Introduction
This article brings together data from both sides of the Atlantic to explore how widowers tell the story of their wife's death. It grew out of a symposium on the topic at the annual meeting of the Gerontological Society of America in 2009. The authors of two papers discovered that their findings overlapped to a surprising extent. The surprise arose because the two authors had previously found differences in the ways in which widowers established their relationship with the interviewer (Bennett Reference Bennett2007; van den Hoonaard Reference van den Hoonaard2010). There were also different social contexts – Atlantic Canada is rural whilst the British data were collected in urban centres and the small group of American widowers was comprised of Jewish men. It was also surprising, although to a lesser extent, because the two authors come from different disciplines – van den Hoonaard is a sociologist and Bennett is a psychologist – and because the authors use different analytic approaches – van den Hoonaard uses symbolic interactionism and Bennett uses grounded theory. As a result of the similarities in our data and in our interpretation of the data, we decided to analyse the widowers' narratives together systematically to ascertain the characteristics and meaning of the ways they told their stories.
Although losing one's spouse is not an infrequent event in later life, it is one which carries with it profound implications for both personal and social life. Late-life widowhood challenges a person's sense of identity, which hitherto has been bound up with that of his or her spouse, and challenges a person's role in society, from married person to a person on his or her own. Most previous work has focused on women (Bennett Reference Bennett2010; Chambers Reference Chambers2005; Lopata Reference Lopata1996; van den Hoonaard Reference van den Hoonaard1997). More recently, scholars have begun to explore how widowhood affects men. They have found that, similar to women, men have to renegotiate their identities to maintain their sense of themselves as real men (Bennett Reference Bennett2007; Moore and Stratton Reference Moore and Stratton2002; van den Hoonaard Reference van den Hoonaard2010). One way in which people establish the kinds of people they are is through telling their stories to others (Becker Reference Becker1997; Chambers Reference Chambers2005). Research interviews provide an occasion for the telling of stories particularly in the relating of the events that led up to a spouse's death (Bennett and Vidal-Hall Reference Bennett and Vidal-Hall2000).
In response to the manner in which our research participants approached the interview situation and how they narrated their experiences, we explored the concept of masculinity, the ways in which widowers worked to maintain their sense of being real men, and the ways in which they reinforced their sense of masculinity in relationship to the interviewer. van den Hoonaard (Reference van den Hoonaard2009) reports that men used a variety of strategies such as taking control of the interview and using endearments when speaking to the interviewer. Bennett (Reference Bennett2007) found that men presented three types of masculine role, drawn from Brannon's (Reference Brannon, Brannon and Davids1976) work: no sissy stuff, the sturdy oak and the big wheel. Simultaneously, but independently, we began to explore the ways that widowers reinforced their masculinity when talking about the specific events leading up to their wife's death, and in particular in the death narrative.
Relatively few scholars have written about identity and masculinity in the context of widowhood (see Moore and Stratton Reference Moore and Stratton2002). Moore and Stratton (Reference Moore and Stratton2002) mention briefly that widowers have agentic orientation (Bakan Reference Bakan1966) in relationship to jobs and volunteering. They also discuss the care-giving undertaken by widowers before becoming widowed in the light of being in control, a typically masculine quality. The widowers, to whom they spoke, presented a stoic identity and emphasised being in charge of their wife's care.
Although few researchers have studied widowerhood, in recent years there has appeared a small, but growing, literature around masculinity and ageing. Previously, much of the gerontological literature had asserted that men and women became androgynous in old age. This literature had, therefore, treated old men as if they had no gender and as if masculinity did not play a role in their lives (Davidson, Daly and Arber Reference Davidson, Daly, Arber, Arber, Davidson and Ginn2003a).
However, more recently, scholars have begun to address masculinity and old men in a variety of areas. Arber, Davidson and Ginn (Reference Arber, Davidson and Ginn2003), for example, have highlighted the importance of examining how men maintain masculinity and autonomy in later life, while Thompson (Reference Thompson2006) has suggested the importance of examining how older men construct gender scripts. As Alan K. White (Reference White2006) and Edward Thompson Jr (Reference Thompson2006) remind us, many aspects of men's lives remain unexplored. White argues that, even in the area of men's health, few studies have addressed how masculinity relates to men's beliefs and behaviours. Similarly, Addis and Mahalik (Reference Addis and Mahalik2003) argue that we need to explore the challenges of masculinity for men facing difficult transitions.
