It is, therefore, an obvious and “rational calculus” to employ a pragmatic and experimental procedure – that is, a policy of redundancy which permits several, and competing, strategies to be followed both simultaneously and separately. Separately, because the moment a plan is put into effect it constitutes an experiment, and unless we introduce “controls” we cannot determine which course of action is best. And as difficult as it may be to apply this policy to an ongoing agency, so is it necessary. It can be seen, then, that any attempt to “program” solutions prematurely is the height of folly. Managements may do this in the interest of economy and control, but the economy will be false and the control a ritual – for we are acting, and organizing, as if we “know” when we do not. It is a striking phenomenon of organizational life (and elsewhere, of course) that we often present the appearance of “rationality” when we do not know what we are doing or why.
Whatever claims are made for programmed decision making, it is to be recognized that if its organizational structure consisted only of the “absolutely minimal number of parts,” error could not be detected.
That was the wise Martin Landau in 1969, “Redundancy, Rationality, and the Problem of Duplication and Overlap.” In Classics of Public Administration, Fifth Edition. Eds. J. M. Shafritz, A. C. Hyde, and S. J. Parkes, Thomson Wadsworth, 2004, pp. 302-313.
This fact should be obvious, and yet it is rarely discussed. In the name of economy, defense and congressional leaders force programming incredibly early with the idea of removing “wasteful” duplication and overlap, while harnessing efficiencies that come from a single development program, a single production line, a single test, training, and logistical function. Yet once that multi-decade plan is embarked upon, error cannot be detected unless there are alternatives to compare outcomes to. This is precisely the predicament DoD finds itself in with a number of programs, but most obviously the F-35.
Consider the effect of competition in markets, sports, and other aspects of life. It would be silly to allow competition if you already knew who was going to win. The point of competition and redundancy is to resolve uncertainty. Imagine evaluating a soccer team by the speed of each individual players, combined with other metrics like accuracy of their kicks, ability to dribble, and so forth. That quantitative analysis tells you something, but really nothing that matters, about the quality of the soccer team. There’s no other way to judge a team’s quality without putting them against the competition and seeing who wins.
Leave a Reply