What is Evidence?

“The sen­tence ‘snow is white’ is true if and only if snow is white.”
—Alfred Tarski
”To say of what is, that it is, or of what is not, that it is not, is true.
—Aris­to­tle, Me­ta­physics IV

If these two quotes don’t seem like a suffi­cient defi­ni­tion of “truth”, read this. To­day I’m go­ing to talk about “ev­i­dence”. (I also in­tend to dis­cuss be­liefs-of-fact, not emo­tions or moral­ity, as dis­t­in­guished here.)

Walk­ing along the street, your shoelaces come un­tied. Shortly there­after, for some odd rea­son, you start be­liev­ing your shoelaces are un­tied. Light leaves the Sun and strikes your shoelaces and bounces off; some pho­tons en­ter the pupils of your eyes and strike your retina; the en­ergy of the pho­tons trig­gers neu­ral im­pulses; the neu­ral im­pulses are trans­mit­ted to the vi­sual-pro­cess­ing ar­eas of the brain; and there the op­ti­cal in­for­ma­tion is pro­cessed and re­con­structed into a 3D model that is rec­og­nized as an un­tied shoelace. There is a se­quence of events, a chain of cause and effect, within the world and your brain, by which you end up be­liev­ing what you be­lieve. The fi­nal out­come of the pro­cess is a state of mind which mir­rors the state of your ac­tual shoelaces.

What is ev­i­dence? It is an event en­tan­gled, by links of cause and effect, with what­ever you want to know about. If the tar­get of your in­quiry is your shoelaces, for ex­am­ple, then the light en­ter­ing your pupils is ev­i­dence en­tan­gled with your shoelaces. This should not be con­fused with the tech­ni­cal sense of “en­tan­gle­ment” used in physics—here I’m just talk­ing about “en­tan­gle­ment” in the sense of two things that end up in cor­re­lated states be­cause of the links of cause and effect be­tween them.

Not ev­ery in­fluence cre­ates the kind of “en­tan­gle­ment” re­quired for ev­i­dence. It’s no help to have a ma­chine that beeps when you en­ter win­ning lot­tery num­bers, if the ma­chine also beeps when you en­ter los­ing lot­tery num­bers. The light re­flected from your shoes would not be use­ful ev­i­dence about your shoelaces, if the pho­tons ended up in the same phys­i­cal state whether your shoelaces were tied or un­tied.

To say it ab­stractly: For an event to be ev­i­dence about a tar­get of in­quiry, it has to hap­pen differ­ently in a way that’s en­tan­gled with the differ­ent pos­si­ble states of the tar­get. (To say it tech­ni­cally: There has to be Shan­non mu­tual in­for­ma­tion be­tween the ev­i­den­tial event and the tar­get of in­quiry, rel­a­tive to your cur­rent state of un­cer­tainty about both of them.)

En­tan­gle­ment can be con­ta­gious when pro­cessed cor­rectly, which is why you need eyes and a brain. If pho­tons re­flect off your shoelaces and hit a rock, the rock won’t change much. The rock won’t re­flect the shoelaces in any helpful way; it won’t be de­tectably differ­ent de­pend­ing on whether your shoelaces were tied or un­tied. This is why rocks are not use­ful wit­nesses in court. A pho­to­graphic film will con­tract shoelace-en­tan­gle­ment from the in­com­ing pho­tons, so that the photo can it­self act as ev­i­dence. If your eyes and brain work cor­rectly, you will be­come tan­gled up with your own shoelaces.

This is why ra­tio­nal­ists put such a heavy pre­mium on the para­dox­i­cal-seem­ing claim that a be­lief is only re­ally worth­while if you could, in prin­ci­ple, be per­suaded to be­lieve oth­er­wise. If your retina ended up in the same state re­gard­less of what light en­tered it, you would be blind. Some be­lief sys­tems, in a rather ob­vi­ous trick to re­in­force them­selves, say that cer­tain be­liefs are only re­ally worth­while if you be­lieve them un­con­di­tion­ally— no mat­ter what you see, no mat­ter what you think. Your brain is sup­posed to end up in the same state re­gard­less. Hence the phrase, “blind faith”. If what you be­lieve doesn’t de­pend on what you see, you’ve been blinded as effec­tively as by pok­ing out your eye­balls.

If your eyes and brain work cor­rectly, your be­liefs will end up en­tan­gled with the facts. Ra­tional thought pro­duces be­liefs which are them­selves ev­i­dence.

If your tongue speaks truly, your ra­tio­nal be­liefs, which are them­selves ev­i­dence, can act as ev­i­dence for some­one else. En­tan­gle­ment can be trans­mit­ted through chains of cause and effect—and if you speak, and an­other hears, that too is cause and effect. When you say “My shoelaces are un­tied” over a cel­l­phone, you’re shar­ing your en­tan­gle­ment with your shoelaces with a friend.

There­fore ra­tio­nal be­liefs are con­ta­gious, among hon­est folk who be­lieve each other to be hon­est. And it’s why a claim that your be­liefs are not con­ta­gious—that you be­lieve for pri­vate rea­sons which are not trans­mis­si­ble—is so sus­pi­cious. If your be­liefs are en­tan­gled with re­al­ity, they should be con­ta­gious among hon­est folk.

If your model of re­al­ity sug­gests that the out­puts of your thought pro­cesses should not be con­ta­gious to oth­ers, then your model says that your be­liefs are not them­selves ev­i­dence, mean­ing they are not en­tan­gled with re­al­ity. You should ap­ply a re­flec­tive cor­rec­tion, and stop be­liev­ing.

In­deed, if you feel, on a gut level, what this all means, you will au­to­mat­i­cally stop be­liev­ing. Be­cause “my be­lief is not en­tan­gled with re­al­ity” means “my be­lief is not ac­cu­rate”. As soon as you stop be­liev­ing “‘snow is white’ is true”, you should (au­to­mat­i­cally!) stop be­liev­ing “snow is white”, or some­thing is very wrong.

So go ahead and ex­plain why the kind of thought pro­cesses you use sys­tem­at­i­cally pro­duce be­liefs that mir­ror re­al­ity. Ex­plain why you think you’re ra­tio­nal. Why you think that, us­ing thought pro­cesses like the ones you use, minds will end up be­liev­ing “snow is white” if and only if snow is white. If you don’t be­lieve that the out­puts of your thought pro­cesses are en­tan­gled with re­al­ity, why do you be­lieve the out­puts of your thought pro­cesses? It’s the same thing, or it should be.