Human Osteology & Diversity: The question of the validity of the concept of race

In our third year Human Osteology and Diversity module, students explore the history of the discipline and its socio-political influences related to the study of racial diversity. This year, our featured essay is by Lauryn Thomas.
Lauryn Thomas
Lauryn’s bio: In the final year of my anthropology degree I’ve been particularly interested in medical anthropology and the diversity (and similarity) across the human species. My world view has been transformed and I hope to deepen my knowledge in postgraduate study in the future.

In his paper ‘On the Non-Existence of Human Races’ (1962), Livingstone argues that intraspecific variation in humans has resulted in breeding populations having differing gene frequencies, and that this does not correspond to racial categories. So-called ‘racial characteristics’ which were envisioned as the qualifying criteria of a specific race, are often found in geographically distant populations. Furthermore, Livingstone states that there is great difficulty deciding which characteristics should be taken into account in order to create racial categories, as the number of characteristics used to categorise races is increased, more exceptions to the rule arise. Livingstone also critiques the fact that the role of natural selection is largely ignored by proponents of the concept of race as an explanation for genetic variation; he instead proposes that ‘cline and morphism’ (1962: 279) provide a better explanation for this. By plotting graphs based on the gene frequencies of different populations, the factors influencing this (mutation, genetic drift, gene flow and natural selection), as well as the way they interact, can be determined.

Dobzhansky (1962) responds to Livingstone agreeing that the dominant concept of race is faulty, in that it’s impossible to categorise humans into ‘discrete units’ (1962: 279), and that race should not be used as an explanation for biological variation. However, Dobzhansky presents an alternative definition of race, rather than supporting Livingstone’s calls to eradicate the concept.  Dobzhansky defines race as being an open category in which individuals belonging to differing races have the ability to interbreed, rather than a discrete category like species. Dobzhansky views natural selection as the basis of genetic variation (as certain genes are favoured in certain environments) which then gives rise to race. He argues that the fact that breeding populations have gene frequencies which are unique to them, means that racial differences can be verified scientifically. To reiterate, the classification of races should not be based on ‘racial characteristics’ themselves, but on the gene frequencies of the race. Dobzhansky does not view clines corresponding to the geographic distribution of certain genes as something which disproves his theory of race, but supports it.

The data of Long et al. (2009) supports Livingstone’s notion of genotypic gradients as an effective way of describing the genetic distribution of different human populations. The genetic diversity present in non-African populations represents a sub-section of the genetic diversity present in the African continent. Long et al. propose a new basis for the categorisation of populations other than diversity between populations, believing their data to demonstrate ancestry and the process of natural selection, whilst revealing the extent of genetic diversity, just as Livingstone (1962) proclaimed genotypic clines as a useful starting point for investigating the causes of genetic variation. The paper also calls attention to the major issues of how races are categorised, demonstrating that Dobzhansky’s (1962) concept of race is redundant due to the fact that the genetic diversity between populations does not adhere to his view of members of a population being more related to one another than other populations. This is shown when Long et al. highlight the possibility of the Sub-Saharan African population not being considered as a group, owing to the fact that the genetic diversity present among Africans is also found in non-African populations. Additionally, Yu et al. (2002) found that there were greater genetic differences between two African DNA samples, than there was between an African and European sample.

Yudell et al. (2016) praise Dobzhansky’s attempt to alter the definition of race away from the pervasively held view of it being a term which denotes hierarchy among different populations. However, they believe that simply changing the definition of the race concept is insufficient due to the negative social and historical context ingrained in the term; as an alternative Yudell et al. support the use of words such as, ‘ancestry’ (p. 565) to categorise groups of people. They point to present day issues in medicine and epidemiology that have arisen as result of the use of the discontinuous concept of race to classify people, for example, the large amount of overlap in genes between populations is often ignored; this has resulted in blood diseases being misdiagnosed as the patient does not belong to the race predominantly associated with the disease: ‘the identification of sickle-cell as a “Black” disease’ (p. 565). Ancestry as a replacement for race would entail taking into account not only an individual’s geographic ancestry, but their ‘culture, socioeconomic status, and language, among other variables, depending on the questions being addressed’ (p. 565), in order to create clearly defined classifications for human populations which account for genetic diversity.

Kaplan and Winther (2014) examine Livingstone and Dobzhansky’s correspondence, stating that the debate was futile due to the fact that both were using different definitions of race. Livingstone had envisioned race as distinct categories which exaggerated the genetic differences between human populations, whilst Dobzhansky acknowledged the existence of genotypic clines among humans. They argue that greater credence must be given to Dobzhansky’s calls to recognise the biological variation between populations and ‘whether, as a matter of biological practice, we ought to recognize human subpopulations as legitimate biological entities’ (2014: 1046).

