On Noticing Transparent States: A Compatibilist Approach to Transparency

. According to the transparency thesis, some conscious states are transparent or ‘diaphanous’. This thesis is often believed to be incompatible with an inner-awareness account of phenomenal consciousness. In this article, I reject this incompatibility. Instead, I defend a compatibilist approach to transparency. To date, most attempts to do so require a rejection of strong transparency in favor of weak transparency . In this view, transparent states can be attended to by attending (in the right way) to the presented world: that is, they are merely translucent . Here, I first argue that this understanding of transparency is too weak to qualify as a compatibilist view. Drawing on insights from Franz Brentano, I then describe a middle road between strong and weak transparency. The crucial idea is that, although transparent states cannot be attended to, they can be noticed (under suitable conditions). This view, I submit, allows supporters of inner awareness to commit themselves to a more interesting understanding of transparency— moderate transparency —that preserves the initial intuition underlying the transparency metaphor.


Introduction
The philosophical literature relating to phenomenal consciousness is replete with conflicting intuitions. A striking example is the apparent conflict between inner-awareness accounts of phenomenal consciousness and what is known as the transparency thesis.
On the one hand, it is often argued that a necessary condition for a mental state M to be phenomenally conscious is that the subject is aware of M. Given that the kind of awareness at issue is usually called 'inner awareness', I shall call this the Inner Awareness Principle (IAP). Historically, a main proponent of IAP was arguably Franz Brentano (Brentano, 1874;1995b).

Inner Awareness
However, several versions of IAP have been advocated (see, e.g., Rosenthal, 2005;Kriegel, 2009;Gurwitsch, 2010). Michelle Montague has recently coined the phrase 'awareness of awareness' and has argued that it is a constitutive feature of phenomenal consciousness, thereby defending her own version of IAP (Montague, 2016;. Despite these variations, IAP proponents usually agree on two crucial points. First, they take IAP to be phenomenologically evident: although it may be motivated on conceptual or theoretical grounds, IAP is not presented as a theoretical construct but as the faithful description of a phenomenological datum. Second, they typically take inner awareness to provide the subject with a certain cognitive contact with their own mental states. In other words, inner 3 awareness is taken to be a prerequisite for introspection and a source of knowledge about the mind.
On the other hand, it is often claimed that some conscious states or experiences 1 are transparent or 'diaphanous'. That is, when you try to attend to them, you 'look right through them' at the presented world, as you would do through a perfectly transparent window. Call this the Transparency Thesis.
Transparency Thesis (First Pass): for every subject S and some conscious state M, when S tries to attend to M, they 'look right through it' at the presented world. 2 This thesis, too, is supposed to capture an important phenomenon-the transparency phenomenon-which imposes a constraint on any satisfying theory of phenomenal consciousness.
Admittedly, the literal meaning of the transparency metaphor is not entirely clear. The literature has presented various ways of unpacking the transparency thesis, and it is far from clear if they all capture the same datum. 3 However, it is widely assumed that if a phenomenally conscious state is transparent, then the subject is not aware of that state.
Based on this assumption, it is often believed that the transparency thesis is flatly incompatible with IAP: they cannot both be true at the same time.
In this article, I reject this view. My goal is to defend a compatibilist approach to transparency and inner awareness that preserves the truth of both IAP and the transparency thesis. Such an approach should do justice to both the intuition that phenomenal consciousness is correctly accounted for in terms of inner awareness and the intuition that 4 at least some phenomenally conscious states are transparent. Arguably, the contention that these two intuitions are incompatible is attributable to a strong reading of the transparency thesis, on which I shall elaborate in Section 1. Therefore, the best method of defending compatibilism is to offer weaker versions of the transparency thesis, which do not conflict with IAP, and to defend one among them. This is what I propose to do here.
My plan is as follows. Section 1 ('Strong Transparency') introduces the strong transparency thesis and contrasts it with alternative understandings of transparency that I call moderate transparency and weak transparency. Section 2 ('Weak Transparency') critically discusses the weak transparency thesis, according to which transparent states may be attended to by attending (in the right way) to the presented world, meaning that these states are merely translucent. Although this view has sometimes been put forward in an attempt to defend compatibilism, I argue that it is too weak to qualify as a compatibilist view. Section 3 ('Moderate Transparency: The Middle Road') draws on insights from Brentano to flesh out a middle road between strong and weak transparency. The crucial idea is that, although transparent states cannot be attended to, they can be noticed (under suitable conditions). This view, I submit, allows supporters of inner awareness to commit themselves to a more interesting understanding of transparency, namely moderate transparency, that preserves the fundamental intuition underlying the transparency metaphor.

