Editorial: Consciousness and Inner Awareness

It is widely acknowledged that there is a connection between consciousness and awareness. One way to cash out this connection is by understanding conscious states to be those states we are conscious with, i.e., states which give us awareness of the world around us (Dretske 1993). But acknowledging this doesn’t seem to exhaust the connection between consciousness and awareness. As well as external awareness, there seems to be some sort of inner awareness connected with consciousness. Exactly what this inner awareness amounts to, what it is of, how and when it is connected with consciousness, and how it works, are topics considered in the papers in this special issue.

What is this inner awareness? One way to begin is by asking how inner awareness is related to introspection where this is understood as the act of reflectively becoming aware of aspects of one’s mental life. On one conception, inner awareness is something we achieve only through introspection: to become aware of our experience of the sunset, or of ourselves as the subject of this experience, we must explicitly reflect on these things. Many, however, have suggested that inner awareness can also characterise our non-reflective experiences. On this conception, we have a kind of implicit inner awareness even when we are not explicitly introspecting. Such a view of inner awareness encourages us to understand introspection not so much as an act that gives rise to inner awareness but as an act that emphasises or makes explicit an inner awareness that was already present in experience prior to introspection.

A second foundational question concerns what inner awareness is awareness of. On one understanding of inner awareness, what we are aware of is our experience. On a second we are aware of the subject of the experience, i.e., of ourselves. On a third understanding of inner awareness, we are aware of ourselves as the owner of the experience we are undergoing.¹

The purpose of this special issue is to deepen our understanding of inner awareness by adopting an interdisciplinary approach to the topic. Philosophers are conceptualising the subjective character of consciousness in new and illuminating ways, empirical researchers are shedding new light on the mechanisms of inner awareness and clinicians are improving our understanding of disorders of subjectivity. These threads, and many others, are skillfully woven together in the contributions to this special issue. In this introduction we survey the key questions surrounding inner awareness and introduce each of the contributions to the collection. We proceed in two halves: first, we consider inner awareness in non-reflective experience, then we consider the issues surrounding inner awareness in reflective experience.

1 Inner Awareness and Non-Reflective Experience

We can distinguish two aspects of a phenomenally conscious state: its qualitative character and its subjective character (Levine 2001; Kriegel 2009). The qualitative character of a conscious state constitutes, in Nagel’s (1974) familiar locution, what it’s like to be in that state for the subject. What it’s like for you to perceptually experience a sunset, for instance, might be characterised by a variety of red, orange and pink phenomenal qualities. The subjective character of a conscious state might also be picked out using Nagel’s locution. Whatever the qualitative character of a conscious state, there is something it’s like for you to be in such a state. To say that the sunset is a reddish-pinkish-orange for you is to attribute some special status to you – the subject – in the overall structure of your conscious experience. This ‘for-me-ness’ is the subjective character of consciousness.² Although much discussion of consciousness focuses on the qualitative character of conscious states, it has become increasingly common for philosophers and psychologists alike to emphasise the importance of subjective character.

A major difficulty facing the ever-growing literature on this topic is that subjective character is a particularly difficult concept to pin down. The place of subjective character in our phenomenology is subtle and elusive at best (Kriegel 2009), and many report being unable to isolate any aspect of experience that can be distinguished from qualitative character (e.g. Coleman 2015). A promising starting point is to describe subjective character in terms of inner awareness (Kriegel 2009). When we perceptually experience the sunset we have an outer awareness of the scene before us but we also have an inner awareness of that very experience, or of something associated with it. On this proposal, the subjective aspect of phenomenal character is understood—in part, at least—in terms of an epistemic relation. This claim that experiences involve inner awareness raises a host of deep and important questions.

The first question is about the nature of this inner awareness. What does it mean for an experience to be some way for the subject? A modest understanding of inner awareness is that the subject is the thing-appeared-to. On this understanding, the subject is integral to the structure of experiential episodes but is not itself something that is experienced. A more demanding understanding of inner awareness requires that the subject itself figures in experience, so it is not just a thing-appeared-to but also a thing-that-appears. An even more demanding understanding of inner awareness requires that one is not just aware of oneself in experience but that one is aware of oneself as the owner of that very experience.³

Once one has settled on a particular conception of inner awareness, the next question is how pervasive inner awareness is among experiences. One might claim that outer awareness is only rarely accompanied by inner awareness in non-reflective experience. Alternatively, one might claim that inner awareness is a feature of all typical experiences, absent only in atypical cases such as trauma, drug use or mental disorder. One might go further still and claim that inner awareness is an essential feature of all non-reflective experiences, present even in atypical conditions. Again, the plausibility of any of these claims about the pervasiveness of inner awareness will depend on how demanding a conception of inner awareness one adopts.

The allusion above to inner awareness in atypical situations is of special importance. There are a range of mental disorders that involve a disruption to a subject’s ordinary awareness of herself and her experiences. We might call this loose cluster of conditions ‘disorders of subjectivity’. The questions highlighted above interact with questions about these disorders in important ways. On the one hand, our understanding of such disorders promises to shed light on the nature of inner awareness. If, for example, a set of symptoms is best explained by the absence of inner awareness in a subject’s non-reflective experience, then this would suggest that the non-reflective experiences of typical subjects are characterised by such inner awareness, but that inner awareness is not a necessary feature of experience as it can be absent in pathological cases. On the other hand, our understanding of inner awareness promises to shed light on the nature of disorders of subjectivity. A better conceptualisation of the various aspects of subjectivity, and a deeper appreciation of the role of inner awareness in ordinary experience, could help us to understand how disorders of subjectivity disrupt one’s phenomenology and to pinpoint the source of these disruptions.

