Dan Zahavi

Beyond Empathy
Phenomenological Approaches to Intersubjectivity

Drawing on the work of Scheler, Heidegger, Merleau-Ponty, Husserl and Sartre, this article presents an overview of some of the diverse approaches to intersubjectivity that can be found in the phenomenological tradition. Starting with a brief description of Schelers criticism of the argument from analogy, the article continues by showing that the phenomenological analyses of intersubjectivity involve much more than a solution to the traditional problem of other minds. Intersubjectivity doesnt merely concern concrete face-to-face encounters between individuals. It is also something that is at play in simple perception, in tool-use, in emotions, drives and different types of self-awareness. Ultimately, the phenomenologists would argue that a treatment of intersubjectivity requires a simultaneous analysis of the relationship between subjectivity and world. It is not possible simply to insert intersubjectivity somewhere within an already established ontology; rather, the three regions self , others, and world belong together; they reciprocally illuminate one another, and can only be understood in their interconnection. I: Empathy and the Problem of Other Minds The argument from analogy is a classical attempt to come to grips with the problem of intersubjectivity. One version of the argument runs as follows: The only mind I have direct access to is my own. My access to the mind of another is always mediated by his bodily behaviour. But how can the perception of another persons body provide me with information about his mind? Starting from my own mind and linking it to the way in which my body is given to me, I then pass to the others body and by noticing the analogy that exists between this body and my own body, I infer that the foreign body is probably also linked in a similar manner to a foreign mind. In my own case, screaming is often associated with pain; when I observe others screaming, I infer that it is likely that they are also feeling pain. Although this inference doesnt provide me with indubitable knowledge about others, and although it doesnt allow me to actually experience other minds, at least it gives me more reason to believe in their existence, than for denying it.1

[1] For a classical and slightly different formulation of the argument, cf. Mill (1867, pp. 2378).

Journal of Consciousness Studies, 8, No. 57, 2001, pp. 15167

152

D. ZAHAVI

This account has been met with much criticism, not only from Wittgenstein, but also from the phenomenologists. To take one example, in Max Schelers work Wesen und Formen der Sympathie (originally published in 1912) we find a whole list of counter-arguments (Scheler, 1973, pp. 2324): 1. To assume that our belief in the existence of other minds is inferential in nature is to opt for a far too complex cognitive account. After all, both animals and infants seem to share this belief, but in their case it is hardly the result of a process of inference. In order for the argument to work, there has to be a similarity between the way in which my own body is given to me, and the way in which the body of the other is given to me. But my own body as it is felt proprioceptively for me does not at all resemble the others body as it is perceived visually by me. How can the argument from analogy explain that we can empathize with creatures whose bodies in no way resemble our own, say a suffering bird or fish? Even if all of these problems could be overcome, the argument from analogy would still be formally invalid. Noticing the connection between my own mind and my bodily behaviour, and the analogy between my own bodily behaviour and the behaviour of a foreign body, all that I am entitled to infer is that the foreign body is probably also linked with my own mind.

2.

3. 4.

After this criticism, Scheler goes on to examine two of the crucial presuppositions in the argument from analogy. First, the argument assumes that my point of departure is my own consciousness. This is what is at first given to me in a quite direct and unmediated fashion, and it is this purely mental self-experience which is taken to precede and make possible the recognition of others. Second, the argument assumes that we never have direct access to another persons consciousness. We can never experience their thoughts or feelings. We can only infer that they must exist on the basis of that which is actually given to us, namely their bodily appearance. Although both of these two assumptions might seem perfectly obvious, Scheler rejects both. He denies that our initial self-acquaintance is of a purely mental nature, and that it takes place in isolation from others. As he puts it, the argument from analogy underestimates the difficulties involved in self-experience and overestimates the difficulties involved in the experience of others (Scheler, 1973, pp. 2446). But Scheler also denies that we only perceive the physical appearance of the other, and that we then, in a subsequent move, have to infer the existence of a foreign subjectivity. As he writes:
For we certainly believe ourselves to be directly acquainted with another persons joy in his laughter, with his sorrow and pain in his tears, with his shame in his blushing, with his entreaty in his outstretched hands, with his love in his look of affection, with his rage in the gnashing of his teeth, with his threats in the clenching of his fist, and with the tenor of his thoughts in the sound of his words. If anyone tells me that this is not perception, for it cannot be so, in view of the fact that a perception is simply a complex of physical sensations, and that there is certainly no sensation of another persons mind nor any stimulus from such a source, I would beg him to turn aside from such questionable theories and address himself to the phenomenological facts (Scheler, 1973, p. 254).

On the basis of considerations like these, it has been argued that we, in the face-to-face encounter, are neither confronted with a mere body, nor with a hidden

BEYOND EMPATHY

153

psyche, but with a unified whole. Thus, a solution to the problem of other minds must start with a correct understanding of the relation between mind and body. In some sense, experiences are not internal, they are not hidden in the head, but rather expressed in bodily gestures and actions. When I see a foreign face, I see it as friendly or angry, etc.; that is, the very face expresses these emotions, I dont have to infer their existence. Moreover, bodily behaviour is meaningful, it is intentional, and as such it is neither internal nor external, but rather beyond this artificial distinction. For the very same reason we should realize that the body of the other differs radically from inanimate objects, and that our perception of this body is quite unlike our ordinary perception of objects. That I have an actual experience of the other, and do not have to do with a mere inference, does not imply, however, that I can experience the other in the same way as she herself does, nor that the others consciousness is accessible to me in the same way as my own is. But this is not a problem. On the contrary, it is only because of this difference that foreign subjectivity is at all experienced as foreign. As Husserl wrote, had one had the same access to the others consciousness as to ones own, the other would have ceased being an other, and would instead have become a part ofoneself (Husserl, 1973a, p. 139). To put it another way, the self-givenness of the other is inaccessible and transcendent to me, but it is exactly this inaccessibility, this limit, which I can experience (cf. Husserl, 1973a, p. 144). And when I do have an authentic experience of another subject, I am exactly experiencing that the other eludes me. Thus, the givenness of the other is of a most peculiar kind. As Levinas puts it, the absence of the other is exactly his presence as other(Levinas, 1979, p. 89). The otherness of the other is exactly manifest in his elusiveness and inaccessibility. To demand more, to claim that I would only have a real experience of the other if I experienced her feelings or thoughts in the same way as she herself does, is nonsensical. It would imply that I would only experience an other if I experienced her in the same way that I experience myself; i.e., it would lead to an abolition of the difference between self and other, to a negation of the alterity of the other, of that which makes the other other. There is much of value in this criticism of the argument from analogy, and it can serve well as exemplification of a certain type of approach to intersubjectivity, an approach which I will call the model of empathy or the empathic approach. This approach rejects the idea that the relation between self and other is established by way of analogical inference, and instead argues for the existence of a specific mode of consciousness, called empathy, which is taken to allow us to experience and understand the feelings, desires, and beliefs of others in a more-or-less direct manner. To be more specific, empathy is typically taken to constitute a unique and irreducible form of intentionality, and one of the traditional tasks of this approach has consequently been to spell out the difference between empathy and other forms of intentionality, such as perception, imagination and recollection. This descriptive enterprise has often been associated with the phenomenological tradition, in fact the empathic approach has occasionally been taken to constitute the phenomenological approach to intersubjectivity. It is exactly this claim which I wish to contest in the following. In my view, some of the most interesting and far-reaching phenomenological analyses of intersubjectivity are all characterized by going beyond empathy. Not because they deny the existence of empathy or the validity of the criticism of the argument from analogy, but because empathy understood as a thematic encounter with a concrete

