In Self-Feeling: Can Self-Consciousness be Understood as a Feeling?, Gerhard Kreuch offers a well-expressed theory of self-consciousness that largely builds on insights from the so-called Heidelberg School – including such authors as Manfred Frank, Dieter Henrich, and Ulrich Pothast – and from the theory of existential feelings – developed by Matthew Ratcliffe, Achim Stephan and Jan Slaby, among others. His central claim is that understanding self-consciousness as an affective phenomenon – namely as a self-feeling – helps to overcome problems that beset other prominent theoretical branches in this area of research, such as various higher-order representationalisms or theories of pre-reflective self-consciousness. Besides providing a justification for this claim, Kreuch also aims to develop a model that allows us to build a theoretical bridge between basic, non-conceptual self-consciousness and its more sophisticated forms, such as full-blown first-person thoughts about one’s own character traits or values. In addition, Kreuch explores how feelings in general can have a fundamental impact on our self-interpretations, and analyzes the conditions under which we would evaluate such interpretations and self-feelings as appropriate or misguided.
Before going into further detail, allow me to sketch the overall structure of the text. The book is divided into four parts. The first chapter provides a brief but helpful overview of various current theories of self-consciousness and the alleged challenges they face. The second chapter discusses human affectivity in general and introduces the reader to the theory of existential feelings. The third chapter brings the two fields together. Kreuch explains why the theory of existential feelings can be very useful in research on self-consciousness and how the features that these kinds of feelings apparently exhibit fit the description of pre-reflective (or as Kreuch also calls them, “same-order”) forms of self-consciousness identified in the first chapter. The fourth chapter seeks to provide an adequate description of the relationship between self-feelings and those first-person thoughts which are attempts to explore one’s own identity.
In Part I the book begins with a comparison of what Kreuch terms “Higher-Order and Same-Order Models of Self-Consciousness”. Higher-order models maintain that a mental state can only become conscious if it is represented by a numerically distinct state, whereas same-order models hold that there is no need for such higher-order representation. Kreuch’s terminology could be considered somewhat misleading since many of the theories to which he refers (such as those of Armstrong, Gennaro, Carruthers or Rosenthal) are not primarily concerned with self-consciousness but with consciousness. Of course, some of them – such as Kriegel’s and Williford’s self-representationalism – are also intended to provide an adequate account of what is nowadays known as the subjective character, or the mine-ness or me-ishness of experience, but the central purpose of introducing these ideas is to give an informative explanation of consciousness. Nevertheless, it seems reasonable to talk about consciousness, since self-consciousness is, after all, also a case of consciousness. Kreuch then presents reasons for preferring a same-order view of self-consciousness over a higher-order one. Central to this decision are the various regresses and vicious-circular explanations which, according to Kreuch, beset higher-order approaches. For instance, he claims that if we assume that a mental state (A) can only become conscious or self-conscious (Kreuch shifts between these two descriptions) if it is represented by a different mental state (B), then there also needs to be yet another mental state (C), which represents B to make B conscious/self-conscious, and so forth. Moreover, Kreuch maintains that genuine self-consciousness is not just a consciousness of something that happens to be oneself, but includes an additional awareness of the identity-relation holding between oneself and the object of which one is conscious (he uses Manfred Frank’s notion of De Se Constraint to denote this condition). Yet, the crucial question is this: how can a representation ensure genuine self-consciousness (or to use different terminology, fulfill the De Se Constraint)? It seems that in order to be able to recognize a representation as a representation of myself, I already need to have some prior knowledge about myself which enables me to do this. Hence, theories which attempt to explain self-consciousness in terms of self-representation presuppose what they try to explain, or so Kreuch claims.
