According to a thesis which has become familiar as Revelation, perceptual experiences place conscious subjects in a position to know the essence of the properties presented as instantiated by the experiences. Those could either be subjective, sensory properties; or objective, sensible properties – such as the colors. If Revelation is the case, experiencing such properties places us in a position to know exactly and fully what they are like. For example, supposing Purple is essentially compound in hue, experiencing Purple should put one in a position to know this.
While the originators of the thesis took it to enjoy the support of common sense, more recent discussions of this issue have concluded that Revelation is not thus-supported. In turn, the impression that Revelation is the case gets traced back to the plausible and superficially similar – but more modest – claim that conscious experiences acquaint subjects with such properties.
In my view, the critics are mistaken: Revelation can be shown to be common sense. However, a number of obstacles stand in the way of framing the thesis in a form amenable to this. And within the confines of the present discussion, my goal is to do the preparatory work of removing obstacles. Specifically, I will formulate and remove the most serious obstacle to the view that Revelation is common sense: the difficulty in understanding how our experiences could provide the justificatory basis for beliefs in the essentiality of the manifest higher-order aspects of sensed properties, e.g. of Purple’s complex hue. The argument here does not establish that Revelation is common sense, but shows this to be a live possibility.
What is at stake in the debate over the status of Revelation as a common-sense claim is whether there are significant pre-theoretical constraints on the nature of the relevant families of properties. That it seems so can be the gathered from the immediate plausibility of anti-physicalist arguments concerning such properties. Unless something substantial concerning the nature of, e.g. colors or color experiences were revealed through the experiences manifesting them, arguments to the effect that colors or color experiences are not physical could not get off the ground.Footnote 1
Revelation is a thesis concerning the essences of a family of properties, which I take to be collections of higher-order properties. A natural way to understand it is as requiring for its truth that experiences of a property should – along with the property’s second-order essential aspects – also reveal a third-order aspect: their being essential to the first-order property. Call this the Essentiality Construal of Revelation (EC).
Many find EC implausible, since on the face of it our sense experiences can merely inform us of the way presented items are, not of whether they are thus essentially or not. For this reason, the majority of commentators now favor an alternative, weakened version of Revelation. The latter requires for its truth merely that sensing a property should put the subject in a position to recognize that the property bears the respective higher-order aspects – and not, further, that it bears them essentially. Call this the Weakened Construal of Revelation (WC).
I will argue that only EC deserves attention. WC encounters a serious problem: it manages to maintain Revelation as a live possibility at the cost of depriving the thesis of prima facie plausibility. In this weakened form, the thesis might well be true, but could not be common-sense. For that reason, WC is useless to those who regard common sense as imposing constraints on the nature of sensory and sensible qualities.
Based on this result, I set up a dilemma for friends of Revelation. If I am right, WC could not be part of the common-sense view of the relevant properties. On the other hand, EC seems outrightly implausible. So either way it seems that common sense could not impose significant constraints on the metaphysics of the respective family of properties.
As a way out of the dilemma I argue that the apparent implausibility of EC is due to an assumption that essentiality would get revealed in experience in the guise of necessity (if at all). The latter would saddle experiential contents with an exceptionable degree of complexity. But this assumption is dispensable. I propose an alternative model of how an experience as of a property might provide justification for claims concerning the essentiality of its higher-order aspects – and in a way that would not saddle experiential contents with an inadmissible degree of complexity. On my proposal, the essentiality of the manifest essential aspects of a property would get conveyed in the guise of their apparent intrinsicality. First, just as appropriately attending to a manifest instance of a property, e.g. to a particular shape, can sensibly place one in the position to recognize the shape’s intrinsicality, appropriately attending to an essential higher-order aspect of a property of the relevant sort, e.g. to the distinctive hue of a shade of Purple, would put one in the position to recognize its intrinsicality. Secondly, since – as I argue – with higher-order properties intrinsicality is a priori sufficient for essentiality, experiences manifesting the intrinsicality of a higher-order aspect of a property could also provide a justification for the further claim that the aspect is had essentially.
In brief, with the right conceptual background in place, telling – based on experience – that a manifest higher-order property is essential should work no differently from telling – based on experience – that a manifest first-order property is intrinsic. So despite the initial appearance, EC does not saddle the contents of our experiences with any complexity beyond what many are happy to admit for other purposes.
1. Revelation: the basics
Revelation was contemporaneously introduced by Mark Johnston (Reference Johnston1992) and David Lewis (Reference Lewis1995). Johnston took the thesis to concern our epistemic situation concerning sensible properties: properties presented by sense experiences as instantiated by external items. Paradigmatic among those Johnston takes the colors. By contrast, Lewis took it as concerning our epistemic situation regarding sensory properties: subjectively available properties of the experiences themselves (commonly known as qualia). However, Lewis allowed that Revelation might also (or instead) apply to sensible properties. I will work with the version of the thesis concerning sensible properties, and specifically colors: for simplicity, and because it is more natural to attribute to common sense. Towards the end I will say something about how the proposed account of telling essence could be extended to sensory properties.