154

D. ZAHAVI

other is either taken to be a derived rather than a fundamental form of intersubjectivity because it is taken to disclose rather than establish intersubjectivity or because there are aspects of the problem of intersubjectivity which simply cannot be addressed as long as one remains narrowly focussed on empathy. II: A Priori Intersubjectivity Let me start with Heidegger, whose treatment of intersubjectivity is to be found in the context of his analysis of our being-in-the-world. More specifically, it is in connection with an analysis of our practical engagement in the surrounding world that Heidegger addresses the issue of others, for as he points out, the world we are engaged in is not a private world, but a public and communal one (Heidegger, 1979, p. 255). According to Heidegger, the type of entities we first and foremost encounter in our daily life are not natural objects such as oaks and cod, but artefacts or pieces of equipment, such as chairs, forks, shirts, soap, protractors, etc. And it is a fundamental feature of these entities that they all contain references to other persons. Be it because they are produced by others, or because the work we are trying to accomplish with them is destined for others. In short, in our daily life of care and concern we are constantly making use of entities which refer to others, and as Heidegger points out, this reference is frequently a reference to indeterminate others (Heidegger, 1979, pp. 26061). In fact, in utilizing tools or equipment Dasein (Heideggers technical term for the human being or human existence) is being-with (mitsein) others, regardless of whether or not other persons are actually present (Heidegger, 1989, p. 414). That is, Dasein does not initially exist alone, and does not first acquire its being-with the moment another turns up. On the contrary, qua its engaged being-in-the-world, Dasein is essentially social from the very start. If concrete and determinate others are absent, this according to Heidegger simply means that Daseins constitution as being-with does not attain its factual fulfilment. That is to say, one can ultimately only speak of others as lacking precisely because Dasein is fundamentally characterized by its being-with. Thus, Heidegger ultimately claims that Daseins being-with, its fundamental social nature, is the formal condition of possibility for any concrete experience of and encounter with others (Heidegger, 1986, pp. 1205). According to Heidegger, Dasein and world are internally related, and since the structure of the world contains essential references to others, Dasein cannot be understood except as inhabiting a world which it necessarily shares with others (Heidegger, 1986, p. 116). But in this case, the problem of other minds how can one (isolated) subject encounter and understand another (isolated) subject turns out to be an illusory problem. As the influential phenomenological psychiatrist Ludwig Binswanger put it:
By presenting this ontological connection, Heidegger has banished entire libraries on the problem of empathy, the problem of perceiving the foreign as such, the problem of the constitution of the foreign I, and so on, to the realm of history, for what the latter want to furnish proof of and explain is always already presupposed in the proof and the explanations; the presupposition itself can neither be explained nor proven, but rather only ontologicallyphenomenologically disclosed(Binswanger, 1953, p. 66).

The problem of other minds is exactly the kind of problem that Wittgenstein would later characterize as one of those pseudo problems which for far too long has

BEYOND EMPATHY

155

spellbound philosophers. For the very same reason, the empathic approach loses some of its attraction. Even if it does not commit the same mistakes as the argument from analogy, it does misconstrue the nature of intersubjectivity, since it takes it to be first and foremost a thematic exchange between two individuals, where one is trying to grasp the emotions or experiences of the other (this connotation is particularly clear in the German word for empathy: Einfhlung). In contrast, as Heidegger points out, the very attempt to thematically grasp the experiences of others is the exception rather than the rule. Under normal circumstances we understand each other well enough through our shared engagement in the common world, and it is only if this understanding for some reason breaks down, that something like empathy becomes relevant. But if this is so, an investigation of intersubjectivity that takes empathy as its point of departure and constant point of reference is bound to lead us astray. Heidegger is by no means the only phenomenologist to advocate this kind of approach to intersubjectivity. One finds related arguments in both Husserl and Merleau-Ponty as well. Both of these authors fully recognize that we are embedded in a living tradition and that the notion of empathy cannot account for the kind of intersubjectivity which is at play there. As Husserl put it, one has been together with others for as long as one can remember, and ones understanding and interpretation are therefore structured in accordance with the intersubjectively handed-down forms of apperception (Husserl, 1973d, p. 136). I live in a world which is permeated by references to others, and which others have already furnished with meaning, and I typically understand the world (and myself) through a traditional linguistic conventionality. Thus, already prior to Heideggers analysis in Being and Time, Husserl pointed to the fact that there, next to the tendencies originating from other persons, also exist indeterminate general demands made by custom and tradition: One judges thus, one holds the fork in such and such a way, etc. (Husserl 1952, p. 269). I learn what counts as normal from others and indeed, initially and for the most part from those closest to me, hence from those I grow up with, those who teach me, and those belonging to the most intimate sphere of my life (Husserl, 1973d, pp. 4289, 569), and I thereby participate in a communal tradition, which through a chain of generations stretches back into a dim past.
What I generate from out of myself (primally instituting) is mine. But I am a child of the times; I am a member of a we-community in the broadest sense a community that has its tradition and that for its part is connected in a novel manner with the generative subjects, the closest and the most distant ancestors. And these have influenced me: I am what I am as an heir (Husserl, 1973c, p. 223).

