Caste Wars: The morality of treating individuals as though they are members of groups

Edmonds, David (2003). Caste Wars: The morality of treating individuals as though they are members of groups. PhD thesis The Open University.



This thesis is about individuals and groups - and about judgements made of individuals because of their membership of groups. I argue that the concept of caste can illuminate our understanding of discrimination.

Chapter I adumbrates the properties of a group. Chapters II draws a distinction between two types of statistical discrimination: between extrapolations based on characteristics like race, sex or eye colour and those based on past behaviour. But within the former category we feel more strongly about discrimination based on sex than discrimination based on eye-colour. Chapters III explains why this is so. I attempt to show that the concept of caste is essential for making sense of our intuitions.

Race and sex are not supposed to be relevant in a meritocracy - potential employees and university applicants are supposed to be assessed on their individual worth. And yet there are cases where an individual's worth to a company may be a function of a characteristic like skin colour - such as when customers are racist and would rather not be served by a black person. Chapter IV attempts to reconcile this tension within merit. If we wish to break down caste, I argue, then far from pandering to the racist attitudes of the customers, the employer may have an added reason to employ somebody who is black. This brings us to Chapter V, and some reflections on how caste can illuminate the impassioned debate on affirmative action.

But should we wish to break down caste? What is wrong with a society being rigidly split? Chapter VI offers a tentative answer, in the liberal tradition. The answer has something to do with autonomy. Chapters VII and VIII then examine two final areas where groups are relevant, and where caste can once again shed light: voting rights, and animal rights.

Viewing alternatives

Download history


Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions