| Summary: | The history of scientific understanding of chlorine and ozone chemistry in the stratosphere is reviewed. This includes early work on ozone in the atmosphere, and the development of understanding of the role and importance of the stratospheric ozone layer. A warning was given in 1974 that chlorine compounds were building up in the stratosphere, and that the chemistry of chlorine in the stratosphere could lead to ozone depletion. Release of inert synthetic chlorine compounds to the lower atmosphere was blamed. After a decade of little or no indication of actual ozone depletion, discovery of the Antarctic ozone hole was announced. The associated decreases in ozone concentration were both surprising and large, but also very temporary and local. The subsequent investigation eventually showed convincingly that chlorine chemistry caused the phenomenon. This history includes several dramatic changes of scientific opinion. Some interesting issues arise in the investigation, that can be compared against philosophers' accounts of several aspects of scientific epistemology. A crucial experiment in the investigation of the Antarctic ozone hole is identified. It is argued that there are good grounds for viewing crucial experiments as rather more diverse in their natures than most philosophers have allowed. In one important approach to the epistemology of science, scientific theories have been regarded as generalisations. In this case a single counter-instance can potentially undermine a theory by falsification, while no amount of confirmatory evidence is able to establish it. In this thesis I argue that the generalisation model does not fit many scientific theories. The respective weights of confirmation and falsification in deciding between rival theories of the ozone hole are considered. It is shown that in theories of this type, which consist largely in the attribution of cause, direct confirmation may be more important than falsification. Several problems in scientific epistemology are raised which relate to recent changes in the nature of the scientific enterprise. The problems of overspecialisation in science, and of blind spots at disciplinary boundaries, have already been widely discussed by philosophers. The use of very complicated computer models, which can show that the observations support a particular theory without revealing how they do so, raises new epistemological problems which need careful consideration. There are also new problems associated with modem experiments which generate huge amounts of observational data: manual analysis and appreciation of the data becomes an impossibly large task, and automated data analysis routines work from a series of preconceived assumptions about the nature of the data, and there is a good chance that they will miss any unexpected pattern of anomaly. Scientists consider prediction to be an important part of a successful theory. The emerging theory of chlorine-mediated ozone depletion shows that high regard for a theory does not necessarily depend on its central predictions being correct, though it does need a significant set of correct subsidiary predictions. Finally this thesis examines how the scientific debates around the ozone issues were conducted, and how a scientific consensus formed, and was revised at various stages. A presentation of a dissenting view which opposes the current scientific consensus is examined in some detail, and found wanting as a contribution to the scientific debate. It is concluded that the current scientific consensus on stratospheric ozone is well grounded. The study shows that a careful examination of the history of an episode in recent science can throw a new light on some of the epistemological discussions of philosophers of science, and can also indicate some interesting divergences of recent science from the science of earlier centuries.
|
|---|