Christine L. Fry
Professor Emerita of Anthropology
Loyola University of Chicago
Operational definitions are always difficult. We always fret over issues of validity and reliability. However, when variables involve an evaluation, the challenge becomes even greater. One extreme of the variance is better than the other. Not only do we have to answer the question of what makes for good and bad, but what empirically represents goodness and badness. For human aging the matter would seem to be fairly straight forward. Within the successful aging framework as proposed by Rowe and Kahn (1998), good aging is success and bad aging is failure. But success and failure are not as they would seem. In a world mediated by culture, seldom do we find simple black and white contrasts. Culture both enables us to formulate our hypothesis, but it also can mystify our measurements.
Culture is a phenomenon that lurks within our argumentation. Because we use it to think and interpret the entire research process, we ignore some of its aspects as we try to address our scientific peers. The intimacy of the media pushes the awareness into the background. We never think about breathing air until it isn’t there. Even then we sometimes continue breathing when the oxygen is gone (as in the case of carbon monoxide poisoning).
Our picture becomes more complicated even if we are aware of the formalized thinking and procedures of science. This is especially true when we take those processes into the ordinary human world. The concern for validity is in many respects a leap of faith that we can translate one culture (scientific) into another (ordinary)1. We hope we can, but in doing so we are measuring culture with culture. It is never easy because of a plurality of perspectives. The first is that of the observer (scientific culture). The second is the perspective of the observed (ordinary culture). To further compound the issue, the culture under observation is not monolithic. The knowledge that comprises a culture is self generated. During enculturation and by living, individuals teach themselves that which they need to know to negotiate their social worlds. This knowledge is highly variable across individuals.
Successful Aging as a Cultural Domain
Although notions of successful aging have been around since the times ofCicero(1988 original 44 B.C), how we actually operationalize the concept remains somewhat a mystery. As noted by Depp and Jeste (2006), even in quantitative studies the operational definitions are by no means uniform. Multiple criteria have abounded in the attempts to measure the potential good in old age. Following Rowe and Kahn’s explicit formulation, most studies have employed indicators of health and functionality (both physical and cognitive). In this paper, we turn our attention to a dimension which is not an explicit component in formulations about successful aging. Old age and the successes in old age are clearly cultural, but more often than not culture is assumed as a residual variable in the mixture of variables comprising conceptions about the phenomena. With increasing awareness of culture in our research strategies, we address issues of cultural sensitivity to the knowledge that individuals have about what makes for success (and difficulties) in old age. At issue is the validity of the construct of successful aging for both middle class Americans and Europeans and peoples of diverse ethnicity. In operationalizing variables in a culturally sensitive manner, we need to know how to discover and understand “local knowledge” (Geertz, 1973).
It is well known that respondents do answer the questions in our instruments and that there is a learning curve as they figure out the intellectual tasks required. However, sometimes we have little idea of the rationale behind their responses. The point becomes more salient with variables that call for evaluation such as is the case with life satisfaction, well-being and successful aging. These instruments call for respondents to use their knowledge to make judgments. That knowledge may not be commonly shared between researcher and respondent. Cultural sensitivity is a legitimate issue in operationalizing variables. It is one thing to raise the issue; it is another thing to figure out how to do it.
Even anthropologists, who have long experience in researching cultural diversity, did not immediately formulate a strategy to discover native understandings. A relatively new line of Anthropological research emerged which could be called a paradigm shift and for about two decades was a charismatic movement. Ethnoscience or ethnosemantics is now a long standing research strategy to investigate the knowledge that native peoples have about the world they live in largely through how they classify significant domains that structure understanding. With cross-fertilization from Psychology and Anthropological Linguistics, this paradigm became known as “Cognitive Anthropology” (D’Andrade, 1995).
