1 Introduction

The question of whether we experience the self has attracted a lot of attention in philosophy and psychology. By “self” I mean the subject of experience—that inner perspective on ourselves as ourselves that no one else has access to. Most contemporary philosophers deny that we have any such substantial, inner experience of the self. It is often said that introspection reveals various features or properties of the self (e.g., thoughts) that we can identify and (metaphorically) point to, but that the self is not similarly available as an object of consciousness. We cannot, in other words, find the self to whom experiences manifest, which has led to the widespread thesis that the self is elusive (Cassam 1997; Chisholm 1969; Duncan 2019; Hume 1739/1975; Gottlieb 2022; Zahavi 2014). As one prominent source of this sceptical view, consider Hume’s (1739/1975) oft-cited denial:

For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe any thing but the perception. When my perceptions are remov’d for any time, as by sound sleep; so long am I insensible of myself, and may truly be said not to exist (I.iv.6).

According to Hume, what introspection reveals are various thoughts, emotions, sensations, and perceptions—but nothing further that constitutes a distinctive experience of the self. Although it might naively appear as though we experience ourselves as a distinct experiencer of our experiences, feeler of our feelings, or thinker of our thoughts, turning our attention inward reveals various mental events and states, but no introspectable self that has these mental states. In short, the received wisdom in philosophy is that the self does not show up in experience. I call Hume’s denial the “received wisdom” because his view appears to have commanded widespread philosophical agreement. Many contemporary philosophers of mind, metaphysicians, cognitive scientists, and scholars as diverse as Kant and Wittgenstein have endorsed the phenomenological plausibility of Hume’s scepticism.Footnote 1

However, recent work is challenging the received wisdom by appealing to a symptom of schizophrenia called thought insertion. People living with thought insertion claim that some of the thoughts they experience are not produced by them, attributing the source of these thoughts to other agents instead. Thought insertion is commonly explained as a deficit in the sense of being the agentive thinker of our thoughts, while the sense of being the subject of our thoughts remains intact. Duncan (2019) appeals to the disrupted experiences of those living with schizophrenic thought insertion to offer the first sustained argument in support of the claim that this condition can help us to see that we normally experience ourselves as the agent or author of our thoughts.Footnote 2 The argument is simple and elegant: the best explanation of thought insertion is that the sense of agency is missing, what is missing is missed only because it is not normally missing, therefore, the received wisdom is wrong—we normally experience ourselves as the agent of our thoughts.

In this paper, I argue that schizophrenic thought insertion is a problematic way of vindicating experiencing ourselves as the agent of thoughts. There are two reasons for this. First, appealing to a missing sense of agency to explain thought insertion fails to discriminate it from other cases where agency is lacking. Second, arguments about consciousness and the self that use thought insertion as a key example often exhibit an overreliance on a couple of recycled examples, thus presenting a decontextualized account that is carved out in isolation from its wider psychopathological context—a context that I argue is essential to understanding the condition. Instead, exploring the wider clinical and psychopathological context of thought insertion most naturally leads to construing thought insertion as a disturbing or disrupting of the subjectivity of experience (Henriksen, Parnas, and Zahavi 2019). This still allows thought insertion to vindicate experiencing ourselves as the subject of our thoughts, which is where future efforts should be directed.

Here is the structure of the paper. I begin in §2 by providing a brief overview of schizophrenic thought insertion, including the most common explanation of the condition as a deficit of agency and retention of subjectivity. In §3 I explain Duncan’s argument for why thought insertion vindicates self-experience. In §4 I address the problems with the agency-deficit explanation of thought insertion and methodological problems with offering decontextualized accounts of the condition. In §5 I show that exploring the wider psychopathological and clinical context of thought insertion is essential to understanding the condition and its implications for self-experience. In §6 I offer some directions for further research on self-experience and conclude in §7.