The newer work on men and ageing often addresses challenges old men face in maintaining an identity of manliness, of being a ‘real man’, in the face of age-related changes. Men's ageing affects many aspects of their identities including how they understand their place as masculine men. Coles and Vassaroti suggest that older men operate in a ‘field of aged masculinity’ (2012: 39). Here the criteria do not force older men to compete with younger ones. Instead, they are allowed to identify themselves with dominant aspects of masculinity and their sense of themselves as masculine men. Similarly, Meadows and Davidson have shown that men develop strategies to come as close to hegemonic masculinity as they can get. These strategies include illustrating what they can ‘still do’, ‘being better than others’ at ageing, and ‘exaggerated notions of [their] heterosexual prowess’ (2006: 302–4).
One way that older men maintain their masculine identity is by resisting entering feminine spaces, whether in organisations geared toward older adults (Davidson, Daly and Arber Reference Davidson, Daly and Arber2003b) or by seeking help (Addis and Mahalik Reference Addis and Mahalik2003). Care-giving and widowhood both represent a feminised space that presents particular challenges. In fact, Russell (Reference Russell2007) suggests that it is only recently that researchers have begun to take men's care-giving seriously. He emphasises the tasks that male spousal care-givers must accomplish and the ways in which they approach those tasks from a masculine point of view. Russell argues that there needs to be greater focus on the subjective meaning of masculinity.
Black et al. (Reference Black, Schwartz, Caruso and Hannum2008: 178) note that personal control and masculinities are ‘intricately linked’. They found that when a care-giver is an elderly husband, by exercising autonomy he demonstrates that he is competent both as a husband/care-giver and as a man, and his self-definition approximates the masculine norms of ‘agency and independence’. This expression of personal control or being able to influence events or results or to have an influence on what happens allows older men to continue to see themselves as real men. Thompson (Reference Thompson, Kramer and Thompson2005: 29) argues that a man who cares for his wife is both doing gender and doing care-giving, and in doing so his identity becomes that of care-giving husband. According to Spector-Mercel (Reference Spector-Mercel2006), there is at present no broadly recognised cultural script for being both a real man and an ageing person. This absence leaves older men, in general, and widowers, in particular, to communicate their masculinity by developing their own scripts. Calasanti and King (Reference Calasanti and King2007) identify how the strategies that men use in their care-giving illustrate masculinity. They reveal a number of techniques men use including the blocking of emotions, focusing on tasks, self-medication and exerting force. Their work highlights men's descriptions of care-giving and in particular how the men's descriptions differed from those of women and where they explicitly referred to manhood or womanhood.
These papers focus on the tasks and strategies undertaken by men, but they do not analyse how the men talked about their experiences (see also Meadows and Davidson Reference Meadows, Davidson, Calasanti and Slevin2006). In contrast, Kirsi, Hervonen and Jylhä (Reference Kirsi, Hervonen and Jylhä2000) who examined written accounts of care-giving husbands about their experiences, found that older men wrote in particular ways about caring for their wife who had dementia. They identified four types of, what they described as, speech repertoires: factual, agentic, familistic and destiny. In factual speech, the narrator (or in their case the writer) adopts the ‘identity of an observer’ and a reporter of events (2000: 156). In agentic speech, the narrator presents himself as being in control of his life and his environment and as an independent actor. In familistic speech, the narrator tells his story in terms of his family obligations and his commitment to his wife. Finally, in destiny speech, the narrator tells of events in which he was ‘at the mercy of forces that [were] outside his control’ (2000: 163).
Following from their earlier paper, Kirsi and colleagues later included both written and oral accounts of care-giving in their analysis (Kirsi, Hervonen and Jylhä Reference Kirsi, Hervonen and Jylhä2004). They found that the men used a wider range of voices in the interviews than in written accounts of their care-giving. Thus, even more than a written story, an interview provides an opportunity for a widowed person to tell his or her story to another in person.
There are three reasons why we are interested in the way the story is told rather than only in what the participants do, in relationship to masculinity. First, by the time of the interview, the event has finished, that is the wife has died. Second, the events are inevitable – whatever the man does, his wife will die. Finally, the way in which stories are told allows a widower to present himself to an audience. As Anthony Giddens (Reference Giddens1991) has explained, a person relies on the narrative of his biography to maintain his identity, in our case, an identity intertwined with widowhood and masculinity.