Culture is a map of the world as one knows it. Knowledge about the world is encoded in a complex interplay between language and culture. A challenge facing an ethnographer is how to tap into this world, especially if it is alien to one’s own set of understandings. The most straightforward approach is to ask and to observe on a long term basis. Since the early 20th century, anthropologists have been committed to fieldwork and participant observation. Cognitive anthropology has narrowed this generalized strategy of measurement with an explicit and formalized measurement theory. The intent of cognitive anthropological strategies is to ask the question of “what is it that I need to know in order to do a specific thing?” The investigator asks to learn what a native knows so as to understand the behavior as a native would. Such behaviors as driving to work (Wallace, 1965) or riding the New York Subway (Fry, 1972) have been looked at as a complex of culturally grammatical rules. In this paper we examine measurement strategies as they have developed in the late 20th century. We incorporate the explicit measurement theory of cognitive anthropology since it is the most well developed approach to the measurement of meaning. These methods are discussed in terms of: a) eliciting frames; b) organizing cognitive domains and, c) analysis of meaning.
An eliciting frame is simply a framework for asking questions. A very necessary requirement is a pool of people who will willingly help out by sharing their knowledge. These individuals are “key informants.” They are not selected at random, but are carefully chosen to represent diverse segments of a community. Obvious criteria include age, gender, occupation, and other major divisions in that community. Also these individuals are usually very articulate and patient given the length and number of interview sessions. Two major frameworks have been used. These are 1) grand tour questions; and, 2) questions requiring judgments.
Grand Tour Questions: In a grand tour framework (Spradley, 1980), the investigator is approaching the setting as if he/she were a child or a stranger encountering the setting for the first time. Grand tour questions are commonly used in qualitative research by asking a respondent to explain themselves. For instance, the Cantril Ladder was developed by Hedley Cantril (1965) to investigate well-being in 11 nations. His eliciting frame is a ladder presented to a respondent with instructions to rate their well-being by placing themselves on a rung. This is followed by questions to explain their reasons for placing themselves where they did and what it would take to put them on the extremes (lowest and highest) of the scale. This technique has been used to investigate well-being and life satisfaction in settings where researchers have reservations about the adaptability of psychometric scales. Also this eliciting frame has proved invaluable in other cross-cultural and cross-ethnic research (Keith, et. al., 1994).
Cognitive Anthropology, formalizes the basic grand tour questions using linguistic theory and semantics. The focus is on categories and defining attributes. Our intent is to elicit descriptive statements about the characteristics of the setting and about the similarities and differences in the categories used. A likely series of question frames for successful aging could be as follows:
- What kinds of men/women are considered to be successful around here?
- How does this differ from men/women who are not successful?
- Are there any older people who are thought of as being successful?
- How do these people differ from older people who are not thought of as successful?
- How do successful older people differ from successful people in other stages of life? How do unsuccessful older people differ from unsuccessful people in other stages of life?
In the cross-cultural Project A.G.E.2, respondents were asked to think of other people they thought of as doing well and of as doing poorly for their age group. We asked about their gender, age and the reasons for doing well or poorly. The resulting answers were electronically managed and coded for content with checks for coding reliability and comparability across the research sites. The results are discussed below within the section entitled “Local Knowledge and Successful Aging.”.
Another framework to reveal the attributes and meaning of a cultural category is “freelisting” (Weller & Romney, 1988). This strategy asks informants to list the names of things or adjectives that compose or describe a category. An example of a freelisting request would be; “Name all the types of beans you can think of.” The results reveal the relevant items or characteristics specific to categories of knowledge. Freelisting has been used with considerable success to explore domains of kinship, plants, animals, and foods in many cultures. It also has been used to investigate prison life (Fleisher & Harrington, 1998), and diversity in cultural domains including emotions in theUnited StatesandMexico(Schrauf & Sanchez, 2008).
For success and successful aging a likely series of freelisting question frames are as follows:
- Name all the characteristics that are associated with success that you can think of.
- Name all the characteristics that are associated with the lack of success that you can think of.
- Name all the characteristics that are associated with successful aging that you can think of.
- Name all the characteristics that are associated with lack of success in aging that you can think of.
Resulting data will reveal the attributes of success which then can be used in other eliciting frames that require some information about the defining features of meaning. We identify these frames as knowledge games because the questions asked required involve game like qualities and manipulation.
Questions Requiring Judgments. A number of game-like eliciting frames ask informants to use their knowledge of the attributes of a cultural domain to complete a task. In this section we discuss three such frames: triad testing, guessing games, and sorting tasks.
