How to Infer Metaphysics from Scientific Practice as a Biologist Might

1. From Biological Practice to Scientific Metaphysics?

It is time to address the epistemological problem of how we are to make inferences, from premises involving our analyses of scientific practices to conclusions of metaphysical claims about the world. Our need for a new model of inference stems from conjoining practice-based epistemology of science with scientific metaphysics. On the Quinean picture of scientific metaphysics, we can glean the ontology of the world by analyzing the objects quantified over in the statements of our best scientific theories (Quine 1948). Representing inferences to metaphysical claims from scientific theories is logically simple because both are propositional and truth-apt. The problem for a practice-based scientific metaphysics is how to model inferring from a practice, including tools and methods that are not truth-apt, to a metaphysical claim.

Ian Hacking advocated analyzing how science succeeds in terms of intervening and not representing. His slogan “If you can spray them, then they are real” (Hacking 1983, 23) is an example of what an inference pattern from a practice to a metaphysical claim could look like. However, the form of and support for Hacking’s simple inference rule remain mysterious (Miller 2016).

I am interested in arguments of the following form: ‘The practice goes so-and-so because the world is such-and-such’ or ‘We know the world is such-and-such because the practice investigating it goes so-and-so.’ The problem facing the epistemology of scientific metaphysics here is that we need then a model of inference based on a relation that holds between a practice and the world. Ken Waters recently made a bold metaphysical argument about the structure of the world based on how biologists practice genetics: “My metaphysical claim is that scientific practices in genetics and allied sciences take this form because they are adapted to a reality that has no overall structure” (Waters 2017, 99). I analyze Waters’s reasoning at length in this chapter. In addition to being a clear example of making the kind of argument I am interested in, Waters inadvertently suggests an exciting and promising solution to our problem. What if successful scientific practices were adapted to features of the world? If features of a scientific practice were adapted to the world, then it would perhaps be possible to learn about the world by investigating the practice. This proposal has promise because, rather than developing a novel model of inference for metaphysics based on adaptation, we can draw on analogous inferences used by biologists to solve analogous problems.

The analogy I suggest we take seriously then is: Successful scientific practices are adapted to features of the world for purposes as organisms are adapted to features of their environment for functions. Biologists use traits of organisms as proxies for environmental conditions. My metaphysical methodology is therefore to pursue the use of traits of scientific practices as proxies for metaphysical features of the world. I devote this chapter to developing the beginnings of what this research program looks like and how it functions. Biological practice should inform both the methodology and content for doing scientific metaphysics. We should use this biologically inspired methodology to explore the metaphysics of any parts and aspects of the world we can get a grip on through analyzing the practices of all the sciences.

Paying attention to the biological practice for methodological advice promises to provide guidance for dealing with the formidable, basic problems facing a scientific metaphysician such as Waters. If we take the talk of adaptation seriously, we should also accept that, if the world were structured differently, the practice of genetics would be different. How can the metaphysician give evidence for the ability of a scientific practice to track the structure of the world? If we accept that scientists can shape their investigative practices to fit their environment, why think that the reason a practice is the way it is is because of the metaphysics of the world and not because of the history or sociology of the discipline? Perhaps genetics uses a particular gene concept not because the structure of heredity lacks structure but because of the gene concepts that came before it and because of how new biologists are educated and trained. There are many alternative hypotheses as to why a given scientific practice is the way it is, and the scientific metaphysician must show that the adaptation hypothesis is better supported. In each of these problems and more, biologists face an analogous problem and have developed accepted ways for answering it.

I develop my proposal as follows. In section 2, I present how paleoecologists infer to past environmental conditions from fossils of organisms that lived in them. I use the case of inferring from the shape of teeth to past humidity to draw lessons about inference patterns and support to construct the proxy by adaptation framework. In section 3, I import the proxy by adaptation framework to practice-based scientific metaphysics. I use the case of Ken Waters inferring from geneticists’ use of the gene concept to the structure of heredity, development, and evolutionary change to explore how we can apply the framework. In section 4, I comment on several decisions I made in developing the proxy by adaptation framework. In section 5, I discuss the relationship between science and metaphysics on my proposal for a biologically inspired, practice-based scientific metaphysics.

2. From Organism to Environment

How do scientists learn about inaccessible environments? When they cannot directly measure environmental variables, scientists construct and measure environmental proxies. Proxy variables are proxies for other variables based on some understood physical, chemical, or biological processes. Geologists take ice core samples to measure the relative ratios of gases trapped in bubbles. They use gases as a proxy for atmospheric temperature based on physical and chemical processes of isotopes. Geologists also use the levels of dust in ice as a proxy for dryness of the environment because wind blows the dust produced by erosion around more in dry areas (Bender, Sowers, and Brook 1997). Biologists use tree rings as proxies for past temperature and dryness based on ecological and physiological conditions favoring and disfavoring growth. Paleoecologists use pollen counts in soil as proxies to reconstruct past climate based on ecological dispersal and competition. They must identify, calibrate, and evaluate these proxies in an iterative process of continual refinement.

