that there are two or more things that satisfy it. Consequently, there arise doubts about the possibility of making identifying reference. While, practically speaking, there is seldom, if ever, cause for doubt when the description is sufficiently rich in content and where the knowledge of the speaker and hearer is of the right sort, there remains the theoretical problem that in principle such a possibility - that is, of reduplication - does exist.

In order to meet this problem, what must be done according to Strawson (p.9) is link the description being used to identify some particular outside of purview to another particular which can be demonstratively identified. In fact, Strawson intimates that all non-demonstrative identification may depend on there being at some point an element of demonstrative reference. Is there, then, for every particular for which there is no possibility of demonstrative identification some description which will enable a speaker to link that particular to some other particular for which demonstrative identification is possible, thereby making possible identifying reference to particulars outside the speakers (sensory) purview? More importantly, is there a systematic way of constructing, for any such object out of purview, a description which will link it to another particular which the speaker can demonstratively identify? The answer to this question is, according to Strawson, yes.

The universe might be repetitive in various ways. But this fact is no obstacle in principle to supplying descriptions of the kind required. For by demonstrative identification we can determine a common reference point and common axes of spatial direction; and with these at our disposal we have also the theoretical possibility of a description of every other particular in space and time as uniquely related to our reference point. (1959 p. 10).

(3)   That such a solution to the problem of re-duplication exists tells us something about our conceptual scheme. From the hearer's vantage point it is important not to identify the speaker's topic of discourse in merely "story relative" terms. The speaker, like the hearer, requires an individuating fact that is tied ultimately to something besides a story. As a matter of practical fact, however, it would appear that we always come prepared as speakers, or hearers, to provide what we take to be individuating facts relating the topic of discourse to a place. There is often reliance on context, but if Strawson is right it is not only the case that such individuating facts are possible but a matter of necessity. Strawson goes so far as to describe our ability to make such individuating references as a "necessary truth."

It is a necessary truth that any new particular of which we learn is somehow identifyingly connected with the framework, even if only through the occasion and method of our learning of it. Even when the identification is 'story relative,' connection with the framework remains, through the identity of the story teller. (1959 p. 12)
This meets one of two objections entertained by Strawson, in particular the objection that any individuating fact ought to suffice and that, therefore, bringing in spatial coordination, let alone as a necessary fact, is unnecessary. The reply as we see is that spatial coordination allows for a comprehensive individuating framework, one which not only serves to individuate but serves to relate all individuatable items subject to public discourse. But there remains a persistent question: Why must we have such a comprehensive individuative system? Is it not sufficient that for any possible object of discourse there be some individuating fact, never mind that such facts can be coordinatively linked to one another? Why the need for a "unified framework"? In consideration of this question, Strawson describes a three-fold distinction among definite descriptions: logically individuatng descriptions, pure individuating descriptions, and quasi-pure individuating descriptions.

Some descriptions contain an expression announcing the fact that they are uniquely satisfied, descriptions which begin in some such way as the following: 'the only...' or 'the first'. These Strawson calls "logically individuating descriptions." Typically such descriptions will contain either names or demonstratives: 'the only child of Elmer Fudd' or 'the only child in this room'. But there are other logically individuating descriptions that lack such terms, terms like 'the first dog born at sea' or 'the only dog to be born at sea which subsequently saved a monarch's life'. These are "pure individuating descriptions," but notice that they remain a type of logically individuating descriptions.

Descriptions which lack names or demonstratives may fall outside the category of pure individuating descriptions, owing to their dependence on context, such as 'the first boy in the class'. But within this class of context bound individuating descriptions there is the third category of individuating descriptions: the category of 'quasi-pure individuating descriptions'. These are "restricted to what existed before or what exists at the same time as the time of utterance," and here Strawson means descriptions such as 'the tallest man ever' (to take an example of the first case - note here the possibility of introducing 'so far' as in 'the tallest man to ever live so far'. The use of the superlative 'tallest' would appear to make it, too, a logically individuative description, only not purely individuative, owing to the temporal qualification of contemporaneousness with the assertion containing it. But, even were this the case the second objection persists: given such descriptions why is there a need to require as a necessary that they somehow relate their purported referents to a "unified framework" of space and time?

Let us suppose, along with Strawson, that there is the claim on the part of the speaker and the hearer of having agreed on a pure individuating description and agreement, as well, that this description is satisfied. But, now, let us further suppose that there is, contrary to the hypothesis of a unified framework, no way of relating the referent to any spatial coordinates, directly or indirectly. What are we to say in view of our preceding remarks? At some point, disclaiming any ability to relate the thing described to such a framework is going to result in our inferring that the speaker and hearer "have no grounds, except those of general probability, for thinking that the pure individuating description has application at all" (16). For all practical purposes we might add to the conceptual content of the description, thereby minimizing the possibility of reduplication; but, then, there is an increased risk that the description will fail to be satisfied at all. Further, if we should add to the content of the description by reference to additional objects, to be successful would involve us in the very thing that has been disclaimed; that is, we would draw on knowledge of objects which fall within the very framework we abandoned in our disclaimer. Finally, Strawson notes that even if these problems could be solved, complete detachment from the framework would require that any new knowledge of the particular being referred to would require the introduction of new "general truths." These truths must be general because, as has just been noted, the only alternative is drawing from a knowledge of objects in the framework we rejected ex hypothese.

Strawson does not mention Kant at this stage of his discussion, but he does mention that the spatial-temporal system is necessary to our understanding of the nature of experience of the world and our ability to communicate about it. He entertains the question whether it is possible that there be empirical objects which are located nowhere in space or time. Such objects which occur at no distance from us or at no distance from any other objects are virtually inconceivable. There are many issues that one could raise at this point. I will raise only one. Distance is a metrical notion. We measure it by moving about some "rigid body" through space. The space we move it "through" may not possess metrical properties independent of our doing such a thing at least once. But what is to prevent it being argued that such a measurement presupposes re-identification? But if re-identification depends on coordinates and coordinates on a metric, we are caught in a circle. But might we have a spatial world without a concept of distance? Suppose it were possible to characterize a spatial framework in terms of 'closer' (or 'farther'). Would our common concept of distance be displaced or replaced, or what? Distance may be considered a magnitude associated with an answer to such questions as "How much further from x is y than z?" But should magnitudes of this nature be inessential to space, then we must ask whether spatial coordination required of the sort of framework Strawson describes is necessary for the claims he has made. Knowing distance does not require knowledge of where the things at a distance are. Or is it that in a spatial world without a metric of distance objectivity is no longer possible? And here I mean "objective" in the sense that a rigid body is assumed in physics to possess objectivity. If so, then spatiality may be a necessary but, perhaps, insufficient condition of objectivity, 'objectivity' understood in a particular way. But what are the various senses of this term, and which are philosophically solvent in rendering a solution to the Kantian themes Strawson will later examine? We will have occasion to discuss this matter in some detail as it serves to unite Strawson's work in Individuals to his probing inquiry into the nature of objectivity in his later work, The Bounds of Sense.

(4)   There are two uses of 'identify' relevant to Strawson's discussion. Strawson would have us consider the case of a man standing before us with book in hand making some reference to it. I may identify it as the book to which some man before me is making reference, but fail to identify the book as the very copy I bought yesterday. What I have failed to do is reidentify the book, although I have succeeded in identifying it. Keep in mind that I have failed to identify the book as the very copy I bought yesterday. I may in fact, although Strawson does not mention this, succeed in recognizing it as the very same edition of a rare book I myself own. Strawson maintains the position that criteria of reidentification are necessary, such as when I reidentify the book mentioned above as the same copy as the one I own, if we are going to be able to work with a unified spatial framework. That is, I may not reidentify as qualitatively the same a book I may identify as numerically the same as the one a speaker is making reference to. I may in fact reidentify as qualitatively identical a book which is, unbeknownst to me, numerically the same as the one I bought yesterday. Thus to 'identify' may be to qualitatively identify or identify as numerically the same.

If we are unable to reidentify any particular we must operate from moment to moment with a different spatial framework, as if a series of temporal cross sections of some object(s) implied a different spatial framework for each section. There must be a way of relating such frameworks compositionally if we are to make sense even of a changing world. Later, when Strawson discusses Kant's Transcendental Deduction the question arises as to how we are to conceive of objects, if we are to make judgments about them, insofar as we conceive of them as related to one another within a unified spatial framework (Bounds of Sense. p.83). Before proceeding to discuss criteria of reidentification we need to get clearer on the distinction philosophers make between numerical and qualitative identity. Take the sequence 'bacad'. I may say either that the symbol with 'c' to its immediate left is the same symbol as the one with 'c' to its right, or I may say the symbol with a 'd' to its immediate right is the same symbol as that with a 'c' to its immediate left. If both statements are true then in the first instance of 'same' I am concerned with qualitative identity but in the second instance I am concerned with numerical sameness of the letter 'a' in this sequence of letters. Now the question is: what criteria do I employ when I identify something I once saw and then lost sight of as numerically the same as something I now see? More concisely: how do I reidentify a particular through perceptual gaps, where what is at issue is numerical identity?

The skeptic will claim that in cases of discontinuous observation there is no distinction to be made between numerical and qualitative sameness. All I am entitled to say is that the thing I see now is much like the thing I saw before, and even this depends on my ability to recall qualities. Strawson repeats an earlier point: If reidentification of the particular (here we mean establishing numerical identity across an interval of interrupted observaton) were never possible there would be no way of connecting the spatial frameworks across time; there would be no "unified" framework. But there is one other thing to consider in dealing with the skeptic. A doubt about reidentification is possible only if on occasion we reidentify without doubt. Such doubt makes sense only against the backdrop of possible knowledge. In other words one must accept the existence of the idea of a unified framework in order to doubt purported reidentification "across" such an interruption of observation. The alternative left to the skeptic is to resort to the revisionary metaphysical move of proposing an alternative framework explaining all the facts presented to us.


Given that the identification of particulars is central to the conceptual scheme with which we operate, the question becomes one of whether there is any one class of particulars that is fundamental within the scheme. Strawson's answer is that there is such a class, and that this class consists in material objects, rather than, say events, processes, states or other entities. Seeing how the spatial/temporal framework has figured in previous discussion, this comes as no surprise, but an understanding of why material objects, the existence of which many philosophers have rejected altogether, should occupy such a privileged position is worth pursuing. First, let's examine the conditions Strawson lays down for determining whether there is such a privileged class.

If there should be a class of particulars such that reference to at least one of its members is necessary in order to identify a particular from any other class, but where members of this first class can be identified without reference to members of any other class, then the memebers of that class are "basic" according to Strawson ( 1959. p. 28). The only objects in the spatio-temporal framework that fit the bill are material objects, which have the property not only of occupying space but which have endurance and resistance to touch. The main idea comes down to this: there is a class of particulars upon which the identification of all other particulars ultimately depends. Strawson discusses other sorts of particulars to illustrate his point, although as we shall soon discover he also discusses other cases, taking up objections to his own position on a number of occasions.

It is only occasionally in his discussion of basic particulars that Strawson draws a distinction or makes an important point of immediate consequence to his discussion of the self, or "persons." It is while fleshing out the details of his views on the "basicness," for want of a more descriptive term, that Strawson makes such a point. We shall see in a later chapter that Strawson relies on a distinction between so called "M" predicates and "P" predicates (p. 100). Related to this is the distinction between events and processes, on the one hand, and material bodies on the other. What is important to closely attend to is the distinction Strawson makes within the category of events: there is the category of events or processes that are necessarily "of" material objects and, then, there is the category which not so dependent, e.g. "bangs" (sudden loud noises of a certain acoustic quality).

Thus a death is necessarily the death of some creature. But that a flash or a bang occurred does not entail that anything flashed or banged. 'Let there be light' does not mean 'Let something shine'. (Strawson 1959)
Suppose we take a banging sound as a phenomenal property or as what Strawson described as a "pure sense data experience" (Bounds of Sense). What we first need to note is that we are talking here about something in time, something which is in some sense a particular, as long, of course, as we can speak intelligibly of particular bangs. Such a view may create controversy. If it is correct to understand Strawson as saying that particular bangs are examples of elements of "pure sense-data experience" and as such are properly speaking "phenomenal," then a controversy might arise with a claim made by Rorty:
...the only way to identify the phenomenal with the immaterial is to hypostatize universals and think of them as particulars rather than abstractions from particulars ... (Philosophy and the Mirror of Nature Princeton. 1979. p. 31)
Now it would appear that if only we are thinking of a bang as an event, but of such a kind that no corresponding abstraction can even conceivably be made from particulars; the bang *is* a particular and is not thought of as an aspect of its causal source. What is crucially at issue in this section of Strawson's essay on Bodies is an account of the primacy or "basicness" of material particulars over other categories of objects such as events and processes. The conclusion he will arrive at is that while there are special cases where sets of events, such as flashes and bangs, are such that members can be identified independently of reference to objects of other kinds, this is only in an attenuated sense; that is, this applies only to certain special circumstances, constrained by the particular speaker-hearer situation, where at least one object at issue is directly identifiable and the rest are identified in relation to it by counting. By the "rest" I mean those in the set(s) not in purview. Even in such special cases, however, there is no framework within which all objects of the kind at issue are identifiable. A closer examination of the problem of the dependence of identification is worth pursuing.