Triad testing uses the results of freelisting and similarity/difference questioning. Attributes or descriptors are pooled and then matched in groups of 3. Informants are then asked to use their knowledge of the cultural domain under investigation to select the two items that are most appropriate or most similar for that category. Triad testing has been used successfully in a wide variety of cultural domains across cultures. For aging research Harmon (1998) has used triads to examine the pleasures and work of old age among a hill tribe people inThailand. Triad testing works well with a limited number of attributes. With large numbers the triad pairings increases exponentially. Long triad listings quickly become burdensome to informants. Using Rowe and Kahn’s criteria a likely triad testing frame of successful aging is as follows:
Select two of the following characteristics that best describe successful aging.
- a) disease free; b) good functionality; c) poor cognition
- a) engaged with others; b) good cognition; c) poor functionality
- a) disease; b) engaged with others; c) poor cognition.
Guessing games have great potential as eliciting frames. In formulating a guessing game strategy, the investigator is asking an informant to ask for information with which to make a guess. Thus, there is a degree of role reversal involved with the informant as questioner and investigator as informant. This, obviously, requires that the investigator have sufficient knowledge to provide answers. A likely frame for successful aging is as follows:
- I know an older man/woman. I will give you a number of questions to tell me if this man/woman is aging successfully.
- I know an older man/woman. I will give you a number of questions to tell me if this man/woman is aging unsuccessfully.
- These questions can be formulated in a number of ways to reference people of different ages along with gender and other characteristics.
The object is not to see how fast or correctly an informant can guess success or lack of success in aging, but the questions asked. Also as the informant becomes more habituated to the task, the order of the questions is indicative of the centrality of the attribute to the domain.
Sorting tasks also have a game like quality about them. Here informants are given descriptors and are asked to read (or look at a graphic representation) and to group them into categories. Sorting tasks have an advantage in that a larger number of attributes can be used as compared with triad testing. Because of their flexibility sorting tasks have been used in a wide variety of contexts. In aging research, two sorting tasks were used in Project AGE (Keith, et. al., 1994). First, on the basis of grand tour questions of similarity/difference, attributes were listed on cards. There were given to informants with the instructions to make a person. First they were asked to select the attributes that best described themselves. Then they were asked to think of a real people they knew of different ages (gender and actual age noted) and then to create them using the attributes. Because of the number of attributes, after the attributes were sorted, they were asked to remove the descriptors that were least important. The second sorting task was part of an interview guide administered to a representative sample. These respondents were asked to sort “social persona” into major divisions of the life course (age groups as they saw them). The social persona descriptions were derived directly from the sorting in the “create a person” interviews in order to be culturally appropriate. The age game sorting formed a framework in which to examine age structuring and to ask further questions about the life course.
Organization of Cognitive Domains
A domain implies that there are a finite number of items that make up the domain and that the domain is known by members of a culture. Also cognitive domains have an internal semantic structure placing the items in a meaningful context. Success is a cognitive domain as is aging. Our intent in this section is to examine the organization of the cognitive domain of successful aging. The same logical procedures that are used in theory construction are applied to folk knowledge. In formulating a theory, a very necessary step is to make explicit the conceptual and operational definition of variables. Three basic principles are used which are reflective of the complexity of the structure of cognitive domains. These are indexes, paradigms and, taxonomies. In this section, we use the criteria proposed by Rowe and Kahn to illustrate these structures.
Index: An index is a straightforward scale. Indexes linearly organize a domain running from one extreme (unsuccessful) to another (successful). If we examine the attributes proposed by Rowe and Kahn (p. 39), we see that there are actually four indexes (if we distinguish cognitive from physical functioning). In terms of the logical structure of a multidimensional index, we can create a summative index numerically or consider other organizational principles. The figure below represents an index of successful aging.
Paradigm: Paradigms are a little more complicated than indexes in that the organization takes account of multidimensionality by intersecting the dimensions to define the attributes that make up the category. In the figure below we see the intersection of health status and social engagement. Here successful aging is both good health and engagement with others. Unsuccessful aging is poor health and disengagement from others. An interesting observation is that there are two unlabeled cells in the paradigm. Logically people can be in good health, but disengaged and be in poor health, but engaged.