Strong Transparency
The mandate of this first section is to spell out the strong reading of the transparency thesis that underlies incompatibilism. For the sake of simplicity, I shall focus on the formulation 5 repeatedly offered by Michael Tye (1995;2021), whose understanding of transparency brings the compatibility issue directly to the fore ( §1.1). Next, I shall briefly contrast Tye's reading with weaker versions of the transparency thesis defended in the literature ( §1.2).

Tye's Incompatibilist View
Tye describes the transparency phenomenon based on a phenomenological experiment first presented by Gilbert Harman, who in turn drew on G.E. Moore's refutation of idealism (Moore, 1903). Very roughly, the phenomenological experiment is as follows. You are instructed to put yourself in the situation of having a visual experience and to 'introspect' that experience by trying to 'turn your attention to its intrinsic features'. Then, you are invited to register the fact that when you do that, what you are actually attending to are not features of your visual experience, but, rather, features of what is represented in that experience. Here is Harman's original example: Look at a tree and try to turn your attention to intrinsic features of your visual experience. I predict you will find that the only features there to turn your attention to will be features of the presented tree, including relational features of the tree 'from here'. (Harman, 1990, p. 39) Tye devises his own version of Harman's phenomenological experiment: Focus your attention on a square that has been painted blue. Intuitively, you are directly aware of blueness and squareness as out there in the world away from you, as features of an external surface. Now shift your gaze inward and try to become aware 6 of your experience itself, inside you, apart from its objects. […] The task seems impossible: one's awareness seems always to slip through the experience to blueness and squareness, as instantiated together in an external object. (Tye, 1995, p. 30) Tye draws several lessons from this experiment. The first important lesson concerns inner awareness or, more pointedly, the absence thereof: on performing the experiment, Tye argues, it should become clear to you that (1) you are not aware of your visual experience, and (2) you are not even aware of your visual experience's intrinsic features. Another lesson to be drawn has to do with one's ability, or inability, to attend to one's own current experiences: on performing the experiment, it should become clear to you that (3) you cannot attend to features of your visual experience and (4) you can attend to features only of the thing represented in your visual experience-for instance, the blueness of the square.
Thus, we obtain four distinct claims. These claims, Tye argues, are integral to the transparency thesis (see Tye, 2014, p. 40;2021, p. 33). The upshot of Tye's interpretation is a revamped version of that thesis that I shall call the strong transparency thesis: The strong transparency thesis is but one way of unpacking the initial intuition that underlies the transparency metaphor. As we shall see momentarily, there are also weaker 7 interpretations of the transparency phenomenon. The important point is that the strong transparency thesis brings into stark relief the tension with IAP. According to IAP, all conscious states are states of which the subject is aware. The transparency thesis, as Tye understands it, entails that at least some conscious states-namely, transparent states-are states of which the subject is not aware, as per (2). 4 Indeed, if some experience is transparent, the subject is not even aware of the intrinsic features of that experience, as per (1). Plainly, if the transparency thesis genuinely captures a phenomenological datum and Tye's unpacking of it is accurate, then IAP must be false. Conversely, if IAP genuinely captures a phenomenological datum, then the transparency thesis (as Tye understands it) must be false. In sum, IAP and the strong transparency thesis cannot both be true simultaneously. As Montague neatly summarizes the situation, 'the awareness of awareness thesis and the [strong] transparency thesis provide conflicting phenomenological descriptions of conscious experience. They are mutually incompatible, so we must choose one or the other' (Montague, 2016, p. 69;my emphasis). Simultaneously, Montague acknowledges that the transparency thesis contains a grain of truth. She herself accepts that inner awareness or awareness of awareness 'can never itself be made the focus of attention' and 'can never be transformed into introspection' (Montague, 2016, p. 72).
I believe that Montague is right here, although I suspect that her understanding of the transparency thesis somehow departs from its common interpretation. 5 In my view, the problem is not with the transparency phenomenon per se but, rather, with the way it is described by Tye (and many others). First, more attention should be paid to the role of 'attention' and 'attending' in the proposed phenomenological experiment. Second, Tye's method of unpacking the transparency thesis seems to rest on a series of unargued assumptions. Notably, (3) rests on the assumption that it is utterly impossible to attend to 8 features of one's own visual experience. However, as Amy Kind helpfully suggests, there is room for a much weaker interpretation of transparency. According to this interpretation, it is difficult-but not impossible-to attend to the features of one's own visual experience (Kind, 2003). Next, Tye seems to assume without argument that if inner awareness were constitutive of phenomenal consciousness, then introspection (i.e., the turning of your attention to your experience) should reveal it as distinct from outer awareness or awareness of 'external features'. 6 Given that this is not the case, he concludes that the subject is not aware of transparent states-hence (2). But why should introspection reveal inner awareness as an additional ingredient distinct from outer awareness? What if inner awareness is merely a prerequisite for introspection, rather than something that falls within its scope? Furthermore, although (1) and (3) are presented as distinct claims, Tye seems to assume that if you were aware of your own experience, then you would be able to attend to it. Since, in his view, you are not able to attend to your own experience, it follows that you are not aware of it. But this assumption, too, is disputable. Admittedly, if you are able to attend to something, then you are aware of it. But there is no valid converse inference from 'being aware of O' to 'attending to O'. After all, it is conceivable that you can be aware of certain things without being in a position to attend to them. I shall elaborate on this claim in section 3 below.
Once these assumptions are called into question, it is not at all clear that (1) and (2) should indeed be considered integral parts of the transparency thesis. At this point, it will be helpful to introduce weaker versions of the transparency thesis, which constitute alternative descriptions of the transparency phenomenon. One option is to disregard (1) and (2) while retaining (3) and (4). By this moderate interpretation, the transparency thesis does not rule out the presence of inner awareness. It rules out only the subject's ability to attend to their own transparent states. Another option, suggested by Moore's initial characterization of transparency, is to reject (1)-(4) altogether and argue, instead, that attending to one's own experiences is difficult, but not impossible (see Kind 2003). I shall call these two options the moderate transparency thesis and the weak transparency thesis, respectively.

