EXPERIENCES OF AGING AND THE ROLE OF PERSONAL RELATIONSHIPSIN CHILEAN SOCIETY:PERSONAL COMMUNITIES, SOCIAL CAPITAL AND POLICY CONTEXTbyMARÍA-JOSÉ TORREJÓN CARVACHOB.A., Universidad de Chile, 2005M.A., Universidad de Chile, 2007A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTSFOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES(Sociology)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)March 2017© María-José Torrejón Carvacho, 2017iiAbstractThe rapid process of population aging and the effect of an induced process of modernization since theearly 1980’s in Chile have prompted questions about the quantity and quality of older people’s formaland informal networks of support. Socio-cultural changes reflecting values of an individualized societyput into question the mandatory character of traditional family support. Using a conceptualframework that combines a focus on bonding and bridging social capital, on personal communitiesand that employs a narrative approach to policy analysis, this thesis investigates to what extent andunder which circumstances older people living in an urban area exchange help and complementfamily resources with other types of personal ties (e.g. friends, neighbours, and/or stateorganizations). Paying particular attention to the composition, function and meaning of personal tiesin later life, I draw on 40 in-depth interviews with people between 60 and 74 years old living in thecity of Santiago and analyze the Chilean “Integral Policy for Positive Aging 2012-2025” to answer thisresearch question. The findings show how older people become integrated in society through themanagement of a network of diverse personal ties. They highlight the nuances in the meaning andfunction of these ties in a context of low institutional trust and neoliberal social policies. The researchcontributes to existing literature by: a) clearly differentiating bonding from strong ties, and bridgingfrom weak ties, while stressing the role of bonding ties acting as bridging social capital to connect theolder person to key symbolic and practical resources in a context of low trust; b) offering a conceptualand methodological framework to recognize the normative and cultural aspects of social policies onaging; c) explicitly considering the role of the socio-cultural context of a country of the global south inthe creation of personal communities.iiiPrefaceThis dissertation is an original intellectual product of the author, M.J. Torrejon. The fieldworkreported in Chapters 2-4 was covered by UBC Ethics Certificate number H13-02487.ivTable of ContentsAbstract .............................................................................................................................iiPreface..............................................................................................................................iiiTable of Contents ..............................................................................................................ivList of Tables ....................................................................................................................viiList of Figures .................................................................................................................. viiiAcknowledgements ...........................................................................................................ixDedication..........................................................................................................................x1. Introduction ............................................................................................................... 11.1. Background ...................................................................................................................21.2. Personal relationships in context .................................................................................41.3. Conceptual framework .................................................................................................81.3.1. Traditional perspective of the theory of individualization....................................91.3.2. Critical perspective of the theory of individualization ........................................101.3.3. Personal communities .........................................................................................121.3.4. Social capital ........................................................................................................141.3.5. Narrative policy analysis......................................................................................161.4. Data production and analysis .....................................................................................181.4.1. A word of caution ................................................................................................281.5. Organization of the thesis...........................................................................................282. “It is an intimacy criterion”. Relationships in late life from a personal communitiesapproach..........................................................................................................................312.1. Introduction ................................................................................................................312.2. Conceptual framework ...............................................................................................342.3. Methodology...............................................................................................................382.4. Findings .......................................................................................................................422.4.1. Composition of personal communities ...............................................................422.4.2. Complementary typology ....................................................................................472.4.3. Friendship ties in personal communities ............................................................482.4.4. Family ties in personal communities ...................................................................52v2.5. Discussion ...................................................................................................................582.6. Conclusion Chapter 2 ..................................................................................................653. Bridging and bonding social capital in Chilean older people........................................683.1. Introduction ................................................................................................................683.2. Bridging and bonding social capital ............................................................................703.2.1. Distinguishing between bridging and bonding social capital ..............................763.3. Methodology...............................................................................................................793.4. Findings .......................................................................................................................863.4.1 Resorting to help from others .............................................................................873.4.2. Friends as bonding and bridging ties...................................................................933.4.3. Family as bonding and bridging ties ....................................................................993.5. Discussion .................................................................................................................1053.6. Conclusion Chapter 3 ................................................................................................1104. Public policy and experiences of aging: social relationships and social integration inChilean policy on aging ...................................................................................................1174.1. Introduction ..............................................................................................................1174.2. Public policy and experiences of aging .....................................................................1194.2.1. Public policies, social integration and social relationships in late life ..............1204.2.2. Policy as a frame of meaning.............................................................................1224.3. Methodology.............................................................................................................1254.4. Chilean public policies and institutional framework on aging..................................1334.5. Findings of the policy analysis ..................................................................................1374.5.1. ‘Tone’ of the policy: population aging as a challenge .......................................1374.5.2. Different instances of integration for dependent and independent older people............................................................................................................................1424.5.3. Autonomy and Self-Management in the Policy analysis ...................................1454.5.4. Policy’s assumptions of family responsibilities and involvement in olderpeople’s support ..............................................................................................................1484.6. Comparing and contrasting older people’s experiences with policy narratives ......1524.6.1. Different instances of social integration through (in)formal relationships ......153vi4.6.2. Older people’s experiences and expectations of family solidarity ...................1564.6.3. The importance of autonomy and self-management in older people’sexperiences ......................................................................................................................1604.7. Conclusion Chapter 4 ................................................................................................1625. Conclusion...............................................................................................................1675.1. The role of the socio-cultural context in shaping the composition, meaning andfunction of personal networks ................................................................................................1695.2. Population aging and socio-cultural change in Chile................................................1745.3. Implications of the findings ......................................................................................1755.4. Study limitations and future areas of research ........................................................179Bibliography...................................................................................................................187Appendix A: Interview Schedule .....................................................................................205viiList of TablesTable 1: Participants demographics ..........................................................................................20Table 2: Size of personal communities......................................................................................43Table 3: Typology of personal communities (based on Spencer & Pahl’s typology) ................47Table 4: Proposed complementary typology of personal communities ...................................47Table 5: Emotional support received by participants .............................................................114Table 6: Practical support received by participants ................................................................115Table 7: Primary and secondary questions guiding the analysis of Chilean policy on aging ..128Table 8: Summary of lines of action and their corresponding specific objectives of the IntegralPolicy for Positive Aging 2012-2025 ........................................................................................129viiiList of FiguresFigure 1: Diagram to map personal communities .....................................................................24Figure 2: Example of axial coding using memos and network view for theme of “chosen andgiven ties” in men of 60-66 years old........................................................................................26Figure 3: Personal community members in men and women ..................................................44Figure 4: Typology of personal communities ............................................................................46ixAcknowledgementsI would first like to express my sincere gratitude to my thesis supervisor Dr. Anne Martin-Matthews for her continuous support, patience, motivation, and immense generosity toshare her knowledge. I would like to thank you for always encouraging my research andadvising on my career; your guidance as a mentor has allowed me to grow as a researcher.To my committee, Professor Ralph Mathews and Professor Paulina Osorio, I am extremelygrateful for your support and suggestions throughout my research.To my friend Kavita, my husband Victor, my mother, my brother, and my ‘Canadian’ family,for their support and encouragement during the better and worse of the process of writingthe thesis.Also, special thanks to the study participants. This research was only possible because theywarmly welcomed and shared their private lives with me. Thank you.xDedicationDedicated to the memory of my father Fernando Torrejon (1951-2012).11. IntroductionThe thesis examines the personal networks of Chilean older people, paying particularattention to the composition, functions and meaning of older people’s relationships in thecontext of a society experiencing a rapid process of population aging and the effects ofstructural and cultural changes triggered since the dictatorship period. In this context, we canframe the way in which older people manage and signify the ties that form their personalnetworks as representing a push and pull between the values associated with anindividualized society, the low levels of institutional and social trust that characterize Chileansociety, and the need to depend on the support provided by their personal ties due to thelimitations of neoliberal policies and services directed to older people. The thesis iscomposed of three result chapters (chapters 2, 3, and 4), written in the format of self-contained papers, and a conclusion chapter (chapter 5). The conclusion chapter connects theresults with the overarching issues regarding socio-cultural change in Chile presented in thisintroduction chapter (chapter 1). Three central issues are examined from different angles inthe finding chapters, based on the analysis of 40 interviews conducted with Chilean men andwomen between 60 and 74 years of age. These issues were: (1) the function and compositionof older people’s personal networks; (2) the circumstances under which older peopleexchange diverse kinds of help with different ties; and (3) the meaning of personalrelationships in later life.21.1. BackgroundIn 2008, 62% of the world population aged 65 and over were living in developingregions, and it is expected that this proportion will increase to 71% by 2030 (Higo &Williamson, 2011). Chile is an example of this phenomenon as its older population is the agegroup that is most rapidly increasing. Today Chileans live an average of 79 years, which ismore than 20 years longer than the average in 1960. At the same time, the birth rate hasdecreased from 5.4 children per woman in 1962-1963 (INE, 2006), to 1.8 in 2012 (ComiteNacional de Estadisticas Vitales, 2012). According to the last Census conducted in 2012,Chileans aged 60 and older constituted 13.7% of the population (Comite Nacional deEstadisticas Vitales, 2012). The projections indicate that this proportion will reach 28.2% in2050 (INE, 2003).Traditionally in Chile, family more than any other type of relationship, has had theresponsibility for elderly people care (Lubben & Gironda, 2003). However, demographictrends, along with the insufficient capacities of Chilean public institutions to meet thedemands of an aging population (Murad, 2003), have opened discussions about theavailability of informal support in later life. The central concern is that the future cohorts ofolder people will face a shortage of support and care networks due to changes in familystructures and living arrangements, and in cultural values protecting older family members.Regarding changes in family structure and living arrangement, the available data show thatthe number of older people in Chile who are living alone or with another older person hasincreased while the household size has decreased. Chilean census data show that older3people living only with a partner increased from 15% in 1992 to 18% in 2002 (Herrera &Kornfeld, 2008). In addition, more recent data from the National Socio-economic Survey(CASEN) reveals that the number of older people living alone has increased from 10% in 1990to 14% in 2011 (Ministerio de Desarrollo Social, 2011). Also, the average number of peopleliving in the same household has decreased steadily in the last fifty years. While in 1960 thisproportion was 5.4 and 5.0 in 1970, in 2012 it was 2.9 people per household.Another argument indicates that the shift from “familistic” to “voluntaristic” valuesmay be putting at risk the networks of solidarity among family members (Fine, 2006; vanTilburg & Thomése, 2010). Familistic values are related to traditional norms of parentalresponsibility and filial obligation or “mandate” (Oddone, 2009), and thus to the expectationin both parents and children of exchanging support if needed. Voluntaristic values, on theother hand, are characteristic of societies experiencing processes of individualization. Inthese societies, people reflexively pursue relationships that are in agreement with theirexpected individual biographies (McDaniel & Gazso, 2014), as people negotiate theirpersonal relationships guided more by values of personal fulfilment than by taken forgranted moral mandates. Thus, the exchange of help among family members is not based ontraditional rules of family obligation, but instead is negotiated and guided by personal choice.The main concern regarding older people’s social capital is that this greater freedom ofchoice will decrease the reliability and the capacity of the personal network to providesupport to older family members. The assumption is that traditional social structures andcommunities fulfilled a protective role toward individuals (van Tilburg & Thomése, 2010).41.2. Personal relationships in contextPopulation aging in post-industrial societies is a particularly complex phenomenon ofanalysis, since it represents a context in which key social transformations intersect (Osorio,2006). For the particular topic of social capital in late life, we cannot forget that personalrelationships are dynamic and depend on particular trajectories marked by social, cultural,and historical forces. Social relationships are developed, maintained, modified, and ended byindividuals acting in specific contexts (Adams & Allan, 1999).The coup d’état of 1973 marked in Chile the beginning of major changes in society.The authoritarian state implemented a neoliberal approach to develop an open model ofparticipation in the global market. The new model was based on market competition lead bythe private sector (Torche & Wormald, 2004) while, at the same time, the role of the state insocially focused public policies, welfare programs, and service provision was reduced(Menanteau-Horta, 2006). According to Taylor (2003), the nature of the relation betweenstate and society was redefined through individualization, privatization and decentralization.As a result, health provision, education, the pension system, and public industries wereprivatized. In addition, the instruments for political representation were suppressed and theproblems of citizens were to be solved individually through market mechanisms (Espinoza,Barozet, & Méndez, 2013).The first phase of implementation of the new model was marked by a shocktreatment (Collins, 1995) for the liberalization of Chilean economy. During this period theunemployment rate increased along with the number of people living in precarious5conditions. According to Menanteau-Horta (2006), almost 2 million people (one fifth ofChilean population), were affected. As a result, by the financial crisis of 1982, Chilean citizenswere not only politically but also economically debilitated. During the second phase ofimplementation, a more realistic approach was applied and some adjustments were made tothe model, but the market was still expected to play a central role in social assistance. By thetime democracy returned in 1990, the macroeconomic landscape had positively changed anda new period of sustained growth with low inflation had begun (Portes, 1997). However, therapid implementation of an imposed neoliberal model produced winners and losers (Torche& Wormald, 2007), visible today in the levels of social inequality.Many authors have described the cultural changes resulting from the implementationof the new model and its modernizing objectives (Garretón, 2004; Lechner, 1999; UNDP,1998). The findings of the Human Development Reports of 1996 and 1998 are good examplesof these changes. According to these reports, there is discontent caused by the rapidtransformations that affected people’s daily lives, their forms of sociability (family andcommunity), their values, and identities (UNDP, 1998). Based on the concept of humansecurity1, the report of 1998, titled “The Paradoxes of Modernization”, examined objectiveand subjective conditions that enable people to access the opportunities created by themodernization of the country. According to the findings of the Subjective Human Security1“Human security means protecting fundamental freedoms – freedoms that are the essence of life. It meansprotecting people from critical (severe) and pervasive (widespread) threats and situations. It means usingprocesses that build on people’s strengths and aspirations. It means creating political, social, environmental,economic, military and cultural systems that together give people the building blocks of survival, livelihood anddignity” (CHS, 2003, p. 4).6Index2, the national average score was low (0.33), with a majority of the responsescategorized as low (35%) and medium-low (45%) subjective security. These findings, alongwith the results of in-depth interviews and life history interviews, were interpreted in thereport as an indicator of social dissatisfaction, which was translated in Chileans’ perceptionsas insecurity and uncertainty. The report does not include standard measures of trust, suchas the ones included in the World Values Survey. However, the research evaluated‘sociability’ examining four aspects that connect with social trust, where interviewees couldagree, disagree or say they do not agree or disagree. The four aspects were: people respectothers’ opinions (53% disagree, 28% agreed), it is difficult that others do something for otherpeople without expecting something in return (64% agreed, 23% disagreed), people pushtheir way over you to achieve their goals (76% agreed, 10% disagreed ), and it’s easy to makegood friends (53.8% agreed, 28% disagreed). The authors of the report supported theirfindings citing a study conducted by the Catholic University of Chile in 1995, which showedthat only 8% of the interviewees from different cities of Chile indicated they could trust inthe majority of people. Taking all this information into consideration, we can see that it ismainly the anonymous other who cannot be trusted. The report concluded that the overallresults indicated a weak definition of “Us”, which implies a lack of trust in others, a weaksense of belonging, and a feeling of uncertainty. In other words, the report showed aweakening of the social fabric and the traditional forms of representation.2 The index goes from 0 (low human security) to 1 (high human security). The variables included in theconstruction of the index are: sociability (possibility of receiving help from others), retirement, work,information, health, and security (delinquency).7Lechner (1999) interprets the findings of the aforementioned report as indicating a‘cultural gap’ between individuals’ experiences and socio-cultural structures produced by therapid modernization process:[… ] in Chile occurs a deep restructuring in only ten, fifteen years, that makes people’spractical experiences and mental dispositions obsolete. The dissatisfaction wouldreflect the bewilderment of people who are suddenly thrown into an unknown world(Lechner, 1999, p. 2).There was no time for people to develop proper tools to act in this new context,where the state was no longer the main source of social integration. Thus, the Chileanmodernization processes and the experience of a military dictatorship negatively affectedsocial relationships and trust among people and towards Chilean institutions (Lechner, 1999).The latter is made clear by Valenzuela and Cousiño (2000) when comparing levels of socialtrust of Chile and United States. The 42% of people in US who believed that they could trustin people contrasted with the 14% of Chilean people answering positively to this question.This contrast was also visible in the 86% of Chileans and 54% of United State respondentswho agreed with the statement “you can't be too careful in dealing with people”.Today, the level of trust in Chile continues to be low, being among the 30% ofcountries in the world with the lowest social trust. The World Value Survey showed that in2011 only 12% of Chileans believed that “most people can be trusted” and 70% believed thatthey cannot be too careful in dealing with people. In addition, the Bicentenial Surveyindicated that in only eight years (2006-2014) institutional trust diminished, with politicalparties and the parliament being the least trusted institutions, with only 3% of people statingthey trust much or very much in them (Santander & Centro UC Políticas Públicas, 2005). In8this context of low social and institutional trust, we can ask to what extent and why wouldolder people be willing to invest in relationships outside their close family circle to exchangedifferent forms of emotional and practical support, and what is taken into considerationwhen the older person decides asking and/or accepting help from family and non-family ties?1.3. Conceptual frameworkTo say that individuals have greater freedom of choice does not necessarily implynegative consequences for older people’s social capital, but rather a change of scenario. Inthis new scenario, late life has been described as a stage of risk and choice, where changes inthe inter- and intra-generational contract, at the level of family and society in general, canrepresent an opportunity to reconstruct the fragmented pathways of late life (Phillipson,2013). In this regard, older individuals can actively create their own modes of living (Morgan,1996). That personal ties can be developed based on voluntariness also means that olderpeople can invest in relationships outside their nuclear family. From this standpoint, forinstance, the instrumental and emotional help required by older people could be sought alsoin friendship relations and weak ties (Granovetter, 1973), such as casual acquaintances andneighbours. Nonetheless, the particular Chilean context, characterized by low levels of trust,makes us wonder to what extend people would invest in relationships beyond their closefamily circle, and if they do so, why? Two variants of the theory of individualization(described below) are used as an overarching framework to examine the composition,function, and meaning of older people’s personal relationships in the Chilean context.9In Chile, renowned scholars have argued that families and community life are beingaffected by individualization (Güell, 1999; Lechner, 1999; PNUD, 2004; Valdés, 2007).However, this has been done more as a reflective exercise than an empirical inquiry.Therefore, it is unknown how individualization has actually affected older people’s socialcapital.The theory of individualization states that the individual has become the central unitof social life, transforming human identity into a personal task as actors are charged withboth the responsibility for performing that task and the consequences of their performance(Bauman, 2002, p. xv). Below, I explain first the more traditional perspective of the theory ofindividualization, usually applied to describe European societies. Then, I present a morecritical strand of this theory, commonly used to describe the effects of individualization indeveloping countries. Both variants of the theory of individualization enable us questioningto what extent family roles have broken down, changed, or strengthened, as well asrecognizing different types of ties that can be relevant in late life.1.3.1. Traditional perspective of the theory of individualizationThe decreased importance of traditional institutions and value systems in modernsocieties, or risk societies, as some authors have stated, has also resulted in greateruncertainty. In this context, Beck and Beck-Gernsheim (2002) stress the relevant role offamily (in its traditional and emergent forms) in modern society, “It will be an alliancebetween individuals as it has always been, and it will be glorified because it represents a sortof refuge in the chilly environment of our affluent, impersonal, uncertain society” (Beck &10Beck-Gernsheim, 1995, p. 2). Modern forms of family would work then in a more symboliclevel, constituting today’s way for individuals to be “re-embedded into social life” (Fine,2007: 185) through a sense of certainty, continuity, and belonging. Therefore, family role, farfrom being at risk, might have changed or perhaps it has been even strengthened. In thesame way, we can consider that a greater freedom of choice can broaden older people’scircle of sociability and support.1.3.2. Critical perspective of the theory of individualizationPersonal ties are embedded in historical, cultural, and societal contexts. Thestructural context shapes and gives meaning to the different ties that form older people’ssocial capital even in the context of an individualized society. To say that in individualizedsocieties the individual has become the central unit of social life does not mean thatindividuals are freer. According to Araujo and Martuccelli (2014, p. 26) “individuals arerequired and produced by a sum of institutions that oblige them to develop a personalbiography”. In the same way, Beck indicates that individuals must give biographical solutionsto systemic contradictions, but these solutions are shaped by institutional prescriptions(Araujo & Martuccelli, 2014; Beck, 1992b). On this regard, the complex social, governance,public service and policy context of Chile shape older people’s social capital. The rhetoric ofindividualization encourages people to construct their own biography and to connect withpeople based on their own personal motivations, while at the structural level people are leftto their own devices to get support in the context of a reduced welfare state. In such acontext –also characterized by low levels of trust–, we might be inclined to think that olderpeople will mainly depend on family support.11Critical approaches based on the theory of individualization have been developed byauthors analyzing non-European contexts. These approaches put a particular emphasis in thefact that individuals negotiate their biographies with the opportunities and constraintsprovided by the social context in which they are embedded, which result in different forms ofindividualization depending on the country or the group under analysis. Through the conceptof ‘fourth world’ Castells (2010, p. 73) analyzes diverse forms of social inequality andprocesses of exclusion that generate groups (e.g. unskilled workers, ghetto dwellers, peoplefrom underdeveloped countries) unable to keep up with the rapid pace of informationalchange. Individuals from the ‘fourth world’ would not be able to pursue individualized andflexible life planning due to the structural exclusion they face. In a similar fashion, the Chileansociologist Fernando Robles (2000) makes a distinction between developed and developingcountries in the way in which they experience the process of individualization. According toRobles, developing countries have weaker public and private institutions to support thesocial inclusion of their citizens. In addition, public services and social policies follow aneoliberal logic of individual responsibility, so that individuals must seek help from family,peers and other informal ties to be included into society. As Brodie (2007, p. 103) indicates:Individualization is increasingly embedded in strategies for social policy reform, whichboth promote the illusion of choice and are designed to shape citizens into self-sufficient market actors who provide for their needs and those of their families.The theory of individualization offers interesting areas of inquiry to observe andbetter understand changes at the level of personal relationships in a context of populationaging. The issue under analysis in this thesis is not only whether the traditional nuclear familyis still the predominant form of association and the main source of support, but most12importantly whether new meanings and functions emerge for this and other forms ofassociation (e.g. friends, fictive kinship, interest-specific organizations). The study explores: To what extent and under which circumstances do older people exchange help andcomplement family resources with other types of personal ties (e.g. friend, neighborsand/or state) in a context experiencing a rapid process of population aging andprofound socio-cultural changes?Through this question, the thesis identifies resources embedded in the network ofrelationships, describing instrumental as well as affective functions including subtle meaningsof personal ties. With this in mind, the chapters that form this thesis use complementaryconceptual frameworks that enable us to observe older people’s personal networks fromdifferent perspectives. Below, I will briefly describe the three conceptual frameworks:personal communities, social capital, and narrative policy analysis. These frameworks helpedguide the interview schedule and the operationalization of the research question.1.3.3. Personal communitiesWellman (2001) states that community has become embedded in social networks andit is no longer easily observed in specific groups and public spaces (e.g. neighbourhood).From this perspective emerges the concept of personal communities defined as “networks ofsociability, support, and identity, where each person is at the Ptolemaic center of his/herown universe” (Wellman, Wong, Tindall, & Nazer, 1997, p. 28). Research conducted by Pahland Spencer (2004, 2010; Spencer & Pahl, 2006) also have been based on this notion,capturing the voluntaristic elements of developing and maintaining personal relationships in13the context of individualized societies (Phillipson, 2013). Spencer and Pahl (2006, p. 45)define the concept of personal community as:… a specific subset of people’s informal social relationships –those who areimportant to them at the time, rather than all the people they know no matter howtenuous the connection. Consequently, personal communities represent people’ssignificant personal relationships and include bonds which give both structure andmeaning to their lives.The personal community approach developed by these authors gives particularattention to the content of the relationships (Pahl & Spencer, 2004; Spencer & Pahl, 2006)and the way by which people give meaning to such relationships. Personal communities aredefined as “communities in the mind” (Spencer & Pahl, 2006), as they are not geographicallybounded nor directly observed in an identifiable social grouping, such as a specific club orneighbourhood. In this regard, advancements in communication technologies have beencentral for the creation and maintenance of these communities. The framework to analyzepersonal community is the recognition of people who are or were important in the person’slife (Chua, Madej, & Wellman, 2009b; Spencer & Pahl, 2006). In this manner, the concept ofpersonal communities is very appealing in its ability to recognize and understand the array ofpeople with whom older persons relate and the underlying motivations to create andmaintain these relationships. The analysis of older people’s personal communities enable usto consider the interplay of different types of social ties, while recognizing that people act asmanagers of their networks of relationships (Phillipson, 2013).141.3.4. Social capitalFrom an individual approach, social capital can be defined as “the aggregate of theactual or potential resources which are linked to possession of a durable network of more orless institutionalized relationships of mutual acquaintance or recognition” (Bourdieu, 1986, p.248). This definition focuses on the instrumental features of social capital (Portes, 1998), asthe network of relationships is the result of individual or collective investment strategies thatallow accumulation of profits. Although such investment is not necessarily a conscious andcalculated process, this approach does help us to understand personal ties in late life as anetwork of relationships actively managed by the older person.A sometimes considered complementary approach to the individual perspective ofsocial capital is the one developed by Putnam who defines social capital as a public good.From this perspective, social capital is defined as “features of social organizations such asnetworks, norms, and social trust that facilitate coordination and cooperation for mutualbenefit” (Putnam, 1995, p. 2). Whether trust can be used as a measurement of social capitalis a contested issue and it is not within the scope of this thesis to solve it. However, trust inothers is recognized here as facilitating the development of relationships that can potentiallybecome sources of information and support. The consideration of trust in others is central tounderstanding the role of personal relationships in the Chilean context. Today, Chile isamong the 30% of countries in the world with the lowest social trust3.3 As a reference to understand the issues of generalized trust in Chile, though for the general population, Chilehas a trust index score of 34.4, while Canada scores 85.93 (ASEP/JDS, 2009).15From the perspective previously described, the distinction between “bonding” and“bridging” social capital has been established. Bonding social capital refers to inward-lookingconnections that “tend to reinforce exclusive identities and homogeneous groups” (Putnam,2000: 22). This type of social capital reinforces exclusive identity and tight bonds ofsolidarity, trust and reciprocity (M. Leonard, 2004; Phillipson, Allan, & Morgan, 2004).“Bridging” social capital, on the other hand, is “outward looking and encompass[es] peopleacross diverse social cleavages”(Putnam, 2000, p. 22). While bonding social capital isbeneficial for “getting by”, “bridging” social capital is central for “getting ahead” (de SouzaBriggs, 2003; Putnam, 2000) as it links to external assets and information. The distinctionbetween bridging and bonding properties of personal relationships is particularly interestingto study the ways in which older people manage their personal ties in a context of low socialand institutional trust. In such a context, we may expect that personal relationships in latelife (e.g. children and friends) are central to access specific resources and information,whether they can be described as bonding or bridging social capital is an issue to be exploredin this thesis.The distinction between bonding and bridging social capital is relevant as the bridgingand bonding properties of a tie might be different in late life and in the context of a societyexperiencing a process of individualization. For instance, ties usually described in theliterature as bonding social (based on criteria of similarity between the person and themembers of the network of relationships), can be relevant in late life for their bridgingqualities. That could be the case of children who can help their parents to connect withyounger generations and specific information or services that cannot be easily accessible16after retirement. In the same way, friends can act as bridges as they connect the olderperson with information and activities that cannot be facilitated by family members. In thecontext of an individualized society, these bridging properties of ties are specially relevant formore symbolic ways of social integration. As Phillipson (2003) states, bridging social capitalhelps to generate broader identities and reciprocities and can be crucial to deal with lifetransitions (e.g. divorce, widowhood) and to engage with new lifestyles in late life. In asimilar vein, de Souza Briggs (2003, p. 2) indicates that bridging ties constitute bridges acrossroles, symbolic interests, and worldviews, being important to develop broader identities andcommunities of interest (de Souza Briggs, 2003; Putnam, 2000).1.3.5. Narrative policy analysisA narrative approach to public policies defines policies as a meaning frame thatinfluences individuals’ interpretations and practices. Biggs and Powell (2003, p. 116) state:“policies provide narrative templates within which certain categories of person or groups areencouraged to live out their lives”. In that sense, the issue is not only whether policydefinitions and action identify and foster emergent population trends but, more importantly,how policies legitimate certain identities by creating social spaces and providing thematerials for the enactment of those identities.According to Clark (1993, p. 13), public policy can be defined as the “…attempt tobalance competing notions of the responsibility of individuals, families, and the state indeveloping programs to meet human needs”. The author argues that every policyperspective, statement or recommendation represents a story or sub-story within the17broader narrative discourse about a gripping policy problem (Clark, 2011, p. 84). For instance,public policy reflects assumptions on the responsibility of different actors in meeting theneeds that have been defined. In the context of our study, a narrative analysis of the currentChilean policy of aging allows us to examine the state’s particular definitions and priorities ofthe challenges of an aging society and the role of different actors (public, private, families,and individuals) in facing those challenges.An analysis of the Chilean Policy seems relevant to recognize whether this policyprovides the social spaces and materials for the current ways in which older people getsupport and manage their personal relationships. Also, current Chilean Policy, through itsdifferent programs and regulations, can legitimate and foster certain forms of socialintegration that may or may not be in line with definitions, expectations and practicesrelative to personal relationships and social integration of older people. Social integrationcan be defined as older people’s network of social connections and participation inmeaningful roles (Pillemer, 2000, p. 8). We may ask to what extent Chilean Policy identifiesand promotes meaningful ways of integration through informal relationships (Scharf, 1998),such as family and friendship networks.The next section provides a general overview of the methods used to produce andanalyze the data. More detailed information on these topics is provided later in each of thechapters that form the findings section.181.4. Data production and analysisBetween November of 2013 and February of 2014 forty in-depth qualitativeinterviews were conducted with a group of urban people between 60 and 74 years old livingin city of Santiago, Chile. The research study applied principles of ‘grounded theory’ in thecreation of the interview schedule, the sampling strategy, and the analysis of the interviews.Grounded theory can be defined as an inductive methodology for the construction oftheoretical constructs from qualitative analysis (Corbin & Strauss, 2008). A central issue instudies using grounded theory principles is the construction of concepts through an iterativeprocess of data analysis. The purpose of these concepts is to explain people’s experiencesthrough the recognition of the context in which those experiences are situated (Corbin &Strauss, 2008). The concepts are constructed through the constant comparative method.The ‘constant comparative method’ of analysis was used from the beginning of thestudy. This is an iterative method that enables the researcher to make decisions regardinginitial data production based on the new information that becomes available in the course ofthe field work (Charmaz, 2006; Corbin & Strauss, 2008). Through the constant comparativemethod the researcher compares the different pieces of data seeking similarities anddifferences to classify similar incidents in categories (Corbin & Strauss, 2008). Glasser andStrauss (1967, p. 105) distinguish four stages of the constant comparative method: “(1)comparing incidents applicable to each category, (2) integrating categories and theirproperties, (3) delimiting the theory, and (4) writing the theory”.19It is the iterative qualities of grounded theory that were particularly relevant in thepresent study, as its principles enabled modifications as more knowledge was gainedregarding older people’s relationships in the particular context where the study wasconducted. The possibility of making changes as the study was developed was central toimproving the interview schedule to make it more aligned with Chilean older people’sexperiences, and to explore other characteristics of the participants that could be influencingthe way they managed their personal ties (e.g. include more ‘emergent’ types ofrelationships such as living apart together, and recently divorced participants).Two different stages for selecting the study participants were used. The participantswere initially selected from a cohort of older people who participated in a series of focusgroups conducted by the Domeyko Program on Aging of the University of Chile in 2008. Onlyparticipants who were currently within the age range of 60 to 75 years were contacted.Because the Domeyko study was in 2008, people aged 60 to 65 years old were missing fromthe sample pool. Also, some eligible participants could not be contacted due to change intheir phone numbers. Other people could not participate due to health problems. Inaddition, a relatively balanced number of men and women was desirable to compose thesample but the list of eligible people from the Domeyko participants had a lowerrepresentation of men (40 women and 15 men). In total, 10 people (7 women and 3 men)out of 41 people from the Domeyko study who gave consent to be reached for future studieswere contacted and could participate. The other thirty participants were reached in a second20stage through older people’s clubs and associations4 and by the snowball technique. For thelatter strategy, men and women in clubs and associations where asked to refer relatives,friends or acquaintances who could participate in the study. They were asked to refer peoplewho do not necessarily participate in seniors clubs. The use of this sampling strategy wasintended to reach people, particularly men, who did not participate in clubs and olderpeople’s associations. The different sampling strategies produced a heterogeneous group ofparticipants with respect to their marital status and living arrangements (Table 1).Table 1: Participants demographicsMen Womenn % n %Age 60-66 6 43 14 5467-75 8 57 12 46Marital Status