Sailors use seabirds as a proxy for land when they are at sea based on experience and ecology. Alfred Russel Wallace used the long nectary spur of an orchid from Madagascar as a proxy to infer the existence of a pollinator moth with a long proboscis that sucks it. Wallace (1867) made his inference on the basis of adaptation via natural selection.1

In order to articulate the proxy by adaptation framework of reasoning used by biologists, I examine the case of mammalian teeth as proxies for past humidity. Biologists know an extraordinary amount about mammalian teeth.2 Teeth are the most durable and well-preserved parts of animals, and they are one of the most morphologically diverse. Because of their functional importance, they are highly specialized and distinct across even closely related taxa and strongly subject to adaptation by natural selection.3

Mikael Fortelius (1985), professor of evolutionary paleontology at the University of Helsinki, is an expert on Cenozoic (last 66 million years) ungulate (hoofed animals from horses to hippos to deer) teeth. Fortelius leads a multifaceted research program using teeth and other traits to learn about past climate change. I focus here only on the group’s use of one trait of ungulate teeth—their hypsodonty, or height (Figure 2.1). The following is my reconstruction of their reasoning.

To begin, a team digs up teeth and records their location. They date the teeth using the geological context they find it in and other proxies for age, itself a complex procedure. At first, Fortelius and team look for unworn teeth, ideally second upper molars, of known species. They measure features of the teeth including the molar crown width and crown height and use these to classify the species into one of three classes of increasing height: brachydont, mesodont, hypsodont. Species with high molars are called hypsodont and show more hypsodonty than species with lower molars. They then assume that all members of a given species have teeth in the same height class. This allows them to track changes in tooth size using other databases of species, not necessarily based on tooth specimens. They also track the variation in found tooth specimens to observe evolution change and to limit their uniformity assumption.4

With this information in a database, biologists track changes in ungulate teeth shape over space and time. They have global data for the Neogene period from 24 to 2.5 million years ago. They then generalize from the data.5 One particular generalization is that mean hypsodonty increased in herbivores in the late Miocene (10.5–5 mya) in Europe (Fortelius et al. 2006).

Next, using a form of abductive inference, they hypothesize that increased hypsodonty in herbivores is an adaptation to eating plants in an environment with decreased water6 (Fortelius et al. 2002; 2006). Using this hypothesis, they infer to the condition: Water in the environment decreased in the late Miocene in Europe. Their full inference runs as follows:

The inference pattern used here is strong. Its basic form is:

A trait is adapted to an environment if it arose in that environment by natural selection. This is why adaptive traits are suited to being proxies, the only problem being that organisms can continue to carry traits after their environment changes. This issue shows that both the first premise and conclusion in an adaptation inference such as the teeth-water inference need to be carefully dated. The cost of a historical definition of adaptation is increased evidence needed to justify the adaptation hypothesis used as the second premise in an adaptation inference.7

Putting aside the significant empirical justification needed for the descriptive claim, the main scientific work in learning about the past using teeth is justifying the adaptation hypothesis that they abduced. This abduction does not itself give evidence, but rather guides their future research. To support the adaptation hypothesis and to make it exportable to other scientific contexts, Fortelius’s group tries to establish the following four claims:8

1. Establish that a correlation holds between mean hypsodonty and amount of water in other places and times.
2. Support that increased hypsodonty is functionally adaptive to eating plants when humidity decreases.
3. Support that increased hypsodonty was selected for in herbivores eating plants in Western Europe in the late Miocene given decreased humidity.
4. Establish the robustness of their conclusion using other proxies.

In the remainder of this section, I explain how they go about investigating these four claims. In addition to presenting the form of a strong proxy inference, I present enough detail about Fortelius’s research program to function as an exemplar to be followed when we move to scientific metaphysics. Together, this inference schema and these four claims form the basis of the proxy by adaptation framework that will be our guide to making good inferences in scientific metaphysics, as I show in section 3.

3. From Practice to Metaphysics

I began this chapter with a research proposal based on an analogy between the adaptation of organisms to the environment they live in and the adaptation of scientific practices to the world they investigate. We learned from analyzing paleobiological practice that there is a straightforward way to make inferences from traits of organisms to their environments using adaptation. And I presented four claims A–D that, when established, justify the adaptation inference: showing correlation, functional adaptation, selection, and robustness. My task in this section is to apply the analogy and its lessons to doing metaphysics: we should seek to establish metaphysical proxies in scientific practice on the basis of adaptation.

To show how to do this, in this section I fit as well as possible a case of scientific metaphysics from biological practice into the proxy by adaptation framework. My case study is Ken Waters’s (2017) argument for the no general structure thesis.

Waters (1994; 2004) has long investigated the practices of classical and molecular genetics from an epistemological perspective. He is interested in why genetics succeeds and proposes that genetics succeeds not because of a core theory but because of its investigative practices (Waters 2019). But more recently he went beyond an epistemological understanding of scientific success and entered the realm of metaphysics. In “No General Structure,” Waters (2017, 99) argues that the success of genetics is explained in part by the no general structure thesis: “Reality has lots of structure, but no overall structure.”

What exactly metaphysicians mean when they use “structure” is notoriously slippery. To gain enough traction to make progress fitting Waters’s reasoning for his conclusion into the proxy by adaptation framework, here is one way of understanding what Waters means by “structure.”18 We can distinguish between two kinds of structure: horizontal and vertical structure. Consider these concepts using an object that has one version of perfectly general structure—a Sierpinski triangle (Figure 2.2). A Sierpinski triangle is a fractal that is self-similar: if we divide the largest triangle into four triangles of equal area, each of the subtriangles has the same structure as the whole. And this continues at smaller and smaller scales because the pattern is repeated downward infinitely.

