Queen Victoria was 42 years old when her husband, Prince Albert, died of typhoid. She went into seclusion for many years after his death and went through a period of considerable unpopularity with her subjects. She wore black for the rest of her life, and refused to wear the crown or robes of state, but wore a humble bonnet that symbolised widowhood. She constantly kept a picture of her husband at her bedside, and constantly grieved his loss. On her own death some 40 years later, she was buried along with his dressing gown, photographs, and a plaster cast of his hand.
In this section we will examine an extreme though normal variant of sadness that almost everybody will experience at some time: grief. In fact, it may be the individual's failure to show emotion following a major loss or trauma that is the key to an atypical outcome (e.g., Parkes, 1972), although it is clear that the situation regarding expression of emotion in grief is complex (Kennedy-Moore & Watson, 1999). The importance of sadness and grief in both childhood and adulthood can probably best be understood in terms of attachment theory (e.g., Bowlby, 1988), although it must be noted that later work has broadened this focus to emphasise more general aspects of stress and coping (e.g., Stroebe, Stroebe, & Hansson, 1993). One of the key innate systems is that of attachment to the primary caretaker; threats to this relationship are experienced with considerable distress by the child. There is also evidence that problems can develop early on in attachment style and that these problems may be associated with later psychopathology; thus, so-called anxious attachment, disordered attachment, and avoidant attachment styles may be associated with a range of later adult attachment problems.
In an attempt to understand both the lengthy nature of the normal grief reaction and the even lengthier nature of abnormal reactions, studies of children's separations from their primary caretakers provide a wealth of information. Bowlby's (1980) excellent summary of this work showed that the child initially goes through stages of protest and despair because of the separation, but eventually, if the mother (or other primary caretaker) returns, the child may treat her as if she were a stranger. In contrast, a non-primary caretaker (perhaps the father) may be greeted with warmth and relief over the same length of separation. Bowlby's interpretation of these different reactions is that the child reacts to the separation or loss of the mother eventually by "defensively excluding" or inhibiting the negative emotions. Even with very short separations of a few minutes studied in the laboratory, for example, with the "Strange Situation Test" (Ainsworth, Blehar, Waters, & Wall, 1978), a proportion of children as young as 12 months show ambivalent positive and negative reactions to the mother's return.
A second related area from work with children focuses not so much on whether children express ambivalent feelings, but on whether they can conceptualise such ambivalence. Harter (1977, 1999) for example noticed that many children she saw clinically were unable to admit to ambivalent feelings particularly towards primary caretakers. In her subsequent work with normal children, she found that only at about 10 years of age can children acknowledge and describe these ambivalent feelings (Harter & Buddin, 1987). The following quote from Bowlby captures the same idea:
In therapeutic work it is not uncommon to find that a person (child, adolescent or adult) maintains, consciously, a wholly favourable image of a parent, but that at a less conscious level he nurses a contrasting image in which his parent is represented as neglectful, or rejecting, or as ill-treating him. In such persons the two images are kept apart, out of communication with each other; and any information that may be at variance with the established image is excluded. (1980, pp. 70-71)
This statement from Bowlby concurs remarkably with our proposals for SPAARS of two routes to emotion generation, one of which may be conscious and the other automatic, and which may lead to the generation of contrasting or conflicting emotions (Power & Dalgleish, 1999). In relation to grief, the conclusion that we draw from these findings and our previous proposals is that the loss of the main attachment figure whether in childhood or adulthood leads to not only the emotion of sadness, but frequently also anger at that person for abandoning the individual to an uncertain fate. However, the expression and conceptualisation of such ambivalence requires a developmentally sophisticated level of maturity that many individuals may fail to negotiate or may only partially do so; thus, the individual may feel sadness following such a loss, but feel extreme guilt about feelings of anger or, psychologically, go one step further and idealise the lost person so that no feelings of anger could even be imaginable towards such a perfect individual. In addition, as noted above, the combination of pressures on an individual in Western and other cultures to inhibit the expression of both sadness and anger following loss may all lead to the result that grief runs an atypical course (Parkes, Laungani, & Young, 1997).
In terms of our general SPAARS model presented in Chapter 5, the experience of extreme grief consequent on the loss of an attachment figure can be seen as the major loss of mutual goals, roles, and plans that the loss of attachment figures entails. Because of the evolutionary base to attachment, the universal experience of grief across cultures must in part have an innate basis and must therefore involve operation of the direct access route to emotion within SPAARS. However, the impact of the loss of a significant other is so wide-ranging on the individual's life that multiple and continued appraisals will accompany any automatic reactions. For example, studies of the impact of bereavement in our culture show that the nature of the impact and its consequences may differ for widows and widowers; Wortman, Silver, and Kessler (1993) reported that widowers were particularly vulnerable to limited social relationships and problems with taking on tasks that their wives had handled, whereas widows were more vulnerable to financial strain following the death of their husbands. The net effect of these and other problems is that bereaved individuals may make appraisals that they will be unable to cope with the practical and emotional burden that they have been left with. In addition, if the bereaved appraise the grief reaction itself as weak or not allowed, because of a more general rejection of negative emotions, then a more atypical course may be likely for grief.
