I’m pretty promiscuous when it comes to evidence. Especially when it comes to reconstructing the past. Not to put too fine a point on it, but I’m basically open to all-comers.
This promiscuity has led me to argue that philosophers (and some scientists) think about historical evidence in impoverished ways. They often begin and end with ‘traces’: the downstream consequences of past events. Traces include a dinosaur and its fossil; an erratic rock and its glacier; cosmic background radiation and the big bang; a past thing and its present remains. Historical inferences are supported by linking the way the world is now with the way the world was then. Some philosophers who think things like this are Carol Cleland, Derek Turner and Aviezer Tucker.
Traces are definitely powerful. For instance, on the discovery of a funny-shaped rock…
… we might posit the existence of a sauropod dinosaur.
We can do this because we understand the processes of fossilization in virtue of which animal remains can, under very specific circumstances and over long periods of time, be converted into rock. And because we understand vertebrate anatomy: the owner of such a bone would have a set of other bones common to animals of that type. We might even name the critter, in this case (rather wonderfully) Dreadnoughtus schrani*. And so, by linking the present to the past—via ‘traces’—we can find out a lot about the former. This is where most accounts of historical reconstruction stop.
However, such ‘tracecentrism’ doesn’t do justice to the range of support we can draw on in reconstructing the past. For one thing, ‘coherence’ matters: what I posit about D. schrani must fit with our knowledge of the Mesozoic’s ecological, phylogenetic, environmental and climatological regime. The richer our understanding of that context becomes, the more empirically constrained our hypotheses about D. schrani. For another thing, scientists utilize a range of proxies in reconstructing the past. These include the use of analogies (see here for instance) and the construction and deployment of simulations and other models (which, I’ve argued, provide something very much like empirical evidence about the past)— for instance Sellers et al use simulations to understand titanosaur gait. As I said, I’m promiscuous about historical evidence.
This promiscuity leads pretty directly to an objection: well, Adrian, is there anything you don’t like? Recently, Derek Turner wrote a lovely piece on Extinct which encouraged us to re-chew an old bone: the distinction between true and pseudo-science. He did this by putting pressure on the difference between cryptobiology—the supposedly ‘pseudoscientific’ hunt for critters like the elusive Pacific Northwest Tree Octopus (Octopus paxarbolis)—and the discovery of living fossils—extant organisms otherwise only known from age-old fossils such as the Wollemi pine or the Horseshoe Crab. The climax of Derek’s discussion is a thought experiment which I’ll quote in full:
“The Living Fossil Hunters. Suppose that in the 1980s and 1990s, a group of cryptobotanists coalesces around the shared hope that the wollemi pine (which at that time was known only from fossils) might not be extinct. This group sponsors expeditions to remote areas to try to find the wollemi pine. And they provide a discussion forum for enthusiasts to share tips about where to look next, and how to identify the tree when they found it. Of course, the scientific community dismisses these cryptobotanists as engaged in mere wishful thinking, a pseudoscientific exercise that’s not much different from hunting for living pterodactyls.”
Derek asks whether—given that the cryptobotanists would be vindicated—we can defend the scientists’ dismissal. This challenge is particularly pressing for me: if I’m willing to emphasize the messiness of historical reconstruction, be so expansionist about our epistemic resources, and glorify speculation, how can I tell the difference between the proper science and the crazy? Well, (like Derek) I can’t give you an easy answer: there’s no magic bullet to solve the demarcation problem. However, I can at least give you some thoughts: my Crazy-Spotter’s Guide, if you will. To get there, let’s think about conspiracy theories**.
Now, I love a good conspiracy theory. Aliens built the pyramids? Of course: haven’t you read the epic of Gilgamesh? It’s obvious! The Chinese discovered Australia? Naturally. Bring it on. An all-time favourite is the claim that Francis Bacon—the Royal Society’s beloved Lord Verulam and (depending on who you talk to) father of modern science—wrote the plays normally attributed to Shakespeare. Bacon’s not the only supposed culprit, and has fallen out of favour since the 1950s, but I love the thought nonetheless.
Why should we think that Bacon, rather than ‘Shakespeare’ the historical dude, wrote the plays? The reason is summed up in the following quote:
“... There is nothing in the writings of Shakespeare that does not argue the long and early training of the schoolman, the traveller, and the associate of the great and learned. Yet there is nothing in the known life of Shakespeare that shows he had any one of the qualities.”
The quote is attributed to James Corton Cowell of the Ipswich Philosophic Society in 1805. However, according to James Shapiro’s delightful (and highly recommended) Contested Will: who wrote William Shakespeare? (the source for my discussion here) this was apparently a forgery made later in the 19th Century. Regardless, 19th Century ‘Baconian’ theorists such as Mark Twain and Helen Keller (!?) argued that the historical Shakespeare: that “money-lender and malt-dealer”; couldn’t possibly have known enough about the noble, the legal and the transcendent, to write like the author of Shakespeare’s plays.
In addition to this, the Baconians were pretty sure that Bacon was a wily code-master, and had for some unknown reason (perhaps due to his somewhat tense relationship with Queen Elizabeth, because why not?) filled Shakespeare’s plays with cyphers (check it out). This created a cottage industry dedicated to trawling through Shakespeare’s works looking for secret messages from Lord Verulam himself!
Now, as you can tell by my tone, I think all of this is rather ridiculous. And not ridiculous in the fun way that the history of science often is—when a bit of context dissolves the ridiculousness by revealing the sensible thought underlying it all—but ridiculous in the fun way that ‘aliens-built-the-pyramids’ hypotheses are. It’s ridiculous because its pseudoscience – it sorta looks like science if you squint your eyes right, but isn’t really. And it’s fun because in this instance it’s relatively safe. Who *really* cares who wrote Shakespeare's plays? And often pseudoscience ain’t fun because it really ain’t safe (hi anti-vaxxers…).
But what justifies my hubris here? Why am I happy with all of that speculative reconstruction of sauropod dinosaurs but feel justified in sniggering at 19th-century conspiracy theorists? After all, both seem to make claims which outrun their evidence—and that evidence is very indirect, and both tell, what seem to be, coherent stories. In a certain light, then, their warrants are similar. So what’s the difference? To put it more aggressively: why aren’t our reconstructions of sauropods on the same footing as our conspiracies about Will Shakespeare?
The rub, as Will would say, is that there’s no easy answer to this question. Scientific justification is (in my view) highly context-sensitive. So in telling science from pseudoscience, the details are going to matter - and these local details will wash-out our attempts to provide a one-size-fits-all demarcation criteria. But I think we can compare the Shakespeare conspiracy to sauropod reconstruction in revealing ways. Which at last brings us to...
The Crazy-Spotter’s Guide
First, to understand how science produces reliable knowledge, we should emphasize its social nature. Scientific knowledge is filtered through a powerful set of population-level processes, and these make it a fairly conservative beast. Specifically, long institutional training and testing, various kinds of review, and various kinds of publishing practices all lead science to err on the side of caution. It’s more likely for good stuff to go unpublished than it is for bad stuff to be published; false negatives are preferred over false positives. Of course, as Kyle Stanford has made clear, this conservatism comes with its own problems. Regardless, one difference between the Shakespeare-conspiracy and the sauropod-science is that the latter has gone through those social filters (the classic work on how social processes promote scientific objectivity is Helen Longino’s).
Second, consider the background theory which underwrites the studies. Evidence is evidence in virtue of background theories; taphonomy, for instance, tells us how fossilization works, and thus justifies our taking funny shaped rocks as evidence of D. schrani. Theories of taphonomy are very well understood due to years of careful examination of the processes of fossilization. By contrast, the conspiracy theory about Shakespeare, according to Shapiro, is based on the romantic idea that great art can only be produced by people who have lived the topic of said art. Shakespeare clearly didn’t lead the life of the nobles he often wrote about, so therefore he simply cannot have written so well about it. (What this suggests about the actual writer of the play’s interactions with mad wizards on islands and mischievous fairies I’m not sure…) I hope it’s clear that this is a pretty crappy piece of background theory. Historical scientists are always probing, testing, and extending their background theory—and this sets them apart from conspiracy theorists.
Third, think about the lines of evidence underwriting the hypotheses. It’s true that in the Shakespeare case quite a lot of evidence is drawn: various scraps about Shakespeare’s life, information about Bacon, all of those juicy cyphers contained in the writing. However, all of this evidence—in order to be evidence—depends on the same thing: the idea that you have to have experienced something to write well about it. By contrast, sauropod reconstruction is not simply based on our knowledge of fossilization. It also depends on our knowledge of the anatomy, physiology and behaviour of vertebrates—and this is built from careful study of living animals; it depends on our understanding of past climates and ecosystems; and it depends on our understanding of the ancestral relations between organisms. For our understanding of sauropods to completely fall, many interlocked, but partially independent, pieces must also fall down. For the Bacon-as-Shakespeare hypothesis to fail, well, that takes much less.
Fourth and finally, consider the nature of the speculation. The speculation of the conspiracy doesn’t lead to further tests—in fact it (as I’m sure old Popper would quickly point out) basically insulates the conspiracy from testing. By contrast, good scientific speculation actually promotes further scientific testing. Above I mentioned Sellers et al’s simulation of sauropod gait. That simulation was based on some fairly speculative reconstruction, and was fairly speculative itself. But the simulation could also estimate what kind of footprints such a critter would leave—and given that we have the remains of sauropod footprints, that leads to a further test. Those speculative tests didn’t insulate the hypothesis, but opened it up to further tests. Good speculation is productive, bad speculation is idle.
None of these features, or their combination, are necessary or sufficient for being good or bad science. Rather, they are diagnostic of the kinds of things which underlie good scientific practice. I take it this kind of point justifies my hubris. At least I hope it does. But then, I’m not too worried about losing bets.
To finish, I should reconsider Derek’s thought experiment. Are his Wollemi-pine cryptozoologists pseudoscientists? In several important respects, yes: their justification for thinking they’ll find the pine is weak, and hasn’t been filtered through the relevant social practices. I’d be tempted to say that in the case he describes, they count as doing the wrong thing, but getting very very lucky. However, I think it’s worth recognising that some sciences – paleontology most obviously – owe a lot to amateurs, sometimes quite adventurous amateurs. Fossil hounds certainly aren’t big-foot hunters, but in some ways do look quite similar to Derek’s hypothetical cryptozoologists. So, I’m not willing to write them off immediately. Moreover, in principle I think it’s a great idea for people to go into the woods and look for weird, unexpected stuff. They almost certainly won’t find Sasquatch, but they could find other unexpected, interesting, and important things.
So,a little crazy - if the cost isn't too high - can be helpful then. if you want to find Bigfoot, go for it. But, just for me, keep your eyes open for other stuff too okay?
*Of course, as Leonard has recently pointed out, we should tread carefully here: he would argue that D. schrani actually refers to the fossil, not to the extinct animal…
**There’s actually a fair bit of philosophy about conspiracy theories these days. I haven’t read much of it, but it tends to argue that conspiracy theories aren’t bad by definition because conspiracies sometimes happen. Fair enough.