1.1 A Taxonomy

As highlighted above, inner awareness might take a number of different forms. It will be helpful to adopt different terms to refer to the different varieties of subjective character. Although a number of different phrases are deployed in the literature – for-me-ness, me-ishness, subjectivity, mineness, subject-ish phenomenology, the sense of ownership, pre-reflective self-consciousness, and more besides – their use is somewhat confused, with some authors using different terms interchangably and uses of the same term differing between authors. Guillot’s contribution to this special issue imposes a regimentation on this terminology that helpfully traces the most prominent contours of conceptual space. Assuming that the subjective character of experience involves the subject having some particular status in experience, she explains that the subject might be understood:

• as one of the two relata in the relation of phenomenal awareness to her experiences, i.e. as the one who is appeared to;

• as also appearing to herself in being aware of the other relatum (the experience);

• as appearing to herself as the owner of the experience. (Guillot 2017, this special issue)

Guillot uses the label ‘for-me-ness’ for the first of these, ‘me-ishness’ for the second and ‘mineness’ for the third. It is worth highlighting that there are no entailment relations between these three forms of inner awareness – at least not without the addition of substantive further premises. If a subject is the one to whom an experience is presented, it doesn’t follow that the experience is presented to her as her own (see Howell and Thompson 2017, this special issue). Similarly, if the subject is presented with herself in experience, it doesn’t follow that she is presented to herself as the owner of her experience. Thus neither the for-me-ness of an experience nor the me-ishness of an experience (if such there are) entail that the experience has mineness. One might think that the entailment does run in the other direction: that if one is phenomenally aware of one’s ownership of a conscious state, one must be phenomenally aware of oneself and phenomenally aware of one’s experience. However, if we are phenomenally aware of some relation it isn’t necessarily the case that our awareness of its relata is itself phenomenal. Thus:

…even if some form or other of awareness of the experience, and of myself, is a necessary condition for the phenomenal awareness of myself as owner of the experience (mineness), it doesn’t follow that an a priori relation of implication holds from mineness to for-me-ness (a phenomenal awareness of the experience) and me-ishness (a phenomenal awareness of the experiencer). (Guillot 2017, this special issue)

In discussing inner awareness, we must be careful to distinguish these three forms and should be open to making even more fine-grained distinctions if required. Having made these distinctions, we can move on to consider the possible views one might take on how pervasive these forms of inner awareness are among non-reflective experience. Remember, the question we’re considering in this section is not whether introspective experience involves an awareness of one’s experience, or oneself, or one’s ownership of one’s experience but whether such awareness is present during non-reflective episodes. Although there is a vast array of possible views one can take on the prevalence of inner awareness, we suggest that the following three options capture the most important views in the literature:

Universalism: Inner awareness is present in all non-reflective experiences without exception.

Typicalism: Inner awareness is present in all non-reflective experience, except in certain atypical cases in which ordinary experience is disrupted.

Absentism: Inner awareness is never present in non-reflective experience and features only in introspective experience.⁴

Of course, the precise content of each of these three theses depends on the conception of inner awareness one adopts. Combining the three main forms of inner awareness with the three main theses about the scope of inner awareness, we get nine target theses: universalist theses about for-me-ness, me-ishness and mineness respectively; typicalist theses about each those three forms of inner awareness and absentism about each of them.

1.2 Mineness and Me-Ishness

Much of the recent literature on the subjective character of consciousness revolves around a particularly bold understanding of subjective character: universalism about all three forms of inner awareness. Consider the following passage from Zahavi:

When I am aware of an occurrent pain, perception, or thought from the first-person perspective, the experience in question is given immediately, non-inferentially and non-criterially as mine. That is, the experience is given (at least tacitly) as an experience I am undergoing or living through. (Zahavi 2004, p. 78)

This passage and others point to a conception of experience in which all conscious states are characterised by an inner awareness that has each of the three features isolated: for-me-ness; me-ishness and mineness. Zahavi holds that all the major figures in the phenomenological tradition - Husserl, Scheler, Heidegger, Sartre, Merleau-Ponty, Henry, Ricoeur - advocated some version of the view that experience brings with it an implicit awareness of one as the owner of that very experience (2004, p. 82). There is, of course, room for debate on Zahavi’s reading of these figures but it’s fair to say that this bold understanding of inner awareness has a long and influential history.

Perhaps the most interesting objections to universalism about mineness – the claim all experience is characterised by a sense of ownership over that experience – come from empirical investigation of disorders of subjectivity. Zahavi is quite clear that these features are essential to the structure of consciousness, so cannot be absent even in pathological cases (Zahavi and Parnas 1998). But many have argued that an honest examination of the relevant disorders cannot be reconciled with such a view.

First, consider thought insertion—cases in which subjects report that their thoughts do not seem to belong to them and, in many cases, report that they seem to belong to some other person. The following are some representative reports cited by Billon (2013, pp. 291–292):

Thoughts are put into my mind like “Kill God.” It’s just like my mind working, but it isn’t. They come from this chap, Chris. They are his thoughts. (quoted in Frith, 1992, p. 66)

The thoughts of Eamonn Andrews come into my mind. There are no other thoughts there, only his...He treats my mind like a screen and flashes onto it like you flash a picture. (quoted in Mellor, 1970, p. 17)

It is tempting to say that the experiences of these patients are lacking a feature that is present in the experiences of neurotypical subjects. Some have suggested that the experiences of neurotypical subjects are characterised by a sense of mineness, and that patients suffering from thought insertion have experiences that lack this sense of mineness. This is not to say that the patients do not experience their inserted thoughts.⁵ The suggestion is that although they do experience their thoughts, they are not phenomenally aware of these experiences as their own. On this view, cases of thought insertion count against universalism about the sense of mineness: the claim that every experience is characterised by a sense of ownership over that experience.⁶

A similar interpretation has been given to a pair of closely related disorders of subjectivity: depersonalisation disorder and Cotard’s syndrome. Billon offers the following description of depersonalisation:

Depersonalisation is a complex condition involving a broad modification of experience. Its core, however, is probably the feeling that the self or some of its significant parts are estranged, missing or nonexistent. (2016, p. 371).