This atypical course may be more likely to occur, we would speculate, if the coupling of the basic emotions of sadness and anger occurs, as shown in Figure 7.4, given that many grief reactions have in addition to the primary emotion of sadness other appraisals that lead to anger. The anger may be directed at the lost individuals themselves, be directed at others who caused the loss or did not offer enough to prevent the loss; anger is more likely to occur where deaths are sudden and unexpected (e.g., Vargas, Loye, & Hodde-Vargas, 1989) and may thereby explain in part why sudden and unexpected deaths may lead to more chronic or difficult grief reactions, though not always (Stroebe & Stroebe, 1993). We argued above and in Chapter 5 that once such couplings of activated basic emotions occur, they may reciprocally activate each other and thereby prolong the emotional state. In the case of grief, however, we must also emphasise the internal and external pressures to inhibit not only the expression of sadness, but especially the expression of anger towards the lost individual. The greater the attempts to inhibit one or both of the emotions of sadness and anger involved in grief, therefore, the longer the course that the grief reaction is likely to run. The loss of the main attachment figure whether in childhood or in adulthood involves nothing less than a redefinition of the self; an individual who prevents, whether consciously or unconsciously, such a process from occurring, who attempts to deny the loss of the key other, will be left with a model of the self that is maladaptive, inaccurate, and out-of-date (e.g., Parkes, 1993).
In summary, we propose that the experience of grief can run a number of possible courses depending on a variety of factors. First, the cultural and familial pressures that inhibit the expression of sadness in men ("Boys don't cry"), and anger in both men and women ("Don't speak ill of the dead"). Second, the developmental history of the individual in which early childhood ambivalence towards the primary caretaker may have been outlawed, because the adult was, for example, unable to
contain the child's anger. Third, the nature of the relationship with the lost individual; that is, the degree to which the relationship was secure or ambivalent. Fourth, the type and suddenness of the loss, for example the degree to which an individual has been able to prepare for the loss of an ageing parent, versus the premature and unexpected loss of a healthy child. And, fifth, the quality and the nature of the support from significant others in the individual's network: "We don't mention her, so as not to upset him." We will now briefly examine some of the empirical studies of bereavement in order to see in more detail how these factors influence outcome.
Most studies of bereavement show that it is a long process and that even after 12 months only approximately half show reasonable recovery (e.g., Glick, Weiss, & Parkes, 1974; Parkes, 1972), with 2 to 3 years being the more typical timescale. The Glick et al. (1974) study included both widows and widowers, and demonstrated that the emotional reactions such as yearning for the lost person, intense sadness, emotional isolation, and anger were similar in both groups, as was the rate of recovery. However, more widowers were found attempting to exert control over their sad and angry feelings, although they also evidenced more self-reproach initially and found their energy and ability at work considerably reduced. Stroebe and Stroebe (1993) in a study of younger bereaved adults reported that at 4 to 7 months after the loss 42% of their sample met Beck Depression Inventory criteria for at least mild depression; 2 years after the loss this figure had fallen to 27%, a value considerably greater than would normally be expected. Indeed, Rubin (1990) found that an especially painful loss, that of a parent losing a child, may lead to sadness being unabated some 13 years after the loss. Rubin also found that it may be harder for older parents to come to terms with loss than it is for younger parents. Whereas the younger parents may eventually come to develop new plans and goals, older parents of older children have fewer options, for example to have additional children to replace the lost child, or to come to value some other role or goal in the place of the lost child. More recent work, summarised in the excellent Handbook of Bereavement Research (Stroebe, Hansson, Stroebe, & Schut, 2001) has questioned the assumption of "grief work"— the idea that all grieving must go through a necessary series of stages after which the individual has to let go of the deceased individual. Instead, more complex models are needed that look at possible interactions between attachment style (e.g., Parkes, 2001), the type of relationship to the deceased especially where this has been positive rather than negative or ambivalent (e.g., Bonanno, 2001), and the type of coping methods engaged in for coping with grief (Folkman, 2001), factors that Stroebe, Schut, and Stroebe (2005) have integrated into a model of bereavement. The importance of what has been called either "traumatic grief" or "complicated grief" (e.g., Prigerson & Jacobs, 2001) must also be noted because, as we argued earlier in Chapter 6, traumatic reactions traditionally seen to be fear-based can in fact be based on other basic emotions such as sadness, anger, and disgust (Dalgleish & Power, 2004b). The case of traumatic grief highlights what we would argue is a non-fear-based post-traumatic disorder following bereavement and which may occur in up to 20% of bereaved individuals (Prigerson & Jacobs, 2001) especially if the death was unexpected and violent (Kaltman & Bonanno, 2003; Murphy, Johnson, & Lohan, 2002).