At the same time, however, both Husserl and Merleau-Ponty also argue for a place for intersubjectivity in the very intentional relation to the world. But instead of anchoring it exclusively in the social character of tool-use, as Heidegger tends to do, they also focus on the public nature of perceptual objects. As they put it, the subject is intentionally directed toward objects whose horizontal givenness bears witness to their openness for other subjects. My perceptual objects are not exhausted in their appearance for me; rather, each object always possesses a horizon of coexisting profiles, which although being momentarily inaccessible to me I cannot see the front and the back of a chair simultaneously could very well be perceived by other subjects. Since the perceptual object is always there for others too, whether or not such other subjects do in fact appear on the scene, the object refers to those other subjects,

156

D. ZAHAVI

and is for that very reason intrinsically intersubjective.2 It does not merely exist for me, but refers to intersubjectivity and so do my intentionality whenever I am directed at intersubjectively accessible objects (Merleau-Ponty, 1960, pp. 23, 215; Husserl, 1962a, p. 468).3 As Husserl writes:
My experience as mundane experience (that is already each of my perceptions) does not only entail others as mundane objects, but also and constantly in existential co-validity as co-subjects, as co-constituting, and both are inseparably intertwined (Ms. C 17 36a).4

As a consequence, already prior to my concrete perceptual encounter with another subject, intersubjectivity is present as co-subjectivity. This is what Husserl is referring to when he, in the still unpublished manuscript C 17, writes: When empathy occurs, is perhaps community, intersubjectivity, likewise already there, and does empathy then merely accomplish the disclosure of it?(Ms. C 17 84b).5 A question he then goes on to answer in the affirmative.6 III: The Transcendence of the Other So far we have encountered arguments to the effect that there are other types of intersubjectivity than the one addressed by the model of empathy. Intersubjectivity cannot be reduced to the concrete encounter with another subject. However, Heideggers account in particular also seems to suggest that the concrete encounter with another simply unfolds or articulates what was already there from the very start a priori, rather than adding anything new. But is this really satisfactory? According to Sartre, the answer is no. In his discussion of intersubjectivity in Being and Nothingness, Sartre at first seems to accept Heideggers observations concerning the social character of the
[2] In Being and Nothingness, Sartre nicely sums up Husserls position in the following manner: Thus,

each object, far from being constituted as for Kant, by a simple relation to the subject, appears in my concrete experience as polyvalent; it is given originally as possessing systems of reference to an indefinite plurality of consciousnesses; it is on the table, on the wall that the Other is revealed to me as that to which the object under consideration is perpetually referred as well as on the occasion of the concrete appearances of Pierre or Paul(Sartre, 1943, p. 278). [3] For a more extensive presentation of this line of argumentation cf. Zahavi (1997; 2001).
[4] I am grateful to the Director of the Husserl Archives in Leuven, Belgium, Prof. Rudolf Bernet, for per-

mitting me to consult Husserls unpublished manuscripts. Following the standard practice, I will append the original German text in those cases where I quote from the manuscripts (the translations are my own): Meine Erfahrung als Welterfahrung (also jede meiner Wahrnehmungen schon) schliet nicht nur Andere als Weltobjekte ein, sondern bestndig in seinsmiger Mitgeltung als Mitsubjekte, als Mitkonstituierende, und beides ist untrennbar verflochten.(Ms. C 17 36a). [5] Wenn Einfhlung eintritt ist etwa auch schon die Gemeinschaft, die Intersubjektivitt da und Einfhlung dann blo enthllendes Leisten?(Ms. C 17 84b). [6] Husserls discussion of intersubjectivity is very comprehensive and contains quite a number of different suggestions. Let me just mention one more, which incidentally illustrates an aspect of Husserls thought which is still virtually unknown outside the narrow circles of Husserl-scholars: Here I can only briefly point out that it is not only through social acts that such connectedness can be brought about. Just as individual subjects develop their activity on the basis of an obscure, blind passivity, so the same also holds of social activity. But passivity, the instinctual life of drives, can already bring about intersubjective connection. Thus at the lowest level, a sexual community is already established through the instinctual sexual life, even though it may only disclose its essential intersubjectivity when the instinct is fulfilled(Husserl, 1962b, p. 514).

BEYOND EMPATHY

157

equipment, for as he writes (with an emphasis that at the same time indicates a characteristic lacuna in Heideggers own account), it is undeniable that tools and artefacts contain references to a plurality of embodied others (Sartre, 1943, pp. 278, 389, 391). Just as Heidegger, Sartre consequently argues that our daily activities are intrinsically social and reveal our participation in a community of subjects, even in the absence of an encounter with concrete others.
To live in a world haunted by my fellow men is not only to be able to encounter the Other at every turn of the road, it is also to find myself engaged in a world in which instrumental-complexes can have a meaning which my free project has not first given to them (Sartre, 1943, p. 567).

But despite this congruence, Sartre nevertheless ends up directing a pointed critique at Heidegger. According to Sartre, Heideggers concept of being-with completely fails to capture our original and fundamental relation to others. There are several different steps to Sartres criticism. At first he simply points out that it would never occur to me to distinguish between a manufactured piece of equipment and a natural object unless I already had a prior experience of an other. It is exactly in and through my interaction with others that I learn to handle an object as a manufactured tool, as something that is designed for a specific purpose, as something that one uses in a particular manner. For this very reason, the reference to others contained in tool-use is a derived reference. More generally, being-with understood as a lateral relation to others is not the most fundamental type of intersubjectivity; on the contrary it presupposes a more original and quite concrete encounter with others (Sartre, 1943, pp. 4789). As Sartre writes: The we is a certain particular experience which is produced in special cases on the foundation of being-for-others in general. The being-for-others precedes and founds the being-with-others(Sartre, 1943, p. 465). Thus, in Sartres view, Heidegger made the mistake of interpreting our original relation to others as an oblique interdependence rather than as a frontal confrontation.
The empirical image which may best symbolize Heideggers intuition is not that of a conflict but rather a crew. The original relation of the Other and my consciousness is not the you and me; it is the we. Heideggers being-with is not the clear and distinct position of an individual confronting another individual; it is not knowledge. It is the mute existence in common of one member of the crew with his fellows, that existence which the rhythm of the oars or the regular movement of the coxswain will render sensible to the rowers and which will be made manifest to them by the common goal to be attained, the boat or the yacht to be overtaken, and the entire world (spectators, performance, etc.) which is profiled on the horizon (Sartre, 1943, p. 292).