Married5 11 79 9 35Divorced/separated/annulled 2 14 7 27Widowed 1 7 6 23Single 0 0 4 15Number of Children 0 0 0 2 81 2 14 4 152-3 11 79 16 624-5 1 7 4 15Living arrangements Study participant living alone 1 7 5 19Study participant without spouse living withchildren. Study participant is the household head1 7 8 31Study participant without spouse living withchildren. A child is the household head0 0 3 12Study participant living only with spouse/partner 4 29 4 15Study participant with spouse living with children.Study participant or spouse is the household head8 57 3 12Study participant with spouse living with children. Achild is the household head1 7 1 4Other 0 0 2 8Educational level(last level completed)Elementary school incomplete 0 0 1 4Elementary school 0 0 3 12High school 3 21 8 31Technical education 4 29 1 4University 7 50 9 354 People were contacted from a retired teachers club and municipal associations for older adults.5 Only 2 dyads of married couples participated in the study.21However, despite the efforts to have a varied sample, more women than menparticipated in the study. Also, the study participants presented higher than averageeducational level when compared with Chilean older population (60 years and more). Finally,the experiences described in this thesis represent the particular reality of urban older people,as all the study participants lived in Santiago, the capital of Chile, where 40% of Chileanpopulation lives according to Census data from 2002.Gender and cohort criteria were important for the sampling strategy and the processof analysis. Gender was central as personal relationships of men and women are differentlystructured. As in other countries, marital status varies by age and gender in Chile. Womenhave longer life expectancy than men and their average age at marriage is lower than that ofmen. While widowers tend to remarry, widowed women tend to remain single. This alsocreates cohort differences, as being alone and widowed becomes more common amongolder women than men. Chilean data for the population of 60 years of age and more showthat 35.1% of older women are widowed versus 12.3% of men. There are also more singleolder women (11.6%) than older men (6.2%) (Gobierno de Chile, 2012).Differences between cohorts and gender are also important because of the lifetransitions experienced by Chilean older people. In the case of the younger cohort of olderpeople (60-66 years of age) considered in this study, people have recently retired or are inthe process of retirement6. In addition, studies on personal relationships in late life haveshown that women receive more help from and have better relationships with friends andfamily members than men (Cornwell, Laumann, & Schumm, 2008; Tomassini & Glaser, 2007).6 The retirement age in Chile is 60 years old for women and 65 years old for men.22In Chile, there is little information on this regard, but a qualitative study conducted by Barros(1991) showed that older men give more importance to family relations, while older womengive similar importance to friends and family ties.This research focuses on a cohort of older people in the earlier stages of aging, thosefrom age 60-76 years, which was divided in two sub-cohorts: people between 60 to 66 yearsold and those between 67 and 75 years old. The rationale for this distinction is that theyounger cohort includes the baby boomers, a cohort that has not yet been studied in Chile. Inaddition, due to the dictatorship period, Chilean people experienced drastic changes in theirpersonal relationships as well as the institutional context in which such relationships weredeveloped and enacted. In 1973, the younger cohort of older people was at the beginning oftheir university studies or first jobs. The older cohort, on the other hand, was at the end oftheir university studies, advancing their professional careers and/or forming or consolidatinga family7.The interviews consisted of one session of from one to two hours in a place mutuallyagreed by the researcher and the study participant. The interviews were conducted atpeople’s homes, coffee shops, seniors’ organizations, people’s workplace, and at theUniversity of Chile. The interview schedule included open-ended questions adapted from astudy on personal communities conducted by Spencer and Pahl (2006), and a resourcegenerator (Van Der Gaag & Snijders, 2005) especially created for this research with thepurpose of exploring older people’s social capital. A resource generator is an instrument for7 According to Census data, in 1970 35% of people between 15 and 24 years old (current cohort of 60-69 yearsold) and 61% of those between 25 and 29 years old (current cohort of 70-75 years old) were married; 43% ofthe former cohort and 62% of the latter were working.23the measurement of individual social capital that helps to identify accessed social resourcesand the ties by which such resources are accessed.The categories of people who provided or received support and the items describingsituations where help was received or provided, were based on the resource generatordeveloped by Wellman and colleagues for the Connected Lives Project (2006). However, theitems of the resource generator are context-based, as ideas about the usefulness of specificsocial resource vary among countries and populations (Van Der Gaag & Snijders, 2005). Thus,to ensure that the list of situations were meaningful to the life stage and socio-culturalcontext of the participants, the items of the questionnaire were adjusted based on thesecondary qualitative analysis of the Domeyko focus group and the preliminary analysis ofthe first five interviews conducted at the beginning of the fieldwork.The questionnaire sought to identify specific types of emotional and instrumentalhelp exchanged with different groups of people (household members, close family, otherrelatives, neighbours, friends, organizations, and other non-kin ties). Adjustments to theinterview schedule, including the addition of a resource generator questionnaire, were madeafter conducting five initial interviews. These changes were intended to improve the scopeand clarity of the questions and to ensure a good flow to the conversation. In the case of theparticipants of those five initial interviews, a short follow-up by phone was added so as toobtain some information that was not obtained in the initial interview schedule.The interviews included a method of ‘mapping’ personal relationships proposed bySpencer and Pahl (2004; 2006). This method is based on the hierarchical mapping technique24developed by Antonucci (1986), in which concentric circles are used to represent thepersonal network of the participant (figure 1).Figure 1: Diagram to map personal communitiesIn the centre of the circles is the word “I” (Ego). The participants were asked to locateon the diagram the name of the people who were close to them. For the name elicitation theparticipants were asked to first make a list of people “who are important to you now”. Theywrote names or nicknames of the people, kinship, age, and geographical distance. Thismethod was suitable to elicit the names of people who were relevant for reasons beyond thehelp exchanged and aided in exploring the meaning of personal relationships. Thedevelopment of the list and the construction of the map was an activity that was verymotivating for the participants. They took time to decide who should be included on thediagram, and where. During the interview, the diagram helped the flow of the conversation,connecting personal experiences and reflections on the history and qualities of participants’personal relationships (for examples of completed diagrams please refer to chaper 2, p. 46).As each interview developed, participants were allowed to include additional members onthe diagram. All the participants elaborated the list of important people, but thirty-seven out“I”25of the forty participants actually constructed the diagram of concentric circles. Only threeparticipants could not complete the diagrams as two of them were blind and the other wasilliterate.The interviews were first fully read to recognize general themes. In a second stagedata were segmented through open coding (Benaquisto, 2008). The codes that resulted fromthis process were of two types: in-vivo codes, which use participants’ language to name thecodes, and significant codes, which use the researcher’s language to reflect the focus-of-inquiry of the study (King, 2008). The analysis of the data was aided with the softwareATLAS.ti.In a third stage, a more focused coding was conducted. Here, some codes wereincorporated into broader categories and others were refined, thus creating and linking newcodes that described more specific dimensions. Some of the codes created during the secondstage had the purpose of identifying actors of the personal network as they were named atdifferent moments of the interview. Other codes created in the second and third stage weremore analytical as they sought to identify qualities of relationships and help exchanged,reflections and discussion regarding daily life experience, and conceptual themes. The ideawas to link the identification codes with the analytical ones to reflect who was giving orreceiving certain types of help to supplement the findings of the resource generator. In afourth stage of the analysis, axial coding8 was performed integrating the different codes8 “Axial coding relates categories to subcategories, specifies the properties and dimensions of a category, andreassembles the data you have fractured during initial coding to give coherence to the emerging analysis”(Charmaz, 2014, p. 147).26created by establishing relationships amongst them with the aid of memos9 and conceptualnetwork maps.Figure 2: Example of axial coding using memos and network view for theme of “chosen and giventies” in men of 60-66 years oldFigure 2 illustrates the process of axial coding using a conceptual network map,where the memo “chosen and given ties” is used as the central theme to integrate codes,memos and quotations. In the figure, nodes with yellow icon represent quotations, pinknodes correspond to codes created during the second stage of analysis, green nodesrepresent codes created in the third stage of analysis; nodes with red icon symbolize memos.Verbatim transcriptions of the interviews were in Spanish and therefore the quotations are inthat language. Selected quotations were then translated into English to be used in thefindings chapters.9 Memos allow comparing data, exploring ideas regarding codes, and directing further data production(Charmaz, 2006). Memos can go from very descriptive memos to more abstract and integrative ones. They canbe used, for instance, to describe properties of a category, summarize ideas, link concepts, and includeobservations about the research process. In short, memos can be defines as “written records of analysis”(Corbin & Strauss, 2008, p. 117).27The diagrams of concentric circles that represent the personal communities of thestudy participants were classified in two forms: using Spencer and Pahl’s typology,considering the number and type of ties included on the diagram, and applying a visualcriterion based on how all the ties were located in relationship to the centre of the diagram.This visual criterion differentiated between hierarchically and clustered arrangement of tieson the diagram (for examples please refer to chapter 2, p. 46). Hierarchical personalcommunities are characterized by ties clearly organized using the different circles. Inclustered communities, ties are located at the centre of the map, even overlapping oneanother.In addition to the interviews, the document Integral Policy for Positive Aging 2012-2025 (SENAMA, 2012) was analyzed using an interpretative narrative approach that focuseson the expressions of social meaning present in policies, such as beliefs and values thatmotivate specific policy definitions and actions. The analysis conducted on Chilean policy onaging was framed by three primary questions (please refer to chapter 2, Table 7) on thedefinitions, practices, assumptions and unexpressed themes regarding social relationshipsand social integrations in late life (Biggs & Powell, 2003; Grenier, 2012). For each primaryquestion, secondary questions were added to examine specific topics. These secondaryquestions are based on the emergent trends and issues identified by Reichert and Phillips(2009) in their analysis of policy debates. Based on these secondary questions, I createdcodes to identify the main definitions and topics of interest discussed in the Policy. Althougha thematic analysis was conducted, the main purpose of the codes was to provide a startingpoint to identify explicit narrative threads. The contrasting and comparison of these explicit28narrative threads with the available literature and the experiences of the study participantsenabled the identification of assumptions and neglected issues.1.4.1. A word of cautionThe findings presented in each chapter are accompanied by quotations from theinterviews and, in the case of the fourth chapter, by quotes from the Policy document. Theoriginal language of all the data is Spanish and therefore the excerpts used in the chaptersare translated from that language to English. In doing this translation I have faced thechallenge of making the quotes understandable in English while preserving, as much aspossible, the original quality of the expressions used by the study participants (rich incolloquialisms, idiomatic phrases, and metaphors). The final result is a patchwork of“correct” English with not-so-idiomatic English expressions. I take responsibility for this finalresult, recognizing that the reader will need a fair amount of imagination to make sense ofthe quotes. However, I also feel committed to making the “voices” of the study participantsheard as closely as possible to their original uniqueness. After all, this thesis was onlypossible thanks to them, who so gently and willingly allowed me access to their social worlds.1.5. Organization of the thesisThe next three chapters present the results of the study. Each chapter should be readas self-contained papers that provide more detailed information on the specific methodologyand conceptual framework used. The chapters have been organized to present the findingsfrom the micro experiences of personal relationships in old age to the way in which suchexperiences intersect with more macro processes.29The second chapter “It’s an intimacy criterion”: relationships in late life from a personalcommunities approach, is framed by the perspective of individualization that sees personalrelationships as a source of meaning and sense of belonging. It analyzes the diversity of tiesthat form older people’s networks although more specifically attending to their emotionaland meaning function. The analytical framework used in this chapter is based on theperspective of personal communities proposed by Spencer and Pahl (2006) to emphasize asubjective identification and definition of ties that are considered important by the olderperson.The third chapter Bridging and bonding social capital among Chilean older people,complements the first, giving a broader picture of older people’s personal networks and theresources embedded in those networks. The aim of this third chapter is to examine whetherand under which circumstances different types of personal ties are used by older people asbonding or bridging social capital. The experiences of the study participants weresummarized in three analytical themes that enabled us to explore the function of friends andfamily ties and the circumstances under which these ties are selected to exchange particularkinds of emotional and practical help.The fourth chapter Public policy and experiences of aging: social relationships andsocial integration in Chilean policy on aging, uses a narrative approach to examine whatforms of social integration are fostered by Chilean policy on aging in comparison to the ideasand practices relative to personal relationships and social integration as perceived by olderChilean people. Based on Scharf’s (1998) distinction between integration through formal andinformal relationships, the forms of social integration explored in this chapter are: social30integration trough formal relationships (e.g. participation in the labour market) and socialintegration through informal relationships (represented by family and friendship networks).The document Integral Policy for Positive Aging 2012-2025 was analyzed using aninterpretative narrative approach that focuses on the expressions of social meaning presentin policies, such as beliefs and values that motivate specific policy definitions and actions.Finally, the conclusion chapter connects chapters 2, 3 and 4, stressing the practical,methodological, and theoretical implications of the research findings. The findings of thesechapters are interpreted in the conclusion by connecting the nuances in the meaning andfunction of personal ties in later life with the particular socio-cultural context of Chileansociety characterized by low institutional trust and neoliberal social policies.Using different angles of analysis to interpret the data, this research contributes toexisting literature on social capital, personal communities and policy analysis by: a) clearlydifferentiating bonding from strong ties, and bridging from weak ties, while stressing the roleof bonding ties acting as bridging social capital to connect the older person to key symbolicand practical resources in a context of low trust; b) offering a conceptual and methodologicalframework to recognize the normative and cultural aspects of social policies on aging; c)explicitly considering the role of the socio-cultural context of a country of the global south inthe creation of personal communities.312. “It is an intimacy criterion”. Relationships in late life from a personal communitiesapproachWhen I met Claudia she apologized for making me wait in the lobby of the apartmentbuilding. She told me that one of her granddaughters, a 6 months old baby, had somehealth problems during the night and she was talking by the phone with her daughter tocheck that everything was fine. The interview was constantly interrupted by phone calls(Claudia always kept her Smartphone close) from her daughter, an older granddaughterwho needed something from Claudia, and a friend who wanted to discuss a personalproblem and confirm that Claudia and she would meet next day to go dancing. Alsoanother friend called to confirm that Claudia and she would meet for lunch soon. Claudia’slittle granddaughter was in the hospital and she offered to go there to see her and supporther daughter. Then Claudia remembered she had also committed to go to her daughter’shouse so that the cleaning lady could enter. Immediately, she took her cell phone andcalled a friend, Ramon. Everything worked out well; Ramon will go to look after Claudia’sdaughter’s house so that Claudia could go to the hospital. While all these arrangementswere made by the phone, I checked the interview documents and noticed that Ramon hadnot been mentioned as part of the support network, though Claudia located him almost atthe centre of her map of close relationships. While apologizing for all the interruptions,Claudia commented what a good friend Ramon was, always there available to help her.2.1. IntroductionThis field note excerpt shows the challenges of capturing the complex processesthrough which people support one another and the varied functions of personalrelationships. In my interrupted interview with Claudia, I had the opportunity to experiencehow social capital is mobilized to address daily and unexpected issues. And not only that, Icould also see how difficult is to recognize the diverse functions and meanings of personalrelations. Why was Ramon not present in Claudia’s support network? Was it just amethodological issue regarding the questions or are there more subtle functions and32meanings at play to make some ties more salient in certain situations? A reflection shared byanother participant gave some clues regarding this issue:[...] for instance, one day Nancy [daughter] was ill, I took her to the supermarket, butthat has nothing to do with it [to ‘help’ Nancy] , [...] if you talk to me about thosethings [giving help], I would think about bigger stuff [...] because the other things aredaily life stuff. If you are family, if you are ill, you go there, you take her there, butthat is like, no, I don't count it [as help]. (Carmen, 66 years old)Carmen’s quotation shows how problematic it is to identify what can be consideredas help. It was not only an issue of recalling instances, but a matter of what should becounted as help. The absence of Ramon in Claudia’s support network and Carmen’s reflectionprovide several insights into the different functions and meanings of personal relationships.First, regular exchanges of support tend to remain relatively invisible, being considerednormal parts of daily life interactions. Second, the fact that some ties are not identified in thesupport networks does not mean they are less important. The question then is how can wemake visible the ties that have become part of ‘taken-for-granted’ reality and how are theseties relevant to older people?The literature on social capital, social support and social networks in late life usuallydescribes and measures the resources available in the network of the older person (Bowling,Banister, Sutton, Evans, & Windsor, 2002; Gray, 2008; Tomassini & Glaser, 2007; van Tilburg& Thomése, 2010). Although some of that literature includes the concept of emotionalsupport (Krause & Borawski-Clark, 1994; Krause & Rook, 2003) or measures to recognizeemotionally close ties (Lang & Carstensen, 1994; Nyqvist, Gustavsson, & Gustafson, 2006),the studies rarely recognize the emotional nuances and the subjective meaning of thoserelationships (Forsman, Herberts, Nyqvist, Wahlbeck, & Schierenbeck, 2013; Roseneil &33Budgeon, 2004; Takahashi, Tamura, & Tokoro, 1997). This chapter starts from thedescriptions of the function of social relations in modern societies (Beck & Beck-Gernsheim,1995a; Fine, 2006; van Tilburg & Thomése, 2010) to stress the role of subjective meaning ofpersonal relationships in late life. I seek to recognize the diversity of ties that form olderpeople’s emotionally close network of relationships or personal communities, as defined byPahl and Spencer (2010), and the meanings and more subtle functions of personalrelationships in late life, with particular attention to family and friendship.The relevance of this research is rooted in the studies that have shown that havinggood family and friendship networks, as well as participating in the community, are at thebasis of good health and wellbeing in late life (Antonucci, Birditt, & Ajrouch, 2011; de Belviset al., 2008; H. Litwin, 2009; Howard Litwin, 2006; Lubben & Gironda, 2003; Nyqvist,Forsman, Giuntoli, & Cattan, 2012; Takahashi et al., 1997). In modern societies, not onlyfamily relationships but also ties of choice may be particularly relevant to maintain physicaland mental health (Allan, 2010). Personal relationships increase older people’s capabilities tomanage their own lives (Fiscella, Rivera, & Román, 2009), giving the possibility of accessingand using existing resources of their own network (Fiscella, Rivera, & Román, 2009). Thefunctions of relationships in late life go beyond the provision of social support, theycontribute on a more symbolic level with shared life events and common social activities(Forsman et al., 2013). The recognition of the emotional nuances and the meanings ofrelationships could be especially relevant in contexts of rapid change and modernization,where personal ties gain importance in providing a sense of certainty, continuity, andbelonging.342.2. Conceptual frameworkSome authors indicate that we should understand the macro processes affecting thesocial relations of older people in the context of “accentuated modernization” (Allan, 2001;van Tilburg & Thomése, 2010). This context is characterized by a loss of influence oftraditional social structures and communities in individuals’ lives and the greaterresponsibility that the individual has in managing risks and shaping her or his life course. Thisdescription is in line with what Beck and Beck-Gernsheim (2002) have defined as‘individualization’. At the level of personal relationships, individualization is a move awayfrom solidaristic social forms, such as family, social class and community, in which peoplewere tied to each other by instrumental interdependence to meet individual and groupneeds. In a different way, in modern societies, social relationships are voluntary and basedon affection. Thus, personal security and social integration are no longer given by traditionalmeans of collective identity and meaning (Baars & Phillipson, 2014).In individualized societies, risk has to be managed individually, creating greateruncertainty for individuals. In this context, the function of personal relationships isparticularly important as these relationships appear as an important resource to be re-embedded into society, offering a sense of meaning and belonging. In other words, in thecontext of late-modernity, personal ties provide not only different types of instrumentalhelp, but they also work as a benchmark for identity construction (Allan, 2001; Spencer &Pahl, 2006).35Different family structures are becoming more common and accepted. The cohortsthat initiated and experienced changes such as cohabitation, multiple marriages, singlehood,non-marital childbearing, and childlessness are now entering old age (Hughes & Waite,2007). These changes were accompanied by transformed family roles and relationships thatchallenge the definition of what constitutes a family and the obligations among familymembers (Hughes & Waite, 2007; Phillipson, 2013). In current times the limits of kinship arebecoming blurred and the networks of potentially meaningful relationships enlarged(Phillipson, 2013; Spencer & Pahl, 2006).Traditional notions of community, solidarity and family have obscured the potentialof other types of relationships that the current demographic and cultural contexts give(Phillipson, 2013). According to Wellman (2001), community has become embedded in socialnetworks and it is no longer easily observed in specific groups and public spaces (e.g.neighbourhood). From this perspective emerges the concept of personal communities, whichare “networks of sociability, support, and identity, where each person is at the Ptolemaiccenter of his/her own universe” (Wellman et al., 1997, p. 28). The concept of personalcommunities, as developed by Pahl and Spencer (2004, 2010; Spencer & Pahl, 2006),captures the voluntaristic element of developing and maintaining personal relationships inthe context of individualized societies (Phillipson, 2013). It also adds a conceptual insight tobetter understand personal relationships in modern societies:we use the term ‘personal community’ to refer to a specific subset of people’sinformal social relationships –those who are important to them at the time, ratherthan all the people they know no matter how tenuous the connection. Consequently,personal communities represent people’s significant personal relationships and36include bonds which give both structure and meaning to their lives. (Spencer & Pahl,2006, p. 45)The personal communities perspective is different from the “whole network” viewthat focuses on the entire set of ties (Chua et al., 2009b), i.e. the connection among all themembers of the network. While a network perspective focuses on the structural andinteractional characteristics of personal ties, such as size, physical distance, and frequency ofcontact, a personal community standpoint attends to the content of the relationships (Pahl &Spencer, 2004; Spencer & Pahl, 2006) and the way by which such relationships are signified.Considering personal relationships as part of personal communities enable to consider theinterplay of different types of social ties, while recognizing that people act as managers oftheir networks of relationships (Phillipson, 2013).The personal communities’ perspective complements other approaches that paymore attention to the instrumental help and more observable resources embedded in thepersonal network, such as social support and social capital, by adding a more symbolic levelof analysis. In that sense, the function of personal ties as constructors of identity and giversof meaning is central to the idea of personal communities. These communities provide asense of stability and continuity to the individual through shared memories, even whenindividuals face change. Spencer and Pahl (2006) have defined them as “communities in themind”, as they are not geographically bounded nor directly observed in an identifiable socialgrouping. Instead, the framework to analyze personal community is the recognition of peoplewho are or were important in the person’s life (Chua et al., 2009b; Spencer & Pahl, 2006). Inthis manner, the concept of personal communities is very appealing to recognize and37understand the array of people with whom older persons relate and the underlyingmotivations to create and maintain these relationships.A priori definitions of relevant ties are avoided by the personal community approachas they could hide the subjective meaning and experience of the relationships. The latter iscentral since family ties, particularly nuclear family ones, have been usually assumed as themore important relationships in late life (Takahashi et al., 1997) and “just knowing the typeof family connection tells us nothing about the quality of the individual tie” (Spencer & Pahl,2006, p. 33). These kinds of assumptions obscure the subjective meaning given torelationships, their more subtle functions, and the importance of other type of ties, such asfriends, fictive kin and extended family members. With a focus on subjective ways ofsignifying relationships, Spencer and Pahl move forward from distinctions of kin and non-kinties and assumptions of personal proximity based on partnership or biological affiliation(Wall & Gouveia, 2014).The authors use the concept of “suffusion” (Spencer & Pahl, 2006) to indicate theblurring boundaries of categories of kin (‘given ties’) and non-kin (‘chosen ties’) relationships.The idea of suffusion puts into question this dichotomy as both kin and non-kin ties haveelements of personal choice and provide –depending on the circumstance- similar resourcesto a person’s life. For instance, some friends are considered as family and provide helpusually given by family members, while a sister-in-law is regarded as friend to whom a personcan confide personal issues.38Recognizing the varied composition and structure of personal networks, Spencer andPahl (2004) distinguish different typologies of personal communities: ‘Friend-like community’: friends have a central role and outnumber family members. ‘Friend-enveloped’: family members occupy the center of the map and friends surroundthat centre. ‘Family-like’: family members outnumber friends. ‘Family-dependent’: family members not only outnumber friends but also provide moresupport. ‘Partner-based’: partner is regarded as the focal point or only significant tie in thepersonal community. ‘Professional-based’: central role played by professional sources of support (e.g.therapist, counsellor, lawyer).Because the method employed by Spencer and Pahl is highly inductive, the categoriesproposed may or may not be found when conducting studies in other contexts.2.3. MethodologyThe analysis conducted for this chapter sought to identify the subjective meaning ofrelationships in a cohort of people between 60 and 74 years old. Forty men and womenbetween 60 and 74 years old living in the city of Santiago, Chile participated in in-depth semi-structured interviews. The participants were selected using a theoretical sampling strategy(Mason, 2002). Gender and cohort criteria were important for the sampling strategy and theprocess of analysis. Gender was central as personal relationships of men and women are39differently structured. While widowers tend to be remarried, women tend to remain single.This also creates cohort differences as widowhood becomes more common among womenas they grow old. Cohort differences are also important because of life transitions. In the caseof the younger cohort, people have recently retired or are in the process of retirement10. Inaddition, studies on personal relationships in late life have shown that women receive morehelp from and have better relationships with friends and family members than men(Cornwell et al., 2008; Tomassini & Glaser, 2007).The focus on a young cohort of older people sought to increase knowledge of acohort that includes baby boomers, which has not yet been sufficiently studied in Chile onmatters concerning their personal relations. In addition, a better understanding of personalties in people of young cohorts of older people is of great importance for public policies inChile as they are and will experience the consequences of the demographic changes thatimpact the structural availability of traditional means of informal support (i.e. family ties).Finally, the young cohort of older people lived in a central life stages during a criticalhistorical period. During the dictatorship period, Chilean people experienced drastic changesin their personal relationships as well as the context in which such relationships weredeveloped and enacted. In 1973, the younger cohort of older people was at the beginning oftheir university studies or first jobs. People of the older cohort, on the other hand, were atthe end of their university studies, advancing their professional careers and forming orconsolidating a family.10 The retirement age in Chile is 60 years old for women and 65 years old for men.40The selection of the participants was initially made from older people thatparticipated in a series of focus groups conducted by the Domeyko Program on Aging of theUniversity of Chile in 2008. Only participants who were currently within the age range of 60to 74 years were contacted. Some eligible participants could not be contacted because ofchange of phone number or health problems. In total, ten people from the Domeyko studywere contacted and agreed to participate in the study. The other thirty participants werereached through older people’s clubs and associations and by the snowball technique.The interviews consisted of one session of one to two hours length. The interviewschedule included open-ended questions adapted from a study on personal communitiesconducted by Spencer and Pahl (2006), and a resource generator (Van Der Gaag & Snijders,2005). The close-ended questions of the resource generator were based on a secondaryanalysis performed on data obtained from the focus groups of the Domeyko study andpreliminary analysis of the interviews conducted for the current study. The questionnairesought to identify specific types of emotional and instrumental help exchanged with differentgroups of people (household members, close family, other relatives, neighbours, friends,organizations, and other). Adjustments to the interview schedule were made afterconducting the five initial interviews to improve the scope and clarity of the questions andensure a good flow of the conversation. In the case of the participants of those fiveinterviews, a short follow-up by phone was needed to complete some information that wasnot contained in the initial interview schedule (see Appendix A for the revised interviewschedule).41The interviews were aided by a method of mapping personal relationships proposedby Spencer and Pahl (2004; 2006). Their method is based on the hierarchical mappingtechnique developed by Antonucci (1986), in which concentric circles are used to representthe personal network of the participant. In the middle of the circles is the word “I” (Ego). Theparticipants were asked to locate on the map the name of the people that were close tothem. For the name elicitation the participants were asked to first make a list of people “whoare important to you now”. They wrote down the names or nicknames of the people, kinship,age, and geographical distance. This method was suitable to elicit the identification of peoplewho were relevant for reasons beyond the help exchanged and also aided to explore themeaning of personal relationships. The development of the list and the construction of themap was an activity that participants found to be very interesting. They took time to decidewho should be included on the map and where. During the interview, the map helped theflow of the conversation, connecting personal experiences and reflections on the history andqualities of participants’ personal relationships. As the interviews developed, participantswere allowed to include members on the map. All the participants elaborated the list ofimportant people, but thirty-seven out of the forty participants actually constructed the mapof concentric circles. Three participants could not draw the maps, as one of them was blindand the other was illiterate.Verbatim transcriptions of the interviews were analyzed using the software ATLAS.tiguided by grounded theory principles (Charmaz, 2006; Corbin & Strauss, 2008). For theanalysis, the interviews were first fully read to identify general themes and create initialcodes. In a second stage, the codes were used to segment the data. New in-vivo and42significant codes were added as the analysis developed. The former type of codes usesparticipants’ language to name the analytical categories (King, 2008). Significant codes, onthe other hand, use the researcher’s language to reflect the focus-of-inquiry of the study(King, 2008). In a third stage of the analysis, the data were integrated by establishingrelationships among the codes with the aid of memos and network views. The maps wereclassified in two forms: using Spencer and Pahl’s typology, considering the number, type, andlocation of ties included on the map. A visual example of the typology can be seen in thefindings section (figure 46).The next four sections describe the findings of the research. First, I show thecomposition of the personal communities, describing the different ties included by the studyparticipants. Then I present the findings by applying the typologies of personal communities.In this section I also explain the rationale used by the participants to construct and explaintheir personal communities. The last two sections elaborate on the characteristics,similarities and differences of the relationship with friends and family members.2.4. Findings2.4.1. Composition of personal communitiesWhen we try to characterize older people’s personal communities, we notice somegender differences in the size of the personal communities and the diversity of members inthe communities. Regarding size, women have larger personal communities than men (table2). Perhaps the most noticeable indicator reflecting the differences between men andwomen is the proportion of people with ten or less members versus those with more than43ten members in their personal communities. While men typically have ten or less members,women are likely to have more than ten members in their communities.Table 2: Size of personal communitiesNetwork Size Women MenAverage # members 16.9 9.9Mode 11 7Min. # members 7 3Max. # members 35 26≤10 members (%) 23.1 76.9> 10 members (%) 71.4 28.6Women not only have larger personal communities than do men, but also morediverse ones. Figure 3 illustrates this, showing all the types of ties that were included by theparticipants in their personal communities. The ties included by women in their maps includealmost every type of tie. This is not true for men. However, when looking at the mostcommonly mentioned ties, results are similar for men and women. They both had friends,children and siblings as the most commonly included ties included in their personalcommunities. It can also be seen that for both men and women descendant nuclear family,children and sibling in-laws and spouses are present in an important number of cases.44Figure 3: Personal community members in men and womenIt is interesting to notice that half of the female participants included theirgrandchildren in their personal communities, while only two male participants did so. On theother hand, men were more likely than women to include a niece or a nephew in theirpersonal communities. The figure of the “niece/nephew” will be discussed later in thischapter.The decision on who should be included on the map and how close to the centre themembers should be located was based on two complementary logics: affection andavailability given by physical closeness. The explanation by Rodrigo (67 years old) for of hismap shows how both logics work and how much is implied in the notion of peopleconsidered as important in one’s life:0 20 40 60 80 100GoddaughterGrandnieceColleagueFictive kinHouseworkerEx-spouse/partnerOther (no family)Step siblingOther relatives (not speacified)Step childPartnerNeighbourChild's parent in lawAunt/uncleCousinChild in lawGrandchildParentsSibling in lawNiece/nephewSpouseSiblingFriendsChildMap members by sexMap members (total women %)Map members (total men %)45[…] It has two meanings, an issue more related to the contact throughcommunication and, on the other hand, for the affection that one feels for theperson, the closeness… and affection that in some way also means necessity ofcounting with them. I mean, they are like part of your life; they are accompanying inyour own journey… eehh… You feel you need them beside you… I believe that talkingabout indispensability is going too far, but today, for instance if I stand today, for methey are very important in my life, for my life to have a meaning…The logic of affection relates mainly to emotional closeness, sometimes also referredas love. In this regard, participants differentiate between being physically close and often incontact from being emotionally close to someone. Such closeness was given by the trustconstructed over the years or by feelings of love toward nuclear family members.It is an intimacy criterion, I would say. I mean, for me to be able to talk about yourselfand listen the other with sincerity […] You don’t necessarily do that with yourchildren, but the children are the children, they are at the center of the heart andwith these people [the ones on the map], I mean all of them, I speak with an openheart and without limits… here I’m placing the people who really are my friends. I’mnot putting people with whom I have to be careful. (Loreto, 64 years old)The second logic was physical closeness in which, though emotional aspects wereconsidered, personal community members were mainly included due to the regularity of theface-to-face or by phone interaction.They [brothers] are more distant because we see each other less, talk less […] But mybrother, Pablo for instance, I would say every other day, every two days I call him byphone or he calls me. (Raul, 67 years old)Something interesting happened with the decision to not include certain ties,particularly in the case of partner and spouses in the personal communities. This did notalways reflect a bad relationship but, rather, the manner in which participants understoodthe exercise of naming the people who were important for them or underlying assumptionregarding the relationship. I will further elaborate this point in the section on family ties.46Although the ties included on the maps vary from person to person, the typologyproposed by Spencer and Pahl facilitates the identification of some categories (table 3).Figure 4 serves as a visual example of how the categories work. To respect anonimity andconfidentiality, the names on the labels have been replaced by the type of tie they represent.Figure 4: Typology of personal communitiesFriend-based personal communities are the most common among the participants, particularlyfriend-enveloped type of communities. However, the older group of men have equal distribution ofFamily-dependent, family-like communities, and friend-enveloped communities. The second mostcommon category in the younger group of men and the older group of women is the family-dependent one.Josefina, 64 years old47Table 3: Typology of personal communities (based on Spencer & Pahl’s typology)Friend-likeFriend-envelopedFamily-likeFamily-dependent Mixed11Women60-66 (n=13) 23% 38% 23% 15% 0%67-74 (n=10) 10% 50% 10% 30% 10%Men 60-66 (n= 6 ) 0% 67% 0% 33% 0%67-74 (n= 8) 13% 25% 25% 25% 13%2.4.2. Complementary typologyIn addition to Spencer and Pahl’s typology, I suggest a complementary classificationthat distinguished between ‘clustered' and 'hierarchical' maps. This classification wasoriginated from the overall visual difference among maps and considers the way in which theparticipants arranged the different ties on it in relationship to the centre. The table belowsummarize the findings using the proposed classification.Table 4: Proposed complementary typology of personal communitiesClustered Hierarchical MixedWomen 60-66 (n= 13) 15% 69% 15%67-74 (n= 10) 20% 70% 10%Men 60-66 (n= 6 ) 0% 100% 0%67-74 (n= 8) 13% 75% 13%The majority of participants designed maps with more or less clear hierarchies, i.e.members were clearly located on the different circles. In figure 4, the maps of Danilo,Fernando and Josefina represent hierarchical maps. However, six cases (one man and fivewomen) have clustered and mixed type of maps. Participants who designed clustered maps11 Friend-enveloped/ Family like48located the majority of their community members overlapped at the centre of the map andhad their communities composed by a variety of kin and non-kin ties. Lorena’s map in figure4 is a good example of a clustered map. The relevance of this complementary classificationwill be further elaborated in the discussion section.A typology of personal communities is a first approach to understanding the meaningand function of relationships as it gives a visual snapshot of the underlying rationale to defineimportant and emotionally close relationships. The next section further analyzes the tiesincluded as part of the personal communities, with particular attention to family andfriendship ties. It also describes the study participants’ reflections regarding the role ofdifferent types of kin and non-kin ties that were not necessarily included as part of theirpersonal communities.2.4.3. Friendship ties in personal communitiesThe study participants described the nature of their relationship with certain friends.In most of the cases, the friends included in their personal communities are people withwhom they have a common history and/or share things in common.a. Friends and common historyFernando (66 years old) uses the distinction between "long-time" and "short-time"friends to categorize the friends included in his network. The former are people whocontinue to be recognized as friends up to the present, despite of the many different lifestages and life experiences that have not necessarily been shared. In that sense, the event or49life period they shared in the past must have been sufficiently strong and important so thatthe friendship has been carried forward in time. In this regard, in the case of Chile, it isunquestionable the important role of shared experiences during the hardest years of thedictatorship, particularly for the group of men and women between 60 and 66 years old. Thatis the case of the friends included in the personal communities of Fernando, Josefina, Blanca,and Paulina. Some friendships were developed before the coup d’état with university matesand were nourished by militancy and experiences of prison and exile. Other friendship tieswere created during the period of exile and include both Chilean and foreign friends.Maria is another friend we always get together. Gloria for instance, is going to be afriend for life because we were imprisoned together, so we have, eh, and Javi is afriend from the university and we have all this history together, you see? I mean, withher family, exile, all that. (Paulina, 65 years old)However, cases in which friendship was created during school or university years, orchildren's school years are more common among the study participants. In fact, the majorityof the women between 67 and 74 years old mention friends they met many years ago, at theschool, in their childhood neighbourhood, and in the neighbourhood they used to live whenthey were just married. The relationship with some of those friends has been sporadic, butothers included among those ties people who have accompanied them through almost theirwhole lives.this are friends from all my life… because I’m from province, so you go to the school,you have worked in the same place together, eh, is like we have grown up together,we don’t see for a while, but we meet again and it’s as if we had never been apart.