Are you a positivist or an idealist about social programmes?


Genevieve Maitland Hudson


If I ever get a tattoo it will probably be a chunk of text from a book by Ian Hacking. I think it would be suitable, tattoos being permanent and Hacking’s texts so endlessly applicable – for instance to the current dynamics at work in philanthropy, charity and social programmes.

1983’s Representing and Intervening is supposed to be an introduction to topics in the philosophy of natural science but, being as versatile as a classic pair of Levis, it does an excellent job of shedding light on the battle that is pitching reformers against traditionalists in this sector. It does that by going over old philosophical ground and exploring the kinds of ideas that differentiate positivists from idealists.

These days no one really likes to be called a positivist, but that doesn’t mean the cast of mind and many of the propositions aren’t still around, in much the same way that no one much likes being called a Blairite but there Blair still is, thriving.

To boil down a complex area of thought, positivism states that ‘propositions count as true-or-false if and only if there is a way of settling their truth values’. Positivists also, broadly, take probability over causality, dislike unobservable entities (things like quarks or empathy) and distrust explanations on the basis that any set of facts can always be explained in a number of different ways (theories are ‘underdetermined’). The best a positivist can hope for is a theory that is ‘empirically adequate’: it pretty much fits the facts.

You could boil all this down even further, to a trickle, and say positivists like:

  • Observability
  • Probability
  • A tight fit between phenomena and explanations

Idealists are a bolder breed, prepared to adhere to much more expansive beliefs with much less grounding of those beliefs in observable phenomena. There are different kinds of idealism, but one of them in particular works both as a good parallel to positivism and as an insight into the way that some kinds of social programmes are run. Hacking takes it from Hilary Putnam and summarizes it like this:

Truth in an internalist view is […] some sort of ideal coherence of our beliefs with each other and with our experiences as those experiences are represented in our belief system.’

You can dig much deeper into these kinds of definitions, but for my purposes here they are intended only to suggest two different kinds of attitude to delivering social programmes: the first looks outward for confirmation of effectiveness, the second looks inward.

Hacking (circa 1983) thinks that the philosophy of science focuses too much on theory and doesn’t give sufficient attention to intervening. We might be guilty of the opposite problem. We think rather a lot about intervening and forget to look at our theories and how they determine what we do and how we go about assessing it.

I think that philanthropists and the charities supported by them have traditionally been strongly idealist, not in the sense of idealistic (though they may of course be that too), but in the sense defined above, confirming experience by coherence with a set of internal beliefs. Their work has been driven by core values, and confirmation that those values are producing good work has come largely from personal experience. Here, for example, is a cogent and well-argued exposition of this kind of view couched precisely in terms of value. You meet it regularly in the charitable world in the form of ‘we know it works’, ‘we see it all the time’, and its variations.

This quite ordinary way of understanding social programmes has come under pressure from a more positivist streak in policy and philanthropic practice in recent years. That kind of approach emphasizes precisely the key principles outlined above and sets out to find the appropriate truth values for evaluating social programmes using external measures.

These two different approaches to delivering social programmes are deeply held, coherent and – in key respects – incompatible with one another. It’s important to note the incompatibility because not to do so risks being unjust to one side or the other, and failing to appreciate the arguments each is making. Trying to persuade can slip all too easily into forcing to convert.

It may be that most of us are broadly inclined to prefer positivism or idealism in social programming. These are philosophical preferences and styles of reasoning that colour how we view the world. It is useful to recognize them. Here, in a very short quiz, you can test your own and see where you stand: with those who measure to determine effectiveness or those who determine effectiveness through action.

Genevieve Maitland Hudson is a researcher and consultant. She works with the consultancy Osca.

Comments (2)

Gen Maitland Hudson

Thank you for your thoughtful consideration of this piece Barry! It wasn't intended to be entirely serious, nor is it a 'real' typology, but rather a light-hearted attempt to point towards different sets of attitudes (and yes, the quiz is only a bit of fun, not a research tool). I entirely agree that there is no inherent opposition between idealism and positivism, but what I often find when working on social programmes here in the UK is something that resembles either unqualified support or determined opposition to 'measurement', so that the theoretical space for multiple views is not always present. Even to say that positivism and idealism are tools is to take a certain perspective on thinking about theory that wouldn't be shared by everyone. People do take very strong positions on, say, external explanations of stability, contingency and 'truth' that don't always make that kind of dispassionate approach possible. So yes, on balance, and based only on your brief comment above, I would classify you as a positivist. But I would do so only very lightly...

Barry Knight

This is an interesting set of thoughts. But it is far too simplistic to be of much value. Our studies based on >10,000 people suggest that there are six types of attitudes (not two). These types have been established by hierarchical cluster analysis. I would want to see much more evidence for the conclusions and - as it stands - the quiz is close to junk. On the basis of this, the author would probably want to classify me as a positivist. And yes, as a statistician, I am proud to wear that label. I also know the limitations of the positivist method. I am also a fan of John Paul Lederach and his "Moral Imagination", which sees creativity as the central tenet of peace building. Different methods produce different solutions and there is no inherent conflict between idealism and positivism. Both are tools. They do different jobs.

Leave a Reply

Your email address will not be published. Required fields are marked *