Transformation, Persistence, and Identity

Guest blogger Matt Haber writes...

Consider the following two questions: “How much can something change before it becomes a new kind of thing?” and “What makes two distinct things the same [kind of] thing?” Questions like these should look familiar not only to philosophers but also to biologists, though they might approach them in very different ways. For this post, I want to explore one way biologists answer questions like these, and how paleontology, in particular, can provide useful insights for philosophers.

This utility is on display in Kevin Padian and John R. Horner’s (2002) “Typology versus transformation in the origin of birds.” In it, they provide an argument defending the hypothesis that birds are dinosaurs. Here I want to focus more on how they make that argument, rather than the substance of it. The goal is to highlight the intersection of paleontology and philosophy of biology. To do that, I’ll first isolate some philosophical arguments in Padian and Horner’s paper, and describe how these exemplify important ways of doing philosophy of biology. I’ll then take a step back to offer a broader picture, and identify a fruitful (though speculative) philosophical question raised by work like Padian and Horner’s. The point, again, is not so much to try an answer this question, but to consider it as an example of how paleontology provides opportunities to make progress on long-entrenched, traditional problems in philosophy. To wrap up, I’ll provide some comments on how this paper represents useful ways we might use paleontology to teach philosophy.

The Philosophy of Biology Project

In talks I give to general audiences, I often start with a description of what I view as four central projects in philosophy of biology: (1) identifying core commitments of biologists; (2) participating in conceptual debates in biology; (3) acting as science critics to help biologists; and (4) drawing on biology for philosophy (see also Haber 2014, p. 886). This isn’t intended to be an exhaustive list, and perhaps we’ll see other examples of the sorts of projects philosophers of biology do in the comments. Yet I find that this is a useful way to provide entry points for audiences that might otherwise be unfamiliar with the sort of work we do, and it’s easy to list examples for each. For some fields of biology these examples can be very subtle, or require serious excavation to help people see the philosophical content. Systematics, especially in the context of paleontology, is often the opposite. In many cases, these biologists can not avoid the underlying conceptual and philosophical debates; to do biology well here, is to also do philosophy.

Padian and Horner’s paper exemplifies this intersection of philosophy and paleontology in the context of the questions I posed above. The proximate discussion is over the origin and evolution of modern birds, and of whether we ought to think of them as distinct from extinct dinosaurs. At stake is both how much (and in what ways) a thing can change before becoming a new thing, and what makes two distinct things the same kind of thing. Philosophers should recognize these as questions of identity and persistence (or how identity is maintained over time).

In both cases, Padian and Horner argue that central to this dispute is a philosophical disagreement over typological and transformational approaches to taxonomy. This explicitly grounds their arguments in a theoretically informed thesis about identity. Namely, they are working in the context of phylogenetic systematics. This also provides a conceptual basis for interpreting the evidence in the dispute over the taxonomic placement of modern birds.

An example of a Ceratopsid cladogram,  from the University of Utah's NHMU (photo by author)

An example of a Ceratopsid cladogram,  from the University of Utah's NHMU (photo by author)

Researchers working in other theoretical perspectives may provide very different interpretations of those very same data. In important ways, this isn’t so much a disagreement over the evidence, but a symptom of having an entirely different discussion of those data. In this case, that would include researchers who have not adopted phylogenetic systematics. At this point, that would count very few systematists (or other biologists working in taxonomy). Yet philosophers, on the whole, are generally ignorant of phylogenetics, and what it entails for the metaphysics of identity and persistence.[1] This isn’t intended to be a knock on philosophers; I’m not sure we should expect, say, metaphysicians to keep up on the biological literature. Instead, it’s to identify that an important contribution we philosophers of biology can make to philosophy more generally is to draw attention to biology in relevant ways. At the same time, we can contribute to biology by identifying cases like these to biologists, namely, those disputes that are really two parallel discussions, where what is at stake is not a matter of evidence, but underlying conceptual differences. Too often that can go unrecognized, which can undermine fruitful research programs.

