## Metadata - Author: [[Dietrich Dorner]] - Full Title: The Logic of Failure - Category: #books ## Highlights - Complexity is the label we will give to the existence of many interdependent variables in a given system. The more variables and the greater their interdependence, the greater that system's complexity. Great ([Location 403](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=403)) - Supersignals reduce complexity, collapsing a number of features into one. Consequently, complexity must be understood in terms of a specific individual and his or her supply of supersignals. We learn supersignals from experience, and our supply can differ greatly from another individual's. Therefore there can be no objective measure of complexity.Dynamics ([Location 419](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=419)) - People desire security. This is one of the (half) truths of psychology (for people sometimes desire insecurity too). And this desire prevents them from fully accepting the possibility that their assumptions may be wrong or incomplete. ([Location 448](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=448)) - Stating a goal comparatively ("better transportation network" or "moreuser-friendly") often indicates that we don't know precisely what we want. ([Location 463](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=463)) - "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," said the Enlightenment aphorist Georg Christoph Lichtenberg. ([Location 515](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=515)) - As we have seen, in complex situations we cannot do only one thing. Similarly, we cannot pursue only one goal. If we try to, we may unintentionally create new problems. We may believe that we have been pursuing a single goal until we reach it and then realize-with amazement, annoyance, and horror-that in ridding ourselves of one plague we have created perhaps two others in different areas. There are, in other words, "implicit" goals that we may not at first take into account at all and may not even know we are pursuing. To take a simple example, if we ask someone who is healthy about her goals, she will not normally name "health" as one of them. It is, nevertheless, an implicit goal, for if we were to raise this point specifically, she would agree that maintaining her health is important. In general, however, health will become one of her explicit goals only if she falls ill.This may sound obvious, but as we shall see, the fact that most people's actions are driven by an excessive (or exclusive) preoccupation with explicit goals accounts for a great deal of bad planning and counterproductive behavior. People ([Location 537](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=537)) - To summarize, goals may be:• positive or negative• general or specific• clear or unclear• simple or multiple• implicit or explicit ([Location 544](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=544)) - A situation is characterized by high efficiency diversity if it offers many different possibilities("diversity") for actions that have a high probability of success ("efficiency"). In chess, examples of such situations are control over the… ([Location 558](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=558)) - "User-friendliness," "comfort," "favorability to labor," "ensuring the peace"-these are all conceptualizations, and when we have a concept, we are inclined to think that there must be something that underlies that concept, some one thing. But the concepts above do not denote one thing. They are… ([Location 563](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=563)) - By labeling a bundle of problems with a single conceptual label, we make dealing with that problem easier-provided we're not interested in solving it. Phrases like "urgently needed measures for combating unemployment" roll easily off the tongue if we don't have to do anything about unemployment. A simple label can't make the complex nature of a problem go away, but it… ([Location 571](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=571)) - Delegation means that we commission other institutions and persons to do detail work for us but that we remain conscious of the role the delegated problem has in the overall problem. We stay in touch with the delegated problem. By contrast, when we dump a problem onsomeone else, we instantly dismiss that problem from our mind; when it reappears in our consciousness, we respond with irritation… ([Location 592](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=592)) - As Brecht observed late in life, advocates of progress often have too low an opinion of what already exists. When we set out to change things, in other words, we don't pay enough attention to what we want to leave unchanged. But an analysis of what should be retained gives us our only opportunity to make implicit goals explicit and to… ([Location 609](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=609)) - A flow situation is one in which tension is built up, then released, a sequence in which the individual experiences fear of failure, triumph over obstacles, renewed fear of failure, another triumph, and so on. Inadequate concretization and elaboration of goals can leave a problem solver vulnerable to this phenomenon. An interim goal happened on by chance may seduce him into a flow situation he is helpless to escape (and perhaps may not even want to escape). ([Location 655](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=655)) - Many social scientists who have set out to write computer programs they could use to evaluate an experiment have woken up years later to find themselves computer specialists. And they will hardly have realized that they have long since lost sight of their real goal and become addicted to the fascination, challenges, and triumphs of working with a computer. ([Location 658](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=658)) - Contradictory goals are the rule, not the exception, in complex situations. In economic systems costs and benefits are almost always at odds. Something that doesn't cost much is rarely of much benefit, and if we want to derive great benefit we usually have to invest a lot. Because businesses want to minimize costs and maximize benefits, we have clearly found a conflict of goals. But this conflict is relatively harmless because it is one we are usually aware of. More dangerous are the situations in which the contradictory relation of partial goals is not evident. For most people who plunge naively into a planning process, the contradictory nature of better transportation links and better shopping opportunities will not be immediately obvious. Nor was it obvious in the days of the French Revolution that realizing the partial goals of liberty and equality at the same time is no easy task.If we understand "liberty" to mean simply a condition under which there are very few constraints placed on the