Consensus Schmosensus: Dead Dinosaurs, Big Rocks & Simple Stories

Adrian Currie writes...

The scene opens on a typical late-Cretaceous day: a herd of ceratopsids munch on cycads; a few large theropods stalk through a nearby copse; a flock of Hadrosaurs make their way to the river, their occasional long, honking calls breaking the otherwise insect-purr-backed stillness. Then, a flash: a brilliant, white light appears in the distance, reflected in quickly upturned, all-of-a-sudden panicked eyes. A few precious seconds pass before an enormous shock-wave hits, the dinosauroid forms melting into azure haze. A few short months later a near-skeletal T. rex stalks the remains of a blighted, blackened and overcast landscape. Unable to find food, she crumples to the ground, breathing her last; becomes food for the small stealthy mammals who inherit the Earth in her stead. Thus the end of the Dinosaurs: the bang of an extra-terrestrial impact followed by the whimper of the long proceeding winter. Geologically speaking, sudden. Below the K-Pg boundary, dinosaurs; above: none (except birds of course but adding ‘non-avian’ every time I mention Dinosaurs is exhausting alright?).

*insert kick-ass and definitely not sensationalized picture of dinosaurs watching a meteor strike*

If you know anything about Dinosaurs, you know that a Big Rock killed them. After all, we’ve a lot of evidence concerning a pretty impressive impact around 65.5 million years ago. No one seriously denies that the Big Rock hit. And that impact does seem to coincide with the disappearance of that most charismatic of fauna. Since being introduced in the early 1980s, the impact hypothesis has generated a rather bewildering amount of work. There is also the impression of a consensus forming. That, scientifically speaking, the Big Rock hypothesis is a done deal. Is there such a consensus? And should we want one? I haven’t really looked in detail at the K-Pg extinction before: for one thing, the literature is enormous and dauntingly complex; for another, like, there’s so much interesting work in paleontology beyond all this dino-extinction business that the case just looked, well, too obvious? Anyway, in turns out there’s some super interesting stuff going on, and I’m going to dip in a toe after all.

A bit of philosophical background: I’m interested in the kinds of explanations paleontologists prefer, and how those explanations progress over time. We often start with (let’s call them) one-shot hypotheses. Such explanations identify a particular trigger for an event and then attempt to explain as much of the phenomena at hand as possible in light of that trigger. There have been plenty of one-shots concerning the dinosaur extinction: changes in climate driven by shifting continents, egg predation from sneaky mammals, mass epidemics, massively increased volcanism and—of course—Big Rocks to name just a few. I’ve previously argued that these simple explanations are often integrated into much more complex, multi-factored explanations as time goes on. And, I reckon, this is a good thing too: history is a complex, messy beast and at least sometimes that calls for messy explanations. But here—apparently—we have a consensus forming around a one-shot explanation. It might just be that the ol’ one-shot is what we want here. Aft all, not everything in history has to be hideously complicated. However, I think on closer inspection the consensus, and the emerging explanations, are not what they seem.

(A quick caveat before diving in: the explanation might perhaps be simple, but the evidence certainly isn’t. There’s a complex interweaving of geological, paleoclimatological, stratigraphic, fossil and paleoecological evidence implicating the impact, extinction, and so forth, and I’m by-and-large going to ignore the ins and out of this. For one thing, it’d make the post *really long*.)

Let’s start in 2010. Up until that point, much of the debate around the role of the impact focused on whether it was a mere coup de grâce for already declining  biodiversity, or whether the dinosaurs and other fauna thrived up until the extinction. If the latter, then there’s more reason to accept a one-shot explanation for the extinction. There was also still a competitor: as recorded in India’s impressive Deccan traps, the late cretaceous was a period of prodigious volcanic activity. ‘Traps’ are step-like geological features formed by continuous eruptions. It is thought that at their height the Deccan traps covered one and a half thousand kilometres squared: a sure sign of incredibly powerful, extended volcanism.


India's Deccan traps: even if they cause mass extinctions, volcanism sure makes for some pretty landscapes...

India's Deccan traps: even if they cause mass extinctions, volcanism sure makes for some pretty landscapes...

Anyway, in 2010 Science published a 41 author review paper on the extinction. It could have been titled something like ‘Okay, a big rock killed all the dinosaurs, okay? Stop going on about it’. They summarize their evidence: “The temporal match between the ejecta layer and the onset of the extinctions and the agreement of ecological patterns in the fossil record with modelled environmental perturbations (for example, darkness and cooling) lead us to conclude that the Chicxulub impact triggered the mass extinction” (1214). They also highlighted the downstream effects of the impact such as “extended darkness, global cooling, and acid rain” (1214). It’s the classic Big Rock scenario: a mistimed impact makes all the difference between a thriving dinosauroid world and a mass blast followed by a withering winter and poisoned seas, ending that world forever.