So far, this has been pretty abstract. Let’s unpack some of these claims a bit by looking at the specifics of Padian and Horner’s arguments.

Philosophy in Paleontology

To bring it back to the concrete, and back to paleontology, let’s take as a starting point a pretty straight-forward observation offered by Padian and Horner (p. 121):

Nowhere is the contrast between typology and transformation as clear as in the time-honored division between ‘birds’ and ‘reptiles’. What could present a clearer dichotomy than a lizard and a warbler — one a cold-blooded creeper, the other a warm-blooded flyer? But include dinosaurs in the mix and the typology breaks down.

They are far from the first to notice this, either in the specific case or as a general rule. Fossils provide compelling evidence of the gradient nature of biological categories. Groups that appeared as clean, distinct categories turn out to be temporal ends of a gradient. That’s just a feature of systems that evolve from common ancestors. Padian and Horner see this as grounds for adopting what they call transformationist approaches against typological ones; philosophers might be more familiar with the terms gradient and categorical, respectively. Regardless, the challenge is the same: how to think about identity and persistence in such systems.

Let’s start with their notion of identity. Put simply, identity gets tied to unique evolutionary histories, rather to any set of characters or traits possessed by a lineage. (And this is putting it simply. Careers have been made unpacking this.) This is a pretty straight- forward metaphysical claim, though one couched in phylogenetic systematics. I don’t think the fact of the latter, though, should tell against the former.

Padian & Horner's figure illustrating the difference between typology and transformation.

Padian & Horner's figure illustrating the difference between typology and transformation.

Let’s consider some of the implications of thinking about identity in a transformational, historical way, rather than as a typological, categorical, static concept. For one, it implies that the perceived boundaries between species will change over the time. This is just what we see when we survey groups of living things. As novel unique characters (what phylogeneticists call derived traits) appear in a distinct lineage, these may be used to differentiate those lineages more clearly and starkly. These derived characters may even become more prominent than those features the lineages share with sister lineages due to a common ancestor (what phylogeneticists call primitive traits). On the other hand, it’s entirely possible that those primitive traits will remain prominent, and that some derived ones will be lost. This might complicate the task of differentiating closely related lineages. Part of the job of systematists is to reconstruct that evolutionary history, and to identify those characters that identify a unique and distinct history.

Of course, it’s not just about our ability to perceive that distinction. Barring transfer between lineages, characters that evolve along a lineage are going to be unique to that lineage.[2] It’s that uniqueness that constitutes an important component of identity; it is part of what makes a lineage distinct from all others. When these traits (or traits that uniquely evolve from those) are carried forward in that lineage, so is that uniqueness. This, in part, provides the criteria for delineating lineages from one another. Together, this delineated unique history constitutes an important notion of biological identity—a notion that extends beyond phylogenetic systematics (e.g., in immunology Pradeu 2012).

What about persistence? Modern birds have clearly transformed enormously from their extinct dinosaur relatives. Why not stipulate that this amount of change is significant enough that we ought to recognize birds as a taxon distinct from and taxonomically equivalent to the dinosaurs, rather than as a unique group of dinosaurs? Surely it’s at least consistent with transformationist thinking that some transformations are so deep that they lead to a change of identity. That is, why not take the evolutionary change we see as marking a change in identity?

Carl Zimmer's Avian + non-Avian Dinosaur Phylogeny.

Carl Zimmer's Avian + non-Avian Dinosaur Phylogeny.

There are a number of ways to answer this question that turn on philosophical commitments. For example, we might consider whether one feature that distinguishes higher taxa from organisms (or evolutionary from developmental change) is a capacity for open-ended change. Or we might recognize that once we link identity to ancestry, that the amount of change in characters is irrelevant to persistence (after all, the one property that matters — ancestry — will never change). Elements of both of these may be found in Padian and Horner’s argument.