For this reason, Kreuch sympathizes with pre-reflective theories of self-consciousness, which share his diagnosis and argue that not all cases of self-consciousness can be self-representations. However, these theories also provide a couple of other negative characterizations, such as the idea that self-consciousness is non-objectifying and non-relational. Kreuch adopts this characterization, and states that “[s]elf-consciousness is a pre-reflective, non-relational, single-digit phenomenon. There is no duality, inner perception, representation, etc., involved. Every conscious state is in itself self-intimating” (20). I am not entirely sure what “non-objectifying” or “self-intimating” mean, but I take to be the core of this approach the claim that some instances of self-consciousness have to be understood as non-relational properties rather than as relations one entertains to oneself (such as specific propositional attitudes). However, this is not a particularly rich description of these types of self-consciousness, as Kreuch rightly stresses: “Looking closer at alternative pre-reflective theories from the Heidelberg School and Zahavi/Gallagher we saw that they suffer from the ‘ex negativo’ challenge. They focus on refuting what self-consciousness is not and give little account on what it is.” (50). This is the theoretical gap Kreuch wants to close in the remainder of his book.
In Part II, Kreuch provides a general introduction to the philosophy of emotions and an account of existential feelings. He claims that one has to investigate what he calls fundamental human affectivity to come closer to an adequate theory of self-consciousness. Fundamental affective states, according to Kreuch, include moods, background feelings, and bodily feelings. He contrasts these with short-term, object-oriented emotions such as anger or fear, on which most of the contemporary philosophical literature focuses (67). While fundamental human affectivity has been discussed occasionally in the literature on emotions (e.g. in the work of Else Voigtländer, Edith Stein, Max Scheler, Bernard Waldenfels or Antonio Damasio), until recently it has never been the focal point. Kreuch maintains that Matthew Ratcliffe’s theory of existential feelings represents the most elaborated attempt to explore this dimension of affectivity and notes that he therefore relies heavily on Ratcliffe’s prior theoretical work on these issues (73ff.). Kreuch describes existential feelings as bodily emotional states which constitute “the affective background of all our experience” (74) and “shape our space of possibilities” (76). According to Ratcliffe, as Kreuch reads him, existential feelings shape our way of being in the world and are in this sense “about the world as a whole and our relationship with it […] They constitute our way of experiencing the world as a whole. Existential feelings are not about specific objects, and neither are they a medium through which something else is experienced” (80). As such, they shape “our sense of what is possible for us. Thus, they are background orientations that structure all our experience and thought.” (82). To illustrate how these existential feelings define our “space of possibilities”, Kreuch discusses a variety of cases: “In depression, the whole world feels deprived of possibilities, nothing is worth pursuing anymore. […] Similarly, a case can be made for more usual everyday situations. Imagine you had a very hard night with your baby crying for hours. You are tired and feel weak. […] Then you go to work where you are about to give an important presentation for a committee that decides on financing your long-time planned project. Under normal circumstances you would feel a bit nervous but fairly confident about it. This time, however, everything feels harder and less doable.” (84–85).
Although I find Kreuch’s idea rather intriguing, it is not entirely clear to me in what ways these feelings are supposed to define our space of possibilities. In the quotation above, for instance, Kreuch uses the words “feels deprived of possibilities” or “nothing is worth pursuing”. These are quite different things: I can believe that almost anything is possible for me while still thinking that none of these possibilities are worth pursuing, and I can believe that I am deprived of all possibilities while having the feeling that any possibility would be better than none. Do these existential feelings make me epistemically blind to certain possibilities, while presenting others as especially salient? Or do they present with a particular valence those possibilities of which I am already aware; that is to say, as possibilities that may or may not be worth pursuing? Later on, it seems as if Kreuch has in mind something along the lines of the first characterization, when he writes: “Self-feeling shapes what appears as possible action for us, it is a sense of what we can do or cannot do. For instance, when we feel vulnerable and rejected, it is unlikely that we find ourselves capable of holding a speech in front of a large audience” (137, my emphasis). However, it would be helpful to have a more precise description at some point of exactly how these feelings work and how they shape a person’s “space of possibilities”.
