Feathered Dinosaurs

6 comments
An artist's rendering of Yutyrannus huali, a feathered dinosaur recently discovered in China.

I wanted to alert everyone to an article that appears in the journal Nature today, which has been causing quite a stir.  (It was even written up in the NY Times!)  The article announces the discovery of a new feathered dinosaur from the Lower Cretaceous in Liaoning Province, China.  Above is an artist's rendering that gives you a sense of how scientists imagine these creatures looked in the flesh.

There are a few things worth noting here.  First, this creature is a fairly close (but older) relative of T. rex.  Second, as the article points out, it is by far the largest feathered dinosaur that has been found so far.  (The next largest was only about 1/40th its size.) 

Since the discovery of Archaeopteryx in the Victorian period, paleontologists have posited a link between extinct dinosaurs and modern birds.  (Indeed, extinct dinosaurs are now usually referred to as non-avian dinosaurs.)  But in the past several decades, scientists have been pushing the evolution of feathers further and further back, both temporally and phylogenetically.  Archeopteryx was a kind of transitional form, whereas Yutyrannus huali is nested relatively deep within the Tyrannosaurid family tree.

But there's another, perhaps even more interesting reason why the relationship between this creature and the famed Tyrannosaurus rex is so important.  Previous discoveries of feathered dinosaurs have resembled a chicken more so than the creatures we usually associate with the name "dinosaur."  This most recent finding changes that completely.  Now we have something that looks very much like a canonical (non-avian) dinosaur, yet it appears to have been covered with a downy plumage!  Part of the reason this animal has made such a stir, I think, is therefore the fact that it gives license to artist's renderings of the type pictured above.

Anchiornis huxleyi, a feathered dinosaur from Liaoning Province, China.

By way of contrast, consider the example of A. huxleyi (pictured above), which was also discovered in Liaoning Province, China.  In an article published in Science, researchers from China and the United States were able to infer the specimen's color pattern from microfossilization of melanosomes.  You've probably encountered the picture before, because it also made a big splash in the popular press.  Here, the animal cannot boast of a particularly impressive size, but scientists were able to endow it with a visually striking color pattern.

All of this obviously looks very different from how we used to think about dinosaurs.  The most famous and prolific visual interpreter of prehistoric animals during the heyday of dinosaur research around the turn of the 20th century was undoubtedly Charles R. Knight.  Unlike these modern, active, colorful creatures covered in feathers, Knight painted dim-witted, slow-moving, scaly, and drab reptiles.  


T. rex battling a Triceratops, by Charles R. Knight. 
   
So, what's the relationship between material evidence and imagination in producing these illustrations?  Why have our visual renderings of dinosaurs changed so much over time?  I think the answer is neither just cultural -- artists are simply making it up as they go along -- nor is it just empirical -- artists are simply following the available evidence.  Rather, the two interact with one another in a very deep way.

In the comments section of the previous post, Hank, ST(res)S-ed out, and I have been arguing about Ian Hacking's ideas about "dynamic nominalism" -- the extent to which our interpretation of the world changes its material constitution.  I don't want to suggest that dinosaurs are historically constituted in precisely the same way that Hacking thinks people are -- that is, I don't think that what we think about dinosaurs actually changes the material fossils buried underground -- but I do think that *something* comparable to Hacking's dynamic nominalism is going on here.

You might ask yourself (as I often do): why were feathered dinosaurs (of the non-avian variety) not discovered until so recently?  Prior to the discovery announced in Nature today, you might have said: perhaps because feathered dinosaurs are relatively modest in terms of their size and appearance.  (Modest until you have the tools to reconstruct their plumage pattern, that is!)  But as I've already noted, part of what makes this discovery significant is that Y. huali is a close relative of T. rex.  The next obvious question, to my mind, is this: if big, impressive, therapod dinosaurs like Y. huali had feathers, why didn't anyone notice until now?  Is it because paleontologists have only begin to research the evolution of dinosaurs in China, which is where feathered dinosaurs tend to be found, relatively recently?  Or is it because paleontologists simply weren't looking for feathered dinosaurs during the early 20th century, when the Western United States was understood to harbor the world's richest dinosaur quarries?  

