Towards a phenomenology of grief: Insights from Merleau-Ponty

This paper shows how phenomenological research can enhance our understanding of what it is to experience grief. I focus specifically on themes in the work of Maurice Merleau-Ponty, in order to develop an account that emphasizes two importantly different ways of experiencing indeterminacy . This casts light on features of grief that are disorienting and difficult to describe, while also making explicit an aspect of experience upon which the possibility of phenomenological inquiry itself depends.

different ways of experiencing indeterminacy. One of these is attributable to partial or complete comprehension of loss. The other consists in an enduring connection with the deceased, which can aid in sustaining and reorganizing one's experiential world in the face of loss. In considering the latter, it becomes clear how a relationship can endure in a certain way and, with this, why the deceased might be said to evade description.
My discussion will draw on three interrelated themes in Merleau-Ponty's work: the habitual world, phenomenological indeterminacy, and interpersonal experience. First, I will critically discuss his remarks on bereavement, phantom limbs, and anosognosia in Phenomenology of Perception, which concern the preservation of a habitual world in the face of events that undermine it (Merleau-Ponty, 1945/2012. Then, I will suggest that a more dynamic conception of grief is required: It is a long-term process of comprehending and negotiating a disturbance of one's world, something that involves a distinctive experience of indeterminacy. Following this, I will show how an enduring relationship with the deceased involves a different kind of indeterminacy and why this is in tension with the project of capturing that person's distinctiveness by appealing to however many of her properties. I will conclude by considering some parallels between the experience of grief and Merleau-Ponty's conception of phenomenological method. 1

| GRIEF AND THE HABITUAL WORLD
In Phenomenology of Perception, Merleau-Ponty addresses grief only briefly, in comparing bereavement, anosognosia (denial of illness and, more specifically, paralysis), and phantom limbs. 2 His discussion focuses on a certain way in which something can, despite its loss, be experienced as present. In his view, neither phantom limbs nor anosognosia involve nonveridical experiences with specific contents. Instead, what is preserved is a habitually and practically organized world. The world, as encountered pre-reflectively, still includes all those salient and significant practical possibilities that it did before injury or loss. 3 What we have is not the localized representation of something as present when it is not, but a diffuse and quite different kind of presence, which involves continuing to experience one's surroundings in ways that presupposes certain capacities. As Merleau-Ponty (1945/2012 writes, "to have a phantom limb is to remain open to all the actions of which the arm alone is capable and to stay within the practical field that one had prior to the mutilation." Abilities and associated expectations are etched into the experienced world, in the guise of things to be acted upon in various ways-used for something or other, obtained, transformed, navigated, avoided, and so forth. However, it should be added that not every potential activity is experienced in this manner, only those that matter. And how they matter reflects not only what one is capable of but also a range of interconnected projects, commitments, and attachments that together comprise the distinctive structure of one's life.
Merleau-Ponty thus conceives of anosognosia and phantom limbs as involving the preservation of a purposively organized life structure that is reflected in one's experienced surroundings and at odds with one's current abilities.
He also construes this in temporal terms, as a privation of autobiographical time: The passage of time does not carry away impossible projects, nor does it seal off the traumatic experience. The subject still remains open to the same impossible future, if not in his explicit thoughts, then at least in his actual being. ( 1945/2012, p. 85) One might say that this predicament involves living in the past as opposed to the present. But it is not simply a matter of experiencing the past as if it were present. There is also a shift in the form of temporal experience. Preservation of the practically organized world involves a lack of openness to future possibilities, those that might draw attention to the loss. One's sense of the immediate and longer term future is therefore altered; it no longer includes the anticipation or actualization of events with the potential to reshape the structure of one's life. According to Merleau-Ponty, this phenomenon is a matter neither of mechanistic causation nor of desires or intentions to which conscious access is limited or absent. We should, he suggests, stop thinking in terms of a contrast between physiology and psychology. Instead, such phenomena are to be construed in terms of a unitary and phenomenologically more fundamental "movement of being in and toward the world" (1945/2012, p. 80).
Merleau-Ponty indicates that bereavement can involve a sense of personal presence with this same structure. In a range of ways, the shape of one's life, and with it the kinds of significant possibilities that one experiences as inherent in things, come to depend on one's relationship with a particular person. When that person dies, the world endures despite the loss, amounting to a diffuse, nonlocalized sense of his continuing presence.