Instead, let’s focus on the phrase “birds are dinosaurs.” In very broad terms, identity claims like this can be read as claims about what group something belongs to. In an introduction to logic course, for example, this sort of sentence would be treated as a claim that all birds have some property that makes them a dinosaur (which would get treated as a predicate). Hopefully by now it’s becoming evident that this interpretation implies typological thinking, i.e., that there is some key feature or intrinsic character that makes something a dinosaur. So that can’t be how we interpret the phrase “birds are dinosaurs.”

Instead, identity claims like, “birds are dinosaurs” need to be read in the context of phylogenetics. But if belonging to a group isn’t about possessing key features, what might it be? Padian and Horner tell us directly (p. 120):

Cladistics groups organisms only by new features that are identified in the descendants of a common ancestor.

This is one of the core commitments of phylogenetics: A taxonomic group is just all and only the descendants of some common ancestor, and that taxa are grouped in nested hierarchies based on common ancestry. That is, belonging to a group is about common ancestry; identity is a relational, temporal property, not a static, intrinsic one. And once you are a part of a group like that, no amount of transformation will remove you from it.[3]

This discussion on language leads to a philosophical observation Padian and Horner make about methodology. Namely, they recognize that the central debate about the taxonomic placement of modern birds is not merely a dispute over evidence, but of how to interpret that evidence. The methodological point is that when this happens, the participants are in danger of talking past one another, rather than directly engaging the respective arguments. Let’s quickly take a look.

Early birds diverged from extinct dinosaurs, as part of the theropod clade. Those very early birds would have resembled other theropods in just about every way; it takes time for a distinct lineage to, well, evolve distinct characters. For Padian and Horner, it’s not about whether a distinct lineage exhibits any distinct characters, but whether it is a unique lineage following its own trajectory. Being a unique taxon is to have a unique historical trajectory, in a highly specified way. In contrast, systematists committed to typology will define unique taxa by the characters they possess. Taxonomic divergence will not be cast in terms of history, but in terms of characters. In this way, transformational and typological taxonomists may agree about what the data tell them about the characters and history of two taxa, but interpret those as supporting conflicting taxonomic hypotheses. Without recognizing the underlying, foundational differences in conceptual commitments, those conflicting interpretations will be misunderstood as straight disagreements, rather than as talking past one another.

Using Fossils to Disrupt Entrenched Views

There are a number of lessons philosophers might take from a case like this in paleontology. There’s a rich literature, for example, on how evolutionary theory informs us about about natural kinds. As mentioned above, paleontology provides a rich source for this discussion, e.g., in the way it undermines categorical thinking. If contemporary views about natural kinds are going to apply to biology, they must account for this in some sophisticated way. I’ve argued elsewhere that this presents a choice to philosophers: we must either modify our views about natural kinds, or recognize that they are simply not well suited for taxonomic units.

That discussion is a rich one, but not one I really want to dig into here. Instead, let me finish by posing a more speculative question, though not one that I intend to try to answer in this post. Rather, it’s an attempt to go broad, and to spur a discussion about how those big philosophical questions posed at the beginning are fruitfully informed by the more precise and narrower metaphysical hypotheses that are part of our best scientific ones.

What we see in the cases like disputes over the origin and evolution of modern birds is that they require a clear conceptual account of identity. But identity is not a theoretically neutral notion here; it’s explicitly grounded in the context of phylogenetic systematics. This comes rather naturally to anyone working in systematics. That was one of the big lessons of 20th century taxonomy. These sorts of disputes are never merely about discovering theoretically neutral, objective facts, but of what theoretical context we ought to be using to interpret the evidence we discover. This was most obvious in debates over how to think about similarity and simplicity, but it’s also the case for things like identity and persistence.[4]

What if the same is true for claims of identity and persistence more generally? Philosophers often present debates over these without an acknowledgment of the theoretical context in which they are occurring.[5]  What if they simply can’t sensibly occur absent any theoretical context? That is, should a lesson from paleontology be the rejection of absolute notions of identity and persistence? What, then, would our discussions on these concepts look like? Alternatively, is it the context-free account of these debates that distinguish philosophy? Do our context-laden accounts of identity depend on some absolute, context-free account? These are the sorts of questions that come to my mind when I read papers like Padian and Horner’s.