In contrast, as we will see in a moment, Sartre himself takes intersubjectivity to be first and foremost a question of conflict and confrontation rather than of peaceful co-existence (Sartre, 1943, p. 481). In the second step of his criticism, Sartre takes issue with Heideggers attempt to understand being-with as an essential, intrinsic, and a priori determination of Dasein, rather than as a contingent and factual feature that only shows up in and through concrete encounters with others. According to Sartre, such a conception ignores what is most crucial in intersubjectivity, the relation to radical otherness. As Sartre points out, any theory of intersubjectivity which attempts to bridge the gap between the self and the other by emphasizing their similarity, undifferentiatedness, and a priori

158

D. ZAHAVI

interconnectedness is not only in constant danger of relapsing into a monism that in the end would be indistinguishable from solipsism, it is also losing sight of the real issue: our concrete encounter with this or that transcendent other. Sartre consequently argues that if solipsism is truly to be overcome, it will not do to neutralize the otherness of the other by positing intersubjectivity as an a priori feature of our being. On the contrary, our being-for-others must be understood as an existential dimension which only arises in and through the concrete encounter with factual others (Sartre, 1943, pp. 2935, 412).7 Sartres criticism of Heidegger, and his emphasis on the importance of understanding intersubjectivity on the basis of a concrete and thematic relation between embodied subjects does not imply, however, that Sartres approach to intersubjectivity is simply a return to the model of empathy. As we have already seen, the empathic approach is characterized by its attempt to specify the particular intentional structure of empathy, and to distinguish it from other intentional acts such as perception, recollection, and imagination. But whereas the pertinent question for this approach has been how it is possible to experience the other in a way that preserves her subjectivity and otherness, Sartre takes this line to be misguided, and instead suggests the following reversal of the traditional direction of inquiry. According to Sartre, it is crucial to distinguish between the other, which I perceive, and the other, which perceives me; that is, it is crucial to distinguish between the other as object, and the other as subject. And what is truly peculiar and exceptional about the other is not that I am experiencing a cogitatum cogitans, but that I am encountering somebody who is able to perceive and objectify me. The other is exactly the being for whom I can appear as an object. Thus, rather than focussing upon the other as a specific object of empathy, Sartre argues that foreign subjectivity is revealed to me through my awareness of myself qua being-an-object for another. It is when I experience my own objectivity (for and before a foreign subject), that I have experiential evidence for the presence of an other-as-subject (Sartre, 1943, p. 3023, 317). This line of thought is forcefully displayed in Sartres renowned analysis of shame. According to Sartre, shame is not a feeling which I could elicit on my own. It presupposes the intervention of the other, and not merely because the other is the one before whom I feel ashamed, but also and more significantly because the other is the one that constitutes that of which I am ashamed. I am ashamed not of myself qua being-for-itself, but of myself as I appear to the other. I am existing not only for myself but also for others, and this is exactly what shame undeniably reveals to me (Sartre, 1943, p. 266). Compared to Heideggers account, Sartres treatment of intersubjectivity emphasizes the transcendent, ineffable and elusive character of the other, and rejects any attempt to bridge or downplay the difference between self and other. Sartre, however, is not the only phenomenologist who has stressed the importance of recognizing the transcendence of the other. In both Levinas and Husserl we find related considerations.

[7] As a curiosity it can be mentioned that Heidegger in a letter to Sartre of October 28, 1945, wrote as fol-

lows: I am in agrrement with your critique of being-with and with your insistence on being-for-others, as well as in partial agreement with your critique of my explication of death (Towarnicki, 1993, p. 84).

BEYOND EMPATHY

159

Just like Sartre, Levinas also takes the problem of intersubjectivity to be first and foremost a problem of radical otherness, and he explicitly denies that any form of intentionality (including empathy) will ever permit us to understand this encounter. Intentionality is a process of objectivation, and it only lets us meet the other by reducing the other to something it is not, namely an object. To put it differently, although intentionality does relate me to that which is foreign, it is, in Levinas words, a non-reciprocal relationship. It never makes me leave home. On the contrary, the knowing subject acts like the famous stone of the alchemists, it transmutes everything it touches. It absorbs the foreign and different, annuls its alterity, and transforms it into the familiar and same (Levinas 1982, pp. 2123). In contrast, foreign subjectivity is exactly that which cannot be conceptualized or categorized: If one could possess, grasp, and know the other, it would not be other(Levinas, 1979, p. 83). My encounter with foreign subjectivity is consequently an encounter with an ineffable and radical exteriority. It is not conditioned by anything in my power, but has the character of a visitation, an epiphany, a revelation. One of the characteristic moves of Levinas is, then, that he takes the problem of justice and injustice to provide us with an original, non-reductionistic approach to the other. The authentic encounter with the other is not perceptual or epistemic, but ethical in nature. It is in the ethical situation where the other questions me and makes ethical demands of me, i.e., when I have to assume responsibility for the other, that he is present as other (Levinas, 1961, p. 33). Husserl has also repeatedly focussed on the transcendence of the other, but for quite different (transcendentalphilosophical) reasons than both Sartre and Levinas. As we have already seen, Husserl argues that my perceptual experience is an experience of intersubjectively accessible being; that is, being which does not exist for me alone, but for everybody. I experience objects, events and actions as public, not as private. It is against this background that Husserl introduces the concept of transcendental intersubjectivity (Husserl, 1973d, p. 110). By this, Husserl means that the objectivity of the world is constituted intersubjectively and that a clarification of this constitution calls for an examination of my experience of other subjects. Husserls thesis is that my experience of objective validity is mediated and made possible by my encounter with a transcendent other, and that this transcendence, which Husserl designates as the first real otherness and as the source of all kinds of real transcendence, endows the world with objective validity.
Here we have the only transcendence which is really worth its name, and anything else that is also called transcendent, such as the objective world, depends upon the transcendence of foreign subjectivity (Husserl 1959, p. 495). All Objectivity, in this sense, is related back constitutionally to what does not belong to the Ego proper, to the other-than-my-Egos-own in the form, someone else that is to say: the non-Ego in the form, another Ego (Husserl, 1974, p. 248).

