2. The Book and its Title, Friendship and its Failures

The narrator begins his story with a relatively detailed memory of his and Paul’s parallel hospitalizations (his for lung disease, Paul’s for ‘madness’) in the contiguous pulmonary and psychiatric wards on Vienna’s Wilhelminenberg. Though it was during this stay in 1967 that the two “sealed” their friendship (56), as the narrator emphasizes, their relationship extends further back. With the apparent aim of bringing their life together into better focus, the narrator sets out from this lengthier opening scene, sometimes by association, often without transition, to fragments of older and newer memories, some pertaining to his shared experiences with Paul, others to events to which Paul was marginal, and yet others where he is absent or seems irrelevant. These memories sometimes give way to isolated anecdotes about Paul in his life apart from the narrator and to the narrator’s often vitriolic musings. Such non-narrative reflections ultimately eclipse the episodic and anecdotal passages. By the end, the narrator leaves us with little more than his melancholic doubts about that friendship with Paul, and the sense that he has, all along, been trying to establish the meaningfulness of the “scraps of memories” (19) with which his now uncertain friendship has left him.

This rather unusual structure of the book finds expression in its title. The first part thereof, “Wittgenstein’s Nephew,” simultaneously designates the narrator’s friend Paul (without naming him); recalls the famous philosopher; ^{Footnote 14} and alludes to Denis Diderot’s Rameau’s Nephew. That book, which, like Bernhard’s, has no fixed genre, identifies a related problem of friendship and self-definition. ^{Footnote 15} The purpose of the second part of the title, “A Friendship,” is less clear. Bernhard scholarship has called it a “subtitle.” ^{Footnote 16} Yet it is German publishing convention to include a generic designation on title pages and (typically) on covers. More notoriously, it was also Bernhard’s own practice, beginning with the first volume of his autobiography and continuing to his final work, to employ ambiguous secondary titles, especially for those works—such as Woodcutters: An Irritation (1985) and Extinction: A Decline (1986)—whose genres, because they sit somewhere between fictional and factual narrative, are his own. Bernhard appears to be giving idiosyncratic names to idiosyncratic forms of writing. We have good reason to suspect, then, that the second part, “A Friendship,” designates, like ‘novel,’ the genre, what the book is, and not the contents, what the book is about.

But an argument from German and Bernhardian convention will not alone support the argumentative weight we put on the second part of the title. Below, we argue more thoroughly against the view that this second part simply describes the content of the book. Indeed, as Wittgenstein’s Nephew progresses, we come to have increasing difficulties taking what the narrator calls his disconnected “notes” (19, 82, 98) as simply a description of a friendship.

At first glance, the narrator’s relationship with Paul appears to conform remarkably well to Aristotle’s ideal and by consequence the conventional notion of ‘best friends.’ This notion involves joint activity oriented by moral and intellectual virtue, practiced by people who so share in sensibility and understanding as to be, in some respects, the other’s self, and with the triple end of pleasure, self-improvement, and the advancement of the other’s good. Especially in the book’s first half, the narrator claims to have gained a partner with whom he shares in aesthetic experience; through whom he expands his consciousness about what he cares most for; and thanks to whom he might understand himself and his life better (see, e.g., 22). He and Paul would spend untold hours critiquing society at cafés, listening to music, and driving around the countryside, and in so doing actualize and intensify this especially authentic form of togetherness. As the narrator muses, even those “who have meant everything and been everything to us at certain critical moments or certain critical periods of our lives” we hesitate to call our friends—at least until they have died, for we fear “being totally, embarrassingly, and ludicrously wrong.” With Paul, by contrast, the narrator proclaims, he “had no such inhibition” (81). Paul was, while still alive, not just an “extraordinary friend” (17), but in fact “a real friend” (22).

