Why Cultural Anthropology Students Should Learn

Quantitative Research Methods

When I accepted the job as Department Head at Muhlenberg, I did it in part because I saw an opportunity to establish something unique: a Sociology and Anthropology Department that was committed to examining human social behavior from a rigorous and rational scientific perspective.  The first goal that I listed on my original curriculum proposal (1999) made my intentions clear :

• To transform the Sociology, Anthropology and Social Work Department from a disjointed collection of courses and programs into a department containing theoretically and methodologically compatible anthropology and sociology majors and minors based on a scientific approach to the study of human social behavior.

As part of that goal, I was determined to establish two Research Methods courses: one devoted to teaching research design and discussing epistemological issues; and one designed to teach quantitative research methods. To be considered scientific, research must include certain minimum methodological requirements:  (1) scientific concepts must be measurable; (2) the relationship between concepts must be measurable; (3) scientific statements must make predictions that are testable and falsifiable; and (4) scientific explanations must be accepted or rejected solely in terms of their relative ability to explain the data (see Science and the Study of Anthropology).  Unfortunately, many prominent cultural anthropologists have failed to meet these minimum standards: certainly not Franz Boas, Malinowski, Radcliffe-Brown, Julian Steward, Roy Rappaport, Mary Douglas, Levi-Strauss, Clifford Geertz or Victor Turner.  In fact, many cultural anthropologists openly reject the applicability of scientific research methods to cultural anthropology, arguing that they are inappropriate for examining most of the topics that are of interest to anthropologists.  Indeed, the popularity of postmodern thinking in cultural anthropology represents the contemporary expression of a longstanding hostility to science among cultural anthropologists (see Cartmill 1994).

However, social science methodology should not be different from that of the so-called “hard sciences.”  Science as a method transcends specific academic disciplines, as should be expected given the arbitrary nature of discipline boundaries.  This does not mean that all anthropological research falls under the scientific mantle.  Many anthropological topics do not lend themselves as easily to scientific research as do others.  The same is true for topic in other areas of research, including in the biological and physical sciences. However, it needs to be accepted that such research is, therefore, not scientific rather than to redefine science in order to accommodate such research.  Moreover, if such topics and the methodologies that are used to examine them dominate cultural anthropology, then it becomes difficult to justify the claim that cultural anthropology is a social science.  There are several critical questions that need to be asked, such as how do cultural anthropologists verify their arguments and how do they choose between competing explanations?  Is cultural anthropology any more scientific in its methodology than history, cultural studies or performance studies, disciplines that do not claim to be sciences?  If non-scientific methodologies dominate the field, is it valid to claim that cultural anthropology is not strictly a social science, but rather a type of social studies?

 Quantitative research methods are important because they provide an objective (intersubjective) basis for evaluating claims of causal relationships.  Based on probability theory, they provide a superior measure for evaluating claimed causal relations and for choosing between two or more competing theories.  Paul Thagard, in his article “Why Astrology is a Pseudoscience,” argues that astrology cannot be considered a science because it does not test its propositions.  Astrologers merely propose causal relationships between planetary alignments and human behavioral characteristics and apply selective and/or anecdotal evidence to support their claims.  This contrasts sharply with what astronomers have achieved using the sophisticated and rigorous quantitative research methods they have developed over the past few centuries.  Astrologers, like psychoanalysts, continue to use highly subjective and interpretive concepts and principles and have resisted attempts to test their claims.  As a result, whereas astronomy today looks nothing like the astronomy of 100 years ago (or even 50 years ago!), astrology has changed little in the past 3,000 years.  One can make a comparable contrast between physics and natural philosophy, between ecology and natural history and between archaeology and biological anthropology on the one hand and cultural anthropology on the other.