However, the comparison is questionable in some respects and unclear in others. For instance, phantom limb experiences are more diverse and multifaceted than Merleau-Ponty indicates. First-person accounts suggest that many of them do include localized, sensory qualities, which are distinct from the more diffuse retention of a world. In fact, some phantoms may be comprised principally or even exclusively of such qualities. Experiences of personal presence among the recently bereaved are similarly diverse (Ratcliffe, 2019). In addition, considerably more work is needed to tease apart the various different ways in which one's world might come to depend on a particular individual, along with the ways in which personal loss can affect that world. The effects of bereavement can encompass habitual expectations, routines, abilities, commitments, and projects of whatever complexity and duration. In the case of a project, it could be that "I do this for her," in which case an unwavering recognition of her irrevocable absence implies the unintelligibility of that project. In cases where "we do this together" or "we are committed to this," the implications of bereavement depend on how the "we" is to be analyzed. Where it is irreducible to a "you" and an "I," the project is similarly rendered unintelligible. In other cases, "I do this" might come to replace "we." And, where a project depends on another person's practical support, doing something need not lose its intelligibility. The prospect of achieving it still makes sense, but one is now unable to actualize that possibility. The inability might be irrevocable (given the irreplaceability of that person's contribution) or, alternatively, something to be compensated for. Different aspects of the habitual world will thus depend on the deceased in different ways. Merleau-Ponty's discussion does not make clear whether all of them are frozen in place or whether some are more malleable than others.
The phenomenological plausibility of the account is also questionable. A relationship with a specific individual is not just a background condition for the integrity of one's world, shaping what is experienced in and anticipated from a surrounding environment. There is also the expectation of encountering that person in a variety of situations.
Merleau-Ponty suggests that, to keep the world intact, we avoid confronting the impossibility of a personal response by not seeking it (Merleau-Ponty, 1945/2012. However, there is a tension here: Preserving those aspects of the practically configured world that implicate the deceased would also dispose us towards anticipating her appearance and, consequently, confronting her absence. This, in turn, would serve to undermine rather than sustain the world. One could maintain that the world is preserved by avoiding those projects, pastimes, and situations in which that person played a prominent role. But this would involve immersing oneself in other pursuits that do not implicate the deceased and therefore do not involve a diffuse sense of her presence. Regardless of such concerns, the sense of presence described by Merleau-Ponty is not what I am looking for here. Sartre does not refrain from addressing his friend in order to preserve a world, and Merleau-Ponty's presence is not to be found underlying a world that has been gutted of openness and potential novelty. Instead, Sartre contemplates Merleau-Ponty explicitly and finds him "alive" in a way that impedes description. Nevertheless, I suggest that Merleau-Ponty's comparison can serve as a fruitful starting point. Tensions between worlds past and present are an important aspect of profound grief. As William Maxwell writes, in his semi-autobiographical novel So Long, See You Tomorrow, "between the way things used to be and the way they were now was a void that couldn't be crossed" (1980/2012, p. 9). However, such experiences should be construed more dynamically. 4 The diffuse sense of presence captured by Merleau-Ponty overlaps and alternates with an equally diffuse sense of absence. In addition, negation of specific patterns of anticipation leads to more localized experiences of absence, as when expecting to find someone in a particular place and consequently encountering it as lacking (Fuchs, 2018;Ratcliffe, 2016). I will now describe how this dynamic also incorporates an experience of indeterminacy, something that is to be distinguished from experiences of presence, absence, conflict, and ambiguity that arise due to tensions between what is and what was.

