The Schwartz Theory of Basic Values and Some Implications for Political Philosophy

The study of basic human values by psychologists is not new. Probably the best-known theory of basic values in psychology is Abraham Maslow’s “hierarchy of needs,” which dates from the early 1940s. But the psychological study of values has been growing, in both volume and empirical quality of research, and philosophers interested in ethics ought to know something about it.

Unfortunately, growing though it may be, the psychological study of values is nevertheless not in a particularly advanced state of development. Accordingly, there are multiple, conflicting theories of human values (and corresponding virtues) in the psychological literature. A sampling that I spent just a few minutes pulling together is: Braithwaite and Law (1985), Cawley, Martin, and Johnson (2000), Crosby, Bitner, and Gill (1990), Feather and Peay (1975), Hofstede (1980), Maloney and Katz (1976), Peterson and Seligman (2004), Rokeach (1973), Schwartz (1994, 2012), and Wicker et al. (1984). My impression is that on the one hand there is considerable loose agreement in the results of these studies, but on the other hand the agreement is indeed loose, and there are significant differences between theories, especially when it comes to the conceptualization of the results.

I myself am not well enough acquainted with this research to comment on these differences. What I want to do in this post is just describe the one of these theories that seems to me to be the most serious, ambitious, well-developed, and well-supported, namely the “Schwartz theory of basic values,” due to Shalom Schwartz (1994, 2012). At the end I will briefly discuss some implications of Schwartz’s theory for political philosophy.

By “values” we refer to beliefs concerning what situations and actions are desirable. However, values for Schwartz are not attitudes toward particular situations or actions, like having a chicken dinner right now or having $20K in my bank account. He restricts the term “value” to broad motivational goals. Schwartz sees values as stable standards by which we evaluate everything else, including the appropriateness of any norms, attitudes, traits, or virtues that may be suggested to us. It is also characteristic of values that some are more important than others. Multiple values are normally implicated in any proposed action, for better or worse, and the all-things-considered evaluation of an action will depend on the relative importance of the competing values it implicates.

Schwartz reasoned that since values are motivational goals, basic human values might be derived by considering the most basic needs of human beings, which he divides into three fundamental categories: our biological needs as individuals, our need to coordinate our actions with others, and the need of groups to survive and flourish. By considering these needs more or less a priori, Schwartz derived the following set of ten basic values. Each basic value is described in terms of its motivational goal. A set of more specific values that express the basic value is given in parentheses after each description.

  1. Benevolence: Preservation and enhancement of the people with whom one is in frequent personal contact [meaning especially family]. (helpful, honest, forgiving, responsible, true friendship, mature love)
  2. Universalism: Understanding, appreciation, tolerance, and protection for the welfare of all people and for nature. (broadminded, social justice, equality, world at peace, world of beauty, unity with nature, wisdom, protecting the environment)
  3. Self-Direction: Independent thought and action—choosing, creating, exploring. (creativity, freedom, choosing own goals, curious, independent)
  4. Security: Safety, harmony, and stability of society, of relationships, and of self. (social order, family security, national security, clean, reciprocation of favors, healthy, sense of belonging)
  5. Conformity: Restraint of actions, inclinations, and impulses likely to upset or harm others and violate expectations or norms. (obedient, self-discipline, politeness, honoring parents and elders)
  6. Hedonism: Pleasure or sensuous gratification for oneself. (pleasure, enjoying life, self-indulgent)
  7. Achievement: Personal success through demonstrating competence according to social standards. (ambitious, successful, capable, influential)
  8. Tradition: Respect, commitment, and acceptance of the customs and ideas that one’s culture or religion provides. (respect for tradition, humble, devout, accepting my portion in life)
  9. Stimulation: Excitement, novelty, and challenge in life. (a varied life, an exciting life, daring)
  10. Power: Social status and prestige, control or dominance over people and resources. (authority, wealth, social power, social recognition, preserving my public image)

Some of the more specific values may seem a little odd (why is reciprocation of favors an expression of security?), but they have been empirically confirmed to express the basic values they were postulated to express. The sort of empirical testing that Schwartz’s theory has undergone is illustrated by the figure below, which shows the result of a type of multidimensional scaling analysis called Simple Space Analysis.

Schwartz values map big 2

The figure was created as follows. A questionnaire was prepared that asked participants to rate the importance to themselves of each of the specific values in the figure on a 9-point scale ranging from 7 to –1, where 7 indicates supreme importance, 0 indicates no importance, and –1 indicates that the participant regards the item as opposed to his own values. The questionnaire was administered to thousands of participants worldwide. For instance, the study reported in Schwartz (1994) included 97 samples in 44 countries from every inhabited continent, for a total of 25,863 participants. Most of the participants in Schwartz (1994) were evenly split between public school teachers and university students, but about 15% were occupationally heterogeneous adults (or, in the case of two samples, teenagers). The ratings were averaged across all participants and then intercorrelated. A Simple Space Analysis then arranged the average ratings in a 2-dimensional space in the way that best represents their intercorrelations as distances, so that points close together in the space are highly positively correlated and points far from each other are highly negatively correlated. The resulting space was then examined to see if the specific values clustered together in groups corresponding to the 10 basic values. Since they did indeed cluster in the predicted way, partition lines were drawn through the space to mark the basic values.

The fit between theory and data observed in the diagram is impressive. This type of study has been replicated many times in the years since Schwartz first presented his theory. The (1994) study is itself a replication and extension of work first presented in 1992. Other instruments have been used to measure basic values besides direct ratings, and specific values than those presented here have been tested. The spaces produced by Simple Space Analysis have been examined by independent raters looking for clusters that might imply basic values other than Schwartz’s ten. But alternative basic values have failed to emerge.