Unlike biological anthropology and archaeology, cultural anthropology has not advanced our understanding of human social behavior by testing competing theories of human social systems, rejecting those that don’t explain the data and accepting those that do, but rather has proposed explanations of human behavior that are largely untestable and heavily influenced by prevailing social and political considerations.  A clear example of the fundamentally unscientific basis of cultural anthropology is the discipline's continuing commitment to the concept of culture.  Culture is a non-operational concept from which only non-measurable and non-testable causal relationships can be proposed.  The persistence of this concept for more than a century and cultural anthropologists' active defense of it for so many years compares with the intensity and manner by which astrologers defend their concepts, by which psychoanalysts defend the use of such concepts as id, ego, superego, and archetypes and by which theologians defend concepts such as the soul, evil, grace and divine providence.  The concept of culture is only able to persist because a large number of cultural anthropologists accept the notion that quantitative research methods are not appropriate to cultural anthropological research.  When the day comes that cultural anthropologists undertake serious quantitative research, the concept of culture will likely be discarded, as was the concept of race a former concept intimately associated with anthropology.  The abandonment of outdated concepts is the normal process in science.  As concepts lose their utility, they are abandoned for newer concepts that, as part of newer theoretical systems, are better able to explain the data.   Some of the "scientific" concepts that have been abandoned in the physical and biological sciences include entelechies, phlogiston and the ether.  Even the species concept has been seriously challenged in evolutionary biology.  No concept is sacrosanct.

It is logically inconsistent for anthropologists to vehemently reject the race concept on the ground that it is unscientific, but to completely embrace the culture concept, which is equally unscientific.  The race concept deserved to be rejected for the same reason that the culture concept needs to be abandoned:  it is not measurable and, therefore, cannot be part of a predictive science whose theories lead to testable propositions.  Like the race concept, the culture concept is irrelevant to a serious scientific explanation of human behavior.  However, it is only by attempting to apply the concept in a rigorous scientific (quantitative) manner and by comparing predictions of models which use the culture concept with those that do not that the validity of the concept can be determined.  This has largely not occurred, and is not likely to occur.   One needs only to look at how cultural anthropologists have approached controversial claims in the past to understand why.  Cultural anthropologists rejected the race concept less on methodological or scientific grounds, than on the basis of sociopolitical considerations.  The same may be said for how the field has approached sociobiology.  Compare, for example, how cultural anthropologists reacted to sociobiological arguments regarding  the role that evolutionary biology plays in human behavior with the way in which physicists responded to University of Utah scientists claims that they had achieved cold fusion.  While the latter conducted independent tests to see if they could replicate the findings of the University of Utah scientists (they could not), cultural anthropologists organized a plenary session in an attempt to get the AAA to publicly condemn sociobiology as racist.  Whether the principles of sociobiology (evolutionary psychology) explain human behavior is a valid issue to raise; however, the issue must be resolved by testing hypotheses derived from competing sociobiological and anthropological principles, not by proposing political action.

Because its research has not been based on rigorous scientific methods, Cultural Anthropology has always been more susceptible to political and social currents than have the other two sub-fields of anthropology.  One need only compare the status of Nancy Sheper-Hughes and others in the AAA who want to politicize the discipline with the American Association for the Advancement of Science's (AAAS) censure of Linus Pauling, who attempted to use his position as a Nobel Laureate to promote his vitamin C theory in the journal Science.  Similarly, one would not find major research sections in Physics, Chemistry or Biology that are comparable to such politically motivated subdivisions of anthropology as Gay and Lesbian Anthropology, Feminist Anthropology, and Anthropology and the Environment.  It has been Cultural Anthropology’s eager politicization and its lack of rigorous scientific methods that has made it an easy target for critics, such as Dan Seligman ,in his article The Science of Self-Preservation in Forbes Magazine.

It is also not valid to claim that quantitative research methods are appropriate for students concentrating in biological anthropology and archaeology, but not for those taking cultural anthropology.  This a false dichotomy.  Scientific methods transcend disciplines and, therefore, sub-disciplines.  One only has to look at the history of anthropology to see the problem with this argument.  When J.B. Birdsell’s first published his textbook, Human Evolution (1972), he subtitled it An Introduction to the ‘New’ Physical Anthropology.  Birdsell used this subtitle because he was doing something that had not been done much in physical anthropology at that time; he was using scientific methods to examine evolutionary processes as they applied to Homo sapiens.  One only needs to compare the physical anthropology that prevailed prior to Birdsell and others to see just how embarrassingly unscientific it was.  While biologists like Mayr, Haldane, Fisher and Dobzhansky were applying sophisticated mathematical models to generate and test concepts and principles of evolution, physical anthropologists such as Hooten, Garn, Coon and others were measuring skulls and proposing racial typologies.  No physical anthropologist was testing evolutionary theory to explain human evolution in the same way that evolutionary theory was being applied and tested in biology.  Indeed, evolutionary theory was largely absent from physical anthropology.  It was as if research and theoretical developments in biology and physical anthropology had nothing to do with one another.  The sophistication of contemporary physical anthropology is the direct result of the introduction of rigorous scientific (quantitative) research methods into biological anthropological research.  It was only with the adoption of rigorous scientific research methods that the race concept could be –and was- legitimately rejected.

