Published in Journal for the Theory of Social Behavior, 2003, 33, pp. 67-94.
Theoretical and Methodological Problems in
Institute for Cultural Research & Education
Humboldt State University
Although cross-cultural psychology has advanced our understanding of cultural aspects of psychology, it is marred by theoretical and methodological flaws. These flaws include misunderstanding cultural issues and the manner in which they bear on psychology; obscuring the relation between biology, culture, and psychology; inadequately defining and measuring cultural factors and psychological phenomena; erroneously analyzing data and drawing faulty conclusions about the cultural character of psychology. This article identifies fundamental theoretical and methodological errors that have appeared in prominent cross-cultural psychological research. Suggestions for overcoming them are then outlined.
Theoretical and Methodological Problems in
Cross-cultural psychology has demonstrated that psychological phenomena are manifested differently in different locales, and it has identified certain cultural factors that foster these diverse manifestations. However, theoretical and methodological limitations have curtailed the progress of cross-cultural psychology. These limitations must be identified and corrected if we are to comprehend the cultural nature, origins, characteristics, formation, and functions of psychological phenomena.
Our present focus is to identify problems in cross-cultural psychological research which impede this comprehension. We consider such a critique to be a constructive step in advancing the science of psychology. We disagree with those who espouse an eclectic acceptance of diverse approaches and who regard critiques as mean-spirited. Miller espouses this position in her statement that "it is important to avoid a position that privileges one particular theoretical approach or mode of understanding as the sole direction to be taken in future research, dismissing other approaches as providing little or no insightÉTheoretical and methodological heterogeneity may be expected to be a permanent feature of cultural approaches to psychology and not merely a reflection of temporary growing pains. Such heterogeneity not only represents a strength of cultural approaches to psychology but constitutes a feature that is required to provide answers to the complex problems motivating work in this field" (Miller, 1997, p. 118).
The problem with eclecticism is that it leaves us bereft of any principles or direction for understanding cultural aspects of psychology. Pluralism forces us to accept any theory and methodology that comes along because we dare not suggest that it lacks insight. Heterogeneity does not encourage a principled, systematic approach that would reduce errors. In fact, it suppresses critical evaluation of errors under the banner of tolerance.
We maintain that scientific disciplines advance through discussion and debate of competing ideas in which some are rejected and others are accepted. In this article we seek to repudiate certain deleterious theories and methods in cross-cultural psychology. We explain how they obscure and overlook important cultural issues that bear on psychology; they misunderstand the manner in which cultural factors shape psychology; they mystify the relation between biology, culture, and psychology; they inadequately define and measure cultural factors and psychological phenomena; they produce faulty analyses of data and erroneous conclusions about the cultural origins, characteristics, and functions of psychological phenomena. While our criticisms are pointed, we do not impugn all cross-cultural psychological research as contaminated by these errors. Nor do we doubt the good intentions of the researchers we critique. We discuss kinds, or categories, of errors which are fundamental and widespread. We are not taking cheap shots at a few unrepresentative minor figures. We select examples from the work of prestigious scholars in leading publications to demonstrate how typical and insidious these errors are. Some of the criticisms we direct at cross-cultural psychology have been directed by others at mainstream psychology (cf. the discussion of operationism in Theory and Psychology, 2001, 11,1, pp. 5-74; Billig, 1996). We utilize these criticisms in a novel manner to address issues concerning culture and psychology. We also broaden and deepen critiques by other scholars which have been leveled at cross-cultural psychology.
Theoretical Problems in Cross-Cultural Psychology
Inadequate theories neglect or misconstrue what culture is, how it bears on psychology, and how it interacts with biological processes.
Flawed Conception of the Interaction between Biological And Cultural Factors
The interaction between biological and cultural influences on psychology is typically presented in two formulations. One approach presents them as a potpourri of diverse factors which are all connected to psychological phenomena. This typically takes the form of a schematic diagram where biological and cultural factors are connected by arrows to psychological phenomena. Berry, et al. (1992, p. 58) present a diagram of variables leading to gender differences in behavior. The variables include sex organs, child rearing, size and weight, sex role ideology, ecological context, and socio-political context. The authors urge that "individual behavior can be understood across cultures only when both cultural and biological features of our species are taken into account" (ibid., p. 13).
In this eclectic approach, diverse factors are listed without any coherent, systematic integration. No discussion illuminates the specific role that each factor plays in psychological phenomena, or the systematic interrelation of culture, biology, and psychology. In Berry, et al., for example, some studies are presented which emphasize cultural variability in psychological processes, while others claim universal psychological phenomena based upon species-wide biological processes. Such a haphazard, fragmented presentation precludes a systematic understanding of the cultural nature, origins, characteristics, formation, and functions of psychological phenomena.
Some cross-cultural psychologists avoid this eclecticism and attempt to integrate biological and cultural factors. This usually takes the form of privileging biological factors over cultural ones. Universal biological mechanisms are deemed to determine essential psychological characteristics while cultural factors are relegated to affecting peripheral aspects. This formulation minimizes cultural aspects of psychology and contributes little to understanding them.
Matsumoto's (2001) piece on culture and emotions articulates this approach. His chapter is heavily oriented toward a universal conception of emotions. Most of the studies cited claim that the feeling quality and the expression of emotions are universal. Thus, humans everywhere are able to recognize a given emotion from its facial expression. According to this universalistic view, emotions are primarily determined by pan-human biological mechanisms. Matsumoto cites Ekman & Levenson's conclusion that distinct autonomic responses exist for each of six emotions (ibid., p. 178).
Matsumoto does acknowledge another orientation in cultural psychology which emphasizes the cultural organization and variability of emotions. However, this position receives considerably less attention than the biological viewpoint -- only 1/4 of the chapter is devoted to the cultural psychological standpoint. In addition, Matsumoto only agrees that culture can affect the manner in which people express and judge/recognize emotional expressions; he claims that the quality, or feeling, or emotions is natural (Matsumoto, 2001, p. 173).
Matsumoto even minimizes the importance of culture in affecting judgments of emotions. He proposes many non-cultural factors to explain differences in the recognition of emotions (ibid., p. 182). He says, for example, that the expression of different emotions contains overlapping elements so that people sometimes judge a given element to represent different emotions. The confusion in judgments is thus due to the natural overlap of expressions, not to any cultural factors. This is like saying that people sometimes judge red as orange because there is some overlap in the wavelengths of red and orange. Matsumoto also proposes that personality biases in social cognition are another source of disagreement ("error") in emotional judgments. This also leaves culture out of the picture.
Matsumoto never explains how biology and culture interact in judging emotions. He suggests a "neurocultural mechanism" which mandates that "judgments of emotions are affected by (a) a facial affect recognition program that is innate and universal, and (b) culture-specific decoding rules that intensify, deintensify, mask, or qualify the perception" (Matsumoto, 2001, p. 185). This formulation raises questions rather than answers them. To what extent do the cultural rules override the innate program? How do they do so (what is the mechanism that triggers the override, how does it suppress the program, doesn't the program resist tampering)? Why do they override it (why doesn't the program exist by itself without cultural alteration)? Is it easy or difficult for cultural rules to override the universal program? Are particular experiences necessary for this to occur? What happens to the program while it is overriden? Matsumoto provides no answers to these questions. He simply states that cultural display rules are invoked in social situations but not when individuals are alone.
Matsumoto allows that culture can intensify-deintensify, mask, or qualify a perception of emotions. These are all quite different from one another. Which of these is more common? Why is it more common? How do cultural decoding rules sometimes intensify, sometimes mask, and sometimes qualify an emotional judgment?
When cross-cultural psychologists consider the relation of biology and culture to psychology they deny a fundamental, systematic role to culture. They typically regard biology as setting basic characteristics of psychology which culture simply moderates. Lonner & Adamopoulos (1997, pp. 76-77) point out the implications of relegating culture to the status of a moderating variable: "When dealing with culture as a moderator variable one provides arguments for those `mainstream' psychologists who have traditionally challenged the necessity of including culture in psychological theory. After all, the essential theoretical variables remain the same, and culture simply adds refinement." This is why "cross-cultural research is normally not very different from so-called mainstream and essentially experimental and reductionistic research" (ibid., p. 55).
