The Logic Of Failure: Recognizing And Avoiding Error In Complex Situations

It's a thought-provoking book for anyone whose job is to tackle complex problems. In one way or another that includes anyone in just about any profession.


Truly a five-star masterpiece!   February 16, 2002
 36 out of 40 found this review helpful

I picked this book up after hearing a book editor I really respect say he re-reads this great book every few years. And, wow, am I glad I did. What this book talks about is decision-making in situations of complexity, uncertainty and intransparence. The author, Dorner, recounts the results of computer simulations that explore how people succeed and fail in decision making and planning. This is one of those "keepers" that I'll read again and again... and get more from the book each time.

Tangential comment: I'm also a writer (my best-selling books have been Quicken for Dummies and QuickBooks for Dummies) and so I have to say that this book is really, really well-written and edited. Wonderful craftsmanship!


"On S'engage Et Puis On Voit!"   December 21, 2006
 14 out of 14 found this review helpful

Napoleon said "On s'engage et puis on voit!" Loosely translated that means "One jumps into the fray, then figures out what to do next," a common human approach to planning. This discussion (page 161) takes on the adaptability of thought and cautions decision makers about the risks of overplanning in a dynamic, multivariate system. Using examples from Napoleon as well as more concrete examples such as the quotation about soccer strategy (also on page 161,) Dietrich Dörner, the brilliant German behavioral psychologist (University of Bamberg) has created a masterwork on decision making skills in complex systems; I find it to be highly complimentary to Perrow's work and also highly recommend his equally brilliant "Normal Accidents."

A strength of this work is that Dörner takes examples from so many areas including his own computer simulations which show the near-universal applicability of his concepts. One of Dörner's main themes is the failure to think in temporal configurations (page 198): in other words, humans are good at dealing with problems they currently have, but avoid dealing with and tend to ignore problems they don't have (page 189): potential outcomes of decisions are not foreseen, sometimes with tragic consequences. In one computer simulation (page 18) Dörner had a group of hypereducated academics attempt to manage farmland in Africa: they failed miserably. In this experiment Dörner made observations about the decision makers which revealed that they had: "acted without prior analysis of the situation; failed to anticipate side effects and long-term repercussions; assumed the absence of immediately negative effects meant that correct measures had been taken; and let overinvolvement in 'projects' blind them to emerging needs and changes in the situation." (How many governmental bodies the world over does this remind you of?)

I am a safety professional, and am especially interested in time-critical decision making skills. Dörner's treatment of the Chernobyl accident is the most insightful summation I have seen. He makes the point that the entire accident was due to human failings, and points out the lack of risk analysis (and managerial pressure) and fundamental lack of appreciation for the reactivity instability at low power levels (and more importantly how operators grossly underestimated the danger that changes in production levels made, page 30.) Dörner's grasp here meshes the psychology and engineering disciplines (engineers like stasis; any change in reactivity increases hazards.) Another vital point Dörner makes is that the Chernobyl operators knowingly violated safety regulations, but that violations are normally positively reinforced (i.e. you normally "get away with it," page 31.) The discussion about operating techniques on pages 33 and 34 is insightful: the operators were operating the Chernobyl Four reactor intuitively and not analytically. While there is room for experiential decision making in complex systems, analysis of future potential problems is vital.

In most complex situations the nature of the problems are intransparent (page 37): not all information we would like to see is available. Dörner's explanation of the interactions between complexity, intransparence, internal dynamics (and developmental tendencies,) and incomplete (or incorrect) understanding of the system involved shows many potential pitfalls in dynamic decision making skills. One of the most important of all decision making criteria Dörner discusses is the importance of setting well defined goals. He is especially critical of negative goal setting (intention to avoid something) and has chosen a perfect illustrative quote from Georg Christoph Lichtenberg on page 50: "Whether things will be better if they are different I do not know, but that they will have to be different if they are to become better, that I do know." A bigger problem regarding goals occurs when "we don't even know that we don't understand," a situation that is alarmingly common in upper management charged with supervising technical matters (page 60.)

Fortunately Dörner does have some practical solutions to these problems, most in chapter six, "Planning." One of the basics (page 154) is the three step model in any planning decision (condition element, action element, and result element) and how they fit into large, dynamic systems. This is extremely well formulated and should be required reading for every politician and engineer. These concepts are discussed in conjunction with "reverse planning" (page 155) in which plans are contrived backwards from the goal. I have always found this a very useful method of planning or design, but Dörner finds that is rare. Dörner argues that in extremely complex systems (Apollo 13 is a perfect example) that intermediate goals are sometimes required as decision trees are enormous. This sometimes relies on history and analogies (what has happened in similar situations before) but it may be required to stabilize a situation to enable further critical actions. This leads back to the quote that titles this review: 'adaptability of thought' (my term) is vital to actions taken in extremely complex situations. Rigid operating procedures and historical problems may not always work: a full understanding of the choices being made is vital, although no one person is likely to have this understanding; for this reason Dörner recommends there be a "redundancy of potential command" (page 161) which is to say a group of highly trained leaders able to carry out leadership tasks within their areas of specialty (again, NASA during Apollo 13) reportable in a clear leadership structure which values their input. Dörner then points out that nonexperts may hold key answers (page 168); though notes that experts should be in charge as they best understand the thought processes applicable in a given scenario (pages 190-193.) This ultimately argues for more oversight by technicians and less by politicians: I believe (and I am guessing Dörner would concur) that we need more inter- and intra-industry safety monitoring, and fewer congressional investigations and grandstanding.

