I would like to thank Thomas Fuchs, Lawrence Kirkmayer, Elisabetta Lalumera, Andreas Brekke Carlsson, as well as two anonymous reviewers for helpful comments and/or discussions on this paper. This work was supported by the Mobility Grant Fellowship Programme (FRICON) funded by The Research Council of Norway and the Marie Skłodowska-Curie Actions programme (project number: 275251)

1. Introduction

1Sadness is often an appropriate emotional response to difficult life events. Such events often lead to a prolonged, albeit non-pathological, sadness. The loss of a close person is one such event and grief is normally taken to be an appropriate response to it. Despite their adequacy, the sadness of the grieving person and the variety of symptoms that grief involves have been a controversial subject in recent discussions in psychiatry (DSM-IV, DSM-5; ICD-11). On the one hand, many emotions and symptoms characteristic of grief seem to overlap with common symptoms of depression. But if grief is an appropriate response to a loss of a close person, then we should avoid medicalizing non-pathological grief responses. In recent debates in psychiatry it has been suggested that, although grief and depression may overlap, we need some rough guidelines for how the two could be distinguished in diagnostic practice (DSM-IV; DSM-5, APA, 2013).

2On the other hand, it has been argued that some forms of grief may, indeed, be in some sense pathological and require clinical attention (Lichtenthal et al., 2004; Zisook et al., 2010). Some people do not recover from the sadness that comes with grief and experience prolonged suffering and dysfunction. Because of that, it has been argued that we might need some reliable criteria for distinguishing between typical forms of grief and those that could qualify for intervention (Lichtenthal et al., 2004). Complicated grief (CG) is a commonly used term to describe the latter (Zisook & Shear, 2009). The need for diagnostic criteria for CG has been recognized in recent and forthcoming editions of diagnostic manuals that provide operationalized criteria for mental disorders (DSM-5; ICD-11 Beta draft). However, the task of distinguishing between uncomplicated and complicated form of grief is daunting and the proposed solutions are still controversial (e.g. Wakefield, 2013).

3In this paper I will show that a phenomenological approach can help with the task of drawing the divide between uncomplicated and complicated grief. This task is important, since complicated grief may require clinical attention, whether or not it should find its place in diagnostic manuals. Drawing on Ratcliffe’s (2017) and Fuchs’ (2018) phenomenological analyses of typical grief responses, I will argue that a promising way to draw a divide between uncomplicated and complicated grief is to look for the presence of reintegration processes geared towards establishing a new relation between the mourner and the deceased (section 4). I will propose an account of the dynamics and etiology of complicated grief, which will be useful in distinguishing between typical grief responses and complicated grief. I start by introducing the notion of grief and summarizing some of the recent discussions in psychiatry concerning the question of how to distinguish it from depression (section 2). I then move on to discuss complicated grief and recent attempts at providing diagnostic criteria of it (section 3).

2. Grief and depression

4Grief can be broadly defined as a reaction to the loss of a close person (this kind of loss is often called bereavement), involving both psychological and bodily experiences (Gross, 2015, p. 5). So conceived, grief is a universal psychological phenomenon. Although grief often involves intense suffering and leads to existential crisis, with time, most grieving people adjust to the new situation and manage to resume their life after the loss. To date there is no evidence clearly indicating that grief requires treatment or professional intervention (Jordan & Neimeyer, 2003). Grief may take many forms, depending on, for example, cross-cultural differences in its expression. Among the painful emotions commonly experienced in grief we typically find: shock, sadness, loss, anxiety, guilt, regret, fear, loneliness, intrusive images and thoughts, depersonalization. It is often emphasized that grief is not a state but rather a process with a particular dynamics (Zisook & Shear, 2009; Ratcliffe, 2017). At first, such painful emotions are often overwhelming. With time, they tend to come in bursts and the mourner can find other activities and situations meaningful and joyful. The overall experience of a grieving person can and often is compatible with experiencing some positive emotions.