Weak Transparency Thesis (First Pass): for every subject S and some conscious state
We are now in a position to better understand the issue of the compatibility between IAP and transparency. Whereas the strong transparency thesis is flatly incompatible with IAP, the same is not true of the moderate and weak transparency theses: both of these theses are utterly silent on the absence or presence of inner awareness. Therefore, the debate in which incompatibilists oppose compatibilists is, in essence, a debate about whether one should accept the strong version of the transparency thesis or one of the weaker versions thereof. It is not my goal here to offer a conclusive argument refuting the strong transparency thesis. In fact, I doubt that such an argument can be offered. I have already suggested that this thesis rests on disputable assumptions. Now, I shall argue that there is a respectable tradition of compatibilists actually defending a weaker version of the transparency thesis.

Compatibilist Views
Interestingly, there is evidence that G.E. Moore, one of the alleged champions of transparency, himself takes it to be compatible with the presence of inner awareness or 'awareness of awareness'. Although he acknowledges that it is difficult for the subject to attend to the sensation of blue as opposed to attending to blue itself, he maintains that 'to be aware of the sensation of blue is […] to be aware of an awareness of blue' (Moore, 1903, p. 449;my emphasis). 7 Put differently, Moore seems to hold an inner-awareness account of consciousness into which the weak transparency thesis is integrated. But Moore is far from being the only compatibilist in town. As far as I know, the first explicit defense of compatibilism is attributable to C.D. Broad (1921;1925). Broad interprets the transparency phenomenon along the following lines: in introspection, we become aware of the entire mental phenomenon, as opposed to its 'act' component. He calls this the 'truth' of the transparency thesis: It is commonly said that the difficulty of introspecting acts is that acts seem to be 'transparent' and that when you look for an act you only find objects. Now I take the truth of this to be that the real objects of such introspection are cognitions, and that these are complexes containing certain non-mental terms. What we become aware of by introspection is primarily the complex, and always at the same time the non-mental terms in it, which are called the objects of the cognition. But we do not seem to become aware of any mental term in such complexes, nor at all distinctly of the relating relation. (Broad, 1921, p. 148) Therefore, in Broad's view, nothing in the transparency phenomenon forces us to deny the presence of inner awareness. We may simply accommodate the transparency phenomenon by saying that, whereas objects of outer awareness can be inspected, transparent states cannot. Insofar as transparent states are 'felt', the subject is innerly aware of these states, even though they cannot possibly be inspected or attended to in introspection. In other words, Broad defends an inner-awareness account of consciousness that integrates the moderate transparency thesis.
More recently, various versions of compatibilism have been advocated by a number of scholars (see Martin, 1998;2002;Chalmers, 2004;Siewert, 2004;Kriegel, 2009;Boyle, 2011;Dorsch, 2013;Nida-Rümelin, 2014). As Chalmers puts it, 'there is no inference from the datum of transparency to the absence of introspective [i.e., inner] awareness' (Chalmers, 2004, 176). Arguably, the most travelled road to compatibilism is the defense of weak transparency. In the next section, I shall briefly present and critically discuss this strategy before fleshing out (in Section 3) what I take to be a more promising option, namely moderate transparency, a middle road between strong and weak transparency.