Throughout I aim to remain neutral among the two most plausible metaphysical accounts of perceptual experience: relationalism and representationalism. I reject non-intentional qualia for independent reasons – having to do with the so-called Transparency of Experience. But I aim my proposal to be compatible with both representationalism and relationalism.
Lewis and Johnston took Revelation’s plausibility with regard to the respective family of properties to be due to its plausibility with regard to specific properties within the family – specific shades and specific phenomenal properties, respectively:
The intrinsic nature of canary yellow is fully revealed by a standard visual experience as of a canary yellow thing (Johnston Reference Johnston1992, 223).
[W]hen I have an experience with quale Q, the knowledge I thereby gain reveals the essence of Q: a property of Q such that, necessarily, Q has it and nothing else does (Lewis Reference Lewis1995, 143).
The overall thesis gets derived by generalizing from particular cases to the respective family of properties.Footnote 2
To understand the thesis, we need to get clear on the terms it contains.
First, essence. Lewis takes the essence of an item to be a property necessarily had by the item. In the case of the essence of a property or a kind, there is an added condition that nothing else has the property. The definition is vulnerable to the standard objections to modal accounts of essence (Fine Reference Fine1994). Many properties meeting Lewis’ criterion fall outside the essence of the item at issue. Johnston seems to get around this problem by choosing to elucidate ‘essence’ in terms of other terms he does not fully define. Specifically, Johnston defines ‘essence’ as ‘intrinsic nature’. The elucidation seems to have the following shape. The essence of an item just is its nature: it is a collection of higher-order properties. The qualification ‘intrinsic’ indicates that those properties are a subcategory of the intrinsic properties of the item.Footnote 3 With this, Johnston might be gesturing towards something akin to Denby’s (Reference Denby and Francescotti2014) analysis of essence in terms of a combination of intrinsicality and necessity. Alternately, Johnston’s aim might be to merely uncover certain metaphysical connections between essentiality and intrinsicality, without a reduction of essentiality.
For my argument’s purposes, it is useful to take Denbys’s view of the matter, but I need not be committed to a reductive view of essence. The claims i. that the essence of an entity is a property the entity must possess to be the sort of thing it is (Fine Reference Fine, Sinnott-Armstrong, Raffman and Asher1995, 53), and ii. that essential properties are intrinsic would suffice. We speak of the essence of an item as what the item is on its own, or in its own right, and it is natural to take the above two claims as implicit in this pre-theoretical understanding.
Additionally, while I assume, for simplicity, that no essential property is extrinsic, and that every apparently essential extrinsic property can be reduced to intrinsic properties, or else eliminated, Footnote 4 since the ostensible metaphysical connection between essentiality and intrinsicality seems vulnerable to counterexamples, it is important to point out that my argument does not hinge on this assumption. While there seem to be robust counterexamples to the view that first-order essences are intrinsic, there do not seem to be similarly robust counterexamples to the view that second-order essences, the essential features of a property, are intrinsic. Strictly speaking, only the latter claim is needed for my argument, and is on a more secure footing than the claim that all essential properties are intrinsic.Footnote 5
Secondly, getting revealed. The precise epistemic role accorded in the thesis to sense experience was specified by Stoljar (Reference Stoljar, Nola and Mitchell2009). He divides Revelation into: 1) a thesis about the sort of understanding we have of the respective properties, and 2) a thesis about the epistemic basis of such understanding. With some sensible/sensory properties then, the claim is that we can learn the totality of essential facts concerning those properties on the basis of experiencing apparent instances of them. It is important here to make the qualification that this could not sensibly mean that a single unattended experience as of an instance of a property in a conceptually unsophisticated subject would give the subject a complete propositional knowledge of the essence of the property. Rather, as Byrne and Hilbert have clarified, ‘[i]t is consistent with Revelation that the full nature of canary yellow will only be apparent after a diverse range of color experiences including, perhaps, experiences as of transparent canary yellow volumes, canary yellow lights, and canary yellow objects against a variety of backgrounds’ (Byrne and Hilbert Reference Byrne and Hilbert2006, 77). So the claim in question is that upon a careful reflection on a sufficient range of experiences in which the property figures, a suitably conceptually-equipped and attentive subject should be able to come to know the collection of propositions determined by the property’s essence.Footnote 6 For instance, in the case of a particular shade of Purple: that it is bluish, that it is reddish, and so on.Footnote 7
2. Revelation and essentiality
Even with the heavy-duty terms out of the way, Revelation remains underspecified. On one specification, the thesis would state that experiencing a property of the relevant sort places the subject in the position to know the essentiality of the essential aspects of the property. Earlier I called this the Essentiality Construal (EC). By contrast, on what I called the Weakened Construal (WC), the thesis would state that having the experience would place a subject in a position to know that the sensed property has the essential aspects, without putting them in a position to know that they are essential.
