The word "philosophy" is commonly associated by non-philosophers with endless discussions about the meaning of life, and nitpicking about the meaning of simple words. This tale is often told by amateurs, such as the famous physicist Richard Feynman. Feynman tells about his encounter with philosophers in his funny autobiographical "Surely you're joking Mr. Feynman!" (page 69 onwards in the Vintage edition). He started following a class on a book "Process and reality" by Whitehead, in which the phrase "essential object" was used often, but not understood by Feynman. Suddenly the professor of the class asked Feynman to consider whether an electron is an essential object. In order to find out what is meant by essential object, Feynman asked in return whether a brick is an essential object. Then one philosopher stood up and gave an answer, another philosopher stood up and disagreed, and so on! Feynman did not finish the class but joined a biology class instead (with equal results).
At least for philosophy of science I know from my limited experience that this flattering image is not correct. Words and their meanings and definitions are very important, but more in a way analogous to the error analysis of scientists. The accuracy of observations expressed in numbers is important in deciding whether these numbers fit a theory, or are significantly different from each other. In the same way, the philosophical essays are carefully built up from well-defined words, of which the definitions can at least be defended as being reasonable. Consequently, the language in philosophical works commonly is more precise than in scientific literature.
On the other hand, the phrases "geology" or "earth science" are commonly associated by non-geoscientists with volcanoes and earthquakes. (Geography's fate is even worse: this is just mapping the position of towns and rivers.) If earth science is mentioned at all in philosophical literature, it usually relates to the story of Alfred Wegeners idea that continents were once attached to each other, and the subsequent discussions and discoveries that lead to the acceptance of the theory of plate tectonics. But obviously there is much more to geoscience than plate tectonics.
One aim of this symposium is the discussion on the fundamental nature of the geosciences and the relations with other sciences. However, almost no literature exists on philosophy of geosciences. It has been neglected by philosophers, while the self-reflections of geoscientists on their own discipline are interesting but limited in scope and depth. A motive for this symposium is that geoscientists can learn from philosophers and vice versa. To focus the presentations and the discussions, the theme of reduction (explained below) has been chosen.
In the introductory presentation a few central features of geosciences are explained. Next, a general overview of the philosophy of science is given to indicate the place of the reductionism debate and to further focus the discussions. In addition some terminology that will probably be used by the speakers will be explained. Elsewhere the structure and programme of the symposium is given and abstracts to the four lectures by the guest speakers.
Biologists expressed the consequences of this type of underdetermination as follows. As a consequence, theories based on laws of physics and chemistry are difficult or impossible to apply or test. Such a complete causal explanation would involve too many variables and initial conditions. According to Van der Steen and Kamminga (1991), the complexity of historical phenomena, such as the evolution of a particular species or the changes in a particular landscape, precludes the possibility of formulating complete explanations, because of the lack of knowledge of laws and initial conditions. As Van der Steen and Kamminga state:
In many cases such events are so complex that a full-fledged explanation would require appeal to many (physical and chemical) laws and an exhaustive description of initial conditions. So, in practice, one has to be content with descriptive work or with explanation sketches. In the latter case one will use laws, but these may stay in the background and they need not be formulated in a very accurate way.
This situation is, in my view, also typical of earth science. An explanation sketch may be seen as an inference to the best explanation, where the (or a) likely scenario is inferred from the observations, and I think it is similar to geomorphological reasoning and geohistorical narratives. (But this is open to discussion!) Often causal explanations are impossible but an explanation sketch can be given based on various clues from the observations. Such clues are often called "smoking guns", and sometimes are strong enough to exclude some likely scenarios or trigger the construction of a new likely scenario.