Figure 2 from the Nature article, images c-h showing "preserved integumentary structures," i.e., fossil feathers, in Y. huali.

If you read the Nature article carefully,  and especially if you examine the Y. huali fossil (pictured above) closely, you'll see that it's not at all obvious, at least not at first glance, that this creature had feathers.  So I could easily imagine someone preparing this specimen, trying to free the bones from the rock matrix, inadvertently destroying the fossilized traces of feathers.  Perhaps the only reason Y. huali was recognized to have had feathers is because by now paleontologists are actively on the lookout for them.  If that's right, then what's in your mind when you prepare a fossil for study and display may well have a significant impact on the material constitution of dinosaur bones.

All of this is to say that I am now anxiously awaiting the day when paleontologists working in the American west announce the discovery of a downy T. rex!

6 comments

"So I could easily imagine someone preparing this specimen, trying to free the bones from the rock matrix, inadvertently destroying the fossilized traces of feathers."

I think that's right. I've refrained from entering into the Hacking hack-off, but I'm very sympathetic to Lukas' take, especially as he illustrates it here.

In my own work on the preservation of blood and other biodynamic substances, mined not from buried fossils but living populations, the intersections between what scientists know at the time and what they imagine possible uses for these materials in the future shapes how those bodily extracts get archived/inscribed.

If for example, contemporary scientists assume that what is most valuable about human blood is human DNA, they may decide to extract and preserve only that DNA and discard the liquid matrix that supports it.

However, in recent years frozen human blood has been revalued for the microbial life it harbors -- genomes and other things that would have otherwise been discarded (inadvertently destroying the evidence . . .)

In this case, we have an opportunity to link the material to the social, in that the role of science in what comes to "make people up" -- human DNA plus lots of microbial DNA plus who knows what else -- may also be the same thing that undoes them as ontologically stable entities.

Joanna provides a nice example of how empirical evidence, is, at least in part, materially constructed in biomedicine as well. It is not inert & passive, simply awaiting analysis.

Since we're getting deeper into the question, though, I would like to prompt that we think through the distinction that Hank raised in our discussion from the previous post in more detail: what, exactly, *is* the difference between the ontology of something like a fossil versus something like a hysteric.

Hank reads Hacking as saying, roughly, the following: hysterics are what we might describe as social kinds. Fossils, on the other hand, are natural kinds. One way to draw the distinction is to point out that the ontology of social kinds is historically constituted, which is not true of natural kinds. Prior to the invention of hysteria as a disease category, it was simply not possible to *be* a hysteric. (For the same reason, it might no longer be possible to be such a person today.)

The same is not true of fossils. I like to tell a story about the simultaneous discovery of dinosaur fossils in the American west during the 1877 field season. Dinosaur fossils have existed in the American west for millions of years. And people have known about them for thousands of years -- it is almost certain that Native Americans took a keen interest in these objects, which have a place in some of their traditional stories and myths. They did not recognize them as being what we now understand them to be -- the material remains of prehistoric life. But they nonetheless had access to the same fossils, meaning the same material objects. Our changing ideas about fossils did not change their ontology -- they were always the same.

Now Joanna and I say, ah, but in fact our changing ideas about fossils (and blood) *do* have a material impact on these objects. What you think about dinosaurs influences how you prepare one for study and display, for example. Or what you think about blood influences how you store it.

But does this amount to a genuine *ontological* intervention? You might say -- of course not: the fact that someone fails to recognize funny markings on a dinosaur fossil as traces of feathers and therefore removes them does not change the fact that those traces were there to begin with. Preparation and storage techniques might have a material impact on the construction of evidence, but they do not change what it is to be a fossil or blood.