This is a superb book; I recommend it highly to any safety professional as mandatory reading, and to the general public for an interesting discussion of decision making skills.


Overcoming Complexity   July 28, 2005
 13 out of 13 found this review helpful

THE LOGIC OF FAILURE: RECOGNIZING AND AVOIDING ERROR IN COMPLEX SITUATIONS by Dietrich Dörner is a great read that has changed my thinking about organizations, decision making and complexity ever since I read it about a month or so ago. I recommend it unequivocally to anyone who makes decisions, and particularly to those in organizations and assessment and management positions.

Dörner, a cognitive psychologist at the University of Bamberg, writes about tendencies in decision making, basing many of his conclusions on experiments and simulations he has run with human subjects. For example, he has had subjects manage a mythical area of Africa, Tanaland, to help the nomadic peoples there. The managers can control all areas of life, public works funding, pesticide control, legal issues of grazing, housing matters, etc. In another, Dörner makes his research subjects the mayor of a mythical town, Greenvale. Again, they can address many issues related to life in this community, education, employment, taxes, public services, business development, tourism, social programming, etc. He also does a brief study of Chernobyl to analyze how that disaster happened through logical failure.

Dealing with how people make decisions, the real and root processes of decision making in humans, Dörner makes the point early on that decision making is not an isolated process. "But I am not concerned with thinking alone, for thinking is always rooted in the total process of psychic activity. There is no thinking without emotion" (p. 8). He also makes the point that thought is rooted in value systems, and that our decisions are generally made to bring us closer to the goals that are based on these values. Given this basis, Dörner shows throughout the book various tendencies individuals have when faced with ambiguity in decision-making situations and with complexity that overwhelms them. Sometimes people will focus so tightly on an area in which they are comfortable or will allow themselves to be distracted by small items to avoid coping with complexity.

Often, we cannot know all we need to know about a topic. Dörner writes about the success rates of participants who utilize planning procedures for a limited time, then go to work, but who revisit their issues and are willing to change, vs. those who just set right in without situational analysis, get confused by unintended consequences and start blaming outside factors or creating "myths" or superstitions about why they are experiencing the problems that they are.

Comfortingly, Dörner's research results show that some planning that is moved into action, with frequent analysis of the results of the decisions is effective. He also writes about other effective methodologies, such as thinking by analogy, to generate better understanding of a process.

The entire book, by enlightening the reader on the factors involved in decision making -- emotion, values, fears -- helps to create greater perception on what we do in our own processes of making decisions and following up on them. The book, divided into chapters called "Some Examples," "The Demands" (which deals with complexity, dynamics, intransparence, ignorance and mistaken hypotheses, and steps in planning and action), "Setting Goals," "Information and Models," "Time Sequences," "Planning" and "So Now What Do We Do" does effectively dissect the processes and issues in complexity and understanding them (how hard it is to really understand exponential growth for example, or time's effects on variables).

The book is conversational, with great examples, and fabulous data charts that illustrate the concepts about which he is writing. And it stays in the top of one's mind. I'm going to keep this handy at every work place I ever inhabit as it is a fabulous reference book. I recommend it for everyone!



Should be required reading in business and engineering schools!   May 31, 2006
 11 out of 12 found this review helpful

The author deals directly with the limits of human beings under the stressful situation of solving dynamic problems---ones with a life of their own. Here's something from the text:

"We don't neglect the "implicit" problems of a situation because thinking about the possible side effects of the measures we are planning would overburden us terribly. Rather we neglect them because we don't have those problems at the moment and therefore are not suffering from their ill effects. In short, we are captives of the moment.
The slowness of our thinking and the small amount of information we can process at any one time, our tendency to protect our sense of competence, the limited inflow capacity of our memory, and our tendency to focus only on immediate pressing problems---these are the simple causes of the mistakes we make in dealing with complex systems."

That sums up the author's points nicely. He presents numerous results of computer trials of people solving problems. The scientific analysis is intriguing and his drive is persistent. These points, excerpted above, are brought home time and again.

If you want to improve your troubleshooting skills, whether in business or engineering, or day-to-day, read this book.

I hope you will take the time to rate this review.