One example of identificatory dependence is the dependence of identifying a pain on identifying the person whose pain it is. By contrast, there are ways of identifying a tree without identifying the person who claims to see it. The case of private particulars is complemented by the case of unobservable particulars of physical theory, which, like private particulars "exhibit the most direct kind of identifiability dependence" (Strawson 1959 p. 34).

Keep in mind for present purposes that although it is a necessary condition for a class of particulars being basic that members of that class can be identified without identifying members of any other class of particulars, it is not sufficient for being basic that a member of it, or members be likewise independent. While some member of the class may be directly in purview, and non-mediately identified, there may be other members, falling outside our purview, which depend on particulars of another kind for their identification; and this is what disqualifies them being "basic particulars." Thus I may be in a position to identify to some hearer a bang we both hear, say; and in this instance, of course, identification is not dependent on identifying reference to particulars that are not bangs, but other bangs may require for their identification (that is, making identifying reference to them) appeal to their causes or locations, locations that fall outside of purview.

To be sure, if a set of alternating flashes and bangs within ear shot of both speaker and hearer were arranged in a series and assigned a a position based on order within the series, then no particular of any other type would be required. In this case what we would have is what Stawson calls a directly locatable series." (p. 37) But what about cases where, say, only one of such a series is within ear shot of the speaker and hearer? We might rely on the fact of there being a series originating with the directly locatable element. So we might be able to say that such and such a bang was the nth bang from the first following the first bang, which was directly locatable. While this may work with some bang sequences, as a practical matter not all bangs are related to a single series; nor is any such series accessible to all speakers and hearers. This may be a practical problem only but it is one which "determines the nature of our identifying reference to flashes and bangs." (ibid) This is his reply to other such cases, including the sequence of deaths and battles (cf. p. 39)Still there appear to be some sequences of events which are such that one member can be directly identified and whic is such that all other members can be identified without reference to particulars of a different sort. One example Strawson discusses is the sequence of nights and days. Unlike any sequence of flashes and bangs, this sequence is common to all of us.

Strawson is careful not to be swayed too easily to his own position by arguments to the effect that because a class of particulars may entail another - for example births entail particular animals born - that the entailed particulars must be basic. Take the case of births. In order for there to be births there must be animals born; but, if this is so, then, it may be reasonably argued that animals are more basic than births. One must also consider that births being events and animals material particulars that the dependency here is not, as he requires, asymmetrical - asymmetrical in the desired sense that identifying the events, births, depends on identifying the material particulars, animals. His opponent may argue that we can identify screams without identifying the screamers. It is in discussing this case that Strawson appears to emend his position, saying:

The original argument errs in trying to infer from a conceptual dependency too direct a kind of identifiability dependence of particulars.

The argument may be replaced by one with a weaker conclusion. Suppose bs are necessarily bs of as (e.g. that births are necessarily births of animals). Then, though on a particular occasion I may identify a particular b without identifying the a it is of, yet it would not in general be possible to identify bs unless it were in general possible to identify as. (1959 p. 41)

There is one objection to be met here, one which Strawson take seriously. We have observed that dependence requires asymmetry but in moving to this more general characterization we run risk of losing it. For now it would appear that in general not only do births require being the births of animals (material particulars) but animals, similarly, depend on being born (events). In this case it can't be argued that material particulars are independent in a way events, births, are not. But this is something Strawson denies. What Strawson claims is this: while our conception of a birth requires the notion of animals which are born, the converse does not hold. That is, we can conceivably retain our concept of what it is to be an animal while not retaining a conception of them that requires the concept of a birth. (Strawson 1959 p. 43) Let's take a closer look at this entire argument.

If there are basic particulars, as Strawson wants to argue there are, then there is an asymmetry in identifiability: the nonbasic particulars must depend on the basic ones within the context of the identification game. But in the case at issue, it would appear that while births depend on animals, 'This birth entails there is some animal of which this is the birth'; animals in turn depend on births, 'This is an animal entails There is some birth which is the birth of this animal'. It would appear, then, that the required asymmetry is lacking. However, Strawson recovers the asymmetry and thereby rescues the notion that animals, that is a sort of material object, is more fundamental than events, such as the births of animals. He achieves this by pointing out an asymmetry with respect to the entailment of existence statements.

In order to see the asymmetry consider the two sentences at issue:

A. This is a birth entails there is an animal of which this the birth.

B. This is an animal entails there is some birth which is the birth of this animal.

The point Strawson makes is that whereas it is possible to paraphrase (B) as (B'):

B'. This is an animal entails this was born

It is not similarly possible to paraphrase (A) as

A'. This is a birth entails this is an animal of which this is a birth.

What Stawson takes this to show is that the existential statement in the case of (B) ("There is some birth...") can be paraphrased out; there is no way of doing the same with the existential statement in (A) ("There is an animal..."). In this way, the asymmetry we thought eluded us has been reinstated, and with it the idea of animals as basic to events such as births; rather than the other way around. This dependence of event-particulars on material objects for their identification can be repeated with processes, events and other types of particulars. What may have been forgotten and what now has to be restated is the dependence of the framework of identification supplied by material objects on the notion of re-identification, something even more fundamental. Put somewhat awkwardly the main idea can be conveyed, perhaps best, this way: the dependence of identification of all other particulars on identifying reference to material particulars entails the re-identification of material particulars. Whereas identification of events or processes, for example, depends on identification of material particulars, so too does re-identification of events or processes depend on re-identification of material particulars.

If, for example, we take any familiar process name, such as 'thaw' or 'battle', we shall find it impossible to give a detailed account of means of identifying a particular process of the kind concerned as the same again, which do not involve any reference to some material bodies or other...(p. 46).
It is absolutely fundamental to Strawson's position that only material bodies can constitute the spatio-temporal framework which is required by our conceptual scheme. The fundamentality of material particulars, their "basicness," is argued for more or less independently of arguments for the primacy of the spatio-temporal framework. If it should turn out that, as Russell, among others, have proposed that the spatio-temporal framework is defined in terms of events, rather than material objects, then the distinction of ontological basicness and identificatory basicness would have to be more sharply delineated (cf. p. 50). So far, we have seen powerful arguments adduced by Strawson to show that events depend for their identification on material objects, and not the other way round. One other point is worth mentioning in this regard. The reidentification of material objects recognizes a distinction to be made between numerical and qualitative sameness. Events (processes etc.) are such that such a distinction is difficult or impossible to make, as no reidentification of numerically the same event across time is possible. This opens up for discussion a number of topics which, aside from one I will dispatch immediately, will simply be passed over. That one topic has to do with the Strawson's recognition that some have maintained that ordinary particulars can be construed as extended processes. His response to this suggestion is to point out that this is not the conceptual system we have in fact but one which has been recommended we substitute for the one we have. What Strawson is saying is that such a proposal constitutes revisionary metaphysics and not the sort of descriptive metaphysics that he has announced as the object of his inquiry.


Material objects have been shown to be basic in the sense that identifying or re-identifying objects of other categories is either directly or indirectly dependent on identifying or re-identifying material bodies. In addition, the fundamentalality of material objects is owing mainly to the possibility of public knowledge of them as objective, that is, as capable of existing unperceived. Without the possibility of re-identification such objectivity is an impossibility. If there were objects that allowed for the possibility of objective knowledge, which were not material, then there would be reason to argue against the privileged place in our conceptual scheme material objects appear to occupy. Whether being primary in the conceptual scheme amounts to ontological primacy is not a necessary feature of Strawson's account as

The meaning given to the term 'basic' is strictly in terms of particular identification. ( 1959 p. 51)
The question that lies at the center of Strawson's examination of sounds is whether one can make intelligible the idea of objective knowledge within a conceptual scheme where material objects are not basic particulars. Keep in mind that the category of material objects is basic to our scheme because identification of objects of other kinds depend on identifying thoughts about them, and not vice versa. How are we to understand the question, unless we come to some general idea as to what it is to be an object of objective knowledge although that object is not itself a material object? What general condition do we want to impose on such an object, which also, one might add includes the material objects of our actual conceptual scheme. Strawson is careful to point out that such objects are those "distinguished by the thinker from himself and from his own experiences or states of mind, and regarded as actual or possible objects of experience" (1959 p. 53).

When Strawson speaks of distinguishing objects from the experiencer of those objects it is suggestive of his proximity to "the tradition"; for example, when he invokes such facts as the difference between a tree being struck by lightening and seeing the tree being struck by lightening. The importance of this will be forthcoming when we undertake a closer examination of his examination of Kant, but even now a few related matters ought to be called to mind. First, material particulars are not events, whereas being struck by lightening is; so, we must ask about the objectivity of events vs. material objects, for example. For if there are significant differences, particularly as pertaining to identification and re-identification, this could become a significant issue. Second, "objectivity" inasmuch as it may mean 'independence of perception' may retain a far narrower construal than is required to thwart the claims of certain idealists, that is, those in particular who may maintain that independence of perceptions of one self may be insufficient to establish the independence of objects from all perception. One is to think of Berkeley's notion of an omniscient Deity. Such an idea, while open to skepticism on the part of those with a "robust sense of reality," is methodologically sound, insofar as all that appears to be required of objectivity is some relation enjoyed by the oject independent of the perceptual relation to a possible knower. What this points to is that the idea of re-identification and the idea of objectivity as entailing a denial of the thesis that esse est percipi, at the very least, ought to be questioned, as we shall soon discover when we turn to critics such as Don Locke who makes a very closely related point. Finally, the notion of the self (and ultimately persons) enters the picture. This is not surprising, given Kant's special employment of the notion in relation to what it is to be an object, and Strawson's connection to Kant. However, what must be more thoroughly examined, before accepting much of Strawson's program, is the relation between what makes a kind of particular "basic" (its relation to space and time) and that upon which our conception of objectivity depends. Material objects are basic for Strawson precisely because they are spatial, as space is required for re-identification to make sense. But now the question is: are there nonspatial schemes possible that would make room for objective particulars?

If we are allowed to suppose a world without space, objectivity will be possible on the account Strawson has provided only if re-identification is somehow possible in such a world. It is this feature that allows us to distinguish ourselves and states of ourselves from the rest of the world. Stawson proposes that we consider an auditory world, that is, a world of sounds. Such a world provides a certain richness absent from other worlds we might try to imagine, such as worlds consisting of smells or tastes. This is so insofar as we can arrange sounds along "dimensions" of pitch, timbre, and loudness: all are degree concepts and taken together provide the raw material for imagining relationships that don't obtain in worlds consisting solely of other sensory objects. But if Strawson is to test for the possibility of objectivity in a sound world he must convince us that the auditory world, if it is a no space world, lacks objectivity. This he does by considering certain properties we consider spatial in terms of relation to sounds, such as distance and direction. He maintains that such properties do not belong intrinsically to sound but are, rather, introduced by association with other objects of our awareness, such as touch and kinesthetic experience - which may yield primitive but real spatial notions. When a particular is re-identified, numerical identity with an earlier appearance of the object is a requirement. We want to know whether this requirement is met in a world consisting of sounds. Strawson lumps together the notions of 'public object' and 'objective particular' (1959 p. 58). What he doesn't discuss is the fact that in the case of a public object a number of observers may independently identify the object even though the object is not re-identified as such. In other words multiple identification does not constitute re-identification. This is something I shall not pursue except to mention it is passing, but even in the context of more traditional sense data philosophy this difference is not always made explicit. For the purpose at hand what is important to note is that in order for two persons to be experiencing the numerically same sound requires not only sameness of "surroundings" but also that "that the causal source" of the relevant auditory experiences should be the same for all of them. At first it seems somewhat surprising that Strawson does not regard sameness of causal source as equally difficult to satisfy as the condition of sameness of surroundings.

It is almost certainly the case that what explains Strawson's relative disinterest in the causal condition is that causes constitute part of the surroundings. If the condition of sameness of surroundings cannot be expressed in terms of sounds, then the causal condition will almost certainly, likewise, remain problematic. In particular, the surroundings include other persons, and to be consistent other persons would have to be given some presence in auditory terms. Yet, if the surroundings, themselves, consist in objective particulars, then we are presupposing in the stated conditions for there being objective particulars the very things we are attempting to account for in this no space world. At this stage of the discussion, Strawson does two things. First, he elaborates on what would be required in order to give objectivity a sense, where it is considered as relying on other persons; and, second, he moves away from this line of approach and takes up another, one not immediately related to the subject of other persons. I will take these up in turn.