Inherent in paradigmatic representations of a cultural domain are the same problems of triad testing. These quickly become cumbersome. For simplicity we reduced cognition, function and disease to health. If we used all of the health attributes proposed by Rowe and Kahn we would end up with a four dimensional paradigm with 32 cells. Making things more complicated, if instead of a dichotomy (good/poor) we allow for an intermediate state of average. The resulting paradigm would still be in 4 dimensions, but now we have 72 cells representing intersecting attributes. Obviously, we may well have a situation where it is difficult to see the forest (successful aging) for the trees (number of intersecting domains which could constitute distinct categories).
Taxonomy: Taxonomies are the most complex structuring of a cognitive domain. The items are placed in a branching dendrogram. This is metaphorically like the branching of tree roots. At the top is the most inclusive category in the domain (the trunk). As one moves down the diagram categories become more specific. In the taxonomy that follows the most inclusive item in successful aging is aging. At more specific levels in the diagram we find life course stages and then success/failure. It becomes apparent from this dendrogram that successful aging is distinctive in old age, but there are successes and failures at earlier life stages.
Why is it important to understand the structure of a cognitive domain? It is an important theory construction issue and operational definition issue. By searching for greater congruency between scientific conceptual classification and native classification we potentially can avoid some theoretical pitfalls. For instance, if it is an index of successful aging we are after, is it not wise to realize that operationally, there may be several indexes at work? Paradigms are great for discovering unmarked categories or categories overlooked, which could well improve our theories. Taxonomies, by placing categories in a broader semantic context, lead to different theories. In our example above, theoretically we see different kinds of success across the life course. Here we are led to consider success theory. Alternatively we can pursue the temporal nature of the life course and the conditions of earlier life on late life success. The most important reason why it is necessary to even consider the organization of a cognitive domain is that it is here where our conceptual and operational definitions link up with statistical frameworks to interpret data, proposed models, and to evaluate hypothesis.
The Analysis of Meaning
Analysis always leads investigators to ask, “What do we do with it?” The qualitative data discussed here is most usually in the form of text. Questions have been asked of informants or tasks have been given to informants. The responses have been recorded and almost always transferred into electronic storage. To make this data accessible to an audience, it must be processed through data reduction procedures and interpreted. In this section we discuss content analysis and the statistical frameworks that appropriate for an analysis of meaning.
Content Analysis: Open-ended textual data are managed through a set of procedures known as content analysis. With the use of computers and a variety of software packages there are a number of different strategies which fall under the rubric of content analysis. This can range from what has been called “auto coding” based on word counts to complex coding protocols of large sections of text. Classic content analysis is based on judgments the researcher makes about the text being analyzed. A fixed number of codes are developed and the text is marked for the theme encoded. Although this may sound simple, it is not. There are all sorts of challenges regarding reliabilities and consistencies in how the codes are used. Computer aided management certainly improves coding procedures, but the problem of reliability never vanishes. Another advantage of computerized management of textual data is that it is relatively easy to retrieve in later stages of interpretation. A very real tendency is to reify the codes and forget the variance that has been reduced into a single theme. By asking, “What did they (informants/respondents) mean by that?” our interpretations are potentially much improved. This is especially critical in cross-cultural research.
Content analysis enables some statistical description of the resulting data in terms of how prevalent the different themes are talked about. This usually takes the form of counts and frequencies. One problem is that one textual response may contain several themes. The multiple natures of responses thus may result in percentages that exceed 100. One is not, however, limited only to descriptive statistics. Qualitative data, it must be noted, often is considered to be difficult to reduce to statistical indices. The data usually are not linear. There are statistical procedures, on the other hand, that permit the analysis of meaning and the analysis of the structure within the data. These are factor analysis, cluster analysis and, multidimensional scaling.
All of these procedures require a matrix (a paradigm) of associations between the items in which we are interpreting. The items may be the results of triad testing, sorting tasks, or even the items of a composite scale such as functionality or ADL. Data within the matrix are scores of similarity or difference. Sometimes the associations are correlations between the items. Zero – One data are also used. Here a zero means the items are not placed together or are different. A one means the items are placed together and are similar. More sensitive indicators of similarity/difference are also used. These scores are averaged (mean, median or mode) across respondents and entered into the matrix.
