Sometimes when a science notices a conceptual inadequacy in its discourse, the problem is not the rigor of its categories but the categories themselves. When epidemiologists as much as say that they are embarrassed by the limitations of a long-employed concept such as race, 1,2 this may be a hint that something more is needed than a better formalization of familiar categories.
Specifically, neither “race” nor “socioeconomic status” nor “education” may adequately explain intellectual and communicative access, or how it can affect health. “Cultural literacy” may be a concept that can help to fill that breach. Cultural literacy is a term that roughly means “a good grasp of general mainstream knowledge.” (The specific term “cultural literacy” is not necessarily familiar to all researchers who nonetheless employ the concept.) Epidemiologists are invited to consider the possibility that cultural literacy—a concept derived not from the social sciences but from the cognitive sciences—may be a practical and productive tool for studying the links between education (or race, or social class) and health.
Cultural literacy is a theory proposed to answer the question: what causes “literacy”? The theory of cultural literacy rests on the “knowledge-competence principle,” a fundamental concept in cognitive research. In essence, the principle emphasizes the importance to intellectual competence of relevant prior knowledge. General discussions of the knowledge-competence principle can be found in textbooks on cognitive psychology. 3
Comprehension is not an abstract “skill.” If our brains lack relevant prior knowledge, they magically no longer demonstrate the “skill” of “comprehension.” According to the theory of cultural literacy, both oral and written communications entail mutual comprehension between parties. Consequently, “literacy” must depend on mutual relevant prior knowledge. Thus, to be “culturally literate” means to share much of the same broad, mid-level knowledge that other literate members of a society have acquired, either in school or from their reading. The theory of cultural literacy asserts that a solid grasp of broad knowledge about the general mainstream culture causes—and to a large extent is—increased communicative access to the culture itself.
The theory was first formulated in the mid-1980s by Hirsch and was widely disseminated via a book-length treatment titled Cultural Literacy. 4 The word “cultural” in this context arises because the theory proposes that communicative access (literacy) requires knowledge that is broadly shared. By the mid-1990s, a number of researchers had provided evidence in support of Hirsch's general argument. 5–7,8(p12)
The words “causes” and “is” are strong words in epidemiologic circles. Epidemiologists should consider well the implications of their use in the description of the theory provided above. Now 15 years since first publication of a book-length treatment of the theory (an eternity in any modern experimental science), not a single competent investigator has cited or performed cognitive research that even quibbles with the general argument of the theory of cultural literacy.
In fact, a test of cultural literacy was shown to provide by itself all the information the U.S. Congress requested from the National Adult Literacy Survey. 9 Hofstetter et al. 10 have demonstrated that in a study population demographically similar to the United States, cultural literacy not only exists, but also matters quite a lot.
The concept of cultural literacy is inherently statistical, rather than normative or prescriptive, but it will acquire a relatively precise and stable meaning in any given complex society. Our brains bluntly refuse to communicate without shared frames of reference. “Mainstream knowledge” is not merely a social construct—thus dispensable, or inherently sinister—but rather an indispensable cognitive artifact of our brains at work in complex, diverse modern societies. Our brains insist on it. People are diverse, specialties are diverse; nonetheless, brains require shared frames of reference. Utopia itself will have “mainstream knowledge.”
In Utopia, of course, any two people may know many different things, because of their different specialties and home backgrounds. But both will share the mid-level mainstream frames of reference—cultural literacy—that make effective communication statistically more likely.
Epidemiologists will find that the theory of cultural literacy is highly controversial in the sense that it is probably inconsistent (even dramatically inconsistent in some cases) with widespread academic, philosophical, and “folk” (homemade) theories of knowledge and communication. It is also inconsistent with political, social, and educational premises that implicitly depend on such theories. On the other hand, if cultural literacy is real, then whatever anyone thinks of it, it will sit there calmly, having effects, patiently waiting for epidemiologists to account for them.
Cultural literacy is part of the basic communications structure of a diverse, modern, knowledge-based society. Thus, questions related to infrastructure are relevant to a discussion of cultural literacy. As an analogy, owning a telephone is probably a good thing for you as an individual, but even if you own a telephone, it still matters whether you're able to dial 1%, 50%, or 100% of your fellow citizens. It matters whether telephones are abundant in some places and scarce in others, or whether variations in the quality of telephones make some communications easy and others arduous or practically impossible. In the same way, having an extensive grasp of mid-level mainstream knowledge may be a good thing for you individually, but at the same time the overall level and distribution of cultural literacy in your society will still matter, independent of your own personal knowledge. Epidemiologists may have to consider such “infrastructure” effects on the public health and on health care costs.