This estrangement from the self seems to underwrite a number of more specific symptoms associated with depersonalisation including: it seeming as if body parts do not belong to one; it seeming that mental states - including emotions, perceptions and imaginings - do not belong to one; and it seeming as if one is dead (see Billon 2016, pp. 371–373).

Subjects with depersonalisation disorder are not delusional: they report that it is as if they are in the alarming situations described, but they recognise that these appearances are non-veridical. In contrast, patients with Cotard’s delusion accept these appearances as veridical. Many Cotard’s patients report, for instance, that they are actually dead. This symmetry between Cotard’s and depersonalisation leads Billon and others to conclude that ‘Cotard syndrome can…be characterised as the delusional form of depersonalisation, depersonalisation being, conversely, the “as if” form of the Cotard syndrome’ (Billon 2016, p. 372) Most pertinent to our current topic are cases in which patients report an estrangement from their own experience. Consider the following two reports:

It was as if it was not me walking, it was not me talking, as if it was not me living […] I can look at me, I am somehow bothered by my body, as if it wasn’t me, as if I lived on the side of my body, on the side of myself if you want. I don’t know how to explain. (Janet and Raymond 1898)

When a part of my body hurts, I feel so detached from the pain that it feels as if it were somebody else’s pain. (Sierra and Berrios 2000)⁷

Some propose that where the experiences of neurotypical subjects are characterised by a sense of ownership over their experience, the experiences of these patients lack such a sense of mineness. Billon, for example, concludes that:

…even if we do not usually pay attention to it, our experiences normally incorporate a certain subjective feature in virtue of which they seem to be ours to us, [and] this feature is attenuated in depersonalisation. (Billon 2016, pp. 373-374)

Depersonalisation disorder and Cotard’s delusion thus motivate some to reject universalism about mineness in favour of typicalism about mineness: the claim that unreflective experience only lacks a sense of ownership over one’s experience when one is victim to a relevant disruption of ordinary experience.

What about universalism about me-ishness - the claim that all experience is characterised by an awareness of the self (irrespective of whether one is aware of oneself as the owner of said experience)? Thought insertion does not seem to threaten the thesis that the subject always figures in experience: it is the patient’s awareness of their own thoughts that is disrupted rather than their awareness of themselves. However, it has been suggested that at least some subjects suffering from depersonalisation disorder and Cotard’s syndrome are not aware of themselves, and so have unreflective experiences that lack me-ishness (Guillot 2017, this special issue). This conclusion is driven by patients’ reports that (it is as if) they don’t exist. The proposal is that the unreflective experiences of typical subjects include an awareness of the self, and that the disruption to this me-ishness of experience explains the reports of such patients.

If one is convinced by these arguments, the only universalist thesis left available is universalism about for-me-ness: the claim that the subject always has a special phenomenal awareness of her experience, irrespective of whether she is aware of herself or of her ownership of said experience. Indeed, after rejecting the other universalist theses in light of disorders of subjectivity, Guillot (2017, this special issue) settles on this form of universalism. We will discuss the merits of such a view in a moment, but first a different question must be addressed. The objections to universalism about mineness and me-ishness discussed have motivated a typicalist view of those two forms of inner awareness. But are there objections to these typicalist views that might motivate absentism about me-ishness or mineness?

Against a typicalist view of mineness, one might argue that we can only gain awareness of ourselves, or of our ownership of our experience, by explicit reflection. Howell & Thompson articulate this view with respect to mineness:

Despite the fact that experiences on the unreflective level don’t have mineness, we can gain a sense of mineness in reflection… In reflection we lay claim to our experiences, and the mineness is a product—not a condition of—that attitude. (Howell and Thompson 2017, this special issue)

They propose that a similar view can be found in some interpretations of the Pali-Buddhist tradition (see Albahari 2010, p. 102) and in the works of Sartre. Looking at the quotations they offer in support of this, it seems that both can be read as targeting not just the sense of ownership but the sense of self. Consider this passage from Sartre:

Contrary to what has been held, therefore, it is on the reflected level that the ego-life has its place, and on the unreflected level that the impersonal life has its place… (Sartre 1993, p. 58 quoted in Howell and Thompson 2017 this special issue)

An experience has me-ishness just in case it presents one with the self or ego. Sartre’s suggestion is that the ego only has a place in reflective experience. He thus seems to advocate absentism about me-ishness in unreflective experience.

If absentism about mineness and me-ishness are correct, then advocates of typicalism and universalism about these forms of inner awareness are dramatically mistaken about the character of their own unreflective experiences. How is it possible for them to have got their phenomenology so wrong? One significant possibility is that they are subject to a ‘refrigerator-light illusion’. It is easy to fall subject to the illusion that the light in one’s refrigerator is always on because whenever one checks, one finds that the light is on. Of course, this evidence is misleading because the act of checking is what causes the light to come on. Perhaps introspection misleads us in an analogous way. If the absentist view discussed is correct, then the act of introspection yields an awareness of the self and of one’s ownership of one’s experience. This means that whenever one reflects on experience to see whether it is characterised by mineness or me-ishness, one invariably finds that it is. Because of this, it would be easy to make the mistake of thinking mineness and me-ishness characterise all of one’s (typical) non-reflective experiences too (see Schear 2009; McClelland 2015; Howell and Thompson 2017 this special issue).