Archaeology underwent a similar transformation to that experienced in physical anthropology.  Prior to people like Binford, Plog, Flannery, Sanders and others, archaeology was more a form of historical reconstruction than a social science.  While rigorous methods were used in the process of excavation, the general goal of archaeology was to reconstruct past lifeways in much the same way as is done by historians.  The “New” (or processual) Archaeology of the 1960s & 1970s introduced the notion of theory testing and model building.  This was extremely controversial; indeed, the initial reaction bordered on outrage by many traditional archaeologists.  However, few people do the “Old” archaeology today, and unless an archaeologist submits a sophisticated research proposal to NSF, involving rigorous quantitative methods for testing hypotheses, he or she is unlikely to be funded.  Those disciplines that are accepted today as sciences -physical, biological, social and behavioral- had to become sciences, and central to all of those transformations was the adoption of rigorous quantitative research methods.  Cultural anthropology is no different.

However, whereas both biological anthropology and archaeology have made substantial changes in their methodology –most notably adopting rigorous and sophisticated quantitative methodologies as the basis of their research- cultural anthropology stands alone as the only sub-field that has failed to adopt quantitative methods.  Indeed, many cultural anthropologists continue to reject such methods, and cultural anthropology continues to be the only academic discipline which claims to be a science but which fails to require that its students learn quantitative research methods -all the way up to the Ph.D.!   For this reason, many archaeologists and biological anthropologists argue that Cultural Anthropology is out of step with archaeology and biological anthropology in that it no longer contributes to the discipline’s core interest to scientifically understand and explain human evolution in all its manifestations.  Many cultural anthropologists even reject the notion of evolution when applied to human societies.  The hostility of cultural anthropologists to science and evolution led Matt Cartmill (1994), in an address as President of the American Association of Physical Anthropologists (AAPA), to characterize physical anthropology’s position within AAA as comparable to functioning “as a sub-discipline within the Society for Creation Research.”

The lack of rigorous quantitative methods also limits cultural anthropology’s ability to make meaningful contributions to other fields of science.  Epidemiologists are largely unable to make use of cultural anthropological research, even though the peoples studied by anthropologists are the very populations whose circumstances would be of greatest interest to them.    Epidemiologists can’t use anthropological research to test causal relationships between disease and human behavior, because cultural anthropologists don’t collect and analyze data, certainly not the kind that are used in epidemiological research.  Epidemiologists are, therefore, forced to rely on problematic national census data instead.

The issue regarding the use of quantitative methods concerns the objective verification of anthropological claims, whatever those claims happen to be.  For example, how do we choose among three different explanations of Yanomamo warfare?  Napoleon Chagnon (based on his informants’ testimony) states that warfare among the Yanomamo results from the men’s desire to steal women.  Harris (presenting an ecological functionalist explanation) argues that warfare among the Yanomamo is a result of protein scarcity and serves, through enhanced female infanticide, as a mechanism of population control.  I disagree with both of these authors and suggest that their warfare is a consequence of the competition over scarce resources that results from continued population growth.  What is the best way to decide who is correct, or at least the most correct?  A test needs to be constructed which outlines the predicted consequences of each explanation and compares those predictions with the available data (or newly gathered data, if necessary) to determine which explanation is the most accurate. To the extent that quantitative research methods can be applied, they would significantly enhance our ability to evaluate the competing explanations, because they would provide an objective measure for the degree to which the data support one or more of the proposed explanations.