Psychological differences among groups are described with little attention to cultural factors that explain them.
Most cross-cultural psychology focuses on describing the psychology of different peoples without attempting to explain the cultural basis of these psychological differences. A typical conclusion is that Liberians recall things in context and have difficulty with "free recall" (remembering things individually, out of context), while Americans readily employ free recall. Descriptions of mnemonic strategies, perception, emotions, mental illness, and modes of conflict resolution are fascinating, however, they are uninformative about how psychological phenomena are cultural. They are silent about what culture is and how it generates psychological differences among people. Culture is construed as a platform on which psychology rests -- which is external to psychology -- rather than as constituting psychology -- i.e., penetrating into psychology. We may say that this approach to cross-cultural psychology studies psychology in culture rather than culture in psychology. The platform model requires little understanding of culture (in general or in particular) because culture is not brought into contact with psychology.
Explanatory cultural factors are misconstrued or misidentified.
When researchers attempt to identify cultural factors which organize psychological phenomena, a number of errors crop up.
The influence of cultural factors on psychological differences is presumed on the basis of faulty logic.
The faulty logic is:
1) Country A & B score differently on a social value.
2) Psychological differences are found between country A & B.
3) Therefore, psychological differences are due to differences in the social value.
For instance when Chinese and Americans have different views of injustice, the difference is attributed to individualism-collectivism (IC) because China and the US score differently on IC. In other words, Chinese are from a collectivistic culture so collectivism explains their psychology. However, China and America differ in many ways besides individualism-collectivism. To put it differently, IC correlates with many attributes such as wealth, educational level, and political system. Consequently, the psychology of individuals from individualistic cultures may be due to any of these other factors. What seems to be a relationship between psychology and IC is spurious; it is really a relationship between psychology and country. IC is a proxy for country and provides no more information than the country's name. To say that Chinese come from a collectivistic culture is to say that they come from China; to say that Americans come from an individualistic culture is to say they come from America. Whatever individualism or collectivism may exist in these two societies has not been differentiated from other factors which also distinguish the two societies (Oyserman, Coon, & Kemmelmeier, 2002, pp. 41-44).
Cultural factors are construed as abstract rather than concrete.
Many of the cultural variables proposed by cross-cultural psychologists are abstract in that they are devoid of concrete content which reflects a specific social system (Lonner & Adamopoulos, 1997, p. 57). Abstraction results when a factor is misconstrued as a discrete variable with a singular, fixed character. (The definition of a variable is a factor that is qualitatively invariant and only varies quantitatively.) In contrast, a cultural factor that interlocks with other factors is modulated by them and possesses a complex, contextualized, concrete quality.
Van de Vijver & Poortinga (2002) laud abstraction in cross-cultural research. They reject "the dogma that the description of the cultural context should be maximized in order to advance our understanding" (ibid., p. 253). Instead, "The description of the cultural context should be minimized as long as this does not make the behavior studied incomprehensible or irrelevantÉIt is only by abstracting from daily contexts and reducing culture to a set of core variables for the construct under study that we can carry out culture-comparative studies" (ibid., p. 253).
The question is, how relevant and comprehensible are abstract variables for illuminating the cultural quality and origins of psychological phenomena? Rogoff (2002) argues that abstract variables are limited because they ignore the concrete cultural quality that factors have by virtue of being integrated within a social system. For example, education has very different qualities depending upon the specific social system of multiple factors in which it resides. Abstracting it as a core variable -- e.g., as transmitting information -- diminishes its cultural significance. Van de Vijver & Poortinga's rationale for abstraction is specious. It is possible to compare the effect of different educational systems on psychology. It is not necessary to reduce education to abstract generalities to make such comparisons.
The inability of abstract variables to capture the rich cultural content of social factors is illustrated in Hofstede's (1980) conception of social values. Hofstede proposed three social values as significant cultural influences on behavior. They are avoidance of uncertainty, collectivism-individualism, and power distance. These three social values continue to be the main cultural factors which cross-cultural psychologists study in relation to psychological phenomena. Consequently, it is important to examine them in detail.
"Avoidance of uncertainty" obscures the myriad ways that people achieve it. One can avoid uncertainty by being outgoing or self-protective; by barricading oneself in a bunker stockpiled with food, forming support groups, eliminating potentially threatening people, pretending to make friends, earning a lot of money, or living simply to prevent a decline in one's living standard. One can seek to avoid economic uncertainty, personal uncertainty, social uncertainty, or intellectual uncertainty. Postulating avoiding uncertainty as a "core variable" stripped of these daily contexts is uninformative about social life and psychology.
Collectivism-individualism is similarly indefinite and uninformative. Collectivism is defined as people emphasizing the interdependence of every human on some collective group, and the priority of group goals over individual goals (Triandis, 2001, p. 36). Collective interactions can be democratic or autocratic. Collectivism can characterize completely different social systems such as feudalism and hunting gathering societies. It can be a real bond, a disingenuous behavior, or a detested obligation that people try to subvert (Fiske, 2002, pp. 82-83). "Collectivism" is uninformed by and uninformative about concrete social life.
Consequently, collectivism explains virtually nothing about peoples' psychology. For instance, Leung & Stephan (2001, p. 393) tell us that "individualistic and collectivistic societies have different views of transgressions. In collectivistic cultures, in-group harmony is regarded as an important goal, and socially deviant behaviors that jeopardize it are likely to be sanctioned." This statement merely says that collectivists value harmonious group interactions. But that is a tautology; it is the definition of collectivism itself. Collectivism is such an impoverished construct that it cannot provide specific information about psychology. It can only be rephrased in different words, which is a tautology.
Power distance is equally abstract and generates tautologies. It is defined as accepting power differences among groups of people. One conclusion from research is that acceptance of power differences lead individuals to accept injustice and abuse from superiors (Leung & Stephan, 2001, p. 391). But accepting power differences means that one accepts the rights of powerful people to dominate powerless people. The research tells us nothing more than what we know from the definition of power distance. Nothing more could be learned because the abstract concept of power distance -- denuded of specific kinds and amounts of power involved and the reasons for accepting power differentials -- is empty (cf. Louch, 1966; Smedslund, 1995 for discussion of the tautological nature of many psychological hypotheses).Researchers are prone to believing that abstract variables have more explanatory power than we have indicated. They claim, for example, that individualism explains the distribution of rewards/resources according to the principle of equity -- i.e., according to the work that one has contributed (Leung & Stephan, 2001, p. 382-383). However, individualism, per se, does not imply this principle. Individualism simply emphasizes personal independence and goals. The equity principle can only be explained by concrete social factors which must be added to individualism. This is revealed in a statement by Leung & Stephan (ibid., p. 382-383): "individualism is related to the preference for the equity norm because equity is compatible with the emphasis on productivity, competition, and self-gain in individualist cultures." Concrete social goals of productivity, competition, and self-gain are necessary to account for equity.
Collectivism also only accrues explanatory power when it is augmented by concrete social factors. Collectivism is said to promote favoring members of one's in-group and disfavoring member's of one's out-group (Leung & Stephan, 2001). However, this psychological/behavioral effect is not the result of collectivism, per se. Many collective societies are equally generous to in-group and out-group individuals. Hunting and gathering tribes distributed food among all the tribespeople without favoring one's own family, or clan. The Naxi people of China extended sexual favors of wives, daughters, and sisters to male visitors passing through the village (Hua, 2001, pp. 22-23, 195). It is only when collectivism includes particular (as yet to be determined) socio-historical factors that the in-group out-group divide occurs (cf. Ratner, 1991, chap. 3 on distinguishing abstract and concrete factors).
In an important statement which is rarely heeded, Triandis acknowledges that individualistic and collectivistic societies are shot through with specific cultural characteristics. "Korean collectivism is not the same as the collectivism of the Israeli kibbutz" (Triandis, 2001, p. 36). Triandis identifies 60 attributes on which collectivist cultures may differ. For instance, East Asian collectivist cultures disapprove of arguing within the in-group, while this is perfectly acceptable to Mediterranean collectivist cultures. This means that focusing on individualism-collectivism, per se, ignores at least 60 important cultural issues.
