Wednesday, July 4, 2012

Making the distinction between science and non-science

To: Yasuko S.
From: Geoffrey Klempner
Subject: Making the distinction between science and non-science
Date: 14 December 2007 13:05

Dear Yasuko,

Thank you for your email of 8 December, with your University of London Methodology essay in response to the question, ''Some sciences differ from some non-sciences. But there is no way of making a general distinction between science and non-science.' - Discuss.'

This is a very good essay, clearly and succinctly argued, which stays on the track of the question.

To start off with, a couple of points:

As an example of the way that good science uses auxiliary hypotheses, you could have given the famous example of Newton, whose predicted orbits of the planets almost, but not quite, corresponded to what was observed through the telescope. He concluded, correctly, that there must be a factor which had not been taken into account -- his auxiliary hypothesis -- rather than regarding the disparate observation as a falsification of his theory.

I have heard it argued that Popper's falsifiability criterion only works because he restricts science to the making of universal generalizations. Suppose I have a theory which predicts that there is an astronomical body with property X which has never been observed in any astronomical body before. This theory cannot be conclusively falsified but it can be conclusively verified -- by the observation of an astronomical body with property X.

In order to defend his demarcation principle, Popper would have to argue that the theory in question can be recast into the form of a falsifiable universal generalization. But why should it be possible to interpret every scientific theory in this way? Why can't there be scientific theories which make existence claims rather than assert generalizations? Popper might answer that an existence claim doesn't explain anything, whereas a universal generalization does. But now we are adding an additional requirement for demarcating science from non-science: a scientific theory must 'explain' something.

The main thing I would say is that the assertion, 'some sciences differ from some non-sciences' is very weak. In your essay, you mount a good case for rejecting the claim that there is a 'way of making a general distinction between science and non-science.' But you also hint that you think that it is possible to make a stronger claim than merely that some sciences differ from some non-sciences. You approvingly quote the claim that 'there may be not one distinction but rather a variety of distinctions'.

Let's look more closely at what this means. The statement that some sciences differ from some non-sciences is consistent with the claim that some sciences don't differ from some non-sciences. You don't want to say that! Whereas, the claim that there is a variety of distinctions between sciences and non-sciences suggests a flexible but effective way of separating every activity which we intuitively regard as 'science' from every activity which we intuitively regard as 'non-science'. Sometimes you apply criterion A, sometimes B, sometimes C, or sometimes in combination, but in every case there is a way to justify our intuitive sense of what a 'science' is.

This is a point that you need to emphasize, because this is where you are showing that you have an original take on the question. You don't agree that what the question proposes are the only two alternatives. There is a third alternative, which the question doesn't take into consideration.

Consistent with this theme, it would be appropriate to give more examples -- perhaps more controversial examples -- of activities which some regard as 'science' and others regard as 'pseudo-science'. Here is just one example. In my activities as a philosopher of business I have come across an increasing number of practitioners in the field of business training who claim expertise in 'emotional intelligence'. Apparently, there is a growing science of emotional intelligence. There are tests you can give someone to determine their emotional intelligence quotient. You can predict how persons of high EIQ will behave in stressful situations, compared with persons of lower EIQ, and so on.

Is this a science? Or a hodge-podge of rough and ready psychological generalizations? Or pure hokum? My own view would be that practitioners of EI are trespassing into the field of philosophy, specifically moral philosophy, even though they do not realize this.

Another issue which is raised by the question is, Why are we so concerned to make a distinction between science and non-science? What hangs on this distinction? Suppose I am a practising psychotherapist and I am perfectly happy with saying that what I do is not a 'science'. Why should that be a problem? -- But in that case, why is it that I still feel sorry for people who fall for the claims of astrology?

You could say something about the role of a distinction between science and non-science in the wider context of epistemology, and the question of what it is rational to believe. There are some things which we believe on rational grounds, which are not scientific grounds. Some non-sciences can be valuable sources of knowledge while others are worthless. We need some way to justify our intuitive sense that there is a distinction to be made between the rational and the irrational, just as in the case of the distinction between science and non-science.

All the best,

Geoffrey