The typicalist theses about mineness and me-ishness were motivated by a particular understanding of disorders of subjectivity. How can the absentist make sense of these disorders? Clearly the way that thoughts appear to a subject suffering from thought insertion differ from the way they appear to typical subjects, and the way that the self appears in depersonalisation disorder and Cotard’s syndrome differs from the way it appears to typical subjects. Typicalists explain this in terms of typical subjects being phenomenally aware of themselves and of their ownership of their experience, and atypical subjects lacking this phenomenology. An alternative explanation, however, is that subjects undergo some other kind of disruption to their phenomenology. These disorders are generally understood to be the product of two factors: an atypical experience and an atypical belief-forming procedure based on that experience. But once the disorders are framed this way, it is hard to justify the conclusion that the phenomenal abnormality is a lack of mineness and/or me-ishness. Thinking about typical subjects, it’s tempting to say that one would only give these unusual reports if a familiar quality of me-ishness/mineness was absent. But if the two-factor account is true, the patients are not typical in this regard. They are making atypical inferences from atypical experiences. But then it is much less clear exactly what character their atypical experiences have. There is something abnormal about their experience that leads them to make unusual claims about inserted thoughts and the like, but we cannot simply read-off their phenomenology from their reports (Howell and Thompson 2017, this SPECIAL ISSUE; McClelland forthcoming).⁸ Some abnormality other than the proposed absence of mineness/me-ishness could be responsible. But then there is no need to claim that typical unreflective experiences are characterised by mineness or me-ishness.

1.3 For-Me-Ness

We’ve introduced universalism about mineness and me-ishness, explained why disorders of subjectivity might motivate one to soften this to a typicalist view of each form of inner awareness, and shown why both universalism and typicalism might be rejected in favour of an absentist view of each. Now, how does for-me-ness fit into the discussion? We’ve seen that for-me-ness is part of the universalist package proposed by Zahavi and others. Are there disorders of subjectivity that might motivate one to move from universalism about for-me-ness to typicalism? It would seem not.⁹ There do not seem to be disorders of subjectivity best explained by subjects coming to be unaware of their own experiences (Guillot 2017, this special issue). If unreflective experience does have for-me-ness, it is most plausibly an integral feature of unreflective experience. The real battle is thus between a universalist conception of for-me-ness on which every experience involves an awareness of that very experience and an absentist conception on which we are aware of our own experience only during reflective episodes of introspection.

Universalism about for-me-ness has been argued for on phenomenological, conceptual, epistemic and recollective grounds. The phenomenological grounds seem to have a special status for advocates of the view. Kriegel suggests that ‘…inner awareness is simply phenomenologically manifest’ (2009, p. 50) and Zahavi suggests that accurate phenomenological description is ‘the best argument to be found’ for thinking that non-reflective experience is characterised by ‘pre-reflective self-consciousness’ of that very experience (2006, p. 24). The problem with these appeals to phenomenology is that many claim to be unable to find such inner awareness when they reflect on their experience. As Gennaro states ‘[i]t does not seem to me that I am consciously aware (in any sense) of my own experience when I am, say, consciously attending to a play or the task of building a bookcase.’ (2008, p. 48). Similarly, Mehta reports ‘[w]hen I conduct phenomenological investigation into any peripheral awareness of my experiences, attempting to set aside any theoretical prejudices, I haven’t a clue whether I’ve ever had such peripheral awareness.’ (2013, p. 364). (Also see Gertler 2012 and Coleman 2015).

This phenomenological disagreement is curious. How could either side be getting their own phenomenology so dramatically wrong? Kriegel attempts to explain the skeptic’s phenomenological reports by noting how hard it is to spot. He notes that it is much less impressive than other aspects of phenomenology, that it is peculiarly unintrospectible and that its presence in all experience means that we are unable to discern it by contrasting it with experiences that lack inner awareness (2009). In the other direction, absentists have attempted to explain the universalist’s error in terms of the refrigerator-light illusion: whenever we introspect we have an inner awareness of our experience, so it is easy to make the mistake of reading this inner awareness into our non-reflective experience (Schear 2009). Accusations of phenomenological error cut both ways, and neither side provides compelling reasons to think that those on the other are in error about their experience (McClelland 2015).

What about the conceptual argument for universalism about for-me-ness? Zahavi and Kriegel seem to suggest that there is something unintelligible about the idea of conscious states that lack for-me-ness. After all, the difference between conscious states and non-conscious states is precisely that conscious states have for-me-ness. The difficulty with this argument is that it trades on a loaded understanding of ‘for-me-ness’ (Stoljar 2016, Guillot 2017 this special issue, Howell and Thompson 2017 this special issue, Farrell, J. The lexical account of “what it is like” talk (submitted for publication), Farrell, J. Higher-order theories of consciousness and what-it-is-like-ness (submitted for publication)). One can acknowledge that conscious states are necessarily states that are for the subject whilst denying that subjects are thereby aware of that conscious state. On this view, when one has an outer awareness of the sunset the sunset is present for you. This special relationship with the subject does not entail that besides one’s outer awareness of the sunset, one has a second inner awareness of the experience itself (Gertler 2012). It seems that there is no quick conceptual route to the conclusion that all experience is characterised by an inner awareness of that experience. If it is true, it is not something that is simply built into our concept of the conscious/non-conscious distinction.

The third kind of argument for universalism about for-me-ness is epistemological. A number of considerations suggest that we always know about the conscious experience we are having. When asked about our concurrent experience, we can report on it without the need for explicit reflection (Zahavi 2004; Kriegel 2009). When we introspect, we are not surprised by new discoveries about the experience we are having (Kriegel 2009). In order to know these things about our unreflective experiences, we must have an inner awareness of that experience. Since this kind of knowledge of experience seems to be present whenever we are conscious, inner awareness must be a universal feature of experience. The problem with this line of argument is that knowledge of something does not presuppose phenomenal awareness of that thing. Consequently, our knowledge of our concurrent phenomenal state needn’t entail a phenomenal awareness of that state.¹⁰ Perhaps when we are phenomenally aware of the sunset we also have non-phenomenal knowledge of that very experience without the need for any inner awareness (Thomasson 2006; Schear 2009).¹¹

A fourth kind of argument for universalism appeals to episodic memory. Chadha (2017, this special issue) describes different interpretations of an argument from memory given by Dignāga. The simplest version starts with the claim that we can only remember some object (or event) if we have experienced that object. This goes for experiences too: we can only remember an experience if we have an experience of that experience. Thus we must experience at least those experiences which we later remember (or could remember). And we can generalise the result by appealing to Husserl’s claim that every conscious state is, or can be, remembered immediately after it occurs (Thompson 2014). A possible objection here is to clarify the initial claim in the following way. The reason why we can’t remember an object without having experienced it is because having an episodic memory of an object just is having an episodic memory of experiencing the object. Strictly speaking, there are no episodic memories of objects, if these are understood as something distinct from memories of experiencing those objects. Talk of remembering objects is just shorthand for talk of remembering experiencing objects. But that gives us no reason to accept a general rule that having an episodic memory of x requires having experienced x, and so no reason to accept that we must have had experiences of any remembered experience.

