TRANSITIONS TO RETIREMENT: PERCEPTIONS OF PORTUGUESE OLDER MEN

In this paper we present results based on the project Old Guys Say Yes to Community . We interviewed 90 men in southern Portugal between 60 and 93 years of age. Our main aim was to understand how older men experienced their transitions from work to retirement. We clarify the concept of transition and the models that explain transitions in the life course perspective. There are a number of factors to consider during the processes of retirement. Also, not much is known about the role of education and learning during the transition process. We try to contribute to the field, reflecting on the learning processes that result from this transition. Our findings show that work is one of the most crucial dimensions to consider in transitions to retirement (despite the fact that we identify many others). But education and learning do have an immense impact on the lives of older adults, either negative or positive.


INTRODUCTION
In the past decades, we have observed major demographic transformations in most European countries in terms of population ageing, with increasing life expectancy and the consequent changes in the later phase of life (Schmidt-Hertha & Rees, 2017). In Portugal, for instance, the ageing index has increased from 98.8 in 2000 to 153.2 in 2017, being one of the highest in the EU (European Commission, 2018), and it is expected to increase further in the decades to come (European Commission, 2015). Projections of the elderly population established in 2015 predict that in 2060, 34.6% of the Portuguese population will be 65 or older, with almost half of the older adults (46.5%) being 80 years old or more (European Commission -Doctorate-General for Economic and Financial Affairs, 2015). This scenario has led to increasing concern about the challenges posed by ageing, especially when we talk about transitions in late life.
According to Grenier (2012), the concept of transition became a central one in the late 1970s and early 1980s, but its use is not always clear, and it can be applied to multiple contexts. On the other hand, the concept of transition has evolved in complexity over the last few decades because of a multiplicity of motives (Fragoso et al., 2013). For example, it is arguable that a traditional, linear biography still exists. The course of contemporary life seems to have become more complex and has lost its former stable contours (Alheit, 1995). Consequently, in this paper, we will first try to clarify the concept of transition and the ones related to it, reflecting on the models that explain transitions in the life course perspective, with a particular focus on retirement. After all, retirement is considered to be a major life transition associated in public discourses with reduced economic productivity and a range of personal irregularities (Moffatt & Heaven, 2017).
We will see below that a number of factors influence transitions to retirement. Gender is one of them and it is safe to state that transitions to retirement may differ considerably between men and women (Duberley, Carmichael, & Szmigin, 2014;Tams & Arthur, 2010). In our case, we will only analyse men. This article results from an investigation carried out as part of the project Old Guys Say Yes to Community 1 . As a part of this study, 90 men were interviewed in southern Portugal between 60 and 93 years of age, from quite diverse contexts, social classes and cultural environments. Whilst the project aims to better understand the non-formal and informal learning of men in community settings, in this article, we will focus on the older men's transitions to retirement. Our main aim is to understand how these men experienced their transitions from work to retirement. Understanding transitions in late life includes comprehending the changing context of 'growing old' (characterised by demographic change and cultural transformation) and the importance of subjective interpretations of ageing (Grenier, 2012).

CONCEPTUAL DIFFERENCES IN TRANSITIONS: KEY CONCEPTS
To understand transitions in life, we must refer to Erikson (1959) and Levinson (1986), who conceived the life cycle in the context of social and developmental psychology. In this context, transition refers to models that describe different stages of maturation and development. The life cycle suggests continuity and regeneration, and has been applied to social and developmental models, although it represents restrictions according to stages (Grenier, 2012). The life cycle is among the most widely used concepts in social sciences: it may be claimed just to indicate temporality; it may be applied to initiate an analysis; or it may refer to the assumptions of research in developmental processes (O´Rand & Krecker, 1990). Rigorously defined, the concept refers to maturational and generational processes in natural populations.
The concept of life span, also used primarily in psychology, shapes the nature and timing of developmental tasks. In this sense, it is used mostly to refer to psychological development and is, therefore, quite limited when we are dealing with discussions about transition (Grenier, 2012).