I don’t know the politics here (and I bet there’s a heap), but a 41 author paper in Science is surely a play for consensus. And consensus, I take it, means something like: case closed, truth found. Don’t bother thinking up more hypotheses vis-à-vis dinosaur extinction, busy yourselves working out the details of how the Big Rock did its killing business. However, along with the review paper were a set of sceptical responses: consensus resisted. The back-and-forth was mostly wrangling about whether the evidence quite told against competitors (volcanism again), whether awkward geological details had been dealt with, and so forth. For instance, the impact ejecta is not really just contained in one geological moment, but spread across strata. So, to get the timing of the impact right, you need to explain away the spread by saying the ejecta was shifted post-impact by various geological forces and perhaps postulating other, small impacts (we’ll return to this in a second). Notably, a group of vertebrate and freshwater paleontologists complained of being left out of the picture (the evidence in the original paper largely sticking to climate, geology and patterns in the invertebrate record). They push back: “… the simplistic extinction scenario presented in the Review has not stood up to the countless studies of how vertebrates and other terrestrial and marine organisms fared at the end of the Cretaceous” (973).

So, is there a consensus here at that stage? Hard to tell. I’m inclined to think that if the geologists and oceanic paleobiologists can’t get the Dinosaur folk on board, it would be a bit rich to claim a consensus on the dinosaur extinction…

In 2015 we hear from dinosaur experts (a different group, and this time in Biological Reviews) again, and things have changed. The paper highlights advances in both an understanding of trends in Dinosaur diversity, more precise dating techniques and in the analytic techniques required to control for biases in those trends. These, they think, build towards “… an emerging consensus on when and why non-avian dinosaurs died out” (629). This consensus looks like a restatement of the one-shot thesis: “…without the impact non-avian dinosaurs probably would not have completely died out” (638). However I think there are crucial differences between the 2010 and the 2015 paper. In 2015 they emphasize that the late cretaceous is, generally speaking, a time of major disruptions; the extinction occurs “amidst a backdrop of massive volcanic eruptions, major changes in temperature and sea level, and the impact of a ~10-km-wide bolide” (629). Although there is lots of evidence that dinosaurs were not in biodiversity decline up to the extinction, the authors suggest that land-based communities could have been destabilized by buffeting from the pretty wild fluctuations in global temperature (from roughly 16 Celsius to 11 Celsius above the present) and sea level (everything from 50-70 meters above current levels to equal with current levels!) over the last 15 million years of the Cretaceous, and—of course—don’t forget those Deccan volcanoes. Such events could change dinosaur communities, increasing their susceptibility to extinction. For instance, there is a decline in major herbivore diversity in the very important Hell Creek formation across ceratopsid, ornithischian and hadrosaurids (there are very few sauropods in Hell Creek), and modelling results suggest that such a change could make communities more vulnerable to the kinds of impacts our Big Rock would have caused. Further, although there isn’t a decline in biodiversity per se there is reason to think populations have become further localized: a feature thought to increase extinction susceptibility.  

Here’s the difference I see: in the 2010 paper the Big Rock hypothesis is presented as a clashing competitor—mutually exclusive—with other hypotheses; in the 2015 paper the Big Rock hypothesis isn’t really put in competition with the others. It is agreed that the Big Rock was causally necessary insofar that if it didn’t hit, then the extinction wouldn’t have occurred. But we’re now asking whether it was sufficient, in particular, whether other events at the time may have weakened or eroded the resilience of late Cretaceous populations such that the Big Rock could do its work. This suggests, it seems to me, that the pattern I’ve noticed in historical science: from simple, one-shot explanations to increasingly complex multifactorial explanations occurs even when there is relative consensus on which ‘simple’ explanation gets things roughly right. That is, the Big Rock hypothesis hasn’t itself been stable, but has been incorporated into richer explanations which include other features of the late Cretaceous. I want to discuss two other developments since 2010.

Deccan volcanism hasn’t gone away. Most dramatically, Gerta Keller argues for a view which approaches the polar opposite of the 2010 paper: “[a] series of at least four massive volcanic eruptions in short order formed the longest lava flows on Earth and probably caused runaway effects, particularly ocean acidification, which resulted in the carbonate crisis commonly considered to be the prime cause for the mass extinction. The Chicxulub impact played no role in this mass extinction” (29). A crucial aspect of Keller’s argument questions the Big Rock’s timing. On her view, the consensus regarding the Chicxulub impact coincinding with the K-Pg boundary is explained by the K-Pg boundary itself being redefined in terms of the impact's traces: “the Chicxulub impact is assumed to have caused the mass extinction, and this belief, rather than statigraphic data, has led to a redefinition of the K-T boundary based on Chicxulub impact ejecta” (26, her italics). According to Keller, (1) the impact occurred tens to hundreds of thousands of years before the extinction, (2) the second period of Deccan volcanism basically coincided with the mass extinction, (3) the third period of Deccan volcanism can explain the slow recovery from the extinction.

Again, I don’t want to get bogged down by the ins and outs of the evidence here (at the very least its waaaay above my paygrade) but it’s worth pointing out that there are still arguments to be had about the timing of the impact—let alone its causal role regarding the extinction. So is the Big Rock to blame? If you put a gun to my head, I’d probably bet against Keller: but as I’ll argue below, I’m not sure why this is the kind of bet we should be making. Why should we have to pick a winner?

