Guest Blogger Carol Cleland writes...
In January I received an intriguing message from Mark Johnson, an American glaciologist working in Sweden, who assigned my paper “Common cause explanation and the search for a smoking gun”, Geology (2013), in his fall 2015 field class. Mark studies drumlins. Drumlins are subglacial, spoon-shaped, linear landforms that develop under glaciers and are exposed when they melt. With the notable exception of the ones in Iceland being studied by Mark and his team, which are actively forming [Fig. 1], all drumlins are “relict features” left behind on continents at the end of ice ages [Fig. 2]. Drumlins are made of many different materials (e.g., till, sand, and/or bedrock) and their internal structure is highly variable from field to field. As Mark explained, there are a variety of theories about how drumlins form and they fall into two categories, “equifinal” theories, which hold that drumlins have multiple causes, and (what I will call) universalist theories, which hold that they are produced in the same way. Universalists defend their position by invoking the following principle: (Pd)
“if you cannot prove/show that there is more than one way to make a drumlin, you must assume that there is only one way.”
Mark, who is sympathetic to equifinalism, asked me what I thought about this argument. As I explain below, the universalist argument is seriously flawed, both logically and empirically.
From a purely logical standpoint, of course, Pd is defective; it commits a well-known logical fallacy known as “Argument from Ignorance.” An analogous and clearly flawed inference is: If you can’t prove that there is more than one way to “make” life (whether via natural processes or in the laboratory) there must be only one way. Still, the argument of the universalist raises an interesting question about when absence of evidence for a hypothesis provides grounds for rejecting it. For although one cannot in general infer the falsity of a hypothesis from lack of evidence for it sometimes this is the right thing to do. At some point, persistent failure to find evidence of life on Mars, for instance, will be used to justify the claim that there isn’t life on Mars, and this seems reasonable even though it doesn’t prove that there isn’t life on Mars (buried deeper below the surface or in some, perhaps fairly small, region that has yet to be sampled). As Elliot Sober remarks, “… the motto—that absence of evidence isn’t evidence of absence—is often wrong” (Sober 2009).
In the case of geologically produced phenomena, however, I think that there are compelling empirical reasons for thinking that the principle invoked by drumlin universalists is mistaken. To appreciate this, consider a more general principle: (Pg)
If you cannot prove/show that there is more than one way to produce a natural phenomenon, you must assume that there is only one way.
Pg is clearly problematic. At one time scientists did not know that the mineral quartz, for instance, is produced in several different ways. But it would have been a mistake to infer that it is produced in only one way. For scientists now know that quartz crystalizes from silica-rich magma as it solidifies, chemically precipitates in certain types of hydrothermal environments, and is produced artificially in industrial autoclaves. The point is lack of evidence that there are multiple ways of making quartz cannot be used to justify the conclusion that there is only one way of making it. Additional justification is required. Quartz is not an exception. Carbon dioxide gas provides yet another illustration. It can be produced abiotically, biologically, and anthropogenically. Natural sources of CO2 include volcanic outgassing, wild fires, and respiration. It is also produced in laboratory demonstrations and as a by-product in certain industrial processes. The reactants and chemical pathways involved in some of these modes of production are quite different. In respiration, for instance, carbohydrates are combined with oxygen through a series of chemical intermediaries to produce CO2 at body temperatures. Some hot springs, in contrast, produced it at high temperatures through the action of acidified water on dolomite (calcium magnesium carbonate). In the case of drumlins, which are distinguished by their morphology and vary in physical composition and inner structure, the suspicion that they may have multiple causes seems especially compelling. Biology is full of examples like this. Birds, bats, and dragonflies have wings but it would be a mistake to infer common ancestry on the basis of this morphological characteristic, especially given the lack of similarity in the composition and architecture of their wings. The wings of these creatures are the product of independent adaptations of their ancestors to similar environmental pressures.