1.4 Introducing the Contributions

Several of the papers in this special issue address these questions about the place of inner awareness in unreflective experience, and we have already touched on some of the contributions that they make to the debate.

Marie Guillot’s paper argues that the three forms of inner awareness introduced above are distinct and should not be conflated. She shows that influential arguments in this area rely on tacit equivocations of these different kinds of inner awareness. A detailed reflection on disorders of subjectivity leads her to conclude that mineness and me-ishness are not universal features of experience. Mineness, she proposes, is plausibly absent in both thought insertion and depersonalisation/Cotards, and me-ishness is plausibly absent in at least some cases of depersonalisation/Cotards. She maintains, however, that for-me-ness is a universal feature of experience.

Martine Nida-Rümelin’s (2017) invited paper argues that self-awareness plays a crucial role in our understanding of the very notion of an experiencing subject. The idea is that when we instantiate an experiential property—i.e., undergo an experience—such as being presented with pinkishness, we are thereby aware of this experiential property, and of ourselves as the subject to whom pinkishness is presented. Nida-Rümelin prefers a different framework to the one we’ve been assuming, but, in our terms, she is a universalist about for-me-ness. She also seems to be at least a typicalist, and perhaps a universalist about something similar to me-ishness and something similar to mineness: in undergoing experiences, we are aware of ourselves as the subject of the experience, but we are not the objects of this awareness. As well as being non-objectual, this self-awareness is pre-reflective and pre-conceptual, and, she argues, it grounds our understanding of what it is to be a synchronically (and diachronically) unified experiencing subject.

Monimha Chadha’s paper argues for a ‘realist representationalist’ interpretation of the Buddhist-Abhidharma account of consciousness. She argues that conscious representations are ‘Janus-faced’: pointing both out into the world, and in towards themselves. Chadha is a universalist about for-me-ness, offering arguments for this claim (such as that from memory described above) ultimately inspired by those of Dignāga. She argues that, despite what has often been claimed, this view does not lead to idealism, i.e., require that the objects of experience are wholly internal, mental entities. On the Buddhist-Abhidharma view she describes, however, there is no subject of consciousness, but only a constantly changing string of self-aware conscious states. Thus there is no (veridical) awareness of the subject of experience, or of the subject’s owning the experience.

Robert Howell and Brad Thompson’s paper offers a skeptical survey of the arguments available for the view that experiences are phenomenally given as mine. They introduce two conditions that any putative quality of me-ishness must satisfy: that it makes some contribution to a subject’s phenomenology and that it presents or refers to the self. They go through a range of arguments for mineness, including phenomenological arguments, arguments from self-reference and self-knowledge, arguments from pathological cases and arguments from self-consciousness. They find, however, that none of these arguments yields a conception of subjective character that satisfies their two conditions. But rather than dismissing the notion of phenomenal mineness outright they propose that there are a number of aspects of experience – including the sense of agency, affective experience and perspectival perception of affordances – that yield a kind of implicit phenomenal awareness of the self, and which might underwrite the mistaken intuition that there is a primitive sense of phenomenal mineness.

1.5 Lessons and Directions for Future Research

Although the first three papers are broadly in favour of some kind of inner awareness, the differences between them highlight the variety of views that can be taken on the place of inner awareness in unreflective experience. And even though Howell & Thompson’s paper offers a host of skeptical considerations, even they concede experience often involves some kind of awareness of the self, albeit of a reducible kind. Together these papers lend further weight to the conclusion that there is something to the idea that we have an inner awareness of our unreflective experiences, but also highlight that a great deal of work is needed to pin down what form this inner awareness takes, and to distinguish it from forms of inner awareness that are not typically present in experience.

The questions raised regarding the presence of inner awareness in non-reflective experience are important for a number of reasons. On a descriptive level, answering these questions will allow us better to capture a subtle aspect of our phenomenology, or else to free ourselves of the mistaken impression that there is such a thing. These questions are also important on an explanatory level. Since subjectivity is thought to play such an integral role in what makes a mental state conscious, different views of subjectivity drive different explanations of consciousness. Kriegel’s self-representationalist theory of consciousness, for example, relies heavily on his claims about the pervasive presence of inner awareness in consciousness (2009). In contrast, Levine’s subtly different take on subjective character motivates his rejection of representationalist theories of what makes a mental state conscious (2015). Since a theory’s ability to accommodate the subjective character of consciousness is so important to the assessment of that theory, it is vital to make sure that subjective character has been properly understood.

The considerations raised, along with the contributions to be found in this special issue, highlight a number of avenues for future research. On the conceptual side of things, there is clearly further need to regiment the various subjective phenomena referred to during discussion of inner awareness. We must isolate the different aspects of subjectivity, map any entailment relations that hold between them, and resist conflating different forms of inner awareness with one another. In particular, more work needs to be done on the notion of experiences being for the subject. The ubiquity of Nagel’s locution gives this phrase special force, yet there clearly remains confusion about what it means for an experience to be like something for the subject. Clearing up this confusion is likely to be crucial to progress on these issues (see Stoljar 2016; Farrell 2017a, b).

On the empirical side of things, there is more work to be done on disorders of subjectivity. Speculative descriptions of the phenomenology of such disorders are constrained by the concepts we have at our disposal. As our conceptual tools become more refined, so too should our ability to describe these disorders. Not only should we be better able to map the relations between the various disorders of subjectivity, we should be able to improve our understanding of the source of such disorders. And as the disorders are better understood, our appreciation of the varieties and prevalence of inner awareness will be improved.

