Top 15 Charles Perrow Quotes

It takes just the right combination of circumstances to produce a catastrophe, just as it takes the right combination of inevitable errors to produce an accident.

Organizational theorists, at least since the pioneering work of Burns and Stalker, 1961 and Joan Woodward, 1965 and others in what came to be called the contingency school, have recognized that centralization is appropriate for organizations with routine tasks, and decentralization for those with nonroutine tasks. For an early statement see Perrow 1967, and Lawrence and Lorch, 1967.

Engineers speak of a “control loop,” in which the “man in the loop” is the problematical element. This is the human component in a series of sequentially interacting pieces of equipment that control or adjust a function. But when the pilot is suddenly and unexpectedly brought into the control loop (in other words, participates in decision making) as a result of (inevitable) equipment failure, he is disoriented. Long periods of passive monitoring make one unprepared to act in emergencies. The sudden appearance of several alarms, all there for safety reasons, leads to disorientation.

The nuclear power industry, for example, lacks a strong union, has random public victims with delayed effects, has no safety board that is independent of licensing and regulatory functions, and does not see an immediate effect on its profits if safety flags (though a far more severe incentive exists to avoid a catastrophic accident which could shut down the industry)

Most normal accidents have a significant degree of incomprehensibility.

The essence of the normal accident [is] the interaction of multiple failures that are not in a direct operational sequence.

As an organizational sociologist, I was interested in the system characteristics of large accidents, but the discipline also made me aware of the organizational genesis of male-made disasters.

I consider all large organizations troublesome, including governmental and nonprofit organizations. They concentrate power in the hands of their top management; the larger the organization, the greater the power being concentrated. There are degrees of concentration of course, and some large organizations are so disorganized that they lose much of their potential power. But size is generally correlated with these kinds of power: By deciding where to locate they determine economic opportunities for some communities and deny it to others. Their hiring decisions affect the life chances of people, and can, unless checked, favor religious, ethnic, racial, and political affiliations. As consumers of resources, they can favor certain producers over others, and not necessarily on the grounds of “efficiency.” They can mobilize political resources to insure favored treatment better than small organizations.

American society today is shaped not nearly as much by vast open spaces as it is by vast, bureaucratic organizations. Over half the working population toils away at enterprises with 500 or more employees--up from zero percent in 1800. Is this institutional immensity the logical outcome of technological forces in an all-efficient market, as some have argued?

This article [entitled A framework for the comparative analysis of organizations], was one of three independent statements in 1967 of what came to be called "contingency theory." It held that the structure of an organization depends upon (is ‘contingent’ upon) the kind of task performed, rather than upon some universal principles that apply to all organizations. The notion was in the wind at the time.

It [a power based model of organization] sees organizations a intentional human constructions but not necessarily rational systems guided by official goals; as bargaining areas rather than cooperative systems; as systems of power rather than coercive institutions reflecting cultural norms, and as resources for other organizations and groups rather than closed systems. If we define organizations, then, as intentional human constructions wherein people and groups within and without the organization compete for outputs of interest them under conditions of unequal power, we have posed the issue of effectiveness quite differently

I once believed that if organizations had a better fit between their technology and their structure, they would be more efficient and thus more profitable.

The hundreds of scientific studies of this phenomenon come to one general conclusion: Leadership is highly variable or "contingent" upon a large variety of important variables such as nature of task, size of the group, length of time the group has existed, type of personnel within the group and their relationships with each other, and amount of pressure the group is under. It docs not seem likely that we'll be able to devise a way to select the best leader for a particular situation. Even if we could, that situation would probably change in a short time and thus would require a somewhat different type of leader.

The burning cry in all organizations is for “good leadership,” but we have learned that beyond a threshold level of adequacy it is extremely difficult to know what good leadership is.

Another discipline began to intrude upon the confident work and increasingly elaborate models of the human relations theorists (largely social psychologists) and the uneasy toying with bureaucracy of the "structionalists" (largely sociologists). Both tended to study economic organizations. A few, like Philip Selznick, were noting conflict and differences in goals (perhaps because he was studying a public agency, the Tennessee Valley Authority), but most ignored conflict or treated it as a pathological manifestation of breakdowns in communication or the ego trips of unreconstructed managers.