Of course not all losses involve loss through death, but losses also happen through separation and divorce both of which have their own psychological costs. Losses can also be less tangible when, for example, they involve the loss of an ideal;
nor do they necessarily involve another person when, for example, they involve the loss of a limb, or the loss of capacities through chronic illness. Such examples highlight the fact that although we have emphasised the interpersonal context in which losses occur, reactions to losses that are not interpersonal must also be considered.
Consider first a complex case that involved the loss of an ideal, though in the context of key interpersonal losses:
Maria was the fourth born of six children, all close together in age. Her father died suddenly and unexpectedly when she was 5 years of age, though she remembered her childhood as extremely happy. Her mother, however, had had little time for her after her father's death and had clearly endured considerable hardship in keeping six children on her own.
Maria met and married her husband when she was 18. Although he had not been her first steady relationship, she was soon sure that he was her ideal match. They had two daughters within the first few years of marriage and Maria stayed at home and devoted herself to her husband and her children. She described herself during this time as a completely happy person, great fun to be with, with the ideal family. She was unable to understand why her friends' relationships were never as good as hers and, equally, she and her husband were the envy of everyone they knew.
Maria came into therapy with her "ideals shattered". She had discovered about 15 months previously that her husband had slept with her best friend, once by his account in a moment of temptation. So she forgave him. However, a year later she arrived unexpectedly at her friend's flat to find her husband there also. Her friend and her husband had carried on an affair for well over a year.
Maria felt overwhelmed by emotions "that weren't hers". She found herself in absolute rages, even though she was someone who had never become angry "even in situations where other people did". She would burst into uncontrollable floods of tears, and so now avoided seeing her other friends or going out, because of her "complete personality change". The thing that she was the most furious about was that he had taken away the perfect relationship and destroyed the perfect family. Needless to say, they were still living together.
Maria provides a dramatic example of someone who to all intents and purposes was going through a double bereavement, even though her husband still lived with her and her former friend still lived in the same town. What she had lost, though, was the idealised relationship and the idealised family that she had created for herself, repeating the pattern of the early loss and subsequent idealisation of her father. The child who had to hide her own grief and be strong and care for others might well never have had to face up to such, but for the failure of others to play their parts in her melodrama. All of Maria's major goals and plans were mutual ones that she believed she shared with her husband and her best friend. In many ways therefore, the affair violated her basic assumptions about the world (Janoff-Bulman, 1992; Wortman et al., 1993), and about the roles that loving partners and best friends should play in that world. Maria's continuing air of shocked disbelief at what her husband and her friend had done to her world demonstrated the strength of the schematic models that she had created, presumably partly in response to the early loss of her father and her consequent caring for her mother at the expense of her own needs. The price of her devoted caring was not returned by her spouse or her best friend. (See also the trauma section in Chapter 6 for more detailed discussion of "shattered assumptions" about the world.)
In conclusion, grief is a powerful and universal reaction in particular to the loss of attachment figures. The experience of loss leads to intense sadness and, in addition, a myriad accompanying appraisals that revolve around the loss of mutual goals, roles, and plans, together with bereaved individuals' appraisals of whether or not they will be able to cope without the lost person. Theorists such as Bowlby (e.g., 1980) who have proposed that the grief reaction follows a series of stages have also emphasised the angry protests that the bereaved individual typically goes through. Although recent theorists present more flexible models with, for example, the possibility that stages overlap (e.g., Shuchter & Zisook, 1993), nevertheless anger is a very frequent accompaniment of grief. We have speculated therefore that, in terms of the SPAARS model, the duration and outcome of grief will depend on whether basic emotions such as sadness and anger become coupled and thereby sustain each other. In order to explore speculations such as these, it will be necessary to consider preexisting characteristics of individuals and their typical styles of coping with emotion. It will also be necessary for bereavement researchers to distinguish more carefully between bereavement and depression rather than treat one as if it might be the same as the other. As Clayton (1990) found, bereavement depressions show lower levels of hopelessness, worthlessness, and loss of interest in friendships, which would be expected if the basic emotion of disgust is an essential part of depression, but not of bereavement. Moreover, Clayton's comparison of a group of recently bereaved individuals and a group of matched community controls showed near identical levels of worthlessness, with 14% in the bereaved and 15% in the controls, even though the overall rates of "depression" were 47% for the bereaved versus 8% for controls. We will examine these differences in more detail in the next section.
Was this article helpful?