First, rather than considering the circumstances and conceptual basis for our saying that more than one person experiences numerically the same particular the issue is brought to the level of thought and not merely cognition. The level of thought is most evidently present in discourse and takes the form of requiring in order for there to be objective particulars in an all sound - no space - world that we are in possession of "the idea of public auditory objects which are also topics of discourse between beings who hear them" (p. 60). Strawson backs off, at least for a time, pursuing this line of thought. But notice that other minds, or more especially other discourse using agents, are retained with the presumption that in a no space world sense can be made of there being other minds; while dismissing a deeper discussion of causation. This will, as we shall see, a fundamental concern for Gareth Evans in his discussion of Strawson's proposals. Second, Strawson moves away from the notion of 'objectivity' in the sense of 'public' and back towards an earlier conception. Instead of speaking of what is objective in terms of what any one of a number of persons could in principle attend to, Strawson returns to the idea of objectivity as a concept belonging to the understanding of a being which can make use of the distinction between himself and his states, on the one hand, and everything else, on the other. A creature able to make use of this distinction will occupy a world in which the "conditions of a nonsolipsistic consciousness" are fulfilled. Given the importance of this characterization, Strawson's own words are worth quoting:

...I shall mean by a non-solipsistic consciousness, the consciousness of a being who has use for the distinction between himself and his states on the one hand, and something not himself or a state of himself, of which he has experience, on the other..(1959 p. 61)
Later (Op. Cit. p. 66) Strawson will describe the position of the "true solipsist" as someone who merely has no use for the distinction between himself and anything else. This, presumably, distinguishes him from the philosophical solipsist who believes everything is himself or a state of himself. Note that the person who entertains a doubt whether there is a world after his death by this criterion is in a philosophical mode, although one suspects that the question occurs naturally and commonsensically to most thoughtful persons. This consideration does not, however, blur the distinction between the "true solipsist" and the philosophical solipsist, for if a person believes that his actions have moral consequences after his death the question of objectivity in the philosophical sense is addressed, notwithstanding what may be a very commonsensical earthly view; and this view can only be held by someone who takes seriously the philosophical question of solipsism and does not merely live through life's experiences committed to "true" solipsism. In other words, the "true solipsist" is neither the philosophical solipsist nor the man of common sense. We have distinguished identification and re-identification. It behooves us to take each into account in understanding the conceptual resources allowed us in a no space, that is, sound, world.

In order to identify a particular, we must distinguish it from others. In order to re-identify a particular we must identify a previously identified particular, again, as such. Suppose within a certain temporal duration I hear a sound, and then it goes away. Suppose within another, nonoverlapping, duration I hear a sound; it goes away, and then a qualitatively indistinguishable sound begins and then stops. On the first occasion one sound is identified; on the second, two sounds. The criterion of identification, that is of distinguishing, different sounds is what Strawson calls "the criterion of interruption" (Op. Cit. p. 62). This sets the stage for raising the question of re-identification in such a sound world.

The important question is not just whether sounds of the same sort can be re-identified but whether literally the same, that is numerically the same, particular sound can be re-identified. Strawson presents the following suggestion for the possibility of re-identification in a sound world. We are to suppose an organized piece of music (M), say a symphony in four movements (A,B,C,D). Imagine that I am familiar with the structure of the piece and that I am hearing A but, then, a noise much louder than A is heard and I no longer hear A. Later, the loud sound disappears and, as it does, I begin to hear D. Compare this case to one where everything else is the same except there is no loud sound, A simply is no longer heard and later D begins to be heard. The contrast suggests this: that in the first case A and D belong to numerically the same piece, that is, M; whereas in the second case A and D may be parts of the same piece of music, that is the same universal; but they are not members of numerically the same piece of music. In the first case a particular instance of M has been re-identified; but in the second only the same universal M has been identified.

The situation where a louder sound seems to drown out another sound, suggesting that the sound drowned out still exists albeit unheard, has not been shown to be a possibility in the no space world of auditory particulars. If, however, we are in possession of the idea of a nonsolipsitic world within this imagined auditory universe it follows, says Strawson that there could be re-identified particulars in such a world. Before considering whether the entailment works the other way; that is, that if we have the idea of re-identifiable particulars in such a world then we also satisfy the conditions for a non-solipsistic consciousness, let us consider an alternative formulation of what the idea of a non-solipsistic world must consist in. If we think of a mind as in some sense composed of mental "actions," then whether a particular can exist independently of one action of a sort (cognitive) is not going to entail that the world exists independently of minds. There may be other mental acts aside from the cognitive ones which may be necessary for objects. Further, if the mind is more than a structure consisting of acts there may be an aspect of the mind essential to the being of objects, perceived or otherwise unrelated to the status of those objects as intentions of mental acts. I am not defending this position but only pointing out that the conditions for a non-solipsistic consciousness may require something other than reidentifiability. Unless I am convinced that there will exist a world when I am gone, I am not convinced that the world in which I exist is non-solipsistic. My ability to re-identify a particular may not be sufficient to make that determination. Strawson does not discuss this matter. Indeed, in a world without me re-identification as so far described is not possible; but we certainly would not want to say that this proves the case for solipsism although it may suggest the possibility that the existence of other minds is, at least, as fundamental to an objective world as re-identification. At a later stage in our discussion we shall discover the importance of the self in the discussion - in particular when we examine Strawson's view of Kant. It is also worth considering at this juncture that what Strawson has been calling the "true solipsist" most likely lacks a conception of himself, insofar as no distinction between himself and all else falls within his understanding. For now it should merely be noted that unless the self is a continuant the notion of a re-identifiable particular is suspect; for if the self is not a continuant we are hard put to define the sense in which such a self can know of an object that it is a continuant. However, this does not mean that the idea of a particular existing unperceived presupposes the idea of a non-solipsistic self. Something we shall shortly discuss. We now proceed to examine the relation, according to Strawson, between re-identification and an objective world.

The possibility of objective particulars in a no space world presents a challenge only if there are certain features of space that provide for the possibility of re-identification which a no space world lacks. What must be shown if the argument for the possibility of re-identification in a no space world is to be successful, is that something takes the place space takes in our world, a world where although an object may not exist in view it may exist outside our purview in space. As Strawson puts it

...the crucial idea for us is that of a spatial system of objects, through which oneself, another object, moves, but which extends beyond the limits of one's observation at any moment, or, more generally, is never fully revealed to observation at any moment. This idea obviously supplies the necessary non-temporal dimension for, so to speak, the housing of the objects which are held to exist continuously, though unobserved; it supplies the dimension for objects which are not themselves spatial, such as sounds, as well as for objects that are. (1959 p. 66)
Our sound world is a no space world; but what we need for purposes of re-identification is something at least analogous to space in respect of the role space plays in "housing" objects unperceived. It should be noted that what is important is not that the self which is unable to perceive the object - and here I am speaking of our actual world, not the no space world of sound - occupy a position in the same space as is occupied by the object. The simplest hypothesis is that in cases of re-identification, when the object returns within purview of the observer, that it occupies a position continuous with other positions that lie outside the visual field in which it was "housed" while not in visual range. What is not essential is that the space of the visual field be continuous with the space which "surrounds" it, which is what Strawson is suggesting. The reasoning behind the basic idea of re-identification does not require space at all; it requires only some warrant for the claim that an object persists unperceived, even if only - as Berkeley suggested - within the awareness of God. The rationale for introducing something analogous to space is that in the common sense world space can be made out to provide the necessary "housing" for unperceived objects. The continuity of space serving as well to account for the continuous existence of the object while unperceived. Indeed, one might argue that continuous existence is not required for re-identification. All that is required for re-identification, it may be argued, is that the same object identified earlier is re-identified at a later date; but in this case it is difficult to be entirely convinced that re-identification would guarantee objectivity. It is important in this regard to keep in mind that objectivity is assumed to obtain in the real world. Strawson is not engaging in debates over realism vs. phenomenalism. He is doing what he calls "descriptive metaphysics," which sets out to describe the world without introducing metaphysical schemes that add to or revise our common conception of it. But he must walk a fine line, as objectivity is tied essentially to the denial of the claims of "revisionary" metaphysicians who make use of the idea that "esse est percipi" and thereby align themselves with phenomenalism, as we shall discover at a crucial point in Strawson's discussion of Kant's philosophy.


Sounds are said to possess three dimensions: pitch, timbre, and loudness. Space is usually thought of as continuous - although this is not beyond scientific question. Continuity is often described in terms of a serial order. Pitch and loudness can be described in terms of a serial ordering of sounds, and so the analogy to space will involve two dimensions. We are to imagine a sound which is constant in timbre and pitch, but varying in pitch. This Strawson calls the "master sound." It is always audible. There are other sounds of varying complexity. Some may determine "unitary sound sequences" just as did the individual movements of a symphony, for example, earlier in our discussion. Suppose we hear a sound of a certain pitch in addition to the master sound. Imagine now that the pitch of the master sound gradually changes and as it does the other sound fades away and another sound begins to take its place. This might be considered as somewhat analogous to walking away from an object in purview and gradually seeing something replace it at the center of one's visual field. Suppose the pitch gradually changes back to what it was before and the new sound is replaced by one qualitatively like the one we heard before. Let's now consider a case of sound re-identification where numerically the same sound is what is re-identified.

One hears a sound of constant pitch, the master sound. One hears in addition another sound of a constant pitch, although this is not necessary. The master sound, begins to vary in pitch, slowly. The other sound begins to fade and soon the only sound of which I am aware, aside from the master pitch, is a different sound altogether. But next suppose the master sound begins to change pitch, returning to what it was before and as it does a qualitatively identical pitch to the one were originally heard is now heard again. This sound we say has been re-identified. Suppose the question is raised as to whether what we hear is actually the same sound or is merely a qualitatively similar sound. How do we know we have numerical identity and not merely qualitative identity. In a real space world this would present little difficulty, once the assumption is made that space individuates objects numerically; that is, that given qualitatively identical objects spatial coordinates will tell them apart. But in the sound world how is this to be achieved? The answer is cleverly constructed. Strawson tells us that if we are dealing with two qualitatively identical sounds, but in one case the master sound is different, whereas in the other the master sound has not changed pitch at all, then in the first instance re-identification is merely qualitative. If the master sound is the same then we have not merely qualitative sameness but numerical identity and, so, re-identification is of a particular and not of a sort or kind; it is particular rather than universal. We see, then, that if Strawson's analogy holds we have a sound world where the master sound supplies something analogous to spatial coordinates. How the master sound provides coordinates supplies the basis for one strong objection to the analogy, an objection Strawson discusses and Gareth Evans takes up at some length. Neither, however, gives full measure to the roots the objection has in more traditional discussions in the theory of knowledge.

Suppose an object pops out of existence when it becomes no longer perceived and exists only when it is perceived. At first, it may be thought that this is sheer nonsense. In fact, however, sounds have sometimes been thought to be such things; there has been, since Berkeley, the question of whether a tree falling in the forest makes a sound. What causes there to be an object to be seen may involve more than the experiencing of the object. Let us suppose, then, that as philosophical as it sounds the question is legitimate and that it is at least possible that an object ceases to exist when it is not perceived and begins to exist only when it is perceived. This introduces the possibility that I may re-identify an object which passed out of existence the last time I ceased to see it and came back into being as I see it now. What this suggests is that re-identification does not entail the notion of the existence of unperceived particulars. In this paper, I am going to ignore the possible implications this may have for such matters as quantified modal logic, which is, nonetheless, relevant.

The idea of space provides a prophylactic against serious practical consideration of such a possibility. Supposing that there is a common space occupied by an observer and an object that "houses" it. There is reason to believe that that object is protected from necessary extinction by its failure to be perceived: an object may exist at a point in space outside my visual awareness simultaneous with the existence of the observer. That is, there are spatial relations between observer and object even when the object is not perceived. As Strawson puts it:

But surely the idea of simultaneous existence of the perceived and the unperceived is linked with this idea of the simultaneous presentation of elements, each of a definite character, but simultaneously exhibiting a system of relations over and above those which arise from the definite character of each. (p. 73)
What is important here is that a visual space world, consisting of colours ordered in the dimensions of hue, brightness, and saturation differs from the world of sounds described by Strawson in the following respect: in such a sound world spatial relations between particulars, such as distance, cannot be known at a single instant. Only over time with the "swelling or fading" of the master sound can one become aware of relations analogous to the spatial relation of, say distance. This consideration may cause some to abandon the no space world of sounds.

There is a sense in which in such a sound world all sounds now given to me are at the same place, whereas not all color given to me in a color world are likewise at the same place. While this asymmetry is important in deciding the ultimate value of the analogy, there is much to be gained by pursuing the larger issue of re-identification and its relation to the possibility of objective particulars in a sound world. There is one important difference between spaces associated with particular senses, visual or auditory, and the space of ordinary life we commonly think of as physical space. The existence of space itself, outside our purview, does not - in the case of physical space - depend on unperceived particulars. But there is in the case of space defined in terms of sensory particulars, that is color patches or sounds, a question whether space exists at all outside the range of our sensory awareness. More especially in this case, the existence of physical space does not depend on the existence of physical objects. I can imagine empty physical space, even a vacuum may suffice. What I cannot imagine is sound space in the absence of sounds. Further, I cannot imagine a physical object that does not "take up" physical space; whereas I can imagine a sound that does not "take up" sound space, at least sound space as described by Strawson. These considerations are fundamental to the problem of objectivity inasmuch as it affords us the opportunity of distinguishing perceived space and the chunk of space determined by the relations between physical objects within our purview. Such a distinction is impossible in speaking of sensory spaces, whether auditory or visual. Let's briefly review the situation up to this point.

The question we began with was whether a being having only information available to him which is available in a no space world can have the idea of an objective particular, where having such an idea amounts to having the idea of a non-solipsistic consciousness. The analogy of sound space was introduced to test whether objectivity in this sense is possible in a no space world. Objectivity depended on the possibility of re-identification, and so the question of re-identification in such a no space world became paramount. Re-identification of sound particulars was understood to entail the existence of unperceived particulars within the auditory, no space, world. Whether re-identification entails a non-solipsistic consciousness is left undecided (p. 74). It would at first appear that it does, however to say that it does demands that the observer be able to distinguish himself from all else. But now a certain difficulty arises having to do with the observer in such a world.