In the 1960s, some sociologists claimed that patterns and stages of human development are not as universal as previously claimed (Cain, 1964, cit. in Hendricks, 2012. In their opinion, we cannot assume that life is comprised of fixed stages that inevitably and invariably explain everything that happens in the process. Also, people do not develop or grow old isolated from the conditions in which they are grounded (Riegel, 1975). This was the context of the appearance of the initial round of life course explanations. Sociologists, among others, formulated an alternative paradigm of life course analyses that emphasised cultural factors, social circumstances, and social interactions as essential to understanding the life course (Hendricks, 2012).
The concept of life course is usually used to refer to an overall trajectory across the entire period of someone's life (Clausen, 1986;Elder, Johnson, & Crosnoe, 2003). Established in the sociological perspective, this term is intended to capture the fluid and changing aspects of experience. Many approaches to the study of transitions and ageing claim to be attached to the life course, a concept that has become fundamental to the study of continuity, change and transition (Grenier, 2012). This is the broadest concept and can be used to represent a fluid pathway, one where changes and development are possible. It can be applied to place older people's experiences over the duration of a lifetime and this is why this concept provides the frame required in this paper to understand continuity and change in the transition to late life (Grenier, 2012;Hendricks, 2012).
Finally, the term late life is a more recent addition to the research of ageing and transition. Embedded in the life course perspective, this concept draws attention to the later parts of the life course. According to Grenier (2012), it refers to a large, quite unspecified period of life that characterises 'growing old'. In the past few years, it has been used in a variety of ways: to refer to people aged 50; older people as defined in terms of social benefits eligibility at 65+; and people in the 'fourth age'.
Recent research has demonstrated that the limits and expectations of transitions throughout the life course and in late life are changing. In the context of gerontology and policy, transitions tend to be associated with relatively fixed social roles, aged-based or social stages, or even marked by types of experiences (Grenier, 2012). It is thus quite common to understand transitions in the context of stages of development, with childhood, middle age and late life describing the expected stages and roles defined in relation to family, work, and society.
The models used to explain transitions most are assumed to be generic and fixed across the life course (Schlossberg, Waters, & Goodman, 1995). Models that do address adult ageing tend to focus on specific types of transitions occurring in quite distinct stages, such as retirement, grandparenthood or widowhood (e.g. Levinson & Darrow, 1978;Erikson, 1982;Levinson, 1986). In this paper, we will focus on the life stages of pre-retirement and retirement but take into account that a number of factors that happen during other life stages have important meanings for most individuals.

FROM WORKING LIFE TO RETIREMENT
Research over the past few years has indicated that moving into retirement is a major life transition, frequently conceptualised as a critical event that can affect the future in significant ways (Kim & Moen, 2002). From a life course perspective, retirement is seen as the transition from adulthood to older age, but the idea of a tripartite standard life course has been challenged (Schmidt-Hertha & Rees, 2017). To Szinovacz (2003), for example, retirement features multiple dimensions and can refer to an institution, a process, or an experience. As an institution, the term refers to social structures that regulate the abandonment of the labour market; as a process, it concerns decisions and patterns of the withdrawal and is considered a more individualised experience in terms of the age in which it occurs, the form it takes (gradual, voluntary, early) and the growing difficulties associated with it (Phillipson, 2004); as an experience, it refers to the multiple life changes that result from this process (Duberley, Carmichael, & Szmigin, 2014).
In this sense, retirement is no longer a single event that marks the end of paid employment (Wang & Shultz, 2010) but a process that changes in timing and duration (Marshall, Clarke, & Ballantyne, 2001). Over the past few years, the transition from working life to retirement has changed significantly in most European countries, with new perspectives being discussed and considered in terms of the different pathways to retirement. Quite recently, authors such as Künemund and Hahmann (2014) have talked about a change from the idea of a traditional model of a life course that considers three phases (learning in childhood and youth, working in adulthood, and resting in later life), to more flexible models that consider mixed phases of working, learning and resting.
Thus, retirement implicates adapting to changes in life contexts, and these changes can go on for various periods of time. Even if desired, it can be more difficult than initially perceived (Fonseca, 2012), although there is not a consensus regarding the impact of retirement (Kim & Moen, 2002). Different perceptions of retirement implicate different adaptive responses to the changing situations, exposing people to different states of vulnerability that interfere in an active ageing (Loureiro et al., 2015).
Work is a central dimension of our lives that functions as a powerful organiser of the remaining life dimensions. Identity and social adjustments are built according to work, which influences the building of various social representations of ageing (Vicente, 2007). Work also allows the conquest of a certain social status, feelings of autonomy and control. Work-life relations determine the course of the year (holidays), the organisation of the week (weekend, workdays), as well as of the day (working hours, time for family, leisure time) (Schmidt-Hertha & Rees, 2017). Going to work every day provides a routine and a structure, making people feel that they are a part of society. It is only natural that work and career status are key factors when we talk about identity development and this is mostly because people define themselves and live accordingly with their professions and their jobs (Schmidt-Hertha & Rees, 2017). Consequently, leaving the labour market can be a negative experience with significant consequences in terms of physical and psychological health, as well as social capital. In this sense, the transition to retirement may be difficult, in particular for people who perceive work to be an important part of their identity. Losing these structures can have a negative impact when it comes to people's feelings.
Another aspect refers to the importance of the workplace: it usually offers social inclusion by making individuals part of a team, sharing years and hours of the day with the same colleagues. For some adults, leaving work may at least imply the risk of losing social relationships, with the consequent shrinking of their social networks. Nevertheless, the transition to retirement can also be experienced as a relief when it means getting rid of obligations and stressful work (Reeuwijk et al., 2013), or limitations on personal freedom. When this happens, leaving work provides the possibility to embrace new opportunities to follow personal interests. In current European societies many people engage in voluntary activities, travel, or grandparenting -usually activities that promote psychological well-being.
Transitions to retirement can be navigated with greater success if older adults put some effort into planning their retirement phase. Preparation for retirement is therefore an important issue within transitions to retirement. Wang and Shultz (2010) consider that the process that leads to retirement begins with informal planning that progresses, with time, to more deliberate and formal planning. In fact, there is a significant variability in the degree to which individuals actively plan for retirement. Two models of in-between phases in transition from work to retirement have been considered in the literature: bridge employment (part-time work before retirement) and re-careering (second career after legal retirement) (Boveda & Metz, 2016).
The term bridge employment has been used to describe transitional work that occurs between full-time employment in a career and complete labour force withdrawal (Wang, Zang, Liu & Schultz, 2008). It can take many forms (e.g. part-time, temporary, seasonal), is typically of short duration (Feldman, 1994), and is more frequent among those with higher education (Wang et al., 2008). Re-careering or an encore career is considered a new phase of work after departure from career development. According to Feldman (2011), an encore career is personal and characterised by a purpose, passion and personal fulfilment. It was also found that women more frequently choose bridge employment, while men are more likely to embrace a re-careering situation. In fact, the transition to retirement may differ considerably between men and women, as well as their occupational pathways and careers, and their perception of this transition (Duberley, Carmichael, & Szmigin, 2014;Tams & Arthur, 2010).
Gender is, therefore, an important factor to consider in transitions to retirement. But health seems to be determinant in this context, and good health predicts working beyond retirement. On the other hand, a poor state of health may force workers to retire before their time. The financial capacity of people is central, both in terms of the decision to retire and the way individuals navigate transitions to retirement. Hence it is not strange that older workers with low incomes continue to work beyond retirement (De Wind et al., 2016).
There are a number of factors to consider during the processes of retirement. Attitudes towards retirement may play an important role, but so can financial constraints, the family context, and other individual obligations, as well as state or employer policies (Schmidt-Hertha & Rees, 2017). Also, not much is known about the role of education and learning during the transition process or about the way adults make use of educational compromises to project their transition to retirement. Schmidt-Hertha and Rees (2017) find that in what concerns the learning processes that result from this transition, research is quite inexistent. In this study, we try to contribute to the field by reflecting on this aspect.

METHODS
This paper seeks to contribute to a deeper understanding of transitions to retirement in older men. We chose a qualitative study since it is the most well-suited approach when we seek to understand people's experiences from their point of view. As Schwandt (2001) states: "Qualitative inquiry deals with human lived experience. It is the life-world as it is lived, felt, undergone, made sense of, and accomplished by human beings that is the object of study" (p. 84).
Semi-structured interviews were conducted with 90 men who agreed to voluntarily participate. The participants were recruited from day care centres, third age universities, welfare institutions, through references, and by encouraging snowballing. The participants' ages range from 60 to 93 and they live in rural (S. Brás de Alportel) and urban areas (Faro and Olhão) of the Algarve. The criteria for participation required that they were not institutionalised (in residential homes, for example) and aged 60 years old or more. The participant sample is diverse in terms of work situation (retired, active, unemployed), educational background, income, marital status (single, married or cohabiting, divorced, and widowed), and level of activity in civil society organisations (the type of role and level of participation in these organisations).
The interviews included questions concerning childhood (family background, schooling), working life, the transition to retirement and the current situation (health status, financial situation, family and social networks). Most interviews took place in the participants' residences. However, some were conducted in private spaces in the institutions of the civil society through which the participants were contacted.
Qualitative content analysis (QCA), as described by Schreier (2012), was used in order to analyse the data. Since it allows the defining of categories through an inductive procedure, that is, as emerging from the data, it seemed to be an appropriate method to capture the participants' experiences of the transition into retirement.
Given the diversity of the older adults involved and their specific situations, we separated the men into different groups using education level and rural or urban context of life as the criteria. We defined three groups (with approximately 30 participants each): 1) men from urban areas with a very low level of education (four years of schooling or less); 2) men from rural areas with a low/medium level of education (most with nine years of schooling, but some with secondary level schooling); and 3) men from urban areas with a medium/ high level of education (some with nine to twelve years of schooling, some with a tertiary education degree).
We then created a coding frame for each group, inductively, in a data-driven way. This means that the interview transcripts were read in order to understand how the older men experienced their transitions from work to retirement. This procedure allowed us not only to discover similar patterns within each of the groups, but also that the patterns were different when we compared the groups between each other.

Men from urban areas with a very low educational background
This group includes men with four years or less of schooling, and some of them are illiterate. They have a working history as non-qualified or manual workers, and most of them have a history of changing from one non-qualified job to the next, therefore with high levels of professional instability. A significant number of these men worked in the informal economy at least for parts of their lives, which has even further decreased their pensions, calculated on the basis of the numbers of years of contribution to the tax system.
These men have a very low financial capital that is revealed by their possessions: it is very rare that we find someone in this group possessing a driver's license and none has a car. Not all have a TV. None has used computers or other digital devices apart from a mobile phone. Still, not all possess a mobile phone and the ones who do make a basic use of it.
We found some men in this group who did not have a place to stay. Most of these men, therefore, live in very precarious conditions or even in poverty. Some of them do not even have enough to eat: they are dependent on the assistance of day care centres or similar institutions to get two meals a day.
These men's transitions to retirement were generally problematic, and this was due to multiple causes. First, some of them need to work even after retirement, simply because they need it to survive: some of the pensions they receive are below 200€, in a country where the minimum wage is about 580€ and renting a very small flat in a peripheral area costs at least 300€/month. But the majority of those who retire still have low financial capital due to a professional life of low salaries with no possibility of saving money or face the inevitable loss of income after retirement. Only a minority of these men have financial support from their families. Their low financial capital (which is primarily caused by a very low educational background) has a direct influence on at least two other life dimensions crucial to successfully navigating the transition to retirement: health status and sociability.