Concluding Remarks

Philosophy is hard. Teaching philosophy is very rewarding, but it can also be very difficult to get students unfamiliar with philosophy to grasp what is at stake. Paleontology can help with that, and Padia and Horner’s paper is a great example of this. Growing up, I was taught that birds and dinosaurs were different taxonomic groups of the same rank. Our students most likely grew up learning something very different, namely, that birds are dinosaurs. That difference in what we grew up learning is an easy fact for students to grasp. Using papers like Padian and Horner’s provides an accessible opportunity to discuss with our students what happened. Though there was certainly empirical evidence that influenced this change, at its core the difference is really a philosophical one. It comes down to what concept of identity biologists adopt, and the consequences this has for thinking about change and identity over time.

I’m lucky. When I teach this paper I can hold class at the University of Utah’s Natural History Museum of Utah. That means I can talk about transformation and typology, or gradient versus categorical thinking, in front of a cladogram constructed from casts of ceratop skulls. Students quickly notice lots of things about this display, and, with very little prompting, quickly start discussing the background assumptions that must have informed the choices made in constructing the hypothesis on display. They want to know how decisions were made over which features of similarity to prioritize, how you mark distinct groups when things change gradually over time, and how we determine when two things are distinct from each other or belong to the same group. These are precisely the sorts of questions philosophers have been asking for a long time, but paleontology provides a great way for students to come to these questions on their own, and have some ownership in carefully thinking through how we might go about answering them.

[For an expansion of some of the arguments here, please see my (2012) “How to Misidentify a Type Specimen,” or my manuscript, “The Individuality Thesis (3 Ways)” (email me for a copy).]

Works Cited

Haber, M. H. (2014). In defense of the organism. Review of: Thomas Pradeu (Elizabeth Vitanza, trans.): The Limits of the Self: Immunology and Biological Identity. New York: Oxford University Press, 2012. ix+302 pp. ISBN: 978-0-19-977528-6, $65 hb. Biology & Philosophy, 29(6):885–895.

Padian, K. and Horner, J. R. (2002). Typology versus transformation in the origin of birds. Trends in Ecology & Evolution, 17(3):120–124.

Pradeu, T. (2012). The limits of the Self: Immunology and Biological Identity. Oxford University Press, New York.

[1] Philosophers of biology, especially those working on issues in systematics, tend to be much more sophisticated about this.

[2] Of course, that’s a pretty big condition! We know that transmission isn’t merely vertical (i.e., through parental generations within a lineage), but, for lots of reasons, can also be horizontal (i.e., across lineages). This can complicate matters, and presents some real empirical and conceptual challenges for 21st  century systematists.

[3] Of course, there might be internal change that generates new identities, e.g., as lineages diverge new species may form. Yet all of those are still part of the same group of common ancestors. This is how larger, more inclusive taxonomic groups form within established ones.

[4] Some of this will sound familiar to philosophers trained in the analytic tradition. Nelson Goodman, as an obvious example, provided compelling arguments in the philosophical literature against absolute notions of similarity and simplicity.

[5] For example, a context provided by set theory or mereology is often taken for granted.


Matt Haber is an associate professor in the Department of Philosophy at the University of Utah (and he's also currently residing as chair).  He also holds an adjunct appointment in the Center for Quantitative Biology.  He is a philosopher of biology, specializing in phylogenetic systematics.  Matt got his PhD in Philosophy at the University of California, Davis, where he was also an affiliate member of the Center Population Biology.  Prior to that he received an MSc in History and Philosophy of Science from the London School of Economics, and double majored in Biology and Philosophy at Grinnell College.  He became a philosopher of biology by refusing to ever choose between those two fields.  Matt teaches a class at the Natural History Museum of Utah called, "How to Make a Mammoth and Engineer a Dinosaur," where students grow somewhat alarmed at how excited he gets each time they look at fossils in class.