Factor Analysis: Factor analysis statistically isolates groups of items which form a cohesive unit. The first factor isolated is the most interrelated and most likely to contribute to variance in study variables. But other factors are of interest as well depending on statistical loadings. Because of the cohesion of factors, factor analysis has emerged as “consensus analysis” in cultural research (Romney, Batehelder, & Weller, 1986). If one is factoring an index such as a functionality scale, factor analysis separates units of variables revealing an internal structure to the scale and different kinds of functionality (Fry, 2000).
Cluster Analysis: Cluster analysis produces a representation of the relationships between the study items as taxonomy. The dendrogram begins by grouping items that are most similar. As it proceeds less similar items are added to the groupings until it reaches unity where all items are connected. By using the defining features of the items, the statistical clusters become meaningful categories. For instance in interpreting the sorting task involving social persona in Project AGE, age structuring of the life course is made meaningful. Instead of a flat index of chronological age, the interplay of changing roles (work, education, family, organizations) becomes apparent (Keith, et. al., 1994).
Multidimensional Scaling: Multidimensional scaling (MDS) uses a paradigm in which to view graphically the relationships between study items. These statistical packages usually begin with five dimensional paradigms and then through their iterations reduce the dimensions to only one. The end result is a scaling of items into an index along with a goodness of fit indicator (stress) to tell how well the items scale. Interpretation of the dimensions (most often only 2 or 3 are used) involves examination of the paradigms. These are graphically represented as scatter plots of the study items as mapped in the intersecting space (cells) of the paradigm. Again to assign meaning to the dimensions, the attributes or characteristics of the items are used. Using an example from Project AGE (Keith, et. al, 1994), the MDS solution in two dimensions of the index of chronological age reveals underlying dimensions of social engagement and responsibility for others.
Local Knowledge and Successful Aging
In spite of objections to the biomedicalization of successful aging and the cultural specificity of the criteria, there are surprisingly few studies which bring local knowledge center stage (Bowling & Dieppe, 2005). In a national survey of middle aged and older adults inBritain, 1,701 adults were asked, “What do you think are the things associated with successful aging?” The content analysis of the textual responses, resulted in 9 themes pretty much in order of declining frequency: health and functioning; psychological factors, social roles and activities; financial and living arrangements, social relationships; neighborhood and community; work; independence; and, other (Bowling, 2006; p. 129). In another part of the world, Alaskan Natives (26 in total) were interviewed using an explanatory Model approach to establish an indigenous understanding of successful aging and “eldership” (Lewis, 2011). The instrument included such questions as: “What do you think successful aging means? Why do some Elders age well and some do not? What are the signs of an Elder who is aging well? …. Poorly?” Four elements emerged: emotional well-being; community engagement; spirituality; and, physical health. In Project A.G.E. (Fry, et.al, 1997, 2009) the question of aging well and aging poorly was a part of the “Age Game” interview in each of the seven communities (4 neighborhoods in Hong Kong, two communities in the United States, two communities in Ireland, and two communities in Botswana). For all older age groups distinguished in the “Age Game” sorting task, each respondent was asked to identify a person known to them who was a member of that older group and who was doing well and alternatively another person who was having difficulties or doing poorly. After estimating the respective ages and indicating gender, respondents were asked to explain why a person was doing well or doing poorly in old age. From the thematically coded text, four themes emerged: physical health and functioning, material security, family and kinship, and sociality. Each of these themes contributed to both good and difficult aging, although not with equal emphasis across the seven communities.
It would appear that on the basis of these studies that there is considerable congruence between the biomedical definitions of successful aging and local definitions. Even with the addition of material security and spirituality, we find that they have already been considered as research as filled out the framework (Crowther, et. al., 2002). One could well argue that culture is not as “made up” and arbitrary as all the esoteric barriers to understanding would have us believe. When codifying the phenomenal world, local knowledge reflects that world as does scientific knowledge. Cultural knowledge, however, is not that simple. By limiting ourselves to “grand tour” questions and content analysis, we may obscure the meanings informants provided through data reduction and the themes intended to communicate with scientific peers. Either we can pursue ethnosemantic inquiry further or alternatively return to the text and listen to what informants told us.