Although the national health system is basically free, some medical areas of expertise take time to access and a number of complementary diagnosis methods take time too. To be totally dependent on the public system of health can thus be problematic. Additionally, these men cannot afford the consequences of a severe health problem. If it happens, it is almost certain that their situation turns significantly worse. Health problems can be the cause of a forced early retirement or cause a deficiency. The former and the later further increase their poverty. Health problems can also dramatically reduce the men's mobility and constrain their ability to socialise. The fact that there are still important obstacles to mobility in most public spaces does not help. Just to give an extreme example, we interviewed a man who had not left his own small apartment for about two years because he was in a wheelchair and could not go out without assistance.
Socialisation in the public spaces of a city -marketplaces, parks, small neighbourhood cafés, etc. -is very important to these men because it means a basic connection to their friends and community events. These men are dependent on spaces to socialise because they do not use electronic forms of communication at all. However, some factors negatively impact their ability to socialise: health problems, reduced mobility or deficiency, and very low social capital (some men cannot even afford a cup of coffee and their shame excludes them from that option).
Social networks are therefore important in transitions to retirement. These men's social networks tend to be small in dimension and very unsatisfying. It is true that some have families, but others do not meet family members regularly: João, for example, does not even know where his two nephews live and the only contact with his sister happens when she takes him to the medical centre. Their contact with neighbours tends to be 'civilised' but not deep, and their number of friends is decreasing. Loneliness is a serious problem that affects most of the men in this group, usually deepened if they are widowers. For a small number of men, the death of their wives was a traumatic event that left profound psychological marks (depression or even suicidal thoughts).
It seems natural that education or learning (other than informal learning, at least) play no part in the transitions to retirement of these men. Their patterns of activity include little intellectual activity and no participation at all in structured learning, and no physical activity except if their occupations require it. Socialising, usually in public spaces in the community, is the one more noticeable activity they keep.
The conjunction of all these factors leads the majority of the men of this group to a terrible situation after retirement, marked by poverty and very low expectations towards life itself. Nevertheless, not all the men we interviewed have experienced difficult transitions to retirement and not all of them live in a terrible situation. A minority maintain professional activity (even if this shows a simple financial need), and this is central in their lives. They use public spaces for socialisation purposes and play games with their friends (cards, dominos, etc.). In some cases, we can see that support coming from social networks does make a difference, not only in the way these men subjectively perceive retirement, but in their objective (also financial) situation as well (friends, family, and ex-employers have made a difference in some cases).

Men from rural areas with a low/medium educational background
This group includes some illiterate men, some with a basic level of education and only a few with secondary education. Their geographical space is determinant in various ways. They live in a semi-mountainous context that is characterised by strong isolation and fragmentation. Most live in small places with a reduced number of neighbours and all services are some kilometres away and are only accessible by car because public transportation is scarce. Most of these men were or are subsistence farmers, while others were manual workers. There are a few who performed technical jobs and a high percentage of them worked for the city hall -a main employer in these areas. Only a minority use computers but all of them use mobile phones.
Formally speaking, the meanings of transitions to retirement are very relative among the men in this group. Either they are farmers, and hence never really retire (even if that is the case formally, they keep a daily routine of working in the fields much similar to their usual patterns of activity), or when retired they found something to do in small farming places. So, the main activity they retain is physical (farming and related activities in the rural world), but no learning activity at all. Socialisation is generally a problem due to geographic isolation. They would like to be in closer contact with their families, but their sons and daughters have usually moved away to urban centres or other European countries. They can be in daily contact with neighbours, but conflicts with neighbours can lead to strong isolation. It is only natural that their social networks are small in dimension and not rewarding or even problematic. There are also very few or no cultural opportunities. Socialisation opportunities can be kilometres away and mobility is a barrier. These men consider their main problems to be health and access to public health, lack of places to socialise, lack of cultural events and loneliness.
Similarly to what happened in the first group, education and learning also play a minor or non-existing role among these men. There is a very low level of participation in activities to do with learning or learning associations, partially motivated by their (local) inexistence and difficulties in transport or mobility.