(Catalina, 68 years old)Both men and women, especially those in the cohort between 67 and 74 years old,also mentioned friends met during the time their children were at school. However, in some50cases the relationships have weakened as the children grow-up and the instances to spendtime together are less bounded to children's activities. The case of men in this cohort isinteresting as only three out of eight participants included friends in their personalcommunities. In two cases the circumstances that joined these men with their friends wererelated to the military service. Emilio (71 years old) and Raul (67 years old) mention how theshared extreme circumstances facilitated the creation of a close bond that lasted until today.There is another version of friendship with people with whom participants haveinteracted for a relatively long time. This type of friendship does not necessarily imply aspecific event shared in the further past. They are more about longstanding interaction withpeople who are not necessarily intimate confidants. These friends include neighbours andmembers of a group in which both individuals have participated for a long time (e.g.workplace). In that sense, this type of friendship has a component of a given tie as therelationship is based on the space and time that has continuously been shared. In otherwords, these are friends that are part of participants’ daily or weekly landscape, and fromthat constant –given– interaction a different bond has been developed.[…] short time friends are those friend that I met I would say in the last 10 years, forinstance Jose, or Paolo, or Francisco, ok, who are friends I knew at work and there wedeveloped a type of friendship that we are strengthening even more today, becausewe are more or less in the same page, almost all of us are in the same conditions.(Fernando, 66 years old).The distinction between short-time and long-time friends does not simply imply theamount of years people have known each other. Going back to Fernando’s reflection, short-time friends are those that were met during the current or most recent stage of their life(this participant retired 6 months before de interview). In that sense, long-time friends are51better defined by the amount of life stages that have been shared during people’s lives. Inthis manner, long-time friends represent long lasting relationships with people who havebeen more or less constant companions through different life stages and contexts. Some ofthese relationships may have disappeared for a time to be later reencountered along theway.b. Friends and shared interestsCommon interests gather people together to initiate friendship relationships. In somecases, the things in common are part of a shared history, but in others, new members areincorporated to the network as new interests arise. Loreto (64 years old) reflects on thefriends she incorporated in her personal community when asked about people with whomshe has more things in common.Wow! It’s a hard question, a little piece of each one… (Sigh) the thing is that in thismoment my central friend, with whom I feel more comfortable, is she [pointing to themap] who is relatively new […] because today, today for me is the ecological thing,the plants, that, so we are sharing that, the searching of those things.From the accounts of the participants we can recognize different types of friends whoconnect with different interests and needs: recreational, intellectual, professional, intimateconfessions, and support. A good example of the latter is Luciana (62 years old) whoidentifies “thematic friends” that match the variety of interests she has. Some of Luciana’sfriends are connected to a specific place, e.g. civic organizations, seniors’ clubs, and work.These place-bounded friends are common in men between 60 and 66 years old. The sharedinterest among the study participants and these friends are basically professional andrecreational. Some of the professional-based ties started as workplace interactions and with52time developed into a closer relationship. Other friendships started in a non-professionalcontext and it was the realization of the common interests and areas of work that helped tocreate a closer bond. However, some of these friends are left as organizational-boundedrelations, in the sense of being recognized as workplace friendships that do not cross theboundaries of the professional setting where they were established. The majority of maleparticipants of the younger age group talked during the interview about these types ofrelationships, but only three of them included these friends in their personal communities.The younger group of women participants have similar experiences, with cases oforganizational-bounded relations and relationships that have overcome the specific setting.Women between 67 and 74 years experience something similar with groups formed around aspecific area of interest. Whereas in some cases group participation is a good catalyzer todevelop more permanent friendships, in other cases this type of relationship is bounded tospecific topics and places without developing into a confidant tie. The latter, however, doesnot necessarily have a negative connotation. As Catalina (68 years old) expresses, these typeof friendships bring something different to her life, introducing variety to their relationshipsas they enable her to explore other areas that are appealing to her.2.4.4. Family ties in personal communitiesStudy participants mentioned the different types of relationships they have withdifferent family members included in their personal communities. Men mainly talk about therelationships they have with children, siblings, nephews, and in-laws. The figure of thenephew is interesting, especially in the younger group of men, as sometimes these relatives53take a place that is similar to a child. Two participants mention that they were involved inraising a nephew and receiving him in their home to support their studies. Women’s closerrelationships are usually described with siblings, cousins, children, and grandchildren.Just as people share different things with different friends, they develop a differentbond with different family members. Therefore, some elements of chosen ties can be foundin this given family ties. Issues of common interests, personality, similar world views andshared experiences explain the closeness or intensity of the relationship with one or anotherfamily member. An excellent example is the different affinity that exists with siblings.I meet more with him [pointing to the name of one of the brothers] than, for instance,my other brother [...] with Gabriela less than with the others, because there areaffinities that are or are not, even being such close relatives. (Fernando, 66 years old).Something similar happens with Pamela (60 years old), who describes a good andclose relationship with her sisters, all of them included as members of the personalcommunity. However, Pamela highlights that her more intimate issues are shared with onlyone of her sisters, the one she shares similar ideas, while the other sisters “live their livesdifferently to mine”.As we previously saw, children are among the main members of the personalcommunities and most of the participants with children included all their children asmembers of their personal communities. However, although all the children are regardedwith love, the emotional closeness depends on personality characteristics and life stages. Insome cases geographical distance and co-residence also plays a role in supporting thiscloseness.54My two daughters are very important for me, ok? Obviously, one lives with me, in thesame building, the other lives 15 minutes from here, and obviously I’m closer tothem, with the daughters more than the sons, ok? Then comes Lalo […] he is a lawyerand sees all that [legal] things […] I appreciate him very much but he is more distant,ok? And Andrés, my other son is shyer […] he doesn’t share with nobody; he is apart,because he is single. (Marisol, 73 years old)I have much affinity with Carolina, with Graciela too, the thing is that Graciela is toojealous, so she wants me only for her, with Simon yes, we get along, but he is toosimilar to his mom, so he has her character, he is very reserved with me. (Ramon, 69years old)The case of Cristina (62 years old) is a good example of how the relationship withchildren depends on their stage in life, since the presence of a partner and children alsoinfluences the time available and the priorities. When children are married or cohabitatingwith a partner, the affinity with that partner could influence the relationship with the child:If you see, my children are there too [pointing to the diagram]… both have their ownlives, their own families […] but they’re not so close now, I mean, I know they love meas much as I love them but… they’re not there when I need them. When they needme, yes, the first person they resort to is their mom! But… or they have an excuse orsomething… my daughter, I described her earlier, is quite pretty but unfortunatelyhas a husband who is a cave man, mmm… there was a time where we couldn’t go toher house, he didn’t allow me to call her by phone […] My son, as I told you, heturned 40 today, and has a family that I believe he idolizes.The inclusion of spouses and partners as members of the personal communitiesreflects very complex situations and logics. As mentioned in the previous section, not all thestudy participants included their spouses or partners. The case of Raul (67 years old) andSantiago (68 years old) illustrate one version of the latter situation. Their relationship withtheir wives could be described as bad in the case of Raul and just fine for Santiago, with nomajor complicity or sharing of daily life experiences. A very different version of people whodid not include their spouses is well exemplified by Emilio (71 years old) and Fernando (66years old). In their cases the relationship with their spouses was taken for granted,55representing unquestionably important ties that did not need to be mentioned. In the case ofwomen, on the other hand, some of those who were partnered but not cohabitating decidednot to include their partners explaining that the relationship was still very casual, “anamusement”, as one of the participants said. Another reason for not including the partners iswell represented by Marisol (73 years old), who indicated that they preferred to keep thefamily life fairly separated from their romantic relationships.Interestingly, other study participants gave a place in their personal communities topartners or spouses despite a dim or negative relationship. Cristina (62 years old) and Susana(73 years old) are good examples of the latter situation. They both live with their husbandsand declared themselves to be married, but at the time they began to talk about theirrelationship with their husbands, they clarified being actually separated. The relationshipwith their husbands is limited to mutual instrumental help and reception of economicsupport from their husbands, but the affective part of the relationship has been missing foryears. Cristina explainsWe still live in the same house. At the beginning it was a very bad relationship, butnow is civilized. I mean, there’re weekends we can share the lunch or have tea... andsince I’m pensioned he took care of all the household expenses because, really, Icannot, not that I do not want, I cannot.Another interesting situation is represented by Estefania, Loreto and Pamela from thegroup of women between 60 and 66 years old. They are not currently married, but declare agood relationship with their ex-spouses or ex-partners. For both, Estefania and Loreto thedecision is based on the family best interest, to keep children and parents in contact.However, it is not a forced relationship, and in fact they included their ex-husbands on the56map. The case of Pamela is different, as the ex-partner she included in the personalcommunity is not the father of her children but is part of the family "landscape" and herpersonal history.Most of the participants with children were involved in the care of theirgrandchildren but not without mixed feelings when issues of physical ability or personalityinterfered. Women were more likely than men to include their grandchildren, of differentages, among the members of their personal communities. The decision to includegrandchildren is a good example of the logic underlying the creation of the personalcommunity map. Some participants incorporated grandchildren because they love them andplay an active part taking care of them. Beyond the instrumental support given to theirgrandchildren, the study participants talked about “enjoying” them. For instance, Rodrigo (67years old) mentioned how much he appreciates the opportunity of sharing and witnessinghow their grandchildren grow up, which was not possible with their children in the past.Other interviewees decided to leave their grandchildren out of their personal communitiesbecause of their age. They recognized that they love their grandchildren but preferred tolimit their personal community to people with whom they talk and share common interests.For some participants with older grandchildren that was exactly the reason to include thosegrandchildren on their maps. For instance, Cristina and Pilar (from the younger cohort ofwomen) shared a similar experience, in which the bond with their grandchildren resembledmore a mother-child relationship than a grandmother-grandchild one. In both cases, theytook care of their grandchildren from birth, as their daughters got pregnant at an early age.57Today, their grandchildren are adults, but the special connection with them continues as wellas the exchange of emotional, practical, and material help.Study participants used the distinction of ‘chosen’ and ‘given’ ties to describedifferences between friends and family ties, respectively. These categories are nonethelessmore complex than the binary distinction can describe as components of chosen ties can befound in given ones, particularly in the relationship established with members of theextended family. In fact, while some interactions with extended family members are basedon customary relationships and contexts, such as family birthdays, in other cases, people findfriendship in extended family ties. Even if the tie is not defined as a friend and not includedon the personal community, the bond with some extended family members can becomeunconditional ties of deep affection. In male participants, this is particularly visible in theirrelationship with children's parents in law (‘consuegros’), “very very interesting get to knowmy son’s parents-in-law […] it has developed into a bond of closeness” (Rodrigo, 67 yearsold).Similar experiences are shared by women, although in their case the relationship isnot with the ‘consuegros’, but with their sisters-in-law, cousins, nieces and nephews. Rosa(74 years old) comments on the relationship she has with her sisters-in-law: “my sisters-in-law are friends, absolute friends, not only kinship, the sisters-in-law on my husband’s side,and two sisters-in-law on my brothers’ side”.582.5. DiscussionThis chapter has stressed the role of subjective meaning of personal relationships inlate life so as to recognize the diversity of ties that form older people’s emotionally closenetworks of relationships or personal communities. This discussion section seeks to elaborateon some of the findings in an attempt to explain some gender and cohort differences, as wellas differences in study participants’ experiences and definitions that emerged during theanalysis. Also, we will explore some implications of these findings.Despite the targeted scope of the research, the study findings can be compared withsimilar studies conducted in a different socio-cultural context. A study of personalcommunities conducted by Spencer and Pahl (2004) in England found that ties were selectedaccording to three criteria: whether or not the ties were family connections; the quality ofthe relationship; and, the context of the relationship (frequency of contact, duration of therelationship and sense of continuity). Similarly, the Chilean older people used criteria ofaffection and availability given by physical closeness. Interestingly, the inclusion of familyties, especially nuclear family ties, was explained in the context of the relationship(availability and continuity) and not necessarily by a cultural norm of familism. In that sense,people use a criterion that is not exclusive to family ties when deciding who to include aspart of their personal communities. Being available and having continuous relationships arecharacteristics that can be also found in neighbours and friends who live geographicallyclose. According to McDaniel and Gazso (2014), family norms and values do not disappear inlate modernity; instead, they get blurred. In the case of the Chilean people interviewed in59this study, the blurring of boundaries between kin and non-kin ties is visible not only infictive-kin relations, but also in the decision of including someone in a personal community.Regarding gender differences, we found that women’s communities were bigger insize and more diverse in their composition than those of men. However, when we look at themembers who are more usually included in the personal communities, we see similaritiesamong men and women. Although men and women have similar types of personalcommunities with a majority of friend-based communities, differences by age interactingwith gender are evident. Men between 67 and 74 years old have more family-based personalcommunities. The latter could be explained by a lack of opportunities or willingness toconnect with people outside the family circle after their retirement. The inclusion ofworkplace-bounded friendships as part of men’s personal communities supports thisexplanation. After retirement, the chances of keeping in touch with these friends as well asthe continuation of shared experiences that are mainly bounded to the work place routinemake it difficult for retired men to continue workplace-bounded relationships. Thus, thecomposition of the personal community would be affected, making it more family centered.Nonetheless, more research is needed to address the evolution of men’s personalcommunities and friendship ties after retirement.Another explanation for the prevalence of family-based personal communities inmen between 67 and 74 years old could be found in the logic they use to recognize friends asemotionally close ties, as the friends mentioned by the participants of this group werecharacterized as strong bonds created during key life stages. In that sense, the inclusion ofpeople regarded as friends and, moreover, as emotionally close friends, implies a very60selective criterion that cannot be easily achieved or replaced later in life. These findings alignwith the category of ‘friends as particular individuals’ proposed by Matthews (1983) in herstudy on friendship in old age. According to Matthews, these types of relationships refer tofriends acquired early in life –during middle age or before–, with whom daily life encountersare important. People who only have ‘particular individuals’ type of friendships have fewerresources to buffer the effects of old age and are more at risk of isolation (Matthews, 1983).Chilean older women and the younger cohort of men would be better prepared to adapt toexperiences of change and loss due to their ability to manage their relationships using avariety of strategies.A third significant finding of the analysis relates to the study participants whoincluded in their personal communities spouses from whom they were in fact separated orwith whom they maintained an emotionally distant relationship, such was the case ofCristina, Susana and Raul. This decision may reflect the influence of socio-cultural values,although more research is needed to corroborate the motives for the inclusion theserelationships. Based on secondary information, we can connect the finding to a moralimperative informing the inclusion of family ties. Under this logic, the inclusion of the spouseis done without further consideration as the spouse must be part of the circle in his or herrole of nuclear family member.The latter is reinforced by the highly regarded status of marriage in a country wheredivorce was legalized in 2004 and common law has only been legally recognized as January of2015 (one year after the interviews were conducted). Also, the role played by the spousewithin the household can influence his or her inclusion so as to recognize the degree of61dependency toward that person. Some study participants represent family trajectories thatare consistent with patterns of individualization or deinstitutionalization of marriage andfamily life (with experiences of divorce, singlehood or childless singlehood as a lifelongchoice). However, other participants show more traditional family arrangements and divisionof labour, exemplifying the outcomes of a United Nations Development Programme’s reportabout gender equality in Chile: “Underlying the majority representations is the image thatthe man is the main responsible for providing financial resources through work and ensuringorder through their participation in power, while the woman is in charge of the housework,parenting and the performance of the duties of caring for others” (United NationsDevelopment Programme, 2010, p. 15). In that context, a woman may include her husbandwithin her personal community, despite the quality of the relationship, because he is themain breadwinner of the house, or a man may include his wife recognizing her role as themain family carer.Context-bounded friendship, such as workplace or seniors’ club relationships, isanother interesting finding that can have practical implications. These types of friendshipsare similar to Matthews’ (1983) classification of ‘friends as relationships’. In thesefriendships, a shared biography is not needed and the benefits available in the relationshipare replaceable. Therefore, from a positive stance, these types of relations show that peoplecan develop new relationships in a context that facilitates permanent encounters. Also,context-bounded relationships enable sharing with people who have similar interests, whichare not necessarily present among family members. Thus, this type of friendship adds varietyof topics and opportunities of professional or personal growth.62According to Matthews, people who develop friendships with a focus on therelationship instead of the particular individual are less likely to be at risk of isolation.However, as the Chilean findings indicate, depending on variables such as other types ofrelationships the person has, the level of sociability, or the opportunities to interact withother people, context-bounded relationships are difficult to keep if the person ends his orher affiliation to the particular context. According to Kahn and Antonucci (1981), and in linewith the experiences of Chilean older men, people who mainly develop role-linkedrelationships are at greatest risk if the role is lost. However, as Matthews (1983) states andthe cases of Chilean older women show, some people are able to create new roles that linkthem to others, to capitalize different types of situations to adapt to the changes of old age.In summary, the experiences of Chilean older people demonstrated that currentdemographic context and values offer opportunities to develop different types ofrelationships (Phillipson, 2013). At the same time, this context and values, along with thechanges of growing old, pose challenges that can put at risk some individuals.A fifth outcome of the analysis that requires some further elaboration is the roleplayed by extended family members in older people’s personal communities. Some literaturehas described that ‘western’ values of individualism and secularization have influenced familyarrangements and have given a central role to nuclear family, emphasizing the emotionalbond between spouses, and parents and children (Aboderin, 2005). Although partners andchildren are among the main ties of older people’s personal communities, the results alsoshow that extended family play a particular role. In-laws and cousins are characterized bysome participants as friendship ties.63The relationship with grandchildren, especially in women, and nephews, in the caseof men, is particularly interesting. Interviewees, in their role of grandmothers or uncles, gaveaccounts that showed that extended family bonds have a central function in supplyingemotional, material help, and care. In her review of family ties in old age, Connidis (2009)noticed that uncles sometimes act as surrogate fathers, friends, mentors, or supplement thefunction of the parents. In the case of the Chilean men who commented on their relationshipwith nephews, their role can be described as mentors and supplement to parents. In theirrole as uncles they opened their homes to give their nephews an opportunity to study in thecapital. The cases of women in their role of grandmothers resemble more, at least at thebeginning, of a surrogate mother. They cared for their grandchildren to give an opportunityto their daughters to continue their studies. Their experiences serve as examples ofemergent family roles and obligations in unconventional family structures. A grandmothertaking care of her grandchildren is not new, but the content of the relationship developedwith their grandchildren, due to the focus on affective elements, is more in line withdescriptions of personal relationships in individualized societies. The relationship createdwith their grandchildren have remained close and strong as they grow old, and today thesegrandmothers are not only a potential source of material help, but also an emotionally closetie with whom personal and intimate issues are discussed.Finally, the findings show an interesting notion of friendship. While family is ‘given’and imposed, friendship, on the other hand, is a "constructed relationship" as Rosa (74 yearsold) mentioned. This idea highlights the element of personal choice involved in thedevelopment and maintenance of personal relationships. We can apply the notion of64constructed relationship not only to non-kin friends, but to any personal tie regarded aschosen and developed into a close relationship with features of friendship. The idea ofconstructed relationship involves a degree of complexity to the notion of friendship andchosen ties. Friendship is a constructed relationship that implies constant choices during aperson’s life trajectory within a specific time.Consciously or unconsciously, people weight the amount of involvement needed tokeep a relationship and whether such relationship is worth to be kept. In that sense, and inaccordance to the idea of convoy (Antonucci et al., 2011), networks of relationships are notstatic. These networks as well as particular relationships evolve as people evolve. Some tiesare important in their quality of “particular individuals” (Matthews, 1983), and are kept inrecognition of the similar life paths or the life events that have been shared. Otherrelationships are further developed when common interests or world perspectives arerecognized and thus, context-bounded and interest-specific relations are created. Yet otherconnections are put on hold and remain latent (Matthews, 1986) or are definitelyterminated. Therefore, friendship has different shades and functions that need to berecognized and explored. Not all friendship relationships can be analyzed with the samelenses, as the relationships vary depending on the nature and history of the relationship, thecontext in where it was created, and the reasons for the construction and maintenance ofthe relationship.652.6. Conclusion Chapter 2Matthews (1983) highlighted the importance of subjective definitions in her study offriendship in old age. The author noted that researchers can get a clearer understanding ofopportunities for friendship in old age by attending to the actual definitions used by socialactors. She argued that commonly used approaches to study friendship use features anddefinitions defined in advanced by researchers, obscuring relationships that did not fit withinthe frame defined by the study and ignoring the quality of respondents’ relationships. Thesame argument can be applied to other relationships, such as family ties. In line withMatthews’ comments, the purpose of this chapter was to recognize the subjective meaningand more subtle functions of personal relationships to identify the diversity of ties in olderpeople’s personal communities. Through the analysis, I have aimed to stress the importanceof emotional criteria in identifying the members of personal communities and increaseunderstanding of the underlying reasons for creating and maintaining different types ofpersonal relationships.The concept and typology of personal communities proposed by Spencer and Pahlproved useful to analyze the composition of older people’s network of close relationshipsand to recognize the importance of family members and friends in older people’s lives.Overall, it showed to be a helpful methodological tool to identify both active and latentrelationships. The relevance of the findings are conceptual, methodological and practical asthe assumed role of family as the first line of support (Phillipson, 2013; Shanas, 1979a)ignores individuals’ own definitions and experiences of family life. The approach of personalcommunities used to study Chilean older people helped to identify, using interviewees’ own66definitions, the diverse forms and meanings of personal relationships in late life. In doing so,it has been possible to identify ties that would remain obscured because they are taken forgranted interactions or because they do not supply easily identifiable forms of support. Theattention to meaning allowed the identification of very diverse personal networks thatinclude nuclear and extended family ties, different types of friends, and fictive kinshiprelations.A particular contribution of this chapter is the proposed complementary typologybased on the distinction between ‘clustered’ and ‘hierarchical’ personal communities. Thisdistinction gives additional useful information to depict and define emotional closeness. Thedecision of some participants to represent their communities by clustering relationships atthe centre of the personal communities diagram represents a sharp binary distinction thatdifferentiates between the people who are emotionally close and those who are not.Emotionally close people are included in the community without prioritizing or segmentingthe type or function of the relationship. In combination with Spencer and Pahl’s categories,this additional typology can be a good addition to represent suffusion of personalrelationships. In addition, the suffusion process represented in clustered personal communitycan be better understood through a consideration of the Chilean socio-cultural context.Issues of trust are more extensively discussed in the introduction of this thesis, but can bealso applied in the interpretation of the meaning of clustered personal communities. In theChilean context, where social trust is low, clustered personal communities may also mark adistinctive line between those who can be relied upon and those who cannot. Those who canbe relied upon are at equal level, blending family and non-family ties.67From a practical perspective, the typologies of personal communities as well as theattention to the content of the ties open an opportunity to identifying the strategies used byolder people to construct and maintain their social worlds. It also enables to notice genderand cohort differences in the way in which relationships are managed and defined. The roleof the socio-cultural context and values influencing the rationale of ‘clustered’ and‘hierarchical’ communities is an issue that can be further studied.From a methodological standpoint, the study has shown the complexities of capturingand understanding personal relationships and the importance of considering the meaningthat these relationships have in older people’s lives. The inclusion of ties that were notemotionally close relationships in the personal communities corroborate that not all personalrelationships are positive for older people. The latter should be identified when studyingolder people’s relationships. More so if we consider that even a methodological tool as openas the one applied in this research cannot entirely capture the quality of the relationship, thecomplex socio-cultural frames at work, and the biographical components of relationshipswithout further questioning and analysis.683. Bridging and bonding social capital in Chilean older people3.1. IntroductionTraditionally, in Chile as elsewhere, family members have been the main careproviders for older people. However changes in family composition, reduction of the averagenumber of children per woman, increased participation of women in the workforce, andmigration have affected the quantity and immediacy of the support available in late life;especially that exchanged among family members. In addition to these demographicchanges, some scholars perceive a decrease not only in the quantity but also in the quality ofolder people’s social capital due to the changing values associated with more individualizedsocieties (Fine, 2006; van Tilburg & Thomése, 2010).In a context of individualization, individuals reflexively seek relationships that are inline with their expected biographies (McDaniel & Gazso, 2014) and the exchange of help andsupport among family members is negotiated and guided by personal choice (Phillipson,2003). Some critics argue that this greater freedom of choice represents a problem,negatively impacting the reliability of the personal network to look after older familymembers (van Tilburg & Thomése, 2010; Winter, 2000). Traditionally, research on olderpeople’s social relationships has emphasized the dynamics inside their homes andrelationships with their family, with particular focus on availability of family care. Adominating perspective has been the notion of ‘family first’ with respect to older people’scare and support (Biggs & Powell, 2003; Phillipson, 2003; Shanas, 1979b) . This notion is alsopresent in the work of some social capital theorists, who regard family as a “bed-rock ofsocial capital” (Winter, 2000, p. 5).69Despite this perspective, a growing number of scholars recognize changes in familypractices and the emergence of different types of support and social solidarities that are asimportant as the nuclear family to older people’s lives (Allan, 2001; Phillipson, Bernard,Phillips, & Ogg, 2002). Greater freedom of choice and autonomy does not necessarily meanthe abandonment of practices such as reciprocity, trust and exchange of support (Phillipson,2003, p. 63). Instead, new strategies and networks of solidarity can be developed by elderlypeople to get emotional and instrumental support, social integration, as well as to constructtheir identities.This chapter aims to examine the broad spectrum of relationships that are relevant toolder people by analysing the bonding and bridging properties of their social capitalrelationships. The focus is on examining whether and under which circumstances differenttypes of personal ties are used by older people as bonding or bridging social capital attendingto the subjective experiences of older people. It is not the intention of this chapter toquantify the emotional and practical support provided by different type of ties, nor toprovide a definitive answer about whether family or friends provide more or less specifickinds of support.There are two central points of attention and assumptions in this chapter; the firstone is that we cannot make a clear distinction between bonding and bridging ties whenanalyzing the network of relationships of older people. Whether they provide emotional orpractical support, family ties and friends would act as bonding or bridging social capitalrelationships depending on the circumstance. Connected to the latter, a second key point isto explore how some ties can help the older individual to connect not only with practical70support, but also with emotional support and different life styles. This point is particularlyrelevant to better understand how friends differ from family members when managed asbridging social capital. To address these issues I draw upon the experiences of Chilean menand women between 60 and 74 years old and explore the function of different types ofpersonal ties, as well as whether and why the participants select specific types of ties to getemotional and instrumental support.3.2. Bridging and bonding social capitalFrom a structural approach, social capital has been usually theorized as beneficial;described as the glue and lubricant of community (Barrett, Hale, & Butler, 2013). However,studies on social capital have also shown its negative effects (Lynch, Due, Muntaner, & Smith,2000; Portes, 1998), as it can be coercive, inhibiting individual actions and choices (Cagney &Wen, 2008; M. Leonard, 2004). Taking this into consideration, the structural approach alsooffers an interesting distinction between bridging and bonding properties of social capital,which allows us to addresses the not always beneficial effects of the network of relationshipsof which we are part. “Bonding” social capital refers to inward-looking connections that“tend to reinforce exclusive identities and homogeneous groups” (Putnam, 2000, p. 22). Thistype of social capital reinforces exclusive identity and tight bonds of solidarity, trust andreciprocity (M. Leonard, 2004; Phillipson et al., 2004). The negative aspects of bonding socialcapital are particularly important when studying social capital in late life, where situations ofphysical dependency can occur. Keating and Dosman (2009, p. 314) talk about the costs ofbonding ties, as it can deteriorate family relationships, create feelings of exclusion, and evenlegal struggles in an economic context of limited formal care resources and increasing71demands related to population aging. Of course, bonding social capital can be also beneficialfor older people, as bonding ties are well suited to provide practical and emotional supporton a daily basis due to the intimate relationships among members of a group (Keating,Swindle, & Foster, 2005).Despite the different roles that bonding and bridging social capital can play in olderpeople experiences of aging, the studies addressing the different types of social capital in latelife have tended to focus on their health effects (Ferlander, 2007; Norstrand & Xu, 2011). Forinstance, Murayama and colleagues (2013) found that bonding social capital was negativelyassociated with both poor self-rated health and depressive mood in older people, whilethose who strongly perceived their networks as heterogeneous –used as evidence of bridgingsocial capital– were unlikely to be depressed and presented less cognitive decline than thosewith weak heterogeneous networks (indicative of bonding social capital). Based on a reviewof available literature, Ferlander (2007) observed that bonding and bridging social capitalhave both negative and positive health outcomes due to the different types of resources,support, influence and obligations they entail. For instance, bonding social capital tends toprovide emotional support, impacting mental health through stress reduction, but at thesame time it may create demands for conformity and restrict access to contacts andinformation resulting in unhealthy behaviours (Ferlander, 2007).Although not always applying a social capital perspective, gerontological research hastended to address issues connected with properties described for bonding social capitalwhen examining social support and informal caregiving in late life (Cagney & Wen, 2008).Here, the attention has been on family as the first line of support to care for older people,72stressing the importance of family relationships to provide support and care for old familymembers (Sheets, Bradley, & Hendricks, 2005). In this regard, family has been usuallyindicated as an important source of care for older people, as family ties are usuallycharacterized as emotionally intense and mutually confiding relationships among people whospend high amounts of time together, provide reciprocal services and have a strong sense ofshared identity (Keating & Dosman, 2009).Although bonding social capital can be an important resource for particular groups,its benefits are limited. For instance, in the case of older people, small and intense carenetworks can be ineffective to connect to resources and services outside of the immediatecare network of the older person (Keating et al., 2005). While bonding social capital isbeneficial for “getting by”, “bridging” social capital is central for “getting ahead” (de SouzaBriggs, 2003; Putnam, 2000). “Bridging” social capital is “outward looking and encompass[es]people across diverse social cleavages”(Putnam, 2000, p. 22). Thus, bridging social capitalrepresents open networks of relationships that link to external assets, being useful totransmit information and generate broader identities and reciprocities (Phillipson, 2003).Despite its relevance, bridging social capital has tended to be less discussed inrelationship to older people as it has tended to be more associated with assisting people withaccess to the labour force (Keating et al., 2005). In late life, bridging ties can be crucial to dealwith life transitions and to engage with new lifestyles. That is the case of people facingwidowhood or those who need help to deal with abusive relationships (Phillipson, 2013), asbridging ties can help to access resource and emotional support that are difficult to get frompeople directly connected to the spouse and close family members. Bridging ties can be also73important to adjust to post retirement changes. In this regard, Cornwell, Laumann andSchumm (2008) found in his study on social connectedness that retirement is related tocommunity involvement in late life. Although they do not elaborate about this topic, wecould hypothesize that older people resort to different ties that help them to be connectedto different areas of interest that cannot be covered by family relations.Using the distinction of bonding and bridging social capital, Keating and Dosman(2009) studied the care networks of frail older people. The authors explored care as provisionof practical support (e.g. housekeeping, meal preparation, grocery shopping, andtransportation). In their analysis, the authors distinguished six types of networks: lonespouse, children at home, spouse and children, close kin and friends, older diverse (friendswith smaller proportion of close kin and distant kin) and younger diverse (kin and friends,middle aged and younger, and employed). In line with the previous descriptions, the authorsfound that networks composed of close kin provide higher hours of care in comparison tofriend-and-family networks. This close kin networks were described as tightly knit andformed mainly of co-resident adult children and/or spouses with high normative obligationsto care for the frail older relative. The authors also notice that in intense caregivingsituations, bonding has its drawbacks as close-family networks are less likely to receiveformal support. Moreover, networks of close-family members that lived with the cared-forwere less likely to bridge to formal support. On the other hand, more heterogeneousnetworks, formed of family members and friends, were more likely to bridge to formalservices.74An additional insight that show us the relationship between bonding and bridging tiesis provided by the study conducted by Sixsmith and Boneham (2006) with older women livingin disadvantaged communities. Their approach focused on how such social capital wascreated, maintained and linked to health, attending also to the part played by older womenin helping family and community members. According to Sixsmith and Boneham (2006),bridging social capital can help in the construction of bonding social capital. These authorsstate that women who actively participated in activities that sought to improve theircommunities had the possibility to work in collaboration with other community groups.Through this community work the women not only connected with people that otherwisewould be absent from their personal networks but also developed feelings of empowermentand community belonging that reinforced their bonding social capital.In addition to providing practical support and connection with formal supportnetworks, the distinction between “bonding” and “bridging” can be of particular importanceto understand social capital in late life in a social context where personal relationships arebeing described as influenced by individualization (Beck, 1992a; Beck & Beck-Gernsheim,2002) or detraditionalization (Giddens, 1992, 1994). The latter means that traditional formsof representation and control coexist or are replaced by greater autonomy of action andfreedom of choice. In this context, personal relationships are described as forged based onvoluntariness rather than bounded by norms of moral obligation and familism (Fine, 2006;McDaniel & Gazso, 2014). From the perspective of older individuals, greater freedom ofchoice to manage their relationships and the resources embedded in their personal networkscan be extremely important to maintain their autonomy. According to Cornwell (2011, p.75782), dense social networks do not always appeal to older people as they can presentobstacles to privacy and autonomy due to higher levels of monitoring by the members of thenetwork.The study conducted by Sixsmith and Boneham (2006) also connects with issues ofautonomy when analyzing the role of reciprocity among parents and children. The authorsmention issues of ‘negotiated reciprocity’ to capture the complex feelings experienced byolder people when receiving help from their children. While some interviewees understoodand accepted support from their children as part of the parent-child relationship, others feltmore uncomfortable and needed to be persuaded to accept this support. I highlight thisconcept as it can be useful to better understand the different circumstances under whicholder people prefer to resort to different types of ties, that although categorized as bondingsocial capital, can be managed in such a way that they