2 Schizophrenic Thought Insertion

Imagine that you are standing on a balcony or cliffside, staring at the steep drop below. Suddenly, the thought occurs to you, “I could jump off.” Or suppose you spot another person nearby who is leaning over the railing. As you look at their feet lifting off the ground, you think, “It would be so easy to push them over.” Everyday experience acquaints us with the fact that intrusive thoughts such as these might occur to us from time to time, and we might even puzzle over why our mind produced them. But now imagine that these thoughts did not feel like your thoughts. Instead, suppose it felt like somebody else thought them, and put them into your mind. To illustrate this, here is one commonly recycled description of just such an experience: “Thoughts are put into my mind like ‘Kill God.’ It is just like my mind working, but it isn’t. They come from this chap, Chris. They are his thoughts” (Frith 1992, p. 66). This might give you an idea of what some people living with schizophrenic thought insertion go through.

People living with thought insertion still experience their thoughts in a first-person manner but deny that they are the one forming or producing them. Additionally, they often attribute these thoughts to the authorship of something or somebody else. To make sense of this puzzling phenomenon, the standard approach in the philosophical and psychological literature has been to distinguish between different aspects of our conscious experience, and to argue that some features go missing while others remain intact. Stephens and Graham (1994) distinguish between a sense of agency (an experience of producing, bringing about, or doing one’s thoughts) and subjectivity (that thoughts occur to each of us in a distinctly first personal way), to argue that thought insertion involves a breakdown in the former but not the latter. Many philosophers and psychologists likewise argue that a sense of subjectivity remains intact, but that a sense of agency breaks down (Bortolotti 2010; Carruthers 2007; Duncan 2019; Stephens and Graham 2000; Roessler 2013; Sass 2000). Even accounts that appeal to an additional feeling of alienness as a way of explaining the subsequent externalizing of the thoughts still commonly take a sense of agency to be missing (Gallagher 2015).

In what follows, I will assume that the reports of those living with thought insertion reflect real disturbances on the experiential level. Furthermore, it should be noted that there are other ways of making sense of thought insertion that do not require appealing to a breakdown in agency.Footnote 3 Nevertheless, appealing to a missing sense of agency is a commonly used explanation in the literature on thought insertion, and Duncan makes use of it to support his argument about self-experience.

3 The Self Shows Up in Experience

Duncan (2019) argues that imagining what it would be like to undergo thought insertion (i.e., to cease having it seem like we are the author of our thoughts) can help us to attend to how we normally experience ourselves. Or as he puts it more strongly in his concluding remarks: “It allows us to see that self-experience is true” (p. 315). To begin, let me explain what Duncan says self-experience amounts to, and then how he ties this into thought insertion.

SELF-EXPERIENCE: The self shows up in experience; that is, an inner experience of one’s self has its own proprietary phenomenology that normally constitutes a distinctive component of one’s total phenomenal experience (p. 302).

There are a few key points to note here. First, Duncan claims that the self normally shows up in experience—not that it always does. Second, an experience of the self has its own proprietary phenomenology, which means it adds something to our total phenomenology that would not otherwise be there. Third, an experience of the self is distinctive, in the sense that it is different from other phenomenal experiences. By “self” he means a conscious subject of experiences as opposed to a bundle of experiences (the very thing Hume could not find), which is not reducible to a mental state or property.

Although Duncan accepts (and his argument from thought insertion endorses) the claim that we can experience ourselves as the subject of our thoughts, he states that his primary aim is to “argue that the self normally shows up in experience as the author or agent of one’s thoughts” (p. 303). The distinctive phenomenal quality which is said to characterize an experience of the self as the author of our thoughts is as follows: “To experience oneself as the agent of one’s thoughts is to experience oneself as the individual who is bringing about, producing, or thinking one’s thoughts” (p. 304–305). We have, as he puts it, a “feeling of doing” our thoughts that gives us the sense that we are their author. Experiencing ourselves as the agent or author of our thoughts is supposed to be what is missing in cases of thought insertion, which he argues is suggestive of how we normally experience ourselves.

