The task of time in retirement

Abstract

Retirees' encounter with time has long interested social scientists, especially the negotiation of such an open-ended status. Pursuing theoretical suggestions that daily activities anchor a narrative of self-identity, this project examined the coherence of retirees' representations of their time use. Information is drawn from interviews with 30 retirees in the Midwestern United States of America who were invited to discuss their daily lives and activities. The retirees valued time sovereignty and accounted for their time use by describing schedules of activities in some detail. Daily time was not presented as improvised but rather as structured into routines. Recurring behaviours flowed from situations and structures in which people were implicated, such as body care and living with others. Even in replies to a specific question about the preceding day, people slipped into language about what they typically do. Retirees' ready narratives about routines were also accounts of who they are not. Our findings suggest, first, that daily routines are instrumental for retirees in economising thought and behaviour. Second, the assertion of a routine is an assurance that one's life is ordered and proceeds with purpose, thus solving the task of time. Third, routines can be a means to signal conformity with ideals of active ageing.

Introduction

The experience of retirement is fundamentally about dealing with time. Retirement essentially relieves workday obligations on one's time while offering, for better or worse, the potential for increased sovereignty over personal involvements, activities and pastimes. Whether relief from work was wanted or unwanted, there is nevertheless time to fill.

Retirees' encounter with time has long interested social scientists, especially the negotiation of such an open-ended status. In an early characterisation of retirement, just as it was emerging as a widespread practice, Ernest Burgess (1960) deposited a memorable but dark view into the literature. Modern social trends, in a ‘series of blows’, had reduced the status of older people in the economy, the family and the community, and relegated them to retirement. ‘[T]hey found themselves cursed instead of blessed by leisure time in abundance and little or nothing to do with it’ (1960: 20). Retirees are ‘imprisoned in a roleless role’ with no vital function to perform. Burgess admitted that he might be overdrawing the contrast between traditional and contemporary societies, but maintained that retirement was a typically empty status nonetheless. ‘The increase of leisure time presents a problem, first, of how to fill it and, second, of how to use it for self-expression and for community service’ (1960: 21). In saying this, Burgess recognised that the use of time (doing) would be important to post-work identity (being) in this roleless role.

Burgess was not alone among contemporaries in detecting a problem with the leisure time afforded to retirees (e.g. Kleemeier 1961). Here was an emerging social status with few expectations compared to the adult involvements in work and family. One of the major latent functions of work is a time structure on the waking day. Writing about another context, unemployment, Jahoda (1982: 22) observed that when work's time structure is removed, ‘its absence presents a major psychological burden. Days stretch long when there is nothing that has to be done; boredom and waste of time become the rule’.

Burgess painted retirees as devalued and resigned to their marginality. Yet as things developed in subsequent decades, retirees were not generally discontented with their status (Barfield and Morgan 1978; Fouquereau et al. 2005). This remains true of today's retirees. The Health and Retirement Study, an ongoing survey of older Americans, asks retirees: ‘Would you say that your retirement has turned out to be very satisfying, moderately satisfying, or not at all satisfying?’ Responses among persons who were completely retired and aged 65–74 in 2010 showed 52 per cent saying that retirement has turned out to be very satisfying, 39 per cent found it moderately satisfying and only 9 per cent were not satisfied at all (authors' calculations). ¹ A similar distribution of responses was seen in the 2008 survey. The responses of men and women were virtually identical.

In addition, the ‘leisure time in abundance’ that Burgess disparaged is, to the contrary, valued by retirees. Long-form interview studies identify time sovereignty as one of the chief advantages of retirement in the eyes of retirees (Savishinsky 2000; van Dyk et al. 2013; Weiss 2005). ‘The freedom of retirement is not only a freedom from the need to respond to work's demands. It is also, notably, the freedom to engage in new activities, thoughts, and experiences’ (Weiss 2005: 74). At the same time, such interview studies also note a double edge to the freedom of retirement. According to Weiss, ‘all retirees face the same two challenges of retirement: to manage its threat of marginality and to utilize its promise of freedom’ (2005: 14). Van Dyk et al. (2013) identify competing ‘story lines’ among retirees about the liberation from employment: one is free but not free to do nothing. Said one woman: ‘The great freedom is actually what is most difficult’ (2013: 104).