offer a more suitable or comfortablesolution for getting certain types of support.From a theoretical analysis of the context in which people grow old and its effect onthe inter- and intra-generational contract12, Phillipson (2003, 2013) states that the context ofan individualized or detraditionalized society represents an opportunity to reconstruct thefragmented pathways of late life, where social solidarities are restructured by reconcilingautonomy and interdependence. Also, the function of personal ties is different, as they arerelevant not only for their instrumental features but also because they work in a moresymbolic level. Relationships in the context of an individualized society would constitute12 Agreement of mutual aid between and within generations based on social norms and values (Reichert &Phillips, 2009).76today’s way for individuals to be “re-embedded into social life” (Fine, 2007: 185) through asense of certainty, continuity and belonging. In this manner, personal ties are not only thesource of potential resources to get help and support in times of crisis, but are relevant bythemselves (Victor, Scambler, & Bond, 2009).In fact, the latter descriptions of personal relationships in late life connect with therole of bridging social capital used by some authors. According to de Souza Briggs (2003, p.2), bridging ties establish connections not only for accessing material and more practicalresources, but also constitute bridges across roles, symbolic interests, and worldviews. In thisway, bridging ties are of key importance to develop broader identities and communities ofinterest (de Souza Briggs, 2003; Putnam, 2000), and therefore they play a central role in olderpeople’s life and experiences of aging in the context of modern societies.3.2.1. Distinguishing between bridging and bonding social capitalIn general, the basic distinction between bonding and bridging ties lays in the(dis)similarity between members of a network. However, the indicators of similarity anddissimilarity tend to vary depending on the study. For instance, Murayama and colleagues(2013) used age, gender, and socio-economic status as categories to measure degree ofhomogeneity and heterogeneity between a person and the members of his or her network.Norstrand and Xu (2011) assumed close ties among family members, friends and neighboursas indicators of bonding social capital, and expectations of support from organizations inwhich people participated as indicators of bridging social capital. In fact, it is not uncommonto find in the literature that key indicators of bonding or bridging social capital rest on the77assumption that bonding social capital can be equalized with strong ties (family and friends)and bridging social capital with weak ties (acquaintances and sometimes organizations)(Ferlander, 2007).Despite being conceptually different, the two set of ties act similarly and thus areassumed as equivalent, as both strong and bonding ties are good to provide emotional andinstrumental support, while weak and bridging ties are helpful to accessing to wideinformation (Ferlander, 2007). In gerontological research, the study conducted by Keatingand Dosman (2009, p. 302) also used a similar distinction, stating that “[b]ridging socialcapital is more heterogeneous and is based on weak ties among network members”. To befair, in the operationalization of the concepts Keating and Dosman focus their attention onthe structural characteristics of the care networks (gender, age, proportion of care networkcomprised by kin, geographical proximity, among others). Interestingly, in their analysis andcreation of a typology of six types of care networks, they stress the bridging properties ofnon-kin ties which facilitate the connection of the older person and its care network withformal care services. With this, the authors shift the attention from assuming (and defining)in advance that some ties are bonding or bridging, to describing bridging social capital basedon the likelihood of some ties connecting the older person to resources that are notnecessarily present in their more direct network of relationships.Although the binary distinction between bonding and bridging social capital helps usto capture the complex and diffuse character of social networks, it can also lead us to assumethat these two types of social capital are strictly mutually exclusive (Patulny & Lind HaaseSvendsen, 2007, p. 36). Thus, using a binary distinction can be problematic and not always78productive to advance our understanding of how older people manage their personal ties. AsFerlander (2007) indicates, although a family is usually composed of a network of strong ties,it can be bridging in terms of gender and age. Leonard and Onyx (2003, p. 194) summarizedthe distinctive features of bridging and bonding social capital and theorized that the former isassociated with loose networks, fairly strict reciprocity, thin trust, few shared values, andmore instrumental use. Bonding social capital, on the other hand, is characterized by densenetworks that serve a variety of functions –though they are less instrumentally used – thicktrust, shared values, and long term reciprocity. However, Leonard and Onyx (2003)recognized that in their research, bonding social capital was present in both strong and weakties.In this chapter I support the idea that a clear demarcation for classifying ties as eitherbonding or bridging can be difficult to establish and perhaps not always necessary, asassuming a binary distinction can diminish the richness of the analysis and the complexity ofthe relationships and their qualities. When analyzing social capital in late life, we may findthat attention to apparently homogeneous networks, such as family and friends, candisregard the array of resources they provide to the older person, and vice versa. In thatsense, depending on the stage in life and situation, family and friends can be described assimilar or dissimilar to the older person in terms of lifestyles and/or resources they can offeror help to connect. From a perspective of bridging ties as connecting structural holes innetworks, certain people can bridge gaps between networks that are not necessarily formedby dissimilar people (R. Leonard & Onyx, 2003), or at least not by obviously recognizablydissimilar people. That would be the case, for example, of an older person resorting to a child79(a tie usually referred as bonding and strong) to get a referral to hire or get advice from atrustworthy professional who works with that child.By attending to how (dis)similar ties (e.g. family and friends) act as bonding and/orbridging social capital depending on the circumstances and moment in life, this chapterexamines the broad spectrum of relationships that are relevant to older people. These tiesare relevant due to the practical and emotional support they provide as well as the differentlife styles they help to connect.3.3. MethodologyAs previously stated, a challenge of studying bonding and bridging social capitalrelationships is that there is no standard form of identifying or measuring them. A qualitativeapproach to social capital represents an opportunity to get insights regarding the blurredboundaries between bonding and bridging social capital. In the present study13 I avoided apriori definitions and categorizations regarding personal ties and social capital by attendingto participants’ definitions of who constitutes their network of important relationships. Thischapter explores the function of different types of personal ties (e.g. family, friends,neighbours, and organizations), examining whether and under which circumstances theseties are used as bonding or bridging social capital.A total of 40 men and women between 60 and 74 years old living in the city ofSantiago, Chile participated in in-depth semi-structured interviews. This method provided a13 The study was approved by the ethics committee of the University of British Columbia (UBC BREB Number:H13-02487).80flexible way to explore the different forms in which people identify and signify their personalties and recognize the resources present in their personal networks. The participants wereselected using first a purposive and then a maximum variation sampling strategies (Mason,2002). With a purposive sampling strategy, some early sampling decisions are made based onthe available literature. In the specific case of this study, some early decisions to select theparticipants were based on the literature on personal relationships in late life regardingcohort and gender differences. A maximum variation sampling technique seeks to engagedifferent types of participants to ensure that a variety of experiences and views are beingintegrated in the study. This strategy seeks to gain greater insights into a phenomenon byobserving it from different angles (Mujere, 2016), allowing the recognition of commonpatterns across cases with different characteristics (Patton, 1990). In this research, amaximum variation strategy sought to include the experiences and perspectives of thoseolder people who did/did not participate in older people’s organizations, people withdifferent living arrangements, marital status and socioeconomic status.The focus on people between 60 and 74 years old is relevant because it allows us toexplore the personal relationships of a young cohort of older people that partially includesbaby boomers; a cohort that has not yet been studied in Chile on matters concerning theirpersonal relations and social capital. In addition, this cohort is of great importance for publicpolicies because it would be the most affected by the decrease in family size and its effectson the availability of informal support.The participants were initially selected from a cohort of older people who hadparticipated in a series of focus groups conducted by the Domeyko Program on Aging of the81University of Chile in 2008. Only participants who were within the age range of 60 to 75 yearswere contacted. Because the Domeyko study was conducted more than five years prior tothis study, people aged 60 to 65 years old were missing from the sample pool. Also, someeligible participants could not be contacted due to change of phone number or healthproblems. In addition, to compose the sample a relatively balanced number of men andwomen were desirable but the list of eligible people from the Domeyko participants had littlerepresentation of men (40 women and 15 men). In total, only 10 people from the Domeykostudy could be contacted and were able to participate in the study. The other thirtyparticipants were reached through older people’s clubs and associations and then by thesnowball technique. The use of the snowball sampling strategy intended to reach people,particularly men, who did not participate in clubs and older people’s associations.The interviews consisted of one session of one to two hours length in a placemutually agreed by the researcher and the study participant. The interviews were conductedat people’s homes, coffee shops, seniors’ organizations, people’s workplace, and theUniversity of Chile. The interview guide included open-ended questions adapted from a studyon personal communities conducted by Spencer and Pahl (2006), a resource generator andquestions regarding generalized trust. The complete interview guide can be found inAppendix A. Questions about trust were based on those included in two surveys that havebeen conducted in Chile: the World Value Survey and Latinobarometro (e.g. Generallyspeaking, would you say that most people can be trusted or that you need to be very careful82in dealing with people?)14. The resource generator is a questionnaire to measure individualsocial capital that facilitates the identification of specific social resources (e.g. information,transportation, jobs) and the ties (e.g. family, friends, colleagues) by which such resourcesare accessed (Van Der Gaag & Snijders, 2005). An advantage of the resource generator is thatit does not require long interview sessions composed of multiple rounds of questions. Theitems work as a checklist that can be easily and quickly filled during the interview. However,the items of the resource generator are context-dependent, as ideas about the usefulness ofspecific social resources vary among countries and populations (Van Der Gaag & Snijders,2005). The latter means that the items included in the questionnaire need to be theoreticallyand empirically grounded in order to include a list of relevant and meaningful ties andsituations of help exchange (Ichirō Kawachi, Subramanian, & Kim, 2008).The categories of people who provided or received support and the items describingsituations where support was received or provided, were based on the resource generatordeveloped by Wellman and colleagues for the Connected Lives Project (2006). Thisquestionnaire was selected as the basis for our study due to the overall suitability of its itemsfor different types of populations and the simplicity of its application, as it was conceived tobe self-administered. To ensure that the list of situations were meaningful to the life stageand socio-cultural context of the participants, the items of the questionnaire were adjustedbased on the secondary qualitative analysis of the Domeyko focus groups and thepreliminary analysis of the first five interviews conducted at the beginning of the fieldwork. A14 This question has been also used in the General Social Survey, conducted by National Opinion ResearchCentre. The indicators of trust included in the GSS were developed by Kawachi and colleagues (I. Kawachi,Kennedy, Lochner, & Prothrow-Stith, 1997).83central issue was to create a questionnaire that could be easily understood and quicklyanswered, with situations to which participants could relate. Instead of giving all the possibletypes of kin and non-kin ties, general categories were used. These categories distinguishedbetween co-resident people, close family members, other relatives, neighbours, friends,organizations, and others.The first two categories of people who provided or received support are similar, asthey are mainly used to capture nuclear family members. However, these categoriesrepresent different people. The category of ‘co-resident’ comprises the people who live withthe older person in the same household. The study participants, usually included spouses andchildren, but others also included parents and friends, depending on the living arrangements.The category of ‘close family members’, on the other hand, refers to nuclear family membersfrom the family of orientation and procreation that live in a different household. Thiscategory was used by participants to include children, siblings and parents that were notliving in the same household.The questionnaire has a total of fifteen items that describe situations where the studyparticipants have received support in the last year (e.g. ‘advice on important issues’, ‘help tofix something at home’, and ‘information about recreational activities’)15. Nine of those itemsare about instrumental and practical support, including situations where the studyparticipants had received practical and material support: health issue, house chores, use acomputer, legal advice and paperwork, transportation (including accompanying to a place),15 For full questionnaire, please refer to Appendix A84emergency, watch over the house while the person was away, and little and greatereconomic help. Six items represent situations where the study participants receivedemotional help as well as support and information regarding recreational activities: counselon important issues, counsel on work, daily talk, information on recreation, support in familyissues, and company to events.The ‘constant comparative method’ of analysis was used from the beginning of thestudy. This method enabled making decisions regarding initial data production and makechanges as the research process developed (Charmaz, 2006; Corbin & Strauss, 2008).Verbatim transcriptions of the interviews were analyzed using constant comparative analysiswith the help of the software ATLAS.ti. The interviews were first fully read to recognizegeneral themes. In a second stage, data were segmented through open coding (Benaquisto,2008). A previous study conducted in Chile (Guzmán, Huenchuán, & Montes de Oca, 2003),indicated that there are gender differences in the type of ties with which older peopleexchange help. Therefore, the analysis of the interviews compared the experiences of menand women in order to identify differences in the ties involved in the exchange of supportand the situations in which support was exchanged.The codes that resulted from this process were of two types: in-vivo codes, which usethe language of the participants to name the categories, and significant codes, which use theresearcher’s language to reflect the focus-of-inquiry of the study (King, 2008). In a thirdstage, a more focused coding was conducted. Here, some codes were incorporated intobroader categories and others were refined creating and linking new codes that described85more specific dimensions. Some of the codes created during the second stage had thepurpose of identifying actors of the personal network (e.g. friend, child, and spouse) as theywere named at different moments of the interview. Other codes created in the second andthird stage were more analytical as they sought to identify qualities of relationships (e.g.confidant, companionship), situations of help exchanged, reflections and discussionregarding daily life experience, and conceptual themes. A specific code was created toidentify instances where the older person or a tie acted as a bridging tie (link someone with aspecific service, information, or other more knowledgeable or skilled person). Bonding socialcapital was recognized while analyzing the connection of the codes that identified actors withthe analytical codes. Specifically, bonding social capital was identified during the analysis oftypes of support given or received by the older person (how did they access or help to accessa specific resource), the quality of the relationship and how emotionally close thisrelationship was considered, and the subjective assessment of the influence of geographicalcloseness in turning to a specific person for support. The analysis of these codessupplements the findings of the resource generator.As previously discussed, identifying bonding and bridging ties presents somedifficulties. The aim of the analysis was not to develop self-contained categories, but insteadto recognize the instances in which certain types of ties could be considered as bondingand/or bridging. Ties were classified as bonding and/or bridging considering: the function ofthe tie by examining the type of support provided (e.g. information that require lookingbeyond the immediate support network, practical support with household chores, emotionalsupport with advice regarding family problems), whether the tie connected to resources that86were not directly present in the personal network, strength of a tie given by its emotionalcloseness to the older person (as defined by the participants16), and similarity or differencebetween the older person and the tie (e.g. kinship, age) with whom the help was exchanged.3.4. FindingsJust as we saw in the literature on bridging and bonding social capital, in our data wecan distinguish instances in which ties take a bonding or bridging role by attending to thesimilarity or difference of the ties regarding the older person in different aspects. Thecriterion of similarity, however, presents certain challenges. When examining older people’ssocial capital, we find that friends, children and spouses are the main providers of differenttypes of support (see tables 5 and 6 at the end of this chapter), as well as the people moreemotionally close to them. Depending on the situation, we may consider that friends andfamily share an array of characteristics with the study participants (criteria of similarity), thuscategorizing them as bonding ties. However, the analysis loses depth when we group theseties and label them, as other studies have done it, as the same type of bonding social capital.In fact, as indicated in chapter 2, the participants clearly recognize differences betweenfamily and friends when describing them as ‘chosen’ and ‘given’ ties with whom theyexchange different aspects of their lives. Not surprisingly, friends and family are usuallyseparated groups that are only (or could be potentially) connected through the older person.A qualitative approach is suitable then to recognizing the nuances of the support provided by16 The findings of the ‘personal communities’ methodology were particularly relevant to identifying the ties thatwere considered emotionally close by older people (see chapter 2: “It is an intimacy criterion”. Relationships inlate life from a personal communities approach”).87different types of ties and whether apparently similar kinds of ties connect the older personwith resources that could not be otherwise accessed.The following sections summarize the study findings divided in three topics addressedin the analysis: resorting to help from others, friends as bonding and bridging social capital,and family as bonding and bridging social capital. After this section, a discussion of the mainfindings in connection with the literature is provided.3.4.1 Resorting to help from othersBefore distinguishing the circumstances under which certain ties act as bonding andbridging, it is necessary to understand the frames used by study participants to talk aboutthose ties and the help exchanged through them. The central idea of this section is todescribe the problematic character that asking for or accepting help from others has in thediscourse of the participants. This is important as it allows us to better contextualize andunderstand the circumstances under which they use a tie considering its bonding or bridgingproperties to mobilize different types of resources.Talking with the study participants about help and support received was not astraightforward situation. The answers that people decided to share are a partial reflection ofa more complex process where personal experiences, family circumstances, and culturalcontext are at play. It is an issue that goes beyond the matter of who do I ask for what orwho do I help in which situation. Speaking about help during the interview was problematicfor some participants because the social status of being independent influences the way inwhich people value receiving help from others. Women and men both find asking for or88receiving help as problematic, as the idea of resorting to help from others is associated withnotions of material and/or physical dependency and laziness.There are different points of view regarding the issue of asking for help. While, forsome participants this is not at all complicated, for others is almost problematic and requiressome reflection due to the burden they may put on the relationship with other people, thestatus of dependency they might be seen to have, or the commitment this help may bring inthe future. As Patricio (60 years old) explains, "I don't want to complicate others' lives".Considering oneself as an independent person plays a role in the way people face theissue of help. In addition to trying not to bother others, some participants mention that theyneed to be consistent with how they have acted their whole life or that they need todemonstrate they can deal with things by themselves, “I can do this by myself” or “I camehere to be stressed alone”.In two cases, the importance of independent arrangement to solve problems wasstronger and asking for help to others was not an option. In this regard, women made notonly reference to their feelings toward receiving or asking for help, but also reflected in avery complex way on the circumstances and conditions for that. Paulina (65 years old) usesan informal Chilean expression (pedigüeña) to stress that she is not one of those people whoare always asking for something. Noticeably, despite their very different backgrounds,Paulina’s expression also exemplifies very well Estrellas's (61 years old) reflection "You haveto scratch with your own nails". However, this time the expression is not completely used tocharacterize an individual decision, but to stress how the context (particularly the state) has89not helped when she has faced hardships and how she does not trust the formal structuresto support her in the future. Estrella talks about this:They say [politicians] 'when you're 70 they will give you this and that'. Look, if youdon't save for your retirement, you don't have pension [...] You have to scratch withyour own nails, because if I don't work, I’ve always worked my backside off, I hadn'thad what I have.Three men share a similar preference to solve things by themselves. Interestingly,both support this preference with similar statements "I don't ask for favours" (Rene, 67 yearsold), "I'm reluctant to ask for help" (Juan, 74 years old). Interestingly, it is not that these menrejected help, but their conditions for being helped are based on an unspoken norm ofreciprocity:Look, I’m reluctant to ask for help. I offer my help, but it’s different, and as aresponse, I have received [help]. Because I’m not one of those who is asking forthings (Juan, 74 years old).Similarly, other study participants alluded to how reciprocity, mutual understanding,and some degree of empathy eventually act so that they can receive help, particularlyemotional support, without directly asking for it. All these characteristics seem to describe abonding tie:I: So, about the help issue. How does it work between these friends and you?Or with the people of the family? Do you tend to help with things, or you ask them forhelp?Emilio: In general no… let’s say, the family is not like that. Generally one knows that[he or she] has a problem and what [he or she] needs. That’s the concept at least forus. With the friends, sometimes they shout, ‘and what happen to you, tell me’. Andmany times one doesn’t even need money, but someone to listen to you. And goodthing is that they listen to you and nobody gives advice, unless one asks for it. Andone does the same, I mean, a true friend, and when there is also love in the family,the richest part of the human being: time. That is really yours; you give time and itshows. (Emilio, 71 years old)90When my husband died this brother-in-law paid all of my daughter’s universitystudies […] Being that we are not that close because he is a very public person, verybusy professionally, but when that happened, he said he was going to pay Lorena’suniversity up to the last year, I have always counted on them. Then Ricardo, let’s say,the other brother of my husband, ehh, he, with one of the brothers who died andwith my brother Alan, they took charge of the funeral, I knew nothing (…) no, I cancount on them, I’m very bad asking for help, they come […] When I’m complicated[with something] they come to help me, I’m very bad asking for help, it’s very hard forme. (Rosa, 74 years old)Boneham and Sixsmith (2006) used the notion of negotiated reciprocity. In line withthe findings of this research, Boneham and Sixsmith found that some participants expresseddifficulty in asking for or accepting support from children. Reciprocity was negotiated in twoways: in the expectation of mothers to be helped now by their children and in the need ofbeing persuaded by children to accept their support. In the case of the Chilean respondentswe can recognize similarities with the second type of negotiation. In fact, despite theproblematic character of asking for help, study participants do resort to others if need be. Insome cases, asking for help is a less problematic issue, but some conditions may apply:I would say that the offer of help eh, within the, the network of relationships one hasis always welcomed, always welcomed, and the offer that one can make to thisnetwork should always be welcomed, and I think that it is. I think that it always is, eh,there are no individual worlds, independent and isolated. No, there is not; one canalways give and receive, that’s the thing. (Fernando, 66 years old)From Fernando’s perspective, receiving help was unproblematic when such helpcame from close ties. Five other men made a similar comment, indicating that they have noproblem asking for help, alluding that people need others in order to accomplish things.Reciprocity plays a role not only in facilitating the reception of help, as Juan’s quoteshowed earlier, but it is also at the core of how asking for or accepting help is conceptualized.Study participants do not recognize an obligation between family members to exchange help.91They understand help exchange among close family members, and in some cases amongfriends, as part of a normal interaction that is not governed by a social norm. This is not thesame as saying that help exchange is taken for granted. Instead the constant flow of helpgiven and received acquires a natural character, similar to the notion of generalizedreciprocity (Sahlins, 1972; Swartz, 2009). I will comment more on these findings in thediscussion section.However, when the tie is weaker or the help required is negatively perceived(particularly asking for money), the participants become more self-aware of the need toreciprocate:Of course, so, no, no, excuse me, I think differently, to me it’s not a commitmentasking for help, it’s not commitment, less to the children, less to the children […] It’ssomething natural, and I feel committed to nothing, to nothing, less with them. Now,if a neighbour, for instance, asks me something that... sometimes I rather not to. Ihave to be willing to let the neighbour asks me something, but it’s not that I feelcommitted. (Danilo, 71 years old)I think that in front of the difficulties that one faces in life, the first thing is to do yourbest effort trying to solve your problems, but we are in front of a situation where youcannot solve, ehhh… I believe that one has to ask for help, one has to go and ask foradvice or help, and for that are the ties with who you feel you have the samecommitment. I mean, the other way around, no? I mean, one has to be open forsomeone who is close to you, is going to ask something, and you have to be open,and you also have to do the best effort to give support […], I believe that things aremutual, Ok?, I mean, one has to be willing to help, but also has one has to be willingto be helped by others, I mean, I believe that one cannot be selfish facing the other,of giving him the chance to show love, affect, appreciation […]. (Rodrigo, 67 years old)In both Rodrigo and Danilo asking for or accepting help opens the door for a twoways interaction. Despite Danilo’s reluctance to recognize this help as a type of commitment,at the end the reception of help implies that the person must be willing to return the favour92in the future. The latter mostly applies to people who are not part of the emotionally closenetwork of relationships, for example neighbours.These findings regarding when and with whom to exchange help are particularlyinteresting as the value given to independence can encourage people to use and cultivateties with bridging potential (Cornwell, 2011). However, as indicated by Leonard and Onyx(2003, p. 192), the expectation of return are more obvious, more immediate, and moreexplicit for bridging ties than they are for bonding ones. For the study participants, thismeans that using bridging ties could imply obligations to reciprocate, thus discouraging themfrom activating these ties. It is here, then, that resorting for support to ties that are bothbonding and bridging becomes necessary. As a result, family members and friends are at thecentre of older people’s bonding and bridging social capital. In the discussion section, I willelaborate on these findings in connection with the literature.The next sections examine whether and under which circumstances friends andfamily ties are used by older people as bonding or bridging social capital. In line with theliterature, the bridging capacity of family and friends considers situations where these tiesenable the older person to be connected with external resources present in other networks(Keating et al., 2005), such as referral, information and professional advice. In addition, weconsider that in some circumstances the bridging capacity of friendship ties is different fromthat of family ties, as the former connect the older person with different life styles as well asemotional support in specific situations.933.4.2. Friends as bonding and bridging tiesAlthough friends are part of the strong ties (perceived as emotionally close) that formolder people’s personal networks, they are not necessarily equivalent to bonding ties. AsFerlander (2007) indicates, some scholars frequently use bonding and bridging ties asequivalents of strong and weak ones. We can see in the study participants’ experiences thatfriends, particularly those of long duration or those emotionally close who live nearby, act asstrong and bonding ties, providing emotional and practical support. As such, some friends areregarded by study participants as similar to them, in terms of values and interests:The family, we like it or not, we have to accept it, the friends we choose them. Andusually we tend to choose people, except if it's a very special person, who will agreeto our tendencies, we can talk, we can chat, we can have small differences, but Ibelieve that we are close in values and ideals, in the same line. But at the same time,they have experiences and points of view different to us. (Julia, 74 years old)Of course there are differences [between family and friends], because there with thefamily… for example one doesn’t… have complicities and links of common lifeexperiences, complicities, no? That, that we have lived experiences together and thatone doesn’t, doesn’t have as topics with the family because no… they are friends’topics […]. (Luciana, 62 years old)Particularly for women, friends play an important role providing help in case ofemergencies and when transportation is required; 38% of women have received help withtransportation from friends in comparison to only 7% of men, and 35% of women and 14% ofmen have received help from friends in case of an emergency. Friends are also an importantsource of emotional support for both men and women, with 42% of women and 29% of menindicating that friends provide them with daily talk. When looking at the number, we see thatthere are similarities in the type of support provided by friends and family members.However, when we attend to the experiences of the study participants we can see that the94type of counsel and daily talk exchanged between friends and the older person are notnecessarily the same as those exchanged with family members. Luciana and Julia’s quotesare example of the ways in which family members and friends are regarded and how thisimpacts the types of topic that are shared when seeking emotional support.At the same time, friends work as bridging ties, facilitating access to information,providing an array of advice in practical and professional issues, and acting as a bridge toaccessing services. Friends, in comparison to other ties, are the preferred source ofinformation and counsel for both men and women: 42% of women and 36% of men haveasked for or received counsel on important issues from friends, and 31% of women and 36%of men have received or asked for counsel on work from them. Especially for women, friendsare the preferred source of information on recreational activities, with 35% of women and14% of men who indicated they have received or asked for such information to friends.Friends are also among the most common ties with which study participants have attendedto events, with 42% of women and 29% of men indicating that they have asked friends toaccompany them to events (e.g. movies, theatre, concerts, and celebrations).According to Phillipson (2013, p. 123), non-kin ties are central to deal with periodssuch as retirement and old age, as those ties help older men and women to find and connectwith new lifestyles. In many cases, the exchange of information on recreational activities isdone with friends from organized groups and older people’s clubs. For instance, Catalina (68years old) comments:95Yes, my friends, because here everybody, very good ties are created here, so there isalways a friend ‘hey!, there is a concert, I got tickets, do you want to come?’, Ok?, doyou know what I mean?, so in that circle you create very good ties.Also, friends’ particular skills and expertise make them important to obtaininformation and counsel on legal and financial issues.When there is something like that we told him, we ask him ehh… for advice. Let’s saythere is an investment we can make, something, something to buy… I don’t know,there is so much legal stuff […]. (Alberto, 71 years old)In a similar vein, Paulina (65 years old) explains how a friend of hers helped to fix aproblem with the social security contributions she made to a former house worker:[…] a home worker I used to have, it looks as if I hadn’t paid the social securitycontributions, for some months, but I always paid her because I, myself, signed herinto the AFP […] what happens is that I gave the money to her [home worker] and shehas to go to… […] So, and it happens that they charge you an interests rate […] so I’llgo now that I have free time, with this I will go and I will solve this issue, because afriend told me, this friend told me ‘you know that they don’t sign you in’, socialservices don’t sign you in, but I have all the emails I sent, everything, so…I: So this friend, she knows about this topic or is she a lawyer?P: She worked, no, she works at the labour inspectorate…The referral that friends can make is relevant when the help required involvesdelicate issues, for instance, information that will be disclosed or the amount of money thatwill be spent:[…] before my brother from Talca did that, but not anymore, but now I’m, I want tomake all that again, so I asked friends to recommend me people. My cousin too whois an architect made the design… a friend is helping me, to see a handyman who hasmade some work for him […]. (Paulina, 65 years old)I: And there has been a situation where you have resorted first to a friend, to someparticular acquaintance or neighbour before asking for help to your close family?Mercedes: Yes.Danilo: At least we have done it…Mercedes: To Diego when we first…96Danilo: How is the name of this guy I am telling you, Diego who work at the hospitaland everything. When we were ill from this, to him I tell “hey what doctor, whatspecialist, this and that, where’, where because to know, to unclog the…Mercedes: coronary…Danilo: the coronary, we asked him where, ‘what do you think is the best, here andthat’, so he said ‘no, no, no, you cannot get lost, J.J. Aguirre that is from theUniversity of Chile’, he said. (Mercedes, 71 years old and Danilo, 65 years old)From these experiences, there is another way of understanding the bridging qualityof friendship ties in the context of old age. As Souza Briggs (2003, p. 2) indicates, bridging tiesnot only help to access practical resources and information, but also to connect across roles,symbolic interests, and worldviews. In that sense, friendship ties help to address the diversityof personal and professional concerns and interests that are not necessarily shared withfamily ties. Study participants have different types of friends that connect with differentneeds, such as recreational, intellectual, professional, and intimate confessions:Everybody needs, besides having a family, even if it is a nice relationship, but needshaving a circle of friends, either labor or social [friends]. (Juan, 74 years old)it’s the case that the family cannot give you, let’s say, the knowledge eh, work-relatedor cannot give you the possibility of a job, that is what the friends give, that you gotto know in that realm […] they supplement, let’s say, what others cannot give you andthat is the knowledge. I tell you for instance, here with him, with Jose, with Fabian,there is a knowledge, even technical, that we have […]we have the same profession,except Claudio who is a mechanical engineer, no, but eh, Jose, Fabian, and I arecommercial engineers, so we form a, there is a relationship, let’s say, that is also arelationship of technical knowledge, that no one of those within the family circle canprovide. (Fernando, 66 years old)Something similar occurred in other realms of life. Some participants preferred, forinstance, to share and discuss contested topics, intimate areas of their life, and/or themes ofparticular interest with their friends. For both men and women, friends occupied a moreintimate place in their lives as they give non-judgemental advice and offer an objectiveperspective when personal problems are shared:97Friend are more objective than the family […] obviously, and that gives them a pointin favor, eh, the family has affection but also subjectivity, that’s it, advice is alwayspreventive, are, always useful, the friends, uh, the friends tells you things straightforward. I mean, if you like it good, if you don’t good too. (Matias, 60)… let’s see, with these friends I have shared the activism, life and intellectually. Mybrother is a physician, so he is boring, no, he is a private doctor, well, he can talkabout everything, about architecture, theatre, all, but I share affectively in actualitywith my family circle, they are, we have, I have a similar cultural matrix, so let’s say,we have fundamental agreements. Now, the other friends they are there to sharemore things, ah, it’s more, more, yes more relaxed regarding the things that are atplay in the family. That is more complicated, the friends are more, one has thefreedom, one has chosen them, and has chosen them again, plus in general I havelong histories with everybody […] I share my affect with them all but of course, theyhave more political dimensions, more cultural, eh, and complicities, let’s say, of otherkind. (Josefina, 74 years old)[…] but the trust, the intimacy, sometimes the family does not talk truly, but they talksocially and that is boring for me, the ‘small talk’ … That thing of talking for the sakeof it, I… so, the friends you can also talk for the sake of it for a moment, but then youcan talk, so I like it when people show themselves authentic, and that cannot befound always in the family, is like, that routine gets set and you cannot go further… sothat for me is frustrating, is like a frustration, and that does not happen with friends,except with the friendship relations that are dying, are deteriorated, that you getfrustrated […]. (Loreto, 64 years old)Taking into account the type of perspectives and advice that friends offer comparedwith family members, we can understand the preference for friends as a source of emotionalsupport, providing counsel regarding important issues. Friends also offer an outsider’sperspective when study participants face problems within their families. For both men andwomen friends are among the preferred ties to resort for support in case of family issues;31% of women have resorted to friends in these cases, being friends the second preferredsource of support after close family members (54%). Men share their preference betweenhousehold members, close family members, and friends, each one with 21%.98From the perspective of the participants, friends offer leeway (“a broad sleeve”) that allowssharing intimate insights that cannot be shared with family members:There are things you cannot talk and you share the, only with friends. Things that in…you don’t tell normally, you are not going to tell. (Emilio, 71 years old)Not all things can be talked with the family, because it’s bounded, it’s conditioned,freedom gets in the middle. So, with more reason, if there is affinity in what we aretalking, just as you search a specialist in health, also one has to know who to resort…As I was telling you, I among therapist friends that I have, we tell our health problemsand help one each other, we exchange therapies, but I have friends who areaccountants-auditors, economist, that are topics that I like too and we meet to talkabout that. (Juan, 74 years old)Friends are the preferred source of advice when unbiased and straightforwardopinions that cannot be provided by family members due to the specific topic underdiscussion and/or for being the family members too closely involved in the problem. That isthe case of certain life transitions, such as divorce or separation. For instance, Luciana (62years old) explains that she has a very close relationship with her daughter and son, and theyusually share intimate topics. However, when she decided to divorce, she preferred tochannel the emotional problems associated with that process with different people:Sure, all the divorce situation I didn’t resort to my daughter, because she was alsoaffected […] and there my son, with my son, like we live that together, with them wewere very… we contained each other until we could, because he also had a process,like fear, anger and… but there my friends, my work, my friends and my work weretherapeutic.Another reason for preferring friends over family members to discuss certain types ofpersonal and family problems is based on the avoidance of conflicting topics and situations.Because family is given for life, there are more opportunities for frequent meeting (e.g.holidays, birthdays, family lunch) and therefore maintaining a conflict-free relationship is ofcentral importance. More about this issue is discussed below.993.4.3. Family as bonding and bridging tiesFamily ties are usually regarded as strong ties and bonding social capital, as family iscomposed of people of similar background and socio-demographic characteristics. However,as Julia’s quote showed in a previous section, the ‘given’ quality of family members does notimply that they would be similar to one’s interests and lifestyles, nor they would benecessarily the closest and preferred ties to get emotional support. The previous sectionsshowed that it is for this reason that friends add variety in older people’s network. Friendsalso act as bridges that connect the older person with different lifestyles, as wells as adviceon issues that are not connected, or should not be discussed, with close family members.Thus, friends not only give practical help and information but also variety of topics andexperiences to draw upon when advice is needed. Of course, family members share acommon background with the older person, particularly those of the nuclear family.However, the preference of family ties for certain types of help has to do sometimes morewith their availability than with the emotionally closeness among relatives:Let's say in case of a medical emergency, I was alone in Santiago at that moment so Icalled my brother, because my children and my partner were out of town. I called mybrother, but... mainly my brother, my partner, my children, yes […]. (Alma, 60 yearsold)I: And in those situations, to understand how it Works, who do you think that people,for instance from your close family, resort to you for some types of help, such as incase of an emergency?Patricio: Let’s see, because we have quite a bit of communication with the members,eh, I’m seeing them daily, every day I see them. (Patricio, 60 years old)Thus, for everyday help or urgent unexpected issues, the participants ask the peoplewho live close to them or have the possibility to reach them quickly. Close family (i.e.100children, spouse, and siblings) members are the preferred source of support in case of anemergency or when they themselves face or a household member faces health issues; 50% ofwomen and 36% of men have received help from close family members when facing healthissues, and 46% of women and 36% of men have received support from close familymembers in emergency situations. A total of 23 study participants have co-resident childrenand 20 live with their spouse or partner. One person who does not have co-resident childrenhas co-resident adult grandchildren. These household members had provided help inresponse to health issues to 31% of women and 36% of men, and to 42% of women and 21%of men in cases of an emergency. In the provision of emotional support, family members arein daily communication with the study participants; 50% of women and 64% of men rely onhousehold members for daily interaction, and 38% of women and 14% of men rely on closefamily members.The preferred ties to ask for or receive help are similar for both men and women,who mention spouses, and children. However, we can find gender differences as women, butnot men, also include friends among the preferred ties for support. In the case of men, theyprefer resorting to close family members due to their physical proximity and because familymembers are able to perceive when they need help without needing an explicit request orfurther explanation. As mentioned in the section on the issue of help, reciprocity also plays arole in the reception of support, as an unspoken reciprocity norm enables men to get helpwithout the need to asking for it.There is not only a preference for certain ties to receive or ask for help depending onthe type of support needed, but also some situations are more problematic than others,101particularly asking for economic help or care related support. Regarding the latter, a tensionexists when ties with close people need to be used. On one hand, close ties, such as children,are preferred due to their availability as well as the possibility of receiving help even withoutasking for it. On the other hand, the notion of not being a burden usually applies to therelationship between the study participants and their children.Literature on social capital indicates that family members are among the mainsources of bonding social capital, being especially suited to provide care and functional help(Gray, 2008; Phillipson, 2013; Sheets et al., 2005). However, the aforementioned tension putinto question how suitable some bonding ties are to provide certain kind of help. Theexperiences of study participants regarding close family ties and provision of help indicatethat the physical and emotional closeness between older people and some family members,particularly children and spouses, facilitate the reception of support. However, the studyparticipants are at the same time reluctant to compromise their own and their children’sindependence. In that sense, the bonding social capital present in close family ties is, fromstudy participants’ experiences, best suited to get support with daily practical things andemergencies.Despite the latter, some participants do seek assistance from their children foreconomic help and to solve family problems. These participants are in general more familyoriented and with smaller networks of close friends than other interviewees. They also seemmore reserved about their personal problems and prefer to discuss them with their closefamily circle:102I: Have you received any kind of help in case of a family issue, or you have solved thatby yourself, or you help each other?Marisol: We help each other; dirty laundry is washed at home…I: And has anyone helped you in case of an emergency, something urgent?Marisol: Yes, of course… […] Lisa, the youngest one […] let’s see, first of all monetary,she is the one who helps me monetarily, and… I help her sometimes with the children[…] (Marisol, 73 years old).[…] I get along better with the girls, the girls, the ones when one has a problem I tellthem, but with other people no, even the uncles, they are far and they are old sowhat are they going to give me […]. (Cristian 73)I believe that it is to not complicate others’ existence, so we sit at the table every day,we meet around 8 pm, my daughter arrives from the university, my wife from herwork, and there we always have something to talk, and we open up, we always talk atthe table and it has always been like that. We have never exteriorized our problems.(Patricio, 60 years old)Issues of boundaries can be found in the account of study participants, both men andwomen. Some of the characteristics used to describe the difference between friendship andfamily relations have in common the belief that some aspects of life cannot be shared withfamily members or with friends. In the case of family, whether a person likes it or not, theconnection is continuous. Even when the relation with a specific family member hasdeteriorated, chances are that that member will be present in a family gathering. Also, thereare some relationships that are worth keeping in good shape or at least neutral because theirdeterioration may imply distancing from a loved one. Marisol (73 years old) mentions howshe prefers not to get involved in issues related to raising his grandson to avoid problemswith her daughter:It’s a very difficult kid [grandson], very, very […] he makes me nervous, very nervousbecause the kid is, he’s not spoiled […] he has no hierarchy, he doesn’t consider you,he is like me, me, me, everything. He gets tantrums, is, he can tell you any bad word,he can tell you anything, so he needs a lot of contention, you know? So I don’t, if Iwant to get involved because ‘look this brat, this and that’, so I don’t want to haveproblems with her [daughter]…103Lorena (65 years old) takes a similar stance in her relationship with her son-in-law:Javier [son-in-law] is not a bad person, he is childish though, very childish, but I sayone thing, well if I see them fine it’s OK. I close my eyes and look to other place, andthey manage as they want but, I don’t, well, in every married couple, I don’t getinvolved. If I can help in what I can, but getting involved in their discussions, theirproblems, that’s their problem […]Situation as the ones described by Marisol and Lorena shed light on the complex andsometimes negative aspects of bonding relationships, which health effects have been studiedby different scholars (Ferlander, 2007; Murayama et al., 2013).So far, I have described the bonding properties of family ties. As the quotes of Danilo,Rosa and Emilio indicated in the first part of this result section, emotionally close familymembers are attentive to their needs. This closeness facilitates the natural exchange ofcertain kinds of support, allowing participants to receive help without feeling that they haveacquired a debt that must be repaid. In that sense, certain family ties, particularly childrenand spouse, can be regarded as bonding ties based on tight bonds of trust and reciprocity (tothe point of making reciprocity invisible) and shared values.Nonetheless, family ties can also be analyzed as playing a bridging role in olderpeople’s personal networks. Although Lancee (2010) indicates that bridging ties imply acrosscutting network of thin trust, Ferlander (2007) observed that some family members,regarded by the participants as emotionally close ties, can act as bridging ties, for instancesin terms of gender and age. From the experiences of the study participants we can observethat family members are an important source of counsel, as well as helpful ties to accessinformation and services. Here, the bridging quality of children has to do with their age, butnot only in terms of connecting different generations within the family, but also because they104link the older person to public spheres of life from which they are more distanced afterretirement.Although friends are the preferred ties when advice about work issues is needed,close family members are also important with 12% of women and 21% of men indicating thatthey have asked for or received help from close family members on work-related issues.Similarly, close family members are central when help with legal paperwork is required, with27% of women and 29% of men resorting to these ties. Some study participants mentionedthat they have children and siblings who are lawyers or are acquainted with legal issues,which make them a preferred source of support to provide or connect with a service:Recently no, because I have not needed [help with legal issues], but when myhusband passed away yes, there, my brother who is a lawyer and everybody, they didthe inheritance paperwork17, this brother-in-law did it too […]. (Rosa, 74 years old)Well, now, recently I was hospitalized, and a friend of my daughter, who is aprofessional of that field, came to take care of me. She cares for elderly people andhas taken some courses at the University where she studied (Susana, 73 years old).The professional knowledge and position of close relatives was also relevant to accessingservices. One study participant, Julio (66 years old) talks about this:[…] I had some support from my brother because, because he was a worker at thehospital J. Aguirre, so is like, like the contacts there […] no economic [problem], butonly making the contacts, in the case, to get me a doctor […]Similarly, Lorena (65 years old) mentions how the fact of having her daughter working at aprimary health service helps her to have a better quality medical attention:Trust in primary care centres, well, my daughter works in a public clinic, so that youcan trust but, they don’t treat me bad17 She actually says “posesión efectiva”, which is a legal procedure to dispose of the assets and debts of adeceased person.105Interestingly, Lorena makes reference to the role that having her daughter working atthe public clinic plays in enhancing her own experience using that service. Due to that,Lorena also mentions that she, recognizing her advantage, can trust in public clinics. Bondingties that act as bridges to access certain services are particularly important in context of lowinstitutional trust; such is the case of Chile. The connection of these findings with theliterature is further elaborated in the discussion section.3.5. DiscussionThe experiences of this particular group of Chilean older men and women weresummarized in three analytical themes that enabled us to explore the function of friends andfamily ties and the circumstances under which friends and family ties are selected toexchange particular kinds of emotional and practical help. In general, the results indicate thatfriends and family ties play a double role as bonding and bridging ties depending on thesituation. However, an interesting finding was that friends, in comparison to family ties,represent a different type of bridge that connects with specific interests and broaderidentities of the older person. In that sense, friendship ties add variety to the studyparticipants’ networks, not only in term of practical resources, but also emotionally. In fact,the interviewees have different types of friends that support them in different forms, e.g.with recreation, counsel on family and intimate issues, and professional advice.The latter finding is in line with perspectives that understand social capital as aresource that helps bridging across roles, identities, worldviews and symbolic interests (deSouza Briggs, 2003; Putnam, 2000). Our findings corroborate those of others research106indicating that relationships with family and friends work differently. Both types of tiesprovide emotional support, but family relations play a more prominent role in providingpractical support, while friends provide companionship, reaffirmation of self-regard,resilience, and social integration (Huxhold, Miche, & Schüz, 2014; Messeri, Silverstein, &Litwak, 1993; Wiles, Wild, Kerse, & Allen, 2012). In addition to the bridging qualities offriendship ties, the study findings show that friends also support with help usually attributedto bonding ties, as they constitute an important source of help for something as key asassisting in case of an emergency and as providers of emotional support.It was stated when reviewing the different definitions of social capital that themobilization of resources embedded in the networks of relationships is not necessarily arational calculated action. In fact, as Bourdieu (1986) and Small (2010) state that, theunderlying objectives or the strategies used by individuals are not necessarily consciouslypursued. Our results regarding altruistic forms of help show that the latter is even more so atthe level of family relationships and it is perhaps one of the more characteristics features offamily members when used in their quality of bonding ties. In fact, our study findings showthat reciprocity is not necessarily an expected outcome of exchanging help. The experiencesshared by Danilo, Rosa and Emilio showed that practical help exchanged among emotionallyclose family members acquires a natural quality, which is not the same as saying that thishelp is taken for granted nor it is anticipated based on a notion of debt. Indeed, exchange ofhelp in these cases flows without the need of asking for it and even without expecting anyspecific form of support.107Regarding the latter, chapter 4 shows how study participants are against normative(mandatory) forms of family obligation. In addition, the findings of the current chapter showin some cases the type of negotiated reciprocity (Boneham & Sixsmith, 2006) where parentshave to be persuaded by their children to accept support. What we observe in our findings isthat bonding social capital, particularly that present in family ties, enables mobilization ofresources based on an unspoken rule of reciprocity. This rule is similar to the notion ofgeneralized reciprocity, defined as an altruistic form of help in which the time, quantity, orquality of the expected return is indefinite (Sahlins, 1972; Swartz, 2009). Obtaining support inthis manner results more comfortable for those people who value their autonomy and/orwant to protect their privacy. The findings show that particularly older men depend more onthis type of reciprocity to access and mobilize resources.The importance of having social capital with bonding and bridging qualities had beenstated by Putnam (2000) when mentioning their importance to “getting by” and “gettingahead”. The aforementioned findings can be connected with other studies have supportedthe importance of both types of social capital indicating that dense social networks are notalways beneficial in late life or might be not always appealing to older people (Cornwell,2011; Ferlander, 2007; Murayama et al., 2013). Cornwell (2011) indicates that maintainingautonomy and privacy can be problematic in dense networks. According to this author, olderpeople value their independence so receiving excessive support causes feeling ofvulnerability, decreases self-esteem, and emotional distress. The latter creates a tensionbetween the value given by older people to being independent and the benefits that dense108networks have in terms of care and support (Cornwell, 2011, p. 783). Interestingly, thistension, discussed in chapter 3, is also present in the study participants.Both bridging and bonding social capital are important not only to give access todiverse kinds of resources. As observed by Victor, Scambler, and Bond (2009), personal tiesnot only represent a source of potential resources to face crisis or emergencies, but also arerelevant by themselves. Personal ties help to reconcile autonomy and interdependence(Phillipson, 2013), by providing support and a sense of belonging, while enabling people toconstruct and negotiate their social identities. The latter is essential when considering theimportance of the resources embedded in older people’s networks, as they are not onlyrelevant to provide instrumental help but also, and particularly, to give affective support andcounsel. In that sense, bridging ties, and specifically friends acting as such, are central to getahead when facing important changes. In line with Phillipson’s (2013) statement, our findingsshow that the outward looking character of friendship ties enables the study participants todeal with life transitions, so that they can get emotional help in areas and situations that aredifficult to address with family members. That was the case of Luciana when dealing with theprocess of divorce. Not surprisingly, research has shown that the role of friends as providersof emotional support and trusted confidants is central for the psychological wellbeing of theolder person (Huxhold et al., 2014; Takahashi et al., 1997). The latter means that even inthose cases where health issues can complicate the exchange of instrumental help, friendsare still a central resource to ensure the wellbeing of older people.109The role of friends and family ties (siblings and children) as bridges to access promptand better services is connected with issues of trust. Leonard and Onyx (2003) found thatpeople are more willing to take risks to access other networks for information and resourceswhen they used trusted intermediaries. In that sense, family members and friends are farfrom being a crosscutting network of thin trust, in the sense of bridging ties asconceptualized by Lancee (2010). In the case of this group of Chilean interviewees, friendsand family members represent trusted intermediaries who improve older people’s chancesto access the right professional or to get better public services, while reducing concerns ofbeing over charged or not being considered as a priority client or beneficiary. This findingshould be understood as part of a major context, as in Chile the levels of institutional andgeneralized trust are low (ICSO, 2014; Latinobarometro, 2013). In such a scenario, olderindividuals rarely would resort to people and organizations that are completely disconnectedfrom their strong ties, which explain the importance of using family members and friends asbridges.The way in which the analysis has been conducted makes it difficult to establish cleargender differences. Attention was paid to how older people mobilized the resourcesembedded in their personal networks. In that sense, the logic and process used by older menand women are similar. However, the bridging potential of their networks is different.Chapter 2 discussed the gender difference in size and composition of older people’s personalnetworks (in the form of personal communities), indicating that women have bigger andmore diverse networks, which also include more friends, than men. The resource generatoroffers similar results, with women exchanging more types of help with more diverse kind of110people than men. According to Cornwell (2011), these type of network structures wouldincrease the bridging potential of women’s networks. When analyzing the instancesmentioned by study participants where they mobilized bridging social capital, we find thatsimilar proportion of men (36%) and women (31%) have received help by using bridging ties.However, in actuality more proportion of women (31%) than men (14%) have acted asbridges to help family, friends and even acquaintances, most of the times linkingunconnected ties (i.e. kin with non-kin). In this manner, Cornwell ‘s (2011) statement seemsto apply to the case of this particular group of Chilean older people. According to Cornwell(2011) socio-emotional selectivity theory could explain the bridging potential of olderwomen’s networks in contrast to older men’s. Older men’s networks become moreemotionally oriented as they grow old, focusing more on strong rewarding relationships thatprovide emotionally fulfilment (Cornwell, 2011, p. 791). Our findings offer only a snapshot ofa particular group of Chilean older people. More research is needed to identify how bondingand bridging ties change over time and may differ in men and women as people facedifferent life stages and transitions..3.6. Conclusion Chapter 3This chapter aimed to examine the relationships that were relevant to older peoplethrough an analysis of their bridging and bonding social capital relationships. The experiencesof this particular group of Chilean older men and women enabled us to research into thecircumstances under which some specific types of ties are selected to get particular kinds ofemotional and practical help, and how under certain circumstances bonding ties can also actas bridging ties.111In the case of this particular group of Chilean older people, both family and friends,usually perceived as emotionally close ties by the older person, act as bridging and bondingties. As bonding ties, they represent trustworthy people with whom help exchange is lessproblematic due to an unspoken rule of reciprocity and shared values that allow helping oneanother without the need of asking for it aloud. If we consider their function, family tiesrepresent a different form of bonding social capital. Family members appear as the preferredsource of practical help due to their availability (geographical closeness and continuity ofmeeting). Friends, on the other hand, are usually described as people with whomparticipants share a history, values and/or interest. This gives to friends a specializedknowledge of study participants’ needs and likes, facilitating the exchange of support.Both friends and family members–including the older person– act as bridging tieswhen specific skills and knowledge are needed. This shows that bridging qualities can be aproperty of similar people, as friends and family members connect gaps in information (R.Leonard & Onyx, 2003; Widmer, 2006) as well as benefit the older person by helping them toaccess professional advice and specific services. As observed by Leonard and Onyx (2003),children bridge across structural categories in terms of age. Here, nonetheless, they not onlybridge an older generation with a younger one. Children also link people connected to publicspheres of life through work and active professional engagement, with those who havewillingly or unwillingly withdrawn from those spheres due to retirement or a preference tobe more actively involved in domestic issues.The findings of this study have conceptual relevance. In line with other studiesconducted in family and community settings (Ferlander, 2007; R. Leonard & Onyx, 2003), the112experiences of the study participants showed that bonding and bridging social capital werepresent in strong ties (close family members and friends). Although, the distinction of bothtypes of social capital is important, the findings indicate that we cannot assume that in latelife strong ties are more important or useful than loose ones or vice versa. Also, whenattending to social capital in late life, the findings indicate that bridging ties can connectpeople in different ways. The way in which family ties and friends connect older people withdifferent areas of need and interests is a good example of this.Similarly, it cannot be assumed that the help exchanged with family member andfriends is of the same type or is signified as equivalent. As an example, trust works differentlyin older people’s relationships with family and friendship ties. Older people share animportant part of their lives, resources and emotion with emotionally close family membersbut, due to continuous encounters and interaction, these ties are unsuited for sharingintimate issues or family problems. It is in those circumstances when friends appear as moresuitable options for emotional support. Thus, methodologically, it is important to avoidgrouping family members and friends within the same category of ‘strong’ ties. It is theinterplay of their specific qualities and functions what makes them more or less beneficial forolder people, giving them access to a variety of material, practical, and emotional resources.Although this cross-sectional study does not track changes in the type and sources ofsupport across the life of the participants, it provides a starting point to explore how peopleperceive and feel about different ties at different life stages. In that sense, this study makes acontribution to the social capital literature by adding a life course perspective to better113understand the importance and differences of ties described as close in their role of bondingand bridging social capital relationships.114Table 5: Emotional support received by participantsType of emotional support Ties Women (%) Men (%)Counsel on important issues Household members 35 14Close family 35 21Other relatives 12 0Neighbours 4 0Friends 42 36Organizations 4 0Other 12 7Counsel on work Household members 12 0Close family 12 21Other relatives 4 7Neighbours 0 0Friends 31 36Organizations 12 7Other 4 7Daily talk Household members 50 64Close family 38 14Other relatives 19 7Neighbours 4 0Friends 42 29Organizations 0 0Other 0 7Information on recreation Household members 8 7Close family 8 7Other relatives 4 7Neighbours 4 14Friends 35 14Organizations 35 29Other 19 14Support in family issues Household members 19 21Close family 54 21Other relatives 15 7Neighbours 0 0Friends 31 21Organizations 0 7Other 15 7Company to events Household members 23 29Close family 38 21Other relatives 12 14Neighbours 4 7Friends 42 29Organizations 4 0Other 4 0115Table 6: Practical support received by participantsType of practical support Ties Women (%) Men (%)Health issue Household members 31 36Close family 50 36Other relatives 12 0Neighbours 4 7Friends 19 0Organizations 12 0Other 15 7House chores Household members 38 29Close family 27 14Other relatives 19 7Neighbours 4 0Friends 12 0Organizations 8 0Other 15 14Computer use Household members 38 21Close family 38 21Other relatives 15 21Neighbours 4 0Friends 8 14Organizations 8 14Other 12 21Legal paperwork Household members 8 7Close family 27 29Other relatives 12 0Neighbours 0 0Friends 27 43Organizations 19 7Other 8 21Transportation (includingaccompanying to a place)Household members 35 21Close family 19 36Other relatives 4 7Neighbours 4 0Friends 38 7Organizations 0 0Other 8 0Emergency Household members 42 21Close family 46 36Other relatives 23 14Neighbours 8 21Friends 35 14Organizations 8 0Other 15 7Watch house while away Household members 23 14Close family 31 29Other relatives 4 21Neighbours 23 36Friends 8 0Organizations 4 0Other 19 14116Type of practical support Ties Women (%) Men (%)Little economic help(up to Ch$ 50,000 or US$85)Household members 35 21Close family 42 21Other relatives 23 0Neighbours 12 0Friends 12 7Organizations 0 0Other 4 0Greater economic help Household members 12 7Close family 38 14Other relatives 12 7Neighbours 4 0Friends 15 0Organizations 15 14Other 0 71174. Public policy and experiences of aging: social relationships and social integration inChilean policy on aging[…] the older adults themselves, and the society in general, ought totake charge of their aging process. Each one of us is responsible for our lifeand, consequently, of the way in which we decide to live this stage. (RosaKornfeld, former director of SENAMA18)4.1. IntroductionWhereas developed countries experienced a gradual population aging and had manydecades to adapt, the majority of developing countries are experiencing, or are projected toexperience, a more rapid transformation of their age structures. This means that developingcountries, such as Chile, have less time to adjust to the consequences of this change and theywill have to deal with the challenges and requirements of an aging population while facingother persistent social problems that require attention and resources (Higo & Williamson,2011). In fact, according to a report edited by the Chilean researcher Sandra Huenchuan forECLAC (2009), the Latin American aging process raises concern for being faster than theprocess experienced by developed countries, and occurring in a more vulnerable context.Huenchuan argues that Latin America presents persistent inequality, weak institutionalframeworks development, welfare systems with low coverage and quality, as well as a familyinstitution burdened by obligations regarding the safety and protection of its members.Acknowledging the challenges of an aging population, the Chilean governmentdeveloped a formal institutional framework to address the needs of Chile’s older population.This started with the creation in 1995 of the National Commission for Older Adults (CNAM)18 Speech during the presentation of the Integral Policy for Positive Aging 2012-2025 (Ministerio de DesarrolloSocial, 2012)118and was broadened in its areas of action in 2012 when the CNAM became the NationalService for Older People (SENAMA). The areas of institutional concern of SENAMA currentlyinclude economic security, health, quality of life, and autonomy of elderly people (Calvo,2013). In 2012 SENAMA presented a document called Integral Policy for Positive Aging (2012-2025) that recognized the need of a comprehensive understanding of the aging process andlater life by offering a series of guidelines to develop programs and services directed to anincreasing older population.The purpose of this chapter is to examine what forms of social integration arefostered by Chilean policy on aging in comparison to the ideas and practices relative topersonal relationships and social integration perceived by older Chilean people. In thischapter, the concept of social integration is understood as older people’s network of socialconnections and participation in meaningful roles (Pillemer, 2000, p. 8). Based on Scharf’s(1998) distinction between integration through formal and informal relationships, the formsof social integration explored in this chapter are: social integration using formal relationships,e.g. participation in the labour market, and social integration through informal relationships,represented by family and friendship networks. The policy analysis also examines to whatextent the nature and quality of contacts with family members, friends and other relevantties are considered and supported by current Chilean policy on aging.I have taken the document of the Integral Policy for Positive Aging (2012-2025) as thebasis for my analysis because it represents the state’s understanding of the challenges of anaging society and the role of different actors (public, private, families, and individuals) in119facing those challenges. At the same time, public policies represent discourses of a specificcontext and historical time that suggest standards, modes of understanding of old age andthe experiences associated with growing old (Grenier, 2012; Laliberte Rudman, 2006). Theanalysis of the specific policy document is compared and contrasted with the result of 40qualitative interviews conducted with Chilean people between 60 and 75 years old. Inexamining the different forms of social integration, special attention is paid to whether thediverse types of personal relationships relevant to older people’s lives are recognized andsupported by policy definitions and programs.4.2. Public policy and experiences of agingLiterature on the political economy of aging stresses the role of social structures inthe construction of old age (Estes & Phillipson, 2002; Phillipson, 2005; Walker, 2006). Thisapproach understands old age and its problems as socially produced. Attention is paid to therole of public policy and the welfare state in shaping the ‘reality’ of aging and old age, as thatcontext presents opportunities and exerts pressures and constraints that influence theexperiences of growing old (Biggs, 2001; Phillipson, 2005).Public policy can also be understood from a more cultural perspective. Here, theinfluence of policy on the experiences of aging is also recognized, but at the same time policyitself is seen as a socio-cultural product. As Grenier (2012, p. 64) states “policy solidifies aconsensus of the day into practice”. In that sense, public policy is a reflection of the time andparticular socio-cultural context in which it was developed and implemented. According toKraft and Furlong (2012, p. 4), public policy is “a course of government action or inaction in120response to public problems” that represents the most important values of a society andreflects which of many different values have the highest priority for appropriate courses ofaction.In a similar fashion, Oszlack and O’Donell (1995, pp. 112–113) define public policy as“a set of actions and omissions that represent a determined form of state intervention inrelationship to an issue that arouses the attention, interest or mobilization of other socialactors”. These authors state that a normative orientation can be inferred from policies;orientations that are expected to influence the course of the social process previouslydeveloped around a specific issue. Thus, public policies can determine with the force of lawwhat social values –sometimes conflicting social values– will prevail. As a result, policies caninfluence people’s lives since policy definitions and practices create typified constructs, in thephenomenological sense, that influence social and individual interpretations of theexperiences of growing old (Grenier, 2012).4.2.1. Public policies, social integration and social relationships in late lifeIn relationship to the aging population, few studies have analyzed how social policieshelp or hinder the social integration of older people, and those scholars who have addressedthis issue have done it analyzing the policies and evidence of developed countries (Biggs,2005; Biggs & Powell, 2003; Reichert & Phillips, 2009; Scharf, 1998). Therefore, the role ofpublic policy in fostering social integration of older people living in developing countriesremains largely understudied.121Among the available literature, we can distinguish two broad categories of studiesregarding the influence of public policies in the ways in which older people are integrated insociety through, for instance, public and private organizations, and their personal networks.The first category of studies compares and contrasts socio-demographic data with existentaging and family policy or with policy debates. Some research within the first group of studiesidentifies areas for policy debates by investigating the changes in the generational contractand its effects on family solidarity to analyze whether social policy can strengthen familysolidarity to avoid the “war of generations”. Examining British and German policy, Reichertand Phillips (2009) address issues such as: whether the welfare orientation of the Policyaddresses individual needs or fosters family tradition; to what extent voluntaristicrelationships are fostered by policy and whether diverse family structures are recognized;what types of exchange are recognized and promoted by public policy; what policy instancespromote intra and intergenerational solidarity; how are independence and care addressed bypublic policy, and, what measures are available to promote inclusion and preventdiscrimination of older people.We can also classify the work of Scharf (1998) within this group of studies. Scharf(1998) analyzes the role of German social policy in promoting or hindering older people’ssocial integration, contrasting social policy focus with data on formal and informalrelationships, quality of contacts, and feelings of loneliness and isolation in late life. In thismanner, the author assesses whether public policies guarantee social integration of olderpeople or act as a source of exclusion.122The second category does a similar analysis but taking a narrative approach toanalyze public policy. Here, public policy is understood as a meaning frame that influencesindividuals’ interpretations and practices. Biggs and Powell (2003, p. 116) state: “policiesprovide narrative templates within which certain categories of person or groups areencouraged to live out their lives”. In that sense, the issue is how policies legitimate certainidentities by creating social spaces and providing the materials for the enactment of thoseidentities.This chapter is situated in the second category of studies, as it uses a narrativeapproach to compare and contrast older people’s experiences with the definitions andactions of Chilean policy on aging. However, the insights offered by the first group of studiesare also relevant due to the topics under study, which help to specify primary and secondaryquestions used to guide the analysis of the Policy document. This analysis was summarized infour themes: ‘tone’ of the policy, different instances of integration for dependent andindependent people, definitions of autonomy and self-management, and the Policy’sassumptions of family responsibilities and involvement in older people’s support. Theseprimary and secondary questions and their relationship with the themes are betteraddressed in the methodology section.4.2.2. Policy as a frame of meaningAccording to Clark (1993, p. 13), public policy can be defined as “the attempt tobalance competing notions of the responsibility of individuals, families, and the state indeveloping programs to meet human needs”. The author argues that every policy123perspective, statement or recommendation represents a story or sub-story within thebroader narrative discourse about a gripping policy problem (Clark, 2011, p. 84). Policies canbe also defined as selections of reality based on specific frames that select out some parts ofreality while leaving others out (Fischer, 2003, p. 144). Empirical facts and values influencethe definition of a public policy problem and, at the same time, the relationship between thevalue and fact dimension shapes the solution of the problem. Clark (2011, p. 85) states: “Thevalue dimension represents cherished principles or beliefs that are affected in some way bythe empirical state of affairs”.Focusing on the United Kingdom, Biggs and Powell (2003) used a narrative approachto analyze the rhetoric and representation of the relationship between family and olderpeople in social policy. The authors compared neoliberal and social democratic family policyto stress the ideological continuities and discontinuities present in such policies. Biggs andPowell (2003, pp. 103–119) highlighted the contradictory narratives present in these policies.Neoliberal policies represented two contradictory ideas of family and its relationship witholder people. On one hand, policies represented family as independent from thegovernment, as a realm of care supported by traditional family values. On the other hand,neoliberal policies depicted family as a context where obligation toward older members wasavoided, resulting in their mistreatment. Thus, greater surveillance from the government wasrequired.Biggs and Powell (2003) also observed that the subsequent social democratic policieschanged the focus toward older people, who were depicted as active citizens who should beencouraged to participate in and contribute to society as producers and consumers.124However, social policy contained conflictive narratives. On one hand, social policy includedan implied hedonism of aging lifestyles based on consumption and productivity. On the otherhand, the policy discourse and actions assumed that older people can actively, and willingly,participate in society through grandparenting. Despite the differences in focus, the narrativesof both periods of policies had a common strand, which allowed them to coexist. In liberalfamily policies, the focus was not on the aging individuals, but on family and governmentresponsibilities toward them. Then, in the social democratic period, Biggs and Powell (2003)found that the narrative of active citizens gave older people a secondary status, as theircontribution benefited other players who were central to society. In that way, both policiesmasked the “authentic tasks of aging” (Biggs & Powell, 2003, p. 114).Grenier (2012) also used a narrative approach to analyze the way in which lifetransitions and old age were represented in social policies from the United Kingdom andCanada. The author found two competing discourses on aging, one that focused on itsbiomedical or functional features, and other emphasizing its healthy, productive, and socialaspects. These discourses overlapped in the structures and practices related to health andillness, though they gave conflicting messages. On one hand, older people were instructed onhealthy and successful aging while, on the other hand, unavoidable loss of autonomy in oldage is contained in policy discourse. In that way, policy on aging had two polarized targetedpopulations, the younger and healthy older people defined as the ‘third age’, and those whoare older and affected by illnesses –usually classified as forming the ‘fourth age’. Grenier alsoindicated that successful models of aging integrated in policy definitions directed efforts125toward prevention of illness and chronic conditions, focusing more on changing individuals’behavior and lifestyles rather than the social conditions affecting them.Common to the analysis conducted by Biggs and Powell (2003) and Grenier (2012) isthe focus on assumptions informing policy definitions and areas of action, as well as thetaken for granted experiences of their targeted population. Both studies also compared thepolicy discourse and associated actions with the available evidence of older people’spractices and expectations. The identification of taken for granted definitions andexperiences of aging is a key starting point for examining how social integration in late life isconstructed by Chilean policy on aging. This enables us to recognize to what extent thetemplates provided by the Policy regarding formal and informal means of social integrationare similar to older Chilean people’s experiences and ideas of social integration.4.3. MethodologyThe document Integral Policy for Positive Aging 2012-2025 was analyzed using aninterpretative narrative approach that focuses on the expressions of social meaning presentin policies, such as beliefs and values that motivate specific policy definitions and actions.According to Fischer (2003, p. 142), an interpretive approach to policy inquiry examines theprocesses through which the policy meanings are transmitted, the intended audience ofthose meanings, and the interpretations of the readers. The analysis also seeks to makevisible the conflicting frames informing policy narratives and marginalized stories (Fischer,2003; Grenier, 2012). The recognition of the context in which policies are formulated is alsoan important part of narrative analysis. According to Riessman (2008, p. 8), “storytelling126occurs at a historical moment with its circulating discourses”. Similarly, Fischer (2003, p. 146)argues that policy issues arise in contexts that are part of broader political and economicsetting, which are located in a specific historical time.Despite the latter definitions of what a narrative analysis of social policies shouldinclude, there are few models of analysis to be followed. In general, the authors mentionedin the previous section directly state the policy issues, with no clear reference to the sourcesanalyzed or clear description of the steps followed for the analysis. One of the fewexceptions is s the work of Grenier (2012) analyzing life transitions. In many aspects, theanalysis I conducted is based on her work, where the author analyzed the gaps betweenmacro constructions of life transitions and the actual experiences of people.The analysis conducted on Chilean policy on aging was framed by three primaryquestions (first column of table 7) that addressed the definitions, practices, assumptions andunexpressed topics and issues regarding social relationships and social integrations in late life(Biggs & Powell, 2003; Grenier, 2012). For each primary question, secondary questions wereadded to examine specific topics (second column of table 7). These secondary questions aregrounded on the available literature. More specifically, they are based on the emergenttrends and issues identified by Reichert and Phillips (2009) in their analysis of policy debates.By asking primary and secondary questions, I seek to comprehend the Policy. With thismethod I aim to have a better understanding of the content and discourse of the Policy.Similar to how a researcher conducts an interview using guiding questions and follow-upquestions when necessary, the secondary questions —grounded in previous studies— are127used to anticipate areas of analysis connected with the research question. The secondaryquestions may or may not be answered from the content of the Policy, although they help asa guide for the analysis. This analysis is summarized in the four themes previouslymentioned, which address some of the areas examined through the secondary questions.The first theme (the ‘tone’ of the policy) arises from the analysis guided by theprimary question A. Primary question B guides the analysis that gives rise to the other threethemes regarding instances of integration for dependent and independent people,definitions of autonomy and self-management, and policy’s assumptions of familyresponsibilities and involvement in older people’s support. Primary question C allows ahigher level analysis that crosses the four themes, as it is concerned with the policyassumptions and the tensions that arise from those assumptions. In that sense, primaryquestion C is relevant not only to analyze the content of the Policy document but particularlyto acknowledge the tensions between the actual experiences and expectations of olderpeople and the type of late life represented and constructed by the policy definitions andactions (e.g. services and programs).128Table 7: Primary and secondary questions guiding the analysis of Chilean policy on agingPrimary questions Secondary questionsA. How is demographic changeand its impact addressed?What problems are defined and what is the ‘tone’ of that definition (e.g.crisis, challenge, opportunity)?Are effects of demographic change on personal relationships andavailability of care and support addressed?B. What is tacitly expressed inthe Policy regarding: diversity ofrelationships in late life, familysolidarity, intergenerationalcontract, isolation, autonomy,support and care, and formalmeans for social integration?What means for social integration in late life are recognized and fostered?Are personal relationships of different type fostered in any way?How is family defined?What is the role of family members? Are there specific family relationshipsfostered by the policy?Are ‘voluntaristic’ relationships recognized? How are they fostered?How is personal autonomy and independence addressed?How is family solidarity included in Policy definitions? How is it fostered?What type of exchange are recognized or promoted?C. What is assumed or taken forgranted in the policy documentregarding the latter issues? Whatremains unexpressed?What theoretical frameworks are being used? What are their assumptions?How do they inform policy practices?Are there conflicting narrative frames? What are the tensions betweenthem?The policy analysis started with a thorough reading of the document Integral Policyfor Positive Aging (2012-2025). The 122 page document constitutes a framework to guide theactions of different Ministries by outlining a series of measures and programs directed toolder people. The first reading of the Policy permitted the selection of relevant threadspresent in specific sections. The main criterion to select a section was that the content has tobe directly connected to the topic of personal relationships in late life (described or definedcertain ties) and means for social integration.The Policy contains three lines of action and thirteen specific objectives, summarizedin Table 8. I focused the analysis on ten of the specific objectives, as these were related tothe research question and directly involved the participation of older people or theirimmediate benefit for social integration. In addition, I analyzed the introduction section ofthe Policy. This consisted of different presentations written by government authorities and129the actual introduction, where relevant data, main definitions and objectives are stated.Other sections where older people were not immediate beneficiaries have not beenconsidered for the analysis; for example, sections about ‘research’, ‘systematization anddissemination of information’, and ‘training of specialists’. Also, the section that describeshow the programs will be assessed from a technical point of view was not considered for theanalysis.Table 8: Summary of lines of action and their corresponding specific objectives of the Integral Policyfor Positive Aging 2012-2025Lines of action Specific ObjectivesHealthy agingHealth and careHealthy lifestyleSpecialistsActive agingParticipationEducationEconomic securityTransportation & housingPositive agingAbuseAccess to justiceInformation & disseminationIdentityResearchSubjective well-beingAfter the thorough reading of the policy document significant codes were created toidentify the main definitions and topics of interest discussed in the Policy. Significant codesuse researcher’s language to reflect the focus-of-inquiry of the study (King, 2008). Thesignificant codes used in the analysis of the Policy document were based on the sub-questions presented in table 7.Memos were used during and after the coding process. In qualitative analysis, memosallow comparing data, exploring ideas regarding codes, and directing further data production130(Charmaz, 2006). Memos are useful analytical tools to aid the description and integration ofcodes. They can be used, for instance, to describe properties of a category, summarize ideas,and link concepts. In short, memos can be defined as “written records of analysis” (Corbin &Strauss, 2008, p. 117) . During the analysis of the Policy document, memos were used toidentify the main policy narratives as well as issues of interest due to their connection ordisconnection with the emergent trends and problems identified in the literature, and withthe issues addressed by the study participants.Although some thematic analysis was conducted, the main purpose of the codes wasto provide a starting point to identify explicit narrative threads. The contrasting andcomparison of these explicit narrative threads with the available literature and theexperiences of the study participants enabled the identification of assumptions andneglected issues. The interview data used for this comparison were part of a larger study onsocial capital in late life.The study participants were men and women between 60 and 74 years old living inthe city of Santiago, Chile. The focus on people of the younger generation of older people isrelevant because it allows the exploration of the personal relationships of the cohort of babyboomers, a cohort that has not yet been studied in Chile19. In addition, this cohort is of greatimportance for public policies because it would be more effected by the decrease in familysize and the potential problems of availability of informal support than previous cohortsstudied in Chile.19 Studies using data from the Survey on Health, Well-Being, and Aging (SABE) have analyzed older people’spersonal networks. However, it was conducted with people aged 60 years old and older in 2000.131The participants were initially selected from the younger cohort of older people whohad participated in a series of focus groups conducted by the Domeyko Program on Aging ofthe University of Chile in 2008. Only participants who were within the age range of 60 to 75years old were contacted. Forty-one participants from the Domeyko study had given consentto be reached for future studies. Due to changes in phone number and health problems, only10 people (7 women and 3 men) out of those 41 could participate. The other thirtyparticipants who formed the study sample were reached through older people’s clubs andassociations and using the snowball technique.The interview schedule included open-ended questions adapted from a study onpersonal communities conducted by Spencer and Pahl (2006) and a ‘resource generator’. Theresource generator is a questionnaire that asks about specific social resources that areaccessed by the individual through their personal ties (Van Der Gaag & Snijders, 2005). Theclose-ended questions of the resource generator are similar to those developed by Wellmanand colleagues for the Connected Lives Project (2006). This questionnaire was selected asthe basis for our study due to the overall suitability of its items for different types ofpopulations and the simplicity of its application, as it was conceived to be self-administered.To ensure that the list of situations were meaningful to the life stage and socio-culturalcontext of the participants, the items of the questionnaire were adjusted based on thesecondary qualitative analysis of the Domeyko focus group and the preliminary analysis ofthe first five interviews conducted at the beginning of the fieldwork. The questionnairesought to identify specific types of emotional and instrumental help exchanged with different132groups of people (household members, close family, other relatives, neighbours, friends,organizations, and other). The questionnaire is available in Appendix A.Verbatim transcriptions of the interviews were analyzed with the help of the softwareATLAS.ti using grounded theory principles. These principles enabled an iterative approach todata analysis to go from a general and descriptive organization of the information to aninterpretative analysis aided by the creation of codes, memos and diagrams (Corbin &Strauss, 2008). In this manner, during the process of analysis, data were first segmented intocodes and then integrated by establishing relationships among those codes with the aid ofmemos and networks view.The maps of personal communities were classified using Spencer and Pahl’s typology(2004), considering the number, type, and location of ties included on the diagram. In thisresearch four types of communities were identified: ‘Friend-like community’: friends have a central role and outnumber family members. ‘Friend-enveloped’: family members occupy the center of the map and friends surroundthat centre. ‘Family-like’: family members outnumber friends. ‘Family-dependent’: family members not only outnumber friends but also provide moresupport.The information provided by the personal communities method and the resourcegenerator permitted the identification of the different types of ties that form older people’spersonal networks, the kind of help exchanged with those people, the quality of the133relationships, and the people identified as emotionally close. All this information was used toidentify similarities and differences between older people’s experiences of social integrationand the definitions and actions present in the Policy.In the sections that follow I divide the findings in two parts. In the first part, I willdescribe the Policy and the context in which it was developed. I shall present the findings ofthe policy analysis organized in the four themes that summarize the analysis conducted usingthe primary and secondary questions included in table 7. The first theme, ‘tone’ of the policy,responds to the first main question (how is demographic change and its impact addressed?)and it gives a better understanding of the content of the Policy, complementing the sectionthat describes the institutional framework under which the Policy was developed. The otherthree themes respond to the second general question regarding topics tacitly expressed inthe Policy regarding whether and how diversity of relationships, family solidarity,intergenerational contract, isolation, autonomy, support and care, and formal means forsocial integration are addressed. In the second part of the findings section, the Policy analysisis compared and contrasted with older people’s definitions and experiences regardingpersonal relationships and social integration in late life. The gaps between older people’sdefinitions and experiences, and the topics included in the Policy help us to give answer tothe third primary question, making visible the unstated narratives of the Policy.4.4. Chilean public policies and institutional framework on agingSince the 1980’s, aging and old age became a priority in the Chilean governmentalagenda; first applying an assistance-based approach and gradually emphasizing entitlements134and autonomy of Chilean older population (Calvo, 2014; Huenchuan, 2004). During that time,as part of other structural reforms under the authoritarian regime of Augusto Pinochet(1973-1989), a pension system based on individual retirement accounts completely replacedthe old public pay-as-you-go system20.A more formal institutional framework emerged with the creation in 1995 of theNational Commission for Older Adults (CNAM) (SENAMA, 2012), which became the NationalService for Older People (SENAMA) in 2012. Since the creation of this organization, agingpolicies have broadened their areas of concern beyond economic security, to include issuesof health, quality of life, and autonomy of older adults (Calvo, 2013). In view of thechallenges of a rapidly aging Chilean population, in 2012 SENAMA presented a documentcalled Integral Policy for Positive Aging (2012-2025). This policy document constitutes aframe for action based on three general objectives: protect the functional health of olderpeople, improve their integration into the different areas of society, and increase their levelsof subjective well-being. These general objectives are accompanied by thirteen specificobjectives, each of which responds to one or more Positive Aging Goals that serve to assessthe impact of the Policy. Also, connected to the general and specific objectives more or lessspecific actions are proposed for the short (2012-2014), medium (2015-2019) and long term(2020-2025).20 Chilean pension system is a mandatory individual retirement account system where workers are free to choose anyPension Fund Manager (AFP) and may change from one AFP to another at any time. Employers are not required tocontribute to their employees' accounts and participation is voluntary for the self-employed (Kritzer, 2008). The system iscomprised of three pillars (tiers): solidarity, mandatory individual capitalization contributory and contributory voluntary(also individual). The solidarity pillar finances the poorest 60% of the population that have not contributed to the mandatorypillar and complements the pensions of those who have sporadic participation in the formal labor market and/or hadrelatively low wages (Bernstein, 2010).135According to the document, the Policy is integral “as it considers the contributionsthat different ministries and public services in collaboration with other social actors canmake” (SENAMA, 2012, p. 43). In that sense, the integral quality of the Policy points more tothe coordination between actors through the policy guidelines than the consideration ofactions directed to an array of issues related to population aging and heterogeneousexperiences of growing old in the Chilean context. The document also indicates that it is anaging policy as “it answers to the dynamic processes that occur along the course of life andhistory, and not only to the static condition of being an older person” (SENAMA, 2012, p. 43).Therefore, at least in its foundation, the policy assumes aging as a process rather than aspecific stage of life as its focus for action.The idea of positive aging is part of the foundational principles of the Policy as it doesnot limit itself to solving problems, “but to search for a desirable future, where the countryfaces with success the challenges of the new demographic structure, and where older peopleare healthy, integrated and able to report levels of subjective well-being as high as those ofthe younger people” (SENAMA, 2012, p. 43). Therefore, policy guidelines and practicesattempt to go beyond reactive actions to particular situations. It aims toward a morecomplex and long term perspective that consider both objective and subjective indicators toassess the adequacy and impact of its actions.The targeted population of the programs included in the Policy are generally meanstested and directed to the most vulnerable people: people with very low income, withoutinformal support networks (family), and/or, with physical or mental health issues. In most of136the cases these are the eligibility criteria to be part of a welfare program such as “Vínculos”(Links) or “Puente” (Bridge) and/or a public service (i.e. public health insurance FONASA). Thetargeted character of the programs and services included in the Policy responds to the actionof a liberal-residual state (Olmos & Silva, 2010).In the liberal-residual model, the state is limited to be an agent that stimulates themarket to create conditions for the productive activity of the private sector, so it can becompetitive nationally and internationally (Olmos & Silva, 2010). According to Esping-Andersen (2006, p. 162) “the [residual welfare] state assumes responsibility only when thefamily or the market fails; it seeks to limit its commitments to marginal and deserving socialgroups”. Examples of targeted programs are: Home Care; Protected Housing; Long Term CareFacilities; Social Tourism;, Courses of Courses of Food Safe; Industrial, Sanitary andHousehold Cleaning; Micro Business Administration; and, Day Centres, among others. As ageneral requirement, the person has to be part of the income quintile I, II, or III21. This meansthat the monthly income must not exceed $193,104 Chile peso (US $280 approximately).The different programs promoted by the Policy are of an individualistic character, asthe individual represents the unit of need and service around which policy is formulated anddeveloped (Clark, 1993, p. 27). In practice, this means that the individual bears the mainresponsibility for meeting his or her needs. Only when the individual fails does the21 Chilean social policies refer to income quintiles (‘quintiles de ingreso’) to target different benefits. Incomequintiles are obtained by ordering the households surveyed with the Social Protection tab (Ficha de ProteccionSocial), from lowest to highest scores, grouped into five sections of equal size. Thus, the I quintile groups the20% of households with the lowest score et (most vulnerable group) and the V quintile groups the 20% ofhouseholds with the highest score (least vulnerable).137government steps in to guarantee a minimal level of social assistance (Clark, 1993). In thecontext of many of the programs contained in the Policy, this failure is determined by theincome level, i.e. to have a monthly income that qualifies the person as part of the quintiles I,II or III.The following sections present the findings of the analysis of the Chilean “IntegralPolicy for Positive Aging 2012-2025”. The findings are divided in two sections. The firstsection corresponds to the analysis of the Policy. This analysis is presented using the fourthemes that have been previously described. The second section of findings integrates theresults of the interviews, comparing and contrasting older people’s experiences with thenarratives present in the Policy document. Also, some policy implications are outlined.4.5. Findings of the policy analysis4.5.1. ‘Tone’ of the policy: population aging as a challengeThe more official language used to introduce the issue of population aging and thepolicy goals identify the former as a challenge, the challenge of transforming old age into anexperience of positive aging. This contrasts with the language of the "problem" of populationaging that has been used in other influential documents (e.g. "Averting the Old Age Crisis:Policies to Protect the Old and Promote Growth" from the World Bank, 1994), as well as itcontrasts with the needs-based and assistance-giving22 approach of previous Chilean policies22 The concept of “asistencialismo” is usually used to describe previous policies and programs directed to olderpeople. This concept refers to policies that conceive to be different as a defect or lack of something that needsto be supplied. These types of policies also consider older people as mere recipients of help and passivesubjects who do not have the ability to control their lives (Huenchuan, 2012).138and programs directed to the elderly population. The notion of aging as a problememphasized the social and family costs of supporting an older population. According to thecurrent Policy, the actions developed recognize what older people have contributed tofamily, community and society, and seek to support and recognize older people's dignity. Inthe Executive Summary of the Policy we can read:[The Policy] seeks to create a desirable future where the country successfully facesthe challenges of the new demographic structure and where seniors are functionallyindependent and integrated into the various sectors of society, and report levels ofsubjective wellbeing as high as the younger population. (p. 10)Later, the document follows:Faced with this inexorable demographic change, SENAMA maintains that it isappropriate to anticipate the challenges that Chile will face in 2025 and turn theminto opportunities. Population aging will have a strong impact on our country and thelives of the Chileans. Life experience and wisdom of the elders will certainly be abenefit. At the same time, the country will have to make additional efforts to protectthe functional health of older people, to improve their levels of integration andparticipation in various areas of society, and to increase their subjective well-being orhappiness. (p. 17)However, underlying this positive approach there is a language of population aging asproblematic. The narrative of aging as a challenge is supported by data that contrast theactive older population with those who are dependent:In fact, by 2025 it is estimated that older people over 60 years old will be, for the firsttime, more than those under 15 years old. Today only 20% of those over 60 years oldremain occupationally active; and 22% are dependent, estimating that thispercentage will increase to 30% within the next ten years. ("Positive Aging: thechallenge of restoring prominence to older people", presentation of the Policywritten by Bruno Baranda, Ministry of Social Development, p. 8)Interestingly, in the aforementioned section the contrast between active anddependent people does not refer to occupationally active and occupationally dependentpeople. In that part of the document it is not completely clear but the contrast seems to be139between occupationally active people and physically dependent ones. This distinction is clearlater in the Executive Summary where we can read:The risks of inaction include prohibitive economic costs and negative social impactsassociated with the increase in the proportion of older people with problems toperform everyday activities such as walking or sitting. (p.10)Thus, the argument of the Policy Document focuses on the assessment of anincreasing number of physically dependent older people in contrast to the active workingolder population. In making this argument, the underlying idea of aging as a problememerges. The problem is the increasing population of dependent older people; people whodo not fit with the notion of active and productive people presented in the Policy. Althoughthe stress is on working older people, the idea of activity also includes people who, due to noor few functional limitations, are volunteering either in helping younger generations oractively engaged in interest groups that are part of active aging lifestyle.Therefore, two motivations characterize the Chilean policy on aging. The firstmotivation is the one officially stated, in which late life is depicted as a time to receive backwhat has been given to family, community and society in general. Nonetheless, the beginningof this period of receiving back is not clear. If we consider that this idea is presented close tosupporting data of older people in the labour force, retirement could be considered as themarker. The second motivation is more subtle, as it does not respond to a formally presentedobjective of the Policy. It represents the concern for an increasing proportion of dependentolder people. Here, late life is represented as a time of physical and economic dependency.Thus, two contrasting narrative strands coexist in the Policy, one in line with the successful-140models of aging, including the active aging perspective, and another informed by thedisengagement theory, strongly influenced by a biomedical discourse.In this manner, although avoiding the language of population aging as a problem or acrisis, the duality between dependent (infirm, not working and/or actively engaged in groupsof interest) and independent people is supported by data and topics of interest. When weexamine the programs and services proposed in the Policy, we notice that two types ofbeneficiaries are distinguished. A proportion of these programs and services are directed todependent, frail, and/or abused (or at risk of being abused) people and, on the other hand,to their opposite: older people who are vigorous, independent, and participating at differentlevels of society. Another subtle representation of late life that could reinforce the notion ofaging as a problem depicts this life stage as a period of crisis and risk:At this stage of life in which changes of identity occur, a crisis exists, there is a risk ofloneliness. Social connections, participation are necessary because health is physical,psychological and social, three components intertwined. Participation allowsdeveloping an active, productive, socially protagonist old age. (p.56)The individual experience of aging is characterized as a period of risk and crisis due tochanges in identity and connectedness. Whereas the previous notion of old age representeda social problem due to the social and family costs, the problem here occurs at a moreindividual level. It is the individual who is at risk of isolation and it is the individual who canprevent and counteract the negative changes by participating in society. However, the causesof these changes are not mentioned nor the specific changes that occur in older people'sidentity. Therefore, questions arise regarding who or what determines those changes, andwhen and at what level the transformations occur. For instance, do these changes depend on141biology, individual attitude, and/or age stratification? Depending on the answer the issuegoes beyond individual action and intersects with the structures of society that facilitate orprevent older people's social integration and participation.The ideas that support Chilean policy on aging are based on ‘keeping active’ theoriesthat, according to Marshall & Bengtson (2011), are based on values of autonomy andindividualism . This policy approach confers upon the individual the responsibility foraddressing the consequences of the demographic shift. According to Grenier (2012), success-based models of aging focus on preventive efforts that seek to change individuals’ behaviorsrather than social conditions or structured disadvantages that shape their experiences ofaging. For instance, in the Chilean Policy actions directed to social integration through work,most of the initiatives give the responsibility to find or create a job position to olderindividuals.Most of the programs (e.g. Job Training Program for Older Women; Courses of FoodSafe, Industrial, Sanitary and Household Cleaning; and Micro Business Administration) fosterentrepreneurship and participation in different types of trades that involve self-employmentor informal jobs. Thus, core problems such as age discrimination and lack of infrastructure forpeople with mobility issues are not addressed or are expected to be solved by older people’sown means. To be fair, the Job Creation Program tries to encourage potential employers tohire people from specific groups, among them unemployed women between 25 and 60 yearsold and unemployed men between 50 and 65 years old. However, incentives to employers142are only applicable to very specific groups, which additionally have to be classified asvulnerable people (quintiles I, II and III) who, among other conditions, have never worked.The distinctive narratives of healthy and successful aging on one hand, and aging as aproblem distinguished in this study are in line with the findings of Grenier’s (2012) analysis ofCanadian and English policy on aging. These competing narratives target two differentgroups: younger, healthy and active older people; and those who are affected by illness.These groups are further polarized by the instances of social integration fostered by Chileanpolicy actions (e.g. programs and services). The next section elaborates on this issue,identifying the differences between the programs targeted to dependent and independentolder people.4.5.2. Different instances of integration for dependent and independent older peopleIntegrating older people in society is part of the objectives of the Chilean IntegralPolicy for Positive Aging. Different areas are mentioned as related to social integration:prevention of abuse against older people, improvement in education and training, use ofinternet, participation in the labour force, housing, urban infrastructure and transportation,and cultural integration. The latter includes promoting a positive image of elderly people(e.g. in school textbooks) and opening opportunities for recreation and socialization. Allthese areas correspond to formal ways of social integration.The distinction between dependent and independent people is central for the actionsdirected to promote social integration. Different programs and areas of integration are143implemented depending on whether the beneficiaries are physically and/or economicallydependent people or are recognized as active.[...] the promotion of the rights of the elderly has been approached from differentfronts. One front has focused on participation, aimed at semi-able and able people,through the National Fund for Older Adults [...] created to finance projects designed,developed and implemented by organizations of older persons through competitions.The second component of the program, the Competitive Fund for initiativesdeveloped by intermediate implementing partners that work with seniors [has as itsbeneficiaries] dependent older people. [...] SENAMA, is responsible also for thetechnical operation of the Links Program, in alliance with MIDEPLAN [Ministry ofDevelopment and Planning], currently the Ministry of Social Development, in whichpsychosocial interventions are developed for poor and vulnerable older adults,providing subsidies and social benefits integrated to the community network of socialprotection, strengthening or developing skills through community monitors. (p. 37)From this description, integration through participation in a broad spectrum of socialactivities is only possible for relatively independent people who are already participating in agroup. Eligible projects include: volunteering, productive activities, infrastructure, self-care,training, and recreation23.On the other hand, dependent older people have a more restricted area ofparticipation, as recipients of care through specific activities and in defined places. Forinstance, the Long Term Care facilities (ELEAMs for their name in Spanish) indicate as part oftheir objectives "to promote social integration of residents, promoting their participation andinvolvement with the community in which the center is inserted" (p. 50). However, specificforms in which such participation is achieved are not elaborated. Another example is theHome Care Program, which seeks to maintain people living in their homes as much aspossible, as part of their community and exercising their social and family roles. In practice,the possibility of being part of the community depends on the infrastructure and the23 http://www.senama.cl/FondoNacional.html144instances for social interaction created by the older person and/or the primary caregivers. Asthe background information included in the Policy shows, this infrastructure is insufficient.Thus, the realm for participation is mainly older people's home and their private social world(i.e. close family members).An area where the Policy programs target social integration is that associated withparticipation in the workforce. As previously stated, these programs are mainly directed tohealthy older people, who are expected to be not only physically independent, but alsoeconomically independent:[...] currently, there are five economically active people for every elderly person, butin 2025 will have only three. Older people not only will have fewer people who cansupport them financially, but also will have to finance longer periods of retirement,due to the increase in life expectancy. ("Positive Aging: the challenge of restoringprominence to older people", presentation of the policy written by Bruno Baranda,Ministry of Social Development, p. 7)The motivation to encourage participation through work is mainly economic, as theincreased longevity may economically affect not only the individual, but also the stateexpenditure. The focus on healthy older people is not stated in the Policy documents.However this focus is given by the absence of recognition of the challenges faced bydependent older people to be integrated in the labour force such as: offices’ builtenvironment, adequate transportation, and remote connection with internet use.The specific programs that promote integration of older people in the labor forceoffer training to increase their possibilities to find a job or become entrepreneurs. Thedifferent programs are under the objective of productive aging and seek to "improve theeducational level and the job training of older people" (p. 43). By not recognizing the145difficulties faced by dependent older people and by not offering options, the Policydocument is unintentionally neglecting their participation as productive members of society,reinforcing an image of problem and burden.4.5.3. Autonomy and Self-Management in the Policy analysisA second relevant theme identified in the policy analysis is the centrality of theconcepts of self-management and autonomy in the definitions of the Policy as well as in theprograms outlined. In the Policy, self-management is promoted through prevention andhealth actions as it influences personal autonomy which in turn allows to be sociallyintegrated:Hospitalizations in long-term care facilities as well as hospitalization can have veryhigh costs for the country. In addition, functional health is a fundamental pillar of thequality of life and subjective well-being among older people. For both reasons, it is ofvital importance to promote self-management and prevent dependency, with the aimof preserving autonomy of older people for as long as possible, contributing to asatisfactory and socially integrating aging. (p.25)Here, autonomy is defined as the opposite of physical dependency and thus it isrelated to preventing health conditions that can increase this dependency. There is animplicit two sided distinction between autonomous/integrated and dependent/un-integratedpeople. Except for the “Protected Housing” program (Viviendas Tuteladas), it appears as ifthe responsibility to stay in the "positive" side of the distinction is at the individual level, bybeing able to exercise self-management. People who fail to stay healthy and autonomoushave no other choice but to be classified as dependent. When the expectations of activeaging are not met, the individual enters into the biomedical realm where issues of treatment146and acute care are the main concerns. In this sense, the definitions of the Policy, as well asthe actions fostered by it, give no place to ‘grey areas’.This binary distinction also influences the rationale for allocating state resources.Active aging, then, is a cost effective idea that fosters prevention. In the actions of theChilean Policy, the state develops instances that facilitate social participation and healthyhabits of independent older people. On the other hand, dependent people need acute care,which requires greater public expenditure. The latter is in line with Grenier’s (2012) findingsas she states that “success-based models of aging are used to justify funneling resources intoearlier periods of the life course in the name of prevention”.In the Policy document, autonomy is also defined as the "ability to make decisions,managing one's life, exercising an active and fulfilling aging" (p. 28). Although the notion ofautonomy presented in the Policy involves much more than lack of physical dependence, itpresupposed a favorable health condition. We can see this in the way autonomy is connectedto the notion of "good aging":Good aging is directly related to autonomy, the ability of the elderly person toextend, optimize and use the favorable health conditions to communicate his/herhistory, demonstrate his/her expertise, participate, have quality of life and maintainwell-being in old age. (p. 15)Together, the ideas of self-management and autonomy point to an individual who isable to prevent and avoid the negative effects of growing old and, in doing so, multipleinstances for social participation and integration that influence the construction of identityare opened. Most of the programs directed to autonomous active people address differentareas of life. On the other hand, people who are unable to fit in that model have more147limited choices, as most of the programs are health related, promoting rehabilitation orproviding acute care.An example of programs that are informed by the idea of self-management and self-care is the program "Choose a healthy life". In the Policy document, the accent is on theindividual element of choice and the state's role as a facilitator to motivate individuals tomake informed smart choices regarding eating and exercising. The program assumes a levelof activity and mobility on the part of the older person. In that sense, it is directed to alreadyactive and mostly independent older people who can access the activities that are beingcarried out. Members of the older person’s support network are not considered as a meanfor the older person to get information or to reach the place where the activities are carriedout.Regarding the idea of choice, but applied to older people’s care and support, thePolicy does not elaborate on the different ties that could provide them. Nor does the Policyfoster actions or provide services to promote older people’s choice in cases where care isneeded (e.g. having the option of using public services, or accessing to community care,instead of resorting to family care). In that sense, the available options are older people’spersonal resources –including their family ties– and then, if family fails or needs support, thestate provides options to specific eligible people through institutionalization (ELEAMs) andthe Home Care Program. This represents a residual approach of state’s intervention, as publicservices are considered as a last resource, only to be used once families are overwhelmedafter having used up their own resources (Marshall & Clarke, 2010, p. 27).1484.5.4. Policy’s assumptions of family responsibilities and involvement in older people’ssupportA third theme identified in the policy analysis was whether family members –particularly children– bear responsibility on older people’s support and care. In other words,the issue here is how the notion of intergenerational contract is expressed in the Policy, andhow the Policy document includes this notion in its definitions and reinforces familysolidarity. By using the language of duty toward older people, family obligation is not clearlystated in the Policy, but it is part of its assumptions. The penalization of family members forabandonment is one instance where the obligation is assumed.At the macro level there is the notion of rightful return to older people of what theycontributed early in their lives (e.g. love, affection, caring, spending quality time as a familytogether). Based on this, the state takes responsibility to ensure the safety and care of olderpeople who are in vulnerable situations due to physical and economic dependency. Thisvulnerability is connected to a complete or partial breach of the intergenerational contract atthe family level, due to family inability or unwillingness to care for its older members.Programs such as day centres, home care delivered by women and long term care centres(ELEAMs) seek to supplement or partially replace the availability of care from familymembers.Reichert and Phillips (2009) indicate that the agreement of mutual aid betweendifferent generations (i.e. intergenerational contract) can be based on social values andnorms or determined by law. In Chilean Policy on Aging, at the level of family, the149intergenerational contract is expresses in the Policy as a failure of family solidarity, in theform of family members neglecting older people’s care. Here, we find language of abuse andabandonment of older family members. The role of the state is to ensure, through lawenforcement, the rights of older people to receive family care. This stance toward familyobligation is defined by Reicherd and Phillips (2009) as a traditional view of theintergenerational contract, similar to what they found in Germany policy.An example of this in Chilean policy is the modification to the "Law of Intra-familyAbuse" (Article 5 Paragraph 1 of Law No. 20,066) to include as instances of family abuse notonly specific actions of violence, but also cases in which care has been neglected for omissionor abandonment. Although the boundaries of what is considered family are implied in thelaw –recognizing nuclear and extended family members, as well as in-laws–, it does notconsider other emergent type of ties involved in caregiving activities. For instance, it does notmention fictive kin and friends. This is reinforced on the background given to support thechange in the law:Omissions should be considered as constituting family violence, since older people,especially those with advanced dependency, require care that can only be providedby one or more persons (caregiver) and if it fails to perform (giving medications,changing diapers and/ or positions in case of those bedridden) will generate damageto the dependent person, establishing domestic violence if the care is provided by thefamily. (p.69)However, other programs indicate friends and neighbours as primary caregivers. Forinstance, having a primary caregiver, defined as "a relative, friend, or neighbour whoprovides care in a regular and permanent basis"24, is among the requirement to apply to the24 http://www.senama.cl/CuidadosDomiciliarios.html150Home Care Program. Therefore, there is a possibility of abuse and neglect from other type ofties not included in the law.Beside the concern for older people's wellbeing and rights, there is a central issueinfluencing the importance of family solidarity in the form of care. From the state'sperspective one major concern regarding older people's care is how to protect theirfunctional health while also avoiding increasing public expenditure. Hospitalizations andELEAMs are too costly so a tiered model of care is proposed:The care service must take a form of pyramid-based home. Functionally independentolder people can live independently in their own home or in their family home, but asthe level of functional dependence increases it is necessary to go up steps andincorporate a more active role of the family, tele-care mechanisms, assistance, andhome care (ranging from simple assistance and delivery of prepared food or help toclean and tidy up, to more complex care such as assistance with activities,rehabilitation exercises, and taking medicine), rehabilitation in day centers, andfinally, institutionalization in long-term care facilities. (p.26)In this model, the role of the family is central. In relationship to care, Sheets, Bradley,and Hendricks (2005) indicate that the 'family first' model proposed by Shanas (1979b) hasbecome embedded –and we can add reinterpreted and naturalized– within neoliberal socialpolicies, so that family support is assumed as the first line of defense in the provision of carefor older people.This notion can be found in the background for the construction of Long-Term Care Facilities:This measure [the construction of Long-Term Care Facilities] responds to the highdemand for ELEAM, which cannot be met completely either by private organizationsor by non-for-profit foundations. Ideally, the "potential" beneficiaries of an ELAM arecared for at home, but in extreme cases, given the above requirements [people olderthan 60 years; have a Social Protection file; belong to quintile I, II or III, according tothe Housing Lack Score; have housing and/or support and care needs, accredited by aSocial Report; have no physical or mental illness that requires acute care in a hospital151setting, as demonstrated by a medical certificate; have some degree of dependence,according to the functional assessment performed by SENAMA; have a representativeand informedly consent their admission, the State shall provide them with these carefacilities. (p. 50)Despite the taken for granted role of family, Chilean social policy reflects whatReichert and Phillips (2009) have called an ‘individualistic approach’. As a result, at theimplementation level, family role is excluded from explicit discussions. Nonetheless, thecentral role of family is largely assumed and the targeted character of state’s services makesfamily more necessary. In that sense, the individualistic approach of policy intersects with thestate’s expectation of a ‘collectivist’ pattern of caregiving. According to Biggs (2005),‘collectivist’ families offer greater instrumental support than ‘individualistic’ families, in thatthey mix instrumental family support with use of welfare services. Thus, the state’s practices–allocation of resources in the form of subsidies and programs- is contradictory with its ownexpectation.The main indicator of the assumption of family obligation to support its elderlymembers is that in many programs the lack of family support is part of the criteria to applyfor a service. In that sense, informal networks are assumed as part of the resources thatolder people have at hand. Thus, Chilean policy on aging not only takes an individualisticapproach, but also takes what Clark (1993, p. 27) calls a residual approach to familism.According to this approach, public services are only justified when families have failed tomeet members’ needs with its (the family’s) own resources. Thus, state’s support is onlyapplicable once individuals have used up all their resources at hand, being these material orsocial resources.152Assumptions of family responsibility toward older members are at the basis oftargeted programs. Family as a “reserve of support” (Kemp & Denton, 2003, p. 756) thatmakes up for the services and protections not provided by the state is taken for granted.Included in this view is the assumption that in the absence of economic resources on the partof the older person, the family would have the ability and willingness to provide care.4.6. Comparing and contrasting older people’s experiences with policy narrativesIn this part of the findings section the experiences and definitions of the studyparticipants are compared and contrasted with the four themes outlined in the policyanalysis. We have to keep in mind that the study participants were healthy people who didnot need special physical care, and thus this section represents their particular experiencesand definitions of social integration through informal ties, family solidarity, and notions ofautonomy.This second part of the findings sections is directly connected with three of thethemes addressed in the previous sections (i.e. different instances of social integration,autonomy and self-management, and assumptions of family responsibilities and involvementin older people’s support). Also, the experiences of the study participants presented hereallow identifying different issues that remain unexpressed (third primary question), offeringinsights on the tensions between the actual experiences and expectation of older people andthe old age that is assumed in and constructed by the Policy.1534.6.1. Different instances of social integration through (in)formal relationshipsThe different dimensions for social integration described in the Policy correspond toformal means of integration. Informal forms of social integration, i.e. family and non-kinnetworks, are not explicitly recognized, despite their centrality in older people’s experiences.According to Scharf (1998), integration through informal relationships includes consideringthe nature and quality of contacts with family, friends, and neighbours. The centrality ofinformal relationships has been highlighted by different scholars that have noticed theirimportance for help exchange, engagement with new life styles, and identity construction(Pahl & Spencer, 2004; Phillipson, 2013).Study participants’ experiences indicate that exchanges of emotional and practicalsupport occur between them and their circle of relationships in different situations: lend andborrow money, in case of an emergency, transportation, daily conversation, recreation, forlegal paperwork, and access to information. Family and friends, because of their professionalskills and knowledge or acting as a bridge, are particularly important when legal proceduresand prompt access to health services are required. The importance of informal relationshipsin late life are better elaborated in chapter 2 and 3, which show the role of friends and familymembers in older people’s social networks and personal communities.A point where the Policy and the experiences of the study participants present cleargaps is in the notion of community involvement. For instance, the Policy mentions that olderpeople experiencing dependency should maintain their involvement with the community. Ifcommunity is understood as a geographical place such the neighbourhood, this integration154could have little meaning for the person. According to study participants’ experiences, theyhave little or no relationships with their neighbours. The exchange of help with neighbours isvery limited and only 4 participants included neighbours as part of their emotionally closenetwork of relationships. The discrete role played by neighbours in older people socialintegration is in line with Scharf’s (1998, p. 159) findings for the German context, whonoticed that the frequency of visit by and to neighbors were low among older people, andthe interactions were generally to exchange pleasantries.Social integration through informal relationships is different for older men andwomen, an issue not particularly recognized in the Policy. For instance, the size and variety ofthe ties that form the circle of emotionally close people present gender differences. Men’snetwork of emotionally close people had an average of ten people, whereas women’s hadseventeen. While 77% of men had networks of ten or less members, 71% of women hadnetworks of more than ten members. Also, gender interacts with cohort as male participantsbetween 67 and 74 years old have more family-based networks of emotionally closerelationships. Whether these differences are due to lack of opportunities to connect withpeople outside the family circle after retirement and/or to men’s individual abilities todevelop new emotionally close ties requires further analysis. These