| INDETERMINACY AND MEANING-LOSS
Merleau-Ponty has little more to say about grief and bereavement. Nevertheless, wider-ranging themes in his work serve as valuable sources of insight. At one point in Phenomenology of Perception, he briefly mentions grief, in a way that suggests an emotion with a specific object: "he suffers because he has lost his wife, or he is angry because his watch has been stolen" (1945/2012, p. 372). However, it is not clear what Merleau-Ponty takes emotions to be or how he would distinguish them from more subtle and ubiquitous experiences of our surroundings as salient and significant. He does at least claim that an emotion is not an internal mental event, hidden inside a head; it is "not a psychic, internal fact but rather a variation in our relations with others and the world which is expressed in our bodily attitude" (1964b, p. 53). 5 He also indicates that the expression of an emotion, including its linguistic expression, is not secondary to an emotional experience but integral to it. The words, according to Merleau-Ponty, "express" the "emotional essence" of their objects, in ways that are inseparable from how those objects are experienced emotionally (1945/2012, p. 193). But let us focus instead on the suggestion that an emotion consists in a "variation." This points to a distinction between an emotional disturbance and a more mundane experience of something as significant relative to some project or set of projects. For example, as I walk up the steps to the railway station, go through the ticket barrier, and wait for the train during my regular journey to work, these things are experienced as mattering to me in ways that are mundane, unsurprising, and not at all disruptive. However, the announcement that says "all trains cancelled indefinitely" is experienced differently. It disrupts my various projects: I will have to wait in the station for hours; I will miss my meeting; I will need to catch up on work tomorrow; I must inform these people now.
If this is the sort of contrast Merleau-Ponty has in mind, then episodic emotions involve experiencing potential or actual perturbations of one's experiential world, which reflect ways in which their objects impact on our projects and wider concerns. 6 However, when it comes to grief, we can be more specific. Drawing inspiration from Merleau-Ponty's contrast between language and speech (which I will address shortly), Kym Maclaren distinguishes between "emotional clichés" and "authentic passions." In her words, clichés involve "familiar routes and enticing possibilities sketched out by the individual's habits within the sensed situation," whereas authentic passions involve the "realization of unforeseen meanings within the world and new ways of becoming oneself" (Maclaren, 2011, pp. 56-58). This is not to be identified with the distinction I have drawn between a "variation" and a nondisruptive experience of one's surroundings as practically significant, given that emotional clichés equally involve disturbances of habitual arrangements. What distinguishes them from authentic emotions is their involving emotional responses that themselves unfold in familiar ways, akin to scripted performances or routines.
To illustrate the distinction, Maclaren (2011, pp. 60-62) considers intense and profound grief, which she takes to be an authentic passion. It involves, she says, the "crumbling" of a world that can "no longer exist with the meanings that it had," the "breakdown of our habitual negotiation of the world." Importantly, what Maclaren identifies here is different from the tension and ambivalence that characterizes interaction between two worlds, past and present.
Even if one's world were to adapt instantaneously so as to accommodate the loss, a sense of indeterminacy would remain. In fact, it would be maximally pronounced, given that the past world would cease to apply without being replaced by anything new. Practically salient patterns that once specified-to varying degrees-what will happen next and what one will do or ought to do next would be lacking.
However, the kind of indeterminacy I seek to characterize is attributable not only to grief's being an "authentic passion" but also to something more specific. For Maclaren, an authentic passion is uninhibited by a constraining structure and thus harbors the potential to reshape a world more usually taken for granted. But it can be added that grief takes this form because it undermines the very context within which it arises. The person who has died was not only an object within one's world but also a condition of intelligibility for that world. Complete acknowledgement of loss therefore involves a disturbance of the world within which the loss is initially experienced as occurring (Ratcliffe, 2017b). The emotional experience is emancipated from habits and norms would otherwise have shaped its unfolding, as its concrete object (the death of a particular person) implies their inapplicability. There is a profound sense of being lost; it is not that the right path cannot be discovered but that there is no path to follow and nowhere familiar to retreat to. A distinction can thus be drawn between the category of emotions that deviate from established paths and a more specific category, to which grief belongs, of emotions that involve facing the loss of those paths. In the case of grief, it would also be wrong to think of clichés and authentic emotions as mutually exclusive. While undergoing a profound disturbance of the habitual world, one might seek out familiar patterns wherever possible, including familiar emotional patterns, perhaps with the support of other people. So, certain emotional clichés may turn out to be symptomatic of authentic passions and integral to their expression.