But why is it that the other is a necessary condition of possibility for my experience of an objective world? Why is foreign subjectivity so central a condition of possibility for the constitution of objectivity? The explanation offered by Husserl is that objects cannot be reduced to being merely my intentional correlates if they are experienced by others. When I discover that the object I am currently experiencing is also perceived by an other, my relationship to the object is changed. Only insofar as I experience that others experience the same objects as myself, do I really experience these objects as objective and real. The intersubjective experienceability of the object

160

D. ZAHAVI

testifies to its real transcendence or to rephrase the point negatively: That which in principle is incapable of being experienced by others cannot be ascribed transcendence and objectivity and my experience (constitution) of this transcendence is consequently mediated by my experience of its givenness for another transcendent subject; that is, by my encounter with a foreign world-directed subject.8 It is exactly for that reason that the others transcendence is so vital. If the other were only an intentional modification or an imaginative variation of myself, the fact that he experienced the same as me would be just as conclusive, to use an example of Wittgensteins, as if one found the same report in several copies of the same newspaper. Husserl continues his analyses by describing a special kind of experience of the other, namely those situations where I experience the other as experiencing myself. This kind of original reciprocal co-existence where I take over the others objectifying apprehension of myself; that is, where my self-apprehension is mediated by the other, and where I experience myself as other; is also construed to be of decisive importance for the constitution of an objective world. When I realize that I can be an alter ego for the other just as he can be it for me, a marked change in my own constitutive significance takes place. The absolute difference between self and other disappears. The other conceives of me as an other, just as I conceive of him as a self (Husserl, 1973b, pp. 2434). As a consequence, I come to the realization that I am only one among many and that my perspective on the world is by no means privileged (Husserl, 1973d, p. 645). In short, Husserl readily acknowledges that my experiences are changed when I experience that others experience the same as I, and when I experience that I myself am experienced by others. My differentiation between the merely subjective (the hallucinatory) and the objective is made possible by concrete intersubjectivity, by my factual encounter with the other. In fact, the categories transcendence, objectivity and reality as well as the categories immanence, subjectivity and appearance are all constituted intersubjectively. Thus, it only makes sense to speak and designate something as a mere appearance, as merely subjective, when I have encountered other subjects and thereby acquired the concept of intersubjective validity (Husserl, 1973b, pp. 382, 3889). IV: Alterity in Self Whereas Sartre and Levinas chose to view the problem of intersubjectivity as a problem concerning the encounter with a radical otherness, and whereas they both argued that this encounter is so overwhelming that it is something the subject is absolutely unprepared for and has no chance of anticipating, we find the exact opposite approach in Merleau-Ponty and Husserl. Both of them are concerned with an analysis of the condition of possibility for intersubjectivity; that is, with an investigation into the question of how a relationship between subjects is at all possible. Both of them argue that my encounter with the other, my ability to interact with and recognize another embodied subject as a foreign subjectivity, is preempted by and made possible through the very structure of my own subjectivity. Both consequently refuse to take
[8] Whereas the guaranty in every single case is fallible what I took to be a valid experience of another

could turn out to be a hallucination this is not the case when it comes to the fundamental connection between intersubjective experiencability and objectivity.

BEYOND EMPATHY

161

the encounter with the concrete other as a basic and unanalyzable fact (something that Husserl, incidentally, also accuses Scheler of doing (Husserl, 1973c, p. 335)). On the contrary, the genesis and specific presuppositions of this encounter have to be clarified. We need to investigate its conditions of possibility, particularly those which concern the nature of the experiencing subject. If we start with Husserl, one of the issues explicitly emphasized by Husserl in his phenomenological account of the body is its peculiar two-sidedness. My body is given to me as an interiority, as a volitional structure, and as a dimension of sensing, but it is also given as a visually and tactually appearing exteriority. But what is the relation between that which Husserl calls the Innen- and the Aussenleiblichkeit, i.e., what is the relation between the lived bodily inwardness on the one hand, and the externality of the lived body on the other? (Husserl, 1973c, p. 337). In both cases I am confronted with my own body. But why is the visually and tactually appearing body at all experienced as the exteriority of my body? When I touch my own hand, the touched hand is not given as a mere object, since it feels the touch itself. (Had the touched hand lacked this experience, it would no longer be felt as my hand. Anybody who has tried to fall asleep with his arm as a pillow will know how strange it is to wake up with an insensible arm. When it is touched, it doesnt respond in the appropriate way and feels most of all like the arm of another.) The decisive difference between touching ones own body and everything else, be it inanimate objects or the body of others, is consequently that it implies a double-sensation. It presents us with an ambiguous setting in which the hand alternates between two roles, that of touching and that of being touched. It provides us with an experience of the dual nature of the body, since it is the very same hand which can appear in two different fashions, as alternately touched and touching. This relation between the touching and the touched is reversible, since the touching is touched, and the touched is touching, and it is exactly this reversibility that testifies that the interiority and the exteriority are different manifestations of the same (Husserl, 1973b, p. 263; 1973c, p. 75). To put it differently, my bodily self-exploration permits me to confront my own exteriority, and according to Husserl this experience is a crucial precondition for empathy (Husserl, 1973d, p. 652). It is exactly the unique subjectobject status of the body, the remarkable interplay between ipseity and alterity characterizing body-awareness that provides me with the means of recognizing other embodied subjects (Husserl, 1959, p. 62; 1973c, p. 457; 1973b, p. 263). When my left hand touches my right, or when I perceive another part of my body, I am experiencing myself in a manner that anticipates both the way in which an other would experience me and the way in which I would experience an other. This might be what Husserl is referring to when he writes that the possibility of sociality presupposes a certain intersubjectivity of the body (Husserl, 1952, p. 297). The contrast to Sartre is striking. Sartre explicitly denies that my body-awareness contains a dimension of exteriority from the very start. In his view, it is the other who teaches me to adopt an alienating attitude toward my own body. Thus, Sartre claims that the appearance of ones own body as an object is a relatively late occurrence. It presupposes a pre-reflective acquaintance with ones lived body, a consciousness of the world as a complex of instrumentality, and most significantly a perception of the body of the other. The child has used her body to explore the world and examine the

162

D. ZAHAVI

other before she starts looking at her body, and discovers its exteriority (Sartre, 1943, pp. 3856, 4089). For the same reason, Sartre attempts to belittle the significance of the touching touched reversibility. As he writes, it is a matter of empirical contingency that I can perceive and touch myself and thereby adopt the others point of view on my own body; i.e., make my own body appear to me as the body of an other. It is an anatomical peculiarity, and is neither something that can be deduced from the fact that consciousness is necessarily embodied, nor something that can serve as the basis for a general theory of the body (Sartre, 1943, pp. 351, 408). The bodys being-for-itself and the bodys being-for-others are two radically distinct and incommunicable ontological dimensions of the body.
To touch and to be touched, to feel that one is touching and to feel that one is touched these are two species of phenomena which it is useless to try to reunite by the term double sensation. In fact, they are radically distinct, and they exist on two incommunicable levels (Sartre, 1943, p. 351).