Recall the point made earlier that re-identification required the idea of a self which is, also, a continuant? If the world we are considering is sound world, what then is the self? The observer must be aware of himself as existing at different places. The difference between an observer and an unobserved particular is entailed by a non-solipsistic consciousness. But how is an observer to conceive himself at different places in a world consisting entirely of sounds. This is question closely connected to the relation of apperception and objectivity in Kant's philosophy. Strawson's approach to this problem is far more simple in Individuals than in his later work, The Bounds of Sense, which we shall examine with some care on a later occasion. In Individuals the procedure is to reject the notion that the word 'I' has referential force. Moreover, we adopt a rule as observers in such an auditory world that "we are make use of any concepts which derive their function from the fact that this special part of our experience is in fact integrated with our experience at large." While Strawson expresses employing "use of" in the context of writing a report (of a special part of our experience), there are deep implications for the Kantian concept of synthesis. This is evidenced by his rejection of the notion of a Kantian in its entirety while accepting a complex view of the unity of consciousness against the back drop of a rather unique account of persons given in the chapter of Individuals following "Sounds." Coherence within the corpus of Strawson's philosophy of these various positions must be understood, while taking into account the precision of each move he makes. The thing to keep in mind is that the question of the observer's having a realization of his existence distinct from other things arose in connection with the question whether re-identification is a sufficient condition for objectivity; it already have been established that it is necessary. By accepting the rule, Strawson is suggesting that there is nothing that stands in the way of our accepting re-identification as sufficient as well as necessary for objectivity. However, it may turn out that purging our reports of the world in which we live of reference to ourselves is not possible. This may be the way things turn out if we are required to distinguish moving and being moved, that is, the difference between agent and patient. One must keep in mind that re-identification's being sufficient for objectivity is especially important in discussing the sound world because if it is not then it cannot be shown that in a world without material objects objectivity is possible. For this to be shown in the case of the sound world it must be a sufficient condition for objectivity that re-identification is possible. In our actual world we take it as given that there is objectivity and while there may be need for a "deduction" in the Kantian sense to show how this is so; in an imagined world no such "deduction" is in question. This is why it is crucial that Strawson answer the charge that the sufficiency of re-identification for objectivity may beg the question by assuming that the inhabitants of a sound world already possess an understanding of the difference between themselves and the rest of the world. If they were to lack need for such a distinction could hey, in fact, understand it? Nothing more on this will be said on this matter at this point, however.

Varying the pitch of the master sound is intended to be analogous to movement; but if we are to distinguish moving things in the sense of things that are moved and the sense of something that is being done, then it might appear that we need to introduce some concept of self which provides the agent of such movement. While Strawson grants that agency contributes in large measure to our concept of what it is to be a self, he does not believe it sufficient to necessitate introduction of the distinction between my self and the rest of the world. All that is required is a distinction between reports that are predictions and "announcements of intention" on the other. But is it possible to make sense of intention without something that intends? Strawson suggests that only an understanding of how to make use of the difference between active and passive voice is needed without introduction of a self as agent.

If I make use of the sentence 'The wind blew down the tree' I use the active voice; if I make use of the sentence 'The tree was blown down by the wind' I use the passive voice. But there is no coincidence between active voice and the announcement of intention and prediction. In other words, the active voice does not circumscribe the domain of agency or what is, to use Strawson's telling expression "executed." Nor does the passive voice entail neglect of a subject even though no overt subject may appear in a sentence containing a main verb that is passive. Thus, even though in the sentence 'The book was laid on the table' no overt subject is mentioned this is not to say that that no understanding of an agent which is a person need be presupposed. Moreover, a prediction may involve use of sentence about another agent in the passive voice: If I say 'John will roll down the hill' this does not imply that should John role down the hill that it will not be an agent acting. Strawson may not have been unaware of these facts. He remarks that

We need to distinguish what happens by agency from what does not. But we do not need to distinguish agents.
There is, however, a legitimate doubt that an action can be understood as such without at least presupposing an agent. If we neglect the distinction at issue we fail to capture the difference between an event and an action: we risk not being able to distinguish the two senses of 'John rolled down the hill'; one where John does something and the other where something has happened to John. In a sound world the problem is made more difficult, as arguably in our world selves are physical, or at least involve material predicates, whereas in a sound world in which agency is distinct from passivity agents can hardly be considered to be sounds.

We have mentioned that Strawson's world lacks a self in any obvious sense. Towards the conclusion of his discussion he does provide for the possibility of a physical body, one in the form of a sound - just as he provides for the possibility of action other than motion by introducing the ability to produce sounds or changes in them. He entertains the possibility of other features of ordinary life such as conversation; but to pursue such an enterprise indefinitely, he notes, misses the point: which is create a world wherein just a few central claims are made and defended. However, there is a bit of a mystery in what he describes as an objection to the "whole procedure." And that is that he assumes the possibility of the scheme he has detailed. There seems to be an inordinate attention to a merely rhetorical objection he takes as central. What explains this? Recall that the question arose as to the sufficiency of re-identification in establishing the possibility of a nonsolipsistic consciousness in the sound world? What he may be driving at here, at least in part, is that whereas in the actual world such an occasion is commonplace it appears that we must make a leap within the sound world from re-identification to a non-solipsistic consciousness by giving a Kantian sort of "deduction" of the possibility of objectivity. Perhaps it was assuming the possibility of such objectivity that concerned him when he wrote this final reply to his imaginary critic. But then, again, such critics came to be. As we shall see in regards to the sufficiency of re-identification for a non-solipsistic world, Don Locke would raise some interesting questions; Jonathan Bennett and Gareth Evans would press Strawson in some instances even harder. It is to these criticisms that we must eventually attend.

Gareth Evans on Strawson's Sound World

Evans believes ("Things Without the Mind - A Commentary upon Chapter Two of Strawson's Individuals," in Philosophical Subjects: Essays Presented to P. F. strawson, edited by Zak Van Straaten, Clarendon, Oxford, 1980. p. 76. that the difference between

Warm now


Now it's warm

is that the latter unlike the former expresses a judgment of objective significance. By 'objective' he means "something whose existence and operation are independent" of one's experience of it. But is 'it' in 'Now it's warm' an objective something? If so, what? If not, then does the grammatical difference between these two sentences express the difference Evans is striving to articulate? Whether Evans succeeds in this is not of primary significance. What is significant is whether spatiality and objectivity are so related that the former is a necessary or sufficient condition of the latter. If 'physical space' can be defined independently of our use of 'objective' either, implicitly or overtly, then it is arguable, at least that it is a sufficient condition of objectivity that something exists in space. If not then no grammatical contrast will capture the difference between 'objective' and 'subjective' where that difference requires the satisfaction of a spatial condition, viz. having a location. Is it an essential property of spatial objects that they can exist independently of awareness, as some have maintained? If a thing exists independently of such awareness must it have spatial properties? Without answering this question and for the sake of our pursuit of Evan's understanding of the problems involved, let us suppose objective space to provide the conditions both sufficient and necessary for objectivity. As Evans understands it, Strawson constructs an analogue of space in his world of sounds in arguing for a version of the Kantian thesis that space is a necessary condition for objectivity (TWM. p. 77). Evan's penetrating commentary on Strawson is void of any reference to Strawson's work on Kant. Even so, a word on this subject will help clarify matters.

It is of the utmost importance for an understanding of Strawson on the topic of objectivity that he is understood to reject the notion of a Kantian synthesis of experiences in the production of an objective world. What Strawson is in fact denying by relying on spatiality as providing the essence of objectivity is that there is any way of arriving at the concept of an objective world solely by putting experiences together in a special way, determined, say, by the Kantian categories of the understanding. The question will, however, be difficult to silence as to whether a notion of spatial properties as secondary properties, like color, taken together with a synthesis of experiences is sufficient to provide us with the idea of objectivity. If so, then objectivity in a world without objective space is possible and strawson is thereby refuted. We shall keep this in abeyance for the time being, but the gravity of Strawson's rejection of synthesis will be felt when we move from a discussion of Strawson's views in Individuals to his more complete conception of objectivity in The Bounds of Sense.

Evans identifies two stages in Strawson's argument. The first is that re-identifiability is "implicit" in the notion of objectivity, while the second is to show that re-identifiability considered, along with its dependence on the distinction between numerical and qualitative identity, requires the notion of space or something very much like it. Evans will argue that should the first stage be completed the second cannot.

Initial arguments against Strawson's program concern the second stage. The first Evans considers the weaker objection. It consists in pointing out that given the continuing presence of the master sound no two sounds given to consciousness are truly qualitatively identical, for if we take all that is given to consciousness, or at least that much as is includes the master sound, no two auditory experiences will be the same. But there is a second objection which, although weaker than the first, may be sufficient: since Strawson's criterion of numerical identity makes no use of the notion of continuous variation in the master sound, the argument could have been framed in terms of a discontinuous or unordered series of master sounds. Dimensionality is no longer a consideration and the analogy with space collapses. This objection is forceful and is reminiscent of our earlier point that spatiality merely serves a purpose that God's omniscience might serve in an ontology where objectivity does not entail spatiality, at least objectivity in a sense that would require re-identification. On this view, as we saw, there is only need for some thing or property to guarantee continuity, nothing else. What might Strawson say? One thing he might say is this: the notion of occupancy is essential to the notion of space and it is intuitively clear that in doing descriptive rather than revisionary metaphysics, that is, in attempting to capture relevant feature of the real world without dictating to nature on metaphysical principles how it must be, we must consider how certain features are brought together in this world to account of the world as we know it. One feature is the feature of occupancy attributable to objects in space. Motion through space in the absence of continuity would require continuous continous creation in order to overcome the gaps in space should occupancy be a required feature. Further, even if discontinuity is not relied upon this is no reason in itself to believe Strawson's project is affected, as any analogy will be composed of disanalogous elements. What Evans has not done is show how what is disanalogous, that is, the lack of continuity, should be essential to the argument given by Strawson. There is a point, however, Evans does not mention and which impacts Strawson's argument adversely.

Space it has been argued (by Russell and others) can be thought of as a set of perspectives; at least this is true of the world that is presented to our awareness. Further, within space and from a certain perspective two things can be seen to move towards one another. But in Strawson's world of sounds it is impossible to describe a perspective or the fact of two sounds moving closer together from a perspective. It may be adduced that in a sound world, where sound source and sound are indistinguishable, such a phenomenon is never presented to awareness; but if we admit to this, then I think we do have certain problems with accepting Strawson's sound space as spatial in the relevant sense. If we cannot re-identify two sounds qua occupants of space in the same world then motion in space as an objective fact becomes suspect. Where re-identification requires change of perspective (change in relation to the master sound), that is, motion, re-identification within a single perspective is impossible and so movement of two objects, sounds in this case, relative to one another becomes impossible. But most all of these objections may be met with a single reply: the analogy purports to show among other things that re-identification is a possibility in a sound world where there is an analogue of space. If this much is accomplished the other objections can be attributed to features which, while admittedly disanalogous to ordinary space, nevertheless account for one essential feature of objectivity, viz. re-identification. Nevertheless, there is another feature to consider, and that has to do not with re-identification but identification; for if the latter is impossible so is the former.

Before, we mentioned that one component to our idea of space was that in some sense objects are said to occupy it. There is, however, the possibility of a problem that threatens Strawson's position. How is it possible that there might be two sounds qualitatively identical simultaneously audible to a single hearer? If the master sound functions as space how do we account for numerical diversity, rather than numerical identity as in the case of re-identification, obtaining between qualitatively identical sounds given at once. There is a sense in which if we grant the possibility, then we grant that two objects can occupy the same place. But this should be impossible, given the role of space in individuating objects, presumably even in a sound world. If we deny the possibility, then we are saying that in the case where sounds are phenomenally different they cannot be numerically different as they are co-present with the same master sound, and this would appear to contradict the principle of the non-identity of discernibles. I will not venture a final pronouncement on the ultimate disposition of these objections or those proferred by Evans during the early stages of his paper; but it should be clear that should there be fatal problems with Strawson's proposed analogy they will almost certainly trade on the properties associated with the master sound. Evans suggests at one point (p. 82) that there may be ways around using the notion of the master sound while preserving the role of the auditory world whose space it defines.

There may be a way of replacing the master sound with another conception while at the same time blocking the objection Evans raises against completion of the second stage of Strawson's argument. The way out is to construct what Evans calls a 'travel based' conception of sound space. This consists in a set of relations generating an ordering of experiences based on adopting generalizations of the following form:

An experience of kind k will intervene between any experience of kind k' and any experience of kind k'' (TWM p. 82).
This "travel based" conception of sound space, according to Evans, has the advantage of allowing us not only to dispense with the idea of a master sound but also enables a discussion of objectivity without introducing re-identification. All that is required for Strawsonian objectivity is that a phenomenon persist while unobserved, and for this we need not insist on individuatable phenomenona required by a conception of existence based on quantification, as quantification is regarded as possible only over individuatable phenomena. To see how this is possible consider merely that one can easily imagine its continuously raining even though our awareness of its raining is only for a few minutes, say. In such a case the notion of 'a single rainstorm' is not prior to 'its raining' and yet 'its raining' inasmuch as it refers to an event persisting beyond the time of our awareness is, on a Strawsonian conception of objectivity, an objective phenomenon. Reasoning along these lines leads Evans (TWM p. 83) to raise the possibility of objectivity in a world describable using only "feature placing" sentences. I will defer discussing feature placing because it is more appropriate to a discussion of Strawson's views on Kant, which I will discuss in considerable detail in the second part of this essay. However, Evans' example of its raining vs. a rain storm can be addressed with advantage, as it provides the opportunity to make an important distinction.

