brain-made-of-hands.jpg
Dave Snowden

Of practical wisdom

RSS Feed

brain-made-of-hands.jpg I couldn’t have found a better image for today’s post which is about the proper balance of theory with practice; nay the necessity of interaction between the two.  I’m going to illustrate this in the context of Agile in software development but the point has wider application.

One current trope within a part of that community is to reject theory focusing on good people doing good things. Now I can understand that view given that much which purports to be theory is just a set of generalised recipes based on limited cases. Interestingly there is a related debate in aesthetics between the art for art’s sake guys and those who argue that art is not art with our prior evidence of craft. I am firmly in the latter school of thought there which is why I have sympathy for many who reject ‘theory’ in the Agile community.

Now this is a big topic and I am only going to be able to touch on aspects of it in this post. The link above to my earlier post on limited cases was one contribution to that debate and I will assume readers are familiar with my arguments there. To be clear, there is nothing wrong with deriving a hypothesis  from observations of practice provided you are aware of the limitations of that approach. There are aspects of Popper here;  I may not yet have managed to refute the case, but neither have I proved it. But theory is not confined to that derived from observations of cases, even less if the cases do nor arise from the deliberative application of that theory. Theory can also come from other disciplines. So we know a lot about culture from anthropology, of systems from many sources including complex adaptive systems theory, of decision making from the various cognitive sciences and so on.

So there is a different way at looking at the interaction of theory and practice, known as praxis. Here if something happens that I find interesting then I try and find out why it worked by looking at validated knowledge from the sciences and the humanities. If I know why, then I can apply my knowledge (now a combination of theory and practice) in a new or novel situation. Also, in conditions of extreme uncertainty where I have no relevant examples then I can fall back to theory to give be a coherent pathway.  I will also rely on intuition to a degree. Larry Prusak once said that intuition is compressed experience and its something that we should always bear in mind.

Now the problem if only using cases and/or ignoring theory is evidenced throughout management practice but it you can also see it in the Agile community. I once said that Agile is sound practice ill informed by theory and I have seen little evidence to contradict that statement since. One result of this is that we get waves of fads built around accreditation and certification (SAFe is just the latest of many). Anyone with a basic knowledge of complexity science would reject any method that aggregates multiple methods into a complicated flow model and of course it only works because people never fully apply it.  Probably the success of SAFe is that this is actively advocated.  Given that you are allowed to train others once you have been trained sort of gives the game away.  That type of ‘theory’ should be condemned, teaching without practice in the application of a new skill is the province of snake oil sales people.  We survive because good people can always make bad process work, despite itself. It is how companies survive sick stigma and the like.

But this is all a massive missed opportunity. The need to scale Agile is an imperative not just for productivity, good code or satisfied users, although those would be good enough reasons in their own right. There is a real need to radically change the way we use technology in a co-evolutionary way with both articulated, and more importantly unarticulated user needs. That is critical for society as a whole in a world of increasing uncertainty. If we continue cobbling together methods into easily replicated training courses we may get short term success. We may even allow IT leadership to say they are Agile (which is not the same thing as being Agile). But all will happen is that the cycle will come round again and again.  A collection of practices and platitudes not fully informed by theory will work for a bit then fail in whole or in part. Then the whole god damn thing will start again.  But I suspect those who focus on training and accreditation alone will not be too worried about that, in fact they might actively encourage it.

Claiming art without craft is pretension although it may be artful (in the negative meaning of that word). Failing to inform craft by theory is at best a missed opportunity at worst a demonstration of ignorance.

We need to use our brains as well as our hands ….

  • Michael Hill

    I must admit I didn’t really get this one very well at first. Perhaps I lacked enough theory, practice or both. But then in following up on a student critique I think I gained enough to put the pieces together. The short story is one of the best students had recommended NLP in his overall critique, which created dissonance for me. How could the student have seemed to ‘get it’ and then recommend NLP? So I followed up as my field often runs into multiple meanings for the same acronyms or initialisms. Yup, confirmed that NLP. So I did more reading on it. Then reread both his original and follow-on feedback. I came to see NLP as like learning real magic from street magicians: you might learn some useful skills but you won’t learn any real magic. If you continue to believe its real magic you will probably not be very successful – your theory and practice aren’t aligned and you aren’t going to create anything really new and beautiful. And in that I think I finally understood your post. That reread of the original and follow-on feedback revealed the student had learned some skills (mainly to listen, observe and be quiet), discarded (or never learned) the NLP theory (and hadn’t pursued any additional training nor certification for a decade or more), and adapted those skills to more useful pursuit. The danger remains that the student attributes it to NLP, rather than the useful skills that could be better taught without an entangled theory.

    • davesnowden

      Excellent illustration of the point. Thanks for that

Top