gender and cohortdifferences relative to social integration at the level of community and friendships tiesrepresents clear examples of heterogeneous experiences of late life.Phillipson (2013, p. 124) notes that inclusive ties, as the ones found among familymembers, fit well with institutions such as the welfare state and stable intergenerational155contract. However, the current social context is characterized by fragmented andpersonalized pathways, where public provision of help has shifted to more flexiblearrangements. According to Phillipson (2013), in such a context non-blood andmultigenerational relationships would be better suited to deal with the different risks facedin late life. The gender differences in Chilean interviewees show that men’s networks ofemotionally close people, as well as their main sources of practical and emotional help, arebetter suited for a more traditional type of society (as previously described by Phillipson).Women, on the other hand, have a more varied spread of ties, which would give them anadvantage at least at the level of informal forms of social inclusion. However, these genderdifferences and the challenges they represent to promote social inclusion and prevent socialisolation in late life are not acknowledged in the Policy, where programs directed to olderpeople assume old age as a more or less heterogeneous experience. A good example of thisassumption is the important role given in the Policy to involvement in association and clubsas means for social integration. However, male study participants have less involvement thanwomen do, with 50% of men and 81% of women participating in organized activities. Thus,these types of programs would be more beneficial for women than men unless they arespecifically developed to increase the involvement of male participants, attending to theirparticular needs.Informal relationships also play a role in what seems as a formal means of socialintegration. That is the case of social integration through work. Participation in societythrough work is defined in the Policy as a means for economic security. However, from the156experiences of the interviewees, the workplace also represents the context where they canbe socially integrated through the relationships that are created.Particularly for men, work is an instance not only to earn money, but also constitutesa way for self-fulfillment through exchange of professional experience and knowledge. In theexperiences of male interviewees, the workplace appears as one of the main instances todevelop relationships with people outside their nuclear family:They are friends that I've had let’s say, through work, for a working relationship firstand closeness of thought, or situations we have lived […] Claudio has around 66 yearsold... and this one lives in Providencia ... and he is at the same level as friend, uh, andin fact it’s also a labor relationship as we are planning to do things, etc. (Fernando, 66years old)We work a lot together [with his friend Enrique], so the bond was strengthened… wehave worked together for 15 years, done seminars together, so we know what wedo”. (Matias, 60 years old)Predominantly for men, the workplace is a central instance of social integrationthrough friendship and professional ties based on personal interest. The workplace alsoprovides a context where different social identities are at play25. Therefore, the idea of workas an income source to be or to remain economically independent is a narrow approach tothe meaning and experience of work. This leaves aside opportunities to develop strategies toincrease the social connectedness of men through groups and clubs that serve as instances todiscuss professional interests.4.6.2. Older people’s experiences and expectations of family solidarityThe notion of family solidarity is tightly connected to the previous section. Althoughnot explicitly stated, the Policy assumes the central role of the family to facilitate social25 This topic is better developed in the papers on social capital and personal communities.157integration. However, the lack of analysis of informal networks of older people’s in the Policybackground hides the complex role of these networks and the amount of support exchangedfrom the older person to other family member. The experiences of the participants show thecentral role of older people as providers of help to children and grandchildren. As areference, from the total of 26 women and 14 men interviewed, 13 for each group definethemselves as head of the household (main breadwinners). From those, 8 women and 7 menhave children living with them, meaning that the children depended on the material supportof their parents.Another assumption in the Policy document is that the older person would be willingto receive support and care. The experiences of the study participants question thisassumption. In fact, retaining or maintaining independence and being able to solve thechallenges of growing old by themselves is essential for them. Few participants gave theargument that children must help their parents because they are family. The majoritystressed that help should not respond to obligation or duty, but to a decision made by thechildren based on the quality of the relationship and the example given by the parents:I think that the child’s duty is to dedicate himself to his children, and to his family,that’s how I deeply feel it […] but in turn I feel that the children have a specialattachment to their parents, and they show their concern toward the weakness ofthe father or mother. I realize that for the phone calls, when one is in bad health, forthat concern of “take care”, when we are going to go out, no? But there is no duty;there is no obligation from the children to his father […]. (Rodrigo, 67 years old)[…] duty as such does not exist. Because unfortunately, when I have a child thedecision is mine, it’s not of my child. Therefore my duties are to guide my children tohave a good life. Now, if they by… by their own willingness, stuff like that, considerthat I did a good job, they are going to help me. (Julia, 60 years old)158Similar to Kemp and Denton’s (2003) findings relative to the allocation ofresponsibility for later life, there is a strong mismatch between the Policy assumption offamily as a “proper” reserve of support and people’s expectations toward their families. Mostspecifically, the experiences and expectations of the interviewees are in line with the idea of“intimacy at a distance” (Rosenmayr & Köckeis, 1963). Reinchert and Phillips (2009) cite thisconcept to discuss the assumption in public policies of intergenerational co-residence as asign of emotional closeness between generations. According to these authors, co-residencereflected more economic and occupational needs than emotional closeness. Reichert andPhillips add that co-residence may, in fact, prompt tensions among family members. The caseof Raul (67 years old), who has a very complex relationship with his wife and son, is anexample of the latter. This quote stresses how Raul’s son privileges giving economic supportto her mother:Do you want me to tell you the truth? I practically live, we live with my wife atexpenses of my son […] he is not despicable, but not that good also… he sustains us,let’s say, the house. He has a credit card for my wife and she goes to thesupermarket, pays the electricity and the water, [he] pays everything. So the onlything I have is to go buy the bread, the veggies some times, and nothing else. If myson were to die, we would go down to the floor I think […] he is a good son overall.[…] to the mom, he helps her more than to me, because I go, for instance, to thepublic clinic, plain and simple; the mom nope. The mom goes to the private clinic.In the same way, Biggs and Powell (2003) commented on the mythical status of thehappy family, as reflected in England’s policy on aging. According to Biggs and Powell, datashowed that good relationships among family members were based on relatively loose andundemanding exchange between generations. Similarly to Reichert and Phillips (2009), Biggsand Powell note that deep commitment was in fact typically the result of extreme situations.We can observe something similar in the case of the study participants. The high value given159to be independent is accompanied by a strong sense of boundaries between their own livesand their children’s lives:We share when we sat down to lunch. But I do not intervene in the lives of mychildren, or anybody's life, unless they ask for advice. But in turn, when they "mom,do not do this", "do not do that", I say well if I do not mess with you, you do not messwith me, there I mark the limit. (Julia, 60 years old)My children are all very independent. Let’s say that I was also super independent, andI raised them and taught them to be independent […] because she [daughter] isprofessional, has three children, is separated, then I do not get involved in her lifeand she does not get involved into mine. Because I cannot get into her life about howto raise their children, because all I say, once they grow up I'll be under a grave. […] Ihelp her as I can, especially if she asks me for something, obviously, but I will notinvade her privacy and she will not invade my privacy in my department. (Marisol, 73years old)Thus, at the level of current expectations and practices, older people are also inclinedtoward a model of individualistic family care, though the role of subsidies and welfareprograms are unclear as many of the study participants are not eligible for them. However,the notion of additional support is present, and such support seems to be, more by needthan by choice, the family:I believe that here, in this country, we unfortunately have to rely on it [the family], wedon’t have the conditions, because I believe that we can count with the fingers thepeople who can be totally self-sufficient, because they have economic resources thatallow them having someone to assist them, all the time. But in reality I believe thatone needs to resort to the closer family... I believe so. I mean, I believe thattomorrow I would have to rely on my children to say “hey, could you take me to thedoctor?”. (Alma, 60 years old)I believe that the older people are not the older people from the Family AllowanceCompensation Funds [“Cajas”], the one that is travelling, the one they show you, Ibelieve so. I go a little further, because I believe that many adults are abandoned, Imean, it bothers me deeply that deception of the protected elderly. (Patricio, 60years old).160What is particularly interesting about Patricio and Alma’s quotes is how theychallenge the ideal of autonomy, self-management and individual responsibility presentednot only in the policy, but also in the institutional discourse, as exemplified in the openingquote of SENAMA’s former director.4.6.3. The importance of autonomy and self-management in older people’s experiencesThe strong emphasis on autonomy and individual responsibility present in the Policyis in line with study participants’ experiences and expectations. In the language ofparticipants, the ideas of autonomy and self-management are contained in their descriptionsof being independent. Julia (60 years old) talks about late life as a stage of independence andfreedom:For me, independence is being able to rely on things you do economically by yourself,isn’t it? To move, to do your own things, to be able to produce your own income. Andwhen you reach a certain age you are more independent ... You don’t worry about ....,One worry about the children in a way ... You don’t need to worry about teaching… ifthey have bad habits you cannot remove them. I mean, the freedom, that of feeling alittle free from responsibilities. (Julia, 60)Similarly, Catalina (68 years old) highlights the importance of independence, but withattention to its role during the life course. In that sense, she reflects on the value ofindependence for particular groups and cohorts:[…] women of my generation were educated, formed and guided to be dependent onothers, on a patron, a husband. If you got out of the home was to get married. Thenall these women rarely were independent. You could not think of having, at that time,a child being single, neither you thought about to be divorced, the faith, the sin, andthe shame of the family. Then the independence to think and act is super importantfor both the man and the woman. It is a concept of life; independence has to do withthe freedom to choose, to choose certain situations, to say 'no' in somecircumstances, for me that is independence, freedom.161The shared emphasis on autonomy can have positive effects, as they help toempower older people. However, in both the Policy and the study participant’s expectations,the transition from life styles of the third age to those of the fourth age is blurry. More so ifwe consider that state’s initiatives directed to dependent older people are carried out in veryspecific and separated spaces. In this context, older people may fail to envision and plan for afuture that cannot completely meet the expectations of preventing disease, keeping active,and being independent. Matias is currently working and active. His current life style andfuture expectations match those promoted by the successful-model of aging present in thePolicy:[…] I mean I do not see that the picture can change much, uh, in fact I neverconsidered to be dependent, especially now, because the age will not allow me, but...We will continue working and doing the same things we did before... I'm talkingabout working less, to earn more and enjoy the time, which you see that it becomesvery difficult, a bit utopian. (Matias 60).Although Matias tries to articulate the idea of how his always independentpersonality and life style will influence his future, he can only visualize the lifestyle that ispromoted for the third age. In other words, he is acknowledging the utopian character ofkeeping such lifestyle, but without proposing an alternative plan in case he cannot achievethat desired healthy, independent, and active older age.The perspective of the participants matches with the models of successful agingpresented in the Policy. The ideas of keeping active, productive and healthy are considered inthe Policy and in the study participants’ narratives as notions of personal success and failure.However, for the study participants, to be independent is signified as being independentfrom their children. In that sense, participants’ expectations rest almost exclusively on their162own resources and on the help and care they can exchange with their spouses or partners.This represents an important gap within the assumption in the Policy document about familyresponsibility toward older members that is based on a normative notion of theintergenerational contract.4.7. Conclusion Chapter 4This chapter sought to compare and contrast the ideas and practices relative topersonal relationships and social integration of older Chilean people with the definitions andforms of social integration supported by Chilean policy on aging. Particular attention waspaid to identify whether and how the formal and informal ways of social integration relevantin older people’s experiences were included and supported by the Policy definitions andprograms.The quote at the beginning of this chapter anticipated some of the main findings ofthe social policy analysis. One of the main characteristics of Chilean policy on aging is itsalignment with successful models of aging while locating the older person at the centre inboth the policy narrative and actions. Although this should not be necessarily negative, it hasresulted in an invisibility of experiences and conditions of interdependence, such as thecomplex relationships with family members and friends, as well as the diverse motivationsthat underlie different forms of participation in society. The distinction between dependentand independent people gains salience in the Policy narrative when interdependence isneglected.163In this manner, the findings indicate that two narratives coexist in the Chilean Policyon Aging. On the one hand, the document constructs aging as a challenge that needs to betackled by society as a whole. A big part of the responsibility is assigned to the older person.This is done at the narrative level with the notion of active, productive and healthy aging.From this perspective, aging is seen as an issue of lifestyles (Biggs & Powell, 2003), as anoutcomes of choices taken at the individual level that should result in a good aging processand therefore in a good old age. In that sense, the failure or success in achieving the idealizedversion of the third age is, as noticed by Grenier (2012), a matter of individual responsibilityand choice. At the level of state’s programs, the targeted character of the services indicatedin the Policy also points to individuals’ own resources.On the other hand, the Policy constructs old age as a problem. The ‘problem’ isconstituted by people who do not meet the terms of a healthy, active, independent andproductive aging. Therefore, the main characters in this second narrative strand are‘dependent’ older people, who are referred using a language of vulnerability. What are theconsequences then of these coexistent narratives for older people’s instances of formal andinformal social integration?By analyzing the Policy programs we see that very different physical and social spacesare given to people who fulfil the script of successful-models of aging and those who do not.Independent, active, healthy and productive older people are expected and encouraged toparticipate in different spheres of society, as workers, volunteers, consumers, and membersof community groups. On the other hand, people who fit in the model of vulnerable anddependent aging are integrated by the rhetoric and means of the biomedical realm. People164from this group are not constructed as active citizens and ‘subjects’ in charge of their agingprocess, but as ‘objects’ of care, to rephrase Huenchuan’s (2009) distinction between olderpeople as subjects or objects of rights. Interestingly, despite that interdependence has notbeen central in the Policy narrative, here the responsibility for elderly people’s care istransferred to families, and if the family fails, to the state. The latter, nonetheless, would betrue only for those who are eligible. In that sense, a central underlying assumption in thePolicy is the availability of a safety net given by the family material and human resources thatwould support older people once their own personal resources to maintain an independentliving fail.The latter results are in line with Phillipson (2015) and Settersten and Truten’s (2009)argument of later life as being reconstructed in late modernity as a period of potential choiceand risk. In the context of a post-traditional social order, pensions, employment, retirementand intergenerational relations have experienced changes (Phillipson, 2013). Older peoplewould have more choices to live their life more flexibly in ways that are congruent with theirpersonal interests and preferred life styles (Settersten & Trauten, 2009, p. 457). However, atthe same time, people assume these choices with unknown consequences. Any outcomes,including the negatives ones, have to be negotiated by the older person and his or her familyin a context of weakening institutional support from the state:Old people are largely on their own with only the safety nets they can create withthe resources they have, whether through personal and family resources or throughsocial skills and psychological capacities […]. (Settersten & Trauten, 2009, p. 458)The focus of the Policy on individuals’ responsibility to assume the opportunities aswell as the difficulties of old age is an example of how the risks once carried by social165institutions have been placed upon the individual. The creation of contribution pensionschemes (as the one used in Chile), unequal access to secure employment, and stagnation inwages and salary create different forms of inequality in late life, which have to be shoulderedby the older person and his/ her family.From the standpoint of the Policy on aging, the safety net is mainly formed by familymembers. This is clear when analysing the narratives surrounding family responsibility,where two coexistent narratives can be found. For active and independent older people, notonly family, but close and thick relationships in general are at the periphery. In fact, the lackof analysis of changes in living arrangements, family composition, family life and social worldof older people in general, is a clear indicator of how the older individual itself is located atthe centre as an active citizen. He or she can and should contribute to society, and morespecifically to the economy, so that fewer demands on pensions and support are made. Incontrast, for ‘dependent’ older people family is at the centre. Family members have the legalobligation to care for their old members. In that sense, a notion of family solidaritysupported by a traditional form of the intergenerational contract exists. Caring obligationsare enforced by the state based on a traditional normative notion of family solidarity, whichin turn aligns with the state’s aim of reducing social expenditure.According to Biggs and Powell (2003, p. 104) “the success of a family policy can bejudge from the degree to which people live within the stories or narratives of familiescreated by it”. When observing the experiences of the study participants, we can see somematches and gaps with the definitions and practices of the Chilean Policy on aging. Olderpeople’s experiences match with the expectation of autonomy and independence stated in166the Policy, and they visualize themselves living old age under the ideal of a healthy third ageof activity and productivity. However, the interviewees do not place family at the frontline incase of need and care. They do not expect, nor do they want, that family ought to take careof them in the future. The latter, nonetheless, does not mean that family solidarity andwillingness to care has weakened or is disregarded. Instead, as noticed in other studies(Biggs, 2005), familial support and care have moved away from relationships based onnormative obligation toward mutual negotiation.A limitation of this study is the specific cohort of its participants. However, despitethem being part of a young cohort, the study participants’ experiences with family members,friends, and organizations can give some ideas about their personal network and the socialcapital embedded within these networks. Understanding the variety and complexity of olderpeople’s relationships and forms of social integration can open new possibilities to preventsocial isolation and promote different forms of social support that are valued by the olderperson for their emotional and practical benefits. As Grenier (2012) noted, policy interest onidentity and lifestyle choices open an opportunity to recognize how older individualnegotiate and shape their life course, challenging fixed approaches to late life. Theexperiences and expectations of the study participants provide us with some glimpses of howlater stages of their life course may look like and whether the current policy actions andsocio-cultural structures are fostering the type of late life they define in their narratives andexpect to construct in their practices.1675. ConclusionThis study was designed to explore the personal relationships of older people living inthe Metropolitan region of Chile. The literature and data available on this topic for thisspecific context are limited. Little has been studied about the important ties with which olderpeople exchange help, and whether the types of help vary depending on the kind of tie thehelp is exchanged. In the same manner, although the effects of socio-cultural changes in theChilean context have been part of academic and public discussion since the return ofdemocracy, little attention has been given to the effect of those changes in the currentgeneration of older people. Moreover, no previous studies have linked the rapid process ofpopulation aging, its effect on the availability and management of support networks in latelife, and the socio-cultural changes experienced by the Chilean population. Recognizing thegap in knowledge, the study sought to answer the question: To what extent and under which circumstances do older people exchange helpand complement family resources with other types of personal ties (e.g. friend,neighbors and/or state)?The search for answers to this research question has focused on the composition,function and meaning of older people’s personal relationships. These issues have beenexplored in the result chapters 2, 3 and, from different angles. Particularly, chapters 2 and 3represent two different ways of analyzing the data to stress different aspects of personalrelationships in later life. Using a personal communities approach, chapter 2 analyzed thediversity of ties that form older people’s networks, with especial attention to their emotional168function and subjective meaning. The personal communities approach allowed identifying,from the participants’ perspectives, the emotionally important relationships that form theirnetwork of close relationships. Chapter 3 examined older people’s social capital with a focuson whether and under which circumstances family and friendship ties had bridging andbonding properties to provide older people with practical and emotional support and toconnect them with different life styles. This chapter challenged views that equated bondingties with strong ties —commonly represented by family and friends—, and bridging ties withweak ones. Finally, chapter 4 used a narrative approach applied to policy analysis to explorewhat forms of social integration are fostered by Chilean policy on aging to then contrastthem with older people’s actual experiences of social integration in later life.The findings presented in these chapters recognize the influence of the socio-culturalcontext in which older people have developed their relationships. As stated by Osorio (2006),population aging in post-industrial societies, such as in Chile, is a complex phenomenon ofanalysis as key social transformations intersect. In this thesis, the way in which older peoplemanage and signify the ties that form their personal networks can be understood asrepresenting a tension between the values associated with an individualized society and theneed to depend on personal ties due to neoliberal social policies and limited servicesdirected to older people.The current influence of socio-historical processes shaping the particular experiencesof the study participants has been addressed by reviewing the literature (chapter 1)describing the structural and cultural changes experienced by Chilean society. In a more169specific manner, the current social policy on aging —the object under analysis in chapter 4—represents the social norms and values shaped by these changes.To answer the research question, I interviewed 40 older men and women living in thecity of Santiago, Chile. I also carried out an interpretive narrative analysis on the ChileanIntegral Policy for Positive Aging 2012-2025. It is not the intention of this conclusion sectionto offer a summary of the findings, nor to repeat the conclusions already presented in theprevious chapters. My intention in this final section is to go back to the overarching issuesidentified in chapter 1. Here, I focus on how the experiences of the study participantsregarding their personal relationships can be framed by the particular Chilean context. I alsoconnect the findings with the two variants of the theory of individualization (described inchapter 1) to contextualize the particular experiences of Chilean older people with globalsocial processes of cultural change.5.1. The role of the socio-cultural context in shaping the composition, meaning andfunction of personal networksThe different functions of family and friendship ties are shaped by socio-historicalprocesses and structural forces. In the Chilean context, socio-cultural changes triggered bythe dictatorship period greatly and negatively impacted trust in institutions and amongpeople. This negatively influenced how individuals think about other people and institutions,their willingness to develop new ties, as well as older people’s decisions about the types ofhelp that could be exchanged.170In addition to the latter issues, the structural changes introduced in the 1980’s andcontinued after the return of democracy have influenced social policies, which are nowcharacterized by services and programs targeted to the most vulnerable population. As aresult, a diverse network of personal ties plays a very practical role in later life by providingfunctional help and specific forms of connection to services and information.The low levels of institutional trust that characterize Chilean society are central to ourfindings and conclusions. It is in that context that we can better understand the doublefunction of family and friendship ties that work as both bonding and bridging social capital.The recognition of this double function has two implications to the social capital theory. First,the findings support and at the same time expand —by paying attention to the influence ofthe context— what Leonard and Onyx (2003) found in their research: people are more willingto take risks to access other networks for information and resources when they used trustedintermediaries. In other words, in a context of low institutional trust, bridging social capital isnot necessarily formed by weak ties of dissimilar people. In fact, these ties need to be peoplewho are emotionally close. Interestingly, they also need to be similar and at the same timedissimilar to the older person, depending on the situation where different resources need tobe mobilized.Chapters 3 and 4 in particular shed light on the bonding and bridging properties offamily members and friends, where bridging ties assist the access of older people to services,organizations, programs, and relevant information. This finding supports the critical variantof the theory of individualization for developing countries as stated by Robles (1999, 2000),showing that individuals must seek help from family, peers and other informal ties to be171included into society and thereby compensate for weak means of formal social inclusionprovided by public and private institutions. Scharf (1998) distinguishes between integrationthrough formal and informal relationships. Our study findings indicate that, in the Chileancontext, social integration through informal relationships (represented by family andfriendship networks) is central to both supplement formal relationships (e.g. participation inthe labour market) and to connect older persons with those formal relationships.The second implication is closely connected to the first one as it has to do with thedistinction weak/strong and bonding/bridging ties. Family and friends are usually defined asstrong ties in the social capital literature and tend to be presented as equivalent to bondingsocial capital. The findings clearly show that study participants define friends and somefamily ties as emotionally close relationships, which is an important aspect of strong ties.However, depending on the situation, friends and family ties act as bridging social capital.When we attend to criteria of (dis)similarity to differentiate between bonding and bridgingproperties of personal ties, we see for example that a child can be similar in terms ofsocioeconomic status, but is nonetheless from a different generation and plays roles insociety different from those of his or her parents. Thus, a family member or a friend can beboth similar and dissimilar in relationship to the older person depending on the contextwhere the tie is being used. The dissimilar characteristics (e.g. age, profession, skills) can beuseful to connect with resources that could not be mobilized by the older personher/himself.The qualitative approach with focus on the meaning of personal ties used in thisthesis allowed recognizing the weakness of applying binary categories to describe and172analyze older people’s social capital. Specifically, the distinction between bonding andbridging ties, although useful, may hide how a tie can work as bonding or bridging dependingon the situation. In addition, the definition of bridging ties is complex in the context of laterlife, as not all bridging ties work in the same manner. This can be linked back to the researchquestion to better understand the circumstances under which older people complementfamily resources with other types of personal ties. The meaning of personal relationships inlater life, and the needs and expectations of material and emotional support are shaped bycultural norms of an individualized society. These cultural norms give a positive value tobeing independent and self-sufficient, and support the notion that individuals should createtheir own modes of living (Morgan, 1996). It is in this context where the differentiated role offriends and family members described by the study participants can be better understood.From a more practical standpoint, both family members and friends can help theolder person to connect to resources that would be otherwise difficult to access. However,friends represent a particular type of bridging tie given by their similarity, in some aspects, tothe older person. From the study participants’ experiences, we observe that friendsrepresent the preferred ties to exchange emotional support, as the life experiences and/orsimilar interests add a variety that family members do not have. At the same time, they areemotionally close yet outsiders to the family, which make them the ideal ties to ask foradvice in case of family issues. De Souza Briggs (2003, p. 2) indicates that bridging ties areimportant to develop broader identities and communities of interest as these relationshipshelp to bridge across roles, symbolic interests, and worldviews. We could specify the latter bysaying that friends play that particular role for the study participants, as friends connect the173older person with advice, information and activities that cannot be facilitated by familymembers.In addition, the role given by the study participants to friends as preferred sources ofsupport for certain issues, can be also understood as way of being self-sufficient and keepingan independent live style for themselves and their children. In this regard, I used Bonehamand Sixsmith’s (2006) concept of negotiated reciprocity to identify the instances where theinterviewees expressed difficulty in asking for or accepting support from their children. In linewith the values of an individualized society, older people do not exchange help based onnormative moral obligations (e.g. familism), but they actively create, negotiate and managetheir personal relationships. The inclusion of a diversity of ties in their network ofrelationships, as indicated in chapters 2 and 3, allows older people to actively manage theresources embedded within that network. With this, older people try to avoid becoming aburden as they exchange diverse kinds of support with different types of ties.Finally, we can connect the value given to independence and the particular meaninggiven to family and friends by the study participants with the low levels of social trust presentin Chilean society. These low levels of social trust shape to some extent who and why olderpeople develop certain types of relationships to exchange different kinds of support. TheHuman Development Report of 1998 (UNDP, 1998), cited in chapter 1, indicted that 64% ofthe people surveyed believed that it is difficult that others do something for other peoplewithout expecting something in return. Eighteen years later we find similar experiencesamong the study participants, who negatively perceive asking for favours due to thecommitment this might create. I used the concept of altruistic help (chapter 3) to describe174situations where rules of generalized reciprocity were at play. This type of help exchange,characteristic of close ties acting as bonding social capital, seems suitable to get supportwithout the need of asking for it and without acquiring further commitment. In a similar vein,the low levels of social trust present in Chilean society can be described as low trust inanonymous others and it is in this context that we can interpret clustered personalcommunities (chapter 2) as indicating the sharp distinction between those people who canbe trusted upon and those who cannot.5.2. Population aging and socio-cultural change in ChileAs stated in the research question, this thesis connects two form of social change: thesocio-cultural changes previously described, and a rapid process of population aging. Wecould expect that the rapid process of population aging would seriously impact theavailability of support and care in late life, as availability of family ties will decrease.However, in light of the findings, we can conclude that older people have the ability to adaptto different scenarios. Such adaptation can be seen in how they complement the differentresources embedded in their networks of relationships by actively constructing andnegotiating their personal ties. New sources for exchanging help are available, accompaniedby more complex functions of personal relationships. Thus, the findings contribute withresearch based evidence to the literature suggesting that late life today is a period of choiceand risk (Phillipson, 2015; Settersten & Trauten, 2009), where choice representsopportunities to develop relationships and support network that are in line with olderpeople’s actual and expected life styles.175The findings can be framed by the theory of individualization and the literature onsocio-cultural change in Chile. From the theory of individualization, the question aboutcircumstances also relates to more complex functions of personal ties in late modernity. Wecan see, for instance, how elements of voluntariness are present in the construction ofpersonal networks, and how emotionally central those relationships are to give moresymbolic forms of social integration. I have shown in chapters 2 and 3 that these ties play arole in older people’s lives that is close to an individualized society as described for theEuropean context by authors like Beck and Beck & Beck-Gernsheim (1995b). The findingsindicate the important functions of personal ties in later life, where they provide not onlypractical and emotional help, but also support older people’s sense of belonging and identity.In addition, older people do not exchange help based on normative moral obligations (e.g.familism), but they actively create, negotiate and manage their personal relationships.5.3. Implications of the findingsThe previous section has indicated the contribution of the findings to the socialcapital theory, providing a clear distinction between the concepts of strong and weak tiesand bonding and bridging social capital. It has also provided research based evidence toexamine the applicability of the theory of individualization to the Chilean context. Overall,the findings stress the importance of considering the different forms in which older peoplebecome integrated in society and the extent to which such integration is shaped by personalas well as contextual circumstances. In this manner, the study has other conceptual as well aspractical implications that can inform future studies, policy development and professionalsworking with older people:176 The narrative approach to analyze social policy on aging, although perfectible, is anattempt to make visible and provide a clear conceptually and empirically groundedframework to conduct a narrative analysis to social policies to recognize the normativeand cultural aspects of social policies on aging. Studies based on a personal community approach have mainly (if not only) beenconducted in the European and North American context and have not generally focusedon older populations (Chua, Madej, & Wellman, 2009a; Pahl & Spencer, 2010; Phillipson,2003; Wall & Gouveia, 2014). Thus, the research findings also contribute to the availableliterature on changes in family structure and meanings in the context of late modernity,adding the experiences of older people living in the global south. The findings indicate acomplex context where the availability of support and care in late life depends oninterrelated circumstances that include the history and quality of older people’srelationships with family ties and friends, the institutional context and its changes at thelevel of values and structures, and the possibilities of older people to manage theirnetworks of relationships. In the particular Chilean context, the research findings advance knowledge on thequalitative aspects of older people’s personal relationships, the potential that theirpersonal networks have to provide support in late life, as well as the way in which olderpeople contribute to their close family and non-family ties. At the same time, the studyprovides a research based understanding of the ways that the individualization processcan influence the meanings of personal relationships, a topic that has been largelydiscussed but rarely studied in Chile. The consideration of the particular socio-historical177processes of Chile also provides context to advance understanding on the function ofpersonal ties in a context of low social and institutional trust. The recognition of different types of friendships and the gender differences in creatingand maintaining these friendships during the life course is central to successfullypromoting connectedness in late life. The study participants’ experiences of friendshiphave shown that different strategies need to be used if we are to foster friendship in menand women. While the former tend to have work-bound and life-long friends, women aremore inclined to develop new friendships in different contexts. Programs directed toincrease social participation and connectedness in later life should recognize that thesegender differences when developing, for instance, male oriented community programsthat attend to their specific interests and particular experiences of aging. If we considerthe importance of work and professional interests in men’s identity, perhaps a good startis planning workshops that allow them to learn a new skill or practice and share anestablished one. Related to the latter, social policies can benefit from recognizing the diverse forms inwhich older people engage with their families and community, which goes beyond theirparticipation in recreational activities. For instance, the consideration of the specificsituations in which older men tend to interact with new people in a meaningful way isalso central to helping in their preparation and adaptation to post-retirement life andprevention of their social isolation. The findings have shown that not all the ties included in older people’s network ofrelationships are positive, and serious situations of dependency are strengthened due to178insufficient or inadequate support from the state or other organizations. Theconsideration of different situations of physical and psychological violence as well asabandonment, as stated in the Policy, is necessary but not sufficient to foster effectivemeans of social inclusion and integration in late life. Thus, the diverse ties that couldprovide support to older people as well as the quality of the relationships with those tiesneed to be considered when assessing the support networks of older people (e.g.questionnaires used by social services to assess social support). Social policies designed to improve the quality of life in older people could benefit fromconsidering the array of ties that form older people’s personal networks and the differentfunctions that those ties play beyond assumptions of the strength and usefulness offamily ties. For instance, recognition of the important role played by emotionally closeties can be at the basis of programs that seek to decrease social isolation in late life.These programs could include the training in and the use of communication technologiesthat help to keep people connected. The current Policy on aging includes programs totrain older people in the use of computers and internet through the "Biblioredes"Program. The Program is directed to the broad population and a 14% of the participantsare older people. This is a good advance to facilitate the access of older people totechnology. The program could be further improved if the particular needs of olderpeople in risk of isolation are included, such as geographical closeness and accessibility ofthe venues and the possibility of accessing computers and other IT equipment after thecourse is finished. As indicated by study participants, the use of technology (e.g. email,Facebook, Skype, and others) are key to keep friends from the same cohort in touch and179re-connect with old friends. If improving connectedness in later life is included among itsobjectives, programs like Biblioredes could help to reduce the problems of geographicaldistance and/or decreasing mobility as people grow old or problems of increasinggeographical mobility (immigration) of family members.5.4. Study limitations and future areas of researchThe findings and conclusions of this thesis represent only the experiences of the studyparticipants. It is beyond the scope of this study, however, to generalize the findings –forwhich qualitative methodologies are not well suited. Overall my motivations to inquire onpersonal relationships in late life were always related to better understand how older peopledefined their personal ties and how that influenced how they mobilized the social capitalembedded in their networks. Although small, I tried to make the sample as varied as possibleto avoid only reflecting the experiences, for example, of only a young cohort of middle andupper class, educated and married older people. In this regard, I attempted to include theexperiences of men and women of different ages within the selected cohort of people,different marital status, living arrangements, and diverse socio-economic status. However,we must remember that all these participants live in the Santiago Metropolitan Region andpresent a higher than average educational level in comparison with Chilean population aged60 years and older. Thus the experiences of the study participants reflect those of an urbanand relatively well educated population. This influences the quantity and quality of programsand services they can access as well as the specific dynamics they establish with familymembers, friends, and the community in general.180The use of the resource generator to capture some aspects of older people’s socialcapital also had some limitations, as well as strengths. The original idea was to use an openended questionnaire to explore the different types of support exchanged by older people.However, as stated on chapter 2, the complexities of talking about help resulted in a changeof strategy. Thus, while in the process of doing the fieldwork, I decided to create a resourcegenerator with closed ended questions. The strengths of this questionnaire were that it waseasy to use in a short period of time; provided standard questions for all the participants, andmade general analysis easier. It also made sense to the interviewees as the questions werecentred on the Chilean context and contained items that could be expanded throughdiscussion and conversation during the interviews. However, another strategy that couldhave been perhaps more