Triandis further emphasizes that no culture is monolithically individualistic or collectivistic; all cultures are mixes of both. "A culture should not be characterized as individualist or collectivist. That kind of characterization is simplisticÉGermany is more individualistic than Hong Kong, but even in Germany, people select collectivist cognitions 37% of the time and in Hong Kong they select individualist cognitions 45% of the time" (Triandis, 2001, p. 40).
Despite their emptiness, abstract constructs are widely employed in cross-cultural and mainstream psychology alike. Moral and cognitive development are construed in terms of content-free stages (cf. Ratner, 2002, chap. 6); communication is defined as taking turns, offering advice, solving problems, asking questions, managing conflict and stress -- without regard for the content of what is expressed or the social status or the interlocutors; schizophrenia is defined as flat affect and delusions -- without regard for the content of the symptoms or the specific experiences that provoke them; self esteem is defined as valuing oneself -- without reference to what kind of self is involved and the particular ways it is valued. Psychologists invoke abstract constructs because they are hoping to discover universal, fixed psychological characteristics. Since psychological characteristics vary on the concrete level, they only exist as universal on the abstract level. However, this level is uninformative about the cultural organization of psychology.
Particular cultural factors are misconceived.
Certain cross-cultural psychologists seek to identify psychological effects of concrete cultural factors rather than abstract variables. This well-intentioned attempt frequently fails because psychologists are insufficiently informed about sociology and history. Extensive, sophisticated knowledge of these areas is necessary if one is to understand a culture and its effects on psychology. Yet psychologists receive little training in the social sciences, humanities, or philosophy because psychology has been institutionally segregated from them and ideologically oriented toward intra-individual processes. Thus, psychologists' ignorance is not simply due to the vastness and complexity of cultural information which might escape the most diligent researcher. It is due to an entrenched blindness to crucial social issues. This leads to misconstruing cultural factors and their psychological effects.
This problem is found in Peng's cultural analysis of inferential reasoning (Peng & Nisbett, 1999; Peng, Ames, & Knowles, 2001; Ji, Peng, & Nisbett, 2000; Nisbett, Peng, Choi, & Norenzayan, 2001). The analysis traces psychological differences in oriental and Western reasoning to historical differences in social practices and values of Chinese and Greeks (Nisbett, et al. 2001, pp. 293-295, 303). Briefly summarized, Peng and his colleagues argue that in ancient times, in the orient:
A Social System of:
Many social roles,
A Cultural Epistemology of:
A Social Psychology of:
In Ancient Greece:
A Social Structure of:Few social roles
Hunting, fishing, herding, trading
Decentralized, individualized organization
| A Cultural
A Psychology of:
The authors contend that this model persists today: "There is substantial evidence that the social psychological differences characteristic of ancient China and Greece do in fact persist" (ibid., p. 295). Therefore, contemporary Western fragmented, linear reasoning is rooted in the very birth of Western civilization, extending back to philosophy and social conditions thousands of years ago. Contemporary holistic oriental reasoning is similarly rooted in philosophy and social conditions thousands of years old.
This model is erroneous and the cultural analysis of inferential reasoning collapses. In the first place, modern Western societies have little in common with ancient Greece. Modern corporations, factories, consumerism, politics, technology and artifacts, religion, transportation, entertainment, recreation, family structure, and even romantic love between men and women have no parallels in ancient Greece. In addition, Western societies are not decentralized and simple, nor do they have small states. Modern Western societies are at least as complex as China, with states that are as large as any in the orient. Ancient social factors cannot account for Western thinking because they no longer exist.
Peng's account of Western philosophy is also inaccurate and therefore does not explain contemporary inferential reasoning. Peng claims that Western philosophy since ancient Greece has essentially been atomistic. Even dialectical thinking that arose in Western Europe in the 19th century was not true dialectics for it was dominated by linear, formal logic. In Peng's words, "Western thought (including Western dialectical thought)Érests in substantial part on Aristotelian logic which emphasizesÉthe law of identity, the law of noncontradiction, and the law of the excluded middle" (Peng & Nisbett, 1999, p. 744, emphasis added). Western dialectical philosophy and psychology "regard contradiction as a temporary state that will be replaced by integrated thoughts -- using reasoning that is linear, logical, and moves in one direction -- from a contradiction to a synthesisÉThey are still constrained by the laws of formal logic, which do not tolerate literal contradiction" (ibid., pp. 742-743).
Peng is misinformed about Greek philosophy and modern dialectics. Almost any book on modern dialectics emphasizes its difference from formal logic. Modern Western dialecticians such as Hegel, Marx, Marcuse regarded stable integration as a temporary state that is continually undone by conflicting contradictions. Peng, himself, acknowledges this in a statement that unwittingly contradicts his description of Western dialectics: "The dialectics of Marx and Engels can be characterized as aggressive or assertive in that there is constant negation" (Peng & Nisbett, 1999, p. 742).
Western dialectical thinking drew on Greek philosophy which was organic and dialectical. Plato was a great dialectical thinker. And Aristotle espoused dialectical logic in addition to formal logic. His understanding of organic unity and interrelationships is thorough and shows up in his metaphysics, politics, biology, and aesthetics. Although this part of his work has been overlooked in favor of his formal logic, Marx & Engels acknowledged Aristotle's dialectics as a precursor to their own dialectical philosophy. Hegel scholar Findlay similarly says that, "Hegel recognizes the presence of his Dialectic in the ancient modes of argument that went by the same name" (Findlay, 1964, p. 64). A whole literature exists detailing Hegel's debt to Greek dialectical thought and the organic society that spawned it. Hegel and Greek Thought by Gray (1941) states that "Hegel set his ideal of a social ethos developed from the Greek polis" (ibid., p. 53); "Impressed as Hegel was by this Platonic notion of the organic relation of individual and societyÉ" (ibid., p. 64). In fact, Gray goes on to say that Hegel criticized Greek thought and society as too organic because it excluded individual freedom. Triandis (2001, p. 37) states that Greece was a collectivistic culture well into the 20th century.
It is true that Democritus formulated an atomic philosophy in the fifth century B.C. However it was not widely adopted and did not characterize ancient thought as a whole. In fact Plato and Aristotle rejected Democritus' atomic theory as a one-sided account of reality (Van Melsen, 1973, p. 127). Democritus's doctrine "had only a few adherents between the fifth century B.C. and the seventeenth century" (ibid., p. 130).
Peng's notion that: "Most fundamentally, the Greeks, unlike the Chinese, were inclined to see the world as a collection of discrete objects" (Nisbett, et al., 2001, p. 293) is false. So is his characterization of Western philosophy as thoroughly atomistic and anti-dialectical. Organicism-holism was a strong philosophical movement from the 18th to the 20th centuries in Germany. One manifestation of it was Gestalt psychology (and systems theory, in general) which emphasized interrelationships of elements and group processes rather than discrete individual phenomena.
Peng's identification of dialectics with oriental philosophy is confused as well. The mistake stems from a fundamental misconception about what dialectics is. Peng construes dialectics as a relativistic, pluralistic viewpoint that seeks compromise among diverse perspectives. "The dialectical resolution of social contradiction is encouraged by an aspect of Chinese culture, namely the doctrine of the mean, which emphasizes moderation, sincerity, and most importantly a `reasonable' middle-of-the-road approach" (Peng & Nisbett, 1999, p. 746). Dialectics is "a recognition of contradiction and of the need for multiple perspectives, and a search for the `Middle Way' between opposing propositions" (Nisbett, et al., 2001, p. 293).
Peng's conception of dialectics as seeking a middle ground between opposing positions contradicts the established, scholarly meaning of dialectics. Dialectics does not resolve contradictions by accepting contradictory tendencies or compromising among them. Dialectical thinking understands resolution to be a transformation of the elements into a new emergent product. Thus, Marx envisioned the working class taking power and constructing a novel, classless society. Marx never envisioned a compromise or middle way between capitalists and the working classes. Hegel employed that same dialectical reasoning to resolve philosophical conflicts. In his Phenomenology of Spirit he systematically exposes the errors of numerous positions, and he proposes an entirely novel metaphysics that incorporates elements of the other positions in completely new ways.