 If, on the other hand, a cultural anthropologist presents an interpretation of the meaning of a particular ritual, such as Geertz’ interpretation of Bali dance, how do we determine if that interpretation is valid, or even more correct than other interpretations?  What procedures are available to us?  Can quantitative methods be applied to help in choosing between different interpretations or explanations?  For some problems the answer may be yes, while for others it may be no, or at least not yet.  However, it seems highly problematic for a social science discipline to have determined for more than a century that the answer is consistently no. Applying quantitative research methods requires that we examine traditional anthropological problems in wholly new ways, but that is how sciences grow and develop.  The more sophisticated the research methods, the more confidence we can have in the conclusions of anthropological research.  Cultural anthropology needs to learn from other disciplines that have adopted quantitative methods –both social and non-social sciences- to see how these can be applied to enhance cultural anthropology’s ability to understand and explain social behavior. However, to consistently reject such methods is unscientific -and some cases even anti-scientific.  The refusal of cultural anthropology to adopt explicit quantitative research methods, and the degree to which this has left the sub-discipline open to politicization lies at the core of why many archaeologists and physical anthropologists have defected over the years from AAA and have, in some cases, founded separate “Scientific Anthropology” departments, such as at Stanford University and the University of California at Berkeley.  The Department of Anthropology at Penn State University has replaced Cultural Anthropology as one of the three sub-disciplines of anthropology with Demographic Anthropology, reflecting that Department's recognition of the unscientific basis of a cultural anthropological methodology.

Quantitative research methods can be applied in cultural anthropological research just as easily as they are applied in other social sciences, and it is important that cultural anthropologist begin applying them.   In my Mormon research, for example, I examined the ability of a specific model in general ecology regarding the relationship between diversity and stability in ecological systems to explain the evolution of Mormon communities in the Little Colorado River Basin in Arizona.  Several other researchers had previously studied the same communities: two historians and two anthropologists.  However, while the previous researchers provided many valuable insights into the settlement process, none had made any attempt to examine the development of these communities scientifically.  For the most part, both the historians and the anthropologists “explained” the success of these communities in terms of Mormon values.  However, none of the previous researchers could explain why some communities were more successful than others and why some communities even went extinct since, of course, all of the communities possessed Mormon values.  Furthermore, none of the researchers attempted to define values as a measurable concept or attempted to statistically link certain values with specific attributes of community development among these towns.  While one of the historians attributed the success or failure of individual communities to the interplay of religious values, individual personality differences, environmental considerations, etc., such an eclectic explanation is both highly interpretive and completely untestable.  Mark Leone, an anthropologist, provided an equally untenable argument based on such non-measurable concepts as “sanctity”, “mental states” and “collective memories”. In essence, there was no substantial difference between the analytical methodology of the historians and the anthropologists. (Significantly, Mark Leone, an archaeologist, did not feel compelled to apply the same rigorous methods in his “cultural” analysis of Mormon settlements in the Little Colorado River Basin that he applied in his archaeological investigation of prehistoric settlements in the same area.)

In contrast to previous researchers, I applied an explicit predictive model derived from ecological research.  I operationalized the variables of that model to the empirical context of the Mormon settlements and of the Little Colorado River Basin.  All of my theoretical variables –population size, environmental productivity, community productivity, environmental stability, community stability and community diversity-- were operationally defined in terms of measurable variables within the Little Colorado River Mormon communities and their environments, as were the relationships between them.  Furthermore, the model illustrating those relationships and its specific predictions were delineated in a clearly stated flow chart.  I then tested the predictions of the model against the historical outcomes using a Rank-Order correlation between the measures of population, productivity and stability on the one hand and those of community diversity on the other.  This provided an objective basis for evaluating the validity of the model I had proposed.   My research is certainly not the final word on explaining Mormon colonization in this river basin; nor were the statistics I used particularly complicated.  However, I provided a more precise and accurate explanation of the settlement process than had previously existed and one that explicitly linked this local event to a broader testable theoretical framework.  In doing so, I have provided the groundwork from which future researchers can provide even better explanations, using more sophisticated research techniques.

In closing, it is important to point out  that, without an understanding of quantitative research methods, cultural anthropology students could not comprehend much of the research in contemporary archaeology or physical anthropology, let alone the research taking place in related social science disciplines.  Furthermore, by not requiring cultural anthropology students to learn quantitative research methods, academic departments limit these students from applying greater scientific rigor in their research.  In the end, given that today's anthropology students represent the future of the discipline, by not pushing students to apply quantitative methods in their research, we limit the future of cultural anthropology as a social science.

References Cited

Cartmill, M. 1994. Reinventing Anthropology: American Association of Physical Anthropologists Annual Luncheon Address, April 1, 1994. Yearbook of Physical Anthropology 37:1-9.

*     *     *     *     *