By one means or another, time is occupied. There has been extensive research on what retirees do and how much time they spend doing it (Gauthier and Smeeding 2003, 2010; Putnam et al. 2014). For example, detailed information has been available from the American Time Use Survey (ATUS), an annual series based on telephone interviews with large, nationally representative samples (http://www.bls.gov/tus/home.htm). Individual self-reports about the preceding 24 hours are coded into categories such as employment, household work, religious activities, volunteer activities, leisure and sports, sleep, grooming, eating and others, including even finer coding of these general types. Distributions of time use can then be compared by other measured characteristics. Researchers such as Krantz-Kent and Stewart (2007) have used these data to show, for example, gender differences in the amount of time engaged in household work by retired adults. ATUS data also confirm the time re-allocation that retirement entails: more time freed up for leisure and sports, chores and sleeping. In addition to time-use surveys with their comprehensive partitions of the day, other studies of older adults focus on the self-reported frequency of particular activities, such as volunteering, socialising, exercising or care-giving (Dorfman 2013; Horgas, Wilms and Baltes 1998; Morrow-Howell et al. 2014).

To the extent that they have time sovereignty, people at any age, as a form of ‘time work’ (Flaherty 2003), can go on to organise activities into recurring sequences, into routines. There are various disciplinary perspectives on the instrumental or adaptive nature of routine activities. From a psychological perspective, routines meet a trait-like need for personal structure and reduce cognitive load (Neuberg and Newsom 1993). From economics, temporal routines can be seen to streamline individual choice for production and consumption (Hamermesh 2004). In occupational science, routine behaviour is seen to promote health and wellbeing (Clark 2000).

The instrumental nature of routine has also been theorised in sociology, in Giddens's (1991) work on modern life and self-identity. With modernity having eroded traditional social orders in which identities could be lodged, the individual is forced into an ongoing ‘project of the self’. In order to sustain a coherent story or stories about the self, the individual must choose or adopt some lifestyle, ‘a more or less integrated set of practices which an individual embraces, not only because such practices fulfil utilitarian needs, but because they give material form to a particular narrative of self-identity’ (1991: 81). Lifestyles are ‘routinised practices’ and small decisions each day about how to conduct oneself contribute to such routines. ‘All such choices (as well as larger and more consequential ones) are decisions not only about how to act but who to be’ (1991: 81). The routines that individuals follow manage life as normal and predictable, thereby maintaining a sense of ‘ontological security’. In the flux of modern environments and without institutionally ascribed roles to fall back on, the composition of activities across everyday life can anchor stable identities.

Giddens's view on lifestyles and the self echoes that of philosophers such as Aristotle and Dewey regarding habitual acts and character (Cohen 2007). There is also a consonance with Burgess's (1960) characterisation of the roleless role of retirement. Both Burgess and Giddens, each in his own way, see people confronting an open-endedness that can nonetheless be managed by the adoption of lifestyles. As regards retirement, our focus here, the task would be to settle its rolelessness with purposeful uses of time.

Quantitative data about activities and time use in retirement can be informative about average parcels of time per activity, the nature of a synthetic day, and some of the contributing factors and consequences of time allocation. Such data are less informative about the ways that retirees regard their use of time, particularly in negotiating a path between the threat and promise of so much of it. Beyond discrete activities, is there any sort of coherence to people's representations about their time use that, following Giddens, is also a possible representation of one's self as a retiree?