This claim must be questioned, however, not only because it seems to replace an unbridgeable dualism between mind and body with an equally unbridgeable dualism between lived body and perceived body. Rather than dealing with different dimensions or manifestations of the same body, we seem to be left with distinct bodies. And this conclusion is unacceptable, not the least because Sartres position also makes it incomprehensible how we should ever be able to recognize other embodied subjects in the first place. Exactly this criticism is voiced by Merleau-Ponty, whose own position can be seen as a continuation and radicalization of Husserls view. As he asks at one point in Phnomnologie de la Perception:
How can the word I be put into plural, how can a general idea of the I be formed, how can I speak of an I other than my own, how can I know that there are other Is, how can consciousness which, by its nature, and as self-knowledge, is in the mode of the I, be grasped in the mode of Thou, and through this, in the world of the One? (Merleau-Ponty, 1945, pp. 40001).

Merleau-Pontys answer to this question is quite unequivocal. He claims that the self-experience of subjectivity must contain a dimension of otherness. Otherwise, intersubjectivity would be impossible. Thus, in a rather Wittgensteinian move Merleau-Ponty takes self-coincidence and the relation with an other to be mutually incompatible determinations. Or to rephrase his point in a more familiar terminology: Had subjectivity been an exclusive first-person phenomenon, were it only present in the form of an immediate and unique inwardness, I would only know one case of it my own and would never get to know any other. Not only would I lack the means of ever recognizing other bodies as embodied subjects, I would also lack the ability to recognize myself in the mirror, and more generally be unable to grasp a certain intersubjectively describable body as myself.
If the sole experience of the subject is the one which I gain by coinciding with it, if the mind, by definition, eludes the outside spectator and can be recognized only from within, my cogito is necessarily unique, and cannot be shared in by another. Perhaps we can say that it is transferable to others. But then how could such a transfer ever be brought about? What spectacle can ever validly induce me to posit outside myself that mode of existence the whole significance of which demands that it be grasped from within? Unless I learn within myself to recognize the junction of the for itself and the in

BEYOND EMPATHY

163

itself, none of those mechanisms called other bodies will ever be able to come to life; unless I have an exterior others have no interior. The plurality of consciousness is impossible if I have an absolute consciousness of myself (Merleau-Ponty, 1945, pp. 4278).

But as Merleau-Ponty points out, subjectivity is not a motionless identity with itself; rather, it is essential to subjectivity to open itself to an other and to go forth from itself(Merleau-Ponty, 1945, p. 487). As he tells us, it is precisely my own experience as such that makes me open for what I am not, be it the world or the other (Merleau-Ponty, 1966, pp. 1645). Subjectivity is not hermetically sealed up within itself, remote from the world and inaccessible to the other. Rather, it is above all a relation to the world, and Merleau-Ponty accordingly writes that an openness toward others is secured the moment that I define both myself and the other as co-existing relations to the world (Merleau-Ponty, 1964, p. 114). For Merleau-Ponty, subjectivity is essentially incarnated. To exist embodied is, however, neither to exist as pure subject nor as pure object, but to exist in a way that transcends both possibilities. It does not entail losing self-awareness; on the contrary, self-awareness is intrinsically embodied self-awareness, but it does entail a loss or perhaps rather a release from transparency and purity, thereby permitting intersubjectivity.
The other can be evident to me because I am not transparent for myself, and because my subjectivity draws its body in its wake (Merleau-Ponty, 1945, p. 405).

Since intersubjectivity is in fact possible, there must exist a bridge between my self-acquaintance and my acquaintance with others; my experience of my own subjectivity must contain an anticipation of the other, must contain the seeds of alterity (Merleau-Ponty, 1945, pp. 40001, 405, 511). If I am to recognize other bodies as embodied foreign subjects, I have to be in possession of something that will allow me to do so. When I experience myself and when I experience an other, there is in fact a common denominator. In both cases I am dealing with incarnation, and one of the features of my embodied subjectivity is that it per definition comprises an exteriority. To repeat: when my left hand touches my right, or when I gaze at my left foot, I am experiencing myself, but in a way that anticipates the manner in which I would experience an other, and an other would experience me. Thus, Merleau-Ponty can describe embodied self-awareness as a presentiment of the other the other appears on the horizon of this self-experience and the experience of the other as an echo of ones own bodily constitution. The reason why I can experience others is because I am never so close to myself that the other is completely and radically foreign and inaccessible. I am always already a stranger to myself and therefore open to others. Or to put it differently, it is because I am not a pure disembodied interiority, but an incarnated being that lives outside itself, that transcends itself, that I am capable of encountering and understanding others who exist in the same way (Merleau-Ponty, 1964, p. 74; 1960, p. 215). In Phnomnologie de la Perception, Merleau-Ponty calls attention to the fact that an infant will open its mouth if I take one of its fingers between my teeth and pretend to bite it. But why does it do that? It might never have seen its own face in the mirror, and there is no immediate resemblance between its own felt, but unseen mouth, and the seen but unfelt mouth of the adult. But Merleau-Ponty suggests that the infant is able to cross the gap between the visual appearance of the others body and the proprioceptive appearance of its own body exactly because its lived body has an

