October 2009

Is discriminating on the basis of performance acceptable?

Generally in today’s liberal democracies we strive not to discriminate between people on the basis of things which they can’t change or have little hope of changing such as their ethnicity, gender or  sexual orientation.

This has not always been the case. For a long time it was seen as perfectly acceptable to discriminate between people on the basis of these, and other, attributes which they couldn’t change. For most people these changes in current practice would indicate some moral progress in the world.

However it does raise the question: is any discrimination now acceptable? Broadly, while things people can’t change are seen as unacceptable grounds  for discrimination, things over which they have control would seem more acceptable. For example, we discriminate against people who commit crimes by locking them away in prison because their criminal behaviour is seen as something which can be changed (among other reasons for locking them up). Discriminating between clinicians or services on the basis of performance would seem to fall into this changeable category. So why we might ask is there a strong reaction against such discrimination?

I think the problem has some philosophical origins which I would like to tease out in the rest of this column.

The main problem I think is captured by an issue first discussed by Hume, the famous Scottish philosopher from the 18th century, in his treatise. In that work Hume made a point of distinguishing between an is and an ought. What Hume meant by this distinction was that because something is the case in the world it doesn’t mean that it ought to follow that we should do something about it: in other words facts do not commit us to moral actions.

For example, it is a fact that people have different heights but that does not mean we ought to do something about it in the sense of discriminating against people who are short or tall. The ought doesn’t follow from an is: all that follows logically from an is is another is. There will always be a need for a moral decision which weighs up the facts independently.

A further distinct development of this idea arose with the Romantic Movement in the late 18th century when nature was seen as somehow possessing properties which gave rise to goodness and moral worth in people. Anything seen as natural was seen as something we ought to pursue. These romantic notions toward nature, with  the onset of Darwinian developments later in the 19th century, gave rise to social Darwinian ideas, first discussed by Herbert Spencer, that social inequalities are natural and since the natural is equated with what is good, inequality is seen as somehow being good.

The final part of this philosophical development is the work of G E Moore in the Principia ethica in which he undermined the so called ‘naturalistic fallacy’ by showing that what is natural is not synonymous with what is good.

What does this have to do with outcome measurement? I think the is/ought dichotomy (or fact/value distinction as it is often called) and the naturalistic fallacy are still at play in the way people view some activities such as outcome measurement results.

The point which needs emphasising is that following Moore’s undermining of the naturalistic fallacy, the fact that something is the case doesn’t necessarily mean we need to do something about it in the sense of recognising it as morally good. The fact that a clinician or service has better outcome performance results than another doesn’t mean that we have to do something about it. The is doesn’t commit us to an ought.

Clearly it would be disingenuous to say that performance results will not potentially influence decisions taken about  certain practitioners or services, all we are saying is that  there is a separate moral decision that needs to be taken about how we use the factual information we have obtained.

To return to the initial question: is it acceptable to discriminate on the basis of performance? Yes, but the decision to discriminate is a moral one which needs to be separated from the question of whether services or individuals are performing better.

The moral decision will involve looking at the performance results and other information and trying to make the best decision under the circumstances.  

Page last updated: 6 October 2009