Carl Sagan's Imaginary Dragon

In The Demon-Haunted World: Science as a Candle in the Dark (1997), Carl Sagan imagines that he has offered to show me a fire-breathing dragon in his garage, and, when I get there, I see only "a ladder, empty paint cans, and an old tricycle." He'd neglected, it seems, to mention that the dragon is invisible. This thought experiment imagines my proposing a series of ways to detect the dragon. They are good ideas, he imagines replying, but unfortunately won't work. Flour on the floor won't because dragons levitate, infrared won't because dragon fire is heatless, and spray paint won't because dragons are incorporeal. If Sagan insists on countering every test I propose with an evasive redefinition, it's clearly silly for him to then shift the burden of proof onto me and suggest that, since I can't prove the dragon is absent, it may very well be present. There's no practical difference between an undetectable dragon and no dragon at all.

This parallels Russell's Teapot. Bertrand Russell notes that a tiny teapot orbiting between Mercury and Venus is not a genuine astronomical theory because it can't be tested: if it's offered without proof, it can be dismissed without proof. Karl Popper famously describes real scientific statements as falsifiable--that is, open to tests capable of proving them false. Not all important hypotheses can be proven true, of course, but, if they systematically exclude tests that might prove them false, they are fantasies, maybe myth or poetry but not science. Hypotheses of theoretical physics are, of course, typically unproven at their inception--even Einstein's relativity was--but, unlike Sagan's dragon-believer, scientists welcome ways to test their theories.

Science routinely achieves what I call Kalashnikov truth by convincing people of all ideologies with objective results. The 1979-1989 war between the Soviet Union and the mujahideen was a conflict between radically different ideologies--atheistic modernism and religious medievalism--but the  medievalist side fought with Kalashnikovs rifles, not swords, because it would have been suicide to do otherwise. Sagan's dragon could achieve Kalashnikov truth--at least as an invisible presence (if not as a fairytale dragon)--if independent observers were able to detect its heat signature or maybe if it flew around the neighborhood torching houses. Absent such evidence, we can only suspect that Sagan's believer is hallucinating or otherwise unhinged. "Claims that cannot be tested, assertions immune to disproof, are veridically worthless, whatever value they may have in inspiring us or in exciting our sense of wonder."

The main targets of Sagan's Demon-haunted World are pseudoscientific beliefs such as anti-vaxxing, crop circles, and Big Foot, but religions are easy targets of his thought experiment if they claim that their myths and doctrines are true in the same sense as my car is in the garage but block all means of falsifying them. Sagan allows that dragon tales may be valuable if they inspire and excite wonder. He allows poetry but opposes pseudoscience. My car is in the garage can be verified by raising the overhead door and showing title and registration. It's Kalashnikov truth, not a matter of opinion. Such truths trump other kinds of statements when available. This isn't very often though because ordinary discourse is freighted with values, feelings, phobias, priorities, hopes, speculations, and predictions, all of which propose garage-dwelling dragons of varying degrees of visibility--or at least varying distorting resonances in our ideological echo chambers. Epistemologist Joseph H. Shieber argues convincingly that we all tend to outsource verification of our beliefs to social networks. We know very little with Kalashnikov certainty outside the range of our vision and hearing, and even that is clouded by the vagaries of memory.

Sagan expands his thought experiment to imagine that some of my neighbors independently claim dragons in their garages too in spite of a total lack of physical evidence. None of these neighbors appears to be insane. This recalls Christopher Isherwood's claim in Vedanta for the Western World (1946) that an underlying agreement of saints from all cultures and eras demonstrates a spiritual reality underlying world mysticism (pp. 36-40), not to speak of the millions of ordinary believers who profess a living relationship with an invisible God. The classic hymn, "He Lives," comes to mind. Still, even if there were anecdotal scraps of physical evidence, the equivalent of miracles, Sagan concludes that widespread belief in garage-dragons is no reason to suppose that they exist: "Once again, the only sensible approach is tentatively to reject the dragon hypothesis, to be open to future physical data [italics mine], and to wonder what the cause might be that so many apparently sane and sober people share the same strange delusion." 

So ends the three-page Sagan quotation "Dragon in My Garage" posted by Atheists of Silicon Valley (godlessgeeks.com), and this makes sense as far as it goes. But note the term physical data. If a believer supposes that her dragon exists in the same sense that her Subaru does and so she needs to build a second garage, these physical consequences would demand physical data. If, however, her car and dragon occupy the same "space" in different dimensions (or whatever) and the dragon functions to inspire her and excite her sense of wonder--functions that Sagan accepts invisible dragons may valuably perform--then physical data is irrelevant.

                    But let me expand on the thought experiment a bit. Suppose there are measurable effects of invisible dragons, not in garages, but in the persons of their owners. Suppose that (adjusted for other variables) dragon-believers are less depressed than non-believers, or maybe they become significantly less depressed after they begin to believe. What would this prove? Not, of course, that dragons exist physically, but rather that believing, knowing, or assuming a dragon in one's garage was an effective antidepressant. In this case, therapists might try to persuade depressed patients that they had dragons in their garages--or, more realistically, prescribe visualizations, imaginary garages housing imaginary dragons. Suppose this therapy successfully combats depression by inspiring patients with a sense of wonder (Sagan's terms). In this case, wouldn't a depressed person be a fool not to at least try to believe, damn the lack of physical evidence?

You see where this argument is headed. It's not about dragons, but about belief in some particular religion, and it's not a slam-dunk. There are at least two major objections to it. First, the chaotic set of all belief systems called religion doesn't reliably deliver benefits comparable to Zoloft. There are all sorts of religions, some clearly addictive and toxic. Scandalously, Christian churches seem to deliver self-righteousness and hypocrisy as reliably as love and virtue. The Atheists of Silicon Valley post a cartoon of a boy asking his father why he supports a politician who violates the teachings of the Bible. The father says: "Oh, Billy, we don't actually practice these things. We only preach them." The most damning criticisms of religion are often true, like the most damning criticisms of government, medicine, or education. But the teachings of many religions, if actually followed, would lead to good outcomes, and I know people whose daily lives have been improved, even saved, by religious conversion, and many more who draw consolation from its invisible support, an effect at least as powerful for them as Zoloft.

The second objection goes back to Pascal's Wager: everybody should believe in God because, as a gamble, it's more lucrative than its opposite. The problem is that belief faked for profit can't be expected to work, which leads to Pascal's dubious advice to the unbeliever: the theological equivalent of fake it till you make it (Pensees 418). Religion as garage-dragon shares this problem. If you're quite certain there's no dragon in your garage and the idea doesn't turn you on, it may be impossible to believe even if you know it's a good antidepressant and you're blue as hell. You can't force yourself to believe contrary to your understanding any more than you avoid thinking about a purple rhinoceros by trying not to.

            If you really need to believe there's a dragon in your garage, your best bet is to avoid cognitive dissonance by seeing it on a different plane of existence from your Subaru, maybe on a metaphorical (and, of course, spiritual) plane. It's not exactly a dragon-dragon, of course, but something dragonish--not exactly in your garage, but maybe in the Garage of Garages, outside Plato's cave of shadows. Some people do park there. Sagan's dragon is designed to make pseudo-scientifically assured non-scientific belief look silly, and it works. But Hamlet might answer, "There are more things in heaven and earth, Sagan, than are dreamt of in your physical data."