The difference amongst the two construals has been illustrated by Byrne and Hilbert:
“suppose (no doubt contrary to fact) that it is in my nature that I make a certain journey. It may be revealed that I make the journey, without it being revealed that this is in my nature” (Byrne and Hilbert Reference Byrne and Hilbert2006, 76).
Just as the essentiality of one’s making a certain journey need not get revealed when someone is told that they will make it, the essentiality of Purple’s compoundedness need not be revealed by the range of experiences which would allow one to recognize its compoundedness.
Johnston is understood by Byrne and Hilbert to be a proponent of the EC. Lewis, on the other hand, has been read as favoring WC (Stoljar Reference Stoljar, Nola and Mitchell2009). I will show that for their purposes both originators need to endorse EC. To this this effect, I will make clear that adherents of common sense as a guide to the relevant metaphysics generally have a reason to favor EC.
Recall that both Johnston and Lewis consider Revelation to be part of the common-sense view of the relevant families of properties. In my view, the best way to interpret this is as the claim that Revelation is prima facie plausible: that is, plausible given evidence which is commonly-available. Both Johnston’s and Lewis’ arguments hinge on Revelation’s enjoying this standing. Both treat the thesis as imposing constraints on the nature of the respective family of properties. But if Revelation were not prima facie plausible, it would not carry the authority needed to constrain the metaphysician. So Johnston and Lewis might be wrong in this regard, but for their purposes – to treat it as imposing some intuitive constraints on the nature of the respective properties – they must regard it as prima facie plausible.Footnote 8
By contrast, further potential implications of the claim that Revelation is common-sense: that Revelation is commonly believed, or is easy to formulate and endorse with justification – seem considerably less relevant to Lewis’ and Johnston’s arguments, and have the potential to side-track. Being prima facie plausible clearly lacks such implications. It does entail that, with the conceptual background needed to entertain the relevant proposition, and using the skills needed to exploit one’s evidence, a subject should be able to form and endorse the proposition with justification. However, meeting those conditions might be rare.Footnote 9
The significance of Revelation’s prima facie plausibility comes to bear on the eligibility of the two construals of the thesis: EC and WC. If the Essentialist Construal were the case, Revelation would automatically enjoy the needed prima facie plausibility. Experiences would contain enough in them to justify beliefs in the essentiality of the manifest higher-order properties. By contrast, if WC were the case, it does not automatically follow that Revelation would enjoy a prima facie plausibility. For this, there would additionally need to be a common source of evidence for the thesis, other than experience. For on WC, on its own experience could only justify beliefs of the form: i) that a manifest property P has such and such higher-order properties, and not of the form: ii) that the higher-order properties are had by P essentially. But it is not clear that there is a common source other than experience, and better eligible than it to support beliefs of the second form.
This difference in the consequent epistemic standing of the thesis on both construals is probably why EC is favored by Johnston. Consider for instance, the following passage:
[O]ne naturally does take and should take one’s visual experience as of, e.g. a canary yellow surface, as completely revealing the intrinsic nature of canary yellow, so that canary yellow is counted as having just those intrinsic and essential features which are evident in an experience as of canary yellow (Johnston Reference Johnston1992, 223).
One claim Johnston is making here is that Revelation is prima facie plausible by virtue of its being rational for us to treat particular manifest higher-order aspects of particular sensed properties as essential, and generalize from such cases to the entire family of properties. But he also seems to be making the further claim that our perceptual experiences would not just be apt to cause beliefs in the essentiality of the manifest aspects but, further, that this would be a rational response to the underlying experiences in light of the information contained in them.
Some passages in Lewis suggest that he shares this view of the source of Revelation’s plausibility.Footnote 10 True, Lewis does not think that Revelation is the case – but this does not prevent him from according it intuitive plausibility, and tracing this plausibility back to the way the properties are presented in experience. So the originators’ position seems to be that it is rational to take the essence of a particular manifest property to be constituted by its manifest higher order aspects in light of the essentiality of those aspects getting revealed in experience.
Let us take stock. For Revelation to be accorded consideration by metaphysicians, the thesis would need to enjoy prima facie plausibility. The most straightforward and plausible way to make sense of this hinges on experiences’ aptness to manifest the essentiality of the second-order aspects of sensed properties.Footnote 11
3. A challenge for the essentiality construal
Having argued that EC is better suited to vindicate the view that common sense provides a significant constraint on the metaphysics of the respective family of properties, I will now explain why many have tended to steer away from this construal. Anticipating: on the face of it EC requires too much of our experiences, and specifically of their contents.