If the notion of its continuously raining does not introduce individuatable elements what then is it that is being said to be objectively real. In Strawson's discussion, as well as Evans', an important distinction has, I think, been overlooked. This is the distinction between an objectively real thing and an objective fact. It is the fact that it is raining that is objective and not some object over which we can quantify. Strawson is interested in connecting the notions of basic particulars and logical subjects, and thus the connection of physical objects, given his theory of what is basic, and quantifiable subjects. Evans' continuous raining example may be an objective reality but it will not afford any insight of significance into the possibility of an objective world without basic particulars which are physical objects. The fact that I have a toothache is an objective fact; the toothache is not an objective reality. A fact about a mental phenomenon may be objective in the sense of being true independently of our awareness of that fact without being an entity "in" space or time. The nature of the distinction will become clearer as we later become apprised of Strawson's views on Kant, for here the difference between judgment and thing judged about becomes important. Besides the fact/object distinction which Evans apparently ignores in his discussion of objectivity there is another distinction, one of which he is certainly aware, but which he mistakenly believes bears directly on the primary question of Strawson's sound world.

By introducing processes into the discussion of objectivity Evans easily escapes the objectivity/re-identification connection: if I observe a process and then cease to observe it only to come back later and find it still taking place the distinction between qualitative and numerical sameness is not important to re-identifying the "same" process (cf. TWM p.86). But the question that Strawson says he is in fact interested in is whether "the status of material bodies as basic particulars [is] a necessary condition" for our having in our midst objective particulars (cf. Individuals p. 53). While Evans raises an interesting point it simply fails to address the main question(s). Evans makes a stronger case in the third part of his paper in which he discusses the notion of space, beginning with a consideration of the traditional distinction between what have been called 'primary' and 'secondary' qualities. This distinction is far more subtle than many philosophers have taken it to be and how one decides on its status determines an important "metaphysical moment" in one's entire philosophy. For this reason, given Evans introduction of these notions, I will briefly examine how Evans construes the distinction.

The distinction between primary and secondary qualities has been around at least as long as Galileo, but the formulation of the difference which has tended to define the dispute surrounding it we owe to Locke. Among the primary qualities, according to Locke, are solidity, extension, figure, motion or rest, and number. (Essay Bk. II. Chpt. viii). Other qualities, the secondary ones, depend on sensation and exist as properties of objects only insofar as primary properties produce effects in the sensing subject. Only the primary qualities are real, however; as Locke himself remarks

...they may be called real qualities because they really exist in those bodies. But light, whiteness, or coldness, are no more really in them than sickness or pain is in manna...(Bk. II. Chpt. viii. 17).
Berkeley argued the distinction was invalid, as primary properties, themselves, depended on the state of the experiencer, who Evans calls "Hero." For Berkeley there is nothing "secondary" about sensory qualities. Evans, almost imperceptibly does something, very different: he tends to regard all qualities not as uniformly sensible, but, rather, as dispositional and so removed from sensory experience. This will not become a weighty issue, as far as concerns, Strawson, but it is a significant move within the "dialectic" of the discussion, historically conceived. Evans's "spin" on the issue is made more readily evident by contrasting the sensory and the dispositional, and this is not to say Evans abides by the traditional construal.

Dispositional properties understood as "powers" have been around for at least as long as Locke, and they have figured highly in various treatments of what passes as mental for a number of analytical philosophers, such as Gilbert Ryle (The Concept of Mind); but perhaps the most engaging issue with respect to dispositional properties was that of how to treat theoretical terms, such as 'solubility', in an empirical (and logically extensional) language. This matter was addressed in some detail by Carnap ("Testability and Meaning," <Philosophy of Science, 3, 1936 and 4, 1937). Terms for dispositions are here treated as paradigmatically theoretical within the framework of the logic of the language of science. It has not been uncommon to attempt to explain secondary properties, that is empirical qualities, in terms of primary qualities by regarding primary qualities as dispositional, that is theoretical, properties that account for or "ground" the secondary qualities. If I hold up a piece of sugar, its whiteness is a secondary quality; its solubility is a theoretical property that accounts for the sugar's dissolving if placed in, say, water. It seems natural to regard secondary properties as being present to Hero as the exercise of a dispositional property of objects which depends on the primary qualities associated with the material constituency of the object. But what Evans does is, although perhaps not unique, not within the traditional understanding of the distinction. What Evans does is treat secondary qualities as dispositional, such as as is the property of solubility, notwithstanding the disposition's theoretical character; as Evan's remarks, "a sensory property is a dispositional property" (TWM p. 95). This notion of secondary properties being dispositional is not as unorthodox as may first appear. Popper, for one, certainly held such a view and one can present a strong case, although we shall not, that primary qualities are to be contrasted with secondary qualities precisely in the sense that the former are given "directly" and not as a consequence of certain causal circumstances, environmental or of the body.

A great deal may be said about the notion of a disposition or "power" and the idea of a secondary property. Some will argue that such a property is merely the result of primary qualities under the assumption of certain causal conditions and that as such it is merely a disposition of those objects having a particular set of primary qualities. Others will contend that a sensory quality cannot exist unperceived, as a disposition or any other theoretical state. Such philosophers will argue that secondary qualities belong uniquely to sense data and that they do not exist unperceived. This consideration supplies the defining difference between Strawson and Evans on secondary qualities. As we have seen, for Evans such qualities are dispositional, but for Strawson they belong to what traditionalists have called "sense data" and do not exist when unperceived - as do dispositional qualities.

In contrasting Kant's objective world with a world consisting of the "pure sense data experience" (Bounds of Sense p. 100), Strawson will remark on the fact that the objects to which secondary qualities primarily belong, that is sense data, are such as "their percipi seems to be nothing but their esse" (ibid). This will prove to be of fundamental importance not only in understanding Strawson's approach to Kant on objectivity but also in the process of doing so in explaining Strawson's belief that one can retain much of what Kant has to say on the subject while dispensing with the notion of "synthesis." It is convenient, then, to end our discussion of secondary qualities with the single remark that for Evans such qualities are dispositional, and may exist unperceived, whereas for Strawson, who understands the issue in more traditional terms, such qualities exist only when perceived. There is an in between position I will not discuss, which relies on the notion that if sense data can exist unperceived then so can secondary (more properly "sensory") qualities.

We see, then, that for Evans primary qualities are theoretical and that secondary qualities are dispositional and, therefore, in at least one important sense, also, theoretical. This leads us to the question as to why Evans draws a distinction in the first place. The answer appears to go something like this: some qualities cannot be derived from experience; these include the primary qualities. Some, e.g., hardness although experiential, are not exhausted in their content by what is given to us in experience. Nevertheless, since secondary qualities and primary qualities are theoretical his claim to have drawn any kind of distinction between primary and secondary qualities must be met with some degree of skepticism. The issue as it concerns Strawson is whether sensory qualities and the objects of experience which possess them are such that their esse is their percipi. Evans has confidence in Reid's characterization of properties such as hardness and softness independently of experience alone; but even in the passages he cites from Reid, Reid presupposes the existence of bodies. There is every reason to believe that Reid is speaking here of material particulars. But, then, since material particulars are conceived of as objective, that is as entities existing independently of experience the distinction between primary and secondary properties is no longer of relevance to the issue central to Strawson's project, viz, the relation of re-identification and objectivity. Despite such considerations, Evans does make use of the notion of primary qualities in drawing an important distinction between Strawson's sound world and the world of everyday life.

In a sound world the objective existence of an unperceived sound depends on the truth of a conditional. Suppose I am aware of a sound; suppose, further, that I move away from that sound to the point where it is no longer audible. The master sound changes as the sound fades. The truth of the conditional 'If I position myself in such a way, I can return to experiencing a qualitatively identical sound in the presence of the master sound as originally experienced' is the ground of the belief in the persistence of the sound while unperceived. Such conditionals have been invoked by phenomenalists to ground belief in the objective reality of unperceived objects, although the reality itself is guaranteed by God's omniscience. Dispositions, as we have seen reason to believe, are grounded in the truth of subjunctive conditionals such as these, dispositions believed to possess objective reality even though they may not be experienced, such as the fragility of an unbroken glass which possesses the disposition of breaking if stuck with little expenditure of energy. Fragility is a property associated with an object. It is provides the ground of truth for the conditional. Not all conditionals possess a grounding in properties of objects. Primary qualities ground secondary qualities in a way similar to that in which dispositions - which may ultimately be regarded as depending on primary qualities - ground conditionals. But not all conditionals it would appear possess such an objective grounding. This fact was noted by Isaiah Berlin in his criticism of phenomenalism, a view which held reality to amount to nothing more than sense data. Berlin's views are central to Evan's criticism of Strawson. For this reason it will prove advantageous to closely examine Berlin's central claim, which is that there is something "paradoxical in all phenomenalist analyses." Let's consider, then, what it is that about phenomenalism that bears on Evans's approach to Strawson. In discussing phenomenalism Isaiah Berlin remarks on how its advocates deal with objects which common sense regards as existent but outside purview, such as a brown table in the next room.

The statement that if there had been (and there was not), any observer, he would have observed (and no one did observe), certain data, seems to them not equivalent to asserting the past existence of material objects. Categorical propositions about material objects are replaced by unfulfilled "counter-factual" hypothetical propositions about observers, and what troubles the plain man is the thought that if the hypotheticals are unfulfilled, if no observers were in fact observing, then if the phenomenalist analysis is correct, there was - in a sense datum sense - nothing at all, and, moreover, that this sense of 'existence' is basic: because the alleged material object sense in which the non-existence of serial sense data nevertheless can be 'translated into' the existence of material objects, is not a sense in which the word 'exists' is commonly understood. "Empirical Propositions and Hypothetical Statements," Mind (lix) 1950. Reprinted in Robert Schwartz Perceiving, Sensing, and Knowing Anchor Books. p. 373).
[added later] Evans will make interesting use of Berlin's criticism of phenomenalism. Consider such a hypothetical as: If I were to cut this apple I would see two white patches. Here is a hypothetical. If I did not cut the apple would the white patches exist; and even if I don't speak of anything as phenomenal sounding as "white patches" but, instead, speak of parts, do I say that the categorial is supported by a disposition, and what would such a disposition be? Moreover while there may be a categorical basis for the truth of the hypothetical that truth, arguably doesn't enter into its meaning. If I had cut the apple slightly differently the conditional would have been made true by a different fact; and should I not cut it the conditional may be true. Unlike solubility and its corresponding active disposition, 'being a solvent', there is no corresponding duality in the case of the hypothetical in this case. (cf. Sellars Science, Perception, and RealityRoutledge and Kegan Paul, 1963. p. 64) At the time Evans wrote, there was a widely accepted belief that dispositional terms were eliminable in favor of simple conditions. A thing's being called 'fragile', say, meant only that if it were struck with a relatively light blow it would break. There was noting irreducible about dispositions, nor was there an ontological presupposition made in their attribution. There was, at the time, however, circulating among certain "connected" philosophers an argument against the simple conditional analysis of dispositions. This argument was constructed by C. B. Martin and following its publication much later, Dispositions and Conditionals in Philosophical Quarterly Vol. 44. Jan. 1994, a provocative and rich discussion of the details of that argument. Before briefly examining the argument, let's consider the stakes for our discussion of objectivity in a sound world. If dispositions are expressible as conditionals, then there is no basis for Berlin's assertion that conditionals require the grounding of such properties in the world. In addition, Evan's argument for the need of primary qualities would likewise be undercut. This is why it is so important to get to the bottom of the question as to whether there actually are properties which are dispositional, and which are grounded in primary qualities associated with material substances in the Lockean sense. Using a contrivance, and I don't mean here a "mere" contrivance, Martin attempts to show that a conditional will supply neither necessary nor sufficient conditions "for the power ascription of which it is meant to be the analysans" (op. cit. p. 3). The "contrivance" is an imaginary device called an "electro-fink."

