The Proper Use of Doubt

Once, when I was hold­ing forth upon the Way, I re­marked upon how most or­ga­nized be­lief sys­tems ex­ist to flee from doubt. A listener replied to me that the Je­suits must be im­mune from this crit­i­cism, be­cause they prac­tice or­ga­nized doubt: their novices, he said, are told to doubt Chris­ti­an­ity; doubt the ex­is­tence of God; doubt if their call­ing is real; doubt that they are suit­able for per­pet­ual vows of chastity and poverty. And I said: Ah, but they’re sup­posed to over­come these doubts, right? He said: No, they are to doubt that per­haps their doubts may grow and be­come stronger.

Googling failed to con­firm or re­fute these alle­ga­tions. But I find this sce­nario fas­ci­nat­ing, wor­thy of dis­cus­sion, re­gard­less of whether it is true or false of Je­suits. If the Je­suits prac­ticed de­liber­ate doubt, as de­scribed above, would they there­fore be vir­tu­ous as ra­tio­nal­ists?

I think I have to con­cede that the Je­suits, in the (pos­si­bly hy­po­thet­i­cal) sce­nario above, would not prop­erly be de­scribed as “flee­ing from doubt.” But the (pos­si­bly hy­po­thet­i­cal) con­duct still strikes me as highly sus­pi­cious. To a truly vir­tu­ous ra­tio­nal­ist, doubt should not be scary. The con­duct de­scribed above sounds to me like a pro­gram of de­sen­si­ti­za­tion for some­thing very scary, like ex­pos­ing an arachno­phobe to spi­ders un­der care­fully con­trol­led con­di­tions.

But even so, they are en­courag­ing their novices to doubt—right? Does it mat­ter if their rea­sons are flawed? Is this not still a wor­thy deed unto a ra­tio­nal­ist?

All cu­ri­os­ity seeks to an­nihilate it­self; there is no cu­ri­os­ity that does not want an an­swer. But if you ob­tain an an­swer, if you satisfy your cu­ri­os­ity, then the glo­ri­ous mys­tery will no longer be mys­te­ri­ous.

In the same way, ev­ery doubt ex­ists in or­der to an­nihilate some par­tic­u­lar be­lief. If a doubt fails to de­stroy its tar­get, the doubt has died un­fulfilled—but that is still a re­s­olu­tion, an end­ing, albeit a sad­der one. A doubt that nei­ther de­stroys it­self nor de­stroys its tar­get might as well have never ex­isted at all. It is the re­s­olu­tion of doubts, not the mere act of doubt­ing, which drives the ratchet of ra­tio­nal­ity for­ward.

Every im­prove­ment is a change, but not ev­ery change is an im­prove­ment. Every ra­tio­nal­ist doubts, but not all doubts are ra­tio­nal. Wear­ing doubts doesn’t make you a ra­tio­nal­ist any more than wear­ing a white med­i­cal lab coat makes you a doc­tor.

A ra­tio­nal doubt comes into ex­is­tence for a spe­cific rea­son—you have some spe­cific jus­tifi­ca­tion to sus­pect the be­lief is wrong. This rea­son, in turn, im­plies an av­enue of in­ves­ti­ga­tion which will ei­ther de­stroy the tar­geted be­lief or de­stroy the doubt. This holds even for highly ab­stract doubts, like: “I won­der if there might be a sim­pler hy­poth­e­sis which also ex­plains this data.” In this case you in­ves­ti­gate by try­ing to think of sim­pler hy­pothe­ses. As this search con­tinues longer and longer with­out fruit, you will think it less and less likely that the next in­cre­ment of com­pu­ta­tion will be the one to suc­ceed. Even­tu­ally the cost of search­ing will ex­ceed the ex­pected benefit, and you’ll stop search­ing. At which point you can no longer claim to be use­fully doubt­ing. A doubt that is not in­ves­ti­gated might as well not ex­ist. Every doubt ex­ists to de­stroy it­self, one way or the other. An un­re­solved doubt is a null-op; it does not turn the wheel, nei­ther for­ward nor back.

If you re­ally be­lieve a re­li­gion (and don’t just be­lieve in it), then why would you tell your novices to con­sider doubts that must die un­fulfilled? It would be like tel­ling physics stu­dents to ag­o­nize over whether the twen­tieth-cen­tury rev­olu­tion might have been a mis­take, and that New­to­nian me­chan­ics was cor­rect all along. If you don’t re­ally doubt some­thing, why would you pre­tend that you do?

Be­cause we all want to be seen as ra­tio­nal—and doubt­ing is widely be­lieved to be a virtue of a ra­tio­nal­ist. But it is not widely un­der­stood that you need a par­tic­u­lar rea­son to doubt, or that an un­re­solved doubt is a null-op. In­stead peo­ple think it’s about mod­esty, a sub­mis­sive de­meanor, main­tain­ing the tribal sta­tus hi­er­ar­chy—al­most ex­actly the same prob­lem as with hu­mil­ity, on which I have pre­vi­ously writ­ten. Mak­ing a great pub­lic dis­play of doubt to con­vince your­self that you are a ra­tio­nal­ist will do around as much good as wear­ing a lab coat.

To avoid merely pro­fess­ing doubts,1 re­mem­ber:

  • A ra­tio­nal doubt ex­ists to de­stroy its tar­get be­lief, and if it does not de­stroy its tar­get it dies un­fulfilled.

  • A ra­tio­nal doubt arises from some spe­cific rea­son the be­lief might be wrong.

  • An un­re­solved doubt is a null-op.

  • An un­in­ves­ti­gated doubt might as well not ex­ist.

  • You should not be proud of mere doubt­ing, al­though you can justly be proud when you have just finished tear­ing a cher­ished be­lief to shreds.

  • Though it may take courage to face your doubts, never for­get that to an ideal mind doubt would not be scary in the first place.

1See “Pro­fess­ing and Cheer­ing” in Map and Ter­ri­tory.