The Concept of Human Races: Uses and Problems


The use of stereotyped morphophenotypic races as biologically distinct breeding units has continued despite the scientific ambiguities and inherent assumptions associated with them. The following paper will attempt to identify uses for traditional race nomenclature that exists in various fields of study while discussing reasons for its invalidity in others and as a scientific distinction of human subspecies. What the term “race” actually means in a biological sense will be covered in terms of its definition and also in terms of its current use. Next, a discussion of folk taxonomy and its relevance to the race concept will be discussed. In addition, race will be discussed as a device for studying human population divergence patterns.

The discussion of the use of race in studies of human population divergence will be followed by a review of why forensic anthropology is forced to use traditional race concepts in order to facilitate the useful dissemination of information in courtrooms, to police officers, and to the public that provides the source of data and the audience that must interpret forensic findings. The final sections of this paper will deal with the use of the race concept in demography and in the dissemination of information on at-risk groups for particular genetic disorders.

Traditional Racial Schemes and Folk Taxonomy

The term “race” was first applied to humans in the eighteenth century by Buffon, a French naturalist (Molnar 1998: 19). Various and sundry definitions of human races have been proposed, but in general “race” is seen to represent a population that is biologically distinct according to some defined parameter. It is also used interchangeably with the term “subspecies” by most researchers when race is defined by biological parameters (versus sociocultural parameters, which would more accurately define “ethnic groups”).

The question of whether or not various groupings of Homo sapiens are separate species entirely will not be addressed here since it is utterly ridiculous. The idea of “subspecies”, however, is perfectly reasonable in theory since “subspecies are geographic segments of a species, which differ morphologically to some degree from other such segments” (Groves 1989: 6). Thus, a subspecies demarcation is a division of a species into segments with arbitrary physical boundaries. Their boundaries are based more on the degree of the variation of specific traits that do not usually prevent interbreeding.

The degree of difference needed in the specified traits to allow a subspecies delineation is usually the 75 per cent rule. The 75 per cent rule states that “75 per cent of the individuals classified in one subspecies are distinguishable from 100 per cent of the individuals belonging to the other subspecies of the same species, which is statistically equivalent to 90 per cent joint non-overlap” (Groves 1989: 7). Therefore, a particular subspecies may have no functional differences in their genetic makeup or their anatomical makeup, and the only difference may be something as simple as a different colour.

This means that if there are definable differences between definable populations, then a separation of humans into separate races is theoretically sound. The problem is that there are no populations that have 100 per cent definable boundaries. There are no functional RIMs (Reproductive Isolation Mechanisms) to prevent interbreeding between human groups, which does not allow definable populations to form that can be uncritically separated into races. The ambiguity is great enough that no real taxonomic basis for subspecies can be supported, which is why the trinomen (which demarcates through subspecies) for all humans is Homo sapiens sapiens, regardless of one’s perceived race (Keita & Kittles 1997: 535).

Those researchers who separate humanity into distinct races generally admit that the amount of admixture between perceived races has led to difficulty in determining “original” racial stocks (itself an inherently racist idea that assumes different origins of humans). “The proponents of this view tend to assume that even if races are poorly definable today, they were at one time much more distinct, but have since interbred to such a degree that they had merged into a continuum” (Groves 1989: 292).

The number of “original” races these researchers claim existence for ranges from three to thirty-seven (Molnar 1998: 19). Due to the many different races that have been proposed, and the fact that they are all subject to the same arguments against their validity, the races that are important for forensic anthropology will be discussed (specifically North American forensic anthropology). These will be discussed in both terms of their practical usefulness and their philosophical weakness. The racial scheme that will be discussed is one that includes the categories of black (Negroid), white (Caucasoid), East Asian (Mongoloid), Melanesian/Australian (Australoid), American Indian, and Polynesian (Gill 1998: 296).

Folk taxonomy is the discrimination of various organisms (and objects) into broad categories based on the predilections of one’s mother tongue (Clark 1994: 18). Classifications such as “tree” and “fish” are folk taxonomies since an organism does not have to be a tree according to a biological definition to be a tree, or a fish according to a biological definition of what a fish is. Many English speakers call porpoises “fish”, because dolphins fit the folk taxonomic description of a fish.