Duncan’s rationale for why cases of thought insertion show that we normally experience ourselves as the author of our thoughts is that a breakdown in one’s sense of agency best explains the experiential deprivation or disruption. In other words, something is missing in the experiences of those living with thought insertion. Thoughts are not missing, the subjectivity of experience is not missing, nor is a belief of authorship missing (thoughts are reported as being authored, just by other people). Instead, what Duncan argues is missing is an experience of oneself as the agent or author—and that this is missed only because it is normally not missing (p. 306). He claims that the best explanation for the difference between people with and without thought insertion is that we must normally experience ourselves as the author of our thoughts otherwise there would be no sense in speaking of a deficit of agency. Therefore, the self must normally show up in our phenomenally conscious experiences in the role of the author of our thoughts.

This argument relies on the claim that a sense of one’s own agency is missing in cases of thought insertion, as appealing to a missing feeling of what he calls “brute authorship”—a kind of generic, non-self-implicating sense of agency in normal cases—fails to explain thought insertion. His reason is that inserted thoughts are claimed to be authored, but by someone else. So, there is either a belief or some sense of authorship present, it is just not one’s own authorship. Therefore, appealing to a breakdown in a brute, non-self-implicating feeling of authorship that is present in normal cases and absent in schizophrenia boils down to understanding thought insertion as a breakdown in that person’s brute feeling of authorship. This means we are still appealing to self-experience to make sense of thought insertion.

On Duncan’s view, we ordinarily attribute our thoughts to ourselves because we experience ourselves as their author. In cases of thought insertion, people fail to attribute their thoughts to themselves because they no longer experience themselves as their author. According to Duncan, self-experience is pervasive, which is perhaps what makes it difficult to attend to. But thinking about thought insertion can help, as it allows us to imagine what it would be like if it stopped seeming like we agentively produce our thoughts. It is concluded that there is no better explanation for the experiential deprivation of those living with thought insertion that does not presuppose that the self normally shows up in experience as the author of our thoughts. Therefore, research on schizophrenic thought insertion gives us good reasons to think that self-experience is true.

4 Problems with the Agency-Deficit Explanation

It is doubtful, however, whether characterizing thought insertion as a breakdown in the sense of agency in thinking really is the best explanation for the experiential deprivation. As Langland-Hassan (2008) aptly puts it: “[t]he unsupported assumption that there is a clearly understood phenomenology or ‘emotion’ of thought agency—one that can explain reports of thought insertion by its absence—is a pervasive weakness in the literature on this subject” (p. 371). Given the comprehensive rebuttals the agency-deficit explanation has received, I will not revisit the topic in thorough detail, and instead direct interested readers to the relevant critiques (Billon 2013; Billon and Kriegel 2015; Howell and Thompson 2017; Langland-Hassan 2008; Mullins and Spence 2003; Ratcliffe and Wilkinson 2015). Nevertheless, some exposition will help to motivate adopting a different explanation of thought insertion, so I will briefly touch on the two most prominent problems with the agency-deficit explanation. Farrell and McClelland (2017) identify them as follows.

In particular, this interpretation is unable to differentiate cases of thought insertion from other cases in which subjects plausibly lack a sense of agency over their thoughts, such as the unbidden thoughts that neurotypical subjects experience routinely, and that subjects with obsessive disorders experience more persistently (p. 6).

Everyday experience acquaints us with the occurrence of unbidden (or intrusive) thoughts that come to mind without us intending to think them. A joke might occur to you in the middle of an exam or funeral, trivial desires and distractions seem to appear en masse upon trying to concentrate, songs get stuck in our heads, anxious thoughts interfere with our ability to sleep, and situations involving heights seem especially generative of thoughts about jumping or shoving. Such thoughts often occur to our own surprise and irritation, they may not even cohere with our values, intentions, and goals, and feel more like mental activities happening to or within us rather than mental actions. Nonetheless, despite possessing little to no sense of agency for unbidden thoughts, we neither disown them nor attribute them to someone else.