To begin with, it should be clear that EC demands more of experience than WC. Specifically, on the former more information would need to be contained in experience than on the latter. In other words, the representational contents of experience would have to be richer. Moreover, the discussion in the previous section made it clear that there is a requirement that the contents should be apt to justify beliefs in essentiality. But on the face of it not all kinds of perceptual content can play an adequate justificatory role. For instance, in cases of blindsight (Weiskrantz Reference Weiskrantz1986), it is plausible to attribute to subjects visual representations of certain conditions of objects in the environment – such as their shape and orientation. This might sometimes lead to the formation of corresponding beliefs. But consistent with this, a blindseer’s visual state does not seem apt to justify those beliefs.Footnote 12 So it seems that to do the needed justificatory work the relevant information would have to be conveyed by the phenomenal character of sense-experience.Footnote 13 This intuition might ultimately be given up, in the context of an externalist account of justification, but since we are working within the vantage point of common sense, we should aim to accommodate it.Footnote 14
Summing up: EC seems to demand richer justificatory powers from experience, and, consequently – richer phenomenal contents than WC. The evident and potentially excessive demand is the inclusion of essentiality in phenomenal contents. To many the proposed inclusion has seemed utterly implausible. But their reasons are either unclear or unpersuasive. The aim of this section is to present an effective argument for the inadmissibility of the proposed inclusion.
To begin with, the inclusion of essentiality in phenomenal contents is not a problem, unless this would make phenomenal contents inadmissibly complex. But the admissibility of experiential contents is tied to their explanatory adequacy. We should prefer simpler contents only if we are able to account with them fully for the phenomenal character and cognitive functions of experiences. In many cases, there is a reason to adopt a view of content which is more complex than an alternative. For example, contents with predicational structure may be more complex than contents without such structure, but it is clearly preferable to take phenomenal contents to have predicational structure, since introspection suggests that in at least some experiences the sensed properties appear to characterize sensed particulars (be it objects or events). So greater complexity of contents does not automatically mean lower plausibility. And in the particular case proponents of Revelation could argue that such contents are needed to make sense of the full epistemic role of sense-experience.
Still, there are some general constraints on the nature of experiential contents. For one, we want to allow that when it comes to low-level properties, such as the colors, naïve perceivers without significant conceptual repertoire can have the same experiences as conceptually sophisticated subjects. And the representation of essentiality might be taken to require capacities that naïve perceivers arguably lack. For example, Stoljar and Lihoreau have claimed that for EC to be the case, phenomenal contents would need to be constituted by the deployment of a concept of essence. Surely this consequence would be inadmissible. We want to allow that subjects who do not possess a concept of essence can enjoy the same sorts of color experiences which we enjoy.
That said, it is unclear why those critics think that the representation of essentiality by experience would involve the deployment of a concept of essence. Many now allow that a property can be experientially represented as instantiated without the subject’s needing to possess a concept of that property. So the critics of the strong construal owe us an argument as to why this could not be the case with Being Essential.
An argument to this effect suggests itself when we consider what is distinctive of essentiality – as opposed to other properties that are admitted in experiential contents. Essentiality has been taken by many to be a modal notion. Additionally, many have taken experiences to be silent on modal matters.Footnote 15 If both claims were true, phenomenal contents could not inform us about the essentiality of the manifest aspects of sensible properties. From here on, one might derive Stoljar’s and Lihoreau’s conclusion by arguing that the only way for the proponent of Revelation to resist this view of the representational limits of experience is to maintain that experiential contents are in part constituted by the deployment of the relevant modal concepts. However, the prior, and more general, criticism is that it may not be open to proponents of Revelation to explain how experiences could represent modal properties, without adopting a view of the contents of experience that has implausible consequences for the sorts of representational capacities needed to have experiences.
If this is the basis of the widely-felt discomfort with Revelation on the Essentialist construal, it is important to point out that the main assumptions behind it are not yet fully secured.
Specifically, the view that some modal properties are represented by experience is becoming common.Footnote 16 So an argument for the inadmissibility of matters of essence from the general inadmissibility of modal matters in experiential contents is bound to be ineffective. Still, the real problem turns out to lie downstream. While it is not out of the question that some unactualized possibilities would get represented in experience, it does not follow that necessities could also get represented in experience. On the face of it, representing possibilities and representing necessities would involve distinct capacities.
To this, the proponent of Revelation might respond by proposing that the second capacity derives from the first. Here is a possible way to do this. Representing necessity is equivalent to representing the exclusion of a particular type of scenario from modal space. One could represent a particular necessity in the guise of the impossibility of the corresponding scenario. So the friend of Revelation might argue that the capacity to represent necessities would fully reduce to the capacity to represent possibilities.
However, this response proves inadequate. To represent necessities, experiences would need to be able to represent not just unactualized possibilities, but unactualized possibilities as impossible. Here is how one might hope to account for this capacity. To represent unactualized possibilities as impossible, all that one would need would be the capacities to represent merely possible and negative states of affairs. But having those two capacities would suffice for having the third only if the representational capacities involved in experience were compositional. And it may well be that experiential representational capacities do not compose in the way that conceptual capacities do. While your ability to conceptually represent possible and negative states of affairs would guarantee that you are able to represent any possibility you can conceive of as impossible, your ability to represent negative and possible state of affairs otherwise than in a system of concepts does not guarantee the further ability to represent a possible state of affairs as impossible.