Note that Schwartz’s strategy of postulating a structure of values derived from basic human motivational goals and then testing it empirically differs from other strategies that have been used, such as the lexical strategy of gathering all the value terms to be found in the dictionary and eliminating redundancies and the cross-classification strategy of gathering lists of basic values from multiple traditions and cultures and looking for commonalities. Cawley et al. (2000) used the lexical strategy, which is also the basis of nearly all work in personality psychology. Peterson and Seligman (2004) exemplify the cross-classification strategy. Each strategy has certain merits, obviously, but the Schwartz approach seems to me to have an advantage in being grounded in the functional role of values as motivational goals rather than in the way people (lexical strategy) or intellectuals (cross-classification strategy) happen to talk. The randomness of the lexical strategy in particular seems unfortunate and may have something to do with why it took so many decades for a dominant theory of personality to finally emerge.

Schwartz originally postulated an 11th basic value, spirituality, encompassing specific values such as a spiritual life, meaning in life, inner harmony, and detachment, but it was dropped from the system due to failure to find cross-cultural validation for it. In other words, it didn’t pass empirical muster as a basic, universal human value. Schwartz (1994) speculates that this may be because spirituality is not clearly related to any of the three fundamental categories of basic human needs identified above. Those categories all depend on human functional needs. It may be that spirituality values are not functionally driven.

Notice that happiness is not represented on Schwartz’s list, either of basic or specific values. This is deliberate. Schwartz sees happiness as the result of attaining one’s values.

Notice also that there are specific values on the chart, such as self-respect and moderation, that are not listed along with any basic value in the basic values list. This is because they are associated with more than one basic value (self-respect with both self-direction and achievement, moderation with both tradition and security). They satisfy elements of the motivational goals of more than one basic value. They therefore tend to sit on the borderline between basic values and to be associated more or less closely with their basic values in different empirical studies.

This brings us to another important part of the Schwartz theory, which is that the basic values do not form a loose and unrelated collection but are systematically connected. The connections are expected and predicted by the theory. They have two sources. First, they result from overlap between motivational goals. For example, in an obvious way both power and achievement involve social superiority and esteem. Achievement and hedonism both involve self-centered satisfaction. Hedonism and stimulation both involve desire for affectively pleasant arousal. And so on. I won’t go through all the pie slices in Schwartz’s diagram, since most of the connections are pretty obvious. (The two papers I’ve cited give all the details for anyone who wants them.) Note that conformity and tradition were originally predicted by the theory to be ordinary adjacent pie slices like the others. But that is not the way things worked out empirically, hence their configuration as a split slice.

Second, the basic human motivational goals represent different and sometimes competing or conflicting interests. Thus, the pursuit of one basic value may often conflict with the pursuit of another. For example, the pursuit of personal power or achievement will conflict with the pursuit of universalist values like equality. People who value both must prioritize and often find separate activities by which to pursue each.

Thus, Schwartz’s ten basic values form a continuous, closed circle. Basic values that are adjacent in the circle have overlapping motivational goals and are mutually supporting, whereas basic values on opposite sides of the circle have competing goals and are mutually opposed. Moreover, the circle has a 2-dimensional opponent structure. One dimension contrasts basic values of self-enhancement (achievement and power) with basic values of self-transcendence (universalism and benevolence). The other contrasts basic values of openness to change (self-direction and stimulation) with basic values of conservation (conformity, tradition, and security). Note that hedonism is positively associated with both self-enhancement and openness to change. The diagram below is a schematic version of the one above that makes explicit the two opponent dimensions and the circular structure of adjacency between the basic values.

Schwartz circular model of basic values color

The 2-dimensional opponent structure of the circle is yet another prediction of the theory. So it is additional confirmation of the theory that the predicted dimensions show up in the diagram produced by the Simple Space Analysis and that a 2-dimensional SSA does the best job of modeling the data. (At least, I assume Schwartz tried SSA models with more than two dimensions. He does not explicitly say.)

Note that openness to change and self-enhancement both focus on the personal side of life, while conservation and self-transcendence focus the interests of others and one’s relation to society. So the left side of the diagram represents values with a personal focus and the right side represents values with a social focus. Again, conservation and self-enhancement both express anxiety-driven motivations, to secure oneself against loss, gain power to overcome threats, maintain the current order, and so on. By contrast, openness to change and self-transcendence both express anxiety-free motivations of growth and expansion. So the top of the diagram represents anxiety-free values, and the bottom represents anxiety-based values.

There is one final aspect of the theory that should be mentioned. Although values obviously differ widely in importance between individuals, Schwartz found, remarkably, that when individual ratings of basic values are averaged over all the members of a society, the priority order that results is more or less the same in all societies. The basic values were listed above in their order of cross-cultural priority (highest listed first): benevolence, universalism, self-direction, security, conformity, hedonism, achievement, tradition, stimulation, and power. That is, in most societies benevolence is the most prized basic value, and power is the least. The ranking is curious, and I would be inclined to pay it little attention if it weren’t strongly supported empirically. It is striking that only one personal value (self-direction) is in the top half of the order. This may reflect a universal tendency for socialization processes to emphasize pro-social values. Schwartz (2012) spends some time speculating about why the values are ranked the way they are. For instance, he takes the primacy of benevolence to reflect the central role of the family in a person’s cooperative relations, social connections, and development of all further values. Recall that in Schwartz’s system, benevolence is based on local, personal relationships—this is the key point of difference between benevolence and universality. Thus benevolence ranks highest, and is higher than universality despite universality’s plausible claim to be the pro-social value par excellence, because local and family relations are fundamental and generally trump relations with strangers and out-group members.