This is a qualitative study of such representations that first checks the meaning of retirement as time freedom and then explores the instrumental character of talk about activities. Previous qualitative studies encourage this direction, having found that older adults (mainly women) use routinised activities and schedules to manage impressions of themselves (Laliberte-Rudman 2002), compensate for loss of former professional roles (Price 2003) and maintain a continuity of identity, though with more relaxed routines (Ludwig 1997). In this report, we will describe features of the ways that retirees present their time use, not merely as lists or schedules of behaviours, but as idealisations of their conduct and themselves.

Methods

Our analysis is based on interviews completed as part of the Altern als Zukunft/Ageing as Future project (http://www.alternalszukunft.uni-jena.de/?lang=en), a collaborative, multi-method, multi-sample study of perceptions of time in later life being carried out in Germany, Hong Kong and the United States of America (USA). Here we draw upon interviews that were conducted in 2013–2014 with 30 individuals in the Midwestern USA. These participants were recruited by posting notices at senior centres and apartment complexes, in community newspapers and by word-of-mouth referrals. We screened volunteers to be at least 65 years old, self-identified as retired, working no more than 20 hours per week, community-dwelling and functionally able to participate in the interview. Prior to interviewing them in their homes or, in two cases, in public places, we told participants that we would be asking about daily life and activities in retirement.

The sample was recruited with sex and age quotas to reflect a range of experience in retirement. Of the 15 men and 15 women, ten were aged 65–74, 14 were aged 75–84 and six were aged 85 and older (age range = 67–97, mean = 76). The number of years since retirement ranged from two to more than 40, although the exact year of retirement is often difficult to pinpoint (Ekerdt and Deviney 1990). All of the participants were White. Five participants had never gone to college, three had some college education and the rest possessed bachelor degrees or higher. Marital status also varied, with two never married, 12 currently married, six divorced and ten widowed. Most did no work for pay at the time of the interview, although three participants (two men and one woman) had occasional part-time work and one man was regularly employed for five hours a week. Compared to the general population of Americans aged 65 and older, our sample was somewhat more likely to be male, older, more educated, less likely to be married and (by design) less likely to work.

Our semi-structured interviews typically lasted 60–90 minutes, although a few extended over several hours. The interview guide for this study was also used in parallel designs at the other international sites. The basic content of interviews centred on how retirement came about, life and activities since having retired, and expectations for the future. This analysis is based on three passages in the interview guide: (a) ‘To begin, what do you think when you hear the word retirement?’ (b) ‘What has retiring meant to your everyday life [or activities], day in and day out?’ (c) ‘May I ask you, what did you do yesterday? Would you be so kind and tell me everything you did yesterday from morning until evening?’ These questions were intended to generate talk about people's status as retired and their use of time as retirees. In the flow of these qualitative interviews, answers to these questions might be volunteered at any time, not only upon our asking.

Each of the authors conducted some of the interviews; none of the interviewers knew the study participants prior to interview. Whenever possible, participants were interviewed alone, although in a few cases spouses were present for at least a portion of the time. The sessions were recorded and transcribed, and in the excerpts that follow we have masked or changed some details to preserve confidentiality, including names. This research protocol was approved by the institutional review board of the authors' university (project number 20375). All ethical requirements of informed consent, confidentiality and protection from harm were met in accordance with US standards for research with human subjects.

In order to examine retirees' accounts of time sovereignty and use, the authors separately read through each transcript in its entirety, then met to discuss impressions, emergent ideas and preliminary categories that captured the focal topics. Then, in successive rounds of independent reading, we aimed to develop an appreciation of elements of expressed ideas, concerns and considerations in our respondents' own words. We formulated a preliminary set of thematic categories and illustrative examples about time sovereignty, time use and patterns of response to specific questions about daily activity. We returned to apply these to several interviews, met to discuss results, identified refinements needed to code categories then conducted coding of the transcripts using NVivo version 10 qualitative data analysis software (QSR International 2012). Concurrent with the selection and condensation of themes, we were alert to disconfirming instances and negative cases so as to guard against interpretive bias (Ryan and Bernard 2000).