2 Inner Awareness and Reflective Experience

In the previous section we focused on the connections between inner awareness and non-reflective experience. In this section we turn to the inner awareness present when we actively attend to our mental lives. In keeping with the papers in this special issue, we will call this reflective inner awareness ‘introspection’. It is worth noting that this term is used in slightly different ways in the literature, even when authors aim to describe, rather than engage in, debates. Sometimes, for example, ‘introspection’ is used to refer to that way (if there is one) we have of knowing about (or becoming aware of) our own minds (Kind n.d.; Schwitzgebel 2016). And sometimes the term is used only for those ways we have of knowing about (or becoming aware of) our own minds that involve some sort of ‘turning inwards’ (Gertler 2009). On the first use, behaviourist (e.g., Ryle 1949) or ‘theory theory’ (Gopnik 1993; Gopnik and Meltzoff 1994; Carruthers 1996) accounts of self-knowledge will count as introspection, on the second they will not. Here we use ‘introspection’ in something more like this second way, with the emphasis on awareness, rather than knowledge. We also assume (noting that this is an assumption, see Schwitzgebel 2012) that introspection is a unitary process or phenomena.

2.1 What Are We Introspectively Aware of?

As we saw above, when it comes to non-reflective inner awareness, we can ask whether this involves awareness of the experience undergone (for-me-ness), of the subject of the experience (me-ishness), or of the fact that the subject is undergoing the experience (mineness). An analogous distinction can be made for introspection. We can ask whether, when we introspect on an experience, we are aware of the experience itself, of ourselves, or of ourselves as having the experience.

Hume (1739–40) famously denied that he was aware of himself when he introspected and many, but not all, contemporary philosophers have concurred (Shoemaker (1996) and Howell (2010), for example, agree, while Chisholm (1976) and Strawson (2009) do not). The papers in this special issue, however, focus on the awareness of experiences and of our having experiences, so we will do likewise in this introduction.

Whether or not we are aware of ourselves in introspection, there is general agreement that we can become aware of at least some aspects of our mental lives in introspection (although some, such as Ryle (1949), are often understood as denying even this). In particular, it is usually taken that we have introspective access to some of our mental states. Few accept the view, associated with Descartes (1641/1984) that we can introspect on all our mental states. Psychological studies (e.g., Nisbett and Wilson 1977) suggest that some of our motivations are unintrospectible, for example, and we are also unaware of the early-stage cognitive processing involved in perception (Marr 1983).

We might think, however, that introspection gives us awareness of attitudes like beliefs and desires (see, e.g., (Nichols and Stich 2003)) or, at the very least, of our conscious experiences (see, e.g., Hill 1991; Goldman 2006): perceptions, bodily sensations, and perhaps occurrent thoughts. Each of the papers in this special issue which focus on introspection accept that introspection can give us awareness of at least some of our conscious experiences. Giustina and Kriegel (2017, this special issue) argue that if we accept a particular theory of consciousness—one on which, amongst other things, the fact that a mental state represents itself as having some phenomenal character is what determines that it has that character—then introspection gives us awareness of all conscious experiences. In contrast, Shiller (2017, this special issue) argues that if we don’t assume that our being aware of our conscious experiences is constitutive of their being conscious, there are good reasons to think that we have conscious experiences that we cannot introspect. Taking for granted Giustina and Kriegel’s conditional claim—that if we accept their favoured theory of consciousness, we can introspect on all our experiences—we face the question of whether to accept this theory, and so acknowledge that we can introspect on all experiences, or to accept Shiller’s arguments that there are experiences we cannot introspect on and so reject this theory of consciousness.

In giving the argument just described, Giustina and Kriegel are careful to distinguish between thing-introspection—our being aware of our experiences—and fact-introspection—our being aware of the fact that we are having our experiences. They do this by appealing to Dretske’s (1993) distinction between fact-awareness and thing-awareness in perception. Just as we can be perceptually aware of things (such as trees), Dretske argues, we can also be perceptually aware of facts (such as that a tree is green). Similarly, Giustina and Kriegel say, in introspection we can be aware of a thing (an experience such as hunger) or of a fact (such as the fact that we are hungry). For this analogy to hold, it must be that introspection is perception-like in the relevant ways. Whether introspection is best understood according to a perceptual model is a controversial matter amongst philosophers (see next section). But accepting this analogy and also Giustina and Kriegel’s claim that fact-introspection requires and builds on thing-introspection means that, if Shiller is right that there are some things (experiences) we cannot introspect, then, for these experiences, we cannot become aware by introspection of our undergoing them.

2.2 How Are We introspectively Aware?

Explanations of how we can introspect our mental lives given in the philosophical literature can be roughly divided into two camps. On one hand, we have views on which introspection is understood as being—to a greater or lesser degree—similar to perception. Examples of this sort of account include “inner sense” views (Locke 1689; Armstrong 1968; Lycan 1997) and “self-scanning” views (Nichols and Stich 2003). On the other hand, introspection can be explained in terms of logical or constitutive relations. On some views of this kind, the introspected state is in some sense contained within or incorporated into the introspecting state (Burge 1988; Gertler 2001; Chalmers 2002; Papineau 2002; Horgan and Kriegel 2007). On others, there is a rational connection between the two states (Shoemaker 1996; Gallois 1996). In yet others, it is the introspecting state which determines the features of the introspected state. An example of such a view is that assumed in Giustina and Kriegel 2017. Here subjects are aware of their conscious states because those states represent themselves, and a conscious state has the phenomenal properties it has because it represents itself as having them rather than the other way around, as is perhaps most obviously the case in perception-like accounts.