To summarize, the Schwartz theory of basic values seeks to identify a core set of basic human values grounded in the motivational goals inherent in (1) our individual, biological needs, (2) our need for smooth coordination and cooperation with others, and (3) the need of groups of people to survive and grow as groups. The system of 10 basic values derived from these goals forms a continuum arranged in a closed circle as in the above diagrams. The space within the circle contains specific values that express various aspects of the basic values that subsume them. Proximity in the space indicates closeness of values in terms of their motivational goals. Proximity to the perimeter indicates strength of commitment to the relevant basic value. Moreover, the basic values themselves are subsumed by four master values arranged on two opponent dimensions: self-enhancement vs. self-transcendence and openness to change vs. conservation. Because of the opponent structure of the dimensions, values on opposite sides of the center of the space will tend to compete with each other for priority. The theory claims that the set of ten basic values and their structural relations are universal. That is, although individuals may differ in their particular value priorities, the basic values and their structural relations are common coin among all humanity in all cultures. The theory has not only intuitive and theoretical plausibility but a very impressive record of empirical support gathered in dozens of studies using multiple measures and employing tens of thousands of participants worldwide.

I promised to conclude by saying something about the implications of all this for political philosophy. Political philosophy commonly arranges political views along a dimension with endpoints designated “left” and “right,” where the defining feature of this dimension is an opponent contrast between equality on the left and hierarchy on the right. If you read a thinker like Allan Bloom, for example, you will get this stark opposition repeatedly (see for instance Bloom 1987). And this dimension admittedly does a powerful job of organizing diverse political positions and explaining many of their similarities and differences. It illuminates many of the differences between American liberals and conservatives, for example, as well as the many social movements in favor of democracy, income equality, racial equality, sexual equality, etc. that became ascendant in the West in the later 18th century and have intensified and spread across the world ever since. But it is irksome to libertarians, who are inclined to think that it treats as primary an issue—equality vs. hierarchy—that does not deserve that status. Libertarians would prefer to focus on an alternative issue, which might be captured by a dimension with endpoints designated “freedom” and “slavery,” or perhaps “individualism” and “collectivism.”

I suggest that the Schwartz theory of basic values can help us to understand this conflict between the libertarian way of analyzing political systems and the standard one. The suggestion, of course, is that the two political dimensions, equality vs. hierarchy and freedom vs. slavery, correspond to the Schwartz dimensions of self-transcendence vs. self-enhancement and openness to change vs. conservation. Concerning the dimension favored by standard political philosophy, equality is the nonpareil specific value of universalism (this is indicated by its position in the first diagram above), and in general the specific values that are grouped under universalism and benevolence (social justice, protect environment, world peace, forgiveness, broadminded, helpful) are suggestive of equalitarian politics. On the other side, the values of power and achievement, which cannot be equal (that is the point of valuing them) suggest a politics of rank. As for the dimension beloved of libertarians, freedom and independence are the premier specific values of self-direction, a basic value whose congruence with a politics of individual liberty couldn’t be more obvious. Other specific values grouped under self-direction and stimulation are among the most celebrated by libertarians: creativity, curious, choosing own goals, varied life, daring, exciting life. At the other end of this dimension, the conservation values of tradition, conformity, and security embody just the sort comfortable obedience and passivity that aligns with a politics that preaches the supremacy of group interests. The person who is at home in this region of the value space values obedience, the sense of belonging, health, social order, humility, self-discipline, moderation, security, and—most strongly, to judge from its position in the diagram—“accepting my portion in life.” Clearly, these are values that encourage political positions that promise safety and good order in the bosom of the group and maintenance of traditions.

Some implications of this analysis are the following. First, libertarians are right to complain that the freedom vs. slavery political dimension is at least as important as the equality vs. hierarchy dimension and that the freedom vs. slavery dimension has been wrongly neglected or ignored by standard political philosophy.

Second, it would be a good idea for partisans of either dimension to drop the habit of reductionism with regard to the other. That is, recognize the other dimension. Both dimensions are real and both are about equally important and illuminating, so do not treat your favored dimension as the only one that really matters.  Furthermore, stop trying to paint all your opponents with a single brush dipped in the color of the opposite end to yours of your favored dimension. The other dimension may be at least as great a source of disagreement. For example, just because someone does not place the same value on freedom that you do does not necessarily mean that his main political impulses are collectivistic. Those who emphasize equality, for example, often do so in part because they see it as essential to individual autonomy. (I believe this was Jean-Jacques Rousseau’s motivation.) They see any sort of draconian collectivistic consequences of the push for equality as incidental and avoidable. Whereas I think a typical libertarian view is to see emphasis on equality as mere cover for a deeper, collectivistic impulse. But that is quite wrong in many cases, if the present analysis is correct.

Third, no political philosophy that wants to have a chance of adequacy can afford to embrace one side of either dimension to the complete exclusion of the other. Equalitarians must make room for the inescapable values of self-enhancement (for details, see “Harrison Bergeron”), and libertarians must make room for the equally inescapable values of security and social order. (And don’t anybody comment to tell me about “spontaneous order.” I know all about it. The point is that not all desirable social order is spontaneous.)

Fourth and last, we should expect there to be no such thing as a pure libertarian or equalitarian (or conservative). Libertarianism stakes out a position on only one dimension. Every libertarian must be expected to have some orientation with respect to the other dimension as well, and so be either a “conservatarian” or “liberaltarian.”And of course, notoriously, this is exactly what we find. The same will be true of liberals and conservatives. Some should really care about freedom, others not. Since the two dimensions seem to be largely orthogonal, extreme devotion to one end of either dimension, freedom vs. slavery, equality vs. hierarchy, should be no help whatever in predicting what a person’s position will be with respect to the other dimension. We must take both dimensions with equal seriousness.