Rosenberg et al. (2002) carried out a study of people from various populations and were able to highlight the presence of genetic clusters which correspond to geographic areas, without knowledge of the participants’ ancestry. Despite the genetic variation between populations being measured only at around 3-5% (versus a 93-95% variation within populations), the study was still able to identify genotypic categories, just as Dobzhansky (1962) posited that differences between races can be confirmed. Moreover, Rosenberg et al. (2002) attribute the majority of human genetic diversity to differences between the genetic frequencies of different populations due to the wide dispersion of genotypes across populations, just as Dobzhansky’s (1962) basis for categorising populations was their respective gene frequencies. Rosenberg et al. point to a potential use of the data in determining past migrations of humans.

Relethford (2009) describes the clinal nature of genetic variation, despite certain geographic areas displaying a high concentration of certain traits, such as skin colour (mainly due to the impact natural selection). He investigates whether phenotypic traits have the potential to inform on the applicability of biological race. In the case of skin colour, it is not deemed possible to create objective clusters due to the distribution being displayed as a gradient; this is supported in Livingstone’s (1962) assertion that gene frequencies across populations can be plotted revealing a cline. Nonetheless, Relethford recognises the usefulness of race as ‘a culturally constructed label that crudely and imprecisely describes real variation.’ (p. 20), definitions of races are interdependent, in that they rely on their relation to one another for their definition. The conception of race is a response to the fact that certain traits are observed in some populations more than others; the conversion of a continuous gradient into discontinuous categories is seen in everyday language, e.g. short, medium and tall. Relethford poses a question similar to Dobzhansky (1962) (should categories should be recognised and if so how many?), asking whether there is any worth in creating racial groups and if they are used, in what circumstances are highly specialised versus generalised classifications best used? Attention is also brought to the loaded negative connotations of race and if the term itself is best replaced.

To conclude, in spite of Dobzhansky’s assertion that race is something which is ‘objectively ascertainable’ (1962: 279) a possible point of criticism is that his definition has the potential to fail in identifying the racial group an individual belongs, owing to the distribution of genes among races. Dobzhansky (1970) later elaborates by defining a race as a group of individuals that share a gene pool as a result of generally interbreeding with one another, yet races should not be recognised as individuals or certain gene but as the group itself. Despite his argument, how can race be justified as a biological classification if its definition, as Relethford (2009) stated, is reliant on the acceptance that there are difficulties categorising certain individuals?

Livingstone’s (1962) alternative categorisation of human populations is based on the concept of breeding populations while taking into account vehicles of genetic variation, through this he sought to explain observable phenomena – similar to the use of the term ‘ancestry’ (p. 565) by Yudell et al. (2016) as a holistic approach to classifying populations. However, his intolerance towards the term race ignores the fact that different conceptions (Relethford, 2009; Livingstone and Dobzhansky, 1962) can make use of the same methods he has used in establishing breeding populations.

The views of Livingstone and Dobzhansky appear to be diametrically opposed at first glance, however, they actually agree on several points such as the existence of genotypic clines and gene frequency as the identifying factor if a population. The main point of contention between them seems to be Dobzhansky’s use of the term ‘race’; this criticism is also seen in later discussions of race. Livingstone’s failure to take into account and properly respond to Dobzhansky’s modified race definition shows the effect language and the historical context of words have on understanding.


  • Dobzhansky, T. (1970). Genetics of the evolutionary process. New York: Columbia University Press.
  • Kaplan, J. and Winther, R. (2014). Realism, Antirealism, and Conventionalism about Race. Philosophy of Science, 81(5), pp.1039-1052.
  • Livingstone FB and Dobzhansky T 1962. On the non-existence of human races. Current Anthropology 3:279-81.
  • Long, J. and Kittles, R. (2009). Human Genetic Diversity and the Nonexistence of Biological Races. Human Biology, 81(5-6), pp.777-798.
  • Long, J. C., Li J, and Healy ME 2009. Human DNA sequences: more variation and less race. American Journal of Physical Anthropology 139: 23-34.
  • Relethford, J. H., 2009. Race and global patterns of phenotypic variation. American Journal of Physical Anthropology 139: 16-22.
  • Rosenberg, N., Pritchard, J., Weber, J., Cann, H., Kidd, K., Zhivotovsky, L. and Feldman, M. (2002). Genetic Structure of Human Populations. Science, 298(5602), pp.2381-2385.
  • Yu, N., Chen, F., Ota, S., Jorde, L., Pamilo, P., Patthy, L., Ramsay, M., Jenkins, T., Shyue, S. and Li, W. (2002). Larger genetic differences within Africans than between Africans and Eurasians. Genetics, [online] 161(1), pp.269-274.
  • Yudell, M., Roberts, D., DeSalle, R. and Tishkoff, S. (2016). Taking race out of human genetics. Science, 351(6273), pp.564-565.

Leave a Reply

Your email address will not be published.