Weak Transparency
This section discusses the viability of the weak transparency thesis from the compatibilist perspective. One way to defend this thesis is to argue, first, that so-called transparent states can be attended to after all, and, next, that they can be attended to by simultaneously attending (in the right way) to the presented world. For the sake of convenience, I shall say that such states are translucent. In this section, I explicate how the translucence view has 13 been invoked to defend compatibilism (in §2.1) before arguing that this defense is unconvincing (in §2.2).

Translucence
Compare the following two cases.
Case 1. Suppose that, while reading this text, you are peripherally aware of a mug on the table. You can shift your attention from the text to the mug, which you can do either by turning your head and physically watching the mug or by keeping your eyes locked on the text and mentally focusing on the mug. 8 Yet, it is plausible to say that, to attend to the mug, you must somehow turn away from the text. There is a sense in which the text and the mug are 'competing for your attention'. You either attend to the text, or you attend to the mug.
But it is difficult to attend to both simultaneously.
Case 2. Suppose that you are watching a tennis match and your attention is directed toward the movements of the ball. Arguably, you can attend to the way the ball moves. But you cannot do so without attending to the ball itself. Indeed, attending to the way the ball moves is attending to the ball in a specific way. At any rate, there is no conflict between attending to the ball and attending to its movement. The movement is not an additional object that would compete for one's attention: you do not have to turn away from the ball in order to attend to its movement.
I have already suggested that Tye's way of unpacking the transparency thesis rests on a series of disputable assumptions. The contrast between these two cases suggests an additional assumption. Incompatibilists such as Harman and Tye assume that if you attend to the presented world, then you do not attend to your mental state. But this assumption, too, is controversial. Indeed, it may be argued that the introspection of transparent states is analogous to Case 2. On the incompatibilist view, for instance, when you try to pay attention to your perceiving the mug, you do not need to turn away from the mug. But this is not because it is impossible for you to attend to your perception of the mug rather than to attending to the mug itself. Instead, it merely shows that it is impossible for you to attend to your perception of the mug without attending to the mug. So, there is a sense in which you can be attending, at once, to the mug and to your perception of it. This is because, arguably, your perceptual experience is not an additional object in your perceptual field that competes for your attention. There is no competition, and no conflict, between attending to the mug and attending to the perception of the mug. This point has been rightly acknowledged in the literature. For instance, Siewert writes: We should recognize that directing attention to experience is not like directing attention from one sensorily apparent thing to some other […].
Thus an injunction to turn your attention 'inward' on experience seems especially misleading. For if there is some thing in public space your experience is of-as there will be if you're not hallucinating-attending to your experience equally will be attending to this thing (Siewert, 2004, p. 36; my emphasis).
Siewert's response to the transparency phenomenon presupposes that there is an object in the external or 'public' space. But the same point can be made by replacing 'object' with 'apparent object'. Consider, for example, the similar suggestion made by Nida-Rümelin: A subject may focus on its being under the impression of there being a certain object with certain properties by focusing on the apparent properties of the apparent object. There is no opposition (or there need not be an opposition) between these 'two' acts of attention (Nida-Rümelin, 2014, p. 268).
Let us now return to the question raised by the phenomenological experiment briefly

A Problem with Translucence
What should we think of the translucence view? Prima facie, the suggestion that a mental state and its object do not compete for the subject's attention is largely convincing. By constructing 'attending to your mental state' and 'attending to the (presented) world' as a disjunction, incompatibilists are oblivious to this fact. Furthermore, attending to M is a voluntary activity that implies some cognitive contact with M. Thus, it arguably allows us to acquire some information that may serve as a plausible basis for introspection.
Still, one might object that the translucence view, rather than successfully accommodating the transparency phenomenon, actually mistakes translucence for transparency. By assuming that the subject can attend to their own current states (if only by attending to the presented world), supporters of translucence are not only bound to reject the strong transparency thesis, they are also bound to reject its moderate form as well. In other words, their account of transparency fails to preserve any part of Tye's initial interpretation. Therefore, opponents of IAP may insist that the translucence view is simply too weak to qualify as a compatibilist view, given that it does not do justice to the initial intuition underlying the transparency metaphor. To substantiate this view, they can appeal to an additional claim-namely, the claim that, when you are 'in' a mental state, that state is not an object to you (see, e.g., Evans, 1982, p. 227). For the sake of clarity, let us call this the no-object thesis: No-Object Thesis: for every subject S and mental state M, if S is in M, then M is not an object to S.