In Part III of his book, Kreuch seeks to develop a theory of self-feelings to overcome the problems outlined in Part I. This section can be seen as the theoretical heart of the book. Here Kreuch claims that self-feelings are not a subclass of existential feelings and existential feelings are not specific types of self-feelings, but “[i]nstead, existential feeling and self-feeling are two aspects of the one unitary phenomenon of fundamental human affectivity. Every existential feeling is always a self-feeling, too. At the same time, every self-feeling is always an existential feeling, too.” (122). As a first step, Kreuch summarizes the essential features these self-feelings are supposed to have. They are pre-reflective, non-reducible, pre-propositional, single-digit and irrelational De Se phenomena (124). As pre-propositional mental states they do not have complete propositions as their intentional objects but they always include the possibility of full articulation in one’s overall self-narration occurring in thought and speech (131). As bodily mental states, self-feelings integrate experiences of one’s own body with a feeling of the world and they also combine bodily aspects with cognitive ones. Kreuch explains: “When you touch the tip of a pencil there are two aspects involved: You feel the changes in your finger (as part of your body), your skin and tissue is deformed by the hard pencil. At the same time, you feel the pencil, you have a feeling of this object in the world. […] The situation is similar with self-feeling. In self-feeling we feel how we find ourselves in the world. It is at the same time a ‘world-feeling’, we feel our being in touch with the world.” (135). Later on, Kreuch introduces two additional features of these self-feelings. They also include a feeling of one’s own existence and one’s individuality. By virtue of entertaining them we become aware of the fact that we exist and thereby gain somehow an awareness of how we exist as this specific individual in the world. Kreuch mentions feeling healthy or welcome as examples of this how-aspect (139).
After this in-depth discussion of the essential characteristics of self-feelings, Kreuch addresses the problems he identified with higher-order (or reflective) theories of self-consciousness in Part I. He takes the following claims as the key conclusions to be drawn from Part I: (A) “mental states must be self-intimating” (i.e. “they must be conscious by virtue of themselves”); (B) “[a]ll feelings are mental states”; and (C) “all feelings are self-intimating” (150). I am not sure whether I understand Kreuch correctly here, but (A) cannot be true unless we assume that all mental states are conscious, since self-intimation is usually considered to be sufficient for that. And because not all mental states are self-intimating, the argument is not sound and provides no justification for the proposition that all feelings are self-intimating. The argument is more likely to work if we substitute “mental states” in (A) and (B) with “conscious states”, though I am not sure that all feelings are conscious (it also depends, I suppose, on one’s preferred terminology). Later on (153), Kreuch maintains that he does not necessarily have to defend (A) as it stands and that it would be sufficient for his theory if only self-feelings were self-intimating. This, however, seems to be in tension with the assumption that something like a pre-reflective self-consciousness is a universal feature of experience. Kreuch then addresses the circularity and regress problems which, he claims, beset higher-order theories of (self-)consciousness. According to Kreuch, theories which appeal to self-feelings do not entail a regress since these feelings are conscious by virtue of intrinsic properties (or as Kreuch puts it, because they are “self-intimating”) and not because they are represented by other mental states (151).
However, he rightly questions whether his theory is able to explain self-consciousness, since it seems to rely on the rather mysterious notion of “self-intimation”, which appears to presuppose some sort of self-consciousness without specifying or explaining it any further (152). Kreuch admits that he cannot completely avoid such circularity problems but that his theory, like all same-order views, is nevertheless superior in many respects to higher-order theories of consciousness and that therefore the advantages outweigh the disadvantages. First, the theory of self-feelings at least seems to avoid the problem of regress, since a conscious state does not have to be represented by numerically distinct states. Moreover, he argues that it “is plausible that it is advantageous for creatures if their most important mental states are self-intimating and do not need an additional mechanism to become self-conscious. For example, a creature that had to undergo a lengthy and exhausting process of self-reflection in order to be able to conclude that it is hungry would be disadvantaged from a creature that had the feeling of hunger right away.” (152–153). I don’t think that this is a particularly strong argument, since it is still rather unclear what consciousness is good for, and one might even hold that an unconscious hunger is better suited to trigger an immediate reaction, as consciousness might only be present during the performance of cognitively more demanding tasks (like complex decision-making or reporting one’s hunger). Finally, Kreuch claims that higher-order theories of consciousness also fail to explain why some mental states are accompanied by meta-representations while others are not, and therefore that they presuppose something they do not explain (just as his own theory of self-feelings does). Hence, even if Kreuch’s theory might contain some explanatory black boxes, this is a feature that it shares with every other currently available theory of self-consciousness (153).