Other scholars have discussed the reasons why people use narrative. McLeod (Reference McLeod1997), working from a counselling perspective, suggests that narratives convey meaning, a sense of self and the emotional context. Becker (Reference Becker1997) also suggests that the use of narrative allows the teller to maintain continuity in situations of disruption (and becoming widowed is one form of disruption). Bennett and Vidal-Hall (Reference Bennett and Vidal-Hall2000) argued that the use of narrative form is significant (not in the statistical sense) because narratives are frequently rehearsed. This rehearsal is important for at least two reasons. First, remembering the narrated events is an end in itself, and second, because these stories are frequently related to other people. This article extends the study of narratives of death to men.
In Bennett's earlier work, narratives differed in a number of ways from the rest of the interview. The intonation changed; there was a shift from the general to the particular; a shift into the past tense (Jakobson Reference Jakobson1971); and a suspension of turn-taking rules (Jefferson Reference Jefferson and Sudnow1972). van den Hoonaard (Reference van den Hoonaard1999) found that widowed women frequently gave detailed narratives about the deaths of their husband. Chambers (Reference Chambers2005) identifies other types of narrative associated with widowhood, and Moore and Stratton (Reference Moore and Stratton2002) focus on content rather than the modes of telling of the death stories of their widowers. To our knowledge these are the only other studies that have identified death narratives told by widowed persons.
Thus, we draw on three literatures, those of care-giving, masculinity and narrative. We argue that the narrative of death allows a widower to signify his masculinity, and when he agrees to talk about a life event, particularly one about which he might become emotional like the death of his wife, the threat to his identity as a real man is significant (see van den Hoonaard Reference van den Hoonaard2009). We show that each widower uses the interview opportunity to explain and make sense of his wife's death and the roles he played during the dying process and to do the identity work that allows him to reinforce a ‘personal identity’ that is congruent with being masculine (Snow and Anderson Reference Snow and Anderson1987). We demonstrate, using Kirsi, Hervonen and Jylhä's (Reference Kirsi, Hervonen and Jylhä2000) typology, that by using narrative, widowers are able to maintain their masculinity in situations that present an enormous challenge to their masculine role.
Method
The data that form the basis of this analysis come from two independent studies of older widowed men conducted in England (one in the East Midlands and one in the North West) and from a third independent study of older widowers conducted in Atlantic Canada and Florida, USA. The first and second studies explored the experiences of older widowed men (and women) with an emphasis upon understanding the emotional and participatory changes that occurred following spousal bereavement (see Bennett, Hughes and Smith Reference Bennett, Hughes and Smith2005a, Reference Bennett, Hughes and Smith2005b; Bennett and Vidal-Hall Reference Bennett and Vidal-Hall2000). The third study focused on the social meaning of being an older widower and its impact on everyday life (see van den Hoonaard Reference van den Hoonaard2010). In total, 86 widowers aged between 55 and 98 participated. Participants were all resident in the community, living in their own homes or in sheltered accommodation, and were recruited via a diverse range of formal and informal groups run for or by older people, media, personal contacts and social service departments (for discussions of selection issues, see Bennett, Hughes and Smith Reference Bennett, Hughes and Smith2005a; van den Hoonaard Reference van den Hoonaard2010).
The studies used interviews that were non-prescriptive and semi-structured as the aims of the original studies were to learn from the widowed people what was important to them using the approach of ‘We are the novices and you have the experience’. The first two studies asked about life before widowhood, around the time of bereavement, one year after and currently. We wanted to know what respondents did and how they felt at these different times, so we used prompts such as ‘what did you do?’ and ‘what did you feel?’ (see Bennett and Vidal-Hall Reference Bennett and Vidal-Hall2000). The third study, from a symbolic-interactionist perspective, focused on social process and what it meant to be a widower (see van den Hoonaard Reference van den Hoonaard2010). Similar to the first two studies, the interviewer encouraged participants to tell their story in their own way. It broadly asked participants to explain what it was like to be a widower and how various aspects of their lives had changed.
In all studies, respondents gave consent, and the research teams tape-recorded the interviews that ranged from three-quarters of an hour to three hours. Most took place in the men's homes. Interviews were transcribed verbatim.
Analysis
In the original studies, Bennett and van den Hoonaard used different analytical approaches. Bennett used a grounded theory methodology (see e.g. Bennett and Vidal-Hall Reference Bennett and Vidal-Hall2000; Charmaz Reference Charmaz, Smith, Harré and Van Langenhove1995; Glaser and Strauss Reference Glaser and Strauss1967). van den Hoonaard used a symbolic-interactionist approach that seeks to understand the world from the standpoint of those being studied. Hence, she used a thematic approach to analyse the interviews and identify the meanings and concepts the participants raised (Becker Reference Becker, Jessor, Colby and Scheweder1996). Once she grouped the data into themes, she continued with line-by-line coding. Through this inductive process, the strategies involving speech repertoires the widowers used to tell the stories of their wife's death became apparent.