Their sense of belonging to the community and identity are strong and based in the general notion that, culturally, the mountainous areas are different from the coastal cities. But a certain type of despair makes up a part of this identity: they believe that there is no future in these areas -almost a human desert today.
So, the geographical and social situation of the population leads us to state that transition to retirement is, in this context, a concept that loses a significant part of its explanatory or even functional functions. The men included in this group are in a difficult situation and find themselves, after retirement (and they do retire), isolated and alone, far away from easy access to public health services.

Men from urban areas with a medium/high educational background
This group includes older men that spent a minimum of nine to 12 years in school and some of them have a higher education degree. It is important to note that 40-50 years ago nine years of schooling was not a low educational level in Portugal; quite the contrary, this was more than enough to allow easy access to employment. It is therefore natural that the men of this group have a wide range of professions, most of them marked by high salaries and even high social status. Their professional trajectories carry significant financial capital as well as higher social, cultural and symbolic capital. These types of capital were crucial during the professional trajectories of these men as facilitators of employment, change or progress in their careers, and to ease access to political functions in local administration.
Most of the men in this group possess a car and drive frequently. The great majority have computers and uses social networks, email and internet (in some cases daily). All of them have mobile phones and use a considerable number of applications. We can summarise by saying they live comfortably in material terms and have no serious problems with mobility.
The men of this group generally performed successful transitions to retirement. This does not mean there were no obstacles at all during the transition; quite the contrary, most of these men had a very active life in professional terms and preparation to transition was mostly absent. The first impacts were hard for a number of them. For example, António, who was a higher education lecturer, stated that the liberty of being able to choose what to do or the freedom from daily professional obligations felt good only for the first few months. After this initial period, it was necessary to find new challenges. Even today, he misses the daily contact with students, as friends of lower age groups are difficult to make now.
Other men had difficulties of some kind during the transition to retirement or after an initial period when everything seemed wonderful. However, a set of factors helped them to face these changes or problems and achieve a better situation. First, considerable financial capital is of utmost importance and is a basic condition that has a positive effect on various dimensions (access to goods and services, a more comfortable life in material terms, a buffer if health problems arise, the ability to travel to other countries, etc.). Some of these men are capable, for example, to help members of their families in financial terms, providing better health care if needed. In short, they have enough financial capital to be able to react adequately to unpredictable events, and to provide financial help to their families if needed.
Second, social networks tend to be bigger and more satisfying, with increased opportunities both to give and to receive social support. To almost all of them family is fundamental and they meet frequently with sisters and brothers, daughters and sons; and the majority have an active role in socialising or helping in the education of grandsons and granddaughters. Globally, they maintain friends from their professional milieus and arrange frequent meetings with them either daily in coffee shops or weekly for lunch, for example. The majority of these men are very active in (or lead, in some cases) local associations, sports clubs, amateur theatre groups, cultural associations, or similar projects. These are, in most cases, long-lasting activities that were once fundamental to the cultural life of the cities of Faro and Olhão; they provide a source of rich and deep socialisation that men use as an important resource after retirement. These men are active members of the community and contribute daily to community life in various roles.
Third, within this group we can see that learning has a very important role, especially informal learning (not considering the fact that their educational background was the main factor that yielded high financial capital and a rewarding career). After retirement, these men have kept a very interesting pattern of activities in various dimensions: they are active physically; walking is an interest of the broad majority (with or without a dog/company). Collectively, they have a wide range of cultural interests: reading, the cinema, the theatre (some belong to theatre groups), singing, learning at universities of the third age, or cultural performances of different kinds. A significant part has hobbies of some kind. Some are volunteers (for example, two teach at third age universities). In addition, the majority not only had the habit of travelling, but have also kept up that habit after retirement, although with slight changes (for example, these days they prefer fully-organised trips). To summarise, most men have been able to maintain their activities after retirement and a significant number have been able to pursue new interests. But informal learning is, globally, a very important dimension of their lives, for most of them in an explicit way.