Findings

Retirement means time sovereignty

After gathering some personal information, the first question of the interview schedule asked retirees what they think about when they hear the word retirement. The most common reply was something about time sovereignty. In 18 of the 28 interviews in which the question was asked, time sovereignty was expressed (by eight women and ten men) either as release from the obligations of working life, or the subsequent availability of time to do as one pleases. Retirement made people think: ‘not having to sign in to a job’; ‘not having work on my shoulders’; ‘freedom to pursue your own schedule’; ‘freedom to spend my time how I want to’; ‘being able to control your own activities and do what you want to do’. Some of the participants mixed this time theme with other sentiments, e.g. ‘I think freedom. Except lack of money’. Also a few referred to the leisure ideal of retirement but then added that this had not worked out in their case.

Participants whose responses to this first question contained no element of time sovereignty instead offered ambivalent or negative views of retirement (financial difficulties, care-giving obligations, loss of status); denied that retirement entailed discontinuity from working life; or answered by beginning the story of how they retired. (We did not pursue further coding of this theme later into the interview, but it is possible that these participants, too, elsewhere prized the self-determination of retirement.) Nevertheless, the prominence of time-freedom themes in our responses confirms other observations in the literature that the availability and control of time is a focal retirement topic (Moody 1988; Savishinsky 2000; Weiss 2005).

Navigating time sovereignty with routines

In response to our asking about what retirement has meant to everyday life, almost all of the participants accounted for their time use by describing usual or typical activities in some detail. These might be recurring behaviours that cycle daily, weekly or periodically. Participants talked of morning and evening rituals, meal practices, health regimens, household chores, reading, television watching and computer use. They had standing social engagements, meetings with community groups, regular interaction with family members, volunteer commitments and political work. Married participants not only talked about their own involvements, but also recited those of their spouses and spoke of joint routines.

Well, he knows my schedule. I know his schedule. On Monday nights we play bridge with one couple. On Wednesday nights we play bridge with another couple. On every other Thursday night we play games with another couple. So we have these things that we do. We have season tickets for the women's basketball games. We go to community theatre. We go out to eat occasionally. (Ingrid Harris, 77 years old)

I have tried to pretty much organise our lives to where we got something going on each day. Up until she broke her leg, here this spring, why we was delivering meals one day a week at the senior centre. And we go to the senior centre on Wednesday and Friday, just about all the time now, to see people there. I play pool. She does chair exercises. Mondays are pretty open. We don't have much going on Monday, or even Tuesday. Tuesday's a big day. We have a little club that meets up here at the grocery store in the morning where we eat breakfast and visit. That's been going on for, I think we decided, 15 or 20 years. (Ed Walker, 86 years old)

Like Mr Walker, participants frequently talked about their routines as intentionally constructed and maintained. George Heinz, an 85-year-old single man, recalled his initial encounter with the unobligated time of retirement:

Well, it was very strange at first because I was so used to a schedule. And you know, I suddenly woke up to the fact that, my God, the day is my own. It was strange at first. I had a lot of friends, still do, and I enjoy that. But you get used to the idea that you have to structure your own day. Which is good, and I love it. That's what I wanted, but nevertheless, it was odd … I mean I just kept on doing it. I just started, I thought well this is the new life. And this is how I'm going to have to make my own structure. And I did.

An accustomed routine, once under way, becomes valued for the way that it stabilises and economises daily life. Quinton Manning (74 years old) characterised his daily regimen of walking his dog, lunch, drinking coffee with friends and napping as ‘almost compulsive’. He explained:

I don't think I think about time, particularly. I think I'm in such a groove, you know, I just kind of – and I think that's part of the reason I'm in that groove is because then I don't have to think about anything. You know, about being somewhere, doing something. I just go along.