As Billon (2013) observes, nor can the difference between unbidden and inserted thoughts be explained as a matter of degree. Subjects with obsessive compulsive disorder persistently experience highly intrusive thoughts. Despite their very best efforts to resist such thoughts, eventually the compulsion to act them out can be overwhelming, engendering radical feelings of passivity. Yet people living with obsessive compulsive disorder neither disown their thoughts nor attribute them to the agency of something or somebody else. Given that thoughts can occur without a sense of agency without seeming to be inserted, something other than a breakdown in the sense of agency must explain what is going on in cases of thought insertion. But before offering a different explanation, let me say a bit more about what I think further motivates authors to accept the agency-deficit explanation.

It is easy to forget that an example offers a thin glimpse at certain aspects of a phenomenon, and it is noteworthy that in many philosophical arguments that make use of schizophrenic thought insertion (including Duncan’s), the most often-cited examples are the following by Frith and Mellor:

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

I look out of the window and I think the garden looks nice and the grass looks cool, but 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 his thoughts on to it like you flash a picture (Mellor 1970, p. 17).

It has been increasingly pointed out that many arguments about self-experience and the structure of consciousness that appeal to thought insertion exhibit an overreliance on these two examples without first having obtained a clear, clinical understanding of thought insertion more generally (Henriksen, Parnas, and Zahavi 2019, Ratcliffe and Wilkinson 2015). Authors are therefore apt to move very quickly from citing a few recycled examples to arguing that instances of thought insertion provide validating (or invalidating) counterexamples to whatever theory (consciousness, the self) they are out to endorse or criticize. The trouble with this approach is that it provides a decontextualized picture of thought insertion which misrepresents the phenomenon by overemphasizing certain features at the expense of others, making the agency-deficit explanation appear more plausible and generalizable than it really is. For example, both quotes involve thoughts that “come into” or are “put into” the subject’s mind by specific agential sources from outside, which might give the impression that these are central features of the experience of thought insertion.

But as the broader clinical literature documents, people experience thought insertion in much more varied ways (Jansson and Nordgaard 2016, Koehler 1979). Some patients will deny experiencing thought insertion if it is defined as alien thoughts being inserted into their head and will instead prefer to describe the thoughts as simply “being there” in the mind (Jansson and Nordgaard 2016). Others are unable to pinpoint the source of their thoughts, and some highly intelligent and reflective patients are unable to tell whether the sense of alienation stemmed from the thought contents themselves or their mode of presentation in consciousness (Henriksen, Parnas, and Zahavi 2019). Of course, variation in patients’ experiences is only to be expected, but this variation is not captured by using the same recycled examples. Moreover, as I will argue in the following section, the varieties of experience are in turn symptomatic of the often-neglected structure and quality of the altered experiential framework out of which thought insertion develops. Paying attention to the details of the symptomatology matters, not least for the fact that doing so challenges the prevailing view that thought insertion is best explained as the complete absence or preservation of certain features of consciousness.

To sum up this section, explaining thought insertion as a deficit in the sense of agency and retention of subjectivity fails to discriminate it from other pathologies of agency that are relevantly different. Moreover, by relying on a few recycled examples and rushing to discuss the philosophical implications without first exploring the wider psychopathological context and altered experiential framework out of which thought insertion develops, philosophical interpretations of thought insertion are apt to be incomplete. Not only is this broader context essential for understanding and offering a clinically relevant account of thought insertion, I will show that the implications for self-experience are philosophically rewarding too.

5 The Psychopathological Context of Thought Insertion

It has long been observed that schizophrenia spectrum disorders are preceded by a host of trait-like, non-psychotic, experiential anomalies before the official onset of psychosis, which empirical studies have consistently demonstrated in the last two decades (Parnas and Henriksen 2014). Clinicians and researchers have made good use of a psychometric checklist known as the EASE scale (Examination of Anomalous Self-Experience), which is a tool inspired by philosophical phenomenology and is used to understand the disrupted self-experiences of those living with schizophrenia. These experiences have been described in various ways, usually involving a nebulous feeling of inner existential alienation, of being different from other people, and of feeling a certain type of solitude, without the person being able to articulate precisely what this amounts to (Parnas and Henriksen 2014). The inner, experiential field is said to become increasingly anonymized, with the patient often experiencing their thoughts as appearing at a distance from them, feeling as though their thoughts are not generated by them, having too many thoughts, interfering thoughts, a sense of bodily estrangement, feeling as if they are distanced or somehow not really present in the world, and solipsistic feelings (Henriksen, Parnas, and Zahavi 2019). As schizophrenia develops further, this sense of diminishing self-presence and increasing self-alienation grows.

