How thick concepts make it impossible to keep the values out of science
Derek Turner writes . . .
We philosophers like to draw a sharp distinction between descriptive and normative claims. A descriptive claim says something about how things are, while a normative claim says something about how things ought to be.
One might think that natural science deals exclusively with descriptive claims—claims about what happened, and what caused what. Normative claims belong to other fields, such as ethics or aesthetics, or maybe theology. The logical positivists of the 1920s-30s perhaps bear a lot of the responsibility for helping to promulgate this idea that science is “value free.” However, most philosophers of science today agree that this notion of “value free” science is deeply implausible.
Paleontology’s Policy Irrelevance
A good deal of science is what you might call policy-relevant, which is to say that scientists’ findings might bear on practical decisions that have to be made. For example, entomologists who study the effects of chemical pesticides on pollinator species are doing work that can inform decisions about whether to ban certain kinds of pesticides. Values can (and should) sometimes influence the framing of research questions and the design of experiments.
However, one striking thing about paleontology is that it would appear to be almost entirely policy irrelevant. Perhaps paleontology could indirectly inform conservation efforts, but policy makers are not exactly waiting around for the next exciting dinosaur find. For this reason, you might think that paleontology gets closer to value neutrality than do many other areas of science.
But human values matter in paleontology, too. One way to bring this into focus is to consider the role that thick concepts play in our thinking about evolution.
Thick vs. Thin Concepts
Consider the differences among the following three claims:
(1) Smith transferred some funds.
(2) Smith acted wrongly.
(3) Smith embezzled money.
Claim (1) is purely descriptive. It’s just a claim about what Smith did, without any evaluation or assessment. Claim (2) is purely normative. It assesses Smith’s action, without conveying any information about what Smith actually did. But claim (3) is both descriptive and normative. We all know that embezzlement is wrong, so describing Smith’s action as embezzlement comes with a built in negative assessment. Claim (3) also has descriptive content, though, because embezzlement is different from other bad things that Smith might have done. Unlike claim (2), claim (3) tells us what Smith did.
One way to capture what’s special about claim (3) is to say that ‘embezzlement’ is a thick concept, where a thick concept is one that has both a descriptive and a normative dimension. By contrast, ‘wrongness’ is what you might call a thin concept: it has a normative dimension, but has no descriptive content at all.
Stasis and Stabilizing Selection
In 1972, Niles Eldredge and Stephen Jay Gould introduced their now famous idea of punctuated equilibria. They argued that stasis is the dominant theme of evolutionary history. Occasionally you see bursts of rapid evolutionary change coinciding with speciation events, but (they claimed) once a new species establishes itself, it doesn’t change much until it eventually goes extinct. Ever since then, stasis has been a central concept of macroevolutionary theory.
Punctuated equilibria generated huge amounts of discussion, both scientific and philosophical. The idea that stasis is an important evolutionary pattern has held up remarkably well, though of course gradual change sometimes happens, too. Not only that, but scientists have proposed a number of creative models for explaining evolutionary stasis (one of which I discussed in an earlier post).
Scientists sometimes use the terms ‘stasis’ and ‘stability’ nearly interchangeably. If one wanted to do some nitpicky philosophical analysis, one could perhaps say that ‘stasis’ refers to a pattern, while ‘stability’ is more of a dispositional notion; stability, you might say, connotes a resistance to change. Nevertheless, the two notions are closely linked in much of the scientific literature.
A great illustration of this linkage is the notion of stabilizing selection. Some evolutionary biologists argue that stabilizing selection alone fully explains the phenomenon of evolutionary stasis. (For the record, I myself do not agree with this view, but that’s an argument for another day.)
How does stabilizing selection work? With respect to any trait in a population that exhibits continuous variation, you will see a small number of extreme variants. A good example might be birthweight in mammals. Having a birthweight that’s either very low or very high could reduce fitness in obvious ways. In a case like this, we might expect stabilizing selection to cut the tails off the distribution, and to keep variation clustered around the mean for the population. This acts as a brake on directional evolutionary change.
Stability as a Thick Concept
Now, the concept of stability is a pretty good example of a thick concept. Shifting the context just a bit, everyone knows, for example, that ecological stability is a good thing. In his classic essay on the land ethic, Aldo Leopold wrote that “a thing is right when it tends to preserve the integrity, stability, and beauty of the biotic community.” Obviously, stability is being treated here as an anchoring ecocentric value. Ecologists also treat stability as a property of ecosystems that needs to be investigated. According to the diversity-stability hypothesis, the greater the biological diversity in an ecosystem, the more stable it will be in the face of external disturbance, and presumably that stability is a good thing.
There are many other nonbiological contexts in which we naturally think of stability as something positive: You want to build your house on stable foundations. When a patient who has suffered a serious injury is in stable condition, that’s good. We want governing institutions to remain stable, even as particular office holders come and go. The American philosopher Charles Sanders Peirce even held that stable belief is the goal of scientific inquiry. Is stability ever bad? If we’re going to make a negative assessment about things staying the same, we’re more likely to use a different thick concept like ‘stagnation.’ In light of that, it's remarkable that evolutionary biologists frequently associate stasis with stability. You rarely if ever see stasis equated with stagnation. No one ever talks about "stagnating selection."
Stability, then, is a thick concept. Stasis probably is too, given the way scientists (a) sometimes use “evolutionary stability” and “evolutionary stasis” interchangeably, and (b) treat stabilizing selection as a leading explanation for stasis. Because thick concepts always have a normative dimension, it’s impossible to keep the values out of paleontology. One cannot talk or think about evolutionary stability without bringing along some normative baggage.
 Eldredge, N., and S.J. Gould. 1972. “Punctuated Equilibria: An Alternative to Phyletic Gradualism,” in T.M. Schopf, ed., Models in Paleobiology. San Francisco, CA: Freeman, Cooper, & Co., pp. 85-115.
 For just one example of a recent paper confirming the importance of stasis, see G. Hunt. 2007. “The relative importance of directional change, random walks, and stasis in the evolution of fossil lineages,” PNAS 104(47):18404-18408.
 See Charlesworth, B. Lande, R., and M. Slatkin. 1982. “A neo-Darwinian commentary on macroevolution,” Evolution 36(3): 474-498. Also Estes, S. and S.J. Arnold. 2007. “Resolving the Paradox of Stasis: Models with Stabilizing Selection Explain Evolutionary Divergence on All Timescales,” American Naturalist 169(2): 227-244.
 For a critique of the view that stabilizing selection is the whole story about stasis, see J.M. Kaplan. 2009. “The Paradox of stasis and the nature of explanations in evolutionary biology,” Philosophy of Science 76: 797-808.
 A. Leopold. 1949. A Sand County Almanac. Oxford University Press, pp. 224-225.
For an overview of earlier research on the diversity-stability hypothesis, see K.S. McCann. 2000. “The Diversity-Stability Debate,” Nature 405: 228-233.