Horizontal structure is the structure at one spatial and temporal scale. About horizontal structure, we ask, Are their basic units at a scale? In a Sierpinski triangle, the answer is clearly yes at every scale. Vertical structure is the structure across scales. About vertical structure, we ask, Does the structure of one scale reduce to the structure of another scale? Again, in a Sierpinski triangle, the answer is clearly yes between any two scales. The question about structure for Waters is whether the parts of the world investigated by geneticists are structurally similar to this idealized fractal structure. While similarity is a quantitative relation, for simplicity I discuss it only as a qualitative relation of having or lacking horizontal or vertical structure. General structure is a claim about a degree of structure and can be applied to both horizontal and vertical structures.

I reconstruct Waters’s road to the no general structure thesis as follows:

• Step 1: Adopt the practice-centered view of science.
• Step 2: Analyze the practice of genetics from the practice-centered view.
• Step 3: Make inference 1.
• Step 4: Conclude first that the domain investigated by the practice of genetics lacks a horizontal structure.
• Step 5. Make inference 2.
• Step 6: Conclude second that the world lacks vertical structure.

Inference 1 to the first conclusion is where Waters infers from biological practice to scientific metaphysics. Inference 2 on my reconstruction is an inference from a metaphysical claim about one domain to a metaphysical claim about the whole domain. Only steps 1–4 centering on inference 1 and its sources of justification concern us here.

In step 1, Waters begins his reasoning from the practice-centered view of science. One purpose of this volume is to better explore and understand the practice-centered view of science. But for Waters, this view is about shifting the analysis of science from foregrounding theories as things to foregrounding practices as activities. A useful slogan for the practice-centered view is not theories but theorizing, not models but modeling, not experiments but experimenting.19

In step 2, Waters draws on his analyses of the practices of both classical and molecular genetics to make his argument. However, the basic form of his argument for the no general structure thesis can be seen from geneticists’ use of the molecular gene concept. Geneticists use the molecular gene concept to investigate the domains of heredity, development, and evolutionary change. I use “heredity” as a shorthand for all of these domains as Waters’s claims are the same for each.

Waters argues that geneticists use the molecular gene concept when they want to be precise: “A gene g for linear sequence l in product p synthesized in cellular context c is a potentially replicating nucleotide sequence, n, usually contained in DNA, that determines the linear sequence l in product p at some stage of DNA expression” (Waters 2017, 95). The practical consequence of using the molecular gene concept to understand heredity is that it parses different linear sequences into the genes for different cellular products. Therefore, there is no set of genes from which everything is constructed. Waters describes this using his epistemological concept of a fundamental unit: “What does it mean to say the gene is the fundamental unit of heredity? Presumably it means that if you could identify every gene and every difference in every gene, and if you could trace the transmission of each gene and each gene difference from one generation to the next, then you would have a comprehensive basis for understanding everything about heredity” (Waters 2017, 92). Waters argues that the molecular gene concept does not function as a fundamental unit of heredity because the molecular gene concept does not uniquely parse linear sequences into cellular products.

The conclusion of Waters’s analysis of the practice of genetics from the practice-centered view of science is the practice of genetics’ use of the molecular gene concept does not provide a single correct parsing of DNA into genes for understanding the domain of heredity.

In step 3, Waters draws on his analysis of practice to infer his first metaphysical conclusion. His conclusion is the domain of heredity lacks a horizontal structure. But we are interested in his argument for this claim. Waters twice invokes adaptation to explain his conclusion: “My metaphysical claim is that scientific practices in genetics and allied sciences take this form because they are adapted to a reality that has no overall structure. The reality has lots of structure, but no overall structure. Practice has been adapted to work in the reality of the world that biologists are engaging” (Waters 2017, 98). While Waters uses the language of “adaptation,” he does not elaborate on what he means by it or what work it is doing in his argument. Such loose talk of adaptation and fit is common among philosophers when discussing the relationships between things and the world. Drawing on the analysis of paleobiological practice though, we can investigate how the reasoning goes if we read “adaptation” as analogous to biological adaptation. Waters’s adaptation hypothesis about genetics and structure then is the practice of genetics’ use of the molecular gene concept is an adaptation to manipulation in a domain that lacks a horizontal structure.

Waters’s inference 1 then runs as follows:

For our purposes here, we can take for granted that the descriptive claim in the first premise is justified and focus on the justification for the adaptation hypothesis in the second premise. Waters himself argues for premise 1 extensively but does not provide any direct evidence for premise 2. Applying the proxy by adaptation framework we developed from paleobiology, four claims need to be evidenced for this inference to run smoothly and the conclusion to be exportable. They are:

1. Establish that a correlation holds between parsing the domain and horizontal structure in other, independent practices and domains.
2. Support that the molecular gene concept is functionally adaptive to manipulation in a domain without a horizontal structure.
3. Support that the molecular gene concept was selected for in genetics for the purpose of manipulation given no horizontal structure.
4. Establish the robustness of the lack of a horizontal structure in heredity, development, or evolutionary change.

In the remainder of this section, I discuss the prospects for showing these four claims. Waters’s argument is my special target, but I also speak more generally about meeting these standards for any metaphysical argument.

4. Commentary

In this section, I comment on choices I have made regarding inference, success, and adaptation in developing this proposal.