Patients express this increasing felt distance between themselves and their experiences in a variety of ways. Some report that the felt distance is extremely vivid, in that their thoughts are experienced like subtitles in a movie, where they need to listen to or read their thoughts to understand what they are. One patient recounted that everything inside his head was dark except a “light spot” in a slanted left position through which he perceived his thoughts and the world, remarking: “it’s like watching a television from a distance of 2m” (Henriksen and Parnas 2017, p. 180). Others claim that intrusive, emotionally neutral thoughts interfere with their primary stream of thinking. Some patients also report that their thoughts acquire spatial qualities, as if they occupy a specific position in the brain. These anomalous experiences often involve not feeling fully present in the world (e.g., lacking a central core), bodily estrangement from the mind, and various other forms of self-alienation (e.g., quasi-solipsistic feelings, ephemerality, and feeling transparent to the point where others can tell what they are thinking).

Thought insertion arises as an end phenomenon out of this altered experiential framework of a pervasively felt distance between the subject and their experiences. The process of increasing self-alienation or felt distance eventually becomes unbridgeable, to the point where the patient no longer recognises some of their thoughts to be their own due to their mental states appearing so unfamiliar and alien to them. Henriksen, Parnas, and Zahavi (2019) describe this psychopathological trajectory as follows: “initially, a patient experiences a ‘distance’ to his own thoughts; then certain thoughts appear not to be generated by the patient; then the patient becomes increasingly convinced that the thoughts are in fact not his own; and finally, he believes that the alien thoughts are generated by someone or something else” (p. 6). Thought insertion is thus an end phenomenon that arises out of this severely disrupted experiential framework.

This experiential framework that gets disrupted goes by many names: ipseity, for-me-ness, mine-ness, the first-person givenness of experience, the minimal self, and pre-reflective self-consciousness. Dissatisfied with the literature on the subjective character of consciousness becoming bloated with such technical machinery and then trading on equivocations, Guillot (2017) has identified important conceptual distinctions between some of these terms. According to Guillot, for-me-ness is the awareness of our experiences as we live through them, me-ness is the pre-reflective awareness that it is we who live through them, and mine-ness is the awareness that the experience is our own. I make note of this only to forestall confusion, as equivocation is only a threat if the self is understood as a part of the representational content of experience instead of pre-reflectively, and if the terms are not clearly defined and their usage made explicit.Footnote 4 Having noted the potential for confusion and equivocation, let me be clear that I will use “for-me-ness” “subjectivity”, “minimal selfhood”, and “ipseity” synonymously, partly to accommodate different authors’ terminological preferences. Nothing differs except the label.

For-me-ness as I will define it has both metaphysical and phenomenological interpretations. The metaphysical (or deflationary) thesis is that experiences presuppose a subject of experience, and that there are in other words no free-floating experiences. Whenever experiences occur, they will necessarily occur to someone, their subject. The phenomenological thesis concerns the subjective feature of our phenomenally conscious experiences, in that our experiences present themselves in a distinctive manner to the subject whose experience it is (Zahavi 2005, 2014). In other words, to claim that my experiences are for-me is to say something more than the merely metaphysical fact that my experiences are in-me. Rather, our experiences also entail a pre-reflective, thin, or minimal awareness of ourselves. Consider the following illustration by Zahavi and Kriegel (2015):

Compare your experiences of perceiving an apple and remembering a banana. In one respect, these experiences are very different. They differ both with regard to their object or content and with regard to their act type or attitude. In another respect, however, the two experiences have something very fundamental in common: in both cases, it is for you that it is like something to have them. Arguably, for every possible experience that we have, each of us can say: whatever it is like for me to have the experience, it is for me that it is like that to have it. What-it-is-like-ness is properly speaking what-it-is-like-for-me-ness (p. 36).