Folk taxonomies are “common sense” discriminations of distinct classes based on a perceived important trait or traits, and are still pervasive in much of modern taxonomy. It is only recently that scientific classification based on evolutionary divergence has supplanted much of traditional classification. However, “folk taxa are not foolish inventions: they are related to the uses we would make of things. For landscaping or woodworking purposes oak and pine alike are trees” (Clark 1994: 19). Folk taxa are a part of the culture by definition. One receives one’s folk taxonomic associations from one’s culture. Thus, folk taxonomy is a representation of the way individuals within societies view themselves and their world. This is an important consideration when discussing the use of outmoded racial types in forensic anthropology.

Race as a Unit of Study in Human Divergence

While most serious researchers would admit that there is little difference in supposed human races, with differences being associated more with the environment each group inhabits versus distinct genetic difference (Molnar 1998: 257), research comparing “races” for the presence of blood and/or genetic markers in order to determine affiliations between presupposed races is common. Research of this type is particularly suspect since one must presuppose racial delineation, and there is no way to tell if the selected racial categories are correctly partitioned.

An extremely simplistic analogy would be trying to determine a political candidate’s support among various groups by a researcher placing people in demographic groups according to his/her opinion of a person’s demographic versus what that person actually is. This type of statistics is totally worthless because it tells nothing more than quantifying what is presupposed. This is something akin to finding a cache of burials of unknown race and sex, separating into categories of sex-based on height, then running statistical analyses and finding that males were taller than females.

Such an analysis does not tell the researcher anything since whatever category was used would simply reinforce the category that was assumed. It is simply quantifying perceived differences versus identifying actual differences. This is seen in studies of racial differences when groups are arbitrarily placed in broad categories like “African” when the various groups of Africa show higher variation than other groups that are separated based on geography (Keita & Kittles 1997: 539-540).

There is a significant problem in the use of genetics to determine racial divergence and subsequent racial affiliations. If races are considered subspecies, then races differ with respect to one or more identifiable traits of morphology by the 75 per cent rule, although “lesser cut-off points are sometimes used, but clearly it is meaningless to award a formal taxonomic designation to a population, or series of populations, of which only half or less of the members are distinctive” (Groves 1989: 7).

This is what goes on in folk taxonomy, except differences are quantified in scientific taxonomy. If this were the case, “black” and “white” would be perfectly acceptable racial categories. However, in that case, the concept of race loses all scientific meaning because it is all a matter of opinion of which trait to use, and where the boundaries between traits exist. The selection of different traits as representative of a population determines the racial affiliations and divergence times associated with that population, and determining which scheme most closely represents actual bio-history is difficult (Keita &Kittles; 1997: 537).

Morphological differences between two specific groups are the result of general reproductive isolation and selection for different allele frequencies based on chance and environmental and social pressures. Such differences are not always based on similar differences in genetic structure. “Racial thinking rests on the belief that visible human variation connotes fundamental deep differences within the species” (Keita & Kittles 1997: 534). The problem with the race concept is really the attempt to perpetuate previously assumed demarcations between groups on a fundamental level when no fundamental differences exist that warrant separation into distinct species.

Whether or not biological subspecies of human existed in some point in the past is irrelevant to the argument of the existence of “race” in modern populations, since there has been so much admixture of ancestral lines in today’s mobile society. Also, perceived differences between populations may be no more than differences in acclimatization to environmental pressure, which would bring into serious question the idea of subspecies as it is applied to humans. This is the reasoning behind the schemes of clinal subspecies themes.

According to the idea of clinal subspeciation, “most described subspecies are really samples taken at points along clines” (Groves 1989: 7). In this scheme, all subspecies that can be shown to be merely clinically arranged (regardless of the degree of difference) are rejected as true subspecies, and the only definitive subspecies are those that are reproductively isolated to a more or less total degree (Grove 1989: 7-8). Since gross morphological characteristics that differ considerably from deme to deme have been generally shown to be due to clinal adaptation in humans (Molnar 1998), human racial categories would have to be dropped.

