The successful organisational models of the past are no longer appropriate in the nervous '90s and organisations must adapt to changing conditions fast or go under.

The idea of the "learning organisation" is attractive but elusive. There are many definitions and it has different meanings in different industries. Reg Revans defined it in his book Action Learning: "For an organisation to survive, its rate of learning must be equal to or greater than the rate of change in its external environment". But how do you measure the rate of learning and the rates of change inside and outside? It's something to do with establishing reality checks between the organisation itself, the people within it and the external world.

Organisations are often described as information processing systems. For example, the bureaucrats operate according to information processing rules, planners and strategists establish organisational targets in information terms. Indeed many of the routine transactions in organisations are to do with the repeated exchange of information. It is argued therefore, that organisations consist of information systems, communications systems and decision-making systems.

The reality is that such apparently rational systems are all carried out by people, and there's the rub. The '50s Nobel prize winner Herbert Simon established that organisations are not rational and do not make rational decisions because people within them do not. His researches concluded that people tended to act upon too little information, limited knowledge of the alternatives and inadequate evaluation of them. So people settle for "good enough" decisions and despite all we have been led to believe about scientific management, the same conclusion applies to organisations.

In such circumstances, the learning organisation holds out new hope. It begins by recognising how flawed is the knowledge we actually have and yet shows how to use it more productively.

For example, in the '70s, concerns about OPEC energy shortages and finite resources dominated the headlines. In the '90s we are worrying about global warming and Chinese imports and no longer have "an energy problem". Things we didn't know then have changed our understanding. Opportunities we couldn't envisage then now give us confidence for the future.

We don't know how little we actually know and so it's difficult to know how to deal with it. We now realise that we live in an age of uncertainty, information explosion and unprecedented technological advances. How can we learn to take advantage of this?

Organisations have evolved to cope with the human dimension by borrowing from the military model familiar to most bureaucratic structures. With sub-divisions and departments, they built hierarchies, chains of command and career ladders. Process rules and information systems were developed. Conformance and compliant behaviours were encouraged. Then they went on to become the most successful economic entities the world has ever known, in many cases more powerful than nation states.

In this process they created great heroes who could do no wrong, and they became complacent and disconnected from markets and customers. Too much time and effort was focused on repeating past successes, winning at internal politics and trying to invent a safer, predictable environment for themselves. Of course these are fatal flaws: one should never expect to be comfortable in business.

It is now becoming clear that the successful organisational models of the past 50 years are not appropriate to the fast-changing "nervous '90s". We are all aware of the transformations in banking, insurance and travel resulting from the use of modern technology. Deregulation and privatisation of telecommunications, transportation and energy sectors have totally changed competitive conditions. In other areas, environmental pressure groups have dramatically upset the reputations and workings of leading international companies.

So how do organisations respond before it's too late? What can we learn from this and how can we do better? In many cases the learning happens very fast, very painfully and often at great cost to the organisations and people involved. Learning takes place very quickly "at the edge of chaos". When the roof is about to fall in, suddenly everything is possible; old prejudices, rivalries and objections are forgotten and people embrace new ideas and work together as never before. All too often it's too late, they don't have enough time to implement their ideas and they pay the price of failure.

In extremis, people and organisations can face the hard facts and respond successfully to difficult challenges. The irresistible questions remain: Why didn't they see this coming? Why didn't they act sooner? One answer is that they need to avoid these catastrophes by establishing better "reality checks" and acting earlier. So what do we mean by "reality checks"?

Within systems thinking there is a concept known as "single and double loop learning". Single loop learning is essentially the ability to defect and correct errors from the norm. This is often referred to as "know how". An example is the thermostat which tells you whether or not the temperature is at the desired level.

Double loop learning, in contrast, goes much deeper and questions the relevance of both the deviation and the norms themselves. This is referred to as "know why". In the example of the thermostat, the questions would be: Why do you want it at that temperature? Is that the temperature you want it to stay at? What do you want to do about it? And so on. Double loop learning is widely seen as the way to work towards the learning organisation.

So the practical approach to organisational learning involves "improving the connections between the outside world, the organisation itself and the people within it". To do this we can think in terms of those factors which block learning and those which enable it.

On the negative side, over-rigid structures which were conceived to help information processing actually inhibit information exchange. Fear of failure makes people avoid the bad news. Past success makes them hang on to out-dated ideas. Over-emphasis on targets creates "tunnel-vision". Internal priorities cut people off from the outside world, and a desire for harmony - a sort of shoot the messenger syndrome - prevents people from looking at the facts and realising that hard actions are needed.

On the other side of the equation are the enablers. Flexible structures permit easy information sharing. An open and trusting culture helps people to have confidence to pass on the bad news, to be tolerant of mistakes and to learn from them. Acceptance of threat, change and discomfort creates positive and adaptive attitudes. "Stretch goals" make people think more deeply and create new opportunities. A clear external focus opens up vital channels of communication with customers, competitors and others.