And yet signs that the narrator’s relationship with Paul might not be so clearly an extraordinary and authentic friendship appear not far into the book. The narrator soon mentions, if obliquely, other relationships that were apparently just as significant. Paul, for instance, had a long and loving second marriage with a woman, Edith, whose death may have precipitated his own demise (34, 40, 49–50, 77). The narrator has his own ultimate relationship, with his “life support” (Lebensmensch, 224), “the one person who has meant more to me than any since the death of my grandfather, the woman who shares my life and to whom I have owed not just a great deal but, frankly, more or less everything” (18–19). ^{Footnote 17} These relationships do not in themselves cast shadows on the relationship with Paul; but they do hint that the centrality of Paul to the narrator’s life may come less from his relative objective importance than from the perspective the narrator is striving to maintain. Further complicating this picture is that the narrator admits that he and Paul needed “repeated proofs of friendship” (65) and that their relationship was “naturally difficult”: “It was the kind of friendship that has to be daily renewed and re-won, and in the course of time proved exceedingly strenuous” (64). This strain comes to a head on the Wilhelminenberg. At precisely the moment when their friendship was supposedly deepened, the narrator reacts to Paul’s fits of affection with actual fear (32–33).

Furthermore, the narrator’s relationship with Paul, if indeed a friendship, reveals itself to be a highly unconventional one. The picture that emerges is one of at best intermittent companionship. The narrator lingers on episodes in which Paul is tangential. The long passages relating the narrator’s prize receptions (65–73) and one particular theatre production (93–98) are presented explicitly as evidence of Paul’s support, understanding, and truthfulness as a friend (65, 97), and yet Paul himself plays almost no role in them. Moreover, we learn that the narrator was hardly ever even invited to Paul’s apartment, despite its being in the same building as a café they frequented. The friend’s absence even characterizes their moments together. One of the narrator’s fondest memories is of the evenings they spent listening to music, “without saying a word” and in separate rooms of the house (34). ^{Footnote 18}

The peculiar and repeated separation of the friends is most evident in the opening passages of the book detailing their shared convalescence, where nearly every attempt to overcome the (small) distance between them is thwarted. Amid lengthy diatribes against doctors and psychiatrists, the narrator describes his attempts and failures to meet up with his friend. Although Paul does at one point show up at his bedside and agrees that they should pay each other regular visits, this “plan was put into effect only once,” namely, when the two meet halfway between their respective wards, “on a bench just inside the chest patients’ territory” (45). This meeting on the bench, however, is an abject failure: Paul, unable to stop weeping, repeats “Grotesque, grotesque!” and the narrator has to bring him back to his ward.

The bench scene is echoed later in the book when, roles reversed, the narrator breaks down in the face of his dying friend, fleeing Paul’s apartment—“I suddenly could not stand being with him any longer” (78)—and taking refuge “on a bench in the City Park” (79). ^{Footnote 19} If the first bench episode “had set a permanent seal [immer besiegelt, 262] on [their] friendship” (56) but also showed its frailty, this second bench episode appears to dissolve it, but also to occasion for it a renewed understanding. This understanding comes in two stages. The narrator explains his response as a fear of death; and he understands Paul’s and his own breakdowns as defence mechanisms triggered by an inability to confront one another’s—and thus their own—dying. “I no longer visited [Paul] because I was afraid of a direct confrontation with death. I constantly repressed this thought and finally expelled it from my mind” (98). But it is precisely by way of this realization that the narrator comes to the most damning assessment of his relationship with Paul: he only used his friend to draw the strength he himself needed to survive. “I had used Paul’s dying for my own advantage, exploiting it for all I was worth” (99). These worries of exploitation pose the most serious threat to a reading of the book that takes it to depict a successful, even true, friendship. It would thus be strained to call “A Friendship” a descriptive subtitle. More reason, then, to take it as a genre-designation. But what would this mean, and why would Bernhard create and name such a genre?

3. Narrating a Life Together

At face value, Wittgenstein’s Nephew is the narrator’s—and even Bernhard’s—attempt to memorialize his friend Paul, and further, to satisfy Paul’s request that the narrator hold the eulogy at his funeral: “Two hundred friends will come to my funeral and you must make a speech at the graveside.” This request frames the narrative, appearing both as epigraph and in the final lines, where the narrator admits that he failed to make the funeral (100); in fact, he says, he has not even been to the grave. ^{Footnote 20} Though he never explicitly claims to feel guilty, some readers have thought that in writing this book Bernhard (as actual friend of Paul Wittgenstein) must have recognized his failing, and used the book to honour, however belatedly, his friend’s request. ^{Footnote 21} While not implausible, this view conflates the narrator and the author, and worse, ignores the complexities we have shown in the narrative of the friendship.