By way of example, in the Project A.G.E. communities, the four themes have different understandings within each community and interact in different ways to promote or detract from a good old age. In the African communities (Ju/’hoansi and Herero), it is actually difficult to separate health and functionality from material security from family and kinship. With good health one can work and provide food for self and family. As functionality is compromised, it is family as the producing unit that provides food. As one Herero man told his ethnographer, “The good thing about growing old is to be able to command your daughter to milk the cows and bring you food.” American and Irish parents certainly receive food from their children. They rarely would phrase food receiving in terms of the ability to command them to bring it. InIrelandand theUnited States, families are not producing units and usually have separate household budgets. Here material security is stabilized in old age through social security and pensions. Declines in health and functional abilities seldom threaten material security again because of pensions or disability benefits. Also labor saving devices are technological compensations. InAfrica, sociality is almost entirely absent as a reason for a good or bad old age. Our Irish informants saw old age as a time of contentment, reduced responsibilities and the freedom to do what you want to do with friends, family and community. In theUnited States, being vital, active and involved with others and the community are the signs of successful aging. On the other hand, inHong Kong, concerns for harmonious social relations within multigenerational households organized by patrilineal kinship and stem families, foster social qualities of tolerance and open mindedness as indicators of aging well. This argument and documentation are more fully developed in Fry, et. al.(1997, 2009).
Another cultural critique that has been raised in conjunction with the successful aging framework is an ethnocentrism inherent in the value orientation of success (Torres, 2002, 2006). Success and achievement are familiar value orientations of American culture and its capitalistic economic system. The advice offered by Rowe and Kahn, promotes healthy, active, and independent older people which extend the activities of middle age into old age along with abilities to be productive. At the same time if we creditCicerowith one of the earliest formulations of successful aging, the idea is not exclusively an American invention. Cicerowas a citizen ofRomeand was writing nearly 15 centuries before Europeans recognized that there were not only one, but two continents across theOceanSea(theAtlantic). He also wrote some 18 centuries before capitalism emerged as a dominant economic philosophy. Values and value orientations are cultural domains about what is desirable and appropriate in the enforcement of norms about how to achieve the desired. In other words, values and norms are very much a part of the negotiation and manipulation of social life.
An old Ju/hoansi woman illustrates how values and norms work to achieve one’s goals. I was introduced to her by Pat Draper (Project A.G.E. ethnographer among the Ju/hoansi) while on a site visit to the Kalahari. She was sitting in front of a hut adjacent to that of her daughter who appeared to be looking after her with considerable care. She immediately announced to all that could hear that she was Richard Lee’s3 mother and that she was in need of a dress from me. Although her logic was not immediately clear, it actually was flawless from her perspective. Since Richard Lee and I are both anthropologists, we must be kin. Although the claim to have birthed Richard Lee is clearly a fiction, the fact that she in a sense fostered his knowledge of the Ju/hoansi, gave legitimacy to her allegation. By making such an assertion, she was instructing me in the Jo/hoansi values of kinship, filial piety and generosity. In making her demand publicly she reminded everyone that the world could be a very wretched place without these values. She was also hoping for a gift exchange. Another woman, shortly before we departed the Kalahari, placed an Ostrich egg shell necklace around my neck. Earlier she had gifted me her name. Since there are only about 30 names in use, people who share the same name stand in a special relationship to one another including reciprocal gift giving. In turn she received items of clothing and glass beads to use in her craft work. The values of generalized reciprocity place one in a social world and maintain relationships.
Values always reflect a bias simply because they shape behavior toward a goal or desired outcomes. A bias forCicero’s idea of successful aging is that he wrote from an advantaged position in Roman society. He was a senator; a citizen; a member of a wealthy family; and, a scholar with access to knowledge. Gerontologists also speak from an advantaged position in their societies. As academics, medical professionals or social service professionals, gerontologists also have access to knowledge. Not only do we have access to knowledge about aging, we create it. Successful aging is a goal oriented framework to demonstrate how life style choices earlier in life can determine a healthy and vital life in old age. As with all value oriented agendas we not only should ask if the outcomes are desirable, but if they are achievable. Did that old woman in the Kalahari ever get her dress?
