On what matters. Personal identity as a phenomenological problem

Abstract

This paper focuses on the connection between meaning, the specific field of phenomenological philosophy, and mattering, the cornerstone of personal identity. Doing so requires that we take a stand on the scope and method of phenomenological philosophy itself. I will argue that while we can describe our lives in an “impersonal” way, such descriptions will necessarily omit what makes it the case that such lives can matter at all. This will require distinguishing between “personal” identity and “self” identity, an idea well-established in the phenomenological literature – for instance, in Husserl’s distinction between the “transcendental ego” and the person -- but I will argue that self-identity is a normative achievement whose clarification requires a move into second-person phenomenology. The argument moves through three sections. First, I will discuss Aron Gurwitsch’s “non-egological” conception of consciousness and will explain the most important reason Husserl rejected this view in his transcendental phenomenology. Second, I will discuss some contemporary approaches to Husserl’s distinction between person and ego (personal identity and self identity). Third, I will argue that these approaches testify to an ambiguity in Husserl’s account of being “true” to oneself that requires us to understand selfhood as having the structure Heidegger called care. The importance of this will be demonstrated phenomenologically in a critical examination of Paul Ricoeur’s ontology of selfhood, particularly his interpretation of the second-person phenomenology of conscience.

I borrow my title from Derek Parfit’s monumental book, On What Matters, because it raises the issue I want to focus on: the connection between “mattering” and meaning, the latter being, as I see it, the specific concern of phenomenology. While the subtitle wants to suggest that personal identity – where meaning and mattering are at stake – is a problem phenomenology can help us untangle, I will argue, in addition, that personal identity is a problem for phenomenology. It is a matter that requires us to take a stand on the scope and method of phenomenological philosophy.

What is at stake in this latter problem can be signaled by recalling Parfit’s earlier book, Reasons and Persons, in which the question of personal identity is raised in order to support an argument for a kind of consequentialism in which, as he puts it, “personal identity is not what matters.” Parfit claims that if we can “describe our lives in an impersonal way,” then the answer to the question of “what makes someone’s life go best” will not require us to consider that life as in any important sense “mine” (Parfit 1986, 217).

Here I will not directly consider the question of what makes someone’s life go best; that is, I will not be concerned with normative ethics. Instead, I want to argue a phenomenological point: while we can describe our lives in an impersonal way, this will always leave something out; indeed, it will leave out what makes it the case that such lives can matter at all. This will require distinguishing between “personal” identity and “self” identity – roughly, Paul Ricoeur’s distinction between idem-identity, equivalent to “sameness” as “permanence in time,” and ipse-identity (ipseity), which “implies no assertion concerning some unchanging core of the personality” (Ricoeur 1992, 2–3). This idea is well-established in the phenomenological literature – for instance, in Husserl’s distinction between the “transcendental ego” and the person.¹ But my argument will also require showing that self-identity (ipseity) is not mere logical identity (A = A) but a normative achievement -- grounded in the ontology of the second-person, or first-person accusative -- which makes a “personal” kind of identity possible. This thesis is not as well established in the phenomenological literature, but it is necessary if we are to address what has been called Parfit’s “Buddhism,” a “no-self” doctrine in which, as he puts it, “in some cases” the question “Am I about to die?” is “an empty question” (1986, 216). Approaching personal identity as a phenomenological problem requires that we understand death phenomenologically, not metaphysically, and this shows where Parfit goes wrong: death turns out to be a necessary condition for meaning, and so for anything “mattering” at all.²

In what follows, I can do no more than sketch the outlines of an argument that would require much augmentation to be fully convincing. That sketch will consist of three sections. First, I will discuss Aron Gurwitsch’s “non-egological” conception of consciousness, drawn from Husserl’s Logical Investigations. After his transcendental turn Husserl rejected this conception, and I will introduce what I take to be the most important reason for his change of heart. Second, I will discuss how some contemporary phenomenologists understand the transcendental Husserl’s distinction between person and ego (personal identity and self identity). Third, I will argue that ambiguities in Husserl’s account of being “true” to oneself require us to understand selfhood as having the kind of structure Heidegger called care. The importance of this will be demonstrated phenomenologically in a critical examination of Paul Ricoeur’s “ontology in view,” more particularly, his understanding of the second-person phenomenology of conscience.

1 Gurwitsch’s non-egological conception of consciousness

In approaching the question of self-identity, which he does not distinguish from personal identity, Gurwitsch follows the Husserl of the Logical Investigations, who recognized an empirical ego (an object for consciousness), but no ego that would essentially belong to every act in the stream of consciousness. Thanks to their intentional nature, acts are intrinsically unified in one stream of consciousness, and though we may call this unity an “ego” if we like, the name adds nothing. Gurwitsch, following Sartre (and much like Parfit), claims that “all these acts are impersonal in the sense that the subject in his dealing with the object, aware as he is of this dealing, is nevertheless in no way aware of his ego” (1966, 290).³ Later, of course, Husserl “changed his mind” and admitted a “transcendental ego,” a “transcendence in immanence,” but on the assumption that such an ego is supposed to “institute unity” among conscious acts without being one of them, there is, Gurwitsch argues, “no function left” for such an ego (1966, 291).⁴ Instead, what we mean when we speak of the ego phenomenologically is merely the “synthetic unity” of the person – i.e., “psychic objects” such as our “dispositions and actions,” transcendent objects that can be studied by psychology.⁵ As was the case when metaphysical questions about material “substantiality gave way to thinking in terms of functions and relations,” questions of selfhood should now be approached likewise (1966, 299).