Hence, something that renders grief both profoundly disorientating and also philosophically interesting is the breakdown of a phenomenological structure that is more usually presupposed. To further clarify what this experience consists of, let us turn to an account of the anticipatory structure of experience developed by Edmund Husserl in some of his later works, which Merleau-Ponty explicitly refers to, appears to endorse, and certainly draws inspiration from (e.g., Merleau-Ponty, 1968, pp. 41-42). 7 According to Husserl, our experience is riddled with possibilities for perceptual access, most of which take the form of anticipation. Ordinarily, it is not a matter of "if I do this, then x might happen." Rather, anticipation takes the form of certainty or unwavering confidence: "x will happen." Thus, when we anticipate and then experience the arrival of x, no conflicting possibilities appear salient. Of course, what we anticipate is variably specific in content, as in "when I enter the room, I will see one or more people" and "when I enter the room, I will see that particular person." However, in referring to "certainty," Husserl is not suggesting that things are anticipated in a precise way and then proceed to unfold exactly as anticipated, but that uncertainty is resolved in a manner at least consistent with pre-reflective expectation. There is no conflict between one's various expectations or between those expectations and what subsequently occurs. So, for the most part, human experience involves the dynamic and cohesive actualization of possibilities, in line with anticipation. Of course, anticipation sometimes involves doubt, tension, and conflict. And, instead of fulfilment, there can be surprise or negation. Nevertheless, events are experienced as potentially or actually anomalous only relative to a wider backdrop of cohesive and consistent unfolding. 8 Although Husserl emphasizes perceptual experience and, more specifically, how activities are experienced as actualizing and having the potential to actualize possibilities for ongoing perceptual access, his approach can also be applied to how we experience goal-directed actions (our own actions and also those of others) and impersonal events as impacting or potentially impacting on possibilities that reflect our concerns.
When anticipation takes this form, there is a sense of completeness and determinacy. As only one possibility presents itself, there are no alternatives to undermine it and so it appears in the guise of certainty. Where that possibility is perceptual in nature and concerns the revelation of properties already inherent in one's surroundings, it contributes to an experience of those properties as already there and of one's surroundings as complete, bereft of ambiguity. Where it concerns something that will happen, it contributes to a sense of determinacy: Only one set of future possibilities is there to be actualized. Merleau-Ponty maintains that we ordinarily take this clearly defined, spatiotemporal world for granted and fail to acknowledge those underlying phenomenological achievements that enable an inchoate soup of possibilities to coagulate into one or another specific arrangement.
Grief, however, affects systems of anticipation that both shape perceptual experience and provide guidance for action, disrupting what was once presupposed. Granted, one can still anticipate the kinds of practically significant states of affairs to be actualized as one walks to the supermarket checkout or squeezes the toothpaste out of the tube. However, larger patterns of interconnected, unfolding possibilities will depend on the deceased in a range of ways. When things cease to offer these possibilities, they no longer relate to one another in stable, unambiguous ways that reflect long-term projects. My claim is not that, with the loss of certain kinds of possibilities involving an entity, that entity itself starts to look somehow fuzzy. Rather, it is the practically meaningful connections between things that are eroded-one's sense of the overall situation in which things are experienced as appearing and changing. Determinate arrangements give way to haziness. What is lacking here is not merely epistemic in nature. It is not just that one cannot find a path to follow; the paths have gone. There is no fact of the matter concerning how things fit together, no pattern, nothing to specify what is to be done. This is not to suggest that our experiences of entities remain unaltered and that only the practically meaningful relations between them are eroded. Loss of possibilities is not ordinarily so profound that things cease to be experienced as things at all. Nevertheless, this is consistent with their appearing oddly bereft of their usual familiarity and significance, decontextualized. Joyce Carol Oates (2011, p. 63) describes the experience as follows: Not yet have I realized -this will take time-that as a widow I will be reduced to a world of things. And these things retain but the faintest glimmer of their original identity and meaning as in a dead and desiccated husk of something once organic there might be discerned a glimmer of its original identity and meaning.