 

WORKS CITED

  • Bloom, Allan. 1987. The Closing of the American Mind. Simon and Schuster.
  • Braithwaite, V. A. and H. G. Law. 1985. “Structure of Human Values: Testing the Adequacy of the Rokeach Value Survey.” Journal of Personality and Social Psychology, 49: 250–263.
  • Cawley, M. J., J. E. Martin, and J. A. Johnson. 2000. “A Virtues Approach to Personality.” Personality and Individual Differences, 28: 997–1013.
  • Crosby, L. A., M. J. Bitner, and J. D. Gill. 1990. Organizational Structure of Values. Journal of Business Research, 20: 123–134.
  • Feather, N. T. and E. R. Peay. 1975. The Structure of Terminal and Instrumental Values: Dimensions and Clusters. Australian Journal of Psychology, 27: 151–164.
  • Hofstede, G. 1980. Culture’s Consequences: International Differences in Work-Related Values. Sage.
  • Maloney, J. and G. M. Katz. 1976. “Value Structures and Orientations to Social Institutions.” Journal of Psychology, 93: 203–211.
  • Peterson, Christopher, and Martin E. P. Seligman. 2004. Character Strengths and Virtues: A Handbook and Classification. Oxford University Press.
  • Rokeach, M. 1973. The Nature of Human Values. Free Press.
  • Schwartz, Shalom H. 1994. “Are There Universal Aspects in the Structure and Contents of Human Values?Journal of Social Issues, 50: 19–45.
  • ———. 2012. “An Overview of the Schwartz Theory of Basic Values.Online Readings in Psychology and Culture, 2(1). http://dx.doi.org/10.9707/2307-0919.1116
  • Wicker, F. W., F. B. Lambert, F. C. Richardson, and J. Kahler. 1984. “Categorical Goal Hierarchies and Classification of Human Motives.” Journal of Personality, 53: 285-305.

10 thoughts on “The Schwartz Theory of Basic Values and Some Implications for Political Philosophy

  1. Pingback: Jury Duty, Moral Horror, Gratitude, and a Promissory Note, All in a Single Post | Policy of Truth

  2. David,

    I’ve finally gotten a chance to set out some of my thought on your post. I actually have several different, somewhat related comments, so what I’ll do is write a separate comment for each, rather than write one long comment for all of the sub-comments. If I wait to do the latter, I’ll never write anything.

    In a previous post, I made a jibe about Jonathan Haidt’s work. It took the form of a joke, but we all know about jokes and their relation to the unconscious–I’ve read only a tiny bit of Haidt, and I’m extremely skeptical about it. Hence the joking attack on him. I have much the same reaction to Schwartz’s work, based on the same second-hand acquaintance with it. I just do not see the relevance of either Haidt or Schwartz to what I think of as ethics or political philosophy. And I say that as a philosopher in a psychology program myself, on the premise that psychology must somehow be relevant. But whether we’re talking Schwartz or dialectical behavior therapy, the relevance relation is either very complex, very indirect, or very obscure.

    (1) Here’s a naive, big-picture objection: Isn’t there just a categorical difference in subject-matter between moral and political philosophy (call it “normative theory”) and what Schwartz is doing? Normative theory is by definition prescriptive. Schwartz’s theory is descriptive. I suppose you can get some sort of normativity out of the theory by saying that the values he’s describing are derived from some antecedent conception of flourishing. But if the antecedent conception is itself descriptive, I don’t see how any part of the theory has normative bite. It simply describes what people happen to think about what they happen to value. (I have a separate objection I’ll state later about Schwartz’s sample. It just does not seem to me to be particularly representative of human beings qua human, so that I’m not even sure it’s properly descriptive in the relevant sense.)

    (2) Here’s an objection about the second, third, and fourth of the implications you infer near the end of the post. I have a worry that all three of these objections attack a straw man, at least with respect to libertarianism. (And I don’t regard myself as a libertarian, so this isn’t special pleading.) Take Rothbardians, Nozickians, and Randians as representative of “libertarian” (-like) positions. As I see it, in none of the three cases do your Schwartzian implications really apply or demand significant revisions.

    (a) Rothbardian libertarians are famous for the view that libertarianism is merely a political position constituted by allegiance to the “non aggression principle.” Their point is that once you accept that one constraint, you’re allowed to espouse any set of ethical norms you like–including strongly egalitarian ones. That’s why there’s such a phenomenon as left-libertarianism: left libertarians are (often enough) Rothbardian libertarians of a strongly egalitarian stripe. But the theoretical point is broader than that observation: the point is that the theory explicitly allows its proponents to do what you’re saying that libertarians typically don’t do. There’s nothing about Schwartz’s theory that demands revision to Rothbardian libertarianism. I don’t think Schwartz’s theory is normatively powerful enough, or even determinate enough, to entail the claim that Rothbardians should discard the non aggression principle. If not, the theory can’t do much to affect the structure or content of Rothbardian libertarianism. (This is not to say that Rothbardian libertarianism makes any sense, by the way, just to say that Schwartz’s theory doesn’t unseat its claims.)

    (b) Nozickian libertarians regard the “non aggression principle” as a side constraint on teleological norms. But roughly the same thing is true of Nozick-type libertarians as is true of Rothbardians. Once you accept the libertarian constraints, you’re free to be as egalitarian (etc.) as you’d like to be. Bleeding Heart Libertarianism seems to me an instance of this view. Granted, we might see Nozick’s later rejection of libertarianism as an instance of what you’d expect of a libertarian whose theorizing underwent a quasi-Schwartzian transformation (Nozick’s discussion of that rejection in The Examined Life is worth looking at it in this regard). But the fact remains: as long as you accept libertarian constraints, you can espouse egalitarian norms, and Schwartz’s findings don’t entail the falsity of libertarian constraints.