Wittgenstein’s Nephew demonstrates how the effort to narrate a life together can constitute the friendship itself. The narrator claims that he and Paul became real friends during their shared stay at opposite ends of the Wilhelminenberg. Yet, despite their proximity, they saw each other only twice. What, then, could have caused the deepening of their friendship? Perhaps the natural desire for nearness while in recuperation made Paul’s company so appealing: “I longed for nothing more intensely than to see Paul again, having for so many months been deprived of all contact with his mind and having almost suffocated among hundreds of other minds” (27). But it may not have been the thirst for communion with this sympathetic man that deepened their friendship, but what it led the narrator, a writer after all, to do: “The day will come when I really will visit Paul, I thought, and I made some notes about things that I intended to discuss with him … At that time I found it quite simply impossible, without Paul, to have any conversation about music or philosophy or politics or mathematics” (27). He wrote himself, albeit sketchily, into the matter of a friendship. Not all of his reflections were positive: “I suddenly hesitated, wondering whether it was really wise to establish a link …, whether it might not do both of us more harm than good. … I was suddenly scared of seeing him … The possibility of Paul’s coming to visit me became a nightmare” (31–32). In both cases, the narrator places himself into conversational activity with Paul; he can do so only because he recognizes the importance of their earlier conversations together. It was not some affective intensity or fresh discovery of shared interests that deepened the friendship; it was the narrator’s making a meaningful proto-literary reconstruction of their past together.

Bernhard had wanted since the mid-1960s to write about Paul Wittgenstein. ^{Footnote 22} The narrator indicates that he began taking notes about Paul around 1967 (98). While these notes referred mostly to “music and crime, to the Hermann Pavilion and the Ludwig Pavilion [the friends’ two wards in the hospital] and the intense relations between the two” pavilions, they also recorded Paul’s having “had some remarkable things to say about politics, about wealth and poverty, all on the basis of his own experiences, the experience of one of the most sensitive men I have known” (99). Indeed, the narrator has an even larger treasury of remembered stories, “thousands of Paul’s anecdotes in which he is the central figure” (60), which he has been hoping, for the duration of the friendship, to use as a basis for writing about Paul and his relationship with him. Here we see most plainly the connection between friendship and narration. Fitting these stories together gives meaning to the men’s shared experiences.

But it is precisely in wanting to write about Paul that the narrator comes to worry that his literary impulse has taken the place of proper affection and attachment—the traditional criteria of friendship. Just as he “had traced [in notes] his dying over a period of more than twelve years,” the narrator claims he “had used Paul’s dying for [his] own advantage, exploiting it for all [he] was worth” (99). On the one hand, then, narrating his shared life with Paul helped constitute the friendship; on the other hand, it risked separating the two. And yet, coming at the book’s end, the narrator’s recognition of this separation—and thus his recognition of the hazard of giving (narrative) form to his friend’s life while trying not to forget what matters, the friend himself—blends into his most human description of his friendship: Paul’s request that the narrator speak at his funeral, and the narrator’s admission not to have yet visited Paul’s grave. The narrator’s fear about transmuting Paul into literary material, and thereby exploiting him, threatened to hinder him from writing about his life together with Paul. Yet it is just such writing that constitutes the friendship and vitiates his worries that their relationship was simply one of selfish exploitation.

The story concludes, as we have just said, with the narrator’s admission that he failed his friend’s final request to “make a speech at the graveside.” Since we naturally imagine this speech to be a eulogy, an accounting of the friend’s life, we are encouraged to think that the narrator’s reflections in Wittgenstein’s Nephew constitute that speech. Were this the case, the book would allow the narrator finally to discharge his obligation, in a belated act of friendship. But the reflections this 100-page book comprise would constitute an absurd graveside speech. ^{Footnote 23} This is because it is not about Paul himself so much as about the narrator’s attempt to articulate the meaningfulness of his relationship with Paul, which necessarily coincides with reconstructing the course of that relationship. Rather than take Wittgenstein’s Nephew as itself a eulogy, we should see it as focusing our attention on the nature and practices of reflecting narratively on a life. In one episode, the narrator, having just expostulated on the importance of his friendship with Paul, says that he persuaded Paul to write his autobiography. Paul modified this advice, saying he intended to write “a thoroughly philosophical account of his life, not just a rigmarole [keine Schwafelei, 265], as he put it.” (As it turns out, he never wrote more than a few pages (59).) ^{Footnote 24} A little later, as the narrator recounts Paul’s waning days, he digresses to his own award speeches, which themselves turn out to be philosophical reflections on life (65–72). Both attempts at recounting one’s life make thematic the importance we accord to telling a life for making one’s life meaningful as a life. The narrator’s attempt to recount two lives together is thus philosophical about friendship; but it is also constitutive of that friendship, making it meaningful, which is to say, a friendship at all.