Following the dialectic thus leaves us in a strange place. On the one hand, it is clear that our descriptions, classifications, etc., clearly have a material impact on lots of objects, some human, some not so human. However, there is a strong intuition that some such material interventions change the object's ontology whereas others leave it in tact. What lies at the root of this intuition? Is is just a desire to separate humans from the rest of the world, or is there some deeper and more meaningful justification? If it's the latter, what, exactly, does it amount to?

Oh no, Joanna! You tried to steer clear of the Hack-fest, and now look where you are. But I like the direction we've gone, so let's leave Hacking (and my reading of him, and Lukas's creative misreading of my reading) behind, and turn to our dissertation objects. Call them bones, brains, and blood.

So, let's take Lukas's question – "what, exactly, *is* the difference between the ontology of something like a fossil versus something like a hysteric?" – and ask it about those things. As I said on the other thread, and as you both agree, *of course* theorizing about all three changes them – in both mundane way (our ideas about them, how we perceive them, &c.) as well as the more interesting material sense you've both just laid out.

But what does that have to do with ontology? I guess that's where I'm hung up. Without going back to Hacking, can I just ask, then: what do we mean by ontology? Or ontologies? Do we mean theories about things that exist, or those things themselves, or what? Phrases like "the object's ontology" leave me confused – the object doesn't subscribe to a theory, like a human might – but maybe we mean the theory we have about it? Or we mean its constitution in the world?

Yes, I think that's exactly the right question to ask, Hank: "what do we mean by ontology."

Traditionally, philosophers (including Hacking, as I read him) understand ontology to be a branch of metaphysics. Ontology is the study of what there is. There is a quote by Quine, from a 1948 article in the "Review of Metaphysics," of which I'm very fond:

“A curious thing about the ontological problem is its simplicity. It can be put in three Anglo-Saxon monosyllables: ‘What is there?’ It can be answered, moreover, in a word—‘Everything’—and everyone will accept this answer as true. However, this is merely to say that there is what there is. There remains room for disagreement over cases; and so the issue has stayed alive down the centuries.”

Traditional ontological questions include "are there such things as universals?" and "do numbers exist?" In trying to answer these questions, philosophers attempt to get at the fact of the matter. Not -- is it useful to assume numbers exist? But -- do numbers *really* exist? It is a kind of presupposition in the ontological project that there are genuine facts of the matter when it comes to such questions.

On this view, the questions most of us are asking are not really ontological. There is a fact of matter about whether dinosaurs had feathers or not. Then there is a separate question about how we might come to *know* this fact of the matter (or fail to recognize it, as the case may be). The latter is an epistemic question, not an ontological one.

My understanding of how the term "ontology" has been thrown around by Historians of Science and STS scholars recently is quite different. Rather than taking the ontological question to be a metaphysical one, it is taken to be an interpretive tool. So, rather than ask "do numbers really exist?" we can ask "did this community of mathematicians believe numbers really exist?" or "does this formalism seem to bias the thinking of certain practitioners about whether numbers really exist?" That is -- rather than go after a fact of the matter, a statement about the world, many of us want to know what our historical actors took to be a fact of the matter, what statements about the world they subscribed to. This is why we can use the plural form of the word -- ontologies -- which would make no sense to the philosopher. There is just one fact of the matter about how the world is, but different people have different views on the details!

If it sounds like this is conflating ontology and epistemology, it is! But that's an old conflation, going back at least as far as SSK -- if not further (I'm thinking of e.g. Peter Winch's Wittgenstinian ethnography of Zande witchcraft practices).

One way to think about the question I've been trying to ask through Hacking is as an attempt to re-inscribe the distinction between ontology and epistemology, but to draw the boundaries in a new way. I say: course there is a fact of the matter about whether dinosaurs had feathers or not. But what we *think* about that question has a lot to do with how the world is. Not just how we see it, but literally what kinds of material objects it contains. So, although our epistemic commitments do not determine what dinosaurs were like back in the mesozoic, they do affect the material constitution of the world we live in today. And this material constitution determines our understanding of what dinosaurs were like -- whether they had colorful feathers or a drab, scaly skin.