In such a situation, even if one could somehow let go of the past and inhabit a present in which the death is fully acknowledged, there would be nowhere to go, no route to follow, nothing to be done. Hence, the dynamism and tension between past and present can be thought of as integral to a process whereby the world is reconfigured over time rather than abandoned in one go and then rebuilt out of nothing. As Helen Macdonald (2014, p. 16) writes in her autobiographical account of bereavement, "my mind struggled to build across the gap, make a new and inhabitable world." The wildness of a world untamed by habit is, to some extent at least, kept at bay. 9 For Merleau-Ponty, experiences of indeterminacy also extend to language and linguistic thought. He distinguishes language as a "sedimented" or habitually entrenched institution from a kind of authentic language or "speech" that somehow transcends the usual possibilities associated with our words, giving rise to new meanings. In his late (and incomplete) work, The Prose of the World, the two are contrasted as follows: We may say that there are two languages. First, there is language after the fact, or language as an institution, which effaces itself in order to yield the meaning which it conveys. Second, there is the language which creates itself in its expressive acts, which sweeps me on from the signs toward meaningsedimented language and speech. ( Merleau-Ponty, 1973, p. 10) Here and elsewhere, Merleau-Ponty draws inspiration from his interpretation of Saussure. However, something approximating the distinction is also present in Phenomenology of Perception, prior to any engagement with Saussure.
Here, Merleau-Ponty distinguishes a "sedimented" language, which demands no effort of expression or comprehension, from a speech that disrupts the usual constraints so as to realize genuinely new meanings (Merleau-Ponty, 1945/2012. Like the possibilities offered by our surroundings, the possibilities of established language are constrained by shared norms, habits, and enduring concerns. Indeed, the two are inextricable. Our words, like our experiences and activities, arise and are experienced within contexts of practice that reflect and relate to stable sets of concerns. Consequently, both are disrupted by forms of experience that impact on the intelligibility of those concerns. 10 One might think of the process whereby a tension between two worlds (past and present) is negotiated as a conflict between bodily, habitual, nonconceptual expectations that take time to shift and an explicit, cognitive, propositional acceptance of the death. But that would be a mistake. What is yet to accommodate the loss is not just a matter of nonconceptual expectation, as illustrated by those moments (that arise during grief and also in many other contexts) when one thinks through a fairly routine pattern of activity and is then struck by its inapplicability due to the altered situation. The rupture encompasses language and thought just as it does experience and habitual activity.
So, the correct distinction to draw is not between conceptual and nonconceptual or thinking and unthinking ways of relating to the world, but between thought that is integrated into a habitual world and thought that is not. Reconciling what one now "knows" with familiar patterns of thinking can take time: And it wasn't until we were standing on Queenstown Road station, on an unfamiliar platform under a white wooden canopy, wasn't until we were walking towards the exit, that I realised, for the first time, that I would never see my father again. ( Macdonald, 2014, p. 106) What I have said so far complements the widely held view that grief typically incorporates an adjustment process, whereby conflicting perspectives are integrated over time. 11 However, it does not answer the question with which I began. How does all of this relate to Sartre's remarks? The answer, I will suggest, is that a sense of connection with the deceased (the nature of which is in tension with a certain kind of descriptive project) can play an important role in navigating the indeterminacy of loss and also endure beyond the point where one might be said to have adjusted to the loss.

| PERSONS AND POSSIBILITIES
Exclusive emphasis on how one's world is affected by bereavement might suggest a conception of grief that is overly self-centered. Insofar as grief concerns how the death impacts on my world, it is not directed towards the other person but towards what has happened to me. Hence, losing a home, a large sum of money, or a job can involve similar phenomenological disturbances. However, there is also something distinctive about the loss of a person. We might seek to accommodate this by emphasizing that the practically meaningful world depends on the deceased partly because of various projects and commitments that one takes up for her sake. While this dispenses with the selfcenteredness objection, it still does not do justice to the interpersonal phenomenology of grief. In particular, it overemphasizes the ways in which other people are implicated in concrete projects. In so doing, it fails to account for the profound grief and sense of absence that one might feel in response to the death of someone who was not integrated into one's practice to such a large extent, perhaps someone who lived far away. However, there is another, more subtle way in which the possibilities of one's world can come to depend on a particular person. In comparing bereavement to phantom limbs, Merleau-Ponty (1945/2012 writes that "we only understand the absence or the death of a friend in the moment in which we expect a response from him and feel [éprouver] that there will no longer be one." He adds that we turn away from the question, avoiding a confrontation with absence.
But what is the question that we avoid asking; what is it to direct oneself towards a particular person and anticipate his distinctive way of responding?
Merleau-Ponty's account of a world preserved in spite of loss indicates that such a world is also lacking. It is insulated from potential scenarios that might reshape it and consequently bereft of a more usual openness to certain kinds of future possibilities. What is missing is an indeterminacy quite different from that so far described, an openness to possibilities that does not compromise the integrity of an experienced situation, but instead constitutes a sense that "this is not all there is," that "things could change for better or for worse in ways not specified by the situation as it is currently experienced." This type of indeterminacy, I suggest, is inseparable from the anticipation and experience of certain kinds of interpersonal relations.
Merleau-Ponty is consistently critical of a tendency among philosophers to construe interpersonal experience in terms of an encounter with observable behavior that leads us to postulate an internal mental life: "I know quite well that back there there is only 'darkness crammed with organs'" (Merleau-Ponty, 1973, p. 133). As an alternative, he indicates that, when we encounter someone else as a locus of experience, we experience her mental life as inherent in her activities, in the guise of a cohesive and dynamic set of potentialities for expressions, gestures, activities, and relations. These possibilities are neither straightforwardly present nor absent: "absence is itself rooted in presence; it is through his body that the other person's soul is in my eyes" (Merleau-Ponty, 1964a, p. 172). Together, unfolding possibilities constitute a temporally organized pattern, which Merleau-Ponty would call a "style" (e.g., Merleau-Ponty, 1945/2012. We might say that encountering someone in a specifically personal way involves anticipating and experiencing a distinctively personal style. 12 Importantly, this also involves being affected, having the possibilities of my world somehow altered by an engagement with her possibilities: My gaze falls upon a living body performing an action and the objects that surround it immediately receive a new layer of signification: they are no longer merely what I could do with them, they are also what this behavior is about to do with them. A vortex forms around the perceived body into which my world is drawn and, so to speak, sucked in…. ( Merleau-Ponty, 1945/2012 Interpersonal encounters thus shape, usually in subtle ways, how we experience our surroundings. The effect is most pronounced in the context of sustained patterns of interaction, where it encompasses not only world experience but also language and thought. Interactions of a certain quality have the potential to dislodge us from familiar ways of experiencing, acting, thinking, and speaking, in ways that are not attributable merely to whichever propositional contents are transmitted between the two parties: Speaking and listening not only presuppose thought but -even more essential, for it is practically the foundation of thought-the capacity to allow oneself to be pulled down and rebuilt again by the other person before one, by others who may come along, and in principle by anyone. ( Merleau-Ponty, 1973, pp. 19-20) A genuine conversation gives me access to thoughts that I did not know myself capable of, that I was not capable of, and sometimes I feel myself followed in a route unknown to myself which my words, cast back by the other, are in the process of tracing out for me. ( Merleau-Ponty, 1968, p. 13) The suggestion is not that another individual possesses the potential to affect one in a particular way. Rather, it is through interaction between the two parties that new meanings are forged: "my words and those of my interlocutor are called forth by the state of the discussion and are inserted into a shared operation of which neither of us is the creator" (Merleau-Ponty, 1945/2012. To this, it can be added that the anticipation of such interactions constitutes an openness to the possibility of one's world being shaped in subtle or more profound ways. To put it another way, other people play a distinctive regulative role, imbuing the experienced world with a degree and type of indeterminacy that is compatible both with having structured systems of projects and with moving into an open future. 13 A sense of the potential for a certain type of interpersonal relation sustains an openness to possibility with which the present is imbued.
Importantly, these points apply not only to the structure of interpersonal experience in general but also to our relations with particular people. Anticipated and actual relations with specific individuals sustain and shape the structure of our lives in distinctive ways. As Eugene Gendlin (1978Gendlin ( /2003

writes of friendship:
We all know people with whom it is best not to share anything that matters to us. If we have experienced something exciting, and if we tell it to those people, it will seem almost dull. If we have a secret, we will keep it safe from those people, safe inside us, untold. That way it won't shrivel up and lose all the meaning it has for us.
But if you are lucky, you know one person with whom it is the other way around. If you tell that person something exciting, it becomes more exciting. A great story will expand, you will find yourself telling it in more detail, finding the richness of all the elements, more than when you only thought about it alone. Whatever matters to you, you save it until you can tell it to that person.