What Peng calls dialectical thinking is actually an amalgam of traditional Chinese philosophies such as Confucianism, Buddhism, and Taoism. These recognized a unity of contradictory elements which balance each other. The 29th motto of Taoism states this:
"The universe is sacred. You cannot improve it. If you try to change it, you will ruin it. So sometimes things are ahead and sometimes they are behind; sometimes breathing is hard, sometimes it comes easily; sometimes one is up and sometimes down."
This motto advocates accepting and balancing diverse aspects of existing factors. Creating novel phenomena is out of the question.
Confucianism was similarly a conservative philosophy which sought to mitigate conflict through compromise that preserved the status quo. It is fundamentally different from dialectics.
In his philosophical essay "On Contradiction," Mao Tse-tung explained that most of Chinese philosophy was "metaphysical," static, and undialectical: "In China there was the metaphysical thinking exemplified in the saying `Heaven changeth not, likewise the Tao changeth not,' and it was supported by the decadent feudal ruling classes for a long time" (Mao, 1937, p. 313). Dialectical thinking emphasizes inner contradictions which produce change. It was invented by Hegel, Marx, Engels, and Lenin: "Hegel made most important contributions to dialectics, but his dialectics was idealist. It was not until Marx and Engels, the great protagonists of the proletarian movement, hadÉcritically absorbed the rational elements of Hegelian dialectics and created the great theory of dialectical and historical materialism that an unprecedented revolution occurred in the history of human knowledge" (ibid., p. 315). Mao's words show that Peng is confused in claiming that dialectical thinking was endemic to Chinese philosophy and was either foreign to, or deformed in, Western thinking. Mao explains how systematic dialectical thinking achieved its highest form in Western philosophy, motivated by the class struggle of the industrial proletariat against the capitalist class.
Even if Peng's terminology is corrected and dialectics is rephrased as traditional Chinese philosophy, it is dubious that the latter is a prevalent cultural outlook in contemporary China which could foster holistic inferential reasoning. Peng & Nisbett (1999, p. 745) attempted to prove that "dialectical" (i.e., Confucian) metaphysics is common in Chinese society and accounts for the oriental style of inferential reasoning. Yet when the authors analyzed sourcebooks of Chinese proverbs for "dialectical thinking" (i.e., Confucianism), they found that only 12% of Chinese proverbs are "dialectical" while 88% are not. If proverbs influence thinking, then Chinese reasoning should be nondialectical more than dialectical! Dialectical Chinese reasoning could not stem from the meager presence of "dialectical" proverbs in the culture.
The assertion that dialectical (i.e., Confucian) thinking is fostered by contemporary harmonious practices in the orient is similarly dubious. Peng, et al. claim that "In daily life, East Asians strive to maintain harmony" (Nisbett, et al. 2001, p. 304) -- i.e., seek to resolve conflict through compromise -- and this fosters "dialectical thinking" that emphasizes a middle ground over extremes. In actuality, China has been marred for well over half a century by bitter political and ideological struggle and brutal suppression of beliefs that contradict official, hegemonic doctrine. The Cultural Revolution is the most obvious example. The entire field of psychology was banned throughout China following the 1949 revolution because it was deemed bourgeois. Books, news, entertainment, internet sites, religious groups, political parties, and independent trade unions continue to be banned in China when they deviate from official doctrine. Where is the effort to maintain moderation, harmony, and compromise among differences in these cases? Contemporary personal and social relationships are also far less harmonious and compromising than Peng (Peng & Nisbett, 1999, p. 746) indicates. Jealousy, vindictiveness, competition, arrogance, suspicion, secrecy, duplicity, egoism, back-stabbing, and acrimonious criticism of others are commonplace in Chinese personal relations. The public domain is rife with corruption and cheating that are motivated by a desire to enhance personal wealth and power.
The social practices, conditions, or concepts which Peng and his associates Invoke to explain cross-cultural differences in inferential reasoning are fictitious. (We shall see below that Peng's empirical research claiming national differences in reasoning is invalid as well.) Western civilization is not a continuous pattern of social conditions and ideology which promote analytical, fragmental, linear thinking. Holistic-organic thinking was characteristic of Ancient Greece and it was common throughout the Enlightenment and into the 20th century. Peng's attempt to deny dialectical thinking in the West is pure sophistry. Equally apocryphal is his claim that (genuine) dialectical philosophy was popular in China. Confucianism and Taoism were popular however they have nothing in common with dialectics. Moreover, Peng's own research demonstrates that they are not popular now and are not a source of holistic inferential reasoning. It is equally fanciful to claim that contemporary Chinese social conditions and practices promote moderate, holistic thinking.
A deeper understanding of history, philosophy, sociology, and politics is necessary to accurately identify cultural factors that promote particular forms of inferential reasoning.
Methodological Problems in Cross-Cultural Psychology
One of the defining features of cross-cultural psychology is its methodology. Cross-cultural psychologists accept positivistic methodology which their mainstream colleagues endorse. (In contrast, cultural psychologists come from a more humanistic methodological tradition grounded in hermeneutic philosophy outside mainstream psychology.) This methodology leads to researching cultural-psychological phenomena in ways that are amenable to particular principles of measurement and analysis. Unfortunately, positivistic principles of measurement and analysis often lead to obscuring the cultural origins and characteristics of psychological phenomena (Ratner, 1997, chap. 1).
Many testing conditions and instruments are ecologically invalid.
Many cross-cultural psychologists have recognized that test materials and conditions must simulate peoples' social environments if research conclusions are to accurately reflect, and generalize to, their real psychology. Despite this caution, researchers often employ artificial, unfamiliar materials. The reason is that these materials are readily controlled and calibrated. Positivist principles of measurement and analysis demand simple, unambiguous, manipulable, quantifiable stimuli which elicit simple, quantitative responses. Artificial testing conditions and instruments fit these criteria more than natural circumstances do.
An ecologically invalid measure was used by Ji, et al. (2000) in a study on holistic vs. linear perception. They hypothesized that Chinese would be more sensitive to environmental relationships than Americans are. They tested this by presenting Chinese and American subjects with schematic items on a computer screen (e.g., a light bulb and a coin). Objects appeared together in varying frequencies. After several trials, subjects were asked to estimate the frequency of pairings which had occurred. The estimate was the operational definition of paying attention to environmental relationships. Mean differences in the estimates led Ji to conclude that "East Asians are more attentive to relationships in the environment than Americans" (ibid., p. 952).
This conclusion is unwarranted because Ji's material does not represent sensitivity to environmental relationships. Awareness of relationships is awareness of the real relationship among significant objects -- how they resemble each other, affect each other, depend on each other. Ji et al. did not investigate these issues. They used meaningless, deformed stimuli that appeared momentarily on a computer screen. "All the figures were schematic to ensure that there was little cultural-specific symbolic meaning" (ibid, p. 945). Employing culturally meaningless stimuli to draw conclusions about the cultural character of meaningful psychological processes would seem to be oxymoronic.
Ji's artificial conclusions are contradicted by everyday examples. In everyday life Americans are very aware of contextual relationships. American teenagers are obsessed with manifested or anticipated reactions of their peers. Food, clothing, music, entertainment, sexual behavior, and even friends are selected with an eye to how they will affect social relations with peers. American investors who are deciding which stock to buy similarly consider contextual factors such as interest rates, political stability, and unemployment rates. American business executives try to control a broad range of contextual events such as government policies and candidates for political office. Every American school teacher knows that students' academic performance is affected by influences outside the classroom -- the home, the media, and the job market. Research on social referencing and social comparison proves that American psychological phenomena are based upon on contextual social cues. Ecologically valid tests and responses would have indicated such sensitivity.