4. The Philosophy of Friendship

Many of those who write in the philosophy of friendship seek out friendship’s necessary and sufficient conditions, its definition, or its central and penumbral instances. They take as unsatisfying, even if phenomenologically plausible, Michel de Montaigne’s ‘because it was he, because it was I.’ Such a view seems to make friendship irrational, unexplained, and irresponsive to what we find most important about people. But they also take as unsatisfying those definitions that found friendship on the more explicable traits or essential humanity of the friend. These views risk treating the friend merely as a (replaceable) vehicle for either desirable traits or a respect-worthy core. ^{Footnote 25} An appealing solution to this apparent dichotomy comes in a recent paper, Benjamin Bagley’s “Loving Someone in Particular.” Inspired by the models of Wuthering Heights and jazz improvisation, Bagley argues that a friendship gains both significance and particularity as a mutually dependent exercise in self-constitution, the concerted search for life-guiding norms. He observes that only with certain people can I develop and receive help in developing myself; these are the ones I call my friends. The more I work on myself with another, and thus the more we share in life-guiding norms, the more irreplaceable, and explicable, my friend becomes to me. We begin as incompletely self-knowing individuals, but girded by a sense of promise we stick together, and end up finding ourselves through that friendly other. In this analysis, Bagley does not, somewhat surprisingly, cite Aristotle; but in an important way his work realizes and improves on some key strands of Aristotelian theory of friendship. It is worth setting out these Aristotelian strands. All the same, Bagley’s work does not make certain elements explicit enough, especially the importance of recognizing a friendship as a friendship, and the route by which this may be achieved. Nevertheless, his account allows for, even presupposes, the one we have presented here; we take that fact as useful corroboration.

Aristotle argues that friendship depends on mutually acknowledged goodwill (Nicomachean Ethics 8.2). This means that a friend takes his friend’s good as his own, and finds his own good to be shared by his friend. This reciprocal affection and valuing might appear to demand no actual proximity or performance. But Aristotle notes that inactive friends—those living apart, not communicating, simply holding the other in their good wishes but not doing anything to advance their common interests together—drift away (8.2 1157b12). Inactive friends are not just precarious, however; they represent the mere potentiality for full friendship. Active friendship requires joint activity; Aristotle says that nothing characterizes friends as much as living together (suzên, 8.5, 9.5, 9.9–10). “Whatever it is for each person that serves as the substance or purpose of life, in this he wishes to share with his friends: thus some drink together, some play dice together, others exercise together, or hunt together, or philosophize together” (9.12 1172a2–5). These joint activities pass the time; but at least for reasonable people, they also lead to mutual self-improvement unavailable to those who do not know each other so well (9.9 1170a11), and similarly to an expansion of consciousness and the pleasure taken in life (9.9 1170a12–b16).

All this strikes us as a largely sound view of friendship. Nevertheless, Aristotle does not equally emphasize every key aspect of friendship, and we think he underplays one. When taking our friend to share our good, sometimes we worry that we may end up exploiting him, benefiting from him in a way not adequately justified by our shared activity. Presumably, Aristotle does take this worry seriously—he notes often that many of us strive, even despite ourselves, for self-sufficiency (9.9) and fret about the best way to discharge our obligations in asymmetrical relationships (8.7–8, 8.10–9.2, cumulatively Aristotle’s longest discussion in his work on friendship). Our knowledge of our friend’s attitude toward us is incomplete, especially during certain types of separation (when we may still be yielding goods from the friendship). And, if we are successful in a way that our friend is not—vagaries of luck making this always possible—we can fear that our success in fact depended on his failure. What the two books about friendship in Aristotle’s Ethics do not emphasize is the way in which ameliorating this fear—a conscientiousness for which the reasonable people Aristotle imagines in ideal friendships would be prone—may constitute a large and even essential part of the relationship between friends. In a friendship that spans years, the mutuality of goodwill must at least once be acknowledged, but also cannot be acknowledged once and for all; doubts arise, and with them efforts—of varied success—at re-acknowledgement. ^{Footnote 26} Likewise, the togetherness of living cannot simply be established once and for all; each personal victory or jealous moment demands renewed scrutiny about the share in one’s life played by the friend. Though this self-questioning dynamic points only at the imperfect attainment of friendship, not the inadequacy of Aristotle’s definition, it does seem from the practical perspective central to our experience of friendship. ^{Footnote 27}