As we have outlined in the introduction, we developed the current paper from two back-to-back presentations that presented widowers' narratives about the death of their wife that were extraordinarily similar. In that presentation, van den Hoonaard found that her data fit well into the framework of Kirsi, Hervonen and Jylhä (Reference Kirsi, Hervonen and Jylhä2000), who identified four types of speech used by care-giving men in written accounts of their care-giving experience: agentic, factual, familistic and destiny. Although Bennett and Vidal-Hall (Reference Bennett and Vidal-Hall2000) used a more restricted definition of narrative, and in particular a requirement for five or more lines of continuous text, in this analysis we chose to include all stories that men provided of their wife's death, even if relatively brief. We did so because, firstly, our experience of interviewing suggests that men are often more terse in their utterances than women, and secondly, because we were interested in the content of these utterances, however brief. The shortness of the story can add intensity to its telling. For example, ‘She was diagnosed on Monday and on Friday she was dead’, is very short, but the abbreviated construction communicates the power of the experience. Hence, we included narratives that meet the criteria offered by Rubin and Rubin (Reference Rubin and Rubin1995: 24) that stories or narratives are ‘based on remembered history … events as they happened step-by-step’ and have the quality of having been told before (even if only by the widower to himself). As a consequence, 79 of the 86 widowers gave death narratives.
To analyse our data for this article, we returned to the narratives of death in our interviews and categorised them with respect to the four types of speech. In the first instance we categorised our own interviews and then we considered the other person's interviews. We found that there was overlap in some narratives that met the criteria for more than one category, and we also found that there were two new sub-categories of both destiny and familistic speech that we illustrate.
Because our data come from three studies, when we quote from our participants, we identify each man with the study he came from. Thus men from van den Hoonaard's study are identified by pseudonym + NA (North America); those from Bennett's East Midlands' study are denoted by pseudonym + EM (East Midlands of England) and those from the North West of England by pseudonym + NW (North West of England). We felt that it was important to distinguish between these studies, so we could demonstrate that the findings were relevant to both sides of the Atlantic.
Results
In our analysis of the combined interviews, we were able to identify stories which reflect all four of Kirsi, Hervonen and Jylhä's (Reference Kirsi, Hervonen and Jylhä2000) narrative types: factual, agentic, familistic and destiny. In addition, we discovered a sub-set of familistic speech and one of destiny speech. The sub-set within familistic speech reflected widowers' self-blame if they believed they had let their wife down as a husband. The subset within destiny speech concerned medical negligence, where the men considered themselves or their wife to be a victim of forces beyond their own control. We discuss each of these in turn, illustrated with examples from the three studies.
Factual
In the factual mode of speech, the speaker provides, ‘objective and neutral information about the events’ leading up to and surrounding his wife's death. He adopts ‘the identity of an observer and reporter’ (Kirsi, Hervonen and Jylhä Reference Kirsi, Hervonen and Jylhä2000: 156). This mode of speech often includes causal arguments and medical vocabulary. It allowed the widowers to maintain the unemotional demeanor and appear knowledgeable which are consistent with hegemonic masculinity (Schwalbe and Wolkimir Reference Schwalbe, Wolkimir, Gubrium and Holstein2001).
In some cases, factual speech took the form of very brief reports of when the men's wife died:
It was 10 to 10 on the Tuesday mornin’ when she died. The 8th of October. (Bernard EM)
She died in '84 on the 8th of November. No the 4th she got cremated on the 8th. 1984, that's the 8th of November 17 years ago. (Gregory NW)
She had colon cancer. And she was in the hospital when she died. (Tim NA)
Although quite brief, these stories are notable for the specificity of the events the widowers referred to which communicates an impression of factual reporting.