Everyday life invalidate one of Peng, et al.'s (2001, p. 253) conclusions. The authors cite their own unpublished work (!) which supposedly documents that Americans value what people say as more important than what they don't say while Chinese show the reverse preference. However, in everyday life Americans have a deep concern for hidden motives. Freud's extensive emphasis on unconscious processes is a staple of Western thinking. And what American does not wonder whether other peoples' statements truthfully represent their thinking?
Measures of cultural factors use incomplete or ambiguous scale items. Ensuing conclusions about cultural factors are therefore unwarranted.
This is the case with some of the most widely used cultural constructs such as individualism-collectivism (IC), power distance (PD), and avoidance of uncertainty. It is rarely noted that PD is identified from three questionnaire items, avoidance of uncertainty is also identified from three questionnaire items, and IC is identified from six questions! In other words, an entire society is characterized as (more or less) individualistic on the basis of how individuals respond to a few questions. Other operational measures of IC have been employed, however most studies rely on Hofstede's scales. To demonstrate how impoverished this basis is we shall present the questionnaire items which operationally define two of the factors.
Avoidance of uncertainty (Hofstede, 1980, p. 161):
1) How often do you feel nervous or tense at work? (Always-never on a 5-point Likert scale.)
2) Company rules should not be broken -- even when the employee thinks it is in the company's best interests. (Agreement on a 5-point Likert scale.)
3) How long do you think you will continue working for this company? (2 years, 2-5 years, more than 5 years, until I retire.)
Individualism-collectivism (Hofstede, 1980, p. 220-221): how important (very important-not important" on a 5-point Likert scale) is it to you to
1) Have challenging work to do -- work from which you can get a personal sense of accomplishment.
2) Have considerable freedom to adopt your own approach to the job.
3) Have a job which leaves you sufficient time for your personal or family life.
4) Fully use you skills and abilities on the job.
5) Have good physical working conditions.
6) Have training opportunities to improve your skills or to learn new skills.
(The first three items operationally define individualism, the last three define collectivism.)
The reader will note that all these questions concern attitudes toward work. Yet Hofstede presumed them to measure general social values. A few questions concerning work are clearly insufficient to ascertain whether a country endorses broad values, such as avoidance of uncertainty, which encompass many aspects of life.
Another weakness is that Hofstede's names for the values (factors) are not warranted by the questionnaire items and responses. The three items that Hofstede construes as defining avoiding uncertainty have no ostensible relation to that value. One may feel nervous at work for an infinite number of reasons. There are no grounds for assuming that an affirmative response indicates an attempt at avoiding uncertainty. Similarly, one may plan to work at a job for a long time for many reasons (e.g., he likes it), not necessarily to avoid uncertainty. The same is true for affirming that company rules should not be broken. Labeling responses as "avoiding uncertainty" is entirely arbitrary and subjective.
The same holds for presuming that the first three items on the IC questionnaire measure "actor's independence from the organization" (ibid., pp. 220-221). Developing a personal sense of accomplishment, adopting your own approach to the job, and having time for your personal or family life do not imply that you are individualistic in the sense of being concerned with yourself independently of the organization. You might desire a personal sense of accomplishment by contributing to the company. You may desire to be creative in order to contribute to the company. And a desire for personal and family time may complement your positive attitude toward the company. It does not imply any inclination toward self-absorption or rejection of the group. Hofstede assumes that any reference to self or creativity implies insulation from the group, however the questions do not necessarily imply this and the subjects may not have meant it.
Hofstede is similarly arbitrary in assuming that the last three IC items "stress things the organization should do for the individual: provide him or her with training, with working conditions, allow him or her to use his or her skills" (ibid., p. 221). The questions simply ask how important these issues are to the subject; they do not imply that the company should provide them or that the subject defines herself as dependent on the group.
Hofstede's list of social values is thus dubious. The few, fragmentary questions do not express broad social values. Nor do they express the names and meanings that Hofstede attributes to them. Other operational definitions of IC -- there are 27 of them -- are plagued with the same problems.
Operational definitions of cultural and psychological phenomena employ truncated, fragmentary responses which cannot represent the phenomena. Ensuing conclusions are therefore unwarranted.
The positivistic orientation in cross-cultural research regards psychological phenomena to be transparent in simple, overt forms of behavior. Thus, mere agreement with a questionnaire item is deemed to express a definite social value, as we have seen in Hofstede's work. In other words, the social value "avoiding uncertainty," or "individualism" is transparent in the behavioral response "strongly agree." There is no need to ask subjects what they mean by agreeing with a statement. The social value is assumed to be evident.
Triandis (2001, p. 40) adopted this methodological perspective in researching IC. He asked subjects to complete the phrase "I amÉ" When the response was "father," Triandis concluded that this response is a collective form of identity because father is a social role/relationship. Conversely, when a subject wrote "kind," or "play tennis," this response is an individualistic identity because it describes a personal state or quality. Based on this assumption, Triandis (2001, p. 40) concluded that Chinese students have a more collective self (they averaged more social responses) while University of Illinois American students have a more individualistic self (they averaged more individual responses).
The assumption that psychological phenomena are transparent in, and reducible to, a singular, overt response is false. Any response can indicate virtually any phenomenon, and any phenomenon can be expressed in virtually any response. Consequently, a singular, overt response does not convey which psychological phenomenon is operative. Ironically, the attempt at enhancing objectivity by restricting responses leads to its opposite -- making arbitrary, subjective interpretations about what the responses mean (cf. Ratner, 1997, chap 1, 2).
In the case of Hofstede's research we saw that planning to stay on a job can express many psychological reasons and the restricted response prevents knowing which reason, or meaning, the subject has in mind. Similarly, when a man says that he is a father, he may have any number of values in mind. In order to know that he sees himself as a social person, we must know what being a father means to the man. Is it a burdensome responsibility which he dreads and avoids, or does he relish the relationship and frequently interacts with his children? If the former, then the man would be deemed to possess an individualistic self-concept, not a social one. Operationalizing a psychological phenomenon as a truncated, overt response with fixed meaning obscures its variable psychological and cultural quality and makes any conclusion about this quality arbitrary and speculative.
This can be seen in Peng & Nisbett's (1999, pp. 748-750) study #5. They presented statements to Chinese and American students and asked them to rate the plausibility of each statement on a 7-point scale. Some subjects read a single statement and judged its plausibility alone. Other subjects read that statement along with another that contradicted it, and then judged its plausibility. The authors compared the plausibility ratings of each statement alone and in the pair. One statement read, "A social psychologist studied young adults and asserted that those who feel close to their families have more satisfying social relationships." The contradictory statement read, "A developmental psychologist studied adolescent children and asserted that those children who were less dependent on their parents and had weaker family ties were generally more mature."
The authors interpret the results as indicating that when presented with contradictory information, Americans construe one as right and the other as wrong. This follows the Western epistemology of non-contradiction, pursuit of a single truth and consistency. Chinese subjects dialectically believe that both sides of a contradiction might be right and the truth lies in the middle.
However, the restricted response format makes any such interpretation speculative. The subjects were only allowed to rate plausibility on a 7-point scale. From this limited information it is impossible to infer anything about their psychology. There is absolutely no indication that Americans construed statements in the pair as right and wrong, or that they pursued a single truth or consistency. There is no indication that Chinese believed that both statements may be true and that truth lies in the middle. We don't even know whether the subjects regarded the paired statements as contradictory. They may not have regarded them as contradictory and may not have employed strategies for dealing with contradictions. Peng may have invented conclusions about psychological processes that were not even at work.
Cross-cultural psychologists believe that their proposed meanings and labels are validated through correlations with other data whose psychological significance is clear. For instance, Hofstede (1980, p. 224) argues that IC has construct validity because in countries which score high on individualism subjects tend to endorse more frequently the statement "A larger corporation is generally not a more desirable place to work than a small company." Hofstede believes that endorsing this item "confirms the lesser dependency of the individual on the company in more individualist countries." The problem is that the validating data are plagued by the same ambivalence as the original data. One could endorse Hofstede's statement (about large corporations) for any number of reasons which have nothing to do with dependence on the company. Because we don't know the meaning of the statement, it cannot be accepted as validating the assumed meaning of an individualism index (cf. Ratner, 1997, pp. 204-205; Ratner, 2002a, pp. 140-144). Whatever the correlation may be between responses to the two instruments, the psychological significance of both sets (that produces the correlation) remains a mystery.
Conclusions contradict data
Although cross-cultural psychologists are well-versed in statistical analysis, their conclusions are sometimes driven by theoretical concerns rather than by the data themselves.
Earlier we discussed Ji, et al.'s (2000) conclusion that Chinese are more sensitive to environmental relations than Americans. Yet some of their own data contradict this. In the study, some objects appeared together randomly (Pearson correlation coefficient of 0). Subjects were asked to estimate the association of stimuli on a scale from "not at all" (0 association) to "perfect association (100). Americans estimated the association to be 37 and Chinese Ss estimated it to be 44. Chinese were far from accurate, and they were less accurate than Americans -- they overestimated the association more than Americans did. According to Peng's operational definition of attentiveness, Chinese were less attentive to relationships in the environment than Americans.
Some of Peng & Nisbett's (1999) data similarly contradict their conclusion that Chinese people prefer, understand, and use dialectical proverbs more than Americans. The experimenters presented both groups with dialectical and nondialectical proverbs, appropriately translated, from American and Chinese sources. Subjects rated their preference for all the proverbs from 1 (not at all) to 7 (very much). The results are reconstructed in table one.
The authors claim that the data demonstrate that "Americans greatly preferred nondialectical to dialectical American proverbs, and the Chinese preferred dialectical to nondialectical Chinese proverbs." The finding is contradicted by many data which the authors ignore:
1) Americans liked American dialectical proverbs just as much as Chinese liked them (4.5 vs. 4.2). If Americans are not dialectical while Chinese are, there should have been substantial differences in the ratings.
2) On Chinese proverbs, Americans liked dialectical ones as much as they liked nondialectical ones (3.5 & 3.5). If Americans are nondialectical, they should have preferred the nondialectical ones more.
3) On American proverbs, Chinese liked nondialectical ones as much as dialectical ones. If Chinese are systematic dialecticians they should have preferred the dialectical ones from America as well as China.
4) Chinese like nondialectical American proverbs more than Americans like nondialectical Chinese proverbs (4.2 vs. 3.5). If Chinese are dialecticians, they should dislike nondialectical American proverbs.
5) Chinese prefer nondialectical Chinese proverbs more than Americans do (4.1 vs. 3.5). This contradicts the idea that Chinese are more dialectical than Americans.
6) Americans prefer dialectical American proverbs to nondialectical Chinese proverbs (4.5 vs. 3.5). If Americans are nondialectical, the ratings should be reversed.
7) On the 7-point scale, Chinese only gave Chinese dialectical proverbs a moderate rating of 5. If they are strongly dialectical why didn't they give higher ratings?
8) Chinese gave moderate ratings to American and Chinese nondialectical proverbs (4.2 and 4.1 out of 7 points). Strongly dialectical thinkers would be expected to give very low ratings to these proverbs.
These eight data all contradict the authors' contention that Chinese are dialecticians while Americans think in fragmented, linear terms.
Small quantitative differences are exaggerated as substantive qualitative differences. This stems from relying upon statistical tests of significance. Statistical tests conclude that scores are significantly different even when they are quantitatively quite similar. Group scores of 35.2 and 35.4 can be significantly different statistically, despite the minimal quantitative difference between them. It is well known that statistical significance only means that the numerical differences could not have occurred by chance. It does not mean that the numbers are widely different. Nor does it mean that the numbers represent significantly different psychologies. Yet positivistic cross-cultural psychologists routinely ignore this fact and misrepresent statistical significance as significant psychological differences.
Thus, Peng concludes that cognitive differences between Chinese and Americans are broad, systematic, and qualitative: "The differences we have found, it should be noted, are actual qualitative ones. Chinese participants preferred dialectical proverbs, whereas American participants preferred nondialectical ones" (Peng & Nisbett, 1999, p. 750). "Chinese appear to share a dialectical epistemology that stresses the changing nature of reality and the enduring presence of contradiction. This stands in contrast to a Western linear epistemology built on notions of truth, identity, and contradiction. As a result, some scholars argue, Chinese prefer to seek a compromise in the face of contradiction, whereas Americans pursue more exclusionary forms of truth and resolution" (Peng, et al. 2001, pp. 256-257). However, table 1, above, contains miniscule, contradictory quantitative differences in the subjects' preferences. Peng's conclusion about qualitatively contrasting cognitive styles is inflated.
Sweeping statements about the individualism or collectivism of ethnic groups are similarly out of order. IC is a volatile attribute, easily affected by minor situational changes. An individual is not always, or generally, individualistic; individualism is elicited by particular situations. People from "individualistic cultures" (relatively more individualistic than collectivist) may manifest individualistic cognitions when they are alone, however, when their friends or families are under attack they will employ collectivistic cognitions and behaviors. Individuals can alter their individualism and collectivism in a moment's time: "instructing individuals to think for 2 minutes about what makes them the same as their family and friends results in responses that are more collectivist; instructing them to think for 2 minutes of what makes them different from their family and friends results in responses that are more individualist" (Triandis, 2001, p. 40).
We cannot say that "Collectivists have X psychology." We can only say that "Collectivists in Q country have X psychology in situation A." Collectivists in country Q may have Z psychology in situation B, and W psychology in situation D. Other collectivists in country M may have Y psychology in situation A (ibid., p. 41). For example, Hong Kong subjects (country Q) distribute resources using the principle of equality (psychology X) to in-group members when trying to care for them (situation A). However, these Ss distribute resources on the basis of equity (psychology Z) with in-group members when the objective is to maximize production (situation B), rather than caring for the members. And Hong Kong subjects employ equity (psychology Z) to members of an out-group (situation C).
The volatility of IC is revealed in the fact that different measures of it, which contain different questionnaire items (concerning different situations), generate very different IC responses (Oyserman, Coon, & Kemmelmeier, 2002, p. 42). Leung & Stephan (2001, p. 384) similarly recognized that "the relationship between culture and reward allocation in collectivistic cultures seems complex and is qualified by many situational variables. The individualism-collectivism framework alone is unable to provide a coherent account of the empirical evidence."
As operationally defined and measured by cross-cultural psychologists, individualism and collectivism are not broad, enduring values with definite properties. Yet these psychologists still insist that variables such as IC do have definite characteristics which generate main effects on psychology. Sweeping statements are made about "the psychology of individualists and collectivists" independently of subjects' circumstances and objectives. Triandis, himself, asserts that individualists value pleasure, possess high self-esteem, work better alone, perceive individuals, make internal attributes, and express themselves directly, while collectivists are conservative, self-effacing, work well with in-group members, perceive groups and relationships, attribute behavior to external factors, express themselves indirectly, and use action verbs rather than state verbs (Triandis, 2001, pp. 40-43). Given the qualifications which Triandis introduced earlier, general pronouncements about IC must be scrutinized to determine whether they are overgeneralizations from limited situations in which the findings were obtained.
The fact that theoretical and methodological errors are committed by eminent scholars, accepted by scholarly peer reviewers, and are published by editors of the most prestigious journals and publishing houses suggests that they stem from fundamental, widely shared concepts about human psychology, society, and methods for investigating these (cf. Ratner, 1991, 1997 for a discussion of these). Comprehending the central role that real, concrete culture plays in human psychology requires a theoretical and methodological system that fundamentally departs from the problematical assumptions of cross-cultural psychology described in this article. Reforms that are attempted within these assumptions are never sufficient.
For instance, Sulaiman, et al. (2001) diligently attempted to construct a culturally sensitive symptom checklist for depression in Dubai. They painstakingly solicited 200 community members to produce 400 expressions of emotion. This list was then examined by a second indigenous sample of 50, leading to the identification of a list of "feeling" expressions. This list was then scrutinized to extract 173 words that expressed aspects of clinical depression. A third indigenous sample of therapists identified 96 expressions that described their patients' depression. These 96 were then grouped into 22 categories that comprised main symptoms of clinical depression in the native population of Dubai.