Drawing on this thesis, opponents of IAP may argue as follows: (1) S can attend to M only if
M is an object to S; (2) if S is in M, then M is not an object to S; therefore, (3) if S is in M, then S cannot attend to M. Employing this same line of thought, one may further argue that the tennis-ball analogy, which was offered to introduce the notion of translucence, actually fails.
Let us suppose that attending to M just is attending to how the world appears (or seems) to you when you are in M, not unlike how attending to the movement of the tennis ball just is attending to the ball. Yet, the ball and its movement are alike 'out there in the world'-they are objects to you. Your own mental state, on the contrary, is not an object in any sense of the term. The idea that you can attend to your own mental state inevitably invites the thought that your mental state is an object to you, something you can observe or inspect while being 'in' it. But this, IAP opponents may argue, is blatantly impossible. Therefore, they may insist that if your mental states aren't objects to you when you are in them, then those states cannot be attended to, regardless of whether attending to them depends on attending to how the world seems (or appears) when one is in these states.
I consider the no-object thesis plausible, and, therefore, that the associated objection should be taken seriously. However, I wish to argue that supporters of compatibilism have another option at their disposal. Very approximately, the idea is as follows. Using Brentano's distinction between primary and secondary objects, compatibilists may defend a weaker reading of the no-object thesis: No-Object Thesis*: for every subject S and mental state M, if S is in M, then M is not a primary object to S.
Next, Brentano's distinction between attending and noticing allows compatibilists to argue that, whereas one cannot avoid making one's own mental states primary objects when attending to them, one can notice them without doing so. The upshot is a defense of moderate transparency. In the next section, I shall spell out this approach in somewhat more detail.

Moderate Transparency: The Middle Road
In this section, I draw on insights from Brentano to describe a middle road between strong and weak transparency, namely moderate transparency. The crucial insight is that, although transparent states cannot be attended to, they can be noticed (under suitable conditions).
According to the argument, given that one can notice mental states without making them primary objects, the proposed view accommodates the weak version of the no-object thesis.
First, I introduce Brentano's pre-theoretical distinctions between implicit awareness, noticing, and attending (in §3.1). Next, I put those distinctions to work in formulating Brentano's idea that the subject can notice their own mental states without making them primary objects, although they may not be in a position to observe them ( §3.2).