Imagine that we have a wire and we say that it is 'live' if and only if If the wire is touched by a conductor then electrical current flows from the wire to the conductor. But imagine, further, that the wire is connected into a device called an "electro-fink." What an electro-fink does is this: suppose that the wire is dead at t1 but when it is touched at t2 the electro-fink "ensures that the wire is live" and current passes from the wire to the conductor. In this case, the wire was dead, but nevertheless that wire was such that had it been touched, owing to the electro-fink, it would have passed electrical current to the conductor. So even though the conditional was true at t1 it was, nevertheless, false that the wire possessed the disposition. In other words, the truth of the conditional was not sufficient to warrant ascription of the disposition - indeed, at t1 the wire was dead. The electro-fink works in reverse as well; and what this means is that even when a wire is live the electro-fink comes into play by breaking the circuit, say, whenever it is touched with the effect that the conditional is false. Thus, the truth of the conditional is not necessary for the ascription of the disposition 'live' to the wire. Martin concludes his remarkable little paper, saying:

If counterfactualities and dispositionalities...will not do, and if, as we have seen, counterfactuality or strong conditionality cannot explain dispositions, then there is no place to turn but to actual first-order dispositions or powers (Op. cit. p. 7).
Important here is the idea that while the counter-factual may be true it remains the case that there was no sense datum and, so, no object to be experienced. In other words, the brown table in the other room exists only in the sense that had someone gone into that room he would have had a certain sense datum experience; but, he did not go into that room and so there was nothing - even though the counter-factual is true. The question, then, that Evans latches onto is what makes this counterfactual true, if anything; and, if nothing, how good of an analysis can it be? Before pursuing this question let us take a more careful look at Berlin's thinking on the subject of conditionals. One important reason for doing so is this: Berlin takes the phenomenalist to be employing dispositions that attributable to persons, whereas, almost imperceptibly, Evans will shift the focus immediately to material substances and primary qualities. It is in fact not Berlin that animates Evans's position as much as Mackie's discussion of competing views on dispositions. Although Evans is closer to Mackie on this topic than Berlin, he is, perhaps, even more close to Harre, who held to a realistic view of dispositions and not a mythical intermediary between primary qualities and overt manifestations of the presence of such qualities.

Berlin sees the phenomenalist's employment of dispositions as belonging to persons:

To analyse material objects in terms of hypothetical data is, in effect, to turn the statements about them into statements about the dispositional characteristics of observers. (Schwartz p. 373)
By contrast, Mackie raises the question of the need for having dispositional properties in the first place:
Why should we insert this extra element between the non-dispositional basis and the causal behaviour? (Truth, Probability, and Paradox. Oxford. 1973. p. 137)
The issue of dispositional properties has taken on something of a life of its own. While Locke and Leibniz, just to take a couple of examples, had considerable things to say about "powers" we don't really find the issue of dispositions becoming an issue approximating in form that pursued by Mackie, Harre, Evans etc, until Carnap made them "famous" by selecting them as paradigmatic theoretical properties the predicates referring to which he sought elimination within an empirical and logically extensional language. What Carnap argued in part was that explicit definition of dispositional terms, like 'soluble', was not possible but rather what he called bi-lateral reductions sentences contextually introducing dispositional terms was required ("Testability and Meaning," Philosophy of Science, 3,1936 and 4, 1937. Reprinted (in part) in Readings in the Philosophy of Science Appleton-Century-Crofts. 1951. pp. 47-93). I will refrain from a protracted discussion of this remarkable paper, one which Carnap later more or less abandoned; but, I will examine briefly for the purpose of illustration one example of the sort of "reduction" sentence he used to explicate the notion of a disposition.

According to Carnap, then, if we want to introduce a term like soluble we will not use an explicit definition, but rather, in order to satisfy what the language of empirical science demands, a sentence where he dispositional term occurs as antecedent of a conditional which is itself a consequent of a categorical sentence containing reference to a test condition. Taking 'soluble', then, as our example here is how he did it:

If any thing x is put into water at any time t, then, if x is soluble in water, x dissolves at time t, and if x is not soluble in water, it does not.
I will not go into the details that follow such as the need for sets of "bilateral reduction sentences" and numerous other issues brilliantly set forth by Carnap. Instead, I want to call attention to one or two important facts that bear on Evans's treatment of dispositions in connection with sensory properties associated with sound. First, it should be noticed that this way of introducing dispositional qualities may rely on other dispositional qualities, and so the notion of what a disposition is cannot be given except by saying it a property introduced by terms given a characterization in some way involving conditionals. What I have in mind here is that not only is 'soluble' a dispositional term but so is 'being a solvent'. Compare the following two sentences:

1. If x is put in water, then if x is soluble it dissolves.

2. If I put salt in x, then if x is a solvent the salt dissolves.

We see, above, that (1) assumes water is a solvent (more especially a salt solvent, but I will ignore this as it brings up matters having to do with what I was referring to above in regard to "sets of bilateral reduction sentences"). Similarly, (2) assumes that salt is soluble (more especially water soluble). What is important here is not the detailed account of reduction sentences, nor for that matter whether the conditionals are those of material implication but, rather, this sort of dependency. To see why consider two further sentences:

3. If I go to place y, then if x exists I perceive x . at y.

4. If I go to place y, then if I perceive x at y, then x exists at y.

Assuming this sort of characterization and given the central thesis of idealism that esse est percipi then what would be entailed by the 'objectivity' of x would be the following conditional, one containing a biconditional:

5. If I go to place y, then I perceive x at y if and only if x exists at y.

But now the conditional is true if I neither perceive x at y nor x exists at y. However, time must not be ignored: objectivity can be correctly attributed to x even if x does not exist at the time of the attribution. But it may be said that it is also true if the antecedent is false. Indeed, it was the point of such sentences originally to use only material conditionals. We are, however, not committed to this. In fact, there are persuasive arguments that the materiality of such conditionals is an impossible option (cf. Mackie Truth, Probability, and Paradox p. 125). Suppose, then, that the phenomenalist (of an idealistic persuasion) sought to defend the idea that his position is compatible with 'objectivity' in the sense that objective entities exist independently of human will by stating the objectivity of x, something unperceieved by a human agent, in the following conditional sense:

6. If I had gone to place y, then I would have perceive x at y if and only if x existed at y.

This would be a trivial fact agreed to even by most Lockean realists. Something has gone wrong. Here is what the idealist claims: 7. (If I were to go to place y, then I would perceive x at y) if and only if x exists at y. (3) construed existence as a disposition to be perceived under certain conditions; (4) took it to be a disposition associated with a perceiver (Berlin). The correct view is not to take the unperceived existence as a disposition to be perceived under certain conditions, but as a subjunctive conditional which is grounded not in some underlying primary qualities (Evans) but instead in lawlikeness construed as a connection between actions, such as going to some place, and experience, such as perceiving. Evan's mistake, all along, has been to treat objectivity for the phenomenalist as consisting in a dispositional property to experience under certain conditions. We see from the above discussion that this is not the correct account. Logically the problem was parsing the conditionals. For Evans the difference between Hero and ourselves with respect to sounds occurring at a place is that for us there is a disposition to hear that sound when at the right place, one that is grounded in the categorical features of the material being from which the sound originates.

...for we have, and Hero does not have, the resources to make sense of the idea of the persisting categorical basis or ground of that disposition, in the object, or at the place, to which it is ascribed. Unlike Hero, we have the concept of substance, of space-occupying matter, for we have the concepts of the primary properties of matter.
It must be kept in mind that what motivates Strawson's construction of the sound world is a desire to explore the possibility of a non-solipsistic consciousness that does not rely on material particulars as basic. The purpose of the spatial analogy is to show how this might be argued for without introducing the resources Evans mentions, such as material substance whose properties ground certain dispositions etc. The point is that Evans sees the dispute not as one between dispositional accounts and accounts of objective sounds that have no need for dispositions, but rather of how a disposition is to be grounded; for nowhere in Evans are we given reason for thinking that over and above the categorical features of the sounds and the conditionals that describe our experience we need to introduce dispositional properties in the first place. Instead, the debate centers on different ways of grounding dispositions, particularly in a sound world.

This approach, however, may not be warranted. Once we distinguish three things, the disposition, what grounds the disposition, and the behavior associated with the disposition, the natural question is whether we must accept all three. This is a general problem associated with any discussion of dispositions. Here we are interested in sounds and the places where they can be heard, and here the questions become: What is the disposition and to what does it belong? Evans:

For both Hero and ourselves, the truth of a proposition to the effect that there is a sound at such and such a position must consist in this:...if someone was to go to that position, he would thereby be caused to have certain auditory experiences. TWM. p. 100.
This way of introducing the problem is unfortunate. In the first place there is use made of the notion of a cause. This suggests without argument that for Hero such a requirement is warranted, when in fact it tends to bias the issue of whether some underlying categorical fact is needed to ground a disposition in the first place. But, more importantly, from this conditional he moves immediately to dispositions which he associates with these conditionals. There is nothing, however, analytic about the belief that if you are given a behavioral conditional that a disposition must be posited to give it some basis, particularly when that disposition may not be material (in the sense of 'material implication'). In any case, what we are said to have is a disposition associated with a place, that is, the disposition to cause auditory experiences. If we accept this opening gambit we immediately undercut a systematic investigation of whether objective particulars are possible in an auditory world. In fact, by introducing dispositions as properties of places, and places as sometimes being outside our awareness, we have already rejected what Strawson is attempting to evaluate, namely, the possibility of a non-solipsistic world where basic particulars are not material objects (Individuals p. 53 and 61). Evans may be right to raise the point that Hero is "caused" to have the auditory experience only by being at a place, but this does not compel us to accept the idea that at that place there must be some material substance which is the causal source of the sound, itself. We have remarked that the conditional associated with the having of the experience is not material, and what this suggests is that the lawlikeness associated with the having of the sound at the place is operative; what it does not entail is that there must be some set of underlying primary qualities to explain why that conditional is lawlike, or at least not material. Evans thinks otherwise.

Evans's not only does not believe that Hero can make sense of a thing existing both while being experienced and while not being experienced neither does he believe that he can makes sense of "any unexperienced existence existence at all" (TWM 104). But if Hero can conceive of other experiencers then surely while he is experiencing the sound he can imagine other persons not experiencing the sound, and so clearly he can make sense both of a sound experienced and not experienced. This may not be sufficient to prove he has a notion of objectivity, but it suggest, at least, that being able to conceive of something's existing both as perceived and unperceived is not sufficient to achieve the sense of objectivity that re-identification supplies, if possible, in the sound world. Suppose, however, that we grant Evans's central point that we require some sort of underlying material substance complete with primary qualities that explain the disposition we have if we are to have "such and such experience" (ibid) whenever we return to the same place. This alone is not sufficient to establish objectivity. It is open to Hero or what is in fact his opponent, the phenomenalist, to claim that the material substance comes into existence when experienced and that unless there is some reason to believe that it is numerically the same sound-causing material entity which is its source there is no ground for the disposition which would accommodate Berlin's original objection. More problematic for Strawson is that if he allows that a sound may move there is nothing to rule out another qualitatively identical sound moving to take its place, whereupon a return to that place, even though the experience may be indiscernible from the original, will be insufficient to re-identify the sound. Strawson's reply here would be that it is not that re-identification is always successful but, merely, that it be possible; this is all that is needed, he might well argue, in order that Hero may have a concept of objectivity. Further, since primary properties are just that, properties subject to duplication, there is no reason for maintaining, Strawson might argue, that what we encounter on Evans's account is another qualitatively similar, but not numerically, identical object. Evans will have simply pushed the problem back another and even more difficult step merely, adding properties which do not guarantee individuation. Individuation is no guarantee of objectivity as individuation at a time and place is not the same as re-identification across time or some perceptual "gap."

Evans immediately after introducing the matter of dispositions, goes on to entertain two ways in which dispositions may be grounded. The first is by means of a "complicated generalization": "Whenever in the recent past I have gone to p, I have had a φ-experience." (TWM 101). Also, a generalization, presumably is not an entity existing outside of language, but when he introduces the second support dispositions may receive he reverts to ontology. Not only has he skewed the discussion in the direction of accepting this by his use of 'cause' in describing the conditional, but when he goes further in attacking phenomenalism for lacking an ontological backing for dispositions he relies on Isaiah Berlin who took the phenomenalist as locating the relevant disposition not at a place - which as we have seen assumes a solution to the problem of solipsism - but rather, and I think more appropriately at the person qua experiencer as the cite of the disposition. Recall that according to Berlin,

To analyse material objects in terms of hypothetical data is, in effect, to turn the statements about them into statements about the dispositional characteristics of observers. (Schwartz p. 373)
True, Berlin is speaking of material objects, but adjusting for the fact that we are speaking here of a sound world the point is easily made that deciding on whether the disposition belongs to the knower or to the object is to decide ab initio on the fate of solipsism and the possibility of objectivity in a sound world. It may help clarify the point I am attempting to make if one considers not only the two alternatives of grounding by means of "complex generalization," or disposition associated with a cause at a place, the further alternative of a disposition of a mind to cause an object at a place of which one is aware. Berlin's observations are at the center of Evans's interest in grounding dispositions as a way of thwarting the possibity of objectivity in the sound world:
Sir. Isaiah Berlin has objected to phenomenalism that it reduces catergorial existence to the truth of subjunctive conditionals which are not, in their turn grounded on anything else. (TWM p. 102).
Let's move immediately to the passage in Berlin's paper that most vividly states the objection.
The statement that if there had been (and there was not), any observer, he would have observed (and no one did observe), certain data, seems to them not equivalent to asserting the past existence of material objects. Categorical propositions about material objects are replaced by unfulfilled "counter-factual" hypothetical propositions about observers, and what troubles the plain man is the thought that if the hypotheticals are unfulfilled, if no observers were in fact observing, then if the phenomenalist analysis is correct, there was - in a sense datum sense - nothing at all, and, moreover, that this sense of 'existence' is basic: because the alleged material object sense in which the non-existence of serial sense data nevertheless can be 'translated into' the existence of material objects, is not a sense in which the word 'exists' is commonly understood. "Empirical Propositions and Hypothetical Statements," Mind (lix) 1950. Reprinted in Robert Schwartz Perceiving, Sensing, and Knowing Anchor Books. p. 373).