actions of individuals and if we understand "equality" to mean the right of equal access to the material and nonmaterial resources of a society, then "liberty" will quickly result in great inequality because those who are better equipped for certain activities (for example, those who are more intelligent) will be more successful at obtaining the resources they want, while others will be less successful. Thus, more liberty will mean greater inequality. On the other hand, the attempt to achieve a high degree of equality in a political system will produce a correspondingly high number of constraints on the individual, and that is not liberty. ([Location 690](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=690)) - When we must deal with problems in complex systems, few things are as important as setting useful goals. If we do not formulate our goals well and understand the interactions between them, our performance will suffer. When we do not make overly general or unclear goals specific, we are likely to spend time unwisely on ineffective repair-service behavior. Ultimately, to salve our self-confidence, we may find ourselves choosing projects for their obviousness or ease rather than for their importance. If we overlook implicit contradictions among our goals, we may achieve good results initially, but in the long run we will produce many bad results. We have strategies for dealing with this phenomenon as well-goal inversion makes unintended bad results into good ones, conceptual integration erases the differences between incompatible elements, and conspiracy theories place the blame for our mistakes on others. If we can learn to recognize these tendencies in ourselves, we will be able to examine the failures in goal setting they are meant to disguise and, with work, will have a chance to improve our ability to set goals appropriately. ([Location 750](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=750)) - There is no a priori appropriate level of detail. It may happen that in working with a system we will have to move from one level of detail to another. As a rule, however, we should select the level of detail needed to let us understand the interrelationships among our "goal variables," that is, among the variables that we want to influence. We do not need detailed knowledge of the specific components that underlie those variables. ([Location 836](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=836)) - Presented with the abstract mathematical structures that underlie these simulation games, participants in our experiments would not have displayed certain of the behaviors that the semantic trappings of the variables elicited from them. For example, participant pmosc606, whose record is shown in figure 15, refused for ideological reasons to do anything that would harm the ecology of the Moros' territory, and consequently he declined to take any measures against the tsetse fly or to drill any wells. He would have had no such scruples had he been confronted with an abstract mathematical system. When this participant saw that his proecological intentions produced antiecological results, he was obliged to rethink his position. A similar failure in dealing with an abstract mathematical task probably would have moved him to cite his lack of mathematical ability or some other such cause. While he considered himself knowledgeable about ecological matters, he likely had much less pride invested in the notion of himself as a mathematician. ([Location 886](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=886)) - The person who feels moved for one reason or another to study the nature of our world or at least of our society and who concludes that we live in an "automobile society" or a "service society" or an "information society" or an "atomic society" or a "leisure society" proffers a reductive hypothesis that invites us to extrapolate a structure from it.The fact that reductive hypotheses provide simplistic explanations for what goes on in the world accounts not only for their popularity but also for their persistence. Once we know what the glue is that really holds the world together, we are reluctant to abandon that knowledgeand fall back on an unsurveyable system made up of interacting variables linked together in no immediately obvious hierarchy. Unsurveya- bility produces uncertainty; uncertainty produces fear. That is probably one reason people cling to reductive hypotheses. ([Location 974](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=974)) - A necessary generalization can easily evolve into an overgeneralization. And as a rule we have no opportunity to test in advance whether a concept we have developed has struck just the right degree of abstraction or is an overgeneralization. ([Location 1007](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=1007)) - In complex systems with many interlocking elements, deconditionalizing abstractions are dangerous. The effectiveness of a measure almost always depends on the context within which the measure is pursued. A measure that produces good effects in one situation may do damage in another, and contextual dependencies mean that there are few general rules (rules that remain valid regardless of conditions surrounding them) that we can use to guide our actions. Every situation has to he considered afresh. ([Location 1018](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=1018)) - The more we know, the more clearly we realize what we don't know. This probably explains why we find so few scientists and scholars among politicians. It probably also explains why organizations tend to institutionalize the separation of their information-gathering and decision-making branches. ([Location 1067](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=1067)) - We combat our uncertainty either by acting hastily on the basis of minimal information or by gathering excessive information, which inhibits action and may even increase our uncertainty. Which of these patterns we follow depends on time pressure or the lack of it. ([Location 1108](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=1108)) - Anyone interested in studying a case in which resignation-induced torpor, refusal to gather and analyze information, distortion of information, and sudden, frantic fits of action all occur simultaneously shouldread the sections of Joseph Goebbels's diary written during the last days of the Third Reich.' There we find "vertical flight" into fantasies about Frederick the Great, Czarina Elizabeth, and the final phase of the Seven Years' War. We also find "horizontal flight": concern about trivial everyday matters no longer of any importance and about the designing of new military medals. And we find, finally, the enactment of utterly brutal military measures that, quite apart from moral considerations, were utterly pointless. ([Location 1116](https://readwise.io/to_kindle?action=open&asin=B0010O0CS0&location=1116))