Implicit Awareness, Noticing, and Attending
Brentano defends a fairly sophisticated view of inner perception, but he is far from oblivious to the phenomenon of transparency. Although, unlike Moore, Brentano never employs the transparency metaphor, he comes very close to acknowledging the transparency 19 phenomenon in Psychology from an Empirical Standpoint. In an important passage in which he rejects inner observation, Brentano writes: It is only while our attention is turned toward a different object that we come to perceive, incidentally, the mental processes which are directed toward that object. Thus, the observation of physical phenomena in external perception, while offering us a basis for knowledge of nature, can at the same time become a means of attaining psychological knowledge.
Indeed, turning one's attention to physical phenomena in our imagination is, if not the only source of our knowledge of laws governing the mind, at least the immediate and principal source (Brentano, 1874, p. 36;1995b, p. 30;translation modified, my emphasis).
Here I propose one way in which to understand Brentano's view. First, it is wrong to conceive of the awareness of physical phenomena and the awareness of mental phenomena as two independent, separable experiences. For instance, when you hear a sound, there is only one single experience, which is both of the sound as 'primary object' and of itself as 'secondary object'. Next, and more importantly, you cannot be aware of hearing the sound without being aware of the sound, and often attending to it. 9 In other words, Brentano asserts a one-way dependence relation between (i) the subject's being innerly aware of M and (ii) the subject's attending to M's primary object. This view is not unlike that endorsed by supporters of translucence, with one important difference. While supporters of translucence contend that the subject can attend to their own mental happenings (even though attending to them just is attending to their intentional objects), Brentano maintains 20 that occurrent mental states cannot be attended to while one is 'in' them: they can only be noticed. In Brentano's view, this is true of all mental phenomena, not just perceptions.
How, then, are we to conceive of the difference between attending and noticing? To be sure, attending and noticing are both concerned with attention. They belong to a family of concepts that have been called 'heed concepts' (Ryle, 1949;2009, p. 118) or 'attention concepts' (White, 1964). But what, if anything, is the difference between, for instance, noticing something in your visual field, on the one hand, and attending to it or voluntarily applying all your attention to it, on the other? How does that difference apply to the subject's awareness of their occurrent mental states?
Let us first consider Brentano's conception of noticing. Here, we may begin with a simple idea: there is a difference between merely experiencing something and noticing it.
Chisholm once called this claim 'one of the basic theorems of descriptive psychology' (Chisholm, 1968, p. 168;1976, p. 95 Brentano introduces the notion of noticing as a distinctive mode or modification of awareness and inner perception. 'By noticing', he writes, 'we mean an inner perception, in fact an explicit perception of what was implicitly contained in the perception of our consciousness' (Brentano, 1982, p. 33;1995a, p. 36). According to this view, noticing M is tantamount to being explicitly aware of M, where 'explicitly', I submit, means that M somehow stands out against some background, or bears a remarkable phenomenal salience.
By contrast, you are implicitly aware of M when you perceive a whole (e.g., an entire slice of your stream of consciousness) of which M is a part, while M itself remains unnoticed and lacks phenomenal salience. The same distinction between implicit and explicit awareness 21 applies to sense perception, and Brentano himself often illustrates this with the example of auditory experience. For instance, in a manuscript note edited by Oskar Kraus, he writes: There is a two-fold way in which a thing may be said to be an object of awareness: it may be explicit and distinct or it may be implicit and indistinct. If one hears a chord and distinguishes the notes which are contained in it, then one has a distinctive awareness of the fact that one hears them. But if one does not distinguish the particular notes, then one has only an indistinct awareness of them (Brentano, 1933;1981, p. 117).
I have suggested that, in Brentano's view, noticing is a distinctive mode or modification of awareness. This modification is fairly basic. In particular, noticing O is not to be conflated with (a) noticing that O is thus-and-so, which is a propositional attitude, nor with (b) being struck by O, which is an occurrent emotional state directed at O. This is not to deny that an explicit awareness of O may be-and, in human adults, often is-accompanied by either (a), (b), or both. Rather, Brentano's idea is that there is a more primitive act of noticing, which entails no propositional or emotional attitudes. Consider propositional attitudes, first. To be sure, in certain cases, you perceive or notice that something is distinct from, identical to, or merely similar to something else (Brentano, 1982, p. 34;1995a, p. 37). Yet, in Brentano's view, noticing M is not necessarily a propositional state and it does not entail determining M as such-and-such a thing. Noticing something does not require exercising one's conceptual capacities. To put it bluntly: you can notice O without knowing what O is. 'When I speak of noticing in this context', Brentano points out, 'I have in mind only simple accepting judgments [i.e., non-propositional acts of acceptance]' (Brentano, 1982, p. 34;1995a, p. 37).
Once this has been established for propositional attitudes, it is easy to see that something 22 similar holds true for occurrent emotional states. Although noticing something can occasionally be accompanied by emotional responses, it does not entail being struck by what is noticed. What is more, the presence of emotional responses to O presupposes that O has not gone unnoticed. It would be 'erroneous to believe that something must first strike us in order to be noticed. On the contrary, nothing will strike us which has not already been noticed by us' (Brentano, 1982, p. 35;1995a, p. 38).
We have just seen that, for Brentano, noticing does not entail a predicative judgment.
But, on the other hand, noticing O may be described as having a cognitive import: by noticing O, the subject acquires some information about O (see also White, 1964, p. 23). In other words, noticing O may be described as the initiation of a certain cognitive contact with O on the part of the subject. 10 Once O has been noticed, you can commit it to memory, or, as Brentano puts it, 'take note' of it (Brentano, 1982, p. 35;1995a, p. 38). In many cases, you can also attend to it in order to extract more information about it. How, then, should we conceive of the relationship between noticing and attending?
It may seem as though attending and noticing are both associated with the acquisition of information through attention. However, they are different phenomena: it is perfectly possible to notice something without attending to it. Brentano makes this point explicitly, arguing that 'it is not as if attentiveness could be an indispensable precondition for noticing.
How should one come to notice?' (Brentano, 1982, p. 36;1995a, p. 39). Noticing, he continues, can 'take place without a prior paying of attention' (Brentano, 1982, p. 36;1995a, p. 39). Moreover, an interesting difference between attending and noticing is the fact that the former, but not the latter, comes in degrees. You either notice something or fail to do so, but it does not make sense to say that something has gone half-noticed, let alone more 23 noticed or less noticed than something else. Commenting on Brentano, Chisholm once suggested to put this idea in Rylean terms, claiming that noticing, unlike attending, is an 'achievement word' (Chisholm, 1968, p. 169;1976, p. 96). White offers a more granular analysis, distinguishing between activity-concepts such as 'attending', receptivity-concepts such as 'noticing', and achievement-concepts such as 'detecting' (White, 1964, p. 23). 11 Accordingly, a further remarkable difference-which Brentano does not bring to the fore in his analyses-is that whereas attending is something we can do, or fail to do, noticing is something that happens to us. For our purposes, the crucial point is that noticing M does not presuppose attending to M. White argues to this effect as follows: When we notice something our attention is thereby caught by it; what we suddenly notice suddenly catches our attention. The fact that I may notice one thing while giving my attention to another shows that there can be noticing without attending, though not without attention. (White, 1964, p. 29) Could it be, then, that one comes to notice a mental act by attending to, or perhaps noticing, its object? In the next subsection, I draw on further insights from Brentano to suggest the affirmative answer to this question.