Other participants provided factual reports that included lengthy, explicit descriptions of symptoms that led to eventual terminal diagnoses:
How it came about was from her neck downwards; she had no movement at all. If her hand was there and it fell down there, it stopped there 'til I was … to pick it up. Her head used to fall forward like that … One of the things that happened on Thursday morning that she suddenly couldn't swallow anything … On Wednesday morning about six o'clock in the morning, we got a phone call from the hospital saying she'd had a heart attack and she survived that. When I got to the hospital, they were stood all round her bed and she was stretched out and after that she was out of the game. (James NW)
At Mother's Day a year ago … she was suspicious at the time that she had ALS [Amyotrophic Lateral Sclerosis] … it wasn't confirmed until the 28th of June. [The] neurologist thought it was carpel tunnel, but ALS … you just lost strength, you can't do anything. A lot of patients waste away, but she didn't … She was diagnosed on the 28th of June, and she lasted four months and ten days. The average is 18 months. (Herbert NA)
These longer narratives that use factual speech demonstrate the men's rational understanding of what happened. They provide detailed, chronological accounts that focus attention on their ability to provide accurate reports and on their knowledge of the illness from which their wife suffered and died.
Agentic
Another way that the widowers demonstrated mastery was through the use of the agentic mode of speech. This section demonstrates how our research participants used it to further emphasise their masculinity.
This mode of speech allows the widower to ‘adopt the identity of an independent actor’ (Kirsi, Hervonen and Jylhä Reference Kirsi, Hervonen and Jylhä2000: 161). Here the widower took centre stage in telling the story of his wife's death. For example, some men emphasised the role they took in encouraging their wife to see a doctor or in persuading their doctor to do more tests. The following excerpts illustrate this:
I made her go to the doctors. (Fred NW)
So of course I immediately got her into W hospital and she lived there five weeks. (Terry NW)
Her doctor, at first he thought she had this allergy … he thought it was just the allergy getting worse. After a time, I didn't think so. And I had a conversation with him [and] he sent her for these x-rays … then she started to develop a headache… . I thought it was connected … because they'd found a spot on her lungs … I thought it had spread … So I talked to the doctor about that… . And the cancer specialist said that [my wife] … probably would only have six months, at the most, six months to live. (Winston NA, emphasis added)
Although these men were not able to control the outcome, their contributions to the eventual diagnosis became the centre of the story as we can see by their repeated use of the word, ‘I’ in their stories.
Agentic speech is also noticeable in the way some men talked about the caring work they did while their wife was ill. We already know that men are quite willing to care for their wife when she requires it (see e.g. Calasanti and Bowen Reference Calasanti and Bowen2006; and Davidson, Arber and Ginn Reference Davidson, Arber and Ginn2000). Similar to the men to whom Stratton and Moore (Reference Stratton and Moore2003) spoke, several men in our studies described the caring chores as simply work that had to be done. Rather than focus on personal care, which they surely provided, these men foregrounded tasks that involved traditional masculine skills. This supports Kirsi, Hervonen and Jylhä's (Reference Kirsi, Hervonen and Jylhä2004) research that identifies the absent or scanty reports of personal care. For example:
And I put up ramps here and the other end. And the wheelchair could go up the ramps. So that made it pretty good. (Herbert NA)
The agentic mode of speech the men used to talk about the care they gave their wife sometimes combined with familistic speech which highlighted their identities as husbands who were ‘responsible care-giver[s]’ (Kirsi, Hervonen and Jylhä Reference Kirsi, Hervonen and Jylhä2000: 157). For example, Leroy (NA) straddled the two categories:
And then … I had everything all fixed up here for her before she died. I put in that new septic system back there. I bought that barn … bought that new oven … did a lot of work in the kitchen … She was living then, of course.
In this story, Leroy shows his love for his wife as a husband while demonstrating agency in these masculine skills.
Familistic
Participants used familistic speech to explain their actions as well as to talk about how their wife handled being ill. First, we talk about what the men said they did as husbands.
The men underlined their closeness to their wife through placing themselves in the position of a husband. For example:
It came quite sudden during the night. I heard her crying during the night and I woke up and I couldn't wake her up. I couldn't wake her up at all so I called an ambulance and they took her into hospital and she died the same day. (Harold NW)
As well, some men highlighted the fact that they knew their wife better than anyone else and took care of her in a way no one else could:
She went into chemotherapy, and I took care of her for the year or two … I think I extended her life by making her eat and drink, watching her. I had a programme for her medication, and I even drove her to S. to visit her family. And I think I prolonged her life by attending to her …. And I was there when she did pass away. I stayed with her nights. And I was there when she passed away. Right, right, I was right with her. (Al NA)
I insisted that I stayed in with her to the last … I thought that we'd been together so long that I thought I could look after her as good as anybody else. (Billy EM)
These men insisted on being with their wife in her last days and highlighted their role as husband in staying with their wife until the end.