Not all routines are welcomed. When perceived as being externally driven, they are resented or resisted. Mr Manning went on to explain why he stopped going to exercise classes:

I quit that and I couldn't figure out why because I really enjoyed it. And then it dawned on me it was because it was getting to be the routine. I didn't – I was trying to go three times a week and trying to be there at 1:30 and all this. All of a sudden I said I don't want to do that, you know. Even though, it was a good example of liking to do something when I got there, but not having to, it got to be a chore to have to be there, so.

Of course routines are never entirely a matter of individual agency. At any age they bend to the temporalities of bodies, environment and the rhythms of others (Southerton 2013). Just so, rounds of activities by our retirees flow from recurring situations and structures in which they are implicated. Some of these are voluntary choices and some are external realities that individuals must accommodate. Routines are structured around body care, including morning and evening hygiene, exercise and medications. Household maintenance becomes an organised rotation of chores. Congregate dining standardises mealtimes for those who eat in retirement communities or senior centres. Newspapers arrive in the morning and are thicker on Sundays. Disciplines from childhood and the working past are carried forward to the present. Weekends are still a time for religious services, family gatherings or sporting events. Compromises must be made to accommodate a spouse's hobby or part-time work obligations. To own a pet is to surrender to a determined daily order.

I get up and I take Max out. Well sometimes being a young dog, he wants to do more smelling than doing whatever. And so some days it's playtime and some days it's get-the-business-done time. Well, on Tuesday I have Beginner Spanish lessons at 9:30 at the centre where I teach yoga. So he has to get his business done because I have to be gone and Larry was working yesterday. So I have to kind of arrange my schedule. So I get up, take him out, then have my breakfast, then get ready to go to Spanish. And I teach Spanish for an hour before I teach an hour of yoga. They wanted to do it, and I said, ‘I'll teach Spanish lessons if you'll do it according to my schedule.’ I don't have time to have Spanish lessons here and something there. I said, ‘I'll teach them before yoga.’ (Ingrid Harris, 77 years old)

Despite the fact that Ms Harris's time use is shaped by her dog, husband and volunteer commitments, she presents herself as being in charge of her schedule. Like Ms Harris, most of the participants described themselves as controlling their time even as they are embedded in broader temporalities. These influences on time use tended not to be viewed as constraints. Bodily and household care, accommodation to the regimens of others and organisational programmes were all narrative elements available to combine and aggregate into a reasonably detailed account about how they actively structure their time.

This volubility about routine, however, was not a topic for one woman and one man in the sample. Difficulties were so overwhelming that the recurring use of time was not much discussed at all throughout the interviews. Said Kay Meyers (71 years old): ‘There've been times when, in the last few years, just even in the last year, when I've just – about all I've been able to do is just (pause) do what I need to do every day. And sometimes I have trouble getting that done.’

Routines as stories about the self

As the interviews proceeded, our general questions about the retirement experience shifted to focus, literally, on yesterday: ‘May I ask you, what did you do yesterday? Would you be so kind and tell me everything you did yesterday from morning until evening?’ The specificity of the question about yesterday surprised some of the participants, who asked to be reminded what day of the week it was yesterday or who retrieved calendars to check on activities. All of the 29 participants who were asked the ‘yesterday’ question recalled at least some events of the preceding day, sometimes after interviewer prompts to keep the sequence going (And after that? And what was your evening like?). Only eight of the 29 (two women, six men) were wholly able to confine their answers to the events of the preceding day. The rest slipped over into language about what they usually do. As examples, these two individuals drifted from the specific to the typical in talking about yesterday.