But is it true that Husserl developed the doctrine of a “transcendental” ego in order to account for the unity among acts? That seems doubtful. Instead, the transcendental ego is introduced in order to guarantee, by way of naming it, the methodological precondition for the practice of transcendental phenomenology itself. Husserl insists that this methodological precondition is no mere formal principle established through argumentation (as is Kant’s transcendental unity of apperception); rather, it is a demonstrable (aufweisbar) structure of experience, though one whose mode of being has not been categorially clarified; it is a problem for phenomenology.

Phenomenological philosophy is not psychology; it is a critical reflection on the constitution of the meaning through which all scientific and everyday objects are given in experience. Such reflection stands under the epistemic “principle of all principles” (Husserl 1983, 44–45), which requires first-person responsibility for conforming one’s claims to phenomenological Evidenz and first-person commitment to the idea that this is how philosophy ought to be pursued. In short, “ego” here stands for the being who can take responsibility for the normative claim embedded in the principle of all principles. As we shall see, such commitment, and such responsibility, cannot be understood entirely as a function of dispositions and habits; nor is it something that is temporally extended in the sense of a “transcendent psychic object.” Before getting to that, however, let us consider two consequences of the non-egological conception of consciousness, one negative and one positive.

On the negative side, Gurwitsch, as we saw, speaks of “the subject” doing various things and being “aware” of its doing them without reflection (1966, 290). Even if we agree with him that the subject is not aware of any “ego” in such cases, this sort of self-awareness is hard to square with the idea that what the subject is aware of is some part of a “transcendent psychic object” that “supports” various dispositions and actions as “their permanent synthesis” (1966, 296, 298). That, however, seems to be the only option open to Gurwitsch, who argues that if the ego were not such a “unity of transcendent unities,” then it could “never be committed by its actions and dispositions. In truth, however, it is committed” (1966, 298).⁶ As I will argue in Section 3, however, it is wrong to say that the “subject” (a waffling term Gurwitsch uses here because he has construed the “ego” as the “person,” a constituted unity) is committed by its actions and dispositions. Commitment is what the subject is (if I may put it that way, for now), and its dispositions and actions have the meaning that they do, for the subject, thanks to that commitment. Of course, there is a sense in which my action of promising “commits” me to fulfilling that promise, but the meaning of promising would be altogether nothing to me – promising would not matter – were it not the case that I can commit my self to what such a speech-act means or entails.⁷

On the positive side, the non-egological conception of consciousness is right to reject the idea that the ego is a denizen of consciousness, a simple identity that I can “discover” when I reflect on any given conscious act or experience. By and large, contemporary Husserlians have embraced the idea that consciousness is self-unified by its temporal structure, and they have argued, further, that this self-unifying temporality also accounts for the kind of self-awareness that Gurwitsch’s descriptions of pre-reflective experience entail. Though the details are disputed, the prevailing view – endorsed by John Brough, John Drummond, Dan Zahavi, and many others, including myself – is that experiences in which objects are constituted also involve an awareness of themselves as experiences, without it being necessary that such awareness constitute those experiences as objects. That is to say, Sartre was right to hold that experience necessarily involves a “minimal” sense of self; experience cannot be ontologically “impersonal” since, in its very structure, it is “owned.”⁸ If, following the literature, we call this the “mineness” of experience, the possessive pronoun does not entail anything like ownership by a person. Any conscious being will, by the very fact of being conscious, exhibit this minimal self.

Where does this leave us in regard to the question of personal identity and self identity? One thing we can take away from it is that we should abandon talk of the “ego” – the Ich or I – locutions that raise issues of linguistic self-ascription that will not be our concern here. Instead, we should focus on the distinction between self (to begin with, the minimal self) and person. To this end, I will examine some recent accounts of this distinction that provide a rich store of phenomenological descriptions which retain their validity even after the ambiguity in Husserl’s approach to which they testify has led us beyond Husserl. This, I take it, is how a constructive phenomenological research program should proceed.

2 Formal and expressive identity: Self and person

If the minimal self is not an entity inhabiting the stream of consciousness as an absolute ego, what is it? Here John Drummond provides a compelling Husserlian answer: The self-awareness that belongs to the temporal flow of “immanent” contents is a categorial structure (“form”) of any particular flow (concretum).⁹ That is, the minimal self is what guarantees the “numerical identity” of the flow, though it does not constitute the flow as an immanent object (Drummond 2020). In this sense, “selfhood” is a formal principle of identity – not in the Kantian sense of an unexamined principle, “a thought, not an intuition” (Kant 1968, 168 [B157]), that serves to ground transcendental arguments, but as a structure of the flow that can be evidenced in “categorial intuition.”