Overgaard and Mogensen (2017, this special issue) divide the options up a little differently. They note that accounts of introspection in the cognitive science literature tend to fall into one of two categories. Stage models hold that introspected states are in some sense prior to introspecting states: the former influence the latter, but not vice versa. Reciprocal models allow the influence to go both ways. The philosophical views just described which hold that introspection is perception-like, or involves containment, or is governed by considerations of rationality, perhaps fit most naturally into the category of stage models. But there seems no in principle reason why one could not hold one of these views whilst claiming that influence flows both ways.

Overgaard and Mogensen themselves argue for a more complicated model on which introspection involves parallel networks, one concerned with “first-order” information, the other with information available for meta-cognition, including introspection. The two networks share a common representational content and interact with each other, although introspection itself doesn’t alter the content of the first-order conscious state. Each network can give rise to conscious experiences and to action. This more complicated model is required, they argue, to account for evidence derived from psychological experiments (described below).

2.3 How Reliable Is Introspective Awareness?

Another central question concerning introspection concerns its epistemic features. It is generally agreed that introspection gives us privileged access to our mental lives—at the very least, the access we have to our mental states is different to the access we have to mental states which are not ours. This idea of privileged access can be cashed out in a number of ways. Perhaps the most extreme way of doing so is by holding a view associated with Descartes (1641), namely that introspection is infallible: the inner awareness we gain via introspection is never mistaken.

Taken as a general claim about introspection, the infallibility claim does not look plausible. If introspection is perception-like, then we should expect that introspection can go wrong in the same sorts of ways that perception can (Churchland 1988). And empirical work suggests that we can be mistaken about our reasoning processes (Nisbett and Wilson 1977), and, as research into change blindness shows, about our perceptual states (O’Regan et al. 1999). But this leaves open the possibility that, for some kinds of states, our introspective access is infallible.

Another way of understanding the privileged access that introspection gives us appeals to incorrigibility instead of infallibility (Rorty 1970). The idea here is that the beliefs or judgments which subjects have as a result of introspection cannot be shown false by appeal to other ways we have of learning about mental characteristics. Observing someone’s behavior or scanning their brain, on this view, cannot show that an introspecting subject is mistaken about their own mental states. This does not require introspective beliefs to always be true, but only that introspection is a better guide to our mental lives than anything else.

A third option is to look to the features of the mental states introspected to explain privileged access. One way to do this is to hold that mental states are self-intimating: we are always aware of those mental states that we are in (again, this view is often associated with Descartes (1641)). As we’ve seen, this view is not thought plausible by contemporary philosophers and psychologists, but a restricted version which holds that conscious states are self-intimating is a popular one. The idea that what it is for a state to be conscious—in one sense of ‘conscious’—is for its subject to be aware of it is appealed to by both higher-order (Armstrong 1968; Lycan 1996, 1997; Rosenthal 2005) and reflexive (Kriegel 2009) theorists. But although these authors hold that having an experience involves being aware of it, most would not claim that having an experience requires introspecting it, if that is to be understood as an active reflection on the experience.

In this special issue Giustina and Kriegel argue that conscious experiences are self-intimating, and that at least one kind of introspective access we have to them is infallible. They assume a theory of consciousness on which a subject’s awareness of their experiences determines how (the phenomenal character of) the experience is. Thus these experiences are self-intimating—they are conscious only because we are aware of them—and the awareness we have cannot be mistaken since the awareness determines how they are. Giustina and Kriegel are careful to distinguish this claim from a nearby one, namely that introspective awareness of facts—such as that one is hungry—is infallible. Most objections to the idea that introspection of conscious states is infallible, they claim, are objections to this latter claim. But since the move from introspection of things (i.e., experiences) to introspection of facts (i.e., that one is undergoing a particular experience) involves the application of concepts—a procedure that is vulnerable to error—that fact-introspection is not infallible gives us no reason to infer that the same is true of thing-introspection. As we saw above, Shiller argues that if we don’t assume the sort of theory of consciousness favoured by Giustina and Kriegel, there are good reasons to think that some conscious states are not even accessible via introspection. If the idea that introspection is infallible requires that, whenever we attempt to introspect, we do so veridically, it follows from Shiller’s conclusion that introspection of conscious states is not infallible.

2.4 Reflective and Non-reflective Inner Awareness

Distinguishing reflective and non-reflective inner awareness, as we have done in this introduction, invites us to consider how they are related. One question is which (if either) form of inner awareness is more demanding, or which is a prerequisite for the other. On the one hand, we might think that introspection requires and builds on unreflective inner awareness (Zahavi 2004). If this is so, then showing that we lack non-reflective inner awareness of some aspect of our mental lives suffices for showing that we lack introspective access to it as well. On the other hand if the difference between reflective and non-reflective introspection is more like that between passively casting an eye over a scene, and actively looking at (or for something in) the scene, then it is plausible that being introspectible (if not actually introspected) is a necessary condition for being the object of non-reflective inner awareness. The model of introspection given by Overgaard and Mogensen 2017, and perception-like accounts of introspection would seem to fit this picture of the relation between reflective and non-reflective inner awareness. If this sort of view is correct, then showing that some aspect of our mental lives cannot be introspected is enough to show that we lack non-reflective inner awareness of that aspect. This would mean, for example, that those who accept Shiller’s argument that there are conscious states which we cannot introspect must allow that there are conscious states that we lack non-reflective inner awareness of, thus giving us an argument against universalism about for-me-ness.

A second way in which we can investigate the relations between reflective and non-reflective inner awareness concerns what happens when we shift between these two processes. A common view here is that the shift (or attempted shift) from non-reflective to reflective inner awareness is an unsurprising and undramatic one. This is often taken to show that experiences are transparent or diaphanous (Moore 1903; Harman 1990). The idea is that when we attempt to reflect on the inner awareness we have of our experiences—an experience of a sunset, for example—we find that what we are aware of are not features of the experience, but features of the object of the experience—the pinkishness of the sunset, say. A similar point is made by those who claim that there is no phenomenology distinctive to introspection: any phenomenology undergone is contributed by the introspected state, not the introspecting one (Lycan 1996, Shoemaker 1996, Rosenthal 2000, Siewert 2012).