Applying the Distinctions
Let us recall the question raised by Harman and Tye's phenomenological experiment: what exactly are you attending to when introspecting transparent states? One way to respond to this question is as follows. When introspecting transparent states, the subject attends to how the world appears (or seems) to them. But they cannot possibly attend to their mental state itself. As Broad put it, and as discussed in §1.2, the subject's mental state simply is not an object available for their inspection. In this respect, supporters of strong transparency might well be right. However, they are wrong to conclude that transparent states are states of which the subject is not aware. Once noticing and attending have been distinguished, we may argue that transparent states are noticeable even though the subject may be unable to attend to them. As I understand Brentano, this is in fact his view. He argues that all mental phenomena-that is, all occurrent mental states-are innerly perceived. But this inner perception cannot be turned into inner observation: 'it is a universally valid psychological law that we can never focus our attention upon the object of inner perception [i.e., the ongoing mental state]' (Brentano, 1874;1995b, p. 30; see also Montague, 2016). Thus, in this view, you simply cannot attend to your own mental state while you are in it. However, you can attend to it in retrospect, relying on memory: what we might interpret as cases of introspection are, in fact, cases of retrospection. However, this does not prevent you from noticing changes in your stream of consciousness. We have seen some reasons to believe that noticing something does not presuppose attending to it. If anything, the opposite is true: a necessary, albeit insufficient, condition for attending to O is that O was somehow noticed, because otherwise O is unavailable for observation or inspection. Therefore, it is quite possible for transparent states such as perceptions to be noticed while the subject's attention is directed toward something else-some worldly object such as a mug on the table, say. In fact, Brentano himself argues for the existence of a parallelism between the implicit/explicit character of inner awareness and the implicit/explicit character of outer awareness. He takes this to be a phenomenological datum: 25 It is plain that, when one thinks of a primary object in an implicit way, one also thinks, and must perceive implicitly, that one thinks. Furthermore, it is plain that when one has a presentation of, accepts or rejects, etc., a primary object in an explicit way, one also thinks and acknowledges explicitly, and hence notices, that one thinks, accepts or rejects. Noticing something and at the same time not noticing that one notices it, this would be self-contradictory. (Brentano, 1933, p. 282) In other words: whenever you notice O, you notice your noticing of O. Brentano makes the same point in his lecture notes on descriptive psychology: 'if we present to ourselves explicitly, and not just implicitly, a particular part of a presented whole, then presumably we will always notice this explicit presenting as well' (Brentano, 1982, p. 58;1995a, p. 60).
According to this view, then, when your awareness of the mug is implicit-that is, when the mug is part of an entire perceived scene and is not the target of a specific, dedicated act of apprehension-your awareness of your perception of the mug is implicit as well. However, when the mug is noticed-when you become explicitly aware of it-your inner awareness of the perception of the mug becomes explicit as well. This view turns on the idea that noticing an occurrent mental state is something that happens to the subject when the subject's attention is directed elsewhere. To fully understand this view, one has to recall that noticing, unlike attending, is not an activity and requires no effort. Of course, this does not mean that it is not subject to certain factual conditions. Common prejudices, distraction, or exhaustion, for instance, can prevent the subject from noticing things that they would have noticed in more favorable circumstances. This is why Brentano writes that noticing 'happens only 26 under certain conditions', while 'in certain cases […] is difficult, or cannot be achieved at all' (Brentano, 1982, p. 121;1995a, p. 129).
Admittedly, many details of this view cry out for clarification. However, instead of elucidating those details, I shall now suggest that Brentano's distinction between attending and noticing allows IAP supporters to commit themselves to a version of transparency more interesting than the translucence view. In fact, the distinction puts them in the position to endorse a more sophisticated version of the moderate transparency thesis: This understanding of transparency has many advantages. By way of concluding this paper, I shall mention four of them. First, since 'noticing M' does not entail 'attending to M', compatibilists need not argue that transparent states are states that can be attended to after all. In fact, they can concede that transparent states simply cannot be attended to while one is 'in' them.