This well-motivated, painstaking effort was subverted by requiring subjects to respond with isolated, simple, abstract emotional terms. For instance, guilt emerged as a main symptom and it was associated with the term "it's painful." This limited response does not enable us to comprehend the cultural psychology of guilt in Dubai. Sense of failure was another symptom and it was described as "calamitous," and "I am falling to pieces." These two phrases leave us uninformed as to the cultural quality of failure in Dubai. We would need to know whether failure is construed as a personal failure, fate, a sign of sin, or due to witchcraft by one's neighbors. Similarly, self-dislike was one of the 22 symptoms and it was described as "I hate myself." But the crucial question -- what kind of self is implicated in Dubai depression? -- receives no answer from this truncated, abstract phrase. To say that depressed Dubais hate themselves without indicating what kind of self they have and hate reveals little about cultural psychology. Remaining within the confines of a problematical methodology negated the authors' efforts at solving the methodological problems.
Van de Vijer & Leung, (1997, pp. 264-269) similarly sought to enhance the ecological validity of operational definitions objectified in measurement instruments. The authors offer useful recommendations that Ss be given time to familiarize themselves with the test material, that translated instruments be back translated to ensure accurate translation, that two researchers administer instruments to eliminate the influence of the tester, and that, where necessary, numerical points on Likert scales be replaced with verbal designations (e.g., "moderately like") which the researcher can later convert to numbers. However, these helpful suggestions are made within the framework of positivistic methods such as operational definitions, artificial, simple, fragmentary stimuli, and truncated, superficial responses. For instance, the authors state that to ensure good translatability of questionnaires one should "use short, simple sentences in order to minimize the cognitive load of the instrument" (ibid., p. 266). Short, simple sentences, possessing minimal significance and requiring minimal cognitive exertion to comprehend, cannot represent culturally meaningful stimuli; nor can they elicit culturally meaningful responses.
New methodological and theoretical principles must be developed to explore the cultural organization of psychological phenomena. A theoretical standpoint must articulate a coherent theory of culture that identifies major cultural factors which bear on psychology. These cultural factors must include broad, enduring, structural elements of society. We want to know how psychology takes form in and takes the form of macro cultural factors such as social institutions, cultural concepts, and widely used artifacts. We want to understand how the specific features of these -- the ways they are socially organized -- become embodied in psychological phenomena. This analysis reveals the heterogeneous characteristics of psychological phenomena which are spawned by multiple cultural factors within a society.
Relating psychological phenomena to concrete macro cultural factors overcomes the errors of cross-cultural psychology such as focusing on abstract cultural factors, atheoretically and haphazardly selecting cultural factors, the platform model which fails to relate culture to psychology, sweeping generalizations about the psychology of all members of a society, and misrepresenting the nature of cultural factors which affect psychology.
Relating the characteristics of psychological phenomena to the concrete social organization of macro cultural factors is the thrust of activity theory as formulated by Ratner (1991; 1997; 1999; 2000a, b; 2002a, b). It amalgamates ideas of Vygotsky, Bourdieu, and others. The main emphasis is that psychological phenomena are constructed in the process of forming and maintaining macro cultural factors. Psychological phenomena are therefore part of culture and embody features of macro cultural factors. Originating in, embodying, and perpetuating macro cultural factors, psychological phenomena are a cultural factor also. They are a particular kind of cultural phenomena which can be conceptually distinguished but not divorced from the others. Psychological phenomena such as perception, emotions, motivation, memory, personality, intelligence, and mental illness are conceptually distinguishable from other macro factors such as artifacts, cultural activities, and cultural concepts. However, the form and content of psychological phenomena are imbued with those of other macro cultural factors. This is what gives psychological phenomena a concrete cultural character.
For example, the individualistic self arose as part of the transformation to capitalist economic activity, and it embodied concrete characteristics of economic relations and Puritan religion:
The culture of modern individualism emerged most prominently and pervasively in England in the century leading to the English Revolution. It began with the rise of a Puritan opposition in the 1560's...Its constituents were the product of profound changes in the English economy...The increasing role of individual initiative, business acumen, and responsibility for success in this new market economy generated a rising group of enterprising rural gentry, yeomen, and artisans...The dependence of fortune on an individual's own actions increased the reliance on personal judgment and initiative (Block, 2002, pp. 39-40).
Macpherson (1962) identifies additional concrete features of the individualistic self which derived from emerging capitalist economic and political activities. The individualistic self was geared toward competing against others; a hedonistic quest for sensual, material enrichment/pleasure, self-preservation; accepting continual change and uncertainty; alienation and increasing distrust of other people and social obligations; utilitarianism; unconcern with issues beyond the individual's immediate situation; marketability/commodification of the self; privately owning objects and land; and the sense that the individual was the basis of society rather than its product.
These concrete features of the self are only visible by identifying their occurrence in macro factors where they exist to achieve certain ends in specific ways. In contrast, cross-cultural psychologists' notion of the individualistic self ("emphasizing personal autonomy and self-fulfillment") is amorphous because it is not grounded in real macro factors.
Especially in complex society, individuals are exposed to psychological phenomena embedded in multiple cultural activities, artifacts, and concepts. Individuals select and synthesize psychological from this multiplicity. Some businessmen may bring a strong family self into their work and may minimize the possessive self. Consequently, it is necessary to research individuals to ascertain just what their cultural psychology is.
Ascertaining the concrete cultural quality of psychology as it exists for individuals requires qualitative methods. Qualitative methods elicit extended descriptions from subjects and then elucidate the quality of psychological phenomena through a contextual, hermeneutical analysis. For instance, the particular qualities of romantic love, depression, or problems solving will be elucidated through such an analysis. Qualitative methodology eschews superficial, artificial, fragmented, simplistic tests and responses. It abandons the fetish of converting psychological phenomena to numbers and analyzing these according to mechanical statistical tests.
Once the quality of psychological phenomena is comprehended, it can be compared with the content of macro cultural factors to determine which aspects of the latter people have incorporated into their psychology. The qualities of romantic love, depression, the self, and problem solving can be compared with characteristics of social activities (economic activity, religion, family life), artifacts, and cultural concepts. This comparison is one way of indicating the cultural origins and organization of psychological phenomena.
The cultural origins, characteristics, and functions of psychological phenomena can be compared across social systems. The cultural qualities of self and love in Liberia can be compared with those in Poland. (A qualitative comparison is far more meaningful than comparing quantitative scores devoid of cultural and psychological significance.) Moreover, we can compare the macro factors that generate these psychological qualities in various countries. Cross-cultural research presupposes in-depth cultural psychology in individual societies. It should not presume and emphasize abstract, transcultural factors devoid of concrete character.
Finally, activity theory formulates the relation between biology, culture, and psychology in a coherent way. It recognizes biology and culture as operating on psychology in different ways, at different levels of specificity. Vygotsky explained that biological mechanisms directly determine elementary functions. These include the behavior of lower animals and infants, and certain adult sensory functions such as the just noticeable differences in stimulus properties that we can detect, e.g., between colors, sounds, moving objects, weights. With regard to higher psychological functions, biology establishes the potential, or capability, to learn, think, perceive, reason, speak, organize social activities, and produce tools. The neocortex is critical for establishing the capacity to develop these, however it does not determine that they will develop or what they will be if/when they do. These capacities are only realized or canalized in specific forms and characteristics through cultural processes.
Specific psychological forms and contents are not apportioned into biological parts and cultural parts operating according to completely different principles -- e.g., the form/structure of language is biologically determined while the content is culturally constructed. Rather, the form and content of psychology are both organized by cultural factors and processes (cf. Tomasello, 1995. 1999; Ratner, 1989, 1991, 1998a, b, 2000b, 2003a, b for such a formulation).
A coherent theory of biology, culture, and psychology is more informative than incoherent, schematic notions about this crucial relationship. It also avoids the conundra of conceptual problems that plague formulations such as Matsumoto's. Finally, activity theory avoids subordinating culture to biology as many cross-cultural psychologists do.
In summary, the theoretical and methodological approach of activity theory presents the real, concrete cultural life of people as central to their psychology.