Successful Aging: Myth or Science
Some 500 years ago at the beginning of the 16th century, a Spanish explorer led an expedition to the then recently discovered new world in search of fame, fortune and a fountain. Somewhere in the Caribbean (the Bahamas or Puerto Rico), mythical waters were reported to restore youth, vigor, and potency which would make an old man feel better and whole. Ponce de Leon never found his fountain, but he did search for it. We too follow in his footsteps even in the 21st century. Anti aging medicine seeks not a reversal of aging and restoration of youth, but is searching for a prolongation of life. Successful aging is a near relative in that if an individual has all the qualities that place them in the inner circle of the Venn diagram (good physical and cognitive function; absence of disease; and social engagement) one is doing something right. Perhaps one has imbibed the waters? Given advanced age, even the healthiest and most socially engaged individuals leave the inner circle, as they experience difficulties and eventual mortality. In the empirical world success and aging are an oxymoron.
In the biological world there is no such thing as success. For most life forms, especially those that are multiple celled and reproduce sexually (meiosis), mortality is a built in feature of the specie. Only the specie and its genome would appear to have some degree of immortality. But biologically that immortality is not success. It is only the differential reproduction of variation in the genome which gives the carriers of those alleles the chance of survival and reproduction. More often than not, that too fails as extinction is a common outcome of evolution.
Only in the cultural world do we find ideas and ideals about success. For native English speakers, success encodes goal oriented behavior to achieve what individuals set out to accomplish in any number of cultural domains. As our ethnosemantic analysis has revealed when success as it is used in the cultural domain of old age, means something quite different than when applied to other life stages. In spite of very positive goals, given the biological limitations, in the long run everyone fails. Unfortunately, failure implies responsibility either of the ego experiencing difficulties or of the cultural supports engaging that ego. Why construct a theory of success when it is not attainable? The mythical waters may realistically be untroubled aging and pragmatic strategies to deal with problems.
In spite of the oxymoronic qualities of successful aging as a framework, it has had considerable scientific impact within gerontology. First, success as an outcome has brought a positive frame of reference to research that has focused on fairly negative issues. Positive features and outcomes are more rewarding than the difficult sides of life. In embracing the biological, the psychological, the social and the cultural, successful aging has the potential of theoretically integrating an area of research that is plagued with disciplinary differences and fragmentation. Even if success in aging is not attainable and is only a temporary state, scientifically we can devise strategies to extend the duration of that state and ease difficulties that arise along the way. Impossible dreams are the subjects of myth. Science as a path to knowledge through skepticism and empiricism demystifies the world. Although not science, myths are invitations to discovery. In searching for the unattainable we create lines of inquiry that are testable and produce understandings and practical approaches to improve the world in which we live.
Bowling, A. (2006). Lay perceptions of successful aging: Findings from a national survey of middle aged and older adults in Britain. European Journal of Aging. 3:123-136.
Bowling, A.,Dieppe, P. (2005). What is successful aging and who should define it? British Medical Journal. 331:1548-1551.
Cantril, H. (1965). The pattern of human concerns.New Brunswick,N.J.:Rutgers University Press.
Cicero, M. Tullius, & Powell, J. G. F. (1988). Cato maior de senectute.Cambridge [Cambridgeshire]:CambridgeUniversityPress.
Crowther, M. R., Parker, M. W., Achenbaum, W. A., Latimore, W.L., & Koenig, H. G. (2002). Rowe & Kahn’s model of successful aging revisited: Positive spirituality – the forgotten factor. The Gerontologist, 42, 613-620.
D’Andrade, R. (1995) The development of cognitive anthropology.Cambridge: Cambridge University Press.
Depp C. A. Jeste, D. V. (2006). Definitions and predictors of successful aging, a comprehensive review of larger quantitative studies. American Journal of Geriatric Psychiatry. 14: 6 – 20.
Fleisher, M. S. & Harrington, J. A. (1998). Freelisting: Management at a women’s federal prison camp. In De Munck V. C. & Sobo, E. J. (Eds.). Using methods in the field: A practical introduction and casebook. pp. 69-84.Walnut Creek,CA: Altimira Press.
Fry, C. L. (1972). Rules from the Underground: A Cognitive Ethnography of the New YorkSubway System. Western Canadian Journal of Anthropology. Vol. 3, No. 2. pp. 119‑127.