In the final major section of Part III, Kreuch explores the appropriateness conditions of the self-feelings his theory posits. According to Kreuch, we cannot be mistaken in self-feelings about our own existence but these feelings can, on the basis of a number of criteria, be more or less appropriate or inappropriate (171, 194). In order to be appropriate, self-feelings may have to be balanced, stable over time (176), open to intersubjective reasoning (174), include an awareness of their own contingency (173), need to be biologically (180) or socially effective (181) and consistent, to some extent, with the network of other mental states in which they are embedded (182). However, none of these criteria alone is sufficient or necessary for an adequate evaluation of self-feelings, since this task is highly context-sensitive and in different contexts the distinct criteria are considered to be more or less relevant (183).
The final part of the book, Part IV, provides an account of how self-feelings are connected to first-person thought. It concerns the question of how these kinds of feelings inform our thinking about ourselves and influence our understanding of our own identity. Kreuch first provides some terminological clarifications which form the basis for the claim that a theory of self-interpretation should take theoretical center stage in the philosophy of self-consciousness, rather than the epistemological considerations which are usually associated with the notion of self-knowledge. Referring to the work of Cassam, Schwitzgebel and Lawlor, Kreuch suggests that research should focus more on interpretative, self-reflective activities which are pursued to “mak[e] sense of our own individual lives and create synthesized, general narratives about who we are” (204). This is also, Kreuch maintains, the theoretical field in which the concept of self-feeling can contribute valuable insights. While our self-interpretations or self-narratives are, to a certain degree, open and are unlikely to ever end, they are not based in mere imagining and therefore require a hermeneutical foundation that provides some clues as to which direction the interpretations should take. Building on Krista Lawlor’s concept of internal promptings, Kreuch proposes that self-feelings can provide such a hermeneutical base for our self-narrative thoughts (208). Later on, Kreuch writes: “[A self-feeling] offers a direct, affective experience of one’s overall being in this world. As affective resonance of one’s individual existence it can itself function as source of evidence for self-interpretation” (215). Moreover, self-feelings do not only shape our space of possible actions, as Kreuch has argued in Part II, they also determine the way we think about ourselves: “Thus, given a particular self-feeling there are thoughts that are possible and others that are not. For example, in the case of severe depression our self-related thoughts are predominantly concerned with our insignificance and disability. We believe that we are worthless and unable to live a normal life. Thoughts like ‘Things will become better for me again’ or ‘I have strengths and skills, too’ are likely to be impossible in severe depression” (211). Hence, self-feelings guide, limit, and provide evidence for our self-interpretations.
The next major section of Part IV analyzes how the evaluation of self-feelings and self-interpretations are related to one another (217). For this purpose, Kreuch discusses four different evaluative combinations of these mental states and how we might interpret them (220). According to Kreuch, there are: 1) inappropriate self-feelings with fitting self-interpretations (e.g. depression); 2) appropriate self-feelings with non-corresponding self-interpretations; 3) inappropriate self-feelings with misguided interpretations (as in narcissistic personality disorder); and 4) appropriate self-feelings and adequate interpretations (which Kreuch calls “authenticity”). A self-interpretation is only fully appropriate (or authentic) if it somehow captures the nature of the self-feeling on the basis of which it was formed. In addition, the underlying self-feeling has to be appropriate, too. If I feel a deep sadness for no good reason and understand myself as a melancholic person, both the feeling and the interpretation seem to be misguided, even if the interpretation resembles the feeling. Or, as Kreuch ultimately puts it: “There cannot be appropriateness on the higher levels if something is wrong in the foundation” (251).