To know someone is, in part, to experience and be affected in a particular way by that person's distinctive style, by relational possibilities that are unique to her. To varying degrees, the integrity of one's world can come to depend on how one relates to that person and vice versa. Interpersonal relationships include the potential to affect and be affected by one another in a range of ways. Hence, they can aid in the navigation of upheaval, by sustaining and opening up possibilities. Personal loss is thus to be distinguished from emotional upheaval more generally. With the loss of a particular person, what may be lost too is the prospect of a certain style of relating, not just to her but to the world in general. Included in this are possibilities that one might otherwise have drawn upon in negotiating another form of indeterminacy: that which arises when familiar patterns are rendered unintelligible by events.
It could be that, with a person's death, access to her style is altogether lost, but nothing I have said requires this conclusion. Take a case where one might be said to have fully adjusted to the loss, where one experiences and interacts with the world in a way that no longer presupposes possibilities that depend on the deceased. A sense of her style, the distinctive way in which she shaped one's possibilities, is not exhausted by a comprehensive inventory of all the concrete ways in which she contributed to one's life structure. It is indeterminate and cannot be pinned down to any number of more specific roles in any number of situations. Hence, it retains the potential to endure.
This might seem overly speculative. However, it complements an extensive literature on what have become known as continuing bonds, according to which bereavement ordinarily involves revising and retaining connections and relations with the deceased, rather than simply letting go of that person (e.g., Klass, Silverman, & Nickman, 1996;Klass & Steffen, 2018;Steffen & Coyle, 2011). What I have identified here is one form that a continuing bond might take, or one ingredient that a continuing bond might include. 14 As Kathleen Higgins (2013, p. 173) observes, "one's realistic expectations regarding interaction with another person are irreparably altered by that person's death; but one's sense of identity continues to be constructed in part on the basis of one's relationship to that person." With this, we arrive at the answer to my question: How could Merleau-Ponty remain alive for Sartre, in a way that eludes description? Sartre's world is still affected by a certain style. Its nature cannot be specified by appealing to however many properties that might be attributed to Merleau-Ponty, as it essentially involves a sense of openness to inchoate possibilities. This is one way in which the dead could be said to remain alive. In fact, Merleau-Ponty himself hints at just such a position, in the first of three notes that follow his 1953 Inaugural Lecture at the Collège de France. There, he contrasts a questionable sense of "closeness" to the dead, which involves being able to hold them in our imagination without their ever being able to place us "in question," with a form of memory that "respects them" by retaining "the accent of their freedom and the incompleteness of their lives" (Merleau-Ponty, 1953/1970. 15 This is consistent with my proposal that one can continue to experience the style of a person who has died, something that involves a self-affecting openness to possibilities and cannot be captured in terms of a more determinate image or description of the person. It should be added that a sense of enduring connection might be fragile, incomplete, and-at timesunsustainable, making it compatible with intermittent or constant feelings of loss and absence. Given this, what I have said can also be reconciled with the final sentence of Sartre's long essay: "There is nothing to be concluded from this except that this long friendship, neither done nor undone, obliterated when it was about to be reborn, or broken, remains inside me, an ever-open wound" (Sartre, 1998, p. 624).

| GRIEF AND THE PHENOMENOLOGICAL REDUCTION
I have suggested that phenomenological philosophy can enhance our understanding of grief, but the relationship is also one of mutual illumination. The structure of grief resembles, in certain respects at least, Merleau-Ponty's conception of phenomenological method. As he writes in his late and incomplete work The Visible and the Invisible, philosophy "interrogates the perceptual faithbut neither expects nor receives an answer in the ordinary sense" (p. 103). He did not arrive at this view of philosophy only late in his career. That phenomenology does not provide a straightforward answer is also suggested by a well-known passage in Phenomenology of Perception: Perhaps the best formulation of the reduction is the one offered by Husserl's assistant Eugen Fink when he spoke of a "wonder" before the world. Reflection does not withdraw from the world toward the unity of consciousness as the foundation of the world; rather, it steps back in order to see transcendences spring forth and it loosens the intentional threads that connect us to the world in order to make them appear; it alone is consciousness of the world because it reveals the world as strange and paradoxical. ( Merleau-Ponty, 1945/2012 When the habitual world is somehow disturbed, what was previously taken as given becomes salient. The "intentional threads""loosen," facilitating reflective access to something we and our words might otherwise overlook. This philosophical procedure does not involve adopting some form of theoretical perspective in order to acquire new propositional knowledge about human experience. Instead, glimpsing the phenomenological achievements underlying the habitual world (including habitual language and thought) involves a kind of emotional disturbance. In the above passage, it takes the form of "wonder." And describing what we have glimpsed cannot just involve employing words in their usual ways. By lapsing into sedimented language, we would end up presupposing the very achievements that we seek to reveal. What is needed as well, it would seem, is what Merleau-Ponty calls "speech." This is why Merleau-Ponty comes increasingly to regard the boundaries between phenomenological philosophy, art, and literature as blurred. All of them involve forms of expression that somehow disrupt entrenched patterns, revealing their contingency and opening up new possibilities: "philosophy, like a work of art, is an object that can arouse more thoughts than those that are 'contained' in it" (Merleau-Ponty, 1968, p. 199). 16 If this is how we are to think of phenomenological inquiry, then its structure is much like that of profound grief.