    (c) Rand’s view might be thought the most vulnerable to Schwartz’s findings, or at least the most vulnerable to the charge of one-sidedness and/or a rigid failure to accept what’s legitimate in egalitarian norms. Schwartz aside, I do think Rand is systematically blind to the legitimate appeal of egalitarianism, and that her attempts to discuss egalitarianism are invariably inept–either ridiculously hostile (e.g., her critique of Rawls), or reactionary (e.g., her critique of feminism), or inept (e.g., her discussion of racism).* Maybe Schwartz helps us see that better.

    But that said, a Randian could still say this: nothing in Schwartz requires an Objectivist to give up the Objectivist interpretation of individual rights or the “non-initiation of force” principle; since Objectivism has a virtue of benevolence, Schwartz’s findings don’t require either the addition or subtraction of an element of the theory that isn’t already there. At best, Schwartz’s findings might induce Objectivists to rethink their conceptions of one or all of the following: (i) rights and the content of the non-initiation principle itself; (ii) the content or stringency or scope of the virtue of benevolence. But then, it seems to me that you don’t need Schwartz’s findings to have an incentive to rethink any of these things; common sense (and experience with Objectivists) would dictate it just as urgently. So while I’d admit that Schwartz’s findings do some work here, we don’t need Schwartz to do that work, and the big picture objection I stated in (1) circumscribes the work Schwartz does even more tightly.

    *The critique of Rawls is in “An Untitled Letter” in Philosophy: Who Needs It; the critique of feminism is (among other places) in “About a Woman President” in The Voice of Reason; and the discussion of racism is in the essay “Racism” in The Virtue of Selfishness.

    Like

    • Hi Irfan. Thanks for your reply.

      (1) Here’s a naive, big-picture objection: Isn’t there just a categorical difference in subject-matter between moral and political philosophy (call it “normative theory”) and what Schwartz is doing? Normative theory is by definition prescriptive. Schwartz’s theory is descriptive.

      Yes, I completely agree. It would certainly not do to say anything like, “people value X, therefore people ought to value X,” and this doesn’t change just because you add “most people” or “a representative sample of the world population” or even “all people.”

      I don’t think Schwartz means to make any such inference. He is a psychologist trying to map out a structure of natural human values, “natural” in the sense that we are in fact prone to them and that we are prone to them for understandable reasons. And I think he regards his project as important because values are fundamental in determining our decision making and actions. So understanding “basic human values,” if there are such things, helps us understand ourselves. And surely this is right. If it is, then we already have reason to be interested in Schwartz’s findings.

      Like Schwartz, I also do not want to infer from the values we do have to the values we ought to have. As a way of discerning proper values, I am friendly to the Aristotelian ergon argument I outlined in my post on that topic. Namely, if human life is functionally organized, then that very functionality implies standards of better and worse ways of living, just as the functionality of the eye implies standards of better and worse for an eye (by how well it can see). Therefore, I would not say, “Schwartz finds that everybody (more or less) values X, so these are the right values.”

      But this doesn’t make Schwartz’s work irrelevant to the matter of finding proper values (by which I mean, of course, normatively correct basic values). It is relevant in two ways.

      First, Schwartz’s particular procedure draws on the ergon argument I just outlined. As I described in the post on Schwartz, his theory of basic values—both the list of values and the structural relations between values—was originally derived by reflecting on our functional needs. He writes:

      In order to cope with reality in a social context, groups and individuals cognitively transform the necessities inherent in human existence and express them in the language of specific values… Specifically, values represent, in the form of conscious goals, responses to three universal requirements with which all individuals and societies must cope: needs of individuals as biological organisms, requisites of coordinated social interaction, and requirements for the smooth functioning and survival of groups. (1994, 21)

      Thus, one of the intriguing things to me about the Schwartz theory is that it takes this Aristotelian approach as its launch point. Given my own views about the right way to proceed in finding proper basic values, I think Schwartz’s way of going about his project is very promising. In addition, of course, he seems to have made an impressive empirical success of it.

      This leads to the second point, which is that empirical work on actual basic values can hardly be irrelevant to finding proper basic values, can it? At least on the Aristotelian approach I favor, they must be closely related. Values don’t exist to no purpose, on this view, but to organize successful living. Other things equal, better values should be associated with better lives. And “better lives” here obviously means functionally better. Thus a better life is, in some balanced way, healthier, more prosperous, more well-adjusted psychosocially, more productive, and happier. Proper values ought literally to promote the success of individuals and of the species. Thus it ought to be no accident that they are widespread. By and large, successful individuals together with their traits ought to survive and proliferate better than other individuals. The general Aristotelian normative theory predicts that proper basic values ought to be naturally appealing to most people, on pain of saying that the human race is dysfunctional.

      This obviously doesn’t mean that some optimal set of basic values has to be held by everybody. The organization and conduct of our lives is up to us to a very great extent. It therefore requires that we explicitly consider and judge our basic values (as well as many other things). Naturally, this task is enormously difficult and error-prone. Most people will not get far without the help of culture. The influence of culture is both good and bad. It is a great reservoir of wisdom and experience, but it can also be the conduit of some terrible ideas about living and proper values. It’s not hard to think of examples. Still, proper values ought largely to prevail. Living well naturally feels good and reinforces itself. Nothing succeeds like success. So values that promote success will be reinforced. I also think, for the reason given in the preceding paragraph, that proper basic values are probably to some extent innately appealing.