This position is fundamentally flawed since most researchers already admit that human races do not exist in traditional terms (Keita & Kittles 1997: 536), and the question is really how the race concept is used to demarcate populations into discrete units of study. Attributing physical differences to clinal patterning does not answer the important question of what can be used in order to separate the six billion humans on this planet into units suitable for scientific study and analyses. The genetic analysis used to identify racial divergence as evidence of human population movement in the past is not addressed in clinal schemes, as variation in expressed traits that are affected by clinal patterning do not necessarily reflect actual significant differences in gene frequency.

Race as a unit of Study in Forensic Anthropology

George Gill states that the “usefulness of the race concept…does not necessarily relate so closely to its validity” (1998: 293). Forensic anthropologists that are investigating human remains that are skeletonized, incomplete, and/or decomposed are brought in to try to identify a potential victim. The determination that a particular set of remains is human, but maybe male or maybe female, maybe from a “black” or maybe from a “white” individual is not useful to law enforcement.

“People organize and classify, and it is these perceptions and observations that eventually form the personal records that law enforcement personnel obtain” (Gill 1998: 294). A forensic anthropologist must provide information that is useful in identifying remains, identifying a profile of what the person looked like, how old the person was, their sex, etc.

Forensic anthropologists do not have the luxury of scientific objectivity on the race issue, since “as long as society perceives human variation in terms of discrete races (sometimes even social races), then the forensic anthropologist must be prepared to articulate results of the analysis in those terms” (Gill 1998: 295). In other words, the researcher must present results that determine a probable race, sex, and build. Determining skin colour from small pieces of human bone seems incredulous, but some methods do have an accuracy level in the range of 90 per cent (Gill et al. 1988; Curran 1990; Pierce 1994) in separating one group from another.

Determination of race amounts to a probability game, the same one played when determining sex, and the same one used in all statistical procedures. When several characteristics are examined, a reliable indicator of racial affinity can be determined. How dark or light a person’s skin is not determined, but a high probability of “white” racial affinity indicates light skin; thus, race determination is presented as a high probability of accuracy, rather than precise statements (Klepinger & Giles 1998: 427).

This information is disseminated in this format of outmoded racial categories because the people using this information (law enforcement, juries, judges, etc.) are likely to have this perception of races. “In this way, the forensic physical anthropologist is providing information on ancestry that is comprehensible to law enforcement and the rest of the society in which we all must operate” (Gill 1998: 294). While it may be argued that the use of these racial categories merely serves to help continue their perseverance, a forensic anthropologist has little choice if he/she is going to provide useful information in a criminal investigation and/or prosecution.

The use of traditional racial categories is needed at present in the discipline of forensic anthropology. This use does not validate the categories; it is merely used that is required to present information that is useful, rather than incomprehensible. This is likely the reason why researchers that deny the existence of traditional racial classifications still tend to use races and/or ethnic groups as distinct breeding populations that can be delineated based on some arbitrary trait list (Keita & Kittles 1997: 536).

Until public perception changes with respect to racial categories, the concept of distinct races will not disappear. It is in human nature to categorize and compartmentalize what we perceive as differences between classes of objects and organisms, and that is unlikely to change. It is more probable that folk classifications of the race will subside as populations interbreed more throughout the future millennia, and differences between perceived categories become less. “Therefore, it would seem that whatever the rest of biological science may choose to do about race in the future, forensic physical anthropologists will likely continue to articulate results of their analyses within the framework of the traditional race concept, at least as long as society itself does this” (Gill 1998: 294).

Racial and Ethnic Groups as Units of Study

Whether or not a particular definition of “race” or definitions of particular races are accepted or are philosophically rejected as fallacious, the underlying important question raised by the issue of race is not answered. In a scientific sense, large populations need to be broken down into discrete units that allow for study. Traditional categories of race such as “black” and “white” were discussed above as viable discrimination in forensic anthropology due to the nature of the data being presented and the audience that is using that data. These same types of demarcations were also discussed as problematic and filled with inherent fallacies in their use in genetic studies of population origin. The question that will be addressed in this section of this paper is the use of race in demographic studies and inheritance patterns as a viable medium of information dissemination.