Overall, Kreuch’s Self-Feeling is an important contribution to current debates on self-consciousness and addresses pressing issues which are rarely discussed within those debates. The author succeeds in demonstrating that it is worthwhile to investigate the neglected relations holding between self-consciousness and emotions more deeply. The book provides a novel account in a tremendously complex area of research and generally presents its ideas very clearly. While Self-Feeling covers a wide range of topics and appeals to different analytical and continental traditions, one never feels lost since the author is an excellent guide throughout the book. Yet, the broad scope of the book obviously comes at the cost of more detailed discussion of particular arguments, especially those which are presented as motivations for Kreuch’s own theory. Let me offer a few examples and thereby end this review on a constructively critical note.
The regress argument, which Kreuch presents as a problem for higher-order theories of consciousness, is highly debated in the literature. The crucial claim that needs justification or has to be accepted by higher-order representationalists to get the regress going is as follows: only conscious mental states can make us conscious of the things they represent. This, however, is an assumption that most of these theorists will not accept. David Rosenthal, for instance, writes: “Nor must a HOT [Higher-Order Thought] itself be conscious to make us conscious of its target. We’re conscious of things even when we see them subliminally, though we’re not in such cases aware that we’re conscious of those things. Similarly with HOTs and their mental targets. We’re seldom aware of having HOTs, which is what we should expect” (Rosenthal 2005: 9). And even philosophers who explicitly argue against higher-order models, such as Rowlands (2001), agree that at least some types of unconscious mental states, such as perceptions, can make us conscious of things they represent. Thus, in order to provide a comprehensive justification for a critique along these lines, Kreuch would need to put a lot more work into the argument.
Similar issues apply to the circularity critique Kreuch raises against higher-order theories of consciousness (or reflective theories of self-consciousness). Kreuch writes: “The challenge is then about how should self-consciousness emerge out of a set of completely unconscious states? The only explanation seems to be that self-consciousness was implicitly already there. Otherwise it remains dark how the combination of two unconscious states should make for self-consciousness” (15). I am not sure why there is supposed to be a circular explanation here. Higher-order theories of consciousness maintain that a mental state becomes conscious in virtue of being adequately represented. The meta-representations themselves, they claim, are most often unconscious. Hence, according to these types of theories, the only thing needed for consciousness to emerge is meta-representation. And we might add that quite often when two things stand in appropriate relations (such as a representation-relation), they can gain properties (such as being conscious) that they would not have on their own. Therefore, higher-order theories at least prima facie seem to offer a non-circular explanation of (self-)consciousness and therefore more detailed discussion is required in order for this criticism to work.
Yet, as I explained earlier, Kreuch identifies another circle. According to him, any theory which understands self-consciousness merely in terms of representation presupposes other (unexplained) forms of self-consciousness, because a subject would already have to know something about itself to recognize itself as the object of representation. Kreuch concludes, therefore, that some cases of self-consciousness cannot be representations, but must be understood as non-relational properties or, as he puts it, as “single-digit phenomena”. It is not absolutely clear, I think, whether one really has to know something about oneself to entertain genuine De Se representations (or, as people say, “representations about oneself as oneself”), since we might also be able to have proper De Re representations without having any knowledge about the objects in question. And I am also not sure that we have to accept completely non-relational forms of self-consciousness in order to account for De Se representations, since even those non-representational instances of self-consciousness are supposed to be about me and my mental states and, hence, appear to exhibit some kind of directedness or intentionality. We could simply introduce some form of non-representational self-acquaintance instead. I believe that Kreuch is correct in pressing the question of how genuine De Se (or first-person) representation is possible and that this should be the focus of a theory of self-consciousness. However, it would be useful to be provided with a deeper understanding of exactly how Kreuch’s concept of a single-digit self-feeling can help answer this question.
Rosenthal, David. 2005. Consciousness and Mind. Oxford: Oxford University Press.
Rowlands, Mark. 2001. “Consciousness and Higher-Order Thoughts”. Mind & Language. Vol. 16 (3): 290–310.
 As stated earlier, Kreuch seems to use these names interchangeably. However, it is not clear whether these notions usually denote the same types of theories in the literature, as a theory of self-consciousness may not be expected to give any hints about the conditions of consciousness in general.