      Therefore, although the match will hardly be perfect, we ought to find that the basic values which an Aristotelian analysis identifies as proper are in fact widespread. If they are, this helps to confirm the analysis; if they aren’t, we have some ‘splaining to do. And even where the match is excellent, we need to address the inevitable discrepancies. Do the discrepancies exist because of cultural distortions or because our a priori analysis was not quite right?

      Empirical work of the Schwartz kind may therefore be thought of as an implementation of Aristotle’s method of endoxa. The endoxa are the reputable opinions about any matter, “which are accepted by everyone or by the majority or by the wise” (Topics, I.i 100b21). A proper inquiry about any matter must account for the endoxa: it should “prove, if possible, the truth of all the endoxa about [the matter in question] or, failing this, of the greater number and the most authoritative” (NE, VII.i 1145b5). In the NE, Aristotle pursues the method of endoxa a great deal more than he spends time trying to derive proper values by reflecting on the functional organization of human life, even though the latter method is his official account of normativity. But the method of endoxa is entirely suitable, if what I have said is right. Since proper basic values promote successful lives, they ought to be widespread; i.e., they ought be endoxa. Schwartz can be viewed as merely taking a social scientific approach to the gathering and analysis of the endoxa.

      (2) Here’s an objection about the second, third, and fourth of the implications you infer near the end of the post. I have a worry that all three of these objections attack a straw man, at least with respect to libertarianism. … Take Rothbardians, Nozickians, and Randians as representative of “libertarian” (-like) positions. As I see it, in none of the three cases do your Schwartzian implications really apply or demand significant revisions.

      [For example,] (a) Rothbardian libertarians are famous for the view that libertarianism is merely a political position constituted by allegiance to the “non aggression principle.” Their point is that once you accept that one constraint, you’re allowed to espouse any set of ethical norms you like–including strongly egalitarian ones. …

      I don’t think I mean to be saying anything quite so strong as that libertarian political philosophy prohibits libertarians from embracing egalitarian norms (or norms of hierarchy), and for just the reasons you say.

      But, for libertarianism, embracing any such norms must be something one does on one’s own, mustn’t it, independent of the political order? For a libertarian, once you have arranged a political order that adheres completely to the non-aggression principle or to individual rights, you’re done. Libertarianism has nothing else to say, except that the political order may not go further. The political order (the government or whatever) may not take steps to protect the family or tradition or economic equality or the environment or security or health or benevolence or genius or innovation. And to the extent that individuals concern themselves with such things, that’s their own affair and no concern of libertarian political philosophy.

      I’m suggesting that this won’t do. The political order ought to promote all the values on Schwartz’s wheel, to one extent or another, not just the one master value Schwartz calls openness to change (centering on freedom). I have no doctrine about how much the political order ought to promote any other values or in what balance. I personally would place more political weight on freedom than any other single value. But I am past thinking that freedom is the be-all-end-all or that once individual freedom is secured, everything else necessarily takes care of itself. Therefore, a complete political philosophy should have regard for all basic human values and both dimensions, not just the one.

      Like

  3. OK, here’s another set of comments, focused this time on Schwartz’s account of the values themselves. I seem to have found his account more problematic than you did. Here are three complaints.

    (1) In introducing the list of values, you say this:

    Schwartz reasoned that since values are motivational goals, basic human values might be derived by considering the most basic needs of human beings, which he divides into three fundamental categories: our biological needs as individuals, our need to coordinate our actions with others, and the need of groups to survive and flourish. By considering these needs more or less a priori, Schwartz derived the following set of ten basic values. Each basic value is described in terms of its motivational goal. A set of more specific values that express the basic value is given in parentheses after each description.

    I haven’t read Schwartz (I’ve merely looked over the 1994 paper), so I don’t know the details of the a priori analysis. But it seems to me that one crucial distinction not made is one between values whose achievement is within the agent’s control and those whose satisfaction depends to large degree on factors outside of the agent’s control. The first will tend to map onto virtues (whether as virtues themselves, or aspects of virtue), the second onto states of affairs external to the agent’s character.

    Given that, it doesn’t make sense to me to represent, say, self-direction and power as slices of a two-dimensional pie chart. Whether or not I’m self-directed is something I control; the amount of power I have is not something I control. By representing the two as basic, Schwartz seems to be implying that both are equally basic objects of pursuit, but they aren’t. Something that is in the agent’s control is more basic an object of pursuit than something outside of the agent’s control. We pursue values external to agency by means of values internal to agency. That seems to me the most basic, foundational fact about meta-ethics, and it has no place within Schwartz’s account. In fact, the account seems to flout the fact.

    (2) You say that some of the specific values may seem a little odd, but that seems to me an understatement. It’s a total mystery to me how any plausible a priori analysis of human needs would Schwartz to the conclusion that conformity or tradition were basic human values (or human values at all). The mystery is compounded by the claim that self-direction is a basic value. Self-direction just seems flatly incompatible with both conformity and (a conformist commitment to) tradition. The fact that there is an impressive fit between theory and data on this point doesn’t seem to me to have any normative implications. It just suggests, descriptively, that people are often committed to incompatible values.

    (3) I realize that Schwartz goes out of his way to make his sample as representative of the worldwide population as possible:

    The questionnaire was administered to thousands of participants worldwide. For instance, the study reported in Schwartz (1994) included 97 samples in 44 countries from every inhabited continent, for a total of 25,863 participants. Most of the participants in Schwartz (1994) were evenly split between public school teachers and university students, but about 15% were occupationally heterogeneous adults (or, in the case of two samples, teenagers). The ratings were averaged across all participants and then intercorrelated.