Demography is traditionally seen as those aspects that have to do with the gross population effects of total fertility, mortality, and migration (Smouse & Teitelbaum 1990: 213), but will also be used in this paper to encompass all aspects of society that are measurable statistics of sociocultural and socioeconomic realities. Demography is statistically relevant for a myriad of social reasons that range from religion to soft drinks, and as such is reflective of a society’s view of itself.

This places the same restraints on demography as is placed on forensic anthropology since the information used in demography must come from individuals who use a folk taxonomy and must be presented to individuals who use a similar folk taxonomy. Thus, the language of demography is determined by the population that is studied and the population that is doing the studying. “For demographers, a population can be defined simply as an aggregation of organisms occupying a given area” (Adams 1990: 8). Demography uses outmoded racial categories due to the fact that the general public still uses those outmoded racial categories, and the data that is used in demographic studies of “race” is often collected either firsthand or secondhand from individuals from the general public.

A definition of “race” that closely parallels the concept that is used by human populations (in general), and which is utilized by demographers is given by James King:

What constitutes a race is a matter of social definition. Whatever a group accepts as part of itself is within the pale; what it rejects is outside. Acceptance and rejection are not absolute but can exist in various degrees (1981: 155).

This definition is in obvious opposition to a biological concept of race. A similar dichotomy of classification is that between “sex” and “gender”. Sex is a biologically determined attribute that is discrete and exists in intermediaries only in rare instances of biological errors that usually include functional reproductive absence. Gender is a socially derived trait that is reflective of an individual’s view of himself/herself (itself?) and society’s view of that individual. Rather than using the same term to describe one’s group placement in sociocultural terms and to describe ancestry, the term “ethnic group” would be a better fit in the description of most demographic categories (percentages of ancestry may need “race” statistics versus ethnic group statistics, but if there are no strict genetic differences between “races”, race statistics should not be needed).

The same cannot be said for the use of race in forensic anthropology since the identification of ancestry is dependent on the morphological differences between populations that are a reflection of different patterns of variation in gene frequency. While terms such as “Hispanic”, “African-American”, “Euroamerican”, and “Native American” may reflect sociocultural distinctions that are meaningful to demographers, their ambiguity and failure to indicate distinguishable traits makes them useless to a forensic anthropologist (Gill 1998: 295-296). Therefore, while traditional classifications are useful to demography and are required by the nature of the data, the term “race” should be replaced with the term “ethnic group”, which does not necessarily denote biological difference that is independent of an individual’s perception of himself/herself and/or society’s perception of that individual.

The race concept as a unit of study is also important to in the study of inheritance, as it refers to disease and its treatment and prevention. Many diseases are heritable, and many of these diseases are more or less prevalent within certain populations and their derivatives than in others. Diseases such as sickle-cell anaemia, cystic fibrosis, and phenylketonuria are caused when an individual carries a double recessive for the gene(s) that causes the disease (Molnar 1998: 128).

Many genetic diseases can be controlled if detected early enough (e.g., phenylketonuria can be controlled with a proper diet), and many are more common in some ancestral lines than others. For instance, sickle-cell anaemia is far more prevalent in those persons of African descent since there is a heterozygous advantage for the trait in order to help fight susceptibility to malaria (Molnar 1998: 152-158). In the dissemination of information about susceptibility to certain heritable conditions and early detection programs to high-risk individuals, race can be a useful term. In this use of the concept of race, categorization of distinct groups in order to determine ancestral migration patterns and origin patterns is not necessary. Only the use of the term to describe those who are susceptible according to their biological ancestry is needed. Hence, it is not necessary to use definitive categories such as “black”, “white”, etc. since these terms are not specific enough in most cases anyway.