 In this schema there would be little significance to the universal biological program because people mostly express emotions in social situations, and the expression of emotions is most important in social situations -- where cultural display rules should override biological programs. Moreover, if cultural display (and interpretation) rules are dominant in social situations, then the claim (by Ekman, for example) for universal recognition of expressions is inexplicable or invalid because it was conducted in social situations where an experimenter interacted with a subject. Subjects who posed and judged emotions should have been affected by cultural rules which should have generated dramatic differences in the research findings. In fact, this is exactly what did happen. Reports of universal similarities in emotional displays, feelings, and recognition are overstated (Ratner, 1989, 2000b).
2. There is good reason to suspect that subjects may not have regarded the statements as contradictory. In the example above, the paired statements can both be true. The first one implies that weak family ties lead to unsatisfying social relationships. The second one asserts that weak family ties leads to maturity. There is no contradiction between stating that weak family ties can have two psychological outcomes. When thinking about these statements, the subjects may not have been concerned with handling contradictions because they may not have perceived any. With no information about what psychological function is occurring, Peng has no warrant for drawing psychological conclusions from this data.
3. Even if the psychological significance of one measure were known, its correlation with the other would not illuminate the significance of the latter. The two might correlate for any number of reasons and would not indicate that they had similar meanings.
4. Michell (1999) contends that psychological phenomena are not amenable to quantification and measurement at all. This would call into question all conclusions about culture and psychology that were based on psychological measurement.
5. Van de Vijer & Leung's methodological assumptions never allow new principles to be considered. They never once mention the existence of alternative methodological principles such as qualitative methodology. Their chapter exclusively concerns positivistic methods and refinements thereof. Yet the authors claim that the major goal of their chapter is to provide a comprehensive overview of methodological issues encountered in cross-cultural research (Van de Vijer & Leung, 1997, p. 259). Since qualitative methods do not appear in the "comprehensive overview" they must, by implication, not exist, or not be worth considering. This is a case of defending the status quo through excluding alternatives from consideration rather than through reason.
Berry, J., Poortinga, Y., Segall, M., Dasen, P. (1992). Cross-cultural psychology: Research and applications. New York: Cambridge University Press.
Billig, M. (1996). Arguing and thinking: A rhetorical approach to social psychology. New York: Cambridge University Press.
Block, J. (2002). A nation of agents: The American path to a modern self and society. Cambridge: Harvard University Press.
Findlay, J. (1964). Hegel: A reinterpretation. New York: Humanities Press.
Fiske, A. (2002). Using individualism and collectivism to compare cultures: A critique of the validity and measurement of the constructs. Psychological Bulletin, 128, 78-88.
Gray, G. (1941). Hegel and Greek thought. New York: Harper.
Hua, C. (2001). A society without husbands or fathers: The Na of China. New York: Zone Books.
Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Beverly Hills, CA: Sage.
Ji, L., Peng, K., & Nisbett, R. (2000). Culture, control, and perception of relationships in the environment. Journal of Personality and Social Psychology, 78, 943-955.
Leung, K., & Stephan, W. (2001). Social justice from a cultural perspective. In D. Matsumoto, (Ed.). Handbook of culture and psychology (pp. 375-410). New York: Oxford University Press.
Lonner, W., & Adamopoulos, J. (1997). Culture as antecedent to behavior. In J. Berry, Y. Poortinga, & J. Pandey, (Eds.), Handbook of cross-cultural psychology, vol. 1: Theory and Method (pp. 43-84). Boston: Allyn & Bacon.
Louch, A.R. (1966). Explanation and human action. Berkeley: University of California Press.
Macpherson, C.B. (1962). The political theory of possessive individualism: Hobbes to Locke. New York: Oxford.
Mao, Tse-tung (1937/1967). On contradiction. Selected works of Mao tse-tung, vol. 1 (pp. 311-347). Peking: Foreign Languages Press.
Matsumoto, D. (2001). Culture and emotion. In D. Matsumoto, (Ed.). Handbook of culture and psychology (pp. 171-194). New York: Oxford University Press.
Michell, J. (1999). Measurement in psychology: Critical history of a methodological concept. New York: Cambridge University Press.
Miller, J. (1997). Theoretical issues in cultural psychology. In J. Berry, Y. Poortinga, & J. Pandey, (Eds.), Handbook of cross-cultural psychology, vol. 1: Theory and Method (pp. 85-128). Boston: Allyn & Bacon.
Nisbett, R., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and systems of thought: Holistic versus analytic cognition. Psychological Review, 108, 291-310.
Oyserman, D., Coon, H., & Kemmelmeier, M. (2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3-72.
Peng, K., & Nisbett, R. (1999). Culture, dialectics, and reasoning about contradiction. American Psychologist, 54, 741-754.
Peng, K., Ames, D., & Knowles, E. (2001). Culture and human inference: Perspectives from three traditions. In D. Matsumoto, (Ed.). Handbook of culture and psychology (pp. 245-264). New York: Oxford University Press.
Ratner, C. (1989). A social constructionist critique of the naturalistic theory of emotion. Journal of Mind and Behavior, 10, 211-230.
Ratner, C. (1991). Vygotsky's sociohistorical psychology and its contemporary applications. N.Y.: Plenum.
Ratner, C. (1997). Cultural psychology and qualitative methodology: Theoretical and empirical considerations. N.Y.: Plenum.
Ratner, C. (1998a). The Historical and Contemporary Significance of Vygotsky's Sociohistorical Psychology. In R. Rieber and K. Salzinger (Eds.), Psychology: Theoretical-historical perspectives (pp. 455-474). Washington, D.C.: American Psychological Association.
Ratner, C. (1998b). Prologue to Vygotsky's Collected Works, vol. 5. New York: Plenum.
Ratner, C. (1999). Three approaches to cultural psychology: A critique. Cultural Dynamics, 11, 7-31.
Ratner, C. (2000a). Outline of a coherent, comprehensive concept of culture. Cross-Culural Psychology Bulletin, 34, #1 & 2, 5-11.
Ratner, C. (2000b). A cultural-psychological analysis of emotions. Culture and Psychology, 6, 5-39.
Ratner, C. (2002a). Cultural psychology: Theory and method. New York: Plenum.
Ratner, C. (2002b). Subjectivity
and Objectivity in Qualitative Methodology. Forum Qualitative
Social Research, 2002, 3 (3).
Qualitative Social Research, 2002, 3 (3).
Ratner, C. (2004, forthcoming). Vygotsky's Conception of Child Development. In R. Rieber (Ed.), The Essential Vygotsky. NY: Plenum.
Ratner, C. (2003, forthcoming). Genes and psychology in the news. New Ideas in Psychology, 20
Rogoff, B. (2002). Investigating the coordinated functioning of multifaceted cultural practices in human development. Human Development, 45, 211-225.
Smedslund, J. (1995). Psychologic: Commonsense and the pseudoempirical. In J. Smith, R. Harre, & L. Van Langenhove (Eds.). Rethinking psychology (pp. 196-206). London: Sage.
Sulaiman, S., Bhugra, D., & DeSilva, P. (2001). The development of a culturally sensitive symptom checklist for depression in Dubai. Transcultural Psychiatry, 38, 219-229.
Tomasello, M. (1995). Language is not an instinct. Cognitive Development, 10, 131-156.
Tomasello, M. (1999). The cultural origins of human cognition. Boston: Harvard University Press.
Triandis, H. (2001). Individualism and collectivism: Past, present, and future. In D. Matsumoto, (Ed.). Handbook of culture and psychology (pp. 35-50). New York: Oxford University Press.
Van de Vijer, F., & Leung, K. (1997). In J. Berry, Y. Poortinga, & J. Pandey, (Eds.), Handbook of cross-cultural psychology, vol. 1: Theory and Method (pp. 257-300). Boston: Allyn & Bacon.
Van de Vijver, F., & Poortinga, Y. (2002). On the study of culture in developmental science. Human Development, 45, 246-256.
Van Melsen, A. (1973). Atomism: Antiquity to the seventeenth century. In P. Wiener (Ed.), Dictionary of The History of Ideas, vol. 1 (pp. 126-131). New York: Scribners