Hanne Jacobs makes a similar point in responding to the question of what assures us that our self-identity has been preserved after the flow has been interrupted during (dreamless) sleep: because the minimal self is the form of an intentionally structured concretum, my sense of continuity is not based on “recollection” (Locke) but on the world to which I return when I awaken. Things show up as they would have had I been awake; thanks to my kinaesthetic habitualities, any radical discontinuity will show up in the world (Jacobs 2010, 341–2).¹⁰ As Drummond puts it, “self-identity is a transcendental structure” which “individuates selves but does not individuate persons” (2020). And, as Jacobs also noted, the transcendental structure of self-awareness is “embodied” in the sense of leiblich, that is, possesses a “lived” or “own” body (Drummond 2020).¹¹ Thus, even “formal” self-identity includes a dimension that exceeds consciousness understood as a flow of immanent contents. This is important, for it suggests that the minimal self must involve more structure than is contained in the phenomenological account of the temporality of the absolute flow.

But to identify this further aspect of the self’s structure we first need to consider how the “self” in this sense differs from the “person.” According to Drummond, the latter is at issue in the question Who am I; who are you; who are we? If, according to the doctrine of minimal self, it makes no sense to say “she is not herself today,” it does make sense to say that “she has become a different person than she was.” And if we can say that, then the identity of a person cannot be something fixed and absolute. As Jacobs concludes, “a person that would actually be, once and for all, identical to him or herself is [...] nonsense” (2010, 359). Phenomenologically, this means that the person is a constituted unity and personal identity, much like the identity of a perceptual object, is permanently at issue in this constitution.¹²

The person belongs to the world in the sense that it is bound up with the “experiential content” that is disclosed in the concretum whose categorial form is the minimal self (Drummond 2020). To say that the person belongs to the world is to say that to be a person requires, beyond its transcendental Leiblichkeit, the following: an empathetic reciprocity with other persons, membership in a communicatively constituted community of mutual recognition, and historicality in the form of traditions (2020). Under such conditions, personal identity is, for Husserl, not formal but expressive (Husserl 1989, 248–59). My beliefs, attitudes, and actions do not simply follow one another in time; rather, they are legible as expressions of what Drummond calls the person as “a center of conscious decision-making” (2020). In negotiating its socio-historical world, the person is a “self-responsible” agent whose decisions, grounded in its convictions concerning the true, the good, and the right, can establish themselves as sedimented habits, giving shape to a “character,” a “personal style.” Aspects of my identity as a person that do not directly result from such decisions – for instance, sexual or racial identities, what I inherit from the tradition, or even what Husserl calls “the obscure depths, a root soil” of the person (Husserl 1989, 292) – are “mine” in the way that I take them up: making them my “own” either passively (inauthentically) or actively (authentically) (Drummond 2020) – which, I should add, can also mean rejecting them altogether.

For Drummond, then, the person is defined “teleologically” as a striving for truthfulness in her cognitive, evaluative, and practical “convictions” (2020). When those commitments and convictions change significantly we can say that “Mary is a different person,” but she is still her-self. Thus, according to Drummond, the identity of “myself as person” is an identity “that depends on social constructions” and on the individual “freedom” that accomplishes their uptake (2020).

Here a significant question arises: What is meant by “freedom,” and in what sense does it belong to the person? The person can be an expressive unity only because it is free to engage the world in a norm-responsive way; meaning (the “legibility” of expression) is possible only where a standard of evaluation, internal to the one whose acts are expressive, is at issue in their striving – for instance, the standard of being rational or being truthful. Freedom just is this norm-responsiveness. But if this is so, such freedom seems to be a condition of expressive unity, not a constituted property of it. Mustn’t freedom, then, belong to the categorial form of personhood, and not to the person constituted by means of it?¹³

Something like this point also comes to light in Hanne Jacobs’ treatment of these issues. According to Husserl, the person is constituted in position-takings that are “motivated” by the way things appear meaningfully to it. Such position-takings accrue in a sedimented way; they “endure” as “features of the ego or self that is the agent responsible for all its position-takings” (Jacobs 2010, 345–6). But who or what takes such positions? Jacobs suggests that the “personal ego” is “the product of this positing” and its becoming-habitual (2010, 346). So the person seems to be an expressive unity in something like Gurwitsch’s sense: a “unity of transcendent unities,” i.e., of character, habits, dispositions, and the like. But the personal ego is my ego – “my personal habitualities only individualize me” – because they originated in my decisions as their “author” (Jacobs 2010, 347).

As with Drummond’s appeal to freedom, Jacobs here introduces the notion of the identity of authorship, and it is clear that she and Drummond draw on the same source: Husserl’s idea of an “absolute” or transcendental ego, distinct in some way from the personal ego (Jacobs 2010, 350). Terminologically, then, we should distinguish between self identity and personal identity because, as Jacobs puts it, the self that is the author of its decisions “[does] not endure like any worldly object,” nor does it “abide in the way that my habitualities do” (2010, 351).¹⁴ Hence, the self and the person can come apart, as in situations where I want to say “that wasn’t me.” In such situations I give notice that I am “unwilling to personally identify myself” with the action (2010, 350), but this non-identification is not absolute: I do identify with the self who did it; it’s just that I cannot appropriate it into my sense of myself as a person. It doesn’t express who I am. The one who authored the deed and is responsible for it nevertheless cannot identify with it or “own” it.

Jacobs draws a striking conclusion from this: authorship is (or is the capacity of?) “the formal identity, I” (2010, 350). What is striking about this is that formal self identity, as authorship, must involve far more than the minimal self, the mineness or self-awareness that belongs to any stream of consciousness.