The term “white” is not definitive enough when a condition is inherent to an individual ethnic group, for example, the high incidence of Tay-Sachs in Ashkenazic Jewish populations (Molnar 1998: 137). Other broad categories such as “ethnic group” are also not sufficient since a Central American “mestizo” or “mestiza” may not be made aware of his/her risk for sickle-cell anaemia due to misconceptions of ancestral associations of “Hispanics” (Keita & Kittles 1997: 535). Thus, the race concept (populations showing some discrete trait) is a proper unit for such use and would include specific ancestral populations at risk and ethnic groups that may also be at risk if such a population was in the individual’s genealogy. The sociocultural concept of “ethnic group” is, of itself, not sufficient to be an effective unit of study for heritable disorders.


This paper has attempted to show the uses of the race concept as it applies to humans in both biological and sociocultural terms. The concept is theoretically sound in abstract but becomes problematic when applied to modern human populations. The use of modern populations to trace ancestral human divergence patterns is questionable, as is the idea that there are separate biological subspecies of human. The concept is useful in various fields that must gather data and disseminate information from and to individuals who use a traditional concept of race, even if it is scientifically unsound and somewhat distasteful.

For other applications, ethnicity and distinct cultural, social, and biologically distinct (distinct in respect to a specific allele) terms may be preferable and more useful than traditional race concepts. The bottom line is, as long as there are distinguishable differences between individuals that are the reflection of different variation in different geographical and/or cultural populations, the race concept will be prevalent and must be dealt with in the most efficient manner as possible.


Adams, Julian. 1990. “Introduction: Genetics and Demography and Historical Information.” In Convergent Issues in Genetics and Demography, ed. by Julian Adams, David A. Lam, Albert I Hermalin, and Peter E. Smouse. pp. 3-13. New York: Oxford University Press, Inc.

Clark, Stephen R. L. 1994. “Is Humanity a Natural Kind?” In What is an Animal? ed. by Tim Ingold. pp. 17-34. London: Routledge.

Curran, B. K. 1990. “The Application of Measures of Midfacial Projection for Racial Classification.” In Skeletal Attribution of Race, ed. by G. W. Gill and S. Rhine. pp. 55-57. Maxwell Museum of Anthropology, Anthropology Papers, No. 4.

Gill, George W. 1998. “Craniofacial Criteria in the Skeletal Attribution of Race.” In Forensic Osteology: Advances in the Identification of Human Remains, ed. by Kathleen J. Reichs. pp. 293-317. Springfield: Charles C. Thomas Publisher, Ltd.

Gill, G. W., S. S. Hughes, S. M. Bennett, and B. M Gilbert. 1988. “Racial Identification from the Midfacial Skeleton with Special Reference to American Indians and Whites.” In Journal of Forensic Sciences, vol. 33. pp. 92-99.

Groves, Colin P. 1989. A Theory of Human and Primate Evolution. Oxford: Oxford University Press.

Keita, S. O. Y. and Rick A. Kittles. 1997. “The persistence of Racial Thinking and the Myth of Racial Divergence.” In American Anthropologist, vol. 99, no. 3. pp. 534-544.

King, James C. 1981. The Biology of Race. Berkeley: University of California Press.

Klepinger, Linda and Eugene Giles. 1998. “Clarification or Confusion: Statistical Interpretation in Forensic Anthropology.” In Forensic Osteology: Advances in the Identification of Human Remains, ed. by Kathleen J. Reichs. pp. 427-440. Springfield: Charles C. Thomas Publisher, Ltd.

Molnar, Stephen. 1998. Human Variation: Races, Types, and Ethnic Groups. Upper Saddle River: Prentice-Hall, Inc.

Pierce, L. C. 1994. “Further Testing of the Gill Method of Racial Identification.” Paper presented at the 46th Annual Meeting of the American Academy of Forensic Sciences, San Antonio, Texas.

Smouse, Peter E. and Michael S. Teitelbaum. 1990. “Genetics, Demography, and Epidemiology.” In Convergent issues in Genetics and Demography, ed. by Julian Adams, David A. Lam, Albert I Hermalin, and Peter E. Smouse. pp. 209-217. New York: Oxford University Press, Inc.