Those wishing to resist this picture might appeal to the aforementioned refrigerator-light illusion (see Schear 2009; McClelland 2015; Howell and Thompson 2017 this special issue). In order to make claims about non-reflective inner awareness, it is tempting to think, we must somehow attend to, or reflect on this inner awareness. But the inner awareness we have when we reflect on that inner awareness is, of course, reflective inner awareness. Just as, whenever you open the refrigerator to see if the light is on when the door is shut, the light is on (but the door is no longer shut), whenever you reflect on your non-reflective inner awareness in order to characterise it, certain features are salient (but the inner awareness is no longer non-reflective). We might take this to suggest that, just as the refrigerator light is off when the door is shut, we have reason to doubt that there is such a thing as non-reflective inner awareness. Or, more modestly, we might take it to show only that we need to be careful in attributing features to non-reflective inner awareness on the grounds of what we find when we reflect. Some of the empirical research described in Overgaard and Mogensen (2017, this special issue) which involves behavioural, rather than verbal, responses to stimuli may have a role to play here.

2.5 Introducing the Contributions

In their invited paper, Morten Overgaard and Jesper Mogensen take introspection to be second-order attention to first-order conscious states. They note that introspection is hard to investigate empirically, because the same measure—an introspective report—tends to be used for both conscious states and introspection on conscious states. They describe two models of consciousness and introspection popular amongst cognitive scientists. On the stage model (which tends to be assumed in neuroscience) conscious states influence introspection, but not vice versa; on the reciprocal model influence goes both ways. They argue that there is little data which can help us decide between these models, but that what there is pushes us towards the reciprocal model. For example, there is evidence that we get different results if we ask subjects about the stimulus they are exposed to or about their experience of the stimulus. And when the latter involves a scale (specifying degrees of clarity), rather than a binary choice (‘seen’ vs ‘not seen’), results differ again, and not merely in that the former are more fine-grained than the latter. Overgaard and Mogensen suggest that this is best explained by a third model, the integrative model. Their view is that introspection (which is itself conscious) and conscious experiences should be modelled as parallel networks that engage with the same representational content, and which interact with each other (and with other states, such as proximal or distal goals). Their model is a purely cognitive one, and is based on evidence from psychological experiments and theoretical arguments, but, they note, it gains support from its similarity to a neurocognitive model they have described in previous work (Overgaard and Mogensen 2014, 2015).

In their invited paper, Anna Giustina and Uriah Kriegel argue that there is at least one kind of introspection—thing-introspection—which is reliable. Drawing on Dretske (1993) they distinguish fact-introspection—e.g., my inspection on the fact that I am hungry—from thing-introspection—e.g., my introspection on my hunger—and argue that the latter is more fundamental. Arguments which aim to show that introspection is unreliable tend to aim at fact-introspection. Plausibly fact-introspection stems from thing-introspection plus concept application, and it is only at the latter stage that errors can creep in. Giustina and Kriegel then argue that, if we adopt a particular theory of consciousness, thing-introspection is infallible and fact-introspection is usually reliable. The theory of consciousness is Kriegel’s (2009) version of a self-representationalist theory of consciousness on which a mental state’s representing itself as having some phenomenal character is what determines that it has that character, and on which introspection is just the result of attending to the mental state rather than what the mental state is about.

Derek Shiller argues that there are good reasons to think that we have hidden qualia: qualia that we cannot tell that we have by introspection. He gives four arguments for thinking that hidden qualia are possible. First, since there are partially hidden qualia—qualia that we can introspect on only with effort or in the right context—it would be arbitrary to claim that none could be fully hidden. Second, since some qualia have little effect on our beliefs—perhaps because we are not confident that they are present—it would be arbitrary to deny that some have no effect. Third, some animals plausibly have hidden qualia since they plausibly lack the cognitive sophistication required for introspection. And, fourth, we can appeal to Hume’s dictum—that distinct entities can possibly fail to co-occur—to show that hidden qualia are possible. Shiller then gives two arguments for thinking we actually have hidden qualia. The first notes that, as evolved creatures, it is unlikely that our introspective powers are without limitation, so we should expect that we have some hidden qualia. The second relies on the claim that there are certain kinds of qualia that we plausibly have—including qualia engendered by ‘primitive’ parts of the brain, or qualia produced by neuronal ‘static’—which will not show up in introspection.

2.6 Lessons and Directions for Future Research

The answers to the questions raised here and in the contributions to this special issue are important to the explanation of introspection and to methodological issues in psychology and philosophy concerning the epistemic value of introspection. They are also important because of their interactions with questions about the nature of unreflective experience and its relationship with inner awareness. Exactly what we say about this will have consequences for the scope of the papers featured here.

The papers in this special issue suggest a number of directions which further empirical and conceptual research might take. Overgaard and Mogensen describe a cognitive model of how introspection occurs which respects recent psychological work. As they note, more can be done to flesh out their integrative model, and further empirical work in neuroscience will no doubt shed yet further light on modelling introspection. It is interesting to note that there is no clear mapping between the models of introspection offered in cognitive science and those accounts given in philosophy. Investigation of how these two ways of dividing things up interact is liable to be fruitful for those involved in both disciplines.

On the conceptual side, we can ask whether Giustina and Kriegel’s distinction between fact and thing introspection—motivated by an analogy with perception, but paired with a non-perception-like account of introspection—is sustainable. The disagreement between Shiller and Giustina and Kriegel raises the question of whether we can establish or reject there being an a priori connection between phenomenal qualities and introspectibility without adopting a substantive theory of consciousness. It would be interesting to see whether a more fully described version of Overgaard and Mogensen’s dual-network model favours one side of this debate over the other. A further question concerns which (if either) of reflective and non-reflective inner awareness is prior to the other.