Kołakowski's Definition of Positivism

Within the social sciences, ‘positivist’ has become something of an almost-empty signifier. It’s sometimes used to simply mean ‘quantitative researcher’, as often by quantitative researchers themselves. Once or twice, I’ve heard it be claimed that positivists believe they are studying unchanging laws of reality – which would be a fairly strange assumption for a social scientist to make. More than once or twice I’ve heard ‘positivist’ be used as a term of abuse against quantitative researchers.

I have lying around Leszek Kołakowski’s old Pelican book Positivist Philosophy: From Hume to the Vienna Circle and would like to read it in full. But I have already read the inaugural chapter, within which he offers four rules which broadly define positivism:

1. The rule of phenomenalism: that there is no difference between ‘essence’ and ‘phenomenon’. There are no occult realities beyond what is manifested in experience.
2. The rule of nominalism: insights formulated in general terms cannot have any referents other than concrete individual objects. The study of the behaviour of abstractions (or ‘ideal situations’) such as ‘triangle’ is not the study of Platonic forms with a reality of their own, but only helps us in understanding real entities which approximate the abstractions more or less closely. Kołakowski calls this a consequence of the first rule.
3. The rule that refuses to call value judgements and normative statements knowledge. This is a consequence of phenomenalism and nominalism, because things like ‘good’, ‘evil’, ‘beauty’, ‘noble’ and so on are abstractions or generalisations from experience rather than things in our experience themselves.
4. Belief in the essential unity of the scientific method. Most commonly, this shows up as the belief that the methods for acquiring valid knowledge are the same in all spheres of experience.

I think this is a much better definition than whatever else I’ve come across. There is no requirement in this list that reality be considered unchanging, so it is inclusive both of positivists who thought so (Comte) and positivists who did not (the Vienna Circle, for instance, were not particularly interested in the question). You can see a point of difference between positivists and, say, some normative or critical theorists on point 3. But it’s also very clear that to be a quantitative social scientist is not necessarily to be a positivist.

Some violations of positivism I would anticipate finding among quantitative social scientists are:

1. Against rules 1 and 2, some quantitative social scientists will subscribe to the Platonic reality of numbers, or to realism about some abstraction (e.g. of ‘democracy’)
2. Against rule 3, some quantitative social scientists will see value judgements and normative statements as a kind of knowledge, albeit one distinct from empirical knowledge
3. Against rule 3, some quantitative social scientists will disagree with the idea that empirical knowledge and value judgements can be neatly separated
4. Against rule 4, some quantitative social scientists will disagree with the idea that there is or can be a unified scientific method

I expect that there will be a plurality of views among quantitative social scientists. Violations 2 and 3 implicitly disagree with each other in the way I have defined them. I admit that I myself am sympathetic to positivism, and could accurately be called a fellow traveller. I think it’s regrettable that the anti-metaphysical project of the Vienna Circle is largely dead among professional philosophers. The project had real political stakes, motivated for many members by an opposition to clerical rule and fascism which is no less relevant today. It also wouldn’t hurt to give advocates of nonsense ideas like panpsychism a much harder time. Regardless of our sympathies, we will have much better conversations about the philosophical foundations of the social sciences if we define our terms rigorously.