Download Commentary: Culture in epidemiology—the 800

IS MODERN WESTERN CULTURE A HEALTH HAZARD?
grain—of truth. But these broad themes are said to uniformly
represent the meanings by which thousands—or millions—of
people construct an understanding of the world (despite
Eckersley’s brief aside regarding potential sources of variation).
Such a characterization conflates the aggregate and individual
levels of analysis; fails to take into account intracultural diversity,
both at the group and individual level; and, at the individual
level cannot differentiate the cultural from the socialpsychological.
A cognitive theory of culture can resolve these issues.6
From a cognitive orientation, culture is the knowledge one
must possess to function in a society (a definition that
Eckersley mentions, but does not develop). It is important to
reflect on just what that means. For example, from their
upbringing, many of my academic colleagues ‘know’ quite a bit
about conventional American Protestant Christianity; what they
‘believe’ about that and what their ‘attitudes’ are towards
believers are something else altogether. We know far more
than we believe, and what we believe forms a subset of our
general cultural knowledge. Therefore, culture must be
kept distinct from the conventional constructs of social
psychology.
While the locus of culture is the individual, culture refers to
aggregates, because knowledge is distributed. Each of us has an
incomplete understanding of our own culture, and the whole
truly is greater than the sum of its parts. As Searle7 has pointed
out, even when an individual’s knowledge is incomplete, there
is a sense of the collective, of how ‘we’ do things, so that it is not
a mysterious experience to have a sense that, yes, we Americans
do things a certain way, even though I myself am not quite sure
how it is done.
Briefly, a cognitive theory of culture explicitly allocates culture
to both the aggregate and the individual; it can explicitly take
into account intracultural diversity at the group and the
individual level; and, it clearly separates the cultural from the
social-psychological. Furthermore, it provides a method for
determining: (i) if some domain (e.g. individualism) is salient
within a social group; (ii) if understanding of that domain is
widely or narrowly shared; and (iii) if variation in group or
individual salience of the domain has an impact on other factors,
including health.
259
Theory matters. Using this orientation I have found, in urban
Brazil8 and the African American community in the United
States,9 that shared cultural models of ‘the good life’ resemble
less what Thorstein Veblen10 called ‘conspicuous consumption’
and more what he called ‘a common standard of decency.’
Persons at greatest risk of poor health were those unable to
achieve quite modest cultural goals. This is but one example
of how a nuanced view of culture as local knowledge can
lead to conclusions quite different from a cultural configurationist view.
This is not to say that Eckersley is all wrong. It is to
say, however, that his view of cultural influences does not
lead us to ask questions like: Is the meaning of materialism
(or individualism) broadly shared in a society? How is it
distributed within a society? How strong a directive force in
individual belief and behaviour is this domain? How does
variability in its directive force influence individual adaptation,
including health and a sense of well-being? For a refined view
of cultural influences on health, these questions must be
addressed.
References
1
2
3
4
5
6
7
8
9
Eckersley R. Is modern Western culture a health hazard? Int J Epidemiol
2006;35:252–58.
Kuper, A. Culture: The Anthropologists’ Account. Cambridge, MA: Harvard
University Press, 1999.
Putnam RD. Bowling Alone: The Collapse and Revival of American
Community. New York: Simon & Schuster, 2000.
Edgerton RB. Sick Societies: Challenging the Myth of Primitive Harmony. New
York: Free Press, 1992.
Benedict R. Patterns of Culture. New York: Houghton Mifflin, 1934.
Dressler WW. What’s cultural about biocultural research? Ethos
2005;33:20–45.
Searle J. The Construction of Social Reality. New York: Free Press, 1995.
Dressler WW, Balieiro MC, Ribeiro RP, dos Santos JE. Cultural
consonance and arterial blood pressure in urban Brazil. Soc Sci Med
2005;61:527–40.
Dressler WW, Bindon JR. The health consequences of cultural
consonance. Am Anthropol 2000;102:244–60.
10
Veblen T. The Theory of the Leisure Class. New York: Macmillan,
1899.
Published by Oxford University Press on behalf of the International Epidemiological Association
Ó The Author 2005; all rights reserved. Advance Access publication 22 November 2005
International Journal of Epidemiology 2006;35:259–261
doi:10.1093/ije/dyi237
Commentary: Culture in epidemiology—the
800 pound gorilla?
Thomas A Glass
Johns Hopkins Bloomberg School of Public Health, Department of
Epidemiology, 615 N. Wolfe Street, Baltimore, MD 21205, USA.