5. Concluding Remarks

This chapter proposes a new methodology for practicing metaphysics. To found a research program, we practice-based scientific metaphysicians must work out an exemplar to build on and modify going forward. Thankfully, rather than inventing a methodology de novo, we can co-opt and adapt the proxy by adaptation framework used by biologists to make inferences to unknown environments.

Whether the scientific and metaphysical problems should be regarded as analogous or rather the same but in different domains depends on the relationship between science and metaphysics. Considering my proposal at the meta-level, we are left with a picture of metaphysics as continuous with science on both epistemic and ontological dimensions. With its emphasis on inference, we can see this proposal in a long line of inferences based on a dualist ontology. For example, Locke asked how we can infer the properties of external objects from our experiences, leaving him open to strong Cartesian skepticism. However, my proposal is dualistic, not in a mind versus matter sense but in an organism and environment sense. A biological organism and its environment are mutually interacting and only pragmatically distinguished. While I have focused on learning about environments from organisms, we might also learn about organisms from environments.

Our picture of metaphysical knowledge is as with normal scientific inquiry: fallible but improving. The main difference is the dearth of data on scientific practice. We lack analyses, but more than that, we lack scientific diversity in the world. Metaphysicians can reason outward from ordinary experience much like how historical scientists reason backward from the present day using tools and methods developed using iteration and actualism. Skepticism is not in principle limiting to us; it is just that there are not enough possible data to know that much. But we will improve.

To conclude, I address three questions for this proposal. First, what kind of metaphysics can scientific practice inform us about? Second, what is the purpose of investigating metaphysical questions? And third, how does scientific metaphysics fit into the larger project of understanding how science works?

First then, what kind of metaphysics can scientific practice inform us about? As a methodological proposal for how to make strong inferences, it says nothing alone about the scope of metaphysics or scientific metaphysics. With the proxy by adaptation framework, we can only gain evidence for metaphysical claims that have some practical consequence for how scientific practice works. If this is the only methodology admitted, then practical consequences are the mark of knowable metaphysical truths. In this way, the proposal has an affinity with the pragmatist and positivist idea, uncontroversial within science, that the meaningful hypotheses are those that can be empirically verified.

What then is the scope of metaphysical claims that we can evidence with the proxy by adaptation framework? A better answer to this question will come from developing the metaphysical research program further, but some broad outlines can already be seen. On the one hand, first-order ontological claims such as “atoms exist” and “mushrooms are biological individuals” must make differences on scientific practices. If proxies for these kinds of claims cannot be developed, then the present proposal has nothing to offer here. On the other hand, metaphysical positions such as idealism and nominalism about the interpretation of first-order claims will probably not make differences to scientific practices. The proxy by adaptation framework is compatible with any metaphysical position that can make sense of first-order ontological claims and of selection operating on scientific practices. But it is unable to tell us about the natures of things that go beyond their practical effects in the same way that science is.

Second, what is the purpose of investigating metaphysical questions? I want to address whether the project of scientific metaphysics from biological practice is dualistic in the sense of Penelope Maddy’s (2007) two-tier views. Maddy distinguishes between philosophers like Descartes for whom philosophical questions are continuous with and even arise from his scientific practice (one-tier view) and philosophers like Kant, Carnap, and van Fraassen for whom philosophical inquiry floats freely above science in asking questions about the interpretation and import of scientific fruits (two-tier views).

The present proposal fits neatly into neither one-tier nor two-tier views of philosophical inquiry as I understand them. It is like a two-tier view in seeming to imply that biologists themselves cannot say what is real in the biological world. It is like a one-tier view in using broadly the same forms of inquiry and asking similar questions as scientists—what is the structure of an organism, ecosystem, or whole domain?

Let us compare my proposal with the use of proxy-based reasoning from linguistic behavior. Anthropologist Caleb Everett and linguists Damian Blasí and Seán Roberts (2016) argue that the vowel structure of some spoken languages is adapted to the physical environment in which the language evolved and operates. In particular, they argue that desiccation of the environment will push languages against having a complex tonal system. They do not argue further for the tonality of a language as a proxy for humidity level in an environment, but it is possible that they could in the future.

Their article is the target article in the first volume and issue of the Journal of Language Evolution, with eleven commentaries on it. The commentaries are divided between internal methodological critiques of their analyses and external critiques about the very idea of language being adapted to the environment. The question I want to raise is whether these scientists should be characterized with a one-tier or two-tier view of linguistics. And my answer is that it isn’t obvious that they fit into either neatly and further that the very question is of limited use. I see this question primarily about disciplinization because whether or not they are extra-linguist and asking meta-linguistic questions, their questions, inquiry, and subject matter are clearly scientific.24

Similarly then with scientific metaphysics from biological practice and biology, these are both different enterprises but both scientific. Further, they might interact the way that many distinct disciplines interact. The two-tier view of science and metaphysics is behind such quips as “philosophy is useful to science like ornithology is to birds.” Metaphysics via a study of scientific practice is partially disconnected from the science. Scientists do not need to understand how science changes over time to conduct normal science, just as language users do not need to understand how language changes over time to have normal conversations. But metaphysics is continuous with science in that it comes from a scientific study of science as in linguistics.