164

D. ZAHAVI

outside and contains an anticipation of the other. The infant does not need to carry out any process of inference. Its body schema is characterized by a transmodal openness that immediately allows it to understand and imitate others (Merleau-Ponty, 1945, pp. 165, 40405; 1960, pp. 213, 221). Merleau-Pontys observation has recently been substantiated by a number of empirical studies concerned with infant imitation. A series of experiments conducted by Meltzoff and Moore demonstrated successful facial imitation in newborn babies (the youngest being forty-two minutes, the oldest seventy-two hours).9 But one of the crucial questions about this imitation is, how do babies know that they have a face or facial features? How do they know that the face they see is anything like the face they have? How do they know that specific configurations of that other face, as only seen, correspond to the same specific configurations in their own face as only felt, proprioceptively, and never seen? (Stern, 1985, p. 51). Meltzoff and Moore suggest that the infant has a primitive body schema that allows it to unify the visual and proprioceptive information into one common supramodal, cross-modal or amodal framework; i.e., that babies have an innate capacity to translate information received in one sensory modality into another sensory modality,10 and against this background they reach a conclusion very similar to Merleau-Pontys:
One interesting consequence of this notion of supramodality is that there is a primordial connection between self and other. The actions of other humans are seen as like the acts that can be done at birth. This innate capacity has implications for understanding people, since it suggests an intrinsic relatedness between the seen bodily acts of others and the internal states of oneself (the sensing and representation of ones own movements). A second implication of young infants possessing a representation of their own bodies is that it provides a starting point for developing objectivity about themselves. This primitive self-representation of the body may be the earliest progenitor of being able to take perspective on oneself, to treat oneself as an object of thought (Meltzoff and Moore, 1995, pp. 534).

In short, if the infant is to experience an other it has to be in possession of a type of bodily self-awareness that permits it to bridge the gap between interiority and exteriority. V: Conclusion I have only presented a few facets from a wide-ranging and ongoing discussion, but it should by now have become clear that the phenomenological tradition, far from presenting us with one single approach, contains rich but quite diverse and even competing accounts of intersubjectivity. In fact, four different phenomenological takes on the problem of intersubjectivity have crystallized in the course of this presentation:
[9] Meltzoff and Moore (1995). However, the timetable has changed drastically. Merleau-Ponty was

referring to a fifteen-month-old child, and following Wallon he believed that the child lacked the neurological capacity to perceive external objects until after the process of myelinization had occurred between the third and sixth month of life (Merleau-Ponty, 1988, p. 313). For a discussion of these issues see also Gallagher and Meltzoff (1996). [10] For a series of examples of amodal perception in children, cf. Stern (1985, pp. 4753). As Merleau-Ponty points out, the connection between the visual and tactile experience of the body is not forged gradually. I do not translate the data of touch into the language of seeing or vice-versa. There is an immediate awareness of the correspondence (1945, pp.1757, 262, 265). In Straus formulation: the manyness of the modalities is controlled by the oneness of sensory experiencing (Straus, 1958, p. 155).

BEYOND EMPATHY

165

As we have seen, one possibility is to focus on the face-to-face encounter, and to

try to account for it in terms of a specific mode of consciousness called empathy. The task then basically consists in analysing its precise structure and to spell out the difference between it and other forms of intentionality, such as perception, imagination and recollection. There is much of value in this approach as long as it does, in fact, involve a resolute showdown with the argument from analogy i.e., as long as it does not contain any reminiscences of the idea that an understanding of the other is based on projection or introjection, and as long as it does not take over any unfortunate dichotomy between mind and body. It must be emphasized, however, that this line of investigation is only able to account for one of the aspects of intersubjectivity, and that it is debatable whether this aspect is the most crucial one. In short, it remains questionable whether a theory of empathy can constitute the base and centre of a theory of intersubjectivity. Another very promising approach is to start out by acknowledging the existence of empathy, but to insist that our ability to encounter others cannot simply be taken as a brute fact, but that it is, on the contrary, conditioned by a form of alterity internal to the embodied self, for which reason the bodily interlacement of selfhood and otherness must be investigated. Insofar as the possibility of intersubjectivity is taken to be rooted in the bodily constitution of the self, we can already at this stage witness a certain reluctance to simply equating intersubjectivity with the factual and concrete face-to-face encounter. The third option goes one step further since it quite explicitly denies that intersubjectivity can be reduced to a factual encounter between two individuals. On the contrary, such a concrete encounter is taken to presuppose the existence of another more fundamental form of intersubjectivity that is rooted a priori in the very relation between subjectivity and world. By managing to disclose entirely new intersubjective dimensions dimensions which a narrow focus on empathy is bound to overlook this line of investigation has its own strength. Its main weakness is that it tends to belittle the relevance of the concrete face-to-face encounter, but thereby it also ignores the constitutive or transcendental significance of the transcendence of the other. And that is unsatisfactory. It is this failure which the fourth approach seeks to overcome. It rightly emphasizes that the confrontation with radical otherness is a crucial and non-negligible aspect of what intersubjectivity is all about. As might be expected, however, the difficulty with this view is that it often starts emphasizing the transcendence and elusiveness of the other to such an extent that it not only denies the existence of a functioning co-subjectivity, but also the a priori status of intersubjectivity. Moreover, in order to emphasize the absolute and radical alterity of the other, this approach has often denied that the encounter with the other is in some way prepared and conditioned by an alterity internal to the self, but as a result the encounter with the other is turned into a mystery (cf. Zahavi, 2000). This listing obviously does involve a certain amount of idealization. Nevertheless, most (but not all) phenomenologists have mainly focussed on one or two of the

166

D. ZAHAVI

approaches at the expense of the others. This is particularly clear in the case of such authors as Scheler, Heidegger, Sartre, and Levinas. In my own view, however, taken in isolation none of these approaches are sufficient, and a systematic synthesis of the different accounts is therefore called for. To what extent do they actually exclude each other, to what extent do they merely complement each other. Ultimately, a convincing theory of intersubjectivity must necessarily be multi-dimensional and draw on considerations taken from all of the four approaches. However, despite this diversity it is still possible to uncover certain significant and quite distinctive features that are more or less common to all of the different approaches. Let me end this contribution by briefly pointing to some of these:

Without ever denying the eminently intersubjective character of language the

phenomenologists have often endeavoured to unearth pre- or extra-linguistic forms of intersubjectivity, be it in simple perception or in tool-use, in emotions, drives, or body-awareness. This emphasis constitutes one of the decisive differences between the phenomenological approach to intersubjectivity, and the approach that we find in, for instance, Habermas, who famously argues that language is the foundation of intersubjectivity (Habermas, 1973, p. 198. Cf. Zahavi, 2001). The phenomenologists never conceive of intersubjectivity as an objectively existing structure in the world which can be described and analysed from a third-person perspective. On the contrary, intersubjectivity is a relation between subjects which must be analysed from a first-person and a second-person perspective. This is why the phenomenological point of departure is the investigation of a subject that is related to the world and to others. It is precisely such an analysis that will reveal the fundamental significance of intersubjectivity. Far from being competing alternatives, subjectivity and intersubjectivity are in fact complementing and mutually interdependent notions. Thus, the introduction of intersubjectivity should by no means be taken to imply a refutation of the philosophy of subjectivity. On the contrary, it makes a genuine understanding of such a philosophy possible for the first time. One of the quite crucial insights that we find in phenomenology is the idea that a treatment of intersubjectivity simultaneously requires an analysis of the relationship between subjectivity and world. That is, it is not possible simply to insert intersubjectivity somewhere within an already established ontology; rather, the three regions self, others, and world belong together; they reciprocally illuminate one another, and can only be understood in their interconnection. Thus, it doesnt matter which of the three one takes as a starting point, for one will still inevitably be led to the other two: the subjectivity that is related to the world only gains its full relation to itself, and to the world, in relation to the other; i.e., in intersubjectivity; intersubjectivity only exists and develops in the mutual interrelationship between subjects that are related to the world; and the world is only brought to articulation in the relation between subjects. As Merleau-Ponty would write: the subject must be seen as a worldly incarnate existence, and the world must be seen as a common field of experience, if intersubjectivity is at all to be possible. Much of what the phenomenologists have had to say on the issue of intersubjectivity is of obvious relevance not only for related discussions in analytical philosophy of mind,

BEYOND EMPATHY

167

but also for empirical disciplines such as developmental psychology and psychiatry. A further exploration of this interdisciplinary crossover would be of great value.11 References
Binswanger, L. (1953), Grundformen und Erkenntnis menschlichen Daseins (Zrich: M. Niehans). Gallagher, S. and Meltzoff, A.N. (1996), The recent sense of self and others: Merleau-Ponty and recent developmental studies, Philosophical Studies, 9/2, pp. 21133. Habermas, J. (1973), Erkenntnis und Interesse (Frankfurt am Main: Suhrkamp). Heidegger, M. (1979), Prolegomena zur Geschichte des Zeitbegriffs, GA 20 (Frankfurt am Main: Vittorio Klostermann). Heidegger, M. (1986), Sein und Zeit (Tbingen: Max Niemeyer). Heidegger, M. (1989), Die Grundprobleme der Phnomenologie (Frankfurt am Main: Vittorio Klostermann). Husserl, E. (1952), Ideen zu einer reinen Phnomenologie und phnomenologischen Philosophie II, Husserliana IV (Den Haag: M. Nijhoff). Husserl, E. (1959), Erste Philosophie II (192324), Husserliana VIII (Den Haag: M. Nijhoff). Husserl, E. (1962a), Die Krisis der europischen Wissenschaften und die transzendentale Phnomenologie, Husserliana VI (Den Haag: M. Nijhoff). Husserl, E. (1962b) Phnomenologische Psychologie, Husserliana IX (Den Haag: M. Nijhoff). Husserl, E. (1973a), Cartesianische Meditationen und Pariser Vortrge, Husserliana I (Den Haag: M. Nijhoff). Husserl, E. (1973b), Zur Phnomenologie der Intersubjektivitt I, Husserliana XIII (Den Haag: M. Nijhoff). Husserl, E. (1973c), Zur Phnomenologie der Intersubjektivitt II, Husserliana XIV (Den Haag: M. Nijhoff). Husserl, E. (1973d), Zur Phnomenologie der Intersubjektivitt III, Husserliana XV (Den Haag: M. Nijhoff). Husserl, E. (1974), Formale und Transzendentale Logik, Husserliana XVII (Den Haag: M. Nijhoff). Lvinas, E. (1961/1990), Totalit et Infini (Dordrecht: Kluwer Academic Publishers). Lvinas, E. (1979), Le Temps et LAutre (Paris: Fata Morgana). Lvinas, E. (1982/1992), De Dieu qui Vient LIde (Paris: Vrin). Meltzoff, A.N. and Moore, M.K. (1995), Infants understanding of people and things: From body imitation to folk psychology, in The Body and the Self, ed. J.L. Bermdez, A. Marcel and N. Eilan (Cambridge, MA: MIT Press), pp.4369. Merleau-Ponty, M. (1945), Phnomnologie de la Perception (Paris: ditions Gallimard). Merleau-Ponty, M. (1960), Signes (Paris: ditions Gallimard). Merleau-Ponty, M. (1964), Le Visible et LInvisible (Paris: Tel Gallimard). Merleau-Ponty, M. (1966), Sens et Non-Sens (Paris: Les ditions Nagel). Merleau-Ponty, M. (1988), Merleau-Ponty a la Sorbonne (Cynara). Mill, J.S. (1867), An Examination of Sir William Hamiltons Philosophy, 3rd ed. (London: Longmans). Sartre, J-P. (1943/1976), Ltre et le Nant (Paris: Tel Gallimard). Scheler, M. (1973), Wesen und Formen der Sympathie (Bern/Mnchen: Francke Verlag). Stern, D.N. (1985), The Interpersonal World of the Infant (New York: Basic Books). Straus, E. (1958), Aesthesiology and hallucinations, in Existence. A New Dimension in Psychiatry and Psychology, ed. R. May et al. (New York: Basic Books), pp.13969. Zahavi, D. (1997), Horizontal intentionality and transcendental intersubjectivity, Tijdschrift voor Filosofie, 59/2, pp.30421. Zahavi, D. (1999), Self-awareness and Alterity. A Phenomenological Investigation (Evanston: Northwestern University Press). Zahavi, D. (2000), Alterity in self, Arob@se: Journal des lettres & sciences humaines, 4/12, pp. 12542. Zahavi, D. (2001), Husserl and Transcendental Intersubjectivity. A Response to the LinguisticPragmatic Critique (Athens: Ohio University Press).

[11] For one attempt cf. Zahavi, 1999.