If the self, and not just the person, has an authorial structure, what does that say about the nature of self identity in relation to personal identity? One thing it suggests is that while the notion of pre-reflective self-awareness allowed us to avoid questions about personal identity arising from the Lockean tradition, those questions now seem to take on new life, since something like a “self” has emerged to which terms like authorship, freedom, decision, and commitment are germane, terms that entail more than the minimal self.

Here we begin to sense an ambiguity that arises in the Husserlian approach. The possibility of what Jacobs calls “self-alienation” – that is, the inability to appropriate something that I nevertheless did – seems incompatible with personal identity as a constituted unity. As she sees it, in the face of such vicissitudes Husserl has two approaches to remaining “true to oneself” (that is, to “who one is,” the person), both of which turn on the self’s capacity to “decide in favor of something with insight into its unshakeable validity” (2010, 353). The two approaches are distinguished by two different senses of “validity.”

In the first approach, Husserl proposes the idea of an “ahistorical ‘true self’” defined by “deeply personal convictions whose validity is immune to vicissitudes” (Jacobs 2010, 354). That such convictions are deeply personal means that they are not necessarily intersubjectively valid; they are valid for me since they make me “me.” They are my calling or vocation. And the validity of such a calling is immune to vicissitudes because, even if I veer from it, I am still “defined” by it. It is my “deeper” or “true self,” and “the power of our conscience seems to depend on this true self being somehow still alive” even when I stray from it (2010, 254).

In the second approach, validity is understood as “universal validity,” and it is impervious to revision precisely because of its universality (ideality). My “true self” is a self “indebted to truth,” i.e., committed, in its orientation toward the future, to bringing its beliefs, evaluations, and actions into conformity with Evidenz so far as possible. Thus Husserl writes that “the ‘I’ of personal identity” is “an apriori idea of possible self-creation towards identity with oneself.”¹⁵ The important point here is that a self “indebted to truth” is oriented toward universal validity, that is, toward what anyone would or should do in one’s life, defined by an Evidenz that makes no reference to something like the “deeply personal convictions” that would define the “true self.” Thus, what “identity with oneself” means remains ambiguous.¹⁶

What we must note here is that on either approach – whether personal identity is understood as involving a vocationally abiding “true self” or as a “true future self” fully constituted in conformity with the demands of reason – there is a certain tension or ambiguity between the self who is “called,” or who “constitutes” its true future self, and the identity at issue, namely, the identity of the person. To put it another way: the formal identity of the self must be conceived either as responsive to a unique, non-universalizable call or as governed in advance by the rational norm of Evidenz. In most of his writings, Husserl seems convinced of the second disjunct; in later reflections, he moved toward the first.¹⁷ Both Drummond and Jacobs seem to prefer the second disjunct, but I will argue that the first is closer to the truth. Still, it is only closer.

This is because Husserl’s transcendental phenomenology does not have the resources to disambiguate the identity of the self at issue in the choice. If we accept the distinction between formal (categorial) self identity and personal identity,¹⁸ and if the term “true self” refers, on either disjunct, to formal self identity (either as the self called or as the self who “freely” commits to the rational norm of Evidenz), then we must admit categories, beyond the mineness of conscious experience, that clarify how the self possesses the capacities attributed to it. For instance, whether we think of the self as commitment to a calling or as commitment to rationalizing one’s choices, we presuppose a self for whom its own identity matters, and precisely in a normative way. But on Husserl’s picture this presupposition – that mattering belongs to the categorial structure of the self – has not been phenomenologically clarified.

Husserl does try to clarify it with his concept of the “transcendental person,” a hybrid notion that transfers all the features of the person (as described within the personalistic attitude) to the as-yet ontologically unclarified “transcendental ego.” As Husserl writes, “the transcendental ego as pole and substrate of its totality of potentialities is, as it were [“sozusagen”; my emphasis], the transcendental person […]” (Husserl 2002b, 200). Sophie Loidolt (forthcoming) argues that, as “the ‘concretion’ of transcendental subjectivity” in its “transcendental life,” the transcendental person is “the inscription of the transcendental into the mundane as its usually hidden dimension.” If, as Husserl argues (Husserl 1970, 186), this inscription or “mundanization” is necessary, it seems that there can be no ontological distinction between self and person. But then we have also presupposed that the sort of meaningful mattering that characterizes the person as a constituted unity in the personalistic attitude is already a capacity of the minimal self.

If the ontology of the person thus includes the categorial form of the self as its “usually hidden dimension,” then the claim that the person “is an issue for itself” must already be true of the minimal self. As we saw above, however, the minimal self belongs not just to persons but to all conscious beings. Thus we must assume that the normative salience (mattering, meaning) of personal decisions – the “striving for truth and clarity (involving procedures of achieving evidence, i.e., the fulfillment of my intentions)” –arises from “the instinct-structure of intentionality” as a modification of the minimal self’s felt conflict between an innate striving and what inhibits it in various ways (Loidolt forthcoming).

From a methodological point of view, however, this seems to get things backward. It starts from a view of what the life of an organism is – an instinctual desire to reduce complexity or to overcome resistance in the struggle for survival – and argues that such “life” can “eventually grow into a capacity for evaluative self-formation” (Loidolt forthcoming). Thus the transcendental capacities of (what I am calling) “selfhood” – the norm-responsiveness necessary for a kind of mattering in which meaning is at issue – are ultimately grounded in life itself. However, even if one agrees that Husserl’s genetic phenomenology of instinct, habitus, affectivity, and so on can contribute to interdisciplinary research in developmental biology and psychology, one might still insist that appeal to concepts like “transcendental person” and “transcendental life” do nothing to clarify the philosophical problem we are concerned with, namely, what categorial structures, beyond those of the minimal self, are necessary to account for the specifically normative being an issue for itself attributed to the person, the specific “freedom,” beyond the “I can,” that belongs to the self or ego transcendentally considered.