Third, how does scientific metaphysics fit into the larger project of understanding how science works? Questions about scientific change and methodology are often characterized as epistemological and opposed to metaphysical questions about scientific realism because the two lines of thought are historically independent.

We should include how the world is in our analyses of scientific change and methodology. With a pragmatic approach to understanding scientific practice, the basic unit of analysis is: Scientists use tools and methods to investigate aspects of the world for particular purposes. This means that understanding scientific change and methodology is a relative significance problem (Beatty 1997) and we need to apportion relative responsibility to the following: scientists and the social structure of science, tools and methods, the world, and purposes. We should cite aspects of the parts of the world being investigated in explanations of why particular scientific practices work the ways that they do. Normal empirical features of the systems being studied influence how to investigate the system. Once we arrive at tentative metaphysical features of the domain being investigated by a scientific practice, we can use these features to explain why the practice is structured the way it is. This is the normal circularity of induction and deduction, and it is only vicious when the potential deductive strength is counted as inductive evidence as in IBE.

For example, Currie and Walsh (2018) have proposed an ontic-driven account of explanations of scientific methodology and used the differences between the motion of massive bodies and light to explain why Newton’s methodology differs in mechanics and optics. A partially ontic account of scientific practice should be the norm, and whether a given practice is ontic- or socially driven is an empirical question. This problem is different from the standard realism versus anti-realism debate in being generally open to realism but considering realism problems about what exists and what it is like.

Acknowledgments

Oliver Lean and I together came up with the initial idea for taking adaptation to be the relation between a successful practice and the world in Basel in August 2016.

I presented earlier versions of this chapter at the Central APA in St. Louis in 2016, to the From Biological Practice to Scientific Metaphysics Taiwan summer school in 2018, and to the Lake Geneva Biology Interest Group in 2019. I thank the audiences for their comments and questions.

I thank Ken Waters, Alan Love, Bill Wimsatt, Marcel Weber, and Ruey-Lin Chen for their feedback and encouragement.

I thank Mikael Fortelius and Indrė Žliobaitė especially for their comments on the use of teeth as an environmental proxy in section 2.

I have talked about this with just about everyone I know over the past seven years, and I thank you all for our conversations.

Notes

1. 1. Wallace’s reasoning and evidence were refined by Darwin (1862) and many others, and conclusive proof that Angraecum sesquipedale is pollinated in the wild by the Xanthopan morganii praedicta actually only came in 1992 (Wasserthal 1997).

2. 2. See MacCord 2017 for a history of morphological research on mammalian teeth.

3. 3. See Currie 2018 for how a new species of platypus was reconstructed from a single tooth.

4. 4. From personal communication with Fortelius, January 2019.

5. 5. They first make a data model that groups the raw data: periodization into the MN system; tooth size into brachydont, mesodont, hypsodont; and so on. But they also use raw data for quantitative analyses. They also make decisions about which species and taxa to include in the analysis.

6. 6. See the commentary that follows for my comment on the use of abduction versus inference to the best explanation and why I favor the former. Briefly, IBE takes the explanatory power of a hypothesis as support for the hypothesis. Peircian abduction, however, takes the explanatory power of a hypothesis only as reason to pursue independent evidence for it.

7. 7. Following Sober’s (1984), “A is an adaptation for task T in population P if and only if A became prevalent in P because there was selection for A, where the selective advantage of A was due to the fact that A helped perform task T.” Gould and Lewontin used an ahistorical definition of adaptation in Gould and Lewontin 1979. See Forber 2013 on different definitions of adaptation.

8. 8. For an excellent analysis of how to study adaptation, see Olson and Arroyo-Santos 2015. These biologists focus on the different methods for studying adaptation: comparative, populational, and optimality. They also implicitly provide a strong defense of why IBE is not used by biologists here because of circularity.

9. 9. Another solution is to find correlations using an independent proxy for the environmental condition being investigated. However, this pushes the problem back and makes the reasoning more complex. It also blurs the distinction I make between establishing a correlation and showing that the conclusion reached is robust.

10. 10. Some such studies on other animals include the following: Schulz et al. 2013; Merceron et al. 2016; Müller et al. 2014; Karme et al. 2016; Winkler et al. 2019.

11. 11. The sense of community here is just the species which live together and are at the same trophic level, a non-equilibrium concept of community.

12. 12. “The main assumption is that there are already superior competitors in those environments which prevent the migrants establishing themselves” (Fortelius, personal communication, January 2019). They discuss aspects of this as macroevolutionary source-sink dynamics in a review paper (Fortelius et al. 2014), and they have modeled adaptation to harshening conditions in the context of the “species factory” (Fortelius et al. 2015).

13. 13. Called “dental senescence,” this has been shown in lemurs (King et al. 2005).

14. 14. Specialization is another possible outcome of the selection pressure. Giraffes eating leaves high up in the trees are brachydonts (Fortelius, personal communication, January 2019).

15. 15. I am not aware of these or any other scientists specially arguing for ecological adaptation via convergence. But not many biologists talk about ecological adaptation at all.

    It would work in the analogous way though. A trait in two communities is probably the result of ecological selection because the communities both live in similar environments and are independent. The trait would be ecologically superior. You would need to find recent examples, before evolution could turn the ecological adaptation into an evolutionary adaptation.

16. 16. Whether it is a convergence argument for using the proxy or a measurement of the proxy relationship is a tricky question that depends on the stage of inquiry, standing evidence, and purpose.