Some men used familistic speech to talk about how their wife fulfilled her role as wife when she was dying by continuing to care for her husband as much as possible. Winston (NA), for example, explained that his wife preferred being in the hospital because she felt that taking care of her was too much work for him:
[It was] her idea to go to the hospital. 'Cause she thought I was rushing around too much trying to help her, and that's what she said.
Similarly, Giles (NW) described how his wife continued to carry out her role as his wife as much as she could:
If she was fit enough for me to cook, she wouldn't let me. As a matter of fact she cooked dinner at 12 o'clock on the day she died. We washed up, she said make a drink of tea and we'll watch telly. Just put the cup by the side of her here, went and sat down, just shouted at me and she'd gone – so quick. (Giles NW)
In a few cases, participants also talked about things their wife did as a mother. For example:
[Wife speaking] ‘do you mind youse going outside while I talk to the boys’. She started off with the eldest one … and said, ‘John look after your Dad’. ‘Yes Mum…’ (Roy NW)
She had many conversations with my daughter that I wasn't privy to until later on … My wife told my daughter that she would like to be buried [rather than cremated] … But I didn't know until afterwards. (Jacob NA)
In this mode, some men expressed appreciation for mothers' instructing their children in how to relate to their father, but others were disappointed that their wife confided in their children more than in them.
Within the familistic discourse, some participants invoked the love they and their wife had for each other. Kirsi, Hervonen and Jylhä identified these expressions of love as the ‘locus of emotion’ to underline that, in their study, men did not express their emotions haphazardly; rather they ‘play an important role in the [narrator's] identity’ (2000: 160). In these examples, the participants call attention to their identity as loving husbands:
Actually she died in my arms really. And her last words she said ‘I'm frightened’. (Daniel NW)
But she did tell me that her last words to me were, ‘I love you’. (Jack NW)
She wasn't hard to care for. She wasn't a complainer. If she'd even done something to make me mad once in a while, [it] wouldn't have been so hard missing her. (Herbert NA)
There were no major problems in the marriage, but there was some forgiving to be done on both sides. So we did that. But mostly it was happy talk about what we'd done, places we'd been. (Grant NA)
Just as the widowers used the familistic mode of speech to emphasise the relationship they had with their wife as a husband, they also used it to identify regrets they had in terms of not fulfilling their marital roles properly. In these cases, they locate self-blame in what they ought to have done. In contrast to men who had been with their wife when she died, others regretted that they had not been with her. For example:
I wish I'd stayed with her. (Ken NW)
I feel recriminative to myself for not staying when I should have done. (George NW)
One time she went out of her head because of the whole medical situation … She was in another world … And I spoke to the nurse … and I said, ‘You know, I'm going to go home.’ So the nurse said, ‘Please go home.’ And I felt so guilty; even though you know, I couldn't communicate with her … I couldn't do much with this mental business. (Ed NA)
Others regretted not taking their wife's illness seriously enough or being afraid to deal with the situation head on. For example,
They're all little things I look back on and I think to myself well she must have known, you know [but you didn't talk about it?] oh no, no, no. You know back at the hospital, I was a bit of a joker, I used to joke and bluff. (Dave NW)
We didn't talk about death at all, you know. I knew she was dying, and I think she knew it, too. We more or less stayed off the subject. You know, as if she were going to get better. But I knew she wouldn't get better. (Tim NA)
One widower expressed his regret that, as a father, he did not tell his children soon enough that his wife was dying even though it was a decision he and his wife made together:
We made one mistake. G. didn't want the kids to know how sick she was until it was necessary… . And [finally] I called them and told them, and of course they came down. [By then] she was in the hospital. (Charles NA)
A few men used the unusual strategy of blaming themselves for lack of courage in facing their wife's impending death. For example, Keith (NA) noted that he was afraid to talk with his wife about her dying:
I was a big baby in that respect. I never said anything. Maybe that was why she didn't mention it. She never mentioned it again. She was very positive all the way through.