[After returning home from a daily breakfast club] I'm not sure what I did yesterday, come to think of it. You know, if I have to run any errands, you know, pick up any meds, go to Walmart, or whatever. Could be any number of things. And then come home and read my mail. And look at the computer, on the computer and see what's happening. I can call up [online auction sites]. Then if I don't make any bids, which I wouldn't necessarily at this time, then you know, it's just cleaning up the house, puttering around like that. Maybe doing some yard work, if necessary. I really hate to do that. So I usually hire the mowing done. There are other things that I do that way. (James Salazar, 75 years old)

So that [yesterday] is a typical day. Get up at 7. Let the dog out. Cook breakfast. I am a news junkie … I read three newspapers a day, typically. Check Facebook. I try to remember to check it once a day, but sometimes if there's an issue I go back twice (laughs). So, and I take a nap. I have a sacred nap hour. Everybody knows not to come over in the afternoon from 2:00 to 3:00. And then I catch up on my reading. I have a glass of wine and watch my sacred News Hour. As my daughter says, you tell me what time of day it is and I'll tell you what my mother is doing. And then I check the emails one more time. Read 'til I go to bed at 9:30. (Teresa White, 72 years old)

The goal of this question about yesterday was not to record a time diary (in the style of the ATUS), but to generate further talk about activities. If our study had needed a detailed reconstruction of the preceding day, our question about yesterday would not have been terribly productive, having yielded many incomplete or general replies. But it was revealing. By turns, we heard that yesterday was typical in the way that it unfolded, or that its events were not memorable enough to divert one's narrative thread away from usual behaviour.

Are these reports about routine time use valid? From our interviews, there is no way to know how much of anyone's 24-hour day was organised as typical or usual – a few hours or most of it – but we respect people's assertions that their time is structured. Also we respect their assertions that routines actually help them to accomplish things, such as exercise and household maintenance. These involvements did appear to be largely self-imposed and flexible because retirees said that they felt free to excuse themselves from practices if the need or opportunity arose. At any rate, the facticity of people's routines is a secondary question. What we hear in these interviews are people's narratives about ordered lives that proceed with purpose and even discipline. The actual routines may change and even slow with bodily ageing – people say that they do – and we are merely sampling the present version. What endures is the presentation of oneself as comfortable and competent with time.

What is also notable about participants' accounts of their routines is how they correspond with the active-ageing paradigm that idealises the health benefits of ‘productive’ activities (Katz 2013). Our participants commonly mentioned exercise, volunteer work, social engagement and lifelong learning as regular activities. Over and above their descriptions of time use, more than half of the participants (12 women, five men) went on to endorse the quality of their lives as active, full and busy, saying such things as ‘I like to keep busy’ and ‘I am as busy as I have ever been’. The salutation of a busy lifestyle has long been observed about retirees (Ekerdt 1986; Roth et al. 2012; Tsuji 2005; van Dyk et al. 2013; Yen et al. 2012). Again, there is no way to know how busy people are, but in claiming to have an active life, people are aligning themselves with one of the universal goods and positive ideals of later life (Katz 2000).

Participants' ready narratives about routines could likewise be accounts of who they are not, or at least who they do not want to be perceived as being. Freedom to decide how to spend one's time may be seen as a benefit of retirement, but not an unconditional good. There is a risk of losing track of time, of falling into idleness (van Dyk et al. 2013). Descriptions of routines and the importance of structure seemed to assure the participants and their listener that, unlike some retirees, they were not passively letting time slip away. For example, Max Lerner (70 years old) described a weekly schedule he and his wife had established:

Well actually, Sunday and Monday are wash days. Okay, and everything gets, the bedding gets washed and things like that. So there's a regular schedule to make sure that gets done. And Monday all the toilets in the house plus she vacuums the upstairs. Tuesday is the goof off day. Today is the whole basement, and then she does her hobbies … Thursday this whole floor gets done. And Friday we go to Cedar Grove and celebrate.

The interviewer then asked: ‘So would that have been a risk, going into retirement and not having that [structure]?’ His reply:

Well if you don't create it for yourself, then without a job nobody's creating pressures on you. Or very few people are … You can go sleep all day if you're not careful. If you don't set up some discipline, you'll just while away your retirement without, you know, and basically at the end of the day wonder what the hell happened to it.