In fact, non-conceptualists about experiential content are nearly obliged to say this, as the best way to make sense of the conceptual/non-conceptual distinction is in terms of representations of different sorts being involved in perception and in thought. Whatever else must be distinctive of perceptual representations, it seems that they are not likely to be subject to the demand for recombinability that concepts are subject to. So to fully develop the proposal, the proponent of Revelation would need to adopt conceptualism about experiential content, and thus forced into an implausible view.Footnote 17
Faced with this problem, the Revelation theorist might take a different route and deny that experientially representing essentiality is a matter of representing necessity. In this they may be sensitive to another problem that emerges when essentiality is taken to be a modal matter. For even if a successful reduction of essentiality in terms of necessity plus some further condition were forthcoming, essentiality would not be wholly a matter of necessity. But the above, problematic proposal could at best establish that sensory experiences reveal necessities. So the Revelation theorist has an independent reason to reject it. From this point on, they might insist that the capacity to represent essentiality is irreducible to other capacities and must be posited if we are to make sense of the full epistemic role of sense-experience. But here they would be on shaky ground, since the revelatory role of sense experiences is in dispute. So on this alternative proposal the inclusion of essentiality in experiential content would be ad hoc. In what follows, I propose a better response, on which the epistemic role of experience could be made sense of otherwise than by positing a primitive capacity to represent essentiality.
To sum up the results in this section: taking the essentiality of the manifest aspects of sensible properties to get revealed in experience commits one to more than the (plausible) claims that experiences represent unactualized possibilities and negative states of affairs. The capacity to represent essentiality would not reduce – and most likely be entirely unrelated – to those two capacities. Consequently, they could not provide the basis for a bootstrapping argument in favor of Revelation’s feasibility.
4. Essentiality through intrinsicality
I have argued that only on the Essentialist Construal does Revelation stand a chance of enjoying prima facie plausibility. On the other hand, the Essentialist Construal faces a challenge in accounting for how our sense-experiences could play the requisite justificatory role – given that essentiality must be revealed in the guise of necessity, if at all.
With any how-possible question, there are two sorts of potential response strategy (see Cassam Reference Cassam2010). The above, failed strategy constitutes an obstacle-overcoming response to the challenge: it is aimed at addressing the challenge on its terms. By contrast, an obstacle-dissipating strategy would seek to address the challenge by rejecting or at least revising some of the assumptions in its basis.
A response of the latter sort to our challenge suggests itself once the assumption that essentiality must be presented by experience in the guise of necessity gets questioned. With this assumption on board, we encountered problems both with the hypothesis that necessities could get represented in experience, and with the suggestion that this should suffice for essentially to get revealed in experience. Supposing that an alternative hypothesis of how essentiality would get revealed in experience were provided on the basis on an undisputed or at least a less controversial capacity to represent a sort of property, the challenge might get sidestepped.
4.1 Taking a new path
Many allow that the intrinsicality of some manifest properties gets revealed in experience. For instance, shapes are commonly taken to be paradigmatically intrinsic properties. Here is David Lewis considering the proposal that shapes are disguised relations, instead of genuinely monadic, intrinsic properties:
This is simply incredible, if we are speaking of the persistence of ordinary things… If we know what shape is, we know that it is a property, not a relation (Lewis Reference Lewis1986, 205).
Lewis is not merely playing with the view that shapes might be intrinsic. His argument from temporal intrinsics could not work under the weaker assumption that shapes are instantiated by some objective items, but might turn out to be extrinsic.
The reason for which Lewis and others consider shapes to be intrinsic properties is that shapes seem to be presented as intrinsic by our experiences of shapes. Shapes are properties of which we seem to have substantial understanding, based on everyday observation (Brewer Reference Brewer2011, 27). If we regard them as intrinsic, this must be because they are revealed as such by our experiences. To sum up, many regard shapes as intrinisic, and this must be because they take shapes to be presented as such in experience.Footnote 18
Here is my proposal, in outline. Suppose that some sensed first-order properties appear intrinsic. This would mean that we have the capacity to represent intrinsicality in experience. Suppose, further, that some sensed first-order properties such as colors and shapes appear to instantiate second-order properties. There is no reason why the capacity to represent intrinsicality should be confined to first-order properties. But if it extended to the second-order properties, I argue that with the right conceptual background the experiencing subject would be able to derive their essentiality from their apparent intrinsicality.