Let’s switch to another interesting question about the extinction: how come the reptiles didn’t undergo a renaissance? That is, why did mammals and birds take over? This might be a pressing question considering that on first blush us Cenozoans are at a distinct disadvantage. Our warm-blooded lifestyles are extremely expensive, so how come those cold blooded critters didn’t get the jump on us? Arturo Casadevall offers a fascinating bit of speculation. One striking thing about us mammals and them birds—those that appear to have won-out post K-Pg—is our endothermy (warm-bloodedness). And endothermy makes us a ricky target for a prominent set of pathogens: fungus. Our higher body temperatures makes an uncomfortable environment for fungi, so perhaps “mammalian resistance to fungi through the combination of vertebrate-level immunity and endothermy could have been the result of selection by pathogenic fungi”. Especially as the geological record suggests that enormous fungal blooms accompanied the K-Pg calamities, we can imagine a fungi filter acting on he survivors of the K-Pg extinction. Only those with warm blood (and thus partly defended) had a chance to thrive. With the (often endothermic) dinosaurs knocked out of the picture, and the ectotherms repressed by mushrooms and spores, it was a mammalian and avian moment.

Note that Casadevall’s theory doesn’t really turn on whether volcanoes or impacts caused the extinction; rather, he’s adding another idea to the mix, and one answering a question which arises post-extinction. So again there isn’t necessarily competition between hypotheses here. This also takes the form of a highly speculative, one-shot hypothesis. I wonder if this new question and hypothesis will follow the same pattern?

Alright, so there’s still a lot—a lot—going on concerning the dinosaur extinction (and oh boy, this has just been a little taster) what should we make of it all? First, I think there’s reason to think that my ‘simple-to-complex’ model for thinking about progress in historical science looks safe here. Even in when a consensus forms that some event is necessary, it is quickly recognised that more detail is required for sufficiency. We’ve seen explanations that involve single-shot hypotheses, explanations involving multiple triggers, and explanations with a single trigger, but a complex multi-factorial account of that trigger’s effects, and the preconditions required for them. These latter two explanations fit with my sense of how historical explanation proceeds, and the generation of new one-shots targeting slightly different phenomena like Casadevall’s hypothesis looks like good news to me.

Second, I think this case teaches us something about the value of simplicity. I suspect that the complaint against the 2010 paper shouldn’t be that it is presenting such a simple explanation per se, but rather that it treats the other hypotheses as opposing. Perhaps we shouldn’t read simple, one-shot explanations as really being intended as such at all, maybe instead we should read them as possible causal factors in the event’s occurrence. There is a kind of obsession with trying to find out what the cause of the K-Pg extinction was, and this leads us to focus on whether or not we’ve got to the truth of the matter—whether consensus is forming.

Well, consensus schmosensus: I think what’s really interesting is how these one-shot hypotheses provide contexts for examining both possible proxies of causal factors, as well as understanding their dynamics, and how they might interrelate with other factors. That is to say, instead of bemoaning the simplicity of one-shot explanations, perhaps we should see them as a strategy which allows paleontologists to isolate causal features. This isolation allows us to build up a profile of how that causal factor might operate, and this allows us to then deploy that factor in more complex explanations.

Casadevall’s explanation is no doubt a one-shot, but it provides a fascinating way into both the conditions post the K-Pg extinction, and into thinking more generally about the relationship between macroevolutionary patterns and immunology. Keller and her co-author’s certainly have an uphill battle, and hey who knows (or even—dare I say—cares) whether it turns out Big Volcanoes or Big Rocks were a more important causal factor in the extinction; she provides a rich program for understanding the proxies associating volcanic activity with paleoenvironments and both geological and paleobiological features.

The basic thought is that simple one-shot narratives provide the scaffolding required to really investigate those causal factors, and thus the eventual raw materials for tackling the complexity of the real world. Furthermore, under such circumstances even if the main hypothesis turns out to be false, the knowledge gains look to me huge. Imagine that Keller is right, imagine that the Big Rock has simply nothing to do with the extinction. That doesn’t invalidate the enormous amount of research we’ve carried out in understanding how extra-terrestrial impacts work, affect the Earth, and leave traces.

Finally, another aspect of the debate concerns the role or otherwise of unity. Folks who are interested in volcanism often highlight the role of volcanism in other major extinctions—most especially the enormous events which formed the Siberian traps implicated in the end-Permian super-mega-mass-extinction. Presumably the implication is that we can unify mass extinctions as being an occasional volcano-fuelled phenomena. By contrast, folks interested in Big Rocks will emphasize the idiosyncrasy of mass extinctions. For instance, from the 2010 review article: “Each mass extinction event should be considered relative to the record for that event, and we stress the unique aspects of the K-Pg boundary record” (975).  I wonder, what virtue is there in having a common kind of event causing mass extinctions? Should it be something we appeal to? I genuinely don’t know. That’s fine, I’m pretty okay with not knowing things, whether it be the value of unity to an explanation, or the cause of the K-Pg extinction. Keeping to theme: Sometimes knowing stuff, like having a consensus, is overrated.