appealing and engaging for the study participants was to make amore extensive use of the personal communities’ method to produce information about theresources exchanged between the study participants and the members of their networks.However, that requires more time to conduct the interview and may have required twointerview sessions. It also would have needed more time to develop an interview schedulethat adapted the personal community method around questions and activities that enabledthe research to ‘map’ the resources of each participant.I achieved a varied sample in terms of gender, living arrangements, marital status,and socio-economic status. However an analysis of the data that takes full advantage of thedifferent variables that this variety provide can go further than what I have accomplished inthis thesis. I focused on gender differences in composition of personal communities andsupport networks, but more variables can be considered to better understand the underlying181mechanisms influencing those gender differences. Interestingly, the analysis of personalcommunities indicated that gender and cohort interact in the case of the study participants,showing that men between 67 and 74 years old are the ones who base their network onfamily ties. This finding might be suggesting not only that this particular cohort of men are atrisk, but also that the younger cohort has adopted strategies similar to the ones used bywomen to develop and maintain their networks of relationships. Of course, more researchneeds to be done to explore this issue, where we could include other variables (e.g. socio-economic status) that might be influencing the strategies to incorporate non-kin ties in thenetwork of support and meaningful relationships.When planning how to conduct the study, I decided to give a broad account ofChilean older people’s social capital instead of focusing on more specific groups or topics.This decision was based on the limited information available in Chile regarding the socialworld of older people, particularly of those who have recently crossed the sociallyconstructed boundary of midlife to enter into older age. Of course, this decision of privilegingcomprehensiveness over focus comes with some drawbacks. For instance, the differentaspects of social capital or the complexity of biographical experiences influencing personalrelationships in late life are only suggested but not fully examined. As mentioned earlier, achallenge in the analysis was to identify and explain gender differences, as the logic used bythe study participants was similar, but their practices showed some differences. For example,both men and women indicated the important distinction between chosen and given ties todescribe kin and non-kin ties. However, in practice, men mostly resorted to family ties toexchange support, while women had more varied networks of support and emotionally close182people. The approach and sample size used in this study did not allow exploring further thecauses and specific implications of these different experiences.The cross-sectional approach of this study also impeded an analysis of whetheryounger cohorts of men are in fact better adapted to maintain or recreate a varied personalnetwork –thus indicating that different values and definitions influence their practices– or,instead, men tend to decrease the size and variety of their networks as they grow older dueto reasons that need to be further explored. We may ask, for example, about how supportexchange and emotional investment are similar or different for men and women entering oldage. If, as in our sample, women are more involved in the labour market, why do they stillhave larger and more varied networks than men?As indicated in the methodology, the findings are not based on intriguing but uniquequotes or cases, but on the experiences of all the participants. Thus, despite the limitations, Ido believe that the decision of privileging breadth over focus proved useful. It allowed us toexplore the shared definitions of personal relationships and support in late life. It alsoenabled us to better understand the rationale of the participants to exchange help withdifferent people and to develop networks of emotional support with more or less presenceof family and non-family ties. In sum, the findings presented here provide several startingpoints to conduct more specific research in the future focusing on emergent aspects of olderpeople’s experiences of aging. Some avenues for such future research based on the researchfindings and the many questions that arise, include:183 Altruistic help among members of personal communities. The personal communitiesmethod was well received by study participants and supported very well by the findingsfrom the study of social capital. Reciprocity among kin and non-kin ties can be studied byusing different methodologies that help us to identify the more subtle types of helpexchanged on a daily basis. For instance, we could use direct observation of olderpeople’s routines and/or personal diaries to identify those ‘invisible’ exchanges of helpthat represent altruistic forms of reciprocity. Independence and negotiated reciprocity. Future studies can compare and contrast thedifferent narratives of independence present among family members and close non-kinties and how these narratives influence the way in which the exchange of help isnegotiated. Some questions to be explored are: When and with whom does reciprocityneed to be negotiated? and, What are the similarities and gaps in different generationsliving in different contexts regarding the meaning and value of help exchange? Thequestion about negotiated reciprocity can also provide a way to better understandgender differences in personal network composition and sources of support. We may askwhether younger cohorts of men and women share the same definitions regardingsupport and independence, as well as preferences and strategies for exchanging helpwith family and friendship ties. Trajectories of friendship. The different types of friendship identified can be analyzedusing a life course perspective, as many friendships are created, activated or kept latentdepending on the life stages and transitions faced by the person. In late life, we canexplore how different friends have accompanied the older person, when and why these184friends have been put on hold or called forward. For instance we can ask: How do key lifetransitions (e.g. retirement, widowhood, divorce) affect friendship trajectories andfriendship ties? Is this the same for men and women? Are the same types of friendsimportant when the transition occurs earlier or later in life? How does the death of a longstanding friend in late life affect personal communities? Friendship developed in older people’s associations. Following Sarah Matthews’ typologyof friendship styles and the classifications of personal communities proposed in thisthesis, we can identify the type of friendships developed and/or strengthened throughparticipation in older people’s associations. Are some types of association moreappropriated for the creation of certain friendship styles? Are some types of associationsmore effective depending on the gender of the participants? Influence of living environment on older people’s social capital and personal communities.The infrastructure of the environment in which older people live and move, and even thesocioeconomic segregation reflected in city spaces, can influence their opportunities tointeract and socialize with others (e.g. affordable and/or efficient public transportation,even sidewalks, accessible parks, perception of secure spaces). These issues are relevantin the Chilean context, as cities and urban infrastructure have not been planned to beaccessible and improvements are largely been successfully implemented mostly inwealthy municipalities. Thus, some questions that connect the living environment withpersonal networks in late life are: How do more or less age friendly communities impactolder people’s opportunities to develop weak ties? Do age friendly communities facilitatethe creation and maintenance of bonding social capital? To what extent does moving to a185new neighborhood and downsizing change older people’s personal communities andsocial capital? How does bonding and bridging social capital work in rural and urbancommunities in a context of centralized services such as Chile? Social capital in late life in context of low and high trust. If we analyze different socialcontexts, for instance a different region or a rural area, would we find similar resultsregarding how bonding and bridging social capital work? How important is bridging socialcapital for accessing public and private services in a context where there is a higher levelof institutional trust?The current study has primarily used a qualitative methodology. Other suggestedstudies could use a mixed methods approach depending on their particular objective. Forinstance, combining a more complex close-ended questionnaire to measure social capital inlate life that allow examining reciprocity among different kin and non-kin ties that form thenetwork of relationships of the older person. This could be complemented by qualitativemethods, such as walking interviews with the purpose of observe daily life interaction withthe social environment, diaries and/or direct observation of daily routine and interactions toexamine more subtle exchanges of help and interactions that are minimally grasped orrecognized by study participants.Another alternative for the suggested could be conducting a longitudinal study thattakes into consideration the impacts on social capital of changing resources regarding use oftechnology, housing alternatives and living arrangements in a context of increasingly agingdemographics. This study could also help to analyze how the role of family regarding older186people’s care and support change over time, and whether different values and social normsshape our networks of relationships. Although ambitious, such a study could be particularlyrelevant for the Chilean context, as well as other countries experiencing social changes atdifferent levels; changes that have not been systematically characterized in interaction withthe rapid process of population aging. The generation of adequate and updated frameworksrequires more empirical research that enable the conceptualization of late life with focus onboth the influence of institutions and social structures on older people and the impact ofolder individuals in society, considering their own definitions, strategies and resources tomanage their experiences of aging.187BibliographyAboderin, I. (2005). Changing family relationships in developing nations. In V. L. Bengtson, P. G.Coleman, & T. B. L. Kirkwood (Eds.), The Cambridge Handbook of Age and Ageing (1st ed., pp.469–475). New York: Cambridge University Press.Adams, R. G., & Allan, G. (1999). Contextualising friendship. In Placing Friendship in Context.Cambridge University Press.Allan, G. (2001). Personal relationships in late modernity. Personal Relationships, 8(3), 325–339.https://doi.org/10.1111/j.1475-6811.2001.tb00043.xAllan, G. (2010). Friendship and ageing. In D. Dannefer & C. Phillipson (Eds.), The SAGE Handbook ofSocial Gerontology The SAGE handbook of social gerontology (pp. 239–248). London: SAGE.Antonucci, T. C. (1986). Hierarchical mapping technique. Generations: Journal of the American Societyon Aging, 10(4), 10–12.Antonucci, T. C., Birditt, K., & Ajrouch, K. (2011). Convoys of Social Relations: Past, Present, andFuture. In K. L. Fingerman, C. Berg, J. Smith, & T. C. Antonucci (Eds.), Handbook of Life-SpanDevelopment (pp. 161–182). New York: Springer.Araujo, K., & Martuccelli, D. (2014). Beyond institutional individualism: Agentic individualism and theindividuation process in Chilean society. Current Sociology, 62(1), 24–40.https://doi.org/10.1177/0011392113512496ASEP/JDS. (2009). ASEP/JDS Interpersonal trust map. Retrieved February 26, 2014, fromhttp://www.jdsurvey.net/jds/jdsurveyMaps.jsp?Idioma=I&SeccionTexto=0404&NOID=104Baars, J., & Phillipson, C. (2014). Connecting meaning with social structure: theoretical foundations. InJ. Baars, J. Dohmen, A. Grenier, & C. Phillipson (Eds.), Ageing, meaning and social structure(pp. 11–30). Chicago: The Policy Press.188Barrett, P., Hale, B., & Butler, M. (2013). Family Care and Social Capital: Transitions in Informal Care.Springer Science & Business Media.Barros L., C. (1991). Viviendo el envejecer. Santiago de Chile: Instituto de Sociología, PontificiaUniversidad Católica de Chile.Bauman, Z. (2002). Foreword by Zygmunt Bauman: Individually, Together. In Theory, Culture &Society: Individualization: Institutionalized individualism and its social and politicalconsequences (pp. xiv–xx). London: SAGE Publications Ltd.Beck, U. (1992a). Risk Society. Towards a New Modenrity (Reprinted edition). Thousand Oaks: Sage.Beck, U. (1992b). Risk Society: Towards a New Modernity. SAGE.Beck, U., & Beck-Gernsheim, E. (1995a). The Normal Chaos of Love. Wiley.Beck, U., & Beck-Gernsheim, E. (1995b). The normal chaos of love. Cambridge, UK : Cambridge, MA,USA: Polity Press ; Blackwell.Beck, U., & Beck-Gernsheim, E. (2002). Individualization: institutionalized individualism and its socialand political consequences. London ; Thousand Oaks: Sage.Benaquisto, L. (2008). Open Coding. In L. Given, The SAGE Encyclopedia of Qualitative ResearchMethods. Thousand Oaks: SAGE.Bernstein, S. (Ed.). (2010). The Chilean pension system. Santiago de Chile: Superintendencia dePensiones. Retrieved from http://www.safp.cl/portal/informes/581/articles-8557_recurso_1.pdfBiggs, S. (2001). Toward critical narrativity: Stories of aging in contemporary social policy. Journal ofAging Studies, 15(4), 303–316. https://doi.org/10.1016/S0890-4065(01)00025-1Biggs, S. (2005). Aging and family policy: a sociological excursion. The Journal of Sociology & SocialWelfare, 32(2), 63–74.189Biggs, S., & Powell, J. (2003). Older People and Family Policy. In V. L. Bengtson & A. Lowenstein (Eds.),Global Aging and Its Challenge to Families (pp. 103–119). New York: Walter de Gruyter.Boneham, M. A., & Sixsmith, J. A. (2006). The voices of older women in a disadvantaged community:issues of health and social capital. Social Science & Medicine (1982), 62(2), 269–279.https://doi.org/10.1016/j.socscimed.2005.06.003Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and researchfor the sociology of education (pp. 241–258). New York: Greenwood Press.Bowling, A., Banister, D., Sutton, S., Evans, O., & Windsor, J. (2002). A multidimensional model of thequality of life in older age. Aging & Mental Health, 6(4), 355–371.https://doi.org/10.1080/1360786021000006983Brodie, J. (2007). Reforming Social Justice in Neoliberal Times. Studies on Social Justice, 1(2), 93–107.Cagney, K. A., & Wen, M. (2008). Social Capital and Aging-Related Outcomes*. In I. Kawachi, S. V.Subramanian, & D. Kim (Eds.), Social Capital and Health (pp. 239–258). Springer New York.Retrieved from http://link.springer.com.ezproxy.library.ubc.ca/chapter/10.1007/978-0-387-71311-3_11Calvo, E. (2013). Antecedentes Y Desarrollos Recientes Del Sistema De Salud Chileno (Background andRecent Developments in Chilean Health System) (SSRN Scholarly Paper No. ID 2527988).Rochester, NY: Social Science Research Network. Retrieved fromhttp://papers.ssrn.com/abstract=2527988Calvo, E. (2014). Comparative-historical Analysis of Aging Policy Reforms in Argentina, Chile, CostaRica, and Mexico (Working Paper No. 61). Facultad de Economía y Empresa, UniversidadDiego Portales. Retrieved from https://ideas.repec.org/p/ptl/wpaper/61.htmlCastells, M. (2010). End of Millennium (Second). Oxford, UK: Wiley-Blackwell. Retrieved fromhttp://doi.wiley.com/10.1002/9781444323436190Charmaz, K. (2006). Constructing Grounded Theory: A Practical Guide through Qualitative Analysis (1sted.). Thousand Oaks: SAGE.Charmaz, K. (2014). Constructing Grounded Theory. Thousand Oaks: SAGE.CHS. (2003). Human Security Now (p. 168). New York: Commission on Human Security. Retrieved fromhttp://reliefweb.int/sites/reliefweb.int/files/resources/91BAEEDBA50C6907C1256D19006A9353-chs-security-may03.pdfChua, V., Madej, J., & Wellman, B. (2009a). Personal Communities as Sources of Social Support. CJNR(Canadian Journal of Nursing Research), 41(3), 11–17.Chua, V., Madej, J., & Wellman, B. (2009b). Personal communities: the world according to me.Toronto, Canada. Retrieved fromhttp://homes.chass.utoronto.ca/~wellman/publications/personal/handbook-community.htmClark, P. (1993). Public policy in the United States and Canada: individualism, familial obligation, andcollective responsibility in the care of the elderly. In J. Hendricks & C. Rosenthal (Eds.), TheRemainder of their days: domestic policy and older families in the United States and Canada(pp. 13–48). New York: Garland.Clark, P. (2011). The narrative frame in discourse on aging: understanding facts and values behindpublic policy. In G. Kenyon, E. Bohlmeijer, & W. L. Randall (Eds.), Storying Later Life: Issues,Investigations, and Interventions in Narrative Gerontology (pp. 84–97). New York: OxfordUniversity Press.Collins, J. (1995). Chile’s Free Market Miracle: A Second Look. Oakland: Food First Books.Comite Nacional de Estadisticas Vitales. (2012). Anuario de Estadísticas Vitales 2012. Santiago ofChile: Instituto Nacional de Estadísticas, Servicio de Registro Civil e Identificación, Ministeriode Salud.Connidis, I. A. (2009). Family Ties and Aging (2nd ed.). Los Angeles: SAGE.191Corbin, J. M., & Strauss, A. C. (2008). Basics of Qualitative Research: Techniques and Procedures forDeveloping Grounded Theory (3rd ed.). Thousand Oaks: Sage.Cornwell, B. (2011). Independence Through Social Networks: Bridging Potential Among Older Womenand Men. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences,gbr111. https://doi.org/10.1093/geronb/gbr111Cornwell, B., Laumann, E. O., & Schumm, L. P. (2008). The Social Connectedness of Older Adults: ANational Profile. American Sociological Review, 73(2), 185–203.de Belvis, A. G., Avolio, M., Spagnolo, A., Damiani, G., Sicuro, L., Cicchetti, A., … Rosano, A. (2008).Factors associated with health-related quality of life: the role of social relationships amongthe elderly in an Italian region. Public Health, 122(8), 784–793.https://doi.org/10.1016/j.puhe.2007.08.018de Souza Briggs, X. (2003). Bridging networks, social capital, and racial segregation in America (KSGFaculty Research Working Paper Series). Cambridge, MA: John F. Kennedy School ofGovernment.Esping-Andersen, G. (2006). Three Worlds of Wealfare Capitalism. In C. Pierson & F. G. Castles (Eds.),The welfare state reader (Second, pp. 160–174). Cambridge: Polity.Espinoza, V., Barozet, E., & Méndez, M. L. (2013). Estratificación y movilidad social bajo un modeloneoliberal: El caso de Chile. Lavboratorio, (25), 169–192.Estes, C. L., & Phillipson, C. (2002). The globalization of capital, the welfare state, and old age policy.International Journal of Health Services: Planning, Administration, Evaluation, 32(2), 279–297.Ferlander, S. (2007). The Importance of Different Forms of Social Capital for Health. Acta Sociologica,50(2), 115–128. https://doi.org/10.1177/0001699307077654Fine, M. (2006). A Caring Society?: Care and the Dilemmas of Human Services in the 21st Century. NewYork: Palgrave Macmillan.192Fiscella, J. M. G., Rivera, J., & Román, H. (2009). Aplicación del modelo de redes personales al estudiode los ancianos dependientes. Revista Multidisciplinar de Gerontología, 19(3), 121–129.Fischer, F. (2003). Reframing Public Policy: Discursive Politics and Deliberative Practices. Oxford ; NewYork: Oxford University Press. Retrieved fromhttp://www.oxfordscholarship.com/view/10.1093/019924264X.001.0001/acprof-9780199242641Forsman, A. K., Herberts, C., Nyqvist, F., Wahlbeck, K., & Schierenbeck, I. (2013). Understanding therole of social capital for mental wellbeing among older adults. Ageing & Society, 33(05), 804–825. https://doi.org/10.1017/S0144686X12000256Garretón, M. A. (2004). Modernities and democracies in the globalized world. A view from LatinAmerica. Presented at the Symposium The Centre and the Peripheries. Divergences andChallenges, Iran: The Center for Dialogue.Giddens, A. (1992). The transformation of intimacy: sexuality, love and eroticism in modern society.Cambridge: Policy Press.Giddens, A. (1994). Living in a post-traditional society. In U. Beck, A. Giddens, & S. Lash (Eds.),Reflexive modernization: politics, tradition and aesthetics in the modern social order.Cambridge: Policy Press.Glaser, B., & Strauss, A. (1967). The Discovery of Grounded Theory: Strategies for QualitativeResearch. Hawthorne: Aldine.Gobierno de Chile. (2012). Examen y evaluación de la declaración de Brasilia en América Latina y elCaribe (p. 202). Santiago de Chile.Granovetter, M. S. (1973). The Strength of Weak Ties. American Journal of Sociology, 78(6), 1360–1380. https://doi.org/10.2307/2776392193Gray, A. (2008). The social capital of older people. Ageing and Society, 29(01), 5–31.https://doi.org/10.1017/S0144686X08007617Grenier, A. (2012). Transitions and the Lifecourse: Challenging the Constructions of “Growing Old.”Bristol; Chicago, IL: Policy Press.Güell, P. (1999). Familia y modernización en Chile. Presented at the Exposición ante la Comisión deExpertos en Temas de Familia, Santiago de Chile: SERNAM.Guzmán, J. M., Huenchuán, S., & Montes de Oca, V. (2003). Redes de apoyo social de las personasmayores: marco conceptual. In Notas de Población (pp. 35–70). Chile: CEPAL.Herrera, S., & Kornfeld, R. (2008). Relaciones familiares y bienestar de los adultos mayores en Chile.En Foco, (131), 1–15.Higo, M., & Williamson, J. B. (2011). Global Aging. In R. A. Settersten, & J. L. Angel (Eds.), Handbook ofSociology of Aging (pp. 117–129). New York, NY: Springer New York. Retrieved fromhttp://www.springerlink.com/content/r577046875l037l7/Huenchuan, S. (2004). Políticas sobre vejez en América Latina: elementos para su análisis y tendenciasgenerales. Notas de Población, 78, 183–204.Huenchuan, S. (Ed.). (2009). Envejecimiento, derechos humanos y políticas públicas. Santiago de Chile:CEPAL.Huenchuan, S. (2012). Igualdad y universalidad de los derechos humanos en contexto deenvejecimiento. In S. Huenchuan (Ed.), Los derechos de las personas mayores en el siglo XXI:situación, experiencias y desafíos (pp. 19–55). Mexico DF: CEPAL.Hughes, M. E., & Waite, L. J. (2007). The aging of the second demographic transition. In K. W. Schaie &P. Uhlenberg (Eds.), Social structures: the impact of demographic changes on the well-being ofolder persons (pp. 179–211). New York: Springer.194Huxhold, O., Miche, M., & Schüz, B. (2014). Benefits of Having Friends in Older Ages: DifferentialEffects of Informal Social Activities on Well-Being in Middle-Aged and Older Adults. TheJournals of Gerontology Series B: Psychological Sciences and Social Sciences, 69(3), 366–375.https://doi.org/10.1093/geronb/gbt029ICSO. (2014). Encuesta Nacional UDP. Santiago, Chile: ICSO-Universidad Diego Portales.INE. (2003). Censo 2002. Chile.INE. (2006). Fecundidad en Chile. Chile.Kahn, R. L., & Antonucci, T. C. (1981). Convoys of Social Support: A Life-course Approach. In S. B.Kiesler & V. K. Oppenheimer (Eds.), Aging: Social Change (pp. 383–405). New York: AcademicPress.Kawachi, I., Kennedy, B. P., Lochner, K., & Prothrow-Stith, D. (1997). Social capital, income inequality,and mortality. American Journal of Public Health, 87(9), 1491–1498.Kawachi, I., Subramanian, S. V., & Kim, D. (2008). Social capital and health. Springer.Keating, N., & Dosman, D. (2009). Social Capital and the Care Networks of Frail Seniors. CanadianReview of Sociology/Revue Canadienne de Sociologie, 46(4), 301–318.https://doi.org/10.1111/j.1755-618X.2009.01216.xKeating, N., Swindle, J., & Foster, D. (2005). The Role of Social Capital in Aging Well. In Policy ResearchInitiative of the Government of Canada (Ed.), Social capital in action: thematic policy studies(pp. 24–48). Ottawa, ON: Canada. Policy Research Initiative.Kemp, C. L., & Denton, M. (2003). The allocation of responsibility for later life: Canadian reflections onthe roles of individuals, government, employers and families. Ageing and Society, 23(6), 737–760. https://doi.org/10.1017/S0144686X03001363195King, A. (2008). In Vivo Coding. In L. Given, The SAGE Encyclopedia of Qualitative Research Methods.Thousand Oaks: SAGE. Retrieved fromhttp://knowledge.sagepub.com/view/research/n240.xmlKraft, M. E., & Furlong, S. R. (2012). Public Policy: Politics, Analysis, and Alternatives. SAGE.Krause, N., & Borawski-Clark, E. (1994). Clarifying the Functions of Social Support in Later Life.Research on Aging, 16(3), 251–279. https://doi.org/10.1177/0164027594163002Krause, N., & Rook, K. S. (2003). Negative Interaction in Late Life: Issues in the Stability andGeneralizability of Conflict Across Relationships. The Journals of Gerontology Series B:Psychological Sciences and Social Sciences, 58(2), P88–P99.https://doi.org/10.1093/geronb/58.2.P88Kritzer, B. (2008). Chile’s Next Generation Pension Reform. Social Security Bulletin, 68(2). Retrievedfrom http://www.ssa.gov/policy/docs/ssb/v68n2/v68n2p69.htmlLaliberte Rudman, D. (2006). Shaping the active, autonomous and responsible modern retiree: ananalysis of discursive technologies and their links with neo-liberal political rationality. Ageing& Society, 26(02), 181–201. https://doi.org/10.1017/S0144686X05004253Lancee, B. (2010). The Economic Returns of Immigrants’ Bonding and Bridging Social Capital: The Caseof the Netherlands1. International Migration Review, 44(1), 202–226.https://doi.org/10.1111/j.1747-7379.2009.00803.xLang, F. R., & Carstensen, L. L. (1994). Close emotional relationships in late life: Further support forproactive aging in the social domain. Psychology and Aging, 9(2), 315–324.https://doi.org/10.1037/0882-7974.9.2.315Latinobarometro. (2013). Latinobarometro online data analysis. Retrieved February 26, 2015, fromhttp://www.latinobarometro.org/latOnline.jsp196Lechner, N. (1999). Desafíos de un Desarrollo Humano: Individualización y Capital Social. Presented atthe Foro Desarrollo y Cultura, Paris: BID.Leonard, M. (2004). Bonding and Bridging Social Capital: Reflections from Belfast. Sociology, 38(5),927–944. https://doi.org/10.1177/0038038504047176Leonard, R., & Onyx, J. (2003). Networking Through Loose and Strong Ties: An Australian QualitativeStudy. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 14(2), 189–203. https://doi.org/10.1023/A:1023900111271Litwin, H. (2006). Social networks and self-rated health: a cross-cultural examination among olderIsraelis. Journal of Aging and Health, 18(3), 335–358.https://doi.org/10.1177/0898264305280982Litwin, H. (2009). Social Networks and Well-being: A Comparison of Older People in Mediterraneanand Non-Mediterranean Countries. The Journals of Gerontology Series B: PsychologicalSciences and Social Sciences, 65B, 599–608. https://doi.org/10.1093/geronb/gbp104Lubben, J., & Gironda, L. (2003). Centrality of Social Ties to the Health and Well-being of Older Adults.In B. Berkman & L. Harootyan (Eds.), Social Work and Health Care in an Aging Society (pp.319–350). New York: Springer.Lynch, J., Due, P., Muntaner, C., & Smith, G. (2000). Social capital—Is it a good investment strategy forpublic health? Journal of Epidemiology and Community Health, 54(6), 404–408.https://doi.org/10.1136/jech.54.6.404Marshall, V. W., & Bengtson, V. L. (2011). Theoretical Perspectives on the Sociology of Aging. In R. A.Settersten, & J. L. Angel (Eds.), Handbook of Sociology of Aging (pp. 17–33). New York, NY:Springer New York. Retrieved fromhttp://www.springerlink.com/content/m311vv5212473487/197Marshall, V. W., & Clarke, P. (2010). Agency and social structure in aging and life-course research. InD. Dannefer & C. Phillipson (Eds.), The Sage Handbook of Social Gerontology (pp. 294–306).Sage Publications (CA).Mason, J. (2002). Qualitative Researching (2nd ed.). Thousand Oaks: Sage.Matthews, S. H. (1983). Definitions of Friendship and their Consequences in Old Age. Ageing &Society, 3(02), 141–155. https://doi.org/10.1017/S0144686X00009983Matthews, S. H. (1986). Friendships through the life course: oral biographies in old age. Beverly Hills:Sage Publications.McDaniel, S., & Gazso, A. (2014). Liminality and low-income aging families by choice: meanings offamily and support. Canadian Journal on Aging = La Revue Canadienne Du Vieillissement,33(4), 400–412. https://doi.org/10.1017/S0714980814000270Menanteau-Horta, D. (2006). El rol del Estado en el desarrollo social y la Reforma de la previsión enChile y EE.UU. Revista Austral de Ciencias Sociales, (10), 5–22.Messeri, P., Silverstein, M., & Litwak, E. (1993). Choosing Optimal Support Groups: A Review andReformulation. Journal of Health and Social Behavior, 34(2), 122–137.https://doi.org/10.2307/2137239Ministerio de Desarrollo Social. (2011). Adulto Mayor Casen 2011. Santiago de Chile. Retrieved fromhttp://observatorio.ministeriodesarrollosocial.gob.cl/casen_publicaciones.phpMinisterio de Desarrollo Social. (2012, December 10). Directora Nacional de Senama difunde PolíticaIntegral de Envejecimiento Positivo. Retrieved October 9, 2015, fromhttp://www.ministeriodesarrollosocial.gob.cl/seremi/libertador-bernardo-ohiggins/prensa/2012/10/01/directora-nacional-de-senama-difunde-politica-integral-de-envejecimiento-positivoMorgan, D. (1996). Family Connections: An Introduction to Family Studies. Cambridge: Polity Press.198Mujere, N. (2016). Sampling in Research. In B. L Mette (Ed.), Mixed Methods Research for ImprovedScientific Study (pp. 107–121). IGI Global.Murad, P. (2003). Transferencias informales de apoyo de los adultos mayores en América latina y elCaribe: Estudio Comparativo de Encuestas Sabe (Notas de Población No. 77) (pp. 175–218).Retrieved from http://www.google.com/Murayama, H., Nishi, M., Matsuo, E., Nofuji, Y., Shimizu, Y., Taniguchi, Y., … Shinkai, S. (2013). Dobonding and bridging social capital affect self-rated health, depressive mood and cognitivedecline in older Japanese? A prospective cohort study. Social Science & Medicine, 98, 247–252. https://doi.org/10.1016/j.socscimed.2013.09.026Norstrand, J. A., & Xu, Q. (2011). Social Capital and Health Outcomes Among Older Adults in China:The Urban–Rural Dimension. The Gerontologist, gnr072.https://doi.org/10.1093/geront/gnr072Nyqvist, F., Forsman, A. K., Giuntoli, G., & Cattan, M. (2012). Social capital as a resource for mentalwell-being in older people: A systematic review. Aging & Mental Health, 17(4), 394–410.https://doi.org/10.1080/13607863.2012.742490Nyqvist, F., Gustavsson, J., & Gustafson, Y. (2006). Social Capital and Health in the Oldest Old: TheUmeå 85+ Study. International Journal of Ageing and Later Life, 1(1), 91–114.https://doi.org/10.3384/ijal.1652-8670.061191Osorio, P. (2006). La longevidad: más allá de la biología. Aspectos socioculturales. Papeles Del CEIC,(2), 1–28.Oszlack, O., & O’Donell, G. (1995). Estado y políticas estatales en América Latina: hacia una etsrategiade investigación. Redes Revista de estudios sociales de la ciencia, 4, 99–128.Pahl, R., & Spencer, L. (2004). Capturing personal communities. In Social Networks and SocialExclusion: Sociological and Policy Perspectives (pp. 72–96). Aldershot: Ashgate Pub Ltd.199Pahl, R., & Spencer, L. (2010). Family, Friends, and Personal Communities: Changing Models-in-the-Mind. Journal of Family Theory & Review, 2(3), 197–210. https://doi.org/10.1111/j.1756-2589.2010.00053.xPatton, M. Q. (1990). Qualitative Evaluation and Research Methods (2nd ed.). Newbury Park, CA:SAGE Publications Ltd.Patulny, R. V., & Lind Haase Svendsen, G. (2007). Exploring the social capital grid: bonding, bridging,qualitative, quantitative. International Journal of Sociology and Social Policy, 27(1/2), 32–51.https://doi.org/10.1108/01443330710722742Phillipson, C. (2003). From family groups to personal communities: social capital and social change inthe family of older adults. In V. L. Bengtson & A. Lowenstein (Eds.), Global aging andchallenges to families (pp. 54–74). New York: Aldine de Gruyter.Phillipson, C. (2005). The political economy of old age. In M. L. Johnson, V. L. Bengtson, P. G. Coleman,& T. B. L. Kirkwood (Eds.), The Cambridge Handbook of Age and Ageing (pp. 502–509). NewYork: Cambridge University Press.Phillipson, C. (2013). Ageing (1st ed.). Cambridge: Polity.Phillipson, C. (2015). The Political Economy of Longevity: Developing New Forms of Solidarity for LaterLife. The Sociological Quarterly, 56(1), 80–100. https://doi.org/10.1111/tsq.12082Phillipson, C., Allan, G., & Morgan, D. H. J. (2004). Social Networks and Social Exclusion: Sociologicaland Policy Perspectives. Ashgate Pub Ltd.Phillipson, C., Bernard, M., Phillips, J., & Ogg, J. (2002). Family and Community Life of Older People:Social Networks and Social Support in Three Urban Areas. London: Routledge.Pillemer, K. A. (2000). Social integration in the second half of life. JHU Press.PNUD (Ed.). (2004). Desarrollo humano en Chile: nosotros los chilenos : un desafío cultural (1. ed, Vol.2). Santiago : LOM Ed.200Portes, A. (1997). Neoliberalism and the Sociology of Development: Emerging Trends andUnanticipated Facts. Population and Development Review, 23(2), 229–259.https://doi.org/10.2307/2137545Portes, A. (1998). Social Capital: Its Origins and Applications in Modern Sociology. Annual Review ofSociology, 24, 1–24. https://doi.org/10.2307/223472Putnam, R. D. (1995). Bowling alone: America’s declining social capital. Journal of Democracy, 6(1),65–78.Putnam, R. D. (2000). Bowling Alone: The Collapse And Revival Of American Community. New York:Simon and Schuster.Reichert, M., & Phillips, J. (2009). The changing generational contract within and outside the family:Britain and Germany compared. In A. Walker & N. Gerhard (Eds.), Social Policy in AgeingSocieties (pp. 142–164). Bedford: Palgrave Macmillan. Retrieved fromhttp://www.palgrave.com%2Fpage%2Fdetail%2Fsocial-policy-in-ageing-societies-alan-walker%2F%3Fk%3D9780230520981%26loc%3DukRiessman, C. K. (2008). Narrative methods for the human sciences. Los Angeles: Sage Publications.Robles, F. (1999). Los sujetos y la cotidianeidad: elementos para una microsociología de loscontemporáneo ; inclusión, exclusión y construcción de identidad ; el caso de las mujeres jefasde hogar en Chile. Talcahuano, Chile: Sociedad Hoy.Robles, F. (2000). El desaliento inesperado de la modernidad. Santiago de Chile: RIL.Roseneil, S., & Budgeon, S. (2004). Cultures of Intimacy and Care beyond “the Family”: Personal Lifeand Social Change in the Early 21st Century. Current Sociology, 52(2), 135–159.https://doi.org/10.1177/0011392104041798Rosenmayr, L., & Köckeis, E. (1963). Propositions for a sociological theory of ageing and the family.International Social Science Journal, 15, 410–426.201Sahlins, M. (1972). On the sociology of primitive exchange. In M. Sahlins (Ed.), Stone age economics(pp. 185–275). New York: Aldine-Atherton.Santander & Centro UC Políticas Públicas. (2005). Confianza, la clave para el desarrollo de Chile.Santiago of Chile.Scharf, T. (1998). Ageing and ageing policy in Germany. Oxford ; New York: Berg.SENAMA. (2012). Política Integral de envejecimiento Positivo para Chile 2012-2025. Santiago de Chile:SENAMA.Settersten, R. A., & Trauten, M. (2009). The new terrain of old age: Hallmarks, freedoms, and risks. InV. L. Bengtson, D. Gans, N. Putney, & M. Silverstein (Eds.), Handbook of Theories of Aging,Second Edition (pp. 455–469). New York: Springer Publishing Company.Shanas, E. (1979a). The Family as a Social Support System in Old Age. The Gerontologist, 19(2), 169–174. https://doi.org/10.1093/geront/19.2.169Shanas, E. (1979b). The Family as a Social Support System in Old Age. The Gerontologist, 19(2), 169–174. https://doi.org/10.1093/geront/19.2.169Sheets, D. J., Bradley, D. B., & Hendricks, J. (2005). Enduring Questions in Gerontology. Springer.Small, M. L. (2010). Unanticipated Gains: Origins of Network Inequality in Everyday Life (Reprintedition). New York: Oxford University Press.Spencer, L., & Pahl, R. (2006). Rethinking friendship: hidden solidarities today. Princeton: PrincetonUniversity Press.Swartz, T. T. (2009). Intergenerational Family Relations in Adulthood: Patterns, Variations, andImplications in the Contemporary United States. Annual Review of Sociology, 35(1), 191–212.https://doi.org/10.1146/annurev.soc.34.040507.134615202Takahashi, K., Tamura, J., & Tokoro, M. (1997). Patterns of Social Relationships and PsychologicalWell-being among the Elderly. International Journal of Behavioral Development, 21(3), 417–430. https://doi.org/10.1080/016502597384721Taylor, M. (2003). The Reformulation of Social Policy in Chile, 1973—2001 Questioning a NeoliberalModel. Global Social Policy, 3(1), 21–44. https://doi.org/10.1177/14680181030030010101Tomassini, C., & Glaser, K. (2007). Gender and support of older unmarried people in Italy and Britain.In A. Pinnelli, F. Racioppi, & R. Rettaroli (Eds.), Genders in the Life Course (Vol. 19, pp. 237–247). Dordrecht: Springer Netherlands.Torche, F., & Wormald, G. (2004). Estratificación y movilidad social en Chile: entre la adscripción y ellogro. (Políticas Sociales No. 98). Chile: UN-CEPAL. Retrieved from http://www.eclac.org/cgi-bin/getProd.asp?xml=/publicaciones/xml/4/20354/P20354.xml&xsl=/dds/tpl/p9f.xsl&base=/tpl/top-bottom.xsltTorche, F., & Wormald, G. (2007). Chile, entre la adscripción y el logro. In R. Franco, A. León, & R.Atria (Eds.), Estratificacion y Mobilidad en America Latina (pp. 339–387). Santiago: LOM.UNDP. (1998). Informe Desarrollo Humano en Chile. Las paradojas de la modernización. Chile: UnitedNations Development Program.United Nations Development Programme. (2010). Desarrollo Humano en Chile 2010. Género:losdesafíos de la igualdad. Santiago: UNDP. Retrieved fromhttp://www.desarrollohumano.cl/informe-2010/PNUD_LIBRO.pdfValdés, X. (2007). Notas sobre la Metamorfosis de la Familia en Chile. Presented at the Reunión deEspecialistas: Futuro de las Familias y Desafíos para las Políticas Públicas, Santiago de Chile:CEPAL-UNFPA. Retrieved fromhttp://www.eclac.org/dds/noticias/paginas/9/30289/Resumen.XimenaValdes.pdf203Valenzuela, E., & Cousiño, C. (2000). Sociabilidad y asociatividad: un ensayo de sociología comparada.Estudios Públicos, 77, 321–339.Van Der Gaag, M., & Snijders, T. A. B. (2005). The Resource Generator: social capital quantificationwith concrete items. Social Networks, 27(1), 1–29.https://doi.org/10.1016/j.socnet.2004.10.001van Tilburg, T., & Thomése, F. (2010). Societal Dynamics in Personal Networks. In The SAGE handbookof social gerontology (pp. 215–225). London: SAGE. Retrieved fromhttp://knowledge.sagepub.com/cite/hdbk_socialgerontology$002fn16.xml;jsessionid=912306741A28962EA19F0F45BA2A9674?nojs=trueVictor, C. R., Scambler, S., & Bond, J. (2009). The social world of older people. Open University Press.Walker, A. (2006). Reexamining the political economy of aging: understanding the structure/agencytension. In J. Baars, D. Dannefer, C. Phillipson, & A. Walker (Eds.), Aging, Globalization AndInequality: The New Critical Gerontology (pp. 59–81). Amityville: Baywood Pub Co.Wall, K., & Gouveia, R. (2014). Changing meanings of family in personal relationships. CurrentSociology, 62(3), 352–373. https://doi.org/10.1177/0011392113518779Wellman, B. (2001). The persistence and transformation of community: from neighbourhood groups tosocial networks. Ottawa: Law Commission of Canada.Wellman, B., Hogan, B., Berg, K., Boase, J., Carrasco, J. A., Côté, R., … Tran, P. (2006). Connected Lives:The Project. In P. Purcell (Ed.), Networked neighbourhoods (pp. 157–211). Guildorf, UK:Springer.Wellman, B., Wong, R. Y., Tindall, D., & Nazer, N. (1997). A decade of network change: Turnover,persistence and stability in personal communities. Social Networks, 19(1), 27–50.https://doi.org/16/S0378-8733(96)00289-4204Widmer, E. D. (2006). Who are my family members? Bridging and binding social capital in familyconfigurations. Journal of Social and Personal Relationships, 23(6), 979–998.https://doi.org/10.1177/0265407506070482Wiles, J. L., Wild, K., Kerse, N., & Allen, R. E. S. (2012). Resilience from the point of view of olderpeople: “There”s still life beyond a funny knee’. Social Science & Medicine, 74(3), 416–424.https://doi.org/10.1016/j.socscimed.2011.11.005Winter, I. (2000). Towards a theorised understanding of family life and social capital (Working PaperNo. 21) (p. 24). Melbourne: Australian Institute of Family Studies. Retrieved fromhttps://aifs.gov.au/publications/towards-theorised-understanding-family-life-and-social205Appendix A: Interview ScheduleDate / /Day/ Month/ YearLast Name:Given Name:Address:Phone number:2061. Introduction: introduce self, restate research purpose, and reaffirm confidentiality, read consentform.2. First, I will ask you some general questions about yourself.Age: What is your marital status? Do you have a partner?YesNoMarriedWidowedDivorcedSeparatedSingleCommon lawWhat was the highest level ofeducation you completed?How many children do youhave?How far do they live from you? (time walking,driving or by bus)Elementary school C IHigh school C ITechnology school C IUniversity C IGraduate school C IHow long have you been living inthis community?What is your current / wasyour previous occupation?Including yourself, what is the total number ofpeople currently living in your household?What is their relation to you? Who is the head ofthe household?Healthinsurance:FONASAISAPREWhat types of income do you receive?Ex: Pension (INP o IPS, AFP), widow's pension,salary, real state, etc.Do you or your spouse receive any support from the state? Ex:Chile Solidario, Programa Puentes, Programa Vínculos, “bonos”,etc?Do you participate in any governmental ormunicipality program?Chile Solidario Programa PuentesPrograma Vínculos “Bono”Otro:3. I would like you to think about the people who are important to you now; important for anyreason. Let’s write a list of those people. Could you tell me who are they? What is the relationship of this person with you? What does this person do for a living? What is his/her age? Does she/he live close to you? (time to reach your place)Date / /Day/ Month/ YearCode2074. Now, I would like you to help me with this ‘map’. You can see that there are 5 concentric circles.The one in the middle is you. Thinking on how important are these people to you now, can youlocate these labels on the map? The closer someone is to you the more important he/she is.5. Now that we have this map, why have people been placed in the way you did?6. How would you label/describe the different circles? or What is the difference betweenrelationships in the different circles?7. Do you feel that this map is complete now? Would you like to add someone else? (If yes, asksame question as in Q3). Also you are free to relocate people or live someone off the map.8. We have talked about people that have helped you in the past, now, thinking about the future,who would you turn for support (Ex: practical stuff, advice, money, information)? Why?208Now, I’m going to name some situations and you have to tell me whether you have received help from someone during the last year (mark allthat applies).HouseholdmembersClosefamilyOtherrelativesNeighbours Friends Organizations OtherCounsel on important issuesCounsel on work, on workopportunities for you or members ofyour familyCare due to health issues (you or aclose relative)Help to fix something at homeHelp with your computerLegal adviceSome who is there for daily talkHelp with transportation to go to aplace far from your home (eg. hospital,grocery shopping)Get information on recreation anddifferent activitiesSupport in case of family problemsHelp you in case of an emergencyWatch your house while you are away209HouseholdmembersClosefamilyOtherrelativesNeighbours Friends Organizations OtherSmall economic support (in Money orgodos, ej. Ch$5,000 –CA$10-, food)Larger economic support Apoyoeconómico más grande (in Money orgoods, eg. Ch$50,000 –CA$100-, arefrigerator)Accompany you to an event(Also ask: Why do you think those people asked you for help in these situations? Could you give any examples?)9. So now let’s look at this question in the reverse. I going to mention different situations and you have to tell me whether YOU have helpedsomeone in the last year (mark all that applies)HouseholdmembersClosefamilyOtherrelativesNeighbours Friends Organizations OtherCounsel on important issuesCounsel on work, on work opportunitiesfor you or members of your familyCare due to health issues (you or a closerelative)Help to fix something at homeHelp with your computerLegal adviceSome who is there for daily talk210HouseholdmembersClosefamilyOtherrelativesNeighbours Friends Organizations OtherHelp with transportation to go to a placefar from your home (ej. hospital, groceryshopping)Get information on recreation anddifferent activitiesSupport in case of family problemsHelp you in case of an emergencyWatch your house while you are awaySmall economic support (in Money orgodos, ej. Ch$5,000 –CA$10-, food)Larger economic support Apoyoeconómico más grande (in Money orgodos, ej. Ch$50,000 –CA$100-, arefrigerato)Accompany you to an event(Also ask: Why do you think those people asked you for help in these situations? Could you give me an example?)21110. Going back to the map, with whom do you feel you have more things in common? What type ofthings?11. Do you have a friend that you consider as a family member?12. Do you have a friendship relation with any of your family members? How is that relationshipdifferent to the one you have with other family members?13. Do you think that your friends give you something different to what your family gives?14. Do you think that your family gives you something that your friends cannot give you?15. Do these people know each other?16. Are they friends of each other?17. Do some of them meet as a group? When? Could you give me an example?18. Are there situation where you would rather to resort to your friends, neighbours or other peoplebefore asking for help to your close family?19. What should a person do for you to change his/her position on the map, to locate him/her furtherfrom the centre or even outside? (Try: what should a person do to you so that you stop trustingon her/him? Are there any differences between family members, friends, acquaintances?)20. Under which circumstances do you believe that children MUST help their parents? Are there anyexceptions?21. Select a of the situations from the previous questions and ask: What did you feel when you received that help? Would you have preferred that somebody else had helped you with that? Why?22. Would you agree with the following statements?StronglydisagreeDisagree Agree StronglyagreeYou can trust most of the peopleYou need to be very careful when dealing withpeopleYou can trust in young people in your communityYou can trust the majority of people in yourcommunityThere are groups in this community I don’t trustYou can trust in young people (in general)You can trust in the police in this communityYou can trust that the major and otherauthorities of the Municipality answer to theneeds of the community21223. Now, thinking about organizations that work on issues related to older people in Chile. Would yousay that you can trust in:Always Fairly often Almost never NeverPublic clinicsThe municipality to get helpwith what I needIn FONASA / IsapreIn the Centre for Older AdultsCaja de CompensaciónNeighbours association24. Have you faced a situation where you have to resort to family or friends to solve a problem thatyou could have solve through an organization?25. Do you think that is hard to meet new friends during old age?26. Finish: Are there anything you would like to add?