When asked about the most common problems that affect older adults, the men answer as if they are not talking about themselves, but about others, which is meaningful. Most of them point out common problems, such as loneliness, apathy, mobility and physical obstacles to mobility, etc. Interestingly, the men think there is a wide range of available educational, physical and cultural activities that older adults could benefit from. The problem is not, thus, the shortage of supply. They also point out that contact with younger people is crucial and enthusiastically defend inter-generational learning (even if the meanings they give to inter-generational learning differ).

CONCLUSIONS
The men we interviewed show three very different contextual situations that have a basic impact over the life course and, consequently, on the way they experience the transitions to retirement and achieve -or not -a good current situation. This first comment reminds us of a very important conclusion when studying transitions to retirement: we should understand not only the specific period of people's predictable transitions but also the wider conditions of their lives. Without the knowledge of the educational background of our interviewees, the types of occupations they had over time, or the way they live or contribute to community culture, our results would be considerably different, or maybe even biased. The life course perspective therefore has a number of advantages. Our results show how important this perspective is for capturing people's experiences, and the life changes occurring along the process (Duberley, Carmichael, & Szmigin, 2014).
From our findings it seems clear that work is one of the most crucial dimensions one must analyse when dealing with transitions to retirement. Non-qualified, low-paid work marked the lives and therefore severely constrained the retirement of the men in our first group. Rural subsistence work marked the lives of the men living in the inland parts of our region in such a way that it is still a dominant category in their lives today -and will probably be until they die. High-status, well-paid jobs conveyed a high financial capital to the men included in our third group in such a way that other dimensions of their lives were positively affected. There is no doubt that work, identity and social adjustments are strongly connected (Vicente, 2007). Work influences autonomy and control over our lives (Schmidt-Hertha & Rees, 2017) and this is crucial while examining transitions to retirement. Our results also show (particularly among the men in the third group) that getting rid of professional obligations can indeed be experienced as a relief, as stated by Reeuwijk et al. (2013). But for some men this lasts only for some time. These experiences remind us of Atcheley's (1976) pioneer work on the typical phases of living retirement and the transitions to retirement. In the author's terms, the feelings of relief would match the honey-moon phase, and the awareness that this is not enough corresponds to Atcheley's phase of disenchantment. Of course, the diversity of our data also shows very clearly that a sequential phasing, equal for all, such as that suggested by Atcheley at the time, cannot adequately describe the complexity of modern biographies, each time less linear and more complex.
The most original contribution of our article, however, lies in our findings regarding education and learning. There are two major conclusions to stress on this issue. First, educational background is, in our case, the most basic explanatory factor regarding a successful transition to retirement and the achievement of a stable post-retirement situation. The educational level of the men deeply influences their access to employment, and further the quality and status of employment as well as the financial and cultural capital associated with it. Over decades, these factors tend to interact and when reaching an older age, this effect is more visible than ever. The results concerning groups one and three show this very clearly. Second, learning and, most specifically, informal learning, can be very important for the way people experience the transition to retirement and achieve post-retirement stability with a good quality of life. For example, many men in group three find a crucial environment in cultural activities or participating in clubs and cultural associations after retirement. But the great majority of activities they participate in are not formal or even non-formal. The men from our third group made this very visible, as informal learning assumed a central, positive role in their lives. In other words, informal learning can be an important instrument for changing community life (McGivney, 1999). We also note that informal learning is often invisible, unnoticed by both researchers and educators. Our findings tell us we should reverse this situation.
As well as the implications, the limitations of this study must also be considered. Because this study only reports men's perceptions, no comparative analysis is possible with the women's context, even if this was not the purpose. Additionally, our research was developed in a particular region, and we have no guarantee that it can explain the national reality.
Nevertheless, the use of a life course approach has advantages and did contribute to the knowledge of educational background, occupations, people's previous lives, etc. The results do seem to have an impact in local and regional policies and the participants' quality of life when we refer to orientations concerning transitions and the preparation to retirement. Knowing the role of work, education and learning in the process, transitions to retirement can be easier and changes are possible in the community context.