    While epistemically philosophers would like proxies and indeed models and instruments to be justified before they are used to measure things, in practice these steps are not temporally sequential but operate in a feedback relationship.

17. 17. This process happens for all scientific instruments. Chang (2004) describes it for thermometers. Van Fraassen (2010) discusses this in terms of the Problem of Coordination or how abstract signs represent what they represent.

18. 18. I developed the following understanding through comments Waters has made informally. They should not be construed as his view.

19. 19. The extent to which the practice-centered view takes us beyond theorizing, modeling, and experimenting is an open question. For many philosophers of science, it should do so, but the points are independent.

20. 20. Gillet’s distinction is related to Carnap’s distinction between internal and external questions, but Gillet does not share Carnap’s view that external questions are either meaningless or practical.

21. 21. Peirce, “Lecture V,” Lectures on Pragmatism (1903, CP 5.189). See Mcauliffe (2015) for how Peircian abduction got confused with IBE. Also, Peirce said a lot of things about abduction, and this formalization of it into inference form only captures some of it.

22. 22. I thank Reuy-Lin Chen in particular for his comments on the importance of emphasizing practical success.

23. 23. The most helpful survey of evolutionary epistemology for me was Renzi and Napolitano 2011.

24. 24. Another example is the Forman thesis (Forman 1971) and its reception by historians of science. I see the idea that the Copenhagen interpretation of quantum mechanics ascended because of the environment in the Weimar Republic as a clear historical hypothesis. Some historians are upset by these kinds of hypotheses, as are some linguists. But if you accept that language and science are historically evolving, then they are valid hypotheses. The issue is evidence.