Suppose we compare unperceived objects to unbroken but fragile dishes.

What Evans does is note that in a sound world objectivity expressed in terms of the truth of conditionals does not have the grounding which fragility supplies conditionals associated with unbroken objects. What grounds the truth of conditionals in a sound world, at least those relevant to belief in the objectivity of unperceived sounds, is merely a generalization: whenever such and such master sound is heard, so, too, is a qualitatively identical sound. We shall see (in Pt. II) that Berkeley's use of conditionals, and that of the phenomenalists, is indifferent to the distinction between sentences making mention only of features and sentences presupposing physical objects. This is important for the simple reason that what Evans takes objectivity to be is considerably different from what either Strawson or Kant take it to be.

Philosophers sometimes have taken 'objectivity' to be closely linked to what has been dubbed the "act/object" distinction, that is, the distinction between awareness and what it is of which one is aware. Some philosophers, then, construe 'objective' as characterizing whatever it is of which we are aware. However, to take this as the sense most relevant to our discussion is to error, and Evans it would appear has fallen into making this mistake. This is suggested by a number of things Evans says, although the following is, perhaps, the most telling:

The notion of objectivity arises as a result of conceiving a situation in which a subject has experience as involving a duality: on the one hand, there is that which there is an experience (part of the world) and on the other, there is the experience of it (an event in the subject's biography). We have been exploring the consequences of this duality. (TWM 103).
Evans appears unaware that what he has been arguing for depends on a sense of 'objectivity' which goes beyond the act/object distinction, duplicating in effect G. E. Moore's mistake in the "Refutation of Idealism" (1903). Either this or he is assuming a distinction between object in this sense and being an object in a deeper more "weighty" sense. But making this charge presupposes that Evans distinguishes the objects of awareness qua representation and object represented. This may not be a valid characterization since what he speaks of when he speaks of " that of which there is an experience" is not a representation but rather "part of the world" (TWM p. 103).

Kant certainly held to such a distinction as evidenced by, among other things, the following:

Everything, every representation even, in so far as we conscious of it, may be entitled object. But it is a question for deeper enquiry what the word 'object' ought to signify in respect of appearances when these are viewed not in so far as they are (as representations) objects, but only in so far as they stand for an object. A190/B235.
Objectivity for Evans is not, then, what it is for Kant. All that is required is that there be a time when the unexperienced object can be said to exist. This may not be sufficiently precise. Those who maintain, such as Russell at one point, that we are acquainted with universals would then be acquainted with objects objective in the same sense as particulars. The issue of basic particulars as material becomes obscured, objectivity and materiality no longer being connected, essentially. Moreover, there is a presumption of the truth of naive realism, the notion that what we experience is the object itself and not some representation or effect of the object. These may not be insurmountable problems, indeed, they may not be problems at all; but they do suggest that Evans is not addressing certain ideas that Strawson takes as central at least in the sense that there is no distinction for Evans from the very beginning between a "weighty" and non "weighty" sense of 'object'. It is worth comparing here Strawson's distinction between the "weighty" and non-weighty sense of 'object' (Bounds of Sense p. 73). Idealism on one construal can be thought of as the claim that there is no "weighty" sense of 'object'. The idealist can make this claim without rejecting the distinction between the act of awareness and its object. It is of some interest that this issue reveals important ambiguities in the formulation of positions long rejected (for the wrong reasons, usually), such as those surrounding Russell's position on neutral monism. That view has been taken as especially important for some not so much for the affirmation of the neutrality of sense data, and so for Strawson an unorthodox take on what a "pure sense data experience" might be, but for its denial of the act object distinction which Moore felt put idealism in its grave forever. Once we distinguish an "ontological object" from an "objective constituent" of a perceptual situation, after the fashion of C. D. Broad (The Mind and Its Place in Nature (1925) (who in fact is following in the steps of Twardowski), then, even though we may care to entertain a neutral monist world in which the act/objective-constituent distinction is no longer viable we might not wish to similarly reject the act/object distinction under threat of making knowledge of objects impossible. This, however, is an extraordinarily complex matter we best leave aside.

If Evans has, in fact, failed to observe this important distinction between two senses of 'object', and it is by no means evident that he did, he would not be alone. G. E. Moore in his "Refutation of Idealism" bases his attack on idealism on precisely the distinction between act and object, not fully appreciating that in "the tradition" 'object' possesses more than one relevant sense. Moore seems to think that on the basis of the act/object distinction one can infer the reality of objects in the "weighty" sense; such appears not to be the case. Let us set aside the issue of what objectivity is and proceed to give Evans's criticism of Strawson a more careful examination. It would be easy enough to state Evans's view only to the extent required to evaluate Strawson's reply to Evans, but Strawson's reply is not strong and the points Evans raises are of persisting value. Evans will make use of the distinction between primary and secondary qualities. In doing so, he will avail himself of more recent discussions of the nature of so called dispositional properties.

Evans understands a sensory property to be a secondary property and a secondary property is to be merely a dispositional property. A dispositional property in turn is understood in terms of hypotheticals. A sensory property, such as loudness is a secondary property; one such that were a certain sensitive agent suitably placed he would experience it (the hypothetical component). But secondary qualities are "secondary" because they depend upon (some would say "supervene upon") primary qualities, which serve as their ground. Thus sensory properties are understood in terms of hypotheticals and inasmuch as they are secondary properties can be identified with their "ground," viz. primary qualities. More simply put, sensory qualities are dispositional properties and these are reducible to primary properties, and, so, sensory properties are reducible to such nonsensible properties as primary qualities, such as having a certain shape, size, motion, extension etc. As such, primary qualities are associated with material substance and persist even while unseen or otherwise unexperienced. This is not to say, as Evans points out, that primary qualities are not sensible. The squareness of a box is one of its primary qualities and it is certainly a visible property of the object. The point is that the idea of a square does not have an entirely sensory basis; we cannot 'distil' the notion from entirely sensory sources (TWM. p. 96). Evans's employment of the distinction between primary and secondary properties brings with it commitment to the idea that objects which possess primary qualities occupy space. I take it that part of what he means to imply is that if there is nothing corresponding to the occupation of space in a sound world then objectivity can no longer be held to be a feature of objects in such a world - if, indeed, it would make sense in this case to speak of objects at all. Evans tells us that

What is important, though, is that the properties constitutive of the idea of material substance as space-occupying stuff should be acknowledged as primary. (TWM p. 103)
There are, then, two questions to consider: first, is there any such thing corresponding to occupancy in a sound world; and, second, is occupancy essential to re-identification or objectivity in a sound world? The answer to the first question will depend, at least in part, on whether the distinction between qualitative and numerical sameness can be sustained in a sound world. An affirmative answer is suggested by the possibility of being able to make sense of a sound moving within a sound world, and here I mean a sound that persists through space. Persistence through space entails an object which occupies space in some sense of 'occupy'. The answer to the second question will depend on whether speaking of a sound as opposed to sound presupposes motion. In other words, if we take the experience of sound as merely from a single place can we make sense of the difference between sound as a feature and sound as a particular? If we cannot then we must ask whether objectivity is possible in a world consisting of features alone. While this may seem impossible notice this much: reidentification of a feature may be possible even though that feature does not remain in awareness, but if there are only features then the difference between qualitative and numerical sameness ceases to be meaningful, and it is this distinction that Strawson relies on in making his case for the place of re-identifiability in providing what is essential to objectivity. Further discussion of this matter will be deferred. What is of immediate interest is that Evans argues that places in a sound world are the source of sounds without anything analogous to primary qualities belonging to objects occupying those places (TWM 103) and that this presents us with some reason for questioning whether objectivity is possible in a sound world. A further question, one we cannot pursue further, is whether we can re-identify a sound in a sound world when that sound has moved. If Johnathan Bennett is right much depends on how fast the sound travels in relation to the master sound. But if we cannot, or if the movement of sound cannot be given sense, then little sense can be attached to sound's occupying places in the space of a sound world. It is occupancy, I would maintain, which contributes to the difference between 'sound' and 'a sound' - a feature of spatial object recognized by Evans but not addressed by Strawson. Evans does not expand on the bearing primary qualities have on the difference between features and things, that is, basic particulars. This shows in how he makes use of Berlin's objection to phenomenalism that when a conditional is unfulfilled there is nothing outside experience which can be said to be real.

So far we have dealt with what Evans (TWM p. 106) identifies as the "first respect" in which Strawson fails to establish objectivity in an auditory world. This has been its alleged failure to accommodate dispositional properties he believes essential to objectivity. We now move to the second basis for doubt, having to do more specifically with the properties of auditory space in Strawson's world of sounds.

Philosophers have long distinguished passive and active powers (dispositions). If we recur to Berlin's argument against phenomenalism, the role of causation becomes of some interest, as dispositions involve the manifestation of powers; while powers introduce a causal nexus which, itself, on prevalent theories of causation presuppose physical objects. If this is so, however, then as physical objects possess objectivity dispositions cannot be invoked noncircularly in criticizing alleged objectivity in a sound world. Moreover, whether the hypotheticals true in such a world demand counterfactuals can be argued against should we accept a theory of dispositions where both the dispositional property and its causal base exist as separate concerns. On related theories, and here I have in mind David Lewis's account of dispositions ("Finkish Dispositions" Papers in metaphysics and Epistemology. Cambridge. 1999) there is a distinction to be drawn between external and intrinsic qualities. We have seen that there is a discrepency between Evans and Berlin as to whether the relevant disposition belongs to the knowing subject or the place where the experience, in this case a sound, is located. Consider the following dispositional hypotheticals:

x is experiences sound z, if x goes to place y.

sound z exists at place y, if x goes to place y.

In the first instance the hypothetical, assuming we accept something like Berlin's argument, supports a disposition belonging to a person. In the second instance the disposition can be thought of as belonging to a place. The problem is that being at a place is an extrinsic property of an object, whereas having a certain experience is intrinsic to the experiencer. If so, then if we are justified in speaking of a disposition it most likely belongs to the person and not, as Evans would have it, the object understood as being at a place. That places possessing causal efficacy, as well as a number of other considerations could inhibit this mode of argument; but the issue of whether the disposition belongs to the object or the person is an important one, as long as inference from a hypothetical to the existence of a disposition is the point, rather than vice versa. Another consideration is that if Evans is right, and we accept the conventional wisdom that dispositions are essentially tied to causation (Mackie, Armstrong e.g.) then since causation presupposes physical objects a certain bias is built into Evans's criticisms - complicating an evaluation of what the final status of the argument is that the issue of causation in a sound world has not been raised.

The second respect in which there is cause for doubt that objectivity obtains in a world of sounds, as we have said, has to do with the concept of space Strawson constructs for such a world. To begin with, according to Evans there are two concepts of space.

On the one hand, we have what I shall call serial spatial concepts - concepts explained in terms of the succession or sequence of the subject's perceptions, and any muscular or kinaesthetic sensations accompanying these changes, whether they arise from the movement of the whole, or merely part, of the subject's body... the corresponding term 'travel based' ought not to be taken too seriously. For, as Poincare said, the 'movements' can be characterized in terms which do not presuppose the existence of space...distinguished from these are what I shall call simultaneous spatial concepts...relational concepts the situation for whose most direct application is one in which the elements related by them are simultaneously presented or perceived. (TWM p. 109)
Evans is a little more specific on the important differences between these two concepts when he says
Someone might be able to tell that the line connecting the objects was straight by means of the kind of bodily movement necessary to pass from one to another, and that b lay between a and c by means of the temporal relation between the experience of a, b, and c. On the other hand, someone who was able to see, might be able simply to see that such an arrangement existed. TWM p. 110.
Even though this is an entirely appropriate description of one difference between the two concepts there is, perhaps, a more intuitive characterization available to us. Suppose we look at the matter another way - a way that introduces the concept of the subject of experience more explicitly and which does not presuppose the reality of physical objects. In a visual experience the objects of which I am immediately aware, whether they be considered physical or merely phenomenal, are such that between any two such objects I can imagine a line which joins them. But now suppose we introduce a rule which says that in order to join them I must first draw an imaginary line between myself (from the perspective point of my location) to one of the object before I am allowed to draw a line joining them. Suppose further that space is just all the ways of drawing such paths. In the first case, the one without the rule, space is given simultaneously in the sense that I just imagine a line; whereas in the second case I must imagine drawing a line being drawn from myself to one of the objects before imagining a line that joins them. This second concept of space is what might be regarded as serial space; the first, simultaneous space. In Strawson's world of sounds space is serial, although there is the issue of location of the self or experiencer. On any characterization of the difference, however, there is this complication, a complication which arises in the case of time when we speak of the "specious present": it is possible to imagine what seem to be two sounds given at a single place where place is defined in terms of the pitch of a master sound. Notice also that the space of the sound world is not mixed as in the case I have described, that is, all lines are "drawn" rather than imagined to simply be there. The reason for introducing the mixed conception is that it allows for a "travel" based space even in what Strawson calls a "purely visual experience" (in his reply to Evans), a notion I will have occasion to examine more carefully soon with these notions of space in mind. My second reason for adding this modification is that on Evans's account of serial space the experiencer does not enter essentially, only kinaesthetic sensations. The point is that these sensations must belong to a subject occupying some perspective in that space. Whether in a sound world the person experiencing the sounds must be at a place seems intuitively clear; he is at the place within hearing "distance" of the master sound. For certain purposes we might wish to conceive of serial visual space as space where all lines have to be drawn in time rather than simply imagined to be there instantaneously. In an auditory world the self may be conceived of as a cursor that travels the length of the lines joining the sounds in auditory space.

