The pace of change is accelerating, driven by rapid and widespread advances in information technology. When combined with advances in transportation and logistics, this radically increases the rate of globalization. The implications of all of this for business organizations are profound. For manufacturers, one of the most important implications has to do with the importance of building empowered learning teams. Only "top-to-bottom" learning organizations can survive and succeed in such an environment. Much has been written about the benefits businesses reap when they evolve into learning organizations. From the work of Douglas McGregor and Chris Argyris at MIT decades ago to the more recent work of Peter Senge (The Fifth Discipline, 1990, Currency/Doubleday) and Jim Barber ("From the Working Class to the Learning Class," National Productivity Review, fall 1994), learning organizations have been recognized as the most effective avenue to sustainable competitive advantage. A major obstacle to embracing rapid change is that change requires people to unlearn old behaviors if they are to benefit fully from new learning. Old habits become the de facto barrier to change until they are unlearned. Unlearning sounds easy: Just forget the old way and use the new one. Abandon old beliefs and habits, and practice the new ones. If resistance to change were not so great, unlearning would be simple. However, change is difficult to bring about -- and frequently more difficult to sustain. Because change creates discomfort, initiatives fail far more often than they succeed. When they fail, the inescapable conclusion is that the learning/unlearning process has not been effective. In many cases, corporate change processes fail because unlearning does not occur. Some don't fail outright, but because of inadequate unlearning they regress or experience periodic relapses. A dramatic example can be found in General Motors Corp.'s Saturn Corp. This totally new division was launched a decade ago with a huge investment and great fanfare. It was to be the model for the future American car company that would compete successfully with Japanese imports. From a consumer, product, and dealer standpoint, the effort has been a success. The "experiment" worked -- at first. Old behaviors of union and management were set aside -- and seemingly "unlearned." New, more effective approaches led to great success in both product quality and cultural change. However, during the last year or two, with the capacity of the original facility exhausted, an unlearning relapse occurred. As GM and the United Auto Workers developed plans to expand, the regression surfaced. Old habits, temporarily unlearned, crept back into the behaviors of both sides. The UAW was not ready to accept the Saturn experiment as the way things would be throughout GM. And company executives failed to realize that they, too, must resist regressing to old behaviors. The struggle for control began to overshadow the success of the new model. When unlearning is truly effective, the likelihood of a relapse is slight, since the new way becomes ingrained as the best way. The newly learned approaches become the accepted norms under which the organization operates. But the pressure for continued change can cause stress levels to rise and resistance to build. This resistance feeds the tendency to regress to the old, comfortable ways, however ineffective they might have been. Ultimately, the newer, more effective behaviors are pushed aside. Saturn has a new leader for 1999. Saturn-brand vehicles now will be produced at places other than Spring Hill, Tenn. How will the workers and union react? How will GM react? Most of all, how will the customers react? Only time will tell whether both learning and unlearning have occurred successfully. John Mariotti, a former manufacturing CEO, is president of The Enterprise Group (www.shape-shifters.com). He lives in Knoxville. His e-mail address is firstname.lastname@example.org.