E-mail: [email protected]
The paper by Eckersley1 in this issue of the International Journal of
Epidemiology is important in several respects. It represents a rare
attempt to deal with a complex subject that has traditionally
stood outside what Thomas Kuhn2 would call ‘normal science’ in
260
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
epidemiology. It surveys a vast body of evidence, concluding, in
response to the question posed by the title, that yes, modern
Western culture is a health hazard. The author concedes that his
core argument is based on evidence that is ‘indirect and
circumstantial.’ Rather than comment on the substance of his
main aim, my interest is in the more general question of why
culture has been so sparsely examined. Eckersley’s paper is likely
to stimulate discussion and controversy. Any discussion of
culture is strikingly unorthodox and hampered by weak
evidence. It is especially significant that culture has (Marmot
and Wilkinson’s idea of a ‘culture of inequality’ notwithstanding)
been largely ignored or denuded of its essential meaning in
contemporary discourse surrounding determinants of population
health. In this commentary, I address two issues: (i) what is
culture, and (ii) why is it so consistently off the radar screen. In so
doing, I will use several analogies to physics.
Defining culture
As is almost always the case, the thorniest issue has to do with
defining the main concept. Defining culture is especially difficult
because it is among a class of concepts for which there are few
analogues in epidemiology. Our field fits neatly within the family
of disciplines characterized by methodological individualism3
(the idea that all causal arguments, to be convincing, must
explain population-level phenomena as the aggregate consequence of individuals exposed to discrete material events,
substances, or organisms). In this sense, modern epidemiology
shares a great deal in common with Newtonian physics. Newton
set forth a set of universal laws of motion governing the
behaviour of physical objects in terms of discrete, linear,
invariant, and exclusively local forces. Objects and particles
move through trajectories that are fully determined by the
history of ‘exposure’ to those forces. Similarly, epidemiology is
the search for individual trajectories through fixed and invariant
space, in which discrete, isolatable, linear forces (exposures) are
necessary and sufficient causes of those trajectories. Culture,
however, is profoundly counter-paradigmatic in this regard. Like
the pre-20th century idea of the luminiferous ether in physics,
culture has no place in a Newtonian vision of cause and effect.
With few exceptions (think of herd immunity) epidemiology has
great difficulty incorporating aggregate-level phenomena that
exist in larger dimensional space beyond what touches or invades
the individual.
Culture is a non-material, largely symbolic, and cognitive force
(which has material consequences to be sure) existing in the
heads of human actors who are so thoroughly soaked in culture
that it ceases to be something they notice. Just as fish take water
as a given feature of their environment, culture is an invisible
medium through which we all swim effortlessly and without
consideration of its properties. As a result, epidemiologists bristle
at the idea of culture much as they do the now discredited
miasma theory of the sanitation movement. That which is
cultural, because it is non-local, non-material, omnipresent, and
largely invisible, has been off-limits in a field that treats
individuals as objects moving through a material world, subject
to linear billiard-ball-like perturbations of their motions.
Culture is not a property of individuals, but rather a hypercomplex, large-scale symbolic process at the societal level, as well
as smaller-scale patches of cultural space within communities,
neighbourhoods, and workplaces. Culture exists both spatially
(France or South Boston) as well as non-spatially (entertainers,
protestants, or republicans). Yet, epidemiologists are trained to
think in terms of individuals. For this reason, any attempt to deal
with culture invariably reduces to discussions of how different
classes of individuals (blacks, women, Hispanics) talk, think,
behave, or relate. The idea of a symbolic field, permeating the
thoughts and actions of a whole population is simply fantastical
and mysterious. The counter-factual model of causation poses
insurmountable obstacles for thinking of culture in any causal
sense. How can we imagine the effect of culture by comparing
two groups (Japanese and Russians) with the idea that one group
can serve as a proxy for what we would have observed in the
other group if one culture had been replaced by another, and
nothing else had changed?
But, I have yet to define culture, beyond the use of analogies.
Instead, I have sufficed to try to explain why culture is so difficult
to define. Many definitions can be given, and Eckersley’s1 is as
good as any. Adding to this definition, culture can be thought of
as the software on which the human brain operates. Just as two
computers, one running windows, and another Macintosh, will
operate in fundamentally different ways, according to encoded
procedures that are internal to the software and invisible to the
user, culture influences behaviour because it is internal to the
workings of the mind.4 The computer user need know nothing
about the operating system, but will be subtly influenced by the
way it structures and delimits the operations that are made easier
or more difficult. The analogy, of course, has limits. Computers
run on one operating system or another (although this is
changing), while people move in and through a dense
interwoven cultural fabric. The software idea though suggests
that individual consciousness comprises programs, patches,
upgrades, and other wanted and unwanted code fragments
that are ‘downloaded’ from exposure to cultural fields. Those
elements combine to create behavioural proclivities and patterns
that may, as suggested by Eckersley, be powerful and as yet
unexamined shapers of the downstream behaviours that lead
towards and away from disease.
Culture and common ground
Among the arguments made in this paper, Eckersley believes that
‘common ground’ has been achieved between those who argue
in favour of ‘material’ vs ‘psychosocial’ explanations of social
inequality. This is a contentious, if hopeful, claim. Plenty of
evidence remains that a grand unification of materialist and
psychosocial accounts of social inequality has eluded us.
Ironically, the barriers are partly cultural.