Should the defender of sound space inject the idea that an occupant of such a world may create a map of his environment and include some out of experience sound particulars, Evans will begin with the criticism that the spatial properties are "once agains conditional in form" (TWM p. 113). But this is exactly the position of an experiencer in our own world who would create or imagine a map of the backside of a physical object. Our belief in the backsides of physical objects is enabled by our belief that if we either walked around the object or rotated it we would have an experience of a, more or less, continuous portion of it, although we would be given a new backside - formerly the front side.

The idea behind Evans's second doubt concerning the possibility of objective reality in a sound world is expressed, perhaps, most succinctly by Strawson himself:

But surely the idea of the simultaneous existence of the perceived and the unperceived is linked with this idea of the simultaneous presentation of elements, each of a definite character, but simultaneously exhibiting a system of relations over and above those which arise from the definite character of either. Surely the former idea is necessarily an extension of the latter, is just the idea of such a system of relations extending beyond the limits of observation. Individuals p. 80
What Strawson's critics, including Evans, have failed to realize, and what Strawson himself may not have appreciated owing to his desire to distance himself from the more traditional metaphysics, is that just as there is the question of how we arrive at the notion of an objectively existing sound in a sound world there is also the question of what ideas are required in order that we may come to believe that in the case of our ordinary space that it is "system of relations extending beyond the limits of observation." Take an ordinary physical object. It's backside is not presented to an observer and yet there is the idea in mind of a simultaneously existing backside located beyond the "horizon" of its edges. How is it that we incorporate this conception into the "givenness" of the object? I think what is essential here is precisely the notion of "travel" space. Were it not for our being in possession of the idea of our capacity to circumnavigate the object there would be no question of there being an object in three dimensional space present to awareness. The points on the surface of the object are, indeed, given simultaneously but this is not sufficient to distinguish a three dimensional object that we take to objectively exist in purview from a two dimensional datum of the imagination. We can walk around objectively existing things in three space; we cannot walk around sense data and scrutinize their backsides. This point is not easy to appreciate in a sound world, since the objects in such a world are sounds and sounds have no backsides. That this is not a frivolous objection to Evans's position is more easy to see following consideration of the main thrust of his criticism of objectivity in the space world, given its most explicit formulation.
The objection is rather this: because serial spatial concepts do not provide us with a way of thinking about simultaneously existing objects, they are not obviously concepts of relations between (independently existing) objects at all.
Again, the proper reply to this is that without serial spatial concepts, expressed in terms of the hypotheticals Evans eschews, there is no available way of thinking of simultaneously existing physical objects in the first place. Patches of color, perhaps, along with the spatial points discernibly present within them; but this is entirely consistent with the phenomenalist world Evans has taken great pains to avoid in his discussion of primary qualities. An issue that needs resolved, but which, because we alone have raised it in this context, is whether the unexamined places are related to examined areas of space in the same way that unexamined space is continuous with examined space in the ordinary world. We may see an object directly or indirectly, in front of us or in a mirror. We do not hear a sound indirectly; nor for that matter do we experience "visibilia" indirectly. This is a notion reserved for things with objective reality. Further, an ordinary thing may be other than it appears, but a sound is only as loud as it appears; a sound does not have properties which differ from those it appears to have.


Strawson's replies to Evans's Simultaneity Argument ("Reply to Evans" p. 277). He has us first consider a "purely visual experience," and by this he means what he terms in the Bounds of Sense a "pure sense data experience." To understand the nature of such experience requires the introduction of simultaneous spatial concepts. Let us assume a basic understanding, only, of such experience. What is to prevent the subject of such purely visual experience from employing a travel based theory of objects existing outside his purview? There is nothing about the Simultaneity Argument that would prohibit this. While Strawson appears to grant Evans's point that a world consisting only of sounds does not supply conditions necessary for having the concept of an objective world, even in a world of pure visibilia, a world without objective particulars, being able to apply simultaneous spatial concepts is not sufficient for possession of the concept of objective particulars in the strong sense of 'objective'. Because a world of pure visibilia, even one which includes simultaneous spatial concepts is void of causally efficacious material particulars, it would by Evan's lights be no more objective than Strawson's auditory world.

As for Evans's causal argument, Strawson (278) maintains "that there is no reason why a subject should not conceive of an objective world of things as constituted of colour variously disposed in space. To the anticipated reply by Evans that objective entities must retain the feature of theoreticity Strawson maintains that the objects for the Seer (who corresponds to "Hero" in the auditory world) in the Seer's world are no less theoretical (279). Strawson is not content to merely parry Evans's criticisms. He goes on the attack, maintaining that Evans's position contains "its own internal instability."

Evans, it will be recalled, argued that conditionals essential to the conceptual underpinnings of the belief in the possibility of an objective auditory world required a ground in dispositions, which, in turn, possessed a categorical basis. What Strawson alleges is that the properties of which the categorical base consists are themselves dispositions and so there is a regress that disables the attempt to ground dispositions.

In concluding his assessment of Evans's criticism, Strawson returns to the problem that Hero is unable to make use of simultaneous spatial concepts. He makes an important concession without conceding his project in its entirety. Owing to Hero's inablity to make use of such concepts, Strawson remarks that "In fact, the case for Hero seems, in the end rather weak." However, such is not the case for Seer who is able to make use of such simultaneous spatial concepts. But one must now raise the question whether the space of Seer is sufficiently analogous to actual space to sustain the belief that in a world of visibilia objectivity is a possibility; that is, whether in a world of pure visual sense data the idea of objects existing out of purview is a reasonable conception. Let's probe just a bit more deeply, making some speculative comments on the way, before drawing this part of the discussion to a close.

The claim (based on Strawson's own remarks Individuals p. 80) is that the idea of a perceived and an unperceived object existing simultaneously is an extension of the idea of two objects spatially related in purview. Are we to regard the idea of two unperceived objects, then, as the extension of the idea of two objects, one out of purview, existing simultaneously in spatial relation to one another? Or, might we take a more radical view, that the idea of there being two objects in purview presupposes the idea that one may travel from one object in purview to another one out of purview? It may, indeed, turn out to be the case that from our self-conscious ability to set two objects within a single field of vision by way of travel we are led to the very notion of space itself! Travel space may, then, turn out to be more fundamental than simultaneous space. Taking this travel based concept as fundamental has one added advantage; the individuation of objects spatially does not have as a sufficient condition their being simultaneously in view. There is an unwarranted assumption that objects out of view present problems solved by simultaneous spatial concepts when as it may turn out this creates problems for the possibility of applying simultaneous spatial concepts. Keep in mind throughout that, in the sense of Poincare, the space of visibia & the space of sounds constitutes a "physical continuum." This conceptualization is particularly useful when considering such things as the circumstance of one sound drowning out another, if such a thing is literally possible.

One must keep in mind that less hangs on this possibility than may be supposed judging by the weight of the arguments that have been offered both in defense and in criticism of Hero. If it should work out that no world other than one consisting in material particulars as basic is possible in support of belief in an objective world, this will only serve to confirm Strawson's original claim, viz. that a metaphysics descriptive of our conceptual scheme requires material basic particulars. Although necessity and sufficiency are to be distinguished, the plausibility of Strawson's claim may survive not withstanding the demise of Hero and the story he had to tell.

We have mentioned the difference between Berlin and Evans on the matter of conditionals. Berlin took the disposition to be that associated with the power of the subject to experience, a passive power; whereas Evans speaks as if what is central is the active power associated with a place which can elicit an experience under certain conditions. Compare

If x goes to y he experiences z

If x goes to y he experiences z-ly.

When it comes to conditionals are we to rule out:

If x goes to y he experiences existentially?

This last sentence reveals the absurdity of treating existence as a power, when that power is expressed in terms of conditionals where the consequent is adverbially related to the existence claim. What is not ruled out, in a similar way, at least, are statements such as:

If x goes to y he experiences an existing y.

Compare the relative scopes of the existential in

There is an x such that if x goes to y he experiences y.


If x goes to y he experiences some z.

The idealist and the phenomenalist must be distinguished. Esse may be percipi for entities other than phenomena. In this case idealism does not entail phenomenalism; nor vice versa. How idealists, and how phenomenalists, view the thesis that esse est percipi differ. Strawson in the Bounds of Sense fails to entertain such a differences. This has the effect of identifying idealism with phenomenalism; thus the idealist thesis makes objectivity an impossibility. This may be in fact the case, but for reasons independent of the dispute now at issue, viz., the status of powers in a phenomenalist ontology.

Wilfrid Sellars offers a different view of what grounds conditionals ("Phenomenalism" in Science, Perceptions and Reality Routledge & Kegan Paul. 1963. 79-80), one differing, that is, from Berlin's and Evans's view that there must be some categorical basis ultimately depending on a material substratum. For Sellars when the phenomenalist speaks of conditionally existing sense data, the conditionals depend on certain generalizations, whether these generalizations are accidental or lawlike does not become an issue, directly. What is important is that the generalizations, if lawlike, connect sense data without any pressing need to introduce physical objects. The connections are between sense data and nothing more; here there is grounding of conditionals without ostensible need for anything like Evans's material basis. However, if one looks at Sellars position more closely one finds that the generalizations at issue involve theoretical terms, at some point, which are correlated by means of correspondence rules to theoretical entities. These theoretical entities may well be thought to be little else than Evans's material object substrata for dispositional terms. But the idea of conditionals connecting only sensations may have met its greatest challenge as a component in the phenomenalists project in the work of Braithwaite and G. F. Stout. Stout who distinguishes our sensible awareness of things like patches of color from perceptual awareness of things having color, follows Braithwaite in this regards and issues a most compelling challenge in the following way:

...material objects are introduced as conditions on which the occurrence of actual sensations depend. I may have no means of indicating a certain train of sensations except by saying they are sensations which would accompany a certain movement of my arm. I may also indicate another sensation by saying that it is that which follows contact of my hand with the table. But the first sense-experience is not by itself the condition on which the second depends. This must include my body and the movement of my arm in relation to the table as part of my physical environment. What is true in this instance is true in all cases in which we attempt to specify the conditions of actual or possible sensations. These always involve physical factors which have not themselves been analysed into senstaions actual and possible. Thus the case for phenomenalism entirely breaks down. Proceedings of the Aristotelian Society 1938-1939. p. 16.
Notice here that in a world of sounds where there is no distinction between sensible awareness and perceptual seeming the important distinction is difficult to make. Once we take into account the two senses of objectivity Strawson locates in Kant we have to ask not only the question how it is that we justify conceptualizing a sort of entity we call a physical as opposed to the pure sense data experience, we must justify a difference between experiences that have an object, such as an awareness of a flash, and experiences that lack such an object, for example the feeling restlessness. It will be recalled that Broad maintained against Russell's neutral monism, a radical view where both physical objects and minds ultimately come down to sense data (which were "neutral"), that even this latter difference could not be maintained in a pure sense data world. One must ask a more general question as to whether objectivity in the weighty sense (which I take to correspond to the sense in which our "epistemological objects" have built into them the conception of the possibility of a corresponding "ontological object") and objectivity in the not so weighty sense, which I take to correspond to the notion of being an "objective consitituent," are mutually dependent or such that the idea of having an objective constituent depends on that of being an object which may exist independently of having experience. (cf. Mind and Its Place in Nature p. 581-82.)

* This work is dedicated to the performers and organizers of the Bach Society Orchestra on the occasion of its 50th. anniversary season.


Berlin, I. "Empirical Propositions and Hypothetical Statements," Mind (lix) 1950. Reprinted in Robert schwartz Perceiving, Sensing, and Knowing Anchor Books.

R. B. Braithwaite, ‘Propositions about Material Objects’, Proceedings of the Aristotelian Society, vol. 38 (1938)

Broad, C. D. The Mind and Its Place in Nature. Routledge & Kegan Paul LTD. 1925.

Evans, G. "Things Without the Mind (TWM) - A Commentary upon Chapter Two of Strawson's Individuals," in Philosophical Subjects: Essays Presented to P. F. strawson, edited by Zak Van Straaten, Clarendon, Oxford, 1980. p. 76.

Mackie, J. L. Truth, Probability, and Paradox. Oxford. 1973.

Rorty, R. Philosophy and the Mirror of Nature Princeton. 1979.

Sellars, W. Phenomenalism" in Science, Perceptions and Reality Routledge & Kegan Paul. 1963.

Stout, G. F. "Phenomenalism" Proceedings of the Aristotelian Society 1938-1939.

Strawson, P. F. Individuals: An Essay in Descriptive Metaphysics. Anchor. 1959.

Strawson, P. F. Bounds of Sense Methuen. 1966.

Strawson, P. F. "Reply to Evans" in Van Straaten. 1980.