Syncresis Limited

Organisational Learning and the Construction of History

About Syncresis


What do you need?

What we do

Profile: Stephen Scott

Business processes

Systems implementation

Change and risk

BPCS support

Interim systems

Business Continuity Management

Project examples

Email Inquiries

Decision time

  • How to love an old ERP system
  • Just enough project management

Thinking out loud

  • Why learning from experience isn't
  • How to succeed by accident
  • The deep reasons behind project failure

Learning from what?

"What will History say?"
"History, sir, will tell lies, as usual."
George Bernard Shaw, The Devil's Disciple.

Why don't we learn?

There is no shortage of valid ideas about how to promote effective change in organisations. There are a number of soundly-based principles with long pedigrees. For example, I would say that "empowerment" was part of a tradition, emphasising the value of involvement, participation etc. that goes back to the 40's. And there is nothing wrong with empowerment if it is done properly.

Also, lots of case studies have been published, apparently demonstrating the success of these methods. But most change initiatives I've seen, in companies I've worked with, fall into two categories:

(a) a reasonably OK approach is adopted in the beginning, but then peters out

or (b) people are aware of and accept the benefits of applying change-promoting approaches, but the subject somehow fails to get a lot of priority during the startup phase. Many managers have been on courses, they appear to believe in the importance of this stuff, but it just doesn't happen.

There are a number of possible reasons why people would not use a "technology" that works. This note focuses on just one of them.

I'll use "history" as shorthand for the kind of shared, agreed-on experience which we need for collective learning. I was alerted to the difficulties of history in organisations by a number of incidents.

Human ingenuity and the evaluation of success

Firstly, in my experience, it's quite normal to define the success of a project in hindsight, without referring back systematically to the original objectives. I've seen people get away with this quite a few times - success is defined based on what eventually happened, rather than what people set out to achieve.

It seems remarkably easy to do this in the strangest circumstances. In one case, the organisation generally complied well with sound standards for running projects. The client's project manager was as good as any I've worked with. Most of the recommended paraphernalia of project management and control were actively used in this situation. The minutes of the post-implementation review stated that project had been reviewed again the original objectives, and had met nearly all of them.

This was not what happened in the meeting. The original objectives statement was on the table, but it was not reviewed systematically. An important objective concerning standardisation, which had been abandoned in the face of business realities, was passed over. There was some tacit process, which was hard to observe, through which the issue was side-stepped. Also, on-time completion was compared with a different benchmark from the original.

Mysterious amnesias

Other incidents seemed to involve a kind of shared amnesia. A particular group of people would have no recollection, or a very partial one, of some series of discussions which were on the record. The issues weren't necessarily explicitly contentious.

In any case, the publicly agreed story of a past project isn't the only possible one. Which one we end up with is shaped by a number of factors. I'll look at three examples: (a) the need to make sense of what happened; (b) self-presentation; and (c) plausibility.

Tidying up a messy past

People want to understand how they've got to where they are. However, real projects are a tangled maze of events and on-the-fly decisions. Many things go on in parallel and interact with each other. Even on successful projects, people go up blind alleys, make mistakes and recover from them, and so on.

And no way can we hold all of that in our minds, with all the connections, after it happens. It is human to put things in a tidy sequence which didn't necessarily exist at the time, and cut things out which can't easily be made to fit.

But when people do this, they use their beliefs about what makes a sensible cause-and-effect chain, their pre-existing theories about how things work. That's happening before people think about whether the pre-existing theories were a good idea or not - the beliefs are justified before they are criticised.

Training grounds for self-presentation

The second factor is self-presentation. Modern organisational life trains people in this. Good appraisal systems are particularly effective. In two of my past employers, I got excellent advice, as part of the appraisal process, on how to tell a good story about the past year. Of course, because the individual in modern society is also responsible for their own self-improvement, it's OK to identify some areas of weakness as long as they are not too major.

But no matter how good someone gets at this, I'm not claiming they can do it in isolation from other people's perceptions, and that's where plausibility comes in.

Accuracy, plausibility and power

A plausible history is one which is much easier to defend than attack. To achieve this, it has to hang together logically. The versions of the various episodes which are included have to support each other.

But defensibility also has a social dimension The more people share the same account of what happened, obviously the less likely they are to challenge each other. Also, it will be more difficult for a dissenter to establish their account as valid.

But there are obstacles to achieving this consensus. Each participant has different things at stake in the choice between alternative versions of events: different costs and advantages. Also, they have different rewards and sanctions at their disposal, to encourage support for versions favourable to themselves. This can work through supporting or challenging aspects of possible histories which are favourable to the other person, or things people can do through their general role in the organisation.

To be robust, the history needs to achieve a kind of equilibrium, where everybody feels they are getting more or less the best available deal in the circumstances.

The history we use for learning

To summarise: it shouldn't surprise us if the accounts of events, which provide the basis for organisational learning, have some of the following characteristics:

  • They are selective
  • They contain cause-and-effect chains which wouldn't have been obvious at the time
  • The story is more straightforward and less troubled than it felt like at the time
  • The accounts at least partly reinforce existing beliefs
  • And finally, they provide a tradeoff between people's self-presentational needs, their power, of various kinds, and the need for logical coherence.

Accuracy isn't essential; the effect is to make the history less challenging to current practices, and this inhibits learning. That makes it much less likely that people will make a big effort to do things differently next time.

Can anything be done?

I've become convinced that most current project management / change management approaches can only influence this to a limited extent. To escape from the pattern, we need to do a number of things.

Places of safety

The first is to create safe ways for people to express their own versions of events, without feeling threatened. This needs great care: organisations with the most severe learning problems often have problems with trust as well. To avoid defensiveness and resistance, it is equally important to create gentle ways of confronting people with alternative versions of events to their own.


In most cases, skilled facilitation is essential. Depending on the situation, the facilitator may work with individuals, carefully-selected groups, or both. The facilitator will often act as an intermediary, smoothing the process by translating versions of events. The use of email makes it easier to check versions of past events against accounts from the time, but this is also very sensitive. However, there is a well-established body of practice in social and psychological research, which is designed precisely to reduce the risks to people in high-risk situations.


Out of this, we would expect people to be able to sort out a shared understanding of the past which still isn't perfect - that is impossible - but provides far more pointers to doing things differently next time. In many organisations, I see few alternatives. General culture change, to increase trust and openness, is very difficult. By applying this approach to the aftermath of a specific change initiative, we can provide a tight focus, create immediately learning, and still open the door to more widespread changes in how the organisation works.

© 2005, 2008 Stephen Scott / Syncresis Ltd

This note is influenced by a number of academic sources. Particular acknowledgements to:
Argyris, C. (1982). How learning and reasoning processes affect organisational change.
Clegg, S.R. (1989): Frameworks of Power.
Hosking, D.-M. & Morley, I.E. (1991): A social psychology of organising: people, processes and contexts.
Levitt, B. & March, M.G. (1988) Organizational Learning,
Weick, K.(1996): Sensemaking in Organisations.


Syncresis Ltd. Company no. 3548391. Registered in England and Wales. Registered office: 14 Wolverton Avenue, Kingston upon Thames, KT2 7QD. Vat No. 711 0818 74

All material on this page is copyright of Syncresis Ltd, 2005-2009, unless otherwise stated.