Of course, people's individual levels of cultural literacy may affect their own ability to understand the requirements of good health, their access to health care, and their ability to pay for care. For example, more knowledgeable parents in industrializing Europe made greater use of new medical advances. This in turn was associated with lower morbidity and mortality rates in their children. 11
Another example: low-literate Americans have poor ability to understand and properly use healthcare resources, which may both decrease their use of helpful resources and increase their unnecessary use of those resources. 12 Client-centered remedies include distributing greatly simplified health messages.
By implication, cultural literacy, which causes high general literacy, is protective in this situation. This is so even though, at an individual level, those with broad general knowledge will not automatically know all information relevant to their use of and access to healthcare.
Another implication of the theory of cultural literacy is that a higher level of cultural literacy increases further intellectual access—on average, it makes you a quicker study. The broader your existing knowledge, the more likely it is that you already know something that will make it easier to learn something else.
In the United States, at least, broad general knowledge appears to be strongly associated with educational attainment (about twice as strongly as socioeconomic status is associated with educational attainment). 13 It is also associated with power—“access” in the most basic sense. Moreover, cultural literacy can be measured by simple checklists that can be used over the telephone. 9,10 Using random-digit dialing in San Diego, California, a study found that possession of a broad range of declarative knowledge about the mainstream culture was associated with achieving and manifesting power (as measured by occupation, income, and political activity). 10 This held even after controlling for age, education, and ethnicity.
Epidemiologists should probably not think about cultural literacy as if it were a category (like socioeconomic status), but more as a continuous variable that can be assessed for each individual. It is something specific enough to measure, and it is well connected to empirically validated and conceptually rigorous theory (albeit a theory developed outside of epidemiology).
Given the above, a clear epidemiologic question can now be posed in a manner amenable to empirical investigation: is there an association between cultural literacy and good health? If so, what might be the causal pathway? Does cultural literacy produce higher income and increased power, with the health benefits that follow? Or does cultural literacy allow persons better access to health care through better communication with health professionals, better understanding of medical directions, more effective advocacy for their own needs, or increased motivation to modify behaviors?
Epidemiologists should also note the possibility that increased cultural literacy may not produce better health. To assume that the consequences of cultural literacy will be purely positive requires at least three other assumptions, all highly debatable. One, it assumes that people will never use their powers to obtain things that worsen their health. Two, it assumes experts are always right, and therefore increased access to experts leads to better health. Three, it assumes that all the information comprising cultural literacy is actually conducive to better health. Finally, there may be circumstances in which cultural literacy is actually punished, to the detriment of those possessing it (see below).
A brief excursus into “years of education” may introduce epidemiologists to a bit of the subtlety behind ideas only broadly sketched here. “Years of education” is probably not an adequate marker for cultural literacy (maybe especially in the United States). For example, when Ferguson controlled for actual educational attainment (as measured on the Armed Forces Qualification Test) rather than years of education, the income gap between blacks and whites in the United States decreased from 16 per cent to less than 5 per cent. 14 Which is to say, the current American way of schooling at the primary and secondary levels may be differentially unhelpful to the educational achievement of African-Americans.
A prominent initial source of most individuals’ cultural literacy is school. Schooling is a specific measurable entity with empirical and theoretical links not only to communicative access but also to general work-related competence. 15 This can account for variance in income among American blacks and whites at least as well as socioeconomic status or race—categories whose conceptual clarity, proper modern scientific foundation, theoretic richness, explanatory value, and formal rigor epidemiologists have recently begun to question.
This is not to argue that other categories should be eliminated. Studies of cultural literacy often also employ the concepts of income and race. The question is simply which categories are most predictive of the outcome under study. In Ferguson's study of income, 14 the category of “race” was not the most pertinent in explaining the observed variance. 16
Should studies of health measure cultural literacy? Could cultural literacy more precisely account for some health effects previously attributed to other factors? Because of differentially poor schooling, could it be that race and socioeconomic status sometimes actually serve as crude markers for lack of general mainstream knowledge? At the moment, questions such as these are almost entirely open.
In some modern societies, there may well be ethnic and class differences in cultural literacy that are relevant to health. By itself, lower cultural literacy does not indicate an absolute “knowledge deficit,” but only a relative deficiency in mainstream knowledge. Moreover, mainstream knowledge, although crucial for intellectual and communicative access, is neither automatically “better” nor even necessarily correct. What to do? Probably, there is no free lunch. At the individual level, real people still are well or ill, and need (and sometimes mistakenly misuse) health care resources, provided by distinctly non-utopian societies.