Moderate Transparency Thesis (Second
Second, in the proposed view, transparent states can be noticed (under suitable conditions). This claim is not just compatible with IAP. In fact, compatibilists may argue that the noticeability account entails IAP. They can argue that it is impossible to notice something without being aware of it, for noticing M is precisely tantamount to being explicitly aware of M. If transparent states can be noticed, and if noticing M entails being aware of M, then the subject can be aware of transparent states, even though these states cannot be attended to while the subject is 'in' them.
Third, a supporter of this final claim would have to give an account of that apparent exception: why is it impossible to turn the explicit awareness (noticing) of one's own mental states into inner observation (attending)? In Brentano's view, this question has a simple answer, related to the fact that attending is an activity, while noticing is not. Being a mental activity, attending inevitably alters the course of one's mental life, thereby replacing-or at least modifying-the very experience to which one tries to attend. This efficiently explains why it is impossible to attend to one's own unaltered experience while one is 'in' it, although it may be possible, under suitable conditions, to notice that experience.
Fourth, noticing, unlike attending, does not entail that you take your own mental state as primary object. Controversial as it may be, this idea is an integral part of the noticeability view: you may notice a modification of your own stream of consciousness, even though you may not be able to observe or inspect it, as you could do with a primary object. However, noticing qua explicit awareness is a cognitive contact that can serve as a source of knowledge about the mind. It provides the kind of cognitive contact that inner-awareness accounts of consciousness seek to provide a theory of.

The Moral
The goal of this article was to defend a compatibilist approach to transparency. I have argued that strong transparency is too strong, and weak transparency too weak, to qualify as a genuinely compatibilist account. The best road to compatibilism, I submit, is defending moderate transparency. That middle road can be fleshed out by invoking Brentano's 28 distinction between attending and noticing. If this proposal is correct, compatibilists can vindicate inner awareness without defending the controversial view that allegedly transparent states can be attended to after all, and hence are merely translucent. Nor, for that matter, must IAP supporters argue that other conscious states such as emotions are opaque. If they are prepared to endorse the distinction between attending and noticing, they can rest content with arguing that, even though transparent states cannot be attended to while one is 'in' them, they can be noticed (under suitable conditions). Thus, I conclude that IAP supporters should not fear transparency.
self-knowledge. Instead, I shall confine myself to inquiring whether and how the transparency thesis can be made compatible with IAP. 4 Note that, whereas Tye's incompatibilist account of transparency rules out de re inner awareness (awareness of), it is consistent with a characterization of consciousness and/or introspection in terms of fact-awareness (awareness that): 'I am aware that I am having an experience but I cannot attend to, or be aware of, the property that makes my experience an experience' (Tye, 2021, p. 72). This view is shared by other transparentists. According to Fred Dretske's displacedperception theory, for instance, your being aware that you are hearing sounds does not depend on your being aware of your hearing sounds, for you typically come to know that you are hearing sounds simply by being aware of the sounds themselves (see Dretske, 1993;1995, pp. 39 ff.;1999). 5 When she says that the transparency thesis contains a grain of truth, Montague seems to understand the thesis as being about inner awareness and not about some 'first-order' conscious states such as, for instance, visual experience. She takes it that inner awareness exists but is itself transparent, to the effect that it cannot become the focus of my attention. See also Kriegel (2009, pp. 182 ff.), for a similar view. 6 Tye's formulation of (4) assumes an externalist conception of the mind, which does not follow from the transparency phenomenon (see Stoljar, 2004;Gow, 2016). Since a discussion of externalism falls beyond the scope of this article, I won't say more about that here. 7 Strawson (2015) has suggested that it is wrong to see Moore as a representative of the transparency thesis. My alternative proposal is to see Moore as a compatibilist.
8 For the sake of the argument, one can assume that attending to O involves turning one's attention toward O, which, in turn, is best understood as one's (peripheral) outer awareness of O becoming focal or mentally salient. See Watzl (2017; 2018) for a detailed characterization of attention along those lines. 9 Here, Brentano's position is reminiscent of Hamilton's: it is 'palpably impossible that we can be conscious of an act without being conscious of the object to which that act is relative' (Hamilton, 1970, 1, p. 212). The same view is neatly put by Brentano's former student, Carl Stumpf: One can perceive a landscape and be engrossed in this object so much that one's own perceiving and all other inner states go unnoticed. In contrast, one cannot conversely perceive any mental state without somehow taking notice of the material or object this state is directed at. I cannot become aware of an act of hearing without simultaneously becoming aware of the sound that I am hearing. I cannot become 33 aware of an act of will without simultaneously becoming aware of what I will, even if it is only something indeterminate, some direction of my acting. (Stumpf, 1939, 1, pp. 341-342; see also Textor, 2017, p. 106)