The issue is not that it is never the case that there is only one way to make a natural phenomenon or that Drumlin universalists are wrong about all drumlins being produced in the same way; they might be right about this. The point is that insofar as they explicitly rest their argument upon principle Pd, their argument is fallacious. Their argument is not supported by logical considerations. There are often several different ways of producing a natural phenomenon. The case for universalism thus depends upon whether there are specific empirical or theoretical reasons for thinking that all drumlins are made in the same way, just as the case for equifinalism rests upon there being specific empirical or theoretical reasons for thinking that drumlins are made in different ways. In short, the burden of proof is not, as universalists seem to think, on equifinalists.
Finally, one might suspect that I would be sympathetic to drumlin universalism, given that I argue in several papers (Cleland, e.g., 2013 and 2011) that—in the absence of mitigating empirical or theoretical factors (Cleland 2013, pp. 576-577)—puzzling associations among traces of the past are best explained by inference to a common cause, as opposed to separate causes. In other words, common cause reasoning—inference to a single, sometimes very complex, cause—is the default. As an illustration, I argue that the seemingly improbable associations of an iridium anomaly and extensive quantities of cross-hatched shocked minerals in K-Pg boundary sediments from around the world provide compelling (but not of course conclusive) evidence that a gigantic meteorite slammed into Earth 66 million years ago. When this evidence is coupled with the available fossil evidence—extensive patterns of change in flora and fauna on either side of the K-Pg boundary—the case for the end-Cretaceous mass extinction being caused by the impact of a gigantic meteorite becomes very compelling. It is thus not surprising that the meteorite-impact hypothesis currently dominates geological thought about the end-Cretaceous mass extinction. What I don’t argue, however, is that all mass extinctions are caused by meteorite impacts. My claim is only that there is strong evidence that this particular mass extinction was caused by a meteorite impact. And this brings me to an important point: While a particular mass extinction or a particular drumlin is highly likely to have a single (sometimes very complex) conjunctive cause—in the case of the K-Pg meteorite, impact in a biologically productive shallow sea (off the coast of what is now the Yucatán Peninsula) vs., e.g., deep open ocean—it doesn’t follow that all mass extinctions or all drumlins are produced in the same way. Geological evidence suggests that the end-Permian extinction, for instance, was precipitated by extensive flood volcanism located on the supercontinent Pangaea. Similarly, the current anthropocene extinction is neither the product of meteorite impact nor flood volcanism. It is being caused by the ecologically destructive activities of a globally invasive species (human beings).
Carol E. Cleland is Professor of Philosophy at the University of Colorado (Boulder). She received her B.A. in mathematics from the University of California (Santa Barbara) and her Ph.D. in philosophy from Brown University. Her current research interests are in the areas of scientific methodology, historical science, biology (especially microbiology, origins of life, the nature of life, and astrobiology), and the theory of computation. Cleland’s published work has appeared in major science journals (Current Organic Chemistry, Origins of Life and Evolution of the Biosphere, Geology, Astrobiology, International Journal of Astrobiology, and Theoretical Computer Science) as well as in leading philosophy journals (Philosophy of Science, British Journal of Philosophy of Science, Synthese, and Biology and Philosophy). She co-edited with Mark Bedau an anthology, The Nature of Life: Classical and Contemporary Perspectives from Philosophy and Science, and is currently finishing a single authored book (The Quest for a Universal Theory of Life; Searching for life as we don’t know it), which is under contract with Cambridge University Press.
Cleland, C. E. 2013. Common cause explanation and the search for a smoking gun, in Baker, V. (ed.) 125th Anniversary Volume of the Geological Society of America: Rethinking the Fabric of Geology, Special Paper 502: 1-9.
Cleland, C. E. 2011. Prediction and Explanation in Historical Natural Science, British Journal of Philosophy of Science 62: 551-582.
Sober, E. 2009. Absence of Evidence and Evidence of Absence: Evidential Transitivity in connection with Fossils, Fishing, Fine Tuning, and Firing Squads, Philosophical Studies 143: 63-90.