At the system level, Hirsch 4,8 has argued that high mean levels and narrow distributions of cultural literacy do not just happen; heterogeneous societies desiring these must painstakingly aim for both, in both theory and practice. Certain national models of universal schooling dramatically reduce ethnic (eg, Third World immigrant) and class variance in general academic knowledge while also producing high mean levels. 17,18 Adult education and literacy programs can succeed, 19 and mothers attending them are better at nurturing literacy in their children. 20 Plain old lifelong general reading increases cultural literacy. 7
For completeness, it is worthwhile to make a distinction between the association of cultural literacy with communicative access, and the association of communicative access with power and health. Although the theory of cultural literacy proposes that firm possession of the shared mid-level background knowledge of the mainstream culture per se causes communicative access to that culture, population increases in cultural literacy per se do not create jobs, democratic participation, and health. Hence, in a given society, increased cultural literacy could merely create large masses of disaffected people whose only increased “power” would be the ability to offer more sophisticated accounts of why they are without jobs, have no say, and suffer ill health. To take an even more extreme example, in the 1970s, the Communist Khmer Rouge “tried to turn Cambodia into a self-sufficient, agrarian utopia. They forced people to move from cities to the countryside and murdered the educated and the skilled.”21 In Cambodia under the Khmer Rouge, to be culturally literate was to be murdered.
Given the examples here, the firmly causal relation between increased mainstream knowledge and increased mainstream literacy does not of necessity produce income, power, and health; such an association can exist and even be causal, but only within the possibilities created by larger societal beliefs and structures.
The theory of cultural literacy may give epidemiologists a more precise and scientifically rigorous way to think theoretically about the issue of intellectual and communicative access. Recent research may provide some hints about how to undertake its practical empirical investigation. 9,10 Interested epidemiologists will find some overlap of knowledge and interest with colleagues in other fields (for example, political scientist Hofstetter 10 is also adjunct professor in a school of public health). Good luck to all the brave pioneers of this new conversation.
I thank Mary Barger, Eugene Declercq, David C. Geary, E. D. Hirsch, C. Richard Hofstetter, and Thomas G. Sticht, who read drafts of the paper and contributed valuable comments.
1. Kaufman JS, Cooper RS, McGee DL. Socioeconomic status and health in blacks and whites: the problem of residual confounding and the resiliency of race. Epidemiology 1997; 8: 621–628.
2. Morgenstern H. Defining and explaining race effects (editorial). Epidemiology 1997; 8: 609–611.
3. Baddeley A. Human Memory: Theory and Practice. Needham: Allyn and Bacon, 1998; 125–143.
4. Hirsch ED. Cultural Literacy: What Every American Needs to Know. Boston: Houghton Mifflin, 1987.
5. Kosmoski GJ, Gay G, Vockell EL. Cultural literacy and academic achievement. J Exp Educ 1990; 58: 265–272.
6. Pentony JF. Cultural literacy. Adult Basic Educ 1997; 1: 39–45.
7. Stanovich KE, West RF, Harrison MR. Knowledge growth and maintenance across the life span: the role of print exposure. Dev Psychol 1995; 31: 811–826.
8. Hirsch ED. The Schools We Need and Why We Don't Have Them. New York: Doubleday, 1996.
9. Sticht TG, Hofstetter CR, Hofstetter TG. Assessing adult literacy by telephone. J Lit Res 1996; 28: 525–559.
10. Hofstetter CR, Sticht TG, Hofstetter TG, Huie C. Knowledge, literacy, and power. Communic Res 1999; 26: 58–80.
11. Geary DC. Evolution and proximate expression of human paternal investment. Psychol Bull 2000; 126: 55–77.
12. Ad Hoc Committee on Health Literacy for the Council on Scientific Affairs, American Medical Association. Health literacy: report of the Council on Scientific Affairs. JAMA 1999; 281: 552–557.
13. Lubinski D, Humphreys LG. Incorporating general intelligence into epidemiology and the social sciences. Intelligence 1997; 24: 159–201.
14. Ferguson RF. Shifting challenges: fifty years of economic change towards black-white earnings equality. Daedalus
15. Scribner BLS, Smith DA, Baldwin RH, Phillips RL. Are smart tankers better? AFQT and military productivity. Armed Forces Soc 1986; 12: 193–206.
16. Jencks C, Phillips M, eds. The Black-White Test Score Gap
. Washington DC: Brookings Institute, 1998:445, 489–494.
17. Erikson R, Jonsson J, eds. Can Education Be Equalized? The Swedish Case in Comparative Perspective. Boulder: Westview Press, 1996.
19. Sticht TG. The Adult Education and Literacy System (AELS) in the United States: Moving from the Margins to the Mainstream of Education. Washington DC: U.S. Department of Education, Division of Adult Education and Literacy, 2000.
20. Van Fossen S, Sticht TG. Teach the Mother and Reach the Child: Results of the Intergenerational Literacy Action Research Project of Wider Opportunities for Women. Washington DC: Wider Opportunities for Women, 1991.
21. PBS NewsHour, 18 June 1997
(transcript). Alexandria, VA: Public Broadcasting Service, 1997.