If you want to anticipate how people and organizations make decisions, you obviously have to be alert to what can go wrong in the decision-making process. It's especially important to look for factors which can lead people to ignore problems, dismiss information, or to delay any response to problems.
One way to do that is to look at how organizations learn. I mentioned Chris Argyris' analysis of organizational problems in the last post.
It helps to have names for the problems. Argyris distinguishes between single–loop and double-loop learning in organizations in a seminal article in the Harvard Business Review in 1977.
Organizational learning is a process of detecting and correcting error. Error is for our purposes any feature of knowledge or knowing that inhibits learning. When the process enables the organization to carry on its present policies or achieve its objectives, the process may be called single loop learning. Single loop learning can be compared with a thermostat that learns when it is too hot or too cold and then turns the heat on or off. The thermostat is able to perform this task because it can receive information (the temperature of the room) and therefore take corrective action. ..If the thermostat could question itself about whether it should be set at 68 degrees, it would be capable not only of detecting error but of questioning the underlying policies and goals as well as its own program. That is a second and more comprehensive inquiry; hence it might be called double loop learning.
Double-loop learning involves questioning the rationale for “the way we do things here”, the set institional policies, procedures and norms. But many of those norms themselves inhibit such questioning, or hide errors, or discourage passing negative information up the chain. And those norms then get embedded within further norms that prohibit noticing that such avoidance is taking place.
In essence, people avoid saying what they know.
The result is organizations become very slow to adapt to change. Early warning signals and alertness are discouraged.
.. the change usually comes long after its necessity has been realized by alert individuals or groups within the organization. The delay teaches these persons that their alertness and loyalty are not valued. Second, those who are not alert or not as involved are reinforced in their behavior. They learn that if they wait long enough and keep their reputations clean, someone else will someday take action. Third, change under crisis and revolution is exhausting to the organization. Fourth, such changes usually reinforce the factors that inhibit double loop learning in the first place. Hence, from the standpoint of organizational learning processes, there would be no change.
Does it matter? Argyris says that most professionals (and many organizations) become very adept at single loop learning. And in the past the environment was predictable enough that rigidity was less costly.
But it amounts to one important set of reasons why organizational and policy decisions often go so catastrophically wrong. Contrary information gets hidden.
One response to this kind of recognition of organizational behavior is to treat it like an overdose of office politics: shabby, a fact of life, but best left to rot in the dark. There's something to be said for that. But another way to see it as a guide to potential blindspots that regularly devastate organizations and careers, that subvert problem-solving and undermine decisions. It's a signpost to avoidable damage and catastrophe. Any responsible decision-maker is going to look out for traps in order to avoid them, not endorse them.