Clearly, not all intrinsic properties are essential. For instance, the shapes of physical objects are not essential to the objects. But recently Denby (Reference Denby and Francescotti2014) has argued that all essential properties are intrinsic. According to him, essential properties are intrinsic properties possessed necessarily. There are a lot of interesting issues surrounding this proposal. Specifically, Denby has argued that it best explains the constraints anything must meet to exist.
On the face of it, taking Denby’s view still brings us back to the earlier problematic proposal about how essentiality could get revealed in experience. For it would seem that experience would have to represent both the intrinsicality and the necessity of a higher-order property. However, I will argue that nothing beyond the representation of intrinsicality is needed from experience, since with higher-order properties intrinsicality is a priori sufficient for essentiality.
A natural way to explain the ontological distinction between particulars and properties is by accounting for the intuitive asymmetry in the relation of instantiation. David Armstrong has proposed that the asymmetry is due to what he calls the Principle of Instantial Invariance: that while a particular can instantiate any number of properties, a property cannot be instantiated by a varying number of particulars (Armstrong Reference Armstrong1978, 94). The principle is intuitive: denying it has a number of odd consequences, such as collapsing the distinction between properties and relations (Armstrong Reference Armstrong2010, 24). In its basis is a more general principle concerning properties: that a property must be strictly identical in all its instantiations, or to put it otherwise: that all its instantiations must resemble each other perfectly. Since intrinsic properties are those responsible for resemblance among instances, it follows from this principle that all possible instances of any real property will have a uniform intrinsic character. Thus a property cannot instantiate a given intrinsic property in one instance, and a contrary intrinsic property in another. From this it follows that the property is had necessarily, and – being intrinsic, essentially.
So with properties, an intrinsic difference means a difference in the property instantiated. This result has an important epistemological consequence. Since the intrinsicality of a higher-order property would a priori entail the essentiality of the property, perceptual evidence for the intrinsicality of a higher-order property would also constitute evidence for its essentiality.
Now suppose that a sensed instance of Purple appeared compound in hue, and that the bluish-reddish hue exhibited intrinsicality. I maintain that an experience of this sort should suffice to put a suitably conceptually-equipped subject in the position to know that the distinctive bluish-reddish hue is essential.
It has been pointed out to me that on the current proposal recognizing essentiality on the basis of the appearance of intrinsicality would need to involve a background understanding of the connection between intrinsicality and essentiality.Footnote 19 It is not fully clear to me why this outcome should be counted problematic. Perhaps it is because the source of justification for claims concerning the essentiality of those aspects would turn out to be partly intellectual. Earlier, I argued that it is implausible that Revelation is justified on a purely intellectual basis. On the current proposal, one’s belief in the essentiality of the reddishness and bluishness of Purple would be justified in part by experience, and in part a priori. The source of justification would not be purely experiential, but it would not be purely intellectual, either.Footnote 20
Or perhaps the proposal seems objectionable because the relevant theoretical background needed to form the relevant beliefs would commonly be lacking. How could Revelation be part of common sense, then?Footnote 21 Here it is important to remind the reader that by ‘common-sense’ I mean prima facie plausible. For a claim to be prima facie plausible, the requisite theoretical background need not be easily available. In the absence of it, the experience would still provide one with a justification for the claims that the manifest higher-order properties are intrinsic, and consequently – essential. Making use of one’s justification in each case would require drawing on relevant conceptual resources: one’s grasp of intrinsicality, in the one case, and of intrinsicality, essentiality, and their connection, in the other.
Perhaps the real worry behind objections of this sort is that it must be a non-negotiable requirement of the strong construal of Revelation that one’s justification for claims like ‘Purple is an essentially compound color’ should be immediate. It is useful here to distinguish between two sorts of immediacy concerning beliefs: psychological and epistemic immediacy.Footnote 22 Many of our beliefs are formed on the basis of automatic inferences. Those beliefs are psychologically immediate in that they do not issue from any conscious inference. But their justification in many cases would derive from the justification of more basic propositions. This would be the case with our beliefs regarding the essentiality of higher-order properties. With the right conceptual background, such beliefs would be formed non-inferentially in response to investigating what sensed properties are like. Still, their justification would derive from the justification available for believing more basic propositions such as ‘Purple is intrinsically compound, ’ and ‘The intrinsic character of properties is essential.’
Still, on my view it might seem implausible that the psychologically immediate beliefs we form should concern the essentiality of the relevant higher-order properties. Should we not expect them to concern the intrinsicality of those properties, instead? I maintain that this is not at all obvious. Consider an analogy: when you observe a tomato-like object, with the right conceptual background you tend to form beliefs attributing the kind Tomato with psychological immediacy. But it is likely that your justification for such beliefs derives from justification for claims attributing a complex observational property, call it Tomato-gestalt, and the a priori association between tomatoes and tomato-gestalts, encoded in the observational concept TOMATO.
The novel proposal of how Revelation could be implemented by experience has promise. For, on the face of it, some properties do appear intrinsic in experience. Additionally, there is no reason why manifest second-order properties could not appear thus, if first-order properties can. Provided that the relevant second-order properties appeared intrinsic, enough information would be conveyed by experience for beliefs attributing essentiality to the second-order properties to come out justified.
4.2 Is the proposed solution genuine?
The most natural objection to the present proposal of how Revelation could be implemented is that it seems to face the same sort of problem as the old proposal – which appealed to necessity. On the old one, it turned out that there is no way to vindicate Revelation, since it was unclear how necessities could get represented in experience. But an analogous problem may seem to arise for the present proposal: namely how the intrisicality of a property could get represented in experience. Consequently, the present proposal would constitute an article of faith, rather than a genuine explanation, as it does not address the problem.
In response, I want to make two points. The first is that both proposals do not face the same sort of challenge. The second is that there is a straightforward way to make sense of the experiential revelation of intrinsicality, based on the capacity to exercise perceptual attention.
A how-possible question would be based on a particular, highly intuitive or empirically supported claim, which serves as a prima facie obstacle to a sort of capacity we might take ourselves to possess. So how-possible questions are obstacle-dependent questions. In the absence of a clear and serious obstacle, how-possible questions are infelicitous:
We ask how x is possible when there appears to be an obstacle to the existence of x. We don’t ask how x is possible if there is no perceived obstacle or no inclination to suppose that x is possible. So, for example, we don’t ask how baseball is possible or how round squares are possible (Cassam Reference Cassam2007, 2).
We have the right to infer that something is possible, when we have evidence that it is actual, and no defeaters. Now suppose that there is independent evidence, as many have claimed, that our experiences represent the intrinsicality of some manifest properties. Absent some reason why it should not be possible to represent intrinsicality, we would be entitled to deem it possible. By contrast, to my knowledge no one has claimed that there is independent evidence that we represent necessities. This is why the capacity to represent necessities needs to be made plausible by being shown to reduce to uncontroversial representational capacities. There is no need for an analogous argument in the case of intrinsicality. What is needed is merely a defense of the claim that there is independent evidence that our experiences can represent intrinsicality. This is analogous to the need to show that there that our experiences represent possibilities. From then on, there would be no further problematic steps in the argument.
One might argue that a further argument is needed to establish that it is possible for us to experientially represent the intrinsicality of second-order properties, such as Purple’s reddishness. I disagree. If you grant me that some first-order properties appear intrinsic, and also that some sensed first-order properties appear to instantiate second-order properties, we should expect it to be possible that second-order properties would also be represented as intrinsic. In the case of representing necessities, there was a genuine gap between the capacities to represent possibility and necessity. But in this case the same representational capacity would be exercised in both the first-order and the second-order case.
So the only way to attack my proposal is to either argue that there is no evidence for the view that we represent intrinsicality experientially, or provide a clear obstacle to the overall possibility of experiential representation of intrinsicality. Concerning the latter option, I am not aware of any genuine obstacle. It might be argued that intrinsicality is also a modal matter, and, therefore, that the proposal is subject to the same worries as the proposal that essentiality gets revealed in experience in the guise of necessity. But even if intrinsicality were fundamentally a modal matter, experiences would not need to represent it on that basis. Compare: even though water is fundamentally H20, we can represent water otherwise, and in a much simpler manner, than as H20. And beyond this, the modal analysis of being intrinsic is roughly: instantiable regardless of accompaniment. So the relevant modality would be possibility, and the relevant possibilities would be nearby. For a shape to appear intrinsic, it would merely have to appear as capable of existing regardless of the existence and state of objects other than its bearer. But it seems to me that shapes appear in experience precisely this way. On the face of it, then, there is nothing problematic about the proposal that experiences can represent intrinsicality.
Hence, the only point at which my proposal can be targeted is the claim that we have evidence that our experiences represent intrinsicality. According to some (see Byrne Reference Byrne2001), this impression arises from mixing up lacking evidence that sensed properties are extrinsic with having evidence that such properties are intrinsic. They are certainly free to hold this view, and have their reasons, but I do not think that friends of intrinsicality are obliged to convince them. For the claim that something falling short of intrinsicality is attributed in experience in the place of intrisicality has a skeptical flavor, Footnote 23 and to defend common sense from skepticism one is not obligated to respond to skeptics on their own terms. The account below of how intrinsicality would get revealed in experience is meant to serve as a reinforcement of common sense, not to convince skeptics.
4.3 Attention and apparent intrinsicality
I have argued that the proposal that the essentiality of the manifest aspects of the colors gets revealed by experience in the guise of their intrinsicality does not face the sorts of obstacles faced by the proposal that it gets revealed in the guise of their necessity. Still, it is useful to explain exactly how experiences would reveal intrinsicality.