As noted (footnote 7), I understand such normatively structured meaning to be phenomenologically primary; on its basis, a kind of meaning and mattering that pertains to life can be “privatively” investigated, but the latter cannot help us to grasp the ontology of a self capable of such norm-responsiveness. Such an ontology will necessarily include the minimal self – consciousness and the pre-reflective self-awareness characteristic of its temporality – but mattering, in the phenomenologically primary sense, is something in addition to this. Hybrid concepts like transcendental person and transcendental life are attempts to solve the problem of what Husserl calls the “paradox of human subjectivity” (1970, 178), but his claims about the “necessary mundanization” of the transcendental depend on naturalistic assumptions and metaphysical arguments, not on phenomenological Evidenz. The relation between self and person cannot be clarified in this way; it is a problem for phenomenolgy, and only a revision in our understanding of transcendental phenomenology can remove the felt need for such solutions.¹⁹

The upshot is that the Husserlian approach, for all its demonstrable phenomenological richness, leaves us with a question: Why should I listen to a call or care about attending to Evidenz? This cannot be clarified by appeal to the minimal self of conscious experience, nor by appeal to “faculties” of this self that somehow arise in the temporal course of its life, since the term “life” here is a regional ontological category borrowed from the natural attitude and so can lend its structure to the minimal self only by begging the question. If the personal ego is to be characterized by “freedom” for rationality or by “responsiveness” to a call or vocation, then the self on which the personal ego depends must be a phenomenologically evident ontological structure – not a striving borrowed from vitalism, but a “care” for how things normatively matter. In this way things – including the person one strives to be – can show up meaningfully as what they “in truth” are, what it means for them to be.

It will come as no surprise, then, if we now turn to the account of self identity found in Heidegger’s “transcendental” phenomenology – his formal (because neutral with regard to instantiation in some worldly kind) analytic of Dasein.

3 Personal identity, self-identity, and the call of conscience

To argue for the importance of Heidegger’s phenomenological account of care (Sorge) in addressing the relation between personal identity and self identity, I will take a detour through Paul Ricoeur’s contrasting interpretation. Himself a master of detours, Ricoeur’s ontology of selfhood concludes a long and winding dialectic that traces the distinction between idem-identity (sameness, personal identity) and ipse-identity (ipseity, self-identity) through a series of knots that arise in analytic treatments of self-ascription, agency, and narrativity. In doing so, he hopes to overcome the abstract opposition between them that we noted at the outset: between idem-identity or “sameness” as “permanence in time” and ipse-identity which “implies no assertion concerning some unchanging core of the personality” (Ricoeur 1992, 2–3). So it is time to provide the argument for my earlier claim that Ricoeur’s ontological way of overcoming the supposedly abstract opposition – which, like Husserl, subordinates it to “life” – papers over what is phenomenologically distinctive about ipse-identity. In considering a turning-point in this dialectic, namely, Ricoeur’s interpretation of Heidegger on conscience, I hope to show the importance of retaining a sharp distinction between idem-identity and ipse-identity.

Having explained the extent to which the person is a “character” in the narrative sense, and having argued (as did Jacobs) that the “self” relates to this character in the form of commitment or self-ownership, Ricoeur confronts something like the normative issue that worried Husserl: does the commitment to truthfulness entail a “deeper true self” or a “true future self”? As Ricoeur puts it: “characterization of selfhood in terms of the relation of ownership […] between the person and his or her thoughts, actions, and passions […] is not without ambiguity on the ethical plane” (1992, 168). To Ricoeur, the ambiguity that attends Husserl’s phenomenology on this point suggests the ontological inadequacy of the very concept of self-ownership. As he puts it, if commitment – adopting certain norms as binding – describes self-ownership in its relation to narrative character, it seems that only “self-interest” could be of intrinsic normative concern for such a self. And if that is the picture, then Ricoeur agrees with Parfit: “ownership is not what matters” (1992, 168).

For Ricoeur, in contrast, self-ownership is dialectically bound up with “dispossession” by the Other who calls me ethically into question. Only if we make this “ethical” moment a phenomenological theme can commitment be distinguished from self-interest. But if we do so, then “the imagined nothingness of the self” in Parfit “becomes the existential ‘crisis’ of the self,” the “secret break at the very heart of commitment” (1992, 168).

To address the knots in narrative identity, then, Ricoeur situates the self within a phenomenology of the person’s “ethical aim,” and he sees commitment, the “certainty of being the author of [its] own acts and discourse,” as a form of self-attestation (1992, 180). “Attestation,” a kind of trust, is Ricoeur’s phenomenological replacement for the foundational “certainty” that, paradigmatically in Descartes and as deconstructed by Nietzsche, traditionally characterized the self-awareness at issue in ipse-identity. As anti-foundational, it is a performative concept – suspended between belief (avowal) and “the permanent threat of suspicion” – designed to addresses the particular knot in the distinction between self and person we have been tracking: how to “preserve the question ‘who’? from being replaced by the question ‘what’? or ‘why’?” (Ricoeur 1992, 22–23). For Ricoeur, this performativity exhibits various guises depending on what aspect of idem-identity (personhood) he is discussing, but here only its ontological (i.e., grounding) role – as conscience – will concern us.