Appropriate examples of smoking guns for the impact of the comet 65 million years ago, which presumably lead to the extinction of the dinosaurs, are a large impact crater and a thin iridium layer. The laws of physics remain in the background, or are used in scenario computations about the amount of energy released in such an impact. In this example, it is impossible to reconstruct the initial velocity and mass of the fatal comet from the smoking guns, but a reasonable range of the parameters can be derived from statistics of observations of present-day comets in the solar system. The aim of a geoscientist is to establish what kind of events possibly took place and to narrow the potential explanations. The "what" is found out by hypothesising and looking for the smoking guns. The "possibly" is found out by confrontation of the hypotheses with other smoking guns and, depending on the availability of initial condition data, with the laws of chemistry or physics. An interesting question is now whether validation of mathematico-physical models is possible in principle? And, how important is this underdetermination in the theory testing of geoscientists?
Summing up, earth scientists are usually confronted with the following predicament: How to obtain non ambivalent and sufficient data for a complete historical narrative and an associated causal explanation of the course of event? They have to infer from present situations to past ones, or from a limited set of observations to a hypothesis or theory. The empirical data gathered by earth scientists often leave room for a wide range of different, incompatible hypotheses. These hypotheses are empirically equivalent (equally supported by the data) but they cannot be true at the same time. In sum, their inferences are hampered by the problems of underdetermination.
"Reducible" can be understood in two ways: ontological (about "what is") and epistemological (about "what we can know"). Ontology roughly is the theory of being and existence, and about the structure of reality. Ontological reductionism is the theory that all phenomena are determined by the behavior of microphysical entities (elementary particles and fields). Epistemology roughly is the theory of knowledge, the study of our right to the beliefs we have. Epistemological reductionism is the theory that all phenomena can be completely understood in terms of the behavior of microphysical entities.
In order to achieve the ultimate reduction, the world can conceptually be divided in levels of organisation, where elementary particles and fields form the lowest (physical) level, sand grains may form a higher level, and rivers an even higher level. The same would apply to biology, from organic molecules to organels, cells, organs, organisms, and communities.
The ultimate reduction would be achieved stepwise: first, derive the theories of the highest levels from the theories of the level immediately beneath it, then derive the theories from that level from the theories one level lower and so on, all the way down from the highest levels through the intermediate levels to chemistry and physics. The derivation of one theory from another is called intertheory reduction.Epistemological reductionists hold that ultimately all theories can be reduced to microphysical theories.
Reductionism has been divided in many other different types. An important division for the present discussion has to do with the scope of epistemological reductionism: whether a whole science is reduced to a more basic level, which is global reductionism, or whether a single theory or description of a phenomenon is reduced to the more basic level, which is local reductionism. Related to the latter is methodological reductionism, which states that it is useful to attempt reductions without stating (as in epistemological reductionism) that this should always be possible. An interesting question for geoscientists is whether the whole geoscience is reducible to physics, or whether the reducibility is limited to certain phenomena or disciplines only, such as meteorology and geophysics. Even more interesting, if so, why is this?
In the past decades serious doubts have been raised about the possibility to reduce all higher-level phenomena to physical laws, which led to the emancipation of biology, psychology, and other disciplines as autonomous 'special sciences'. These arguments commonly have to do with the evolution of species and with consiousness. Since neither evolution nor consiousness refers to geoscientific phenomena, an interesting question is whether geoscience is a special science like biology, or whether it is more akin to physics even though geoscientists are hampered by underdetermination.
Concluding, the theme of reductionism leads to a number of interesting questions on the relation between geosciences, physics and biology, on the fundamental nature of geoscientific phenomena and on the comparison of causal explanations and seemingly more vague ideas of geomorphological reasoning.
In this symposium, a geoscientific discipline focussing on causal explanation from physics (meteorology and climatology) is contrasted with a discipline based more on explanation sketching and hypothesis generation (hydrology and geology). Second, the philosophies of physics and biology are contrasted with much attention for reductionism. Third, the nature of geosciences between these extreme positions will be discussed between the speakers and with the symposium participants.
Suggested reading: see references and links
Acknowledgements: Arno Wouters, Henk de Regt