Similarly, two men in the UK sample expressed regret that they did not address their wife's prognosis directly:
I've regretted it ever since that I didn't take the doctor into another room and said look what's wrong. (James NW)
I had a terrible feeling, really awful feeling that Mary was going to pass and she was fighting, fighting, and I could hear her fighting and I was a coward [this was a premonition]. (Jeremy NW)
In some cases familistic speech was combined with destiny speech, when ‘[the narrator] is at the mercy of forces that are beyond his control’ (Kirsi, Hervonen and Jylhä Reference Kirsi, Hervonen and Jylhä2000: 163). Colin (NW), for example, invokes his love of his wife alongside the inevitability of his wife's death, when he says:
She just simply fizzled away; she was a lovely, lovely person. (Colin NW)
Destiny
Kirsi, Hervonen and Jylhä (Reference Kirsi, Hervonen and Jylhä2000: 163) suggest that destiny speech is the final mode in the repertoire. It appears most frequently in the death story after the singular moment of the diagnosis of terminal illness. Once the widowers got to the point in their narrative when there was nothing else they could do to affect the final outcome, they tended to gloss over the details. These participants described the process as ‘downhill from there’:
Things just went downhill after that. We knew it was terminal right then, and she went right downhill and was in the hospital. (Angus NA)
Worser and worser and worser until she [was] finally diagnosed with cancer in May '95 and died in October and that was all in this house. (Gordon EM)
Some widowers switched from agentic speech to destiny speech when they got to the period after diagnosis, when it was ‘all over’ even if their wife lived for quite a while longer. Ralph (NA) provides an example of this:
By this time, I was parked right alongside her bed in my little [walker] … and then she just slumped right over against me. And the female orderly said, ‘Mrs. S., Mrs. S.’ I said, ‘For God's sake, never mind the Mrs. S’ I said, ‘Hit the button!’ I said, ‘She's not going to answer you; she's unconscious.’ … So anyway, it all ended up with God in His own way, looked after everything because she got what she wanted. She didn't want to go into a nursing home. (Ralph NA)
Others used destiny speech to express the feeling that there was almost no identifiable cause to their wife's death. It just happened:
Although I had a premonition, it came as a shock to me. (Peter NW)
Well it was a mystery. (Brian NW)
We had high hopes in the beginning. And the wound just would not heal … There was never a chance of a cure, but there was a 20 per cent chance of a temporary remission so we decided to take the chance. (Ed NA)
In these narratives, the participants highlight both the unexpectedness of their wife's death and its inevitability. A more negatively charged situation involved medical neglect or error.
Medical blame
Several participants' stories included episodes in which medical professionals either misdiagnosed or under-played their wife's condition. For example, Chad (NA) explained:
Either the doctor didn't tell her the truth or lied to her or she did the same to me. Because with the very same stack of X-rays … that I took to ___ with her and saw the oncologist down there. Her face, her chin dropped right to the desk when she took a look at it. And then I knew something was wrong, but it was too late then.
Other men provided stories that entailed medical professionals' avoiding conversation when the news was bad:
I'm very unhappy about [how he and his wife were treated]. I said, keep still [because the doctor never stopped moving] I want to have a word with you about my wife, and he said, ‘Oh no’, and that's the nearest I came. (Bert NW)
… and I was walking through the door, this side ward door and there was seven, six or seven nurses there and they were all sort of congregated round her and they looked round to look at me as if to say, ‘what the hell are you doing here? Get out, you shouldn't be here’, and as I walked into the room the sister er Deb turned round she says, ‘Oh I'm sorry, Mr Long, er Mary oh I'm sorry, Mrs Long has just passed away.’ (Andrew NW)
In these situations, the husbands thought that their wife's death may have been the result of medical mistakes, and they saw themselves and their wife as victims of this circumstance.
Discussion
With very little prompting, older widowers describe the events which led up to the deaths of their wife. Whether these narratives conform to Bennett and Vidal-Hall's (Reference Bennett and Vidal-Hall2000) initial definition or whether they are quite terse, these stories have much to say about how widowers interpret the events leading up to the death of their wife, how they tell their stories and how they maintain continuity as real men through the recounting of these narratives. As Becker (Reference Becker1997) has observed, varied discontinuities, such as becoming a widower, require individuals to construct narratives that allow them to maintain a sense of continuity that includes core values of manhood such as taking responsibility and exercising judgement. These stories fit very well into the modes of speech discussed by Kirsi, Hervonen and Jylhä (Reference Kirsi, Hervonen and Jylhä2000): factual, agentic, familistic and destiny.