The person’s ethical aim, its normative telos, is to live a “good life, with and for others, in just institutions” (1992, 172). What Ricoeur calls the “image” of such a good life allows the person to range its various practical identities and their possibly competing normative claims into a (temporary) hierarchy. Since the aim is present only as an image and not a technical rule, the ethical is not a matter of consequentialist calculation but of phronesis or “judgment.” And since Ricoeur, like Husserl, links the image of a good life with “vocation” or calling (1992, 177), judgment or phronesis is not simply a question of what I should do; in it, “the phronemos” herself is “at issue” (1992, 178). Finally, because it is always at issue, the question of what authorizes a choice of vocation cannot be answered on the basis of rational Evidenz.

Situated within the ethical aim, however, the person’s being at issue can be described as “unending interpretation applied to action and to oneself in search of adequation between our choices and what seems to us to be best with regard to our life as a whole” (1992, 179). Here commitment is displaced, at the level of ipse-identity, by the “conviction of judging well,” which is measured, as in Husserl, by the norm of adequation or truthfulness (1992, 180). However, Ricoeur is not tempted, as Husserl was, to ground this ethical aim in a metaphysical (rational) teleology, since ethical interpretation is unending and cannot shed its ground in the image of “what seems to us to be best.” Thus while Ricoeur’s account of conviction resembles Husserl’s teleological ethics, he simultaneously tries to do justice to Heidegger’s insight into the finite self’s being at issue. And because the idea that the person has an “ethical aim” is not of Heideggerian provenance (“person” is not one of Heidegger’s words), Ricoeur must revise Heidegger’s “Kantian” (i.e., transcendental) ontology in the direction of Aristotle.²⁰

The most radical revision is that, in replacing commitment with conviction measured against the norm of adequation, ipse-identity is tethered to the idea of a “higher finality” in human life. Is there something that a human life is supposed to be? (1992, 178). For Ricoeur, “life” designates “the person as a whole” (1992, 177), and so what is at issue for the person includes both its “biological rootedness” and the “vocational” unity of its character. Introducing an ethical aim into the dialectic thus entails subordinating ipse-identity to life; or, more generally, abandoning transcendental phenomenology in favor of philosophical anthropology (1992, 310, 313–14). Neither Husserl nor Heidegger could accept this subordination since, for both, life and the person are constituted unities of meaning and so presuppose, as we saw, the “formal capacities” of the self, the theme of transcendental phenomenology.

For Ricoeur – and here the vast majority of contemporary phenomenologists agree with him, though I do not – this is the price we must pay for an “ontology” of the self.²¹ Since “selfhood [ipse-identity] and sameness [idem-identity]” are “two modes of being” of the same being (1992, 309), the “being of the self,” or ipse-identity, must be approached “in conjunction with a ground starting from which the self can be said to be acting” (1992, 308). That ground, for Ricoeur, is life as conatus, “the effort to persevere in being” (1992, 315), an effort that links the biological rootedness of the person to the ramified concerns of human life. Thus ontology cannot start with the care structure, as Heidegger does, but must take a hermeneutic detour through the person. Finally, if person and self belong to the same being but are not identical to one another, there is “otherness” at the heart of this being.

Phenomenologically, this otherness shows up as passivity (1992, 318). Ricoeur identifies and explores three types of passivity that are relevant for ontology because they belong to the attestation of ipse-identity: the passivity of Leib or the “flesh,” passivity with respect to the otherness of others (empathy), and the passivity of conscience, being addressed by a call. It is the passivity of conscience that grounds the others as modes of attestation by “project[ing] after the fact its force of attestation on all the experiences of passivity placed before it” (1992, 318–19; my emphasis). Conscience, then, is what allows the various forms of passivity to matter to me in a meaningful (norm-sensitive) way.

But here we face a question: What is it about conscience that can, “after the fact,” transform conatus (life) and its attendant passivities into a world where things are encountered meaningfully, i.e., as something? Without an answer to this question, assigning an “ethical aim” to human life beyond the “effort to persevere in being” (self-interest) remains groundless. Our detour thus brings us to Heidegger, who provides a straightforward answer to the question.

For Heidegger, care – ipse-identity as the being of the self – names the formal or categorial structure of selfhood, and conscience belongs to this structure. Thus care is not grounded in the conatus of life, and its structure must account for what the minimal self could not, namely, how personhood (freedom and expression) are possible. Ricoeur recognizes his proximity to Heidegger but insists that Heidegger fails to do justice to the claims of “otherness” or passivity that belong to a phenomenology of selfhood. Heidegger might respond that all such passivity, including what Ricoeur (though not Heidegger) calls “conscience,” belongs within the category of Befindlichkeit as facticity. So the dispute between Ricoeur and Heidegger – namely, whether facticity can, in advance, be ontologically determined as “life” – must be decided on the ground of the phenomenology of conscience.

After sketching my own interpretation of Heidegger’s account of conscience, I will return to Ricoeur’s objections. We will then be in a position to understand what it means to say that the person is a normative achievement. At the same time, it will become clear that the self – the ontological structure of ipse-identity – is one as well.

As I read him, Heidegger’s account of the call of conscience is exclusively concerned with the question of how Dasein comes to occupy a “space of meaning”²² or world. Conscience is the phenomenological category that clarifies how things can matter in a meaningful way because it grounds our responsiveness to normative distinctions as normative.²³ Heidegger’s argument for this point begins in Division I of Being and Time, where he articulates the care-structure through the categories of Befindlichkeit, Verstehen, and Rede, the ground of Dasein’s “understanding of being.” Division II then opens with Heidegger’s admission that he has not yet shown how these very categories also allow Dasein to encounter itself in its own being. The sort of self-awareness that accompanies all my practical engagements as a person – which Division I terms acting “for the sake of” being something (1962, 116–17) – does not bring the ontological structure of the self into view in a phenomenologically adequate way.²⁴

For Heidegger, proper phenomenological access to Dasein’s own being is achieved in a “methodologically distinctive” (1962, 227–28) possibility of the care-structure, namely, in the breakdown of Dasein’s personal identity. This breakdown is still a way to be; that is, a unitary phenomenon (or concretum) in which the three existential categories take on a distinct shape: Angst-death-conscience. Angst is the Befindlichkeit in which the world and others do not “matter,” completely “lose significance” (1962, 231–32). The Verstehen – i.e., the ability-to-be (Seinkönnen) – that goes along with Angst is what Heidegger calls “death,” the “in-ability to be” anything (1962, 294), the collapse of all my practical identities. In breakdown, then, Dasein encounters itself as I-myself, solus ipse, without reference to “what Dasein counts for, can do, or concerns itself with in being with one another publicly” (1962, 317). But I-myself am no featureless ego; rather, being I-myself is a phenomenologically evident structure that I am “given to understand” precisely through conscience, because conscience, as the mode of Rede particular to breakdown, “articulates the intelligibility” of what I undergo.

“Guilty,” a “predicate for the ‘I am’” as such (1962, 326), is Heidegger’s term for what I am given to understand here, but such “formalized” guilt is specified without reference to any “law or ‘ought’” (1962, 328). In a passage crucial for all that follows, Heidegger explicates being-guilty (schuldig-sein: being responsible for, being the ground of) as a specific concretum of facticity (thrownness) and ability-to-be (projection): though as factic I am “delivered over” to a ground that I can never get into my “power,” as ability-to-be I “must take over being-a-ground.” (1962, 330). As formalized, this factic ground – “a ground starting from which the self can be said to be acting” (Ricoeur 1992, 308) – cannot be specified in terms of life, conatus, or anything else: “this ground is never anything but the ground for an entity whose being must take over being-a-ground” (1962, 331). Drawing on Vom Wesen des Grundes, I have argued that taking over being-a-ground can only mean taking up factic grounds, whatever they might turn out to be, in light of what is best, standing toward them as possibly justifying grounds, normative reasons (Crowell 2013, 206–213). Factic grounds thereby come to matter – that is, become meaningful in a norm-responsive way – in my acting for the sake of some ability-to-be, i.e., in my practical identity as a person. To put this in Ricoeur’s terms: factic grounds – including the three ontological passivities – are identifiable as something, as mattering meaningfully, only after the fact of being taken up.

If breakdown thus provides phenomenological Evidenz for the ontological structure of ipse-identity, and if conscience “articulates” how this structure is to be understood, such articulation does not take the form of an assertion. Instead, it is a call or “summons,” a kind of discourse that commands or enjoins: You “must take over being a ground” (Heidegger 1962, 318, 330; my emphasis). Here we enter the domain of second-person phenomenology. In relation to the call, ipse-identity appears originally as the you-accusative, and meaning is grounded in my response to the call, in the way that I take over being a ground. In Ricoeur’s terms, we could say that self identity is consciousness (minimal self) insofar as it is the addressee of a call, a you-accusative whose response to the call cannot be evaded and so takes the form of attestation.

On such a view, the formal identity of I-myself (ipse-identity) is nothing but Verantwortlichkeit (answerability), an orientation toward a measure that is always at issue. This means that the measure is never my possession, and certainly nothing whose “content” could be specified in third-person terms. And so the person – the hierarchical ordering of my practical identities and the bodily, social, ethical, and traditional determinations bound up with them – would be a normative achievement, a “constituted sense” (Husserl) governed by an “image” of what I take to be best, an image that remains at issue in how I go on. But this means that my ipse-identity is a normative achievement as well, since it consists in responding to the call. Though self identity is not a constituted sense, answerability is still a way to be, and as such it is normatively modalized: I can be I-myself either successfully (transparency with regard to my being answerable for the commitments that bind me as a person) or unsuccessfully (concealing that answerability by treating normative binding as a given fact – for instance, as inscribed in life, nature, or the “moral law”).

If the mode of being of selfhood is answerability in this sense, then we can say that while the self is not an “ego” that “abides” in the stream of conscious experiences, it does abide or possess an “identity” as the addressee of a call to which it cannot not respond. In its response it is either Selbstständig or Unselbstständig – that is, either it performatively enacts the demand placed upon it (to take over being a ground) successfully, or it fails to do so. Ipse-identity, then, cannot be thought otherwise than as the normative achievement of a categorial structure that itself is defined in terms of possibility. I cannot fail to be answerable, but my “answer” or response can fail to live up to the normative demand contained in the call. In this respect “failure” to achieve what is entailed in taking over being a ground does not mean that I am not a self at all; rather, it must be understood in the sense in which a “failed” work of art is still a work of art. Still, the idea that self identity is a normative achievement might seem phenomenologically obscure. It may become less so if we address Ricoeur’s concerns regarding Heidegger’s transcendental ontology – specifically, his interpretation of conscience.