For-me-ness has also been referred to as “minimal selfhood”. The idea is not that the self shows up as a datum of experience that one could introspect in isolation from the other contents of consciousness. Rather, this form of self-experience is extremely thin, pre-reflective, and can never become an object of consciousness. To use Zahavi and Kriegel’s (2015) terminology, it is the “how” of experience, not the “what” (or content) of experience, not an “object of experience” but a “constitutive manner of experiencing” which can be grasped by appreciating what remains constant across our changing stream of conscious experiences. While this form of self-experience is a matter of some controversy in the philosophical literature (Carruthers and Musholt 2018; Guillot 2017; Lopez-Silva 2019; Schear 2009), it has also had a profound impact on psychiatric research, with many construing the basic experiential disruption in schizophrenia as a disorder of for-me-ness or minimal selfhood (Fuchs 2015; Henriksen, Parnas, and Zahavi 2019; Nelson, Parnas, and Sass 2014; Nordgaard et al. 2021; Sass and Parnas 2003; Varga 2012).

In summary, one way that we can better understand schizophrenic thought insertion is as an increasingly felt distance between the experiencer and their experiences—construed as a disrupted for-me-ness—which is not disturbed in ordinary phenomenology. Instead of thinking about thought insertion as a condition where some features of phenomenal consciousness are present while others are lacking, construing it as an impairing or disturbing of for-me-ness allows us to retain the plausible perspective that something is deficient without oversimplification, for claiming that the subjectivity of experience remains wholly unaffected and intact underestimates how pervasive and disruptive this felt distance really is.

None of this is to say that construing thought insertion as a disturbed for-me-ness is the only plausible option available to us if we want to reject the agency-deficit explanation, so I will briefly make note of how the view compares with similar accounts. For example, Bortolotti and Broome (2009) reject appealing to a missing sense of agency, arguing instead that subjects with thought insertion lack both ownership and authorship by failing to feel committed to their thoughts, which manifests in patients failing to ascribe their thoughts to themselves and being unable to offer justifying reasons for their thoughts. However, this does not obviously account for the difference between ordinary subjects and those with thought insertion. Ordinary subjects can fail to endorse unbidden or intrusive thoughts without claiming their thoughts are inserted (see also Billon and Kriegel 2015). By contrast, construing thought insertion as the end-phenomenon of an increasingly disrupted for-me-ness does offer an account of the differing phenomenology without appealing to a total lack of ownership (where this is understood as subjectivity), in that ordinary subjects do not experience the severe self-alienation that would lead them to attribute their thoughts to others.

Zahavi and Kriegel (2015) argue that for-me-ness is preserved in thought insertion patients as there is still something it is like for them to undergo their inserted thoughts, and that thought insertion may involve an additional phenomenology of thought-alienation that is not ordinarily present, and that nothing is necessarily missing for patients with thought insertion. Similarly, in defending subjectivity theories of consciousness and arguing against the agency-deficit explanation, Billon and Kriegel (2015) argue that thought insertion might be better explained by positing additions to experience (thoughts feeling alien), and that nothing is necessarily missing. I do not deny that thought insertion might involve additions to experience (such as those described by the above authors), but I differ in being sympathetic to preserving the idea that thought insertion involves an experiential deficiency, which I think is consistent with for-me-ness being severely disrupted.Footnote 5 Of course, something being deficient is different from something being missing (one could be iron-deficient without a complete absence of iron), so the dispute here might be verbal if these authors are not using these terms synonymously. At any rate, what is deficient is the otherwise intact or undisrupted for-me-ness in ordinary phenomenology, although again this is not to say that such subjectivity of experience is entirely lacking in pathological cases.Footnote 6

An objection merits a response before concluding this section. It has been objected that appealing to a lack of for-me-ness does not explain what it is for thoughts to lack this, and simply restates the problem. For example, Langland-Hassan (2008) remarks, “Just what is this ‘my-ness’, other than a label for the problem? And why think that we can easily form a clear notion of what it would be like for it to go missing?” (p. 376). But this type of concern is the foreseeable product of failing to consider the wider clinical and psychopathological context in which thought insertion is embedded, which helps us to go beyond the prevailing binary approach involving the entire absence or presence of certain features of phenomenal consciousness. Thus, there need not be a lack of my-ness (for-me-ness) in thought insertion but a deficiency due to the disruptive felt distance that progressively worsens to the point of thoughts appearing so distanced, alien, and unfamiliar, that the subject sometimes forms delusional beliefs to make sense of their disrupted experience. What bears repeating is that there remains something it is like for them.