As we have noted in the introduction, few authors have examined the narratives of death (Bennett and Vidal-Hall Reference Bennett and Vidal-Hall2000; Chambers Reference Chambers2005) and few others have examined masculinity and widowhood (van den Hoonaard Reference van den Hoonaard2009, Reference van den Hoonaard2010; Bennett Reference Bennett2007; Moore and Stratton Reference Moore and Stratton2002). Moore and Stratton linked widowhood with care-giving, and, in turn, care-giving with masculinity. However, whilst Calasanti and King (Reference Calasanti and King2007) and Russell (Reference Russell2007) both focus on care-giving and masculinity, their approaches are to focus on tasks and strategies employed to maintain masculinity. Kirsi, Hervonen and Jylhä (Reference Kirsi, Hervonen and Jylhä2000, Reference Kirsi, Hervonen and Jylhä2004) are, therefore, the only researchers to attend to how care-giving men tell their stories as a means to demonstrate their masculinity. In this paper we take their approach to understand how men demonstrate their masculinity in the narratives they tell of the death of their wife. Thus, the work we present in this paper is unique.
One of the most interesting features of this analysis was how similar the data from both sides of the Atlantic are regardless of the diversity of urban and rural settings. It appears that English-speaking widowers talk about the death of their wife in the same way, and the typology outlined by Kirsi, Hervonen and Jylhä (Reference Kirsi, Hervonen and Jylhä2000) fits both groups of men. It is not the case that one category was more prevalent amongst one group of men. We found excellent examples from both groups for each mode of speech.
The four modes of speech, which helped tie the strands of the men's stories together, are not monolithic. Research participants used familistic speech, for example, to express how they and their wife conformed to their marital roles, but they also used it to express regret when they did not succeed in doing all they might have done as husbands. Hence, even if they did not have a heroic narrative to tell, the research participants were able to use familistic speech to underline their identities as husbands.
Similarly, the men used destiny speech to illustrate that they were the victims of fate but also to bring attention to the mistakes or omissions of medical personnel that made them feel like victims. Perhaps the laying of blame outside themselves provides a sense of comfort and justice to these widowers. These distinctions illustrate the potential of the analysis of modes of discourse to address the diversity widowers use to establish their sense of self through the stories they tell about their wife's dying and death.
There are some limitations to this study. First, the data were not collected specifically with a view to analysing widowers' narratives of their wife's death. It is, therefore, remarkable that these narratives come across so strongly and in such similar ways from three separate studies collected in different parts of the world. Second, if we were to apply the text-length constraints of Bennett and Vidal-Hall (Reference Bennett and Vidal-Hall2000), we would have had fewer narratives to analyse. There were, for example, 48 of 60 interviews which had longer narratives in Bennett's dataset. Even using this tighter definition, the majority of men do have death narratives. Nonetheless, we should be aware of the power of the short, staccato-style story. Third, the original data were collected within different methodological traditions and asked different questions of the data.
The limitations of this study highlight the strengths of the inductive approach to research. Not only has this article applied a very useful typology, but it also suggests modifications that became apparent through a systematic analysis of data. These additions allow for a more nuanced analysis of narrative data. The results also demonstrate that the experiences of widowers and the ways widowers talk about their wife's death are not confined to one country, or to one culture. The analysis shows that English-speaking older widowers share experiences across the North Atlantic and use common narrative forms to describe the events surrounding their wife's death. Further, our collaboration provides an example of how researchers from different disciplines, i.e. sociology and psychology, and different continents, can work together to shed new light on an important subject.
We are just at the beginning of the study of widowers' narratives of the death of their wife. We were struck that the findings of studies in urban England, rural Canada, and with urban, Jewish Americans were so similar in the way the men told their stories. Studies of the death stories of widowers from non-English-speaking cultures would allow us to see whether widowers in those contexts would share similar modes of speech in their narratives.
Acknowledgements
Thanks are due to the men who participated in these studies and to the people who assisted with data collection. We are also grateful to the small community of researchers who study widowers and always encourage each other's work. Thanks are due to the anonymous reviewers whose careful comments allowed us to improve this article. Kate M. Bennett wishes to thank Georgina Hughes, Philip Smith and Steph Vidal-Hall. She also would like to thank the Department of Human Communication, De Montfort University for support. This research study was also supported by the Economic and Social Research Council (Award No. L480254034) and was part of the Growing Older Programme of 24 projects studying the quality of life of older people. However, the findings reported here are entirely the responsibility of the researchers. Deborah K. van den Hoonaard would like to thank the Social Sciences and Humanities Research Council of Canada which provided a Standard Research Grant (2001–2004, Grant No. 410-2001-0014) in support of the research. She also is grateful for the support of her colleagues at the annual (Canadian) Qualitatives and the Gerontology Department of St. Thomas University.
