Science and Truth

Truth, or at least knowledge, can be thought of as ideological. There is knowledge, that is, justified true beliefs that were obvious to our ancestors but are anything but to us. The position of women in society is one of them, although the historical record shows that that seems not to have been quite as robust as the (male) commentators of the time would have claimed.

There are also claims that science is ideological. After all, the ‘scientific revolution’ was a Western innovation and was dominated by males. A certain quantity of scepticism is therefore aimed at the claims of science, with suggestions that it is inherently sexist, racist and so on. I am not expert enough to comment on the sexism and racism that is (or was: let us hope that things are improving) extant in the history of science, but it does sometimes seem that those women who did achieve anything in the sciences are lauded simply by being rare.

Ideological claims about science go beyond the accusations of modern cultural suspicions. Science can start to claim hegemony overall knowledge, and at that point, it either starts to look a bit silly or becomes dangerous. An example of this is some misconceived ‘experiments’ designed to determine whether prayer (presumably to the Christian deity, but I forget the details) ‘works’. This entails a great deal of difficulty and a large number of assumptions about prayer, the deity in question and what on earth a prayer working would look like. Some things, it seems, cannot be subjected to the empirical study the natural sciences are based on.

In fact, consider a few other options. Try designing an experiment to see if your partner loves you. Can this be reduced to a set of variables which can be controlled against? I very much doubt if any such experiment could be conducted and, of course, if you did conduct such an experiment and got a positive result, you would then have to conduct another one to see if, after the experiment, they still loved you.

As another issue, consider the experiments required to determine the truth of ‘I shall meet you under the clock tomorrow at noon.’ In advance, there are few empirical activities you could undertake to establish the truth of this statement. You could, I suppose, evaluate the history of the speaker, rating them for accuracy, reliability and so on. You could enforce upon them a lie detector test. None of these would seem to be as accurate as simply believing them to be telling the truth and waiting for them to turn up tomorrow, under the clock at midday.

It is not, therefore, simply in religious behaviour that the claims of some scientists to be the sole source of knowledge start to look silly. Human relationships, on the whole, cannot be so evaluated. As can be imagined, even science itself is a product of a set of human relationships, and these are, in general, not measured. We are back to the question of ideology in science again. By assumption, and simply to get anything working at all, some stuff has to be excluded – in this case, the humanity of the observers.

The idea of modern science is that experiments can be repeated and, when that is done by a different set of people, they will get the same result. This is quite hard to achieve, due to constraints of time, money, personnel and equipment, but it can be done. Early in my career as a scientist I came up with an unexpected result and published it, without any explanation of how it came about (because I did not know) although it was supported by one set of calculations (although not by two others). Another research group did happen to have the time, equipment and interest to try it out. Doubtless, they also had a degree of scepticism about my results as well. They reported their results at a conference which I attended and confirmed them.

The example is of something which is assumed in modern science but often does not happen. As with the coherence theory of truth, if the new result coheres with the expectation often there is not much interest or confirmation. If the result defies the rest of the subject, then researchers around the world start checking up on it. Sometimes flaws are found; sometimes the new results or ideas are confirmed. That is how science develops.

Bayes’ theorem is in use here. The probability of a hypothesis given a set of evidence is regarded as the probability of the hypothesis without the evidence multiplied by the probability of the evidence being true given the hypothesis, divided by the probability of the evidence being true. The more evidence there is given the hypothesis, the bigger the probability of the hypothesis being true (it is easier to see this in the formula, and hard to type about; see https://plato.stanford.edu/entries/bayes-theorem/).

P(h|e)=P(h)P(e|h)/P(e)

This is of course precisely what happened in the case of my observation: someone repeated the experiment and obtained the same result. The probability of the result being true was therefore increased, and the probability of the calculation which gave the same answer as my result being true was raised as a result. The result is in the literature, although so far as I know it has not been fully explained as yet.

There are, naturally, problems with Bayes’ theorem, but I do not think we need to go into them. What I have described is the normal process of science, but I dare say you have spotted a problem. We are talking in a rather loose way, about the theorems and experiments being ‘true’. The point here is that the science can never make the probability of the hypothesis given the evidence one; the hypothesis can never be said to be absolutely true. There is no proof in science; only in mathematics can things be said to be proved and even then there are problems.

Normally, this lack of proof does not matter one jot. There is sufficient experience, other evidence and coherence with other results and theory around to enable us to make reasonable judgments about what is true and what is not, even if the truth is ‘overwhelmingly probable’ rather than ‘proven’. But often that is not how the rhetoric of science works – words like ‘proven’ and ‘shown’ are bandied around in the media and some scientists fall for their own claims about the power of science.