In response to the ‘yesterday’ question, six people (three women, three men) initially declined to talk about yesterday because it fell short of some personal standard for a good round of activities. ‘I'm embarrassed to tell you about yesterday.’ ‘Yesterday was a goof off day’, perceived as not as productive as usual. ‘Can I tell you about Monday? Because yesterday I went to the movies.’ ‘Okay. Yes, what was yesterday? Tuesday. Okay, I'm going back to Monday [because] things got cancelled.’ ‘Oh, you picked a terrible day, yesterday! That was a very, very lazy day.’ Ernie Green (82 years old) said, ‘Well, you caught me on a bad day.’ This exchange followed:

Mr Green: Yesterday I got up probably about 10 o'clock. I read the paper. I got on the computer to read emails, see what's going on. My son, daughter and son-in-law came over and planted my flowerpots, so I watered them. And then I sat on my butt and watched this soap opera, The Young and the Restless. And then I watched The Bold and the Beautiful. And then I got off my butt and went out and walked the sidewalks all around. Got my walk in. Then I came back and my wife said Janet had called. And that she and Gary wanted to cook hamburgers and wanted us to come over … So I sat at the computer and did some stuff. Then I sat down and watched PBS, and watched the Tommy Dorsey Band. And then we got in the car and drove over to … our daughter's house, and went ahead and ate.

Interviewer: So why do you think that's a bad day? You said: ‘You caught me on a bad day.’

Mr Green: Oh, it's a wonderful day!

Interviewer: It's a wonderful day, yeah.

Mr Green: People probably would think it was bad.

Interviewer: Why do you think people would think it's bad?

Mr Green: Because you just feel like you're bummed out. When you worked all your life, you feel like … you know.

Mr Green's sheepishness about his current lifestyle was consistent with his cultivation of a strong ex-identity (Ebaugh 1988) elsewhere in the interview when he was eager to talk about work-related accomplishments and prominent people with whom he has been connected. Earlier in the interview, Mr Green had criticised other retirees for failing to ‘get off their butt and enjoy life’. By characterising a day watching soap operas and visiting with family as a bad day, he revealed his discomfort with any similarity between himself and these other less-active retirees.

Discussion

The historian Thomas Cole (1992: xx) has asked, ‘Is there anything important to be done after children are raised and careers completed?’ The indeterminacy of the retirement status is a perennial issue in the literature. Retiring removes the time structure of work, but for what purpose? For the individual, the task of retirement time is not only to occupy it, but to do so in such a way as to establish some social identity. This is later-life's ‘project of the self’ (Giddens 1991) for which daily practices are at once choices for action and for being.

Our interviews confirmed that time sovereignty is a dominant, valued meaning of retirement for many. And how is that time occupied? The retirees that we interviewed spoke readily about their activities, but in a certain way, not as improvised or serendipitous but rather as regular and organised in routines. They convey the impression that they control their time, even as their routines necessarily accommodate other people and circumstances. The value placed on orderly lives was also seen in the reluctance of some persons to identify with the events of a particular yesterday (a lazy day, a bad day). This suggests to us that the idea of a routine is more than the chronicle of one's activities. Its value lies in smoothing out, in the telling, the potential disorder of daily life. So talk about retirement routine can accomplish what narrative does generally, which is to compose ‘the semblance of meaning and order for experience’ (Gubrium and Holstein 1998: 166).

The alacrity with which people could recite the regularities of their lives by time of day or day of the week, extending even to spouses' routines, suggests to us that retirees are practised at accounting for their time and the ways that they occupy it. Time is topical. We were not the first persons to have heard these things. If dealing with abundant time is a core task for retirees, then talking about one's activities should itself be routine in their conversational repertoires.

Are there other life circumstances when ‘leisure time in abundance’ is also a problem that is met with routines? Unemployment withdraws the time structure of regular work and is certainly an aversive experience (Klehe, Koen and De Pater 2012). Yet Berger (2006) found that job search workshops and programmes helped mitigate identity degradation by providing a sense of daily purpose and structure. Although disabling chronic illness is another circumstance that undermines the daily order and habitual modes of being, the reconstruction of the self is possible through new routines and habits (Charmaz 2002). However, the major distinction between such time-rich circumstances as unemployment and chronic illness, on the one hand, and retirement, on the other, is that the former roles are experienced as disruptions of adult life, whereas retirement is an age-normative status and its encounter with unstructured time is usually permanent.

Time use serves the sovereign self but it also contributes to cultural reproduction (Flaherty 2003). As our participants describe them, routines are instrumental for the individual in the ways that theory would predict (Clark 2000). They economise thought, simplify decision-making about behaviour and motivate healthy practices. They help make retired life stable and predictable. Routines set courses that can be taken for granted and lend the comfort of ‘ontological security’ to the ongoing ‘project of the self’ (Giddens 1991). Routines are instrumental in a second way by providing assurance. Assertions that one's days are ordered and regular are the antidote to any concern (by oneself or others) about the rolelessness of retirement and how it might induce passivity and decline (Ekerdt 1986; van Dyk et al. 2013). With routine, one projects competence and discipline in meeting the open-endedness of retirement. To be sure, some of the retirees in our sample went on to emphasise how one or another of their involvements was a primary feature of their lives, e.g. being a freelance writer, a community volunteer, a gardener, a bridge player. This is one way to say ‘who I am’. At the same time, the assertion of routine in general makes a more pervasive claim to be conducting oneself in a capable way, solving the task of time. Routines are instrumental in a third way, as a vehicle for conveying conformity with ideals of active ageing. A routine can be referenced to paint a picture of oneself as a busy, active person who is ageing well. The behavioural practices that could validate such claims were not observed in this study, but half of our retirees declared their lives to be ‘busy’. Activities, both actual and aspirational, can be woven together to report a lifestyle that signals to oneself and others that one is a member of the tribe of successful older adults and not among those whose retirement is spent watching television or sitting in a rocking chair (van Dyk et al. 2013).

Our study findings have limits, having been based on a small, regional sample in the USA. At the same time, the consistency of responses – about time sovereignty and the ready resort to narrative about typical activities – recommends the inference that there are shared cultural patterns in the way that American men and women account for their time as retirees. Another limitation is our sample's temporal remove from the event of retirement; all had been retired for at least two years or more. Thus, this is a report of how people are presently conducting themselves as retirees, but the important question about time use and immediate post-retirement adaptation is beyond the view of this study. Finally, we have reported retirees' accounts of time use that were generated in one particular setting: conversational interviews with a curious researcher. Such a setting may exert ‘narrative control’ over the accounts, the interviewees perceiving expectations for certain storylines about time and retirement (Gubrium and Holstein 1998; Pasupathi 2006). Perhaps in other settings (e.g. the doctor's office, meals with friends) people compose the narrative of daily life in other ways. These variations would shed light on the consistency among actual practices, how time usage becomes symbolic of one's self-concept and how identity is then communicated in different contexts through the telling of time-use narratives.

For further studies on the textures of retirement time, we suggest research on a theme that cross-cuts our observations, namely the way that retirees block or categorise time in a later-life analogue of the partition of work and leisure. Our participants talked about productive patches of their days and about hours spent goofing off, about diligent housework four days a week and relaxing on the fifth. Why do retirees give themselves ‘time off’ from a status that is nothing but time off? How do they protect their time off from other potential claimants? We noted earlier that people are implicated in situations and structures from which rounds of activities flow, such as body care, days of the week and living with others. For retirees, a higher form of ‘time embeddedness’ may be the general cultural rhythms of work and leisure that apply to all adults (Kohli 1988) and that retirees attempt to reproduce in their post-employment context. Such research may come to conclude that, whereas retirees are liberated from the constraint of work, they continue to measure themselves against its forms and standards.