References

1. Beatty, John. 1997. “Why Do Biologists Argue Like They Do?” Philosophy of Science 64 (Proceedings): S432–43.
2. Bender, Michael, Todd Sowers, and Edward Brook. 1997. “Gases in Ice Cores.” Proceedings of the National Academy of Sciences 94, no. 16: 8343–49.
3. Campbell, Donald T., and Donald W. Fiske. 1959. “Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix.” Psychological Bulletin 56, no. 2: 81.
4. Chang, Hasok. 2004. Inventing Temperature: Measurement and Scientific Progress. Oxford: Oxford University Press.
5. Chang, Hasok. 2007. “Scientific Progress: Beyond Foundationalism and Coherentism 1.” Royal Institute of Philosophy Supplements 61: 1–20.
6. Currie, Adrian. 2018. Rock, Bone, and Ruin: An Optimist’s Guide to the Historical Sciences. Cambridge, Mass.: MIT Press.
7. Currie, Adrian, and Kirsten Walsh. 2018. “Newton on Islandworld: Ontic-Driven Explanations of Scientific Method.” Perspectives on Science 26, no. 1: 119–56.
8. Damuth, J., M. Fortelius, P. Andrews, C. Badgley, E. A. Hadley, S. Hixon, C. M. Janis, R. H. Madden, K. Reed, and J. M. Smith. 2002. “Reconstructing Mean Annual Precipitation, Based on Mammalian Dental Morphology and Local Species Richness.” Journal of Vertebrate Paleontology 22, no. 3: 48A.
9. Darwin, Charles. 1862. On the Various Contrivances by Which British and Foreign Orchids Are Fertilised by Insects: And on the Good Effect of Intercrossing. London: John Murray.
10. Eronen, J. T., Kai Puolamäki, L. Liu, K. Lintulaakso, J. Damuth, C. Janis, and M. Fortelius. 2010a. “Precipitation and Large Herbivorous Mammals I: Estimates from Present-Day Communities.” Evolutionary Ecology Research 12, no. 2: 217–33.
11. Eronen, J. T., Kai Puolamäki, L. Liu, K. Lintulaakso, J. Damuth, C. Janis, and M. Fortelius. 2010b. “Precipitation and Large Herbivorous Mammals II: Application to Fossil Data.” Evolutionary Ecology Research 12, no. 2: 235–48.
12. Eronen, Jussi T., P. David Polly, Marianne Fred, John Damuth, David C. Frank, Volker Mosbrugger, Christoph Scheidegger, Nils Chr Stenseth, and Mikael Fortelius. 2010. “Ecometrics: The Traits That Bind the Past and Present Together.” Integrative Zoology 5, no. 2: 88–101.
13. Everett, Caleb, Damián E. Blasí, and Seán G. Roberts. 2016. “Language Evolution and Climate: The Case of Desiccation and Tone.” Journal of Language Evolution 1, no. 1: 33–46.
14. Forber, Patrick. 2013. “Debating the Power and Scope of Adaptation.” In The Philosophy of Biology, edited by Kostas Kampourakis, 145–60. Dordrecht: Springer.
15. Forman, Paul. 1971. “Weimar Culture, Causality, and Quantum Theory, 1918–1927: Adaptation by German Physicists and Mathematicians to a Hostile Intellectual Environment.” Historical Studies in the Physical Sciences 3: 1–115.
16. Fortelius, Mikael. 1985. “Ungulate Cheek Teeth: Developmental, Functional, and Evolutionary Interrelations.” Acta Zoologica Fennica 180: 1–76.
17. Fortelius, Mikael, Jussi Eronen, Jukka Jernvall, Liping Liu, Diana Pushkina, Juhani Rinne, Alexey Tesakov, Inesa Vislobokova, Zhaoqun Zhang, and Liping Zhou. 2002. “Fossil Mammals Resolve Regional Patterns of Eurasian Climate Change over 20 Million Years.” Evolutionary Ecology Research 4, no. 7: 1005–16.
18. Fortelius, Mikael, Jussi Eronen, Liping Liu, Diana Pushkina, Alexey Tesakov, Inesa Vislobokova, and Zhaoqun Zhang. 2006. “Late Miocene and Pliocene Large Land Mammals and Climatic Changes in Eurasia.” Palaeogeography, Palaeoclimatology, Palaeoecology 238, no. 1–4: 219–27. http://dx.doi.org/10.1016/j.palaeo.2006.03.042.
19. Fortelius, Mikael, Jussi T. Eronen, Ferhat Kaya, Hui Tang, Pasquale Raia, and Kai Puolamäki. 2014. “Evolution of Neogene Mammals in Eurasia: Environmental Forcing and Biotic Interactions.” Annual Review of Earth and Planetary Sciences 42, no. 1: 579–604. https://doi.org/10.1146/annurev-earth-050212-124030.
20. Fortelius, Mikael, Stefan Geritz, Mats Gyllenberg, Pasquale Raia, and Jaakko Toivonen. 2015. “Modeling the Population-Level Processes of Biodiversity Gain and Loss at Geological Timescales.” The American Naturalist 186, no. 6: 742–54. https://doi.org/10.1086/683660.
21. Gillet, Carl. Forthcoming. “Using Compositional Explanations to Understand Compositional Levels: Why Scientists Talk about ‘Levels’ in Human Physiology.” In Hierarchy and Levels of Organization in the Biological Sciences, edited by Daniel Brooks, James DiFrisco and William C. Wimsatt. Cambridge, Mass.: MIT Press.
22. Godfrey-Smith, Peter. 2007. “Conditions for Evolution by Natural Selection.” The Journal of Philosophy 104, no. 10: 489–516.
23. Gould, Stephen Jay, and Richard C. Lewontin. 1979. “The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme.” Proceedings of the Royal Society of London B: Biological Sciences 205, no. 1161: 581–98.
24. Gould, Stephen Jay, and Elisabeth S. Vrba. 1982. “Exaptation—A Missing Term in the Science of Form.” Paleobiology 8, no. 1: 4–15.
25. Hacking, Ian. 1983. Representing and Intervening: Introductory Topics in the Philosophy of Natural Science. Cambridge: Cambridge University Press.
26. Herschel, John Frederick William. 1830/1987. A Preliminary Discourse on the Study of Natural Philosophy. Chicago: University of Chicago Press.
27. Janis, Christine M., and Mikael Fortelius. 1988. “On the Means Whereby Mammals Achieve Increased Functional Durability of Their Dentitions, with Special Reference to Limiting Factors.” Biological Reviews 63, no. 2: 197–230.
28. Jernvall, Jukka, and Mikael Fortelius. 2002. “Common Mammals Drive the Evolutionary Increase of Hypsodonty in the Neogene.” Nature 417, no. 6888: 538–40.
29. Karme, Aleksis, Janina Rannikko, Aki Kallonen, Marcus Clauss, and Mikael Fortelius. 2016. “Mechanical Modelling of Tooth Wear.” Journal of the Royal Society Interface 13, no. 120: 20160399.
