Alternatives to definition

The skim leaders list serve has picked up recently on fuzziness in definitions of knowledge management (KM). In a recent exchange Steve Denning, formerly of the World Bank and now Chief Squirrel of organisational story telling had started to argue that a better definition of KM might now be required. He thought this might overcome three practical problems: (i) terminological fuzziness might be preventing you from achieving business objectives, (ii) lack of definition will put KM programmes at risk during budget cuts, and (iii) a better definition might reduce problems of under-performance resulting from over hype. Bill Kaplan had argued in return that Steve was talking about issues of knowledge leadership not KM. I responded to extend Bill’s argument as follows.

I think Bill may be getting closer to the essence of a solution here. Definitions tend to be an issue when something is getting started – you will see a lot in the early KM literature but less a decade on. Some of the points below are the same as Bill’s but with a different emphasis.

1 – Its better to focus on understanding than definition. A precise definition will tend to limit the field (premature convergence to use a complexity term) and several do. Any involving the words “tacit” and “explicit” should to my mind now be banned given the restrictions on thinking and practice imposed by that language

2 – A simple definition can cause people to take interest, but will itself be controversial. Of course that can be interesting. Philosophy continues to get value out of debates around Justified True Belief two centuries after they were first uttered as a definition of knowledge.

3 – Its often better to define something by what it is not, rather than by what it is. People have a tendency to impose old patterns of understanding on new things. One way to do this is by the use of antonyms. I recently blogged on this in respect of complexity. People from Learning Organisation and Systems Thinking backgrounds tend to grab Complexity theory, confuse it with Chaos and say that its the same thing as systems. The Antonym list helps establish differences. It also uses a few simple words that people are familiar with and sometimes confuse (for example efficient and effective)

4 – Heuristics are often better than definitions. Heuristics are common sense sayings that make us think about things. When I created three for KM many years ago it became a lot easier to explain what it was about. Each “made sense” and could then be illustrated with cases, or with possible projects. For the record the three are set out below. The first was inspired by Peter Drucker when we spoke at a management retreat some years ago, the third is an extension of Polanyi’s famous quote.:
(i) Knowledge can only ever be volunteered it cannot be conscripted.
(ii) We only know what we know when we need to know it
(iii) We always know more than we can say, and we will always say more than we can write down.

5 – I recently extended the first into a different formulation: In the context of a real need and a direct request it is rare to be refused access to knowledge. However to ask someone to share their knowledge on the basis of some anticipated future need will result in outright refusal or stubborn non-compliance. That sort of statement is a common sense one – it recognises reality. Overall we need to move away from justifying programmes by positing an idealist future state of who things should be, and instead taking a more naturalistic approach and start by trying to evolve from where we are, to some future state that is currently unknowable but can be discovered.

6 -Relating KM to business needs gives understanding without the need for definition. I created a method (which is now part of the open source suit tool kit) for mapping some years ago. This focuses on creating a matrix between intractable problems (derived from Business leaders without talking to them about KM) and Knowledge Objects, derived emergently from people’s stories on the other. The first top down, the second bottom up. The intractable problems are then mapped against knowledge objects and the resultant grid studied for patterns. A vertical run represents a single problem to which many knowledge objects may provide a solution. A horizontal run, represents a single knowledge object which may impact on many problems. The result is a portfolio of knowledge projects, all of which are seen as relevant, rather than a single knowledge programme that has to be justified. (Cognitive Edge practitioners will recognise this as the QQE matrix)

I think by the way that Steve is wrong to suggest that a better conceptual framework would have preserved KM from budget cuts. There is a life cycle for any new management theory, and KM is proving more resilient than most. Humans need novelty to make progress. An existing programme that switched from providing business benefits to defining knowledge management would I think be on a slippery slope to nowhere. Steve is right to link under-performance with too much hype. However I would (with tongue firmly in cheek) argue that too much of the popularising approaches to business narrative are currently committing the same error by derived normative recipes based on retrospective coherence, promising what they cannot deliver. Useful of course, great guidence but not a simple (or even a simplistic) formula. Its difficult mind you, there is a pattern of management text book which requires such an approach and it seems it cannot be avoided.