5Several emotions and symptoms commonly experienced in grief overlap with those that can be found in major depression. This observation has recently led to a debate concerning the diagnostic criteria for major depressive disorder and has drawn researchers’ attention to the nature of uncomplicated grief. Since grief is a commonly experienced, appropriate response to a loss, psychiatry shouldn’t medicalize typical grief responses. Thus, we need a way of distinguishing between depression and grief. Following this observation, it has been argued that bereavement, i.e. the loss of someone close, may function as an exclusion criterion for major depressive disorder (DSM-IV). The so-called bereavement exclusion criterion was intended to discourage the diagnosis of major depressive disorder during the first two months after the loss and prevent the medicalization of grief. However, the bereavement exclusion criterion turned out to be controversial. It has been argued that it may have harmed those grieving people who have full-blown symptoms of major depressive disorder by depriving them of treatment and professional help (Parkes, 2013). In the aftermath of this debate, the bereavement exclusion criterion was removed from the DSM-5, although not without criticism (Wakefield, 2013, p. 171).

6The authors of the DSM-5 decided that in the case of bereavement and when grief symptoms overlap with those of major depressive disorder, a depression diagnosis should still be carefully considered by drawing on individual history and the situational context (DSM-5, p. 161). The following guidelines for distinguishing between grief and depression were suggested. Grief tends to involve “feelings of emptiness and loss”, whereas depression involves “depressed mood and the inability to anticipate happiness or pleasure”. Depression is more pervasive and persistent, while positive emotions still arise during grief. In addition, while depression often involves feelings of worthlessness and self-loathing, grief usually does not affect one’s self-esteem. In both cases, thoughts of dying may occur, but they typically differ in content: the depressed person may feel that she does not deserve to live, while the bereaved person may rather think of joining the deceased (for critical discussion see Ratcliffe, 2017).

7The characteristic dynamics of grief is one important feature distinguishing it from depression. Zisook & Shear (2009, pp. 68-69) distinguish between acute and integrated grief. Acute grief, according to them, occurs early after the death of a close person and can involve a variety of painful emotions that tend to be omnipresent and have a high degree of intensity, as well as various dysfunctional behaviours and relative disinterest in other people and activities. The level of preoccupation with sadness and accompanying emotions in this form of grief may vary and, as Zisook & Shear (2009) observe, the experienced emotions tend to wane with time. Within a couple of months, acute grief tends to transform into integrated or abiding grief. The latter commonly involves sadness, thoughts and feelings concerning the deceased, due to which the loss becomes accommodated in the autobiographical memory of the mourner. Integrated grief is less preoccupying and less overwhelming and the grieving person can find some ways of enjoying other aspects of life and engage with other people. At this stage, the mourner typically finds new and meaningful ways to continue their relationship with the deceased (Zisook & Shear, 2009; Fuchs, 2018).

8The above discussion concerning the divide between grief and depression shows how modern psychiatry concevies of grief and struggles to differentiate it from depression. By putting it in a broader nosological context, the above material draws our attention to the nature of typical grief responses and can thus help understanding how they may in some cases develop into a pathological form. Although the suffering and existential crisis that are commonly involved in grief can be particularly intense and overwhelming, with time most people come to terms with their loss in one way or another. However, in some cases grief does not lead to any kind of resolution, suffering continues and the mourner cannot resume their life. Such cases are often labelled as complicated grief (CG). In the following section I introduce the notion of complicated grief and summarize recent attempts to establish diagnostic criteria for capturing it.

