Continuous improvement, one of the panaceas currently in good currency, is an integral and important part of Total Quality Management (TQM). No one, of course, can oppose improvement, let alone continuous improvement. Unfortunately, however, what is called "continuous improvement" frequently isn't. In fact, it may hurt performance. Continuous improvement begins by examining a current activity, process, product, or service and determining what is wrong with it. It focuses on weaknesses such as clerical errors, product defects, late deliveries, and so on. Improvement is conceptualized primarily as the removal of defects or deficiencies. Unfortunately, the removal of defects and deficiencies provides no assurance of reaching an end point that is more desirable than the point at which one starts. This is apparent from the following examples. If I turn on a TV set in the middle of a weekday, there is a very low probability that the program that initially appears is one that I want. This is a defect that I can easily remove by changing the channel. However, there is still a very low probability that I will get a program I want, and in fact, I may well get one I like even less. On occasion I may go through as many as ten channels without getting a program I want. Thus, the "solut ion," changing the channel, is not necessarily a solution. Moreover, it does not attack the underlying problem, which is the paucity of programs that I would like to see. Recall what happened in the United States when it tried to get rid of the deficiency, alcoholism, by eliminating what it took to be its cause, alcohol. "Prohibition" did not eliminate either alcohol or alcoholism but it did produce large-scale organized crime. In the 1960s the United States tried to eliminate drug addiction similarly, by illegalizing all addictive drugs. Again, the solution intensified the problem rather than solved it and it gave organized crime an even