Bagley’s article shows how friendship develops from a state of partial ignorance. A person lacks complete understanding of his own values, and those of another; but sympathy between the two allows the joint discovery of those values. This occurs improvisationally, a listening to the other, interpreting where he is going, and taking up that direction for oneself. This appeal to joint improvisation gives further definition to Aristotle’s idea of mutually acknowledged goodwill and living together. For Bagley, it is living together that improves one’s ability to acknowledge both the other’s good and one’s own, because it is in the living, and the acknowledging, that there comes to be a definitive good for which both parties seek. This view identifies the uncertainty in which, or through which, much friendship occurs.

All the same, we identify two lacunae in Bagley’s account. The first is the same with which we charged Aristotle. Improvisation may lead us momentarily astray, into conflict and misunderstanding. There may be many points where we simply doubt that we are in sync or in tune with our companion. One person goes off on a long solo; is that part of our joint song, or is it a turning away from our partnership? We are playing dissonantly; will that yield a refreshing sound, or is it just a sign of incompatible musical visions? Jazz improvisation, especially in its free jazz or free improvisatory incarnations, seems even to depend on these constant risks of departure, incongruence, and disaster. ^{Footnote 28} Great music—as well as much terrible music—can be made only in hazard. Just so, friendships based in routine, on a script, as mere ornamentation of a standard, lack a certain authenticity; those with vulnerability, heedless energy, and originality can fail, and often do, but may also be the only real ones. So the chance of doubt, and the desire to redeem (that is, make) the song or friendship from it, seems essential to good improvisation. Indeed, it is perhaps always at the moment of doubt that the song or friendship is reappraised, the direction reconsidered, the effort redoubled.

The second lacuna is related to something that Aristotle makes quite clear: friendship requires the (continued) mutual acknowledgement of that friendship. For Bagley, friendship depends on a history of joint improvisation. But friendship, generally understood, either requires knowing that one is in it or does not require knowing it. It is implausible to think that one could be in joint improvisation without knowing it. So friendship, on Bagley’s improvisational model, requires knowing one is in it. Yet to know that one is in joint improvisation takes not just looking at the current situation but recognizing the history of that joint improvisation. What this recognition amounts to is a vision of living together through time that has been building on what preceded and that aims for some as-yet undecided future. It amounts to an interpretation of the sequence of joint activities as meaningful. This sequence is meaningful in that each moment of reflection is meant to re-orient that joint activity toward continued and better joint activity. Translated into the social sphere, this improvisatory model describes an activity that is essentially narrative: the retrospective interpretation of a continually unfolding sequence of events as meaningful. We argue that it is precisely such meaning-making activity as carried out in the retelling of a shared past that is, ultimately, necessary to friendship.

Putting these two elements together, we find that Bagley’s theory implies, even as it does not state, what our narrative theory claims. The experimentation and indefiniteness that characterizes improvisation means that there will always be junctures of doubt. It is most at these junctures of doubt—and even of ambivalence, frustration, and fear—that we re-evaluate whether we are in fact in joint improvisation rather than just in solitary reverie or exploitative asymmetry. We determine the existence of a history as joint improvisation by means of narrative creation, trying to make meaningful the sequence of activities we have lived through together.

5. Two Final Considerations

In the previous section, we emphasized mutual acknowledgement. Yet the narrative constitution of friendship seems to ignore that mutuality, to the extent that, in our example, only one friend is engaging in narrative reflection. Wittgenstein’s Nephew is not a story told by the narrator and Paul together. It is, rather, the story of a friendship told unilaterally. That it is a story told unilaterally seems to ground the narrator’s very worry that he has failed to be a friend, has exploited Paul for all he’s worth, and has done little but advance his own literary career. In fact, a number of readers of Wittgenstein’s Nephew have taken basically this line. ^{Footnote 29} Joint improvisation has been replaced with solipsistic storytelling.