There's more to say about this, but I'll just add one more thing:

I also think that material objects such as dinosaur bones really are socially constituted in the traditional ontological sense -- that what we think about dinosaurs does indeed change what they are. That's a more complex argument, so I won't spell it out now. But let me just hint at the kind of thing I have in mind:

There are lots of objects that only exist because we think they exist without therefore being any less real. Money is one of them. The President of the United States is another. The US / Mexico border is a third. These are what Durkheim called "social facts." If humans ceased to exist, there would no longer be a border between the US and Mexico. But that does not mean the border is any less real. It just means it's a social kind.

Dinosaur bones are natural kinds. But there is also a sense in which they are social kinds. For example, dinosaur bones are scientific specimens. Objects are scientific specimens by virtue of the fact that they are accorded a certain status by members of the scientific community. You might say they are "consecrated" as a specimen when they enter a museum collection, are published in a scientific journal, are assigned a latin binomial, etc. So if all humans ceased to exist, dinosaur bones would still be out there buried underground, but they could no longer be scientific specimens.

Does this matter? Well, it matters a little bit because whether something is a scientific specimen or, say, a religious icon has a large effect on how we interact with it. It might even have a large effect on it's material constitution -- and in the case of dinosaur bones it certainly has. Scientific specimens are prepared, stored, inventoried, and talked about in very specific ways. And, as Joanna pointed out, how you prepare, store, inventory, and talk about an object goes a long way to determining what you know about it -- and what you know about our world.

As a recent observer and groupie of this blog, it interested me that Joanna kicked off this comment stream’s line of debate.

I think that for anyone with a marginally constructivist view it is hard to take older visions of ontology and, yes, metaphysics seriously—that is, that we are just sorting out the answer to the question “what is there?” without also simultaneously making the very categories on which we are relying.

I find it both exciting and fascinating that many scholars in our peer group agree that there is something (hacked) off about Hacking’s division between people and things. And, yet, we are not driven away from some form of “realism,” for want of a better word. One way of thinking about Hacking’s division is that he wants to be deeply anti-realist and constructivist about humans and the study of them, while desiring to preserve a more stolid realism for thing. I think we are saying that we can take his “dynamic nominalism” deeper with respect to our objects of study—the way he does with humans—without giving up the idea that there really is something there to study, or even that these things are different than humans.

On this score, I think that Lukas’s and Joanna’s discussion on evidence around bones and blood is to cool for words. My guess is that our agreement partly rests on an idea that is inherent in Lukas’s statement that empirical evidence “is not inert & passive, simply awaiting analysis.” We think that people and objects mutually construct many things, including our representations; people and things are both active; things often “push back” against our pictures of them.

We should remember that, in the 20th century, ontology and metaphysics were relegated to the dustbin before philosophers pulled them back out. Some philosophers, like Kripke, found metaphysics helpful for clarifying issues in analytical philosophy. Others worked under the assumption that every intellectual position has an inherent ontology, or set of assumptions about how the world is. This ontology may or may not be explicit. They believed that it was better to be clear about ontology rather than deny its existence. I think it’s this latter idea of ontology that is worthwhile for science studies.

Because it is in this sense that studying someone’s epistemology amounts to studying (an) ontology. As we study their words and actions, we are also interested in their (often unspoken) assumptions about the structure of the world. And I don’t think this idea entails that we have to give up on the belief/instinct that the structure of the world really, really is some way.

Yet, I also agree with Lukas that the word “ontology” has been harassed and abused in all kinds of ways by folks in science studies. For instance, is Winner’s assertion that artifacts have politics about ontology? Do objects have politics in the same way they have atomic structures? My guess is probably not. Winner’s racist bridges are no longer racist if humans die out and the city is inhabited by an Martians who have no notion of race. But the bridges will still have atoms. Similarly, is the notion of “ontological surgery” really about ontology? Or is it about epistemological judgment? Probably the latter.

So is there a way to be more careful with out words? Would this involve developing a new way of speaking? Or just challenging sloppy usage of our ordinary tongue?

Note: Only a member of this blog may post a comment.

back to top