6 Further Research

Having supplanted the agency-deficit explanation with the disruption of subjectivity (for-me-ness) explanation, this allows us to focus more carefully on the kind of self-experience that I think schizophrenic thought insertion might vindicate, namely, experiencing ourselves as the subject of our thoughts. One of the promising ways to locate self-experience is, as Duncan quite correctly points out, to find cases where it is absent. If certain pathological cases are best explained by a lack of self-experience, this provides a plausible abductive argument for the thesis of self-experience. But given my arguments thus far about the trouble with the prevailing binary view, establishing self-experience by appealing to a lack of either the subjectivity of experience or the sense of agency will not suffice. Let us revisit the definition of self-experience one last time.

SELF-EXPERIENCE: The self shows up in experience; that is, an inner experience of one’s self has its own proprietary phenomenology that normally constitutes a distinctive component of one’s total phenomenal experience.

Notice that minimal selfhood (for-me-ness) is quite consistent with self-experience. As explained in the previous sections, for-me-ness has its own proprietary phenomenology in that it makes its own distinctive contribution to our total phenomenal experience by there being something it is like for me to live through or undergo my experiences. There are at least two different ways to cash out the “normally” part of the definition. The weaker claim is that for-me-ness manifests as an experiential constituent in some of our phenomenally conscious states, while the stronger claim is that there are no conscious states whose phenomenal character lacks for-me-ness (Zahavi and Kriegel 2015).Footnote 7 For present purposes the weaker version will do, as self-experience only requires that the self normally shows up in experience, not that it always does. Appealing to minimal self-experience being severely disrupted or disturbed, as it clearly is in cases of thought insertion, is one strategy for drawing attention to the way in which we ordinarily experience ourselves. If what is missing is missed only because it is not normally missing, then what is disrupted and disturbed is only disruptive and disturbing because it is not normally disrupted and disturbed.

One natural objection is that the move from claims about disturbed self-experience in pathological cases to positive claims about self-experience in non-pathological cases is not as straightforward as I (and many others) suggest. For example, suffering an injury can result in an experience of pain, but the normal absence of pain does not correspond to a positive experience of non-pain, there is just nothing there at all. Similarly, we might worry that a disrupted or disturbed minimal self-experience in thought insertion does not entail a corresponding positive, undisturbed and intact minimal self-experience in non-pathological cases. Instead, the apparent ubiquity of for-me-ness might simply result from the ‘refrigerator light effect’ (Schear 2009). Whenever we open the refrigerator door the light is always on, which might give the mistaken impression that the light is on even when the door is closed. Similarly, Schear argues that for-me-ness only appears to always be there in our normal phenomenology because the act of reflection is what generates it.

Admittedly, I am happy to take it as a given that the deviations found in psychopathological disorders make salient certain aspects of normal experience that we ordinarily take for granted. However, the refrigerator light objection to the ordinary pervasiveness of for-me-ness generates more problems than it solves, which our everyday experiences can attest to. Suppose I am driving a habitual route home from work on autopilot, zone out, and twenty minutes later suddenly find myself in the driveway. If someone were to ask me how the drive home from work was, I would typically be capable of offering a reasonably authoritative response. How? One answer is that when reflecting, we pick out what we were pre-reflectively living through prior to the act of reflection, and it is not clear how we can reliably reflect and report on our occurrent conscious experiences if we are not at least pre-reflectively conscious of these experiences as we are undergoing them. If one denies that the reflective response is based on there being something it is like for me to drive home, then the refrigerator light objection leaves it unclear how our subsequent reflection is targeting or illuminating our experiences, appears to attribute error to such reflective reports, and fails to capture the kind of first-person authority we typically exhibit and enjoy.Footnote 8