The heyday of cultural analysis in public health may have been
the ‘culture of poverty’ theory in the 1960s.5 Oscar Lewis’s
provocative thesis gained rapid influence but was just as quickly
rejected for its tendency to ‘blame the victim.’6 In the subsequent
25 years, cultural arguments have been hampered by the
prejudice that they must inherently conflict with arguments
based on differential access to material resources and opportunity. This places modern-day arguments between the ‘neomaterialist’ and ‘psychosocial’ camps within social epidemiology
into historical perspective. Note that neither camp addresses
culture with any depth or conviction. Eckersley may overstate
the extent of common ground, but his paper provides useful clues
IS MODERN WESTERN CULTURE A HEALTH HAZARD?
as to how culture might become the third leg of the social
epidemiological stool.
Conclusion
Eckersley provides a thought-provoking synthesis of a wide
range of findings. Many will probably remain unconvinced.
Showing evidence of changes over time in population averages of
psychosocial factors such as anxiety, depression, and sense of
control does not lead to the conclusion that cultural changes are
responsible. As he notes, changes in culture may be the byproducts, rather than the determinants, of changing rates of
mental distress. There are other non-cultural explanations with
which to compete. However, the stark spatial and social
patterning of all health behaviours is consistent with the
possibility that cultural software is a contributor. Currently,
epidemiologists lack even basic knowledge about cultural trends.
The measurement of those trends will have to move beyond
aggregating survey responses if we are to proceed. More
importantly, social epidemiologists will be required to abandon
261
either/or explanatory frameworks when considering the role
of cultural factors relative to material deprivation. Envisioning
how culture and material conditions interact synergistically will
prove difficult for epidemiology, but may yield significant
improvements in our ability to account for population patterns
of health.
References
1
2
3
4
5
6
Eckersley R. Is modern Western culture a health hazard? Int J Epidemiol
2006;35:252–58.
Kuhn TS. The Structure of Scientific Revolutions. 1st edn. Chicago, IL:
University of Chicago Press, 1962.
Gellner E. Relativism and the Social Sciences. Cambridge, UK: Cambridge
University Press, 1985.
Eisenberg L. The social construction of the human brain. Am J Psychiatry
1995;152:1563–75.
Lewis O. The culture of poverty. Sci Am 1966;215:19–25.
Rodman H. Culture of poverty: the rise and fall of a concept. Sociol Rev
1977;25:867–76.
Published by Oxford University Press on behalf of the International Epidemiological Association
Ó The Author 2005; all rights reserved. Advance Access publication 8 December 2005
International Journal of Epidemiology 2006;35:261–263
doi:10.1093/ije/dyi238
Commentary: ’Culture’, cultural explanations
and causality
Craig R Janes
As a medical anthropologist, and thus representative of a
discipline that has since the beginning of the 20th century
exercised intellectual dominion over the construct of culture as
explanatory of the human condition (writ both large and small),
I find much in Eckersley’s provocative paper to agree with. I also
appreciate his efforts to incorporate cultural explanations into
social epidemiology. In so doing he has given the field a needed
theoretical push to move beyond a narrow focus on social and
economic relationships to consideration of systems of meaning in
the causation of disease.
Yet, like many anthropologists these days, I experience
considerable intellectual anxiety over the general and uncritical
use of culture as an explanatory variable. Empirical work in
diverse settings conducted over the past three decades has shown
that culture is increasingly hard to define, much less apply, to
understanding social practices. The transnational flows of people
and ideas that are part and parcel of globalization, the legacies of
colonialism and, in consequence, a need to take power into
account, have rendered older ideas of culture—as a relatively
homogeneous set of understandings shared among a group of
Faculty of Health Sciences, Simon Fraser University, Burnaby BC, Canada V5A
1S6. E-mail: [email protected]
socially interacting people—conceptually obsolete.1,2 Thus, as I
teach my students, culture should never be used as an
explanatory variable, or not, at least, without some very careful
unpacking. I suspect that Eckersley’s goal3 of moving us towards
a culturally-informed epidemiology will not be achieved without
some consideration, theoretically and methodologically, of what
culture is, exactly, and how it manages to get into the body. It
seems to me that the challenges are 3-fold: addressing the role
culture plays in human social life; understanding how the ‘stuff’
of culture—ideas, symbols, meanings, shared understandings,
morals, values, beliefs—are distributed within and among social
groups within larger, complex social systems; and developing the
conceptual tools and research methods to apprehend the links
between culture as a shared perspective on the world and
individual experience.
As a starting point, it is important to avoid the culturalist trap;
i.e. seeing culture as a thing in and of itself arising sui generis to
govern social life. Culture, in this sense, is viewed as being
autonomous, explainable only via reference to the ‘working out of
its own internal and particularistic logic’.4 Although anthropologists quibble over the degree to which cultural systems develop
independently of other influences, both social and ecological,
most would agree that cultural systems are inseparable from