30. King, Stephen J., Summer J. Arrigo-Nelson, Sharon T. Pochron, Gina M. Semprebon, Laurie R. Godfrey, Patricia C. Wright, and Jukka Jernvall. 2005. “Dental Senescence in a Long-Lived Primate Links Infant Survival to Rainfall.” Proceedings of the National Academy of Sciences of the United States of America 102, no. 46: 16579–83. https://doi.org/10.1073/pnas.0508377102.
31. Kuhn, Thomas S. 1962. The Structure of Scientific Revolutions. Chicago: University of Chicago Press.
32. Lewontin, Richard C. 1970. “The Units of Selection.” Annual Review of Ecology and Systematics 1, no. 1: 1–18.
33. Liu, Liping, Kai Puolamäki, Jussi T. Eronen, Majid M. Ataabadi, Elina Hernesniemi, and Mikael Fortelius. 2012. “Dental Functional Traits of Mammals Resolve Productivity in Terrestrial Ecosystems Past and Present.” Proceedings of the Royal Society B: Biological Sciences 279, no. 1739: 2793–99.
34. MacCord, Katherine. 2017. “Development, Evolution, and Teeth: How We Came to Explain the Morphological Evolution of the Mammalian Dentition.” PhD diss., History and Philosophy of Science, Arizona State University.
35. Maddy, Penelope. 2007. Second Philosophy: A Naturalistic Method. Oxford: Oxford University Press.
36. Mcauliffe, William H. B. 2015. “How Did Abduction Get Confused with Inference to the Best Explanation?” Transactions of the Charles S. Peirce Society: A Quarterly Journal in American Philosophy 51, no. 3: 300–319.
37. Merceron, Gildas, Anusha Ramdarshan, Cécile Blondel, Jean-Renaud Boisserie, Noël Brunetiere, Arthur Francisco, Denis Gautier, Xavier Milhet, Alice Novello, and Dimitri Pret. 2016. “Untangling the Environmental from the Dietary: Dust Does Not Matter.” Proceedings of the Royal Society B: Biological Sciences 283, no. 1838: 20161032.
38. Miller, Boaz. 2016. “What Is Hacking’s Argument for Entity Realism?” Synthese 193, no. 3: 991–1006.
39. Müller, Jacqueline, Marcus Clauss, Daryl Codron, Ellen Schulz, Jürgen Hummel, Mikael Fortelius, Patrick Kircher, and Jean-Michel Hatt. 2014. “Growth and Wear of Incisor and Cheek Teeth in Domestic Rabbits (Oryctolagus Cuniculus) Fed Diets of Different Abrasiveness.” Journal of Experimental Zoology Part A: Ecological Genetics and Physiology 321, no. 5: 283–98.
40. Novick, Aaron. 2017. “Metaphysics and the Vera Causa Ideal: The Nun’s Priest’s Tale.” Erkenntnis 82, no. 5: 1161–76.
41. Novick, Aaron, and Raphael Scholl. 2017. “Presume It Not: True Causes in the Search for the Basis of Heredity.” The British Journal for the Philosophy of Science 71, no. 1: 59–86.
42. Olson, Mark E., and Alfonso Arroyo-Santos. 2015. “How to Study Adaptation (and Why to Do It That Way).” The Quarterly Review of Biology 90, no. 2: 167–91.
43. Paul, Laurie A. 2012. “Metaphysics as Modeling: The Handmaiden’s Tale.” Philosophical Studies 160, no. 1: 1–29.
44. Peirce, Charles S. 1903. “Lecture V.” Harvard Lectures on Pragmatism. MS [R] 312.
45. Polly, P. David, Jussi T. Eronen, Marianne Fred, Gregory P. Dietl, Volker Mosbrugger, Christoph Scheidegger, David C. Frank, John Damuth, Nils C. Stenseth, and Mikael Fortelius. 2011. “History Matters: Ecometrics and Integrative Climate Change Biology.” Proceedings of the Royal Society B: Biological Sciences 278: 1131–40. https://doi.org/10.1098/rspb.2010.2233.
46. Quine, Willard V. 1948. “On What There Is.” The Review of Metaphysics 2, no. 5: 21–38.
47. Renzi, Barbara Gabriella, and Giulio Napolitano. 2011. Evolutionary Analogies: Is the Process of Scientific Change Analogous to the Organic Change? Newcastle upon Tyne: Cambridge Scholars Publishing.
48. Schulz, Ellen, Vanessa Piotrowski, Marcus Clauss, Marcus Mau, Gildas Merceron, and Thomas M. Kaiser. 2013. “Dietary Abrasiveness Is Associated with Variability of Microwear and Dental Surface Texture in Rabbits.” PLoS One 8, no. 2: e56167.
49. Sloep, Peter B. 1993. “Methodology Revitalized?” The British Journal for the Philosophy of Science 44, no. 2: 231–49.
50. Sober, Elliott. 1984. The Force of Selection: Evolutionary Theory in Philosophical Focus. Chicago: University of Chicago Press.
51. Van Fraassen, Bas C. 2010. Scientific Representation: Paradoxes of Perspective. Oxford: Oxford University Press.
52. Vellend, Mark. 2010. “Conceptual Synthesis in Community Ecology.” The Quarterly Review of Biology 85, no. 2: 183–206.
53. Wallace, Alfred Russel. 1867. “Creation by Law.” The Quarterly Journal of Science 4, no. 16: 470–88.
54. Wasserthal, L. T. 1997. “The Pollinators of the Malagasy Star Orchids Angraecum Sesquipedale, A. Sororium and A. Compactum and the Evolution of Extremely Long Spurs by Pollinator Shift.” Plant Biology 110, no. 5: 343–59.
55. Waters, C. Kenneth. 1994. “Genes Made Molecular.” Philosophy of Science 61, no. 2: 163–85.
56. Waters, C. Kenneth. 2004. “What Was Classical Genetics?” Studies in History and Philosophy of Science Part A 35, no. 4: 783–809.
57. Waters, C. Kenneth. 2017. “No General Structure.” In Metaphysics and the Philosophy of Science: New Essays, edited by Matthew Slater and Zanja Yudell. New York: Oxford University Press.
58. Waters, C. Kenneth. 2019. “Presidential Address PSA 2016: An Epistemology for Scientific Practice.” Philosophy of Science 86, no. 4: 585–611.
59. Wimsatt, William C. 1994. “The Ontology of Complex Systems: Levels of Organization, Perspectives, and Causal Thickets.” Canadian Journal of Philosophy 24, no. S1: 207–74.
60. Winkler, Daniela E., Ellen Schulz-Kornas, Thomas M. Kaiser, Annelies De Cuyper, Marcus Clauss, and Thomas Tütken. 2019. “Forage Silica and Water Content Control Dental Surface Texture in Guinea Pigs and Provide Implications for Dietary Reconstruction.” Proceedings of the National Academy of Sciences 116, no. 4: 1325–30.
61. Žliobaitė, Indrė, Janne Rinne, Anikó B. Tóth, Michael Mechenich, Liping Liu, Anna K. Behrensmeyer, and Mikael Fortelius. 2016. “Herbivore Teeth Predict Climatic Limits in Kenyan Ecosystems.” Proceedings of the National Academy of Sciences 113, no. 45: 12751–56.