I think it is noteworthy that while relying on a missing a sense of agency to explain schizophrenic thought insertion is, as several authors have pointed out, more problematic than it appears, no such equivalent problem exists for the subjectivity of experience in the sense I have defined it. There may be a disruption of for-me-ness but never a dissolution, regardless of how distanced, disrupted, alienated, or otherwise anomalous the patient’s experiences are. Philosophers working on vindicating self-experience by appealing to schizophrenic thought insertion might therefore be interested not just in the end-phenomenon of thought insertion itself, but also the prodromal disruptions to minimal selfhood. Examples of thought insertion are of special utility for highlighting this “how” of experience due to their abnormally disturbing phenomenology, which can help us to deal with some of the problems with the elusiveness of self-experience by allowing us to imagine what it would be like for our self-experience to be so severely disrupted.

Indeed, some existing literature supports just such a view of self-experience. Both Varga (2012) and Fuchs (2015) argue that the first rank symptoms of schizophrenia (e.g., thought insertion) can be traced back to the anomalous self-experiences in the prodromal stages (e.g., for-me-ness). Varga (2012) further argues that this pre-reflective self-alienation is itself situated in a further disruption of the pre-reflective immersement of the self in the world, and that the latter deserves more attention than it has hitherto received. Fuchs (2015) makes an interesting distinction between ‘self’ and ‘ego’, where ‘self’ refers to the pre-reflective subjectivity (for-me-ness) of experience, and ‘ego’ to the reflective self-consciousness emerging around the second year of life that allows us to form a sense of identity, distinguish ourselves from other intentional agents, and form a sense of being the initiator of mental and bodily actions. The minimal self-experience is the foundation for the reflectively self-conscious ego, with the disturbance and diminishment of the self at the pre-reflective level leading to the kind of ego-disorders we see in thought insertion.Footnote 9

In a similar vein, Carruthers and Musholt (2018) note that while ipseity (for-me-ness) disturbances are increasingly being put forward as generative of the symptoms of schizophrenia like thought insertion, researchers use the term ambiguously and end up slipping from a pre-reflective minimal selfhood into an explicit self-representation. Again, the relevant sense of self is not to be understood as the subject being represented or appearing as an object in one’s conscious experiences (ego), rather, in pre-reflective self-consciousness the self is experienced as the subject.Footnote 10 But when ipseity is operationalized in studying schizophrenia, it is easy for researchers to slip into targeting changes in patients’ reflective self-judgements and representations (particularly about their bodies) rather than the pre-reflective, non-representational character of ipseity proper. The bottom line is that further progress on self-experience will require more careful conceptual distinctions and heightened clarity about what exactly it is that we are investigating.Footnote 11 For those of us interested in making self-experience less elusive, the advice is to be welcomed.

7 Conclusion

As Duncan (2019) notes in his concluding remarks, self-experience is ubiquitous, which makes it less noticeable and difficult to latch onto. I think part of what explains the elusiveness of self-experience is nicely captured by what I have variously called for-me-ness, subjectivity, ipseity, and minimal selfhood. The reason why Hume could not find himself when he introspected is because the very self that is searching is precisely what is being searched for. Put another way, Hume’s method of looking at his own experiences suggests he is pre-reflectively aware that they are his experiences—and this constitutive manner of experiencing is precisely (minimal) self-experience. So, there is something right and wrong about Hume’s denial. He was right in observing that the self is not an introspectable quale but mistaken in concluding that the self does not thereby show up in experience.

Granted, even this minimal form of self-experience I have advocated for is elusive, and I do not deny it. But thinking about thought insertion in its wider psychopathological context can help, for it is often the case that what we need most will be found in the places we least want to look. We can imagine what it would be like to undergo these progressively worsening, disturbing, and unsettling disruptions to this most intimate aspect of our conscious lives. Doing so can help us to appreciate how we normally experience the self.