We might say that grief resembles an involuntary phenomenological reduction. Granted, one is forced into it without philosophical insight or prior training and left disorientated and bewildered. Nevertheless, a common theme is the revelation of an indeterminacy that lurks beneath the world of everyday experience and is seldom explicitly acknowledged. 17 As the philosopher Susan Dunston (2010, p. 166) writes, reflecting on her own experience of grief, we are "immersed in a fluid world that fixed, representative, and codified facts scarcely touch." The kind of phenomenological insight sought by Merleau-Ponty is something that could be achieved in any number of ways, so long as they serve to disrupt what is more usually taken for granted. Hence, experiencing grief could indeed be integral to phenomenological method. The difference between the two is attributable to a combination of reflective attentiveness and philosophical training, more so than to their respective revelatory capacities. Phenomenological reflection, characterized in this way, need not privilege the first-person perspective. For Merleau-Ponty, phenomenology also involves relating to and being affected by others in ways that disrupt one's habitual immersion in the world. As he remarks in his analysis of hallucination, "the situation of the patient whom I question appears to me within my own situation and, in this phenomenon with two centers, I learn to know myself as much as I learn to know the other person" (Merleau-Ponty, 1945/2012. This also applies to grief: We can be affected in phenomenologically informative ways by the distinctive "styles" of those who are grieving. In addition to this, our own emotional confrontations with indeterminacy can serve as sources of philosophical insight. So, if we follow a philosophical path similar to that laid out by Merleau-Ponty, we find that phenomenological research has much to contribute to the study of grief and vice versa.

Matthew Ratcliffe
https://orcid.org/0000-0003-4519-4833 ENDNOTES 1 The account developed here is fairly abstract and applies to a range of grief-experiences that differ in various other ways.
Furthermore, I do not wish to maintain that it captures something central to all of those experiences we might term "grief." My more modest aim is to identify something that is central to some grief experiences. One implication of my discussion is that there is something distinctive about grief over the death of a person, attributable to the manner in which interpersonal relations shape our sense of the possible and sustain a habitual world. Even so, some of my points apply equally to loss and upheaval more generally, including the deaths of nonhuman animals and the breakdown of interpersonal relationships due to other causes.
others upon whom it depends. Beauvoir proposes this as an alternative to Sartre's emphasis on experiencing another person's possibilities as the extinction of one's own (Sartre, 1943(Sartre, /1989. 14 My account also complements a growing body of research on human emotion regulation. Earlier research in this field tended to focus exclusively on processes internal to the individual (e.g., Gross, 1999). However, there is now an increasing emphasis on the ways in which interpersonal and social processes are integral to the regulation of emotion (e.g., Gross, 2014). It has been proposed that romantic love is similar in certain respects to infant attachment and that losing a partner can involve significant disruption of regulative processes (Hofer, 1984;Sbarra & Hazan, 2008). 15 I am grateful to a member of the audience at the 2019 conference of the International Merleau-Ponty Circle, where I presented an earlier version of this paper, for recognizing the relevance of this note to my discussion and directing me to it. 16 There are similar remarks in The Prose of the World and some of the essays collected in Sense and Non-Sense.
17 Carel (2018) argues at length that experiences of illness can play a similar role.