    But I’m still skeptical that this sample is really representative of human beings as such. For one thing, it’s admittedly lopsided: if I’m understanding you, a full 85% of the sample consists of public school teachers and university students. But what if public school teachers and university students have something in common that they don’t have in common with people outside of that cohort? I suppose that the lopsidedness is supposed to be corrected for by the 15% who are occupationally heterogeneous/teenagers. I have no real grasp of the mathematical procedures involved here; I’d just ask the naive question whether averaging over these cohorts really captures the character of the worldwide population. Can you adequately correct for a lopsided sample by averaging over heterogeneous adults and teenagers? (A literal, not a rhetorical question.)

    The sampling methods can’t by definition capture the views of the unemployed, and probably can’t capture the views of the illiterate–both sizable populations. Nor, for obvious reasons, can it capture the views of generations born prior to the sample–most of the human beings who have ever existed. Interestingly, the sample doesn’t seem to include any Islamic countries (possibly because it might not have been feasible for an Israeli citizen named Shalom Schwartz to do empirical research there: no Muslim country had diplomatic relations with Israel in 1994).

    There’s also no discussion in the 1994 paper of what seem to me obvious issues in translation. Two countries mentioned in the study are the U.S. and India. It really is not clear to me that you can translate Schwartz’s Americanized English without loss into, say, Hindi or vice versa. I speak Urdu, which is a cousin of Hindi, and often have enormous difficulty explaining myself to native speakers, not just through conversational incompetence but due to (something like) semantic incommensurability.

    I don’t mean to be endorsing literal incommensurability in principle; my point is that there are things one can easily say in English for which there are no obvious Urdu equivalents. Sometimes thoughts that can be expressed in two or three English sentences need five or ten minutes of Urdu explanation, at the end of which you can’t be entirely sure that you’ve been fully understood.

    As I mentioned in an earlier post, I encountered huge obstacles in teaching political philosophy to Palestinians, despite having a first-rate translator to help me. Even the simplest passages from Locke or Mill were astonishingly difficult to translate and explain to native Arabic speakers. Issues of translation and commensurability, a staple of philosophical debate from Wittgenstein to Davidson through MacIntyre (and now decades old) don’t seem to have had any significant impact on Schwartz, on psychology, or on the social sciences as such. That strikes me as problematic, and as casting doubt on the external validity of many of these studies.

    I don’t bring all this up merely to pick nits. My point is that if you wanted an account of the basic values characterizing humans qua human, Schwartz’s sample is clearly inadequate. It might be the best we can do given our current empirical and technical resources, but it’s not an account of “human values” as such, despite the title of Schwartz’s paper.

    Like

    • On the particulars of the Schwartz theory, I can’t speak with a ton of confidence. I have myself read just the two papers, which are synoptic treatments of the whole theory and don’t go into the nitty-gritty. But I will say what I can.

      You raise three points.

      First, you distinguish between values that are under our direct control, like self-direction and benevolence, and values that aren’t, like power and security. You say the two types aren’t equally fundamental but are treated as such by Schwartz. This is a problem because only what is under direct control is really basic. Indeed it is the values (or virtues) we can control directly that we employ in our efforts to obtain values we can’t directly control, which are therefore secondary.

      In answer, I would say that although for some purposes the distinction can be quite important, I’m not sure it is important for Schwartz’s. Schwartz is not building a moral theory. I think Schwartz is trying to discover what people consider to be the most important guiding principles of their lives. What are people’s most fundamental motivational goals? What do people want to get or to be to make their lives meaningful, fulfilled, complete, happy, etc.? It seems reasonable that this would include things from different categories. It should include virtues, of course, and many of the specific items are virtues (capable, creative, wisdom, honest, moderate). Others are personal qualities we wouldn’t necessarily call virtues, because they aren’t under our control (intelligent) or aren’t moral qualities (healthy). But there are also material goods (wealth), social goods (social recognition, self-respect), social conditions (freedom, security), relationship goods (friendship), and transcendental goods (environment, unity with nature). Isn’t this about what you’d expect? These are all (more or less) important things that people want and that they guide their lives by.

      Second, you question some of the alleged values on Schwartz’s list, especially conformity and tradition. You doubt whether they are proper values and suggest that they are flatly incompatible with self-direction.

      There are two interesting issues here: (1) the propriety of taking conformity and tradition as values, and (2) the question of compatibility. Taking the first issue first, it may seem strange at first to talk about conformity and tradition being basic values, especially to someone like me (and you too, I think) whose ethical orientation is strongly individualistic. But really, it should not be so strange.

      By “conformity,” Schwartz is talking mainly about self-restraint, about inhibiting disruptive personal inclinations. Specific items are obedient, politeness, self-discipline, honoring parents and elders, loyal, and responsible. By “tradition,” Schwartz means valuing and accepting the customs and traditions of one’s culture. Specific items are respect for tradition, humble, devout, accepting my portion in life, moderate, detachment. Looking at these lists, I think the two together, conformity and tradition, combine three main ideas: self-discipline, subordination of self to group rules and standards, and a sort of unadventurous satisfaction with the familiar and customary.

      Presumably it’s only the latter two that repel you. Self-discipline—responsibility, conscientiousness, integrity, self-control, sôphrosunê—is a classical and eminently respectable virtue whose value to people as individuals is pretty obvious. Perhaps just because it’s so obvious, psychologists have verified it. A particularly striking example is Duckworth and Seligman (2005), who found that a measure of self-discipline outperformed IQ in predicting the academic performance of middle school students. Beating IQ at predicting academic performance is hard to do! So self-discipline, I take it, is a no-brainer.