Central to my account is the proposal that exercising attention can allow you to register the intrinsicality of some perceptually manifest properties. Here by attention I mean the capacity of a subject to focus on an item, even if they are perceiving other items in the scene at the time. This sort of attention would make the item stand out, get highlighted, or foregrounded for the subject. It is also the sort of attention that is needed for demonstrative thought concerning perceived items (see Campbell Reference Campbell2002; Smithies Reference Smithies2011).
I propose that this sort of attention can make you aware of the intrinsicality of some properties. Here is how. In foregrounding an item, you are able to abstract from the fact that it is experienced together with other items. In turn, this allows you to become aware of its independence from the existence and condition of any other item. This view has been held concerning individuals (see Campbell Reference Campbell2002; Campbell and Cassam Reference Campbell and Cassam2014), but can be extended very naturally to some of their perceptually available properties. Through an exercise of the same sort of attentional mechanism, some properties would appear to depend solely on the existence and internal condition of their bearers. In the relevant cases, while one could attend to the property independently of attending to other items in the scene, one could not attend to it independently of attending to its bearer, or at least to the location at which it appears to be instantiated. So the quality would appear to depend for its existence solely on the state of that individual or of that region of space. This, I argue, is exactly how physical shapes and steady colors appear to us.Footnote 24
Besides being able to attend to manifest properties, we also seem able to attend to their higher-order manifest aspects. For instance, I can attend both to the apparent roundness of a perceived object, and to the apparent distinctive symmetry of the roundness. Similarly for colors: I can attend both to the apparent color of an object, and to various aspects of the color, such as hue, saturation, brightness, and so on. As I have indicated, the capacity to attend to such higher-order properties might acquire a lot of other capacities as background: most clearly, having the relevant concepts. But the result in both cases would be that my attention would allow me to isolate the property and its bearer from anything else experienced. Based on this, the bearer would come to seem uniquely relevant to the instantiation of the property: nothing else would appear to matter to the instantiation.
Adam Pautz has made a similar point when aiming to refute views on which color structure is deemed extrinsic to the colors, and dependent further on the character of our visual experiences. He argues that the proposal that color structure depends on color experience can be eliminated on the basis of attending to sensed colors and their apparent structural properties. Pautz claims that following the latter procedure would make it clear to perceivers that the place of a shade in color space does not depend on the phenomenal character of visual experiences. This is because when discerning similarities and other relations amongst the colors, our attention would be fully focused on the particular colors themselves:
It is introspectively evident that my belief [that Purple is more similar to Blue than to Green] is entirely about what I experience – the colours blue, purple, and green – and has nothing to do with experiences… In forming my belief, my attention was focused out, not in (Pautz Reference Pautz2006, 550).
Pautz seems to suggest that when attention of the right sort is exercised, the manifest higher-order similarities amongst the three colors will appear to depend on the existence and internal character of the respective colors, and on nothing besides.Footnote 25
In the context of the above discussion of the relation between the nature of colors and color experiences, it is natural to wonder whether my proposal of how essentiality gets revealed by experience is sufficiently general. Some might argue that we seem to know the essence of properties of experience in the same way as we seem to know the essence of the properties experiences manifest. Their view can be fully accommodated on my proposal, if it is allowed that we are able to attend to the phenomenal properties of our experiences in the way as we can attend to the external properties presented in such experiences (although for this further background capacities would clearly be needed). In fact, I believe that we are able to do this, and that precisely on this basis it has been held that our perceptual experiences are essentially relational or world-involving.Footnote 26
5. Conclusion
I have proposed how our experiences might reveal the essentiality of manifest higher-order aspects of sensible properties such as the colors. I argued, first, that proponents of Revelation need a version of the thesis on which our experiences could play a significant role in its justification. I then showed that the relevant version faces a considerable challenge, as it seems to demand too much of experience. The challenge got overcome by rethinking the way in which experiences would reveal the essentiality of higher-order properties. The highlighted a priori link between intrinsicality and essentiality in the case of second-order properties suggested an alternative. Finally, I proposed that selective attention is the key to understanding how the intrinsicality of a manifest property – including a higher-order property – would get revealed in experience.
It is my hope that the proposal will bring a renewed interest in Revelation. Currently many regard the thesis as outrightly implausible and replaceable by a view of the epistemic import of experience, on which experiences merely acquaint us with sensible properties. I believe they are wrong on both counts: when it comes properties such as the colors, experience has several distinct and complementary sorts of epistemic import, which tend to get mixed up. Additionally, I believe the critics, especially those sympathetic to primitivism about sensible properties, are implicitly relying on Revelation. The proposed account of the revelatory role of experience is intended not as a rival, but as an addendum to their view; and ultimately as a means to reconcile Revelation’s friends and foes.
Acknowledgments
I am grateful for detailed comments from Kris Laasik, Kevin Lynch, Takuya Niikawa, Caleb Perl, Nicola Spinelli, and two anonymous reviewers. I also want to thank Tim Crane, Uriah Kriegel, Roberta Locatelli, Michael Morris and Matt Soteriou for pertinent discussions.
