Accessibility links

The idea of ‘being a realist’ is rather slippery. Realism in everyday parlance is taken for granted as a sort of common sense perspicacity. We can all relate to being told to ‘get real,’ or ‘be realistic,’ but what that actually means, when you stop and think about it, could be all sorts of different things to different people.

In academic circles, realism is even more slippery, with a cascade of categories you need a lot of patience to pick a path through. At the top: philosophical realism, scientific realism, political realism, artistic realism. Beneath each of those, more specifics: critical realism, transcendental realism, naïve realism, so on and so on. There’s a veritable spaghetti junction of realism out there.

In my research training, I studied ontology and epistemology – basically what there is to know, and how we know it. Lots of people think this stuff is boring, but for me, it is in these questions that the real power to effect change lies. In the world of academic research, the paradigm within which you conduct your work is probably the number one defining influence on what you do, even if you don't say so yourself.

For a long time there was a tug of war between quantitative and qualitative paradigms; positivism versus interpretivism. It doesn’t get much popular press, but in the relatively recent past, profoundly significant advances in the philosophy of science have brought us to a point where we have a real opportunity to transform our world by adopting a different stance to what we know about it. And, it’s realism.

only realism can save us from the state of crisis we are in

Personally, I’m hugely influenced by Roy Bhaskar’s Critical Realism, but I have one major criticism, which is that it can be bloody difficult to understand. And, what’s the use in world-changing philosophy, if it isn’t going to make sense to enough people for it to take hold? I find Ray Pawson’s version of realism a little easier to digest, probably because his work is so firmly rooted in in the type of real world research I've done myself, like evaluating educational and public health interventions.

So, it’s rather bold, perhaps, but I’ll risk sounding ridiculous and assert that only realism can save us from the state of crisis we are in. Adam Lent wrote yesterday about ethical revolutions. Whether we are indeed on the cusp of a time of ethical revolution and not further descent into crisis depends on changes happening on many levels. So many of our power structures are built on evidence bases which have non-realist foundations which by their nature try to eliminate complexity and identify clean causal relationships. With this explicit rejection of the reality of the complex world we occupy, we’re shooting ourselves repeatedly in the feet.

With this explicit rejection of the reality of the complex world we occupy, we’re shooting ourselves repeatedly in the feet.

I’ll try to illustrate. In public services, we are constantly trying to do things within complex social systems, and in order to know whether the things we’re doing are effective, we need to measure and evaluate. With something like an educational intervention with the aim of, say, reducing teenage pregnancy, its effectiveness might be measured by counting the number of teenagers who become pregnant after receiving the intervention and comparing that with the number of pregnancies amongst those who didn’t get the lesson.

The research will try to ‘control’ out all of the social factors which influence whether or not teenagers become pregnant, and make a claim as to the causal relationship between the intervention and the outcome. We want to be able to conclude,’ if the kids have the sex education class, they won’t get pregnant’. But of course, the world is messier than that. The best sex education in the world doesn’t stop condoms breaking every now and then, or emergency contraception failing.

The positivist view of the world relies on the exclusion of critical but unpredictable factors. This type of research might lead to nice, neatly analysable statistics, but whether it’s useful or not, or answers the most important questions is another matter. Realist evaluation, by contrast, assumes that in order to infer a causal relationship between two events (i.e. sex education and teenage pregnancy), it is necessary to understand the underlying mechanism that connects them AND the context in which the relationship occurs. Rather than asking simply ‘what works?’, a realist evaluator asks ‘what is it about this intervention that works, for whom, in what circumstances?’ More complex, yes, but more useful also.

Even if you’re with me this far, you might well be thinking, ‘fair enough, but how is this going to save the world, exactly?’ I have a feeling it’s going to take more than blog post for me to fully unpack this and show why it matters so much. I promise I shall try to avoid becoming terribly dull…

4 Comments

Join the discussion

Please login to post a comment or reply.

Don't have an account? Click here to register.