The superficial plausibility of this conceptual worry allows us to clarify our position. It would be misplaced to expect Wittgenstein’s Nephew to have two narrators, or that Paul should have to write his own book—Wittgenstein’s Nephew’s Friend: The Same Friendship? Narratives are told from a particular perspective, and recognition—even mutual recognition—happens from a particular perspective. The narrator seeks to establish that suzên and mutual recognition did exist, which requires him to establish that Paul himself worked to recognize (through narrative reconstruction) a friendship with himself. The narrator’s occasional attention to Paul alone, and himself alone, points to his effort to see the matter from both sides, though of course that effort is his own.

The more urgent problem is ethical, not conceptual. A risk of unilateral narrative constitution of a friendship is that without any constraints from the other’s perspective it runs the risk of no longer being true to the shared experiences of the friends, and of thus becoming both fictional and self-serving. The narrator of Wittgenstein’s Nephew explicitly raises the possibility of such self-serving impulses, as we have noted. These worries are the same doubts we have posited as essential motivators of the narration and constitution of friendship. That the narrator recounts, in his very attempt to reconstruct a life lived together with Paul, his self-incriminating worries that he used Paul merely as the object of a good story, even when Paul was in the worst state imaginable, a state that demanded above any the care for a friend that the narrator did not offer, provides good evidence (even if it does not prove) that the (possibly fictional) story is authentic, and that the narrative impulse is not merely self-serving. That the account presented in Wittgenstein’s Nephew does not take the shape of the conventional story manifests the narrator’s good faith attempt at reconstructing his past with Paul, not reimagining it. It attests to the narrator’s intention to redeem his (at times fraught or reprehensible) memories of living together—an act that refuses to erase or whitewash those doubtful aspects that originally called for this redemption. It is in this way that the telling of a friendship, even when done alone, can validly constitute something shared.

For all that, there is something true in the hope that there might be two narratives told. In seeking the mutuality of recognition, we seek our friend’s effort also to engage in a similar attempt to recuperate that mutuality. Though at first unconcerned, the narrator turns pessimistic as to whether Paul shared his efforts. This pessimism is a source of more than sadness; his worry that he exploited Paul has a basis only if Paul did not see the narrator as his friend, or did not try to. Paul’s exclamation, “Grotesque, grotesque!” (45), marks this ambivalence in his intentions to recuperate his friendship with the narrator. For while with this exclamation Paul appears to dismiss the very idea of a friendship, only by having looked back and accounted for his arrival at this juncture with his friend could he have voiced his melancholic resignation. And it is this effort at looking back that grounds the friendship. “Grotesque, grotesque!” is at once an expression of doubt, fear, and the profound desire to allay these same worries.

A second consideration is related to the first. We have argued that narration constitutes a friendship. Yet the narrator is telling the story of his living together with Paul only after Paul’s death. Might there be something absurd about constituting a friendship after the death of one of the partners, at which point any living together is obviously impossible? Or, by contrast, is it somehow necessary, in the same way that Aristotle says that we can count no man happy until he is dead (Nicomachean Ethics 1.10)? The narrator did, after all, suggest something like this, while making an exception for Paul. ^{Footnote 30}

As it turns out, the narrator seems quite nervous about this question, thinking, perhaps, that his narrative reconstruction comes too late. The death of his friend could be in some sense the source of the ultimate doubt: How could we possibly be friends? Paul is dead. But in a perverse way, the death may have been necessary, at least in this case, for the friendship’s narrative reconstruction. Paul’s dying, the narrator avers, was the source of his own survival. His conviction about this leads him to suspect that he exploited Paul and thus was not in fact a friend, a worry that motivates the narrating of their life together. But Paul also, on this view, kept the narrator alive so as to respond to this very doubt. Now, friendship after death is impossible only if friendship requires two simultaneous narratives. But we have argued against this view. Furthermore, while friendship is constituted by narrating a life together, which extends back in time, whether one’s friend is alive at that moment of narrative reconstruction is relevant only to the continuation of the friendship, not to the friendship itself, which, as it happens, is constituted retrospectively. Sometimes one’s retrospective vantage point is simultaneous with continued living together, but often it is not. We might conclude by saying that the fact that this question—whether friendship outlives death—arises does not undermine our narrative view of friendship, for example, on the grounds that it leaves out the most important part of friendship, two living people; rather, it vindicates it, in that we can explain why the question arises and why we feel pulled to answer in it both directions.