Fry, C. L. (2000). Culture, Age, and Subjective Well-Being: Health, Functionality and the Infrastructure of Eldercare in a Comparative Perspective. Journal of Family Issues. 21 (6) pp. 751-776. Edited byNan E. Johnson and Jacob J. Climo.
Fry, C. L., Dickerson-Putman, Draper, P., Ikels, C., Keith, J., Glascock, P., & Harpending, H. C.. (1997). Culture and the meaning of a good old age. In, J. Sokolovsky (Ed.), The cultural context of aging: Worldwide perspectives. (pp. 99 -124) Westport, CT: Bergin & Garvey.
Fry, C. L., Dickerson-Putman, Draper, P., Ikels, C., Keith, J., Glascock, P., & Harpending, H. C.. (2009). Culture and the meaning of a good old age. In, J. Sokolovsky (Ed.), The cultural context of aging: Worldwide perspectives. http://www.usfsp.edu/~jsokolov/webbook/fry.pdf.
Geertz, C. (1973). The interpretation of cultures. New York: Basic Books.
Harman, R. C. (1998). Triad questionnaires: Old age in Karen and Maya cultures. In De Munck V. C. & Sobo, E. J. (Eds.). Using methods in the field: A practical introduction and casebook. pp. 121-138. Walnut Creek, CA: Altimira Press.
Keith, J., Fry, C. L., Glascock, A. P., Ikels, C., Dickerson-Putman, J., Harpending, H. C., & Draper, P. (1994). The aging experience: Diversity and commonality across cultures. Thousand Acres, CA: Sage.
Lewis, J. P. (2011). Successful aging through the eyes of Alaska Native elders. What it means to be an elder in Bristol Bay, AK. The Gerontologist. 51: 540-549.
Romney, A. K., Weller, S. C., & Batchelder, W, H. (1986). Culture as consensus: A theory of cultural and informant accuracy. American Anthropologist 88: 313 – 338.
Rowe, J. W. & Kahn, R. L. (1998). Successful Aging. New York: Pantheon Books.
Schrauf, R. W, & Sanchez, J. (2008). Using freelisting to identify, assess, and characterize age differences in shared cultural dimensions. Journal of Gerontology Social Sciences, 63B. S 385- 392.
Spradley, J. (1980). Participant observation. New York: Holt, Reinhart and Winston.
Torres, S. (1999). A culturally relevant theoretical framework for the study of successful aging. Aging and Society. 19: 33-51.
Torres, S., 2002, Relational values and ideas regarding ‘successful aging, Journal of Comparative Family Studies, 33(3) 417-430.
Torres, S., (2006). Different ways of understanding the construct of successful aging: Iranian immigrants speak about what aging well means to them. Journal of Cross-cultural Gerontology, 21, 1-13.
Wallace, A. F. C. (1965). Driving to work. In M. Spiro (Ed.). Context and meaning in cultural– anthropology. New York: Free Press.
Weller, S. C. & Romney, A. K. (1988). Systematic Data Collection. Qualitative Research Methods, Vol. 10,Newbury Park,CA: Sage Publications.
1 In Cognitive Anthropology this contrast is often called “etic” and “emic” to distinguish points of view. The “etic” is the observer’s perspective based on disciplinary and scientific training. The “emic” is the native’s perspective based on local knowledge and most likely is based on different assumptions and distinctions than that of the observer. Technically this distinction is derived from linguistics and the differences between “phonetic” and “phonemic.”
2 Some of the discussion of measurement is based on the authors’ experience in a major cross-cultural project (Project AGE) designed to examine the linkages between community contexts and pathways to well-being in old age (Keith et al., 1994). Project AGE selected seven communities around the globe. These are located in Africa (Jo/hoansi and Herero inBotswana);Ireland (Clifden and Blessington);United States (Momence,IL andSwarthmore,PA); and, 4 neighborhoods inHong Kong. Christine L. Fry and Jennie Keith codirected the project with Charlotte Ikels, Anthony Glascock, Jeanette Dickerson-Putman, Henry Harpending and Patricia Draper as PIs. Project AGE was supported by Grant No. AG03110 through the National Institute on Aging.
3 Richard Lee is a well-known anthropologist based at the University of Toronto who has had a long distinguished career of ethnographic research among the Ju/hoansi.