The first thing to note here is that Ricoeur has nothing to say about the passage on which I have placed so much weight. He thus misses the phenomenological structure of the self. He does recognize that conscience is a kind of attestation, and he acknowledges that Heidegger’s view of conscience preserves ipse-identity from Nietzsche’s (or Parfit’s) elimination of it. Further, Ricoeur defends Heidegger’s “de-moralized” concept of guilt and recognizes that, as a “predicate for the I-am,” being-guilty does not fall under his own concept of ethics (1992, 349–51). Ontologically, however, Ricoeur objects to Heidegger’s idea that commitment (“resoluteness”) names the being-true of the self because, like many others, he takes this to entail empty decisionism, an “indeterminate” sense of Dasein’s “ownmost possibilities” that is also “cut off from the demands of others” (1992, 350). However, if the interpretation I offered above is correct, none of this is entailed by Heidegger’s account.²⁵ To see why, we need to follow the analysis of conscience one step further, where Ricoeur argues that the “attestation” belonging to conscience must be supplemented by an “injunction” supposedly absent in Heidegger’s account (1992, 351).

The ground of Ricoeur’s belief that Heidegger’s account is in need of a supplemental “injunction” might seem hard to grasp. As we saw, conscience, as a summons, enjoins Dasein to take over being a ground. To clear up this problem we must take a further step into second-person phenomenology, a descriptive account of what it is to be the addressee of a call.

First, Heidegger: The call comes “from me and yet from beyond me and over me.”²⁶ In his descriptions of conscience, Heidegger repeatedly emphasizes the “indeterminacy” of the caller. Ultimately, however, he alights on the claim that “in conscience Dasein calls itself” (1962, 320). For Ricoeur, this signals a missed opportunity, since Heidegger does not recognize that second-person phenomenology testifies to the “otherness” – the “beyond me” – at the heart of the self, the “self-as-other” in the form of the three ontological passivities. As Ricoeur sees it, conscience enjoins the self not only to take over being a ground but to embrace those passivities as antecedently providing an “ethical aim”: to live well (persevere in the flesh), with and for others (empathy), in just institutions (conscience as the conviction of judging well).

On my view, however, second-person phenomenology yields a different picture. Heidegger’s identification of the caller as Dasein itself is simply a mistake, since a phenomenology of the you-accusative contains Evidenz only for my being the addressee of a call.²⁷ Any identification of the addresser is a subsequent intentional act and so presupposes the self as you-accusative.

In conscience, I am already enjoined to take over being a ground, but this includes no “information” (Heidegger 1962, 319) about who or what enjoins me. Thus, when Ricoeur emblematizes commitment with the claim “Here I stand; I can do no other,” and then objects that such “conviction is not a substitute for the test of a rule” (1992, 352), he assumes that taking over being a ground has to do with a lawless decisionism that resists rule and reason. But on my view, commitment or self-binding is how we enter the space of rule and reason since, in being answerable for what I take to be normatively binding, the performatively established self is at the same time answerable to others, owes them reasons. Heidegger notes that conscience is the “attestation” of Dasein’s ontological structure, the phenomenological Evidenz that grounds an ontology of the self (1962, 312); at the same time, however – though this is not noted by Heidegger – it is the condition under which I can become attuned, and possibly committed, to a life normatively oriented by such Evidenz. But if conscience is how I enter the space of reasons, then all such reasons remain contestable. This precludes any positing of an “ethical aim” that would condition the categorial structure of ipseity in advance, since the “image” of such an aim is always only something at issue for that very ipseity.

In conclusion, let us note one further implication of the ontology of conscience: any identification of the caller of conscience will be caught up in negotiating the space of meaning that I share with others thanks to the call itself. Ricoeur comes close to acknowledging this aspect of second-person phenomenology. While he identifies the caller as the Other, his criticism of Levinas highlights the equivocalness or undecideability of this identification: is the Other the other human being or God? Perhaps, he concludes, “the ultimate equivocalness with respect to the status of the Other in the phenomenon of conscience is what needs to be preserved” (1992, 353).

Ricoeur suggests that this may mark the limit of what can be said about ipse-identity within the discourse of philosophy. If so, we would be left with the disjunct: “anthropology or theology.” But I stubbornly think that the result is best understood in transcendental-phenomenological terms. What Ricoeur calls the “humility” of attestation is the recognition that what establishes the phenomenological space of meaning is not a telos but is responsiveness to a normative injunction that always remains at issue. This is why all and only beings who can be a you-accusative can also be beings for whom, as “selves,” anything at all can matter in a meaningful, i.e., norm-responsive, way.

What matters is nothing that can be found in the third-person (impersonal) description of things; nor is it an ethical aim grounded in philosophical anthropology or theology. It is what happens when there is a being who must take over being a ground. This also means that such “happening” is not an event but the normative achievement constitutive of a self who can ever only be what it is as the “you” already addressed in the call to be answerable.²⁸