      What about subordination of self to group rules and standards? This must give a dedicated individualist the heebie-jeebies. But for one thing, I have argued before that individual values aren’t everything. “Man is a social animal.” Being a dyed-in-the-wool individualist for basically my entire life, taking this point seriously has been a long time coming for me, but I’m getting used to it. For groups of people to function smoothly and occasionally even to survive requires a certain amount of putting the group first. What kind of soldiers do you want? Do you want them disciplined, loyal, and obedient to orders or do you want a bunch of contentious “individualists” who only follow the orders they personally understand and approve of and who are looking out for number one? I think it is obvious that soldiers who are willing to function blindly as a unit in critical situations, such as the heat of battle, will fight more effectively than those who aren’t. And this point doesn’t only apply to soldiers. Whenever groups contend against each other, individual qualities that promote the functioning of the group as a cohesive unit will promote its success (and thereby, indirectly, the success of the individuals in the group). This is the sort of a priori analysis that I expect would lead to identifying subordination of self as a basic value.

      But in addition to group-oriented reasons for subordination or obedience to be a value, there is also an individualistic reason, which is that it promotes knowledge acquisition. It is obvious that blind acceptance of authority has its dangers. It is less obvious that there is a good side as well. The vast bulk of the general knowledge a person possesses, as opposed to personal history and life skills, is acquired from other people—from school teachers, parents, books, articles, movies, Wikipedia, Stephen Colbert, Dr. Oz, and other important authority figures. Especially in childhood, I think people naturally take on faith nearly everything they’re told by authorities. It’s a habit we have to break when we grow up, for obvious reasons. Nevertheless, it’s a good thing on the whole. It promotes efficient learning. Most of what the authorities tell you, after all, is mostly right. Yes, even Dr. Oz speaks more truth than falsehood. Acceptance of authority is unromantic, but its benefit never goes away. The art of growing up is to become a critical hearer of authority, not a knee-jerk challenger.

      What about unadventurous satisfaction with the familiar and customary? This must also seem unromantic and even ignoble compared with its opposite values, daring, adventure, excitement, and variety. But the value of the familiar has basic biological roots. In general, anything a person has been exposed to repeatedly will be preferred to a similar item he has encountered rarely. For example, if you encounter a particular unknown face repeatedly, perhaps by seeing it in pictures, and another face only once or twice, and are then asked to say which you prefer, or which gives you more positive feelings, or which seems friendlier or nicer, other things equal you will name the face you have seen more often. This preference for the familiar goes for goes for pretty much anything: people, trees, polygons, words, objects, you name it. It is called the “mere exposure effect,” and it has been replicated across cultures, across species, and with any number of measures. The effect is found even when the exposures are subliminal and never reach consciousness at all. It is one of those rock solid facts of biology/psychology. And it makes good biological sense: what an organism has encountered repeatedly without mishap is more likely, other things equal, to be safe than what it has encountered rarely or never. And safety is a value. Daring and adventure have their place, but daring and adventurous lives have a way of being prematurely terminated. It can be good to take risks, but nobody is (or should be) a risk taker all the time in all respects. The comfortably familiar is also a value. Moreover, there is something in the idea of cultural wisdom and accumulated experience embodied in traditional ways.

      Now to the second issue, concerning the incompatibility of basic values such as self-direction versus conformity and tradition, or again power and achievement versus benevolence and universalism. Again, I think the fact that certain values compete or conflict with each other is a fact of life we have to accept. Consider safety versus daring. I don’t see how the importance of either can be denied. The organism that ignores safety by taking unnecessary risks will have a short life. But on the other hand, the organism that never leaves the safety of a protected area will run out of forage and starve, and will also never discover any better habitat. So the impulse to seek safety and the impulse to explore compete. Some sort of balance must be struck between them, depending on circumstances and on the characteristics and temperament of the organism. Exactly how to perform this reconciliation strikes me as an important topic, and I wish I had something interesting to say about it, but I haven’t. I think the ultimate standard must be the promotion of one’s overall well-being, where this means not only survival but the excellent functioning of one’s capacities as the kind of being one is. More than this, I’m afraid I have not much to say.

      Finally, as to your third point, you question the representativeness of Schwartz’s study samples. You are quite right to be skeptical. Schwartz himself mentions that the study reported in his (1994) “is limited to contemporary literate cultures” (p. 27). To overcome these limitations will take some doing. I don’t know to what extent he has tried to do so since 1994. The fact that the bulk of participants are found in academic contexts—grade school teachers and university students—is also significant. Educated people are not necessarily like uneducated people. However, a great deal more research has been done since 1994. He claims in the 2012 paper to have administered his measures (he has two separate instruments) in 82 countries, to samples of “highly diverse geographic, cultural, linguistic, religious, age, gender, and occupational groups” (p. 12), and I guess we should take him at his word. Presumably he has at least addressed the problem of using mainly university students as subjects, although he provides almost no details in the 2012 paper.

      On the matter of translation, he does describe the procedure (1994, 27). He had his surveys translated into local languages by his partner researchers in each country. The translations were checked by having them translated back into the original language (English or Hebrew) to check for equivalence. Multiple independent back-translations were performed. It seems like he was pretty careful in this regard.

      Like

  4. Pingback: New Blogger: Hendrik Van den Berg | Policy of Truth

  5. Pingback: Happy 2016 | Policy of Truth

  6. Pingback: Morals and the Free Society: 9. Aristotelianism, Part 1—Natural Human Functions Can Be Investigated Scientifically | Policy of Truth

  7. Pingback: SAY – Spinning a Yarn | To say nothing of the cat

  8. Pingback: Spinning A Yarn – SAY – Development Perspectives

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s