3. Complicated grief and recent nosological attempts

9Drawing on the above sketched distinction between acute and integrated grief, Zisook & Shear (2009) characterize complicated grief as a result of “a failure to transition from acute to integrated grief” (p. 69). They estimate that complicated grief may affect about 10% of all grieving people and tends to last well beyond six months. As in normal grief, complicated grief may take various forms and involve a whole spectrum of painful emotions and behavioural symptoms. These can be captured by the so-called separation distress - continuing preoccupation with the deceased that involves recurrent bursts of painful emotions and intense longing for the deceased. The separation distress symptoms may be accompanied by ruminating thoughts about circumstances or consequences of the loss, anger, bitterness, feelings of estrangement and guilt - the so-called traumatic distress symptoms (Zisook & Shear, 2009; Zisook et al., 2010). A good measure of CG symptoms is provided by the 19-item Inventory of Complicated Grief (ICG, Prigerson et al., 1995). Complicated grief is indicated by the score equal or higher than 30 at the time of six months after the loss. Several studies suggest that the score qualifying for CG on that scale is correlated with impairment and negative health consequences, such as: sleep disturbances (Hardison et al., 2005); daily routine disruptions (Monk et al., 2006); increased risk for cancer, cardiac disease, substance abuse and suicidality (Prigerson et al., 1999; Szanto et al., 2006).

10The unremitting nature of complicated grief poses a serious threat to the well-being and life of the mourner. Because of that, it has been argued that complicated grief deserves clinical attention (Lichtenthal et al., 2004; Parkes, 2013; Jordan & Litz, 2014; Zisook et al., 2010). Whether and how complicated grief should be addressed in psychiatry is a thorny issue and there may be various approaches to how such clinical attention could be fostered. Establishing a set of operationalized criteria for complicated grief may be one, but need not be the best solution. Given the dominant operationalization approach in psychiatry, where the focus has been on providing operationalized lists of criteria required for clinical diagnoses to improve their validity (APA, 2013; Andreasen, 2006), the main approach also in this debate has so far been on whether and how to design the adequate and reliable diagnostic criteria that would capture complicated grief (Lichtenthal et al., 2004; Jordan & Litz, 2014).

11It has been argued that the characteristic phenomenology, behavioural symptoms, trajectory, and clinical correlates of complicated grief form a distinguishable nosological unit that cannot be captured by any other currently established diagnostic unit, such as major depressive disorder (MDD), posttraumatic stress disorder (PTSD) or adaptation disorder (AD) (Lichtenthal et al., 2004). Zisook et al. (2014) argue that establishing a separate nosological unit for complicated grief could foster research and professional help for individuals suffering from it by providing a commonly shared point of reference. Simon (2013) provides some evidence that CG may be responsive to targeted intervention of complicated grief treatment (CGT), where interpersonal therapy is combined with cognitive behavioural techniques.

12Following the above arguments and desiderata, the need for diagnostic criteria for CG has been recognized by the authors of recent and forthcoming editions of the operationalized diagnostic criteria for mental disorders (DSM-5; ICD-11 Beta draft). The DSM-5 (APA, 2013) decided to include diagnostic criteria corresponding with the complicated grief symptoms in the section for conditions that require further study. The proposed criteria for the Persistent complex bereavement-related disorder require (APA, 2013):

1. Death of a close other

2. Since the death, at least one of the following on most days to a clinically significant degree for at least 12 months:
   
        1) Persistent yearning for the deceased
        2) Intense sorrow and emotional pain in response to the death
        3) Preoccupation with the deceased
        4) Preoccupation with the circumstances of the death

3. Since the death, at least six of the following on most days to a clinically significant degree for at least 12 months after the death:
   
        1) Marked difficulty accepting the death
        2) Disbelief or emotional numbness over the loss
        3) Difficulty with positive reminiscing about the deceased
        4) Bitterness or anger related to the loss
        5) Maladaptive appraisals about oneself in relation to the deceased or the death (e.g. self-blame)
        6) Excessive avoidance of reminders of the loss
        7) A desire to die to be with the deceased
        8) Difficulty trusting other people since the death
        9) Feeling alone or detached from other people since the death
        10) Feeling that life is meaningless or empty without the deceased or the belief that one cannot function without the deceased
        11) Confusion about one’s role in life or a diminished sense of one’s identity
        12) Difficulty or reluctance to pursue interests or to plan for the future (e.g. friendships, activities) since the loss

4. The disturbance causes clinically significant distress or impairment in social, occupational or other important areas of functioning

5. The bereavement reaction must be out of proportion or inconsistent with cultural or religious norms

14The debate concerning the need for diagnostic criteria for CG continues (Bandini, 2015; Brinkmann, 2018). The DSM-5 suggestion was criticized for “representing a problematic compromise between competing grief theories that may lead to overdiagnosis and overtreatment” (Wakefield, 2013, p. 172). Other concerns are: pathologizing non-complicated grief (Stroebe et al., 2001); simplifying individual and cultural variability in grief expression (Prigerson et al., 2002); increased costs of complicated grief treatment, if it were provided to the estimated 20–33% of all people (Piper et al., 2007).

15Whether or not complicated grief will find its place in the forthcoming diagnostic manuals for psychiatric disorders, the consensus seems to be that it often requires some sort of clinical attention. The central dilemma in this discussion is whether and how to recognize and appropriately respond to complicated grief to avoid, on the one hand, the threat of medicalization of appropriate grief responses, and on the other hand, provide help to those who suffer from complicated grief (Zisook et al., 2014). The success in this task crucially depends on gaining a better understanding of how a divide between uncomplicated and complicated grief can be drawn. In the following section I show that the insights from recent phenomenological work on grief can be particularly useful for this task and propose one, novel way of drawing the divide based on a conception on the dynamics and etiology of complicated grief.

4. Towards the phenomenological analysis of complicated grief

16Recent research on the phenomenology of grief can be helpful in understanding the divide between uncomplicated and complicated grief. So far, this research has focused primarily on the underlying structure and dynamics of experiences involved in typical grief responses. In this section I go beyond recent discussions on the phenomenology of grief and focus on complicated grief. In particular, I argue that one of the core features of complicated grief is related to the absence or disturbance of reintegration processes geared towards establishing a new relation with the deceased. So understood, complicated grief can be identified by, among others, careful investigation into the dynamics of grief experience. Phenomenological analysis provides means for such an investigation and can foster various forms of clinical help for those who suffer from complicated grief. I start by summarizing recent phenomenological research on grief and then sketch my proposal for distinguishing typical grief from complicated grief.

17In a recent paper on grief, Matthew Ratcliffe (2017) provides a phenomenological analysis of the structural differences between grief and depression experiences and argues that his approach can facilitate differentiating between the two in clinical practice. He proposes the following three differences between experiences of ‘typical’ grief and major depression. First, grief involves losing what he calls systems of possibility. The lost systems of possibility are those that depend on a relationship with a deceased person. “When one is confronted by that person’s irrevocable absence, they collapse” (p. 4). The change in access to such possibilities, Ratcliffe argues, affects the way in which one relates to the world. Despite its overwhelming nature, grief typically does not imply the loss of access to what Ratcliffe calls kinds of possibility, which has a more global character and involves not being able to see future as open and full of possibilities and is characteristic for depression. Second, according to Ratcliffe typical grief involves dynamic perspective-shifting, whereas depression crucially involves an inability to shift perspective. While grief is typically experienced as a process with a certain dynamics, severe depression typically involves experiencing the world as static, unchanging and inescapable (p. 8). In depression, a person feels that they cannot adopt a perspective outside of the state they are in, hence the diminished ability to shift perspectives. This is quite different from experiencing conflicting perspectives in grief, where the presence and the absence of the deceased person and the contrast between the past and present world come to the fore in one’s experience (see also Fuchs, 2018). Ratcliffe suggests that the above conflict between the world before and after the loss is a crucial element of the underlying structure of experience in grief.

18Third, according to Ratcliffe grief typically involves a sustained ability to relate to and feel connected with other people, including the deceased, the capacity for which is substantially reduced in depression (p. 10). Depression often involves insurmountable isolation from people and the sense of not being able to enter into interpersonal relations. Grief, on the other hand, is often centered around the continuing experience of relating oneself to the deceased. In some cases grief and continuing preoccupation with the deceased may lead to deep isolation of the grieving person from other people and activities (p. 10). Nevertheless, the remaining sense of connection with the deceased is what sets experiences related to interpersonal connections in grief apart from depression where experiences of relating to others typically do not occur.

21Fuchs proposes two complementary ways of reintegration with the deceased. One way involves various forms of identification (or incorporation) of the lost person. The mourner learns to preserve and incorporate the persisting presence of the deceased in the memory, instead of searching for it outside. The mourner may also at later stages start exhibiting some of the traits of the deceased in their own behavior. Another way of transforming the relation with the deceased involves various forms of representation. Fuchs mentions culture-specific memorial rites, commemoration days etc., as typical iconic and symbolic forms of representing the deceased that can preserve the continuing relation to them (p. 58). Both types of resolving the experienced ambiguity share the acknowledgement of the loss and provide ways of regaining the relation to the loved one.

24I would now like to propose that the disturbance of the above sketched processes of reintegration may be a key component explaining the occurrence of complicated grief. The dynamic of grief typically implies that reintegration processes, i.e. those that lead to acknowledging the loss and finding ways of re-establishing the relation with the deceased, are at some point initiated and proceed with time (e.g. Klass et al., 2014, Boelen et al., 2006). As a result of this observation, I suggest that the divide between uncomplicated and complicated grief could rely on whether, when and how the grieving person enters the so described reintegration processes, where by re-establishing the relation with the deceased they can resolve, to some extent, the experienced, paralysing ambiguity of their presence and absence. Uncomplicated grief can turn into complicated one, when the work of grief or the typical progress in the reintegration is either not initiated or largely disturbed at some point. Based on these observations, I propose that the stasis experienced in many cases of complicated grief, as described by Ratcliffe, has a particular kind of source or etiology. It is as a result of the disturbance in the processes of reintegration.

26Without medicalizing either uncomplicated or complicated grief, the divide between the two can be drawn in terms of disturbances in the typical dynamics of grief where the processes of reintegration involving various forms of experience and their intensification occur. Phenomenological analysis is crucial for uncovering the presence or absence of individual experiences involved in the process of reintegration and for assessing possible disturbances. This is because phenomenological inquiry is particularly suitable to investigate individual experiences that concern: the mourner’s acknowledgement of the loss, the way in which the deceased is represented by the mourner and the kind of relation that is established after the loss, the emotions that are associated with the continuing bond with the deceased (e.g. Klass et al., 2014). Given the heterogeneity of such relations and grief experiences, in each and every case we might need to investigate the specific dynamics of reestablishing the relation with the deceased in the process of grief. Although disturbances in reintegration processes may take various forms and involve a variety of experiences, we can still see them as the underlying structural feature of complicated grief.

27A natural question to ask this point is when in the course of time we should expect the signs of reintegration processes and when we should be alarmed by their absence and suspect complicated grief? The answer, unfortunately, is far from straightforward. Since the dynamics of grief may depend on various contextual factors, such as mourner’s psychological resources to deal with the loss, the relationship they had with the deceased or the cultural context, this proposal does not provide an answer to the question about the typical timeframe of the reintegration processes involved in grief. Different constellations of these and other relevant contextual factors may place different requirements for the reintegration processes. For example, a mourner who has a propensity for depressed mood and had a particularly complicated relation with the deceased may face different challenges in the process of grief than a mourner who was relatively close with the deceased and could prepare for their loss by following their illness for some time. Grief may be a genuinely universal phenomenon, but the work of grief and experiences involved in readjustment processes may vary quite a lot. Different constellations of contextual factors and different challenges seem to suggest that specific reintegration processes may require different timeframes. For this reason, using a single time threshold to draw the divide between complicated and uncomplicated grief may be artificial and uninformative.

5. Conclusions

29I have discussed recent research in the phenomenology of grief and depression (Ratcliffe, 2017; Fuchs, 2018) and suggested that a phenomenological approach may be particularly useful for the task of drawing a divide between uncomplicated and complicated grief. I have argued for a new account of complicated grief and suggested that a promising way to draw the divide is to look for the presence of reintegration processes geared towards acknowledging the loss and establishing a new relation between the mourner and the deceased. The proposal is a starting point for a much needed, full account of complicated grief.