Try­ing to Try

“No! Try not! Do, or do not. There is no try.”
Yoda

Years ago, I thought this was yet an­other ex­ample of Deep Wis­dom that is ac­tu­ally quite stu­pid. SUCCEED is not a prim­it­ive ac­tion. You can’t just de­cide to win by choos­ing hard enough. There is never a plan that works with prob­ab­il­ity 1.

But Yoda was wiser than I first real­ized.

The first ele­ment­ary tech­nique of epi­stem­o­logy—it’s not deep, but it’s cheap—is to dis­tin­guish the quo­ta­tion from the ref­er­ent. Talk­ing about snow is not the same as talk­ing about “snow”. When I use the word “snow”, without quotes, I mean to talk about snow; and when I use the word “”snow”“, with quotes, I mean to talk about the word “snow”. You have to enter a spe­cial mode, the quo­ta­tion mode, to talk about your be­liefs. By de­fault, we just talk about real­ity.

If someone says, “I’m go­ing to flip that switch”, then by de­fault, they mean they’re go­ing to try to flip the switch. They’re go­ing to build a plan that prom­ises to lead, by the con­sequences of its ac­tions, to the goal-state of a flipped switch; and then ex­ecute that plan.

No plan suc­ceeds with in­fin­ite cer­tainty. So by de­fault, when you talk about set­ting out to achieve a goal, you do not im­ply that your plan ex­actly and per­fectly leads to only that pos­sib­il­ity. But when you say, “I’m go­ing to flip that switch”, you are try­ing only to flip the switch—not try­ing to achieve a 97.2% prob­ab­il­ity of flip­ping the switch.

So what does it mean when someone says, “I’m go­ing to try to flip that switch?”

Well, col­lo­qui­ally, “I’m go­ing to flip the switch” and “I’m go­ing to try to flip the switch” mean more or less the same thing, ex­cept that the lat­ter ex­presses the pos­sib­il­ity of fail­ure. This is why I ori­gin­ally took of­fense at Yoda for seem­ing to deny the pos­sib­il­ity. But bear with me here.

Much of life’s chal­lenge con­sists of hold­ing ourselves to a high enough stand­ard. I may speak more on this prin­ciple later, be­cause it’s a lens through which you can view many-but-not-all per­sonal di­lem­mas—“What stand­ard am I hold­ing my­self to? Is it high enough?”

So if much of life’s fail­ure con­sists in hold­ing your­self to too low a stand­ard, you should be wary of de­mand­ing too little from your­self—set­ting goals that are too easy to ful­fill.

Often where suc­ceed­ing to do a thing, is very hard, try­ing to do it is much easier.

Which is easier—to build a suc­cess­ful star­tup, or to try to build a suc­cess­ful star­tup? To make a mil­lion dol­lars, or to try to make a mil­lion dol­lars?

So if “I’m go­ing to flip the switch” means by de­fault that you’re go­ing to try to flip the switch—that is, you’re go­ing to set up a plan that prom­ises to lead to switch-flipped state, maybe not with prob­ab­il­ity 1, but with the highest prob­ab­il­ity you can man­age—

—then “I’m go­ing to ‘try to flip’ the switch” means that you’re go­ing to try to “try to flip the switch”, that is, you’re go­ing to try to achieve the goal-state of “hav­ing a plan that might flip the switch”.

Now, if this were a self-modi­fy­ing AI we were talk­ing about, the trans­form­a­tion we just per­formed ought to end up at a re­flect­ive equi­lib­rium—the AI plan­ning its plan­ning op­er­a­tions.

But when we deal with hu­mans, be­ing sat­is­fied with hav­ing a plan is not at all like be­ing sat­is­fied with suc­cess. The part where the plan has to max­im­ize your prob­ab­il­ity of suc­ceed­ing, gets lost along the way. It’s far easier to con­vince ourselves that we are “max­im­iz­ing our prob­ab­il­ity of suc­ceed­ing”, than it is to con­vince ourselves that we will suc­ceed.

Al­most any ef­fort will serve to con­vince us that we have “tried our hard­est”, if try­ing our hard­est is all we are try­ing to do.

“You have been ask­ing what you could do in the great events that are now stir­ring, and have found that you could do noth­ing. But that is be­cause your suf­fer­ing has caused you to phrase the ques­tion in the wrong way… In­stead of ask­ing what you could do, you ought to have been ask­ing what needs to be done.”
—Steven Brust, The Paths of the Dead

When you ask, “What can I do?“, you’re try­ing to do your best. What is your best? It is whatever you can do without the slight­est in­con­veni­ence. It is whatever you can do with the money in your pocket, minus whatever you need for your ac­cus­tomed lunch. What you can do with those re­sources, may not give you very good odds of win­ning. But it’s the “best you can do”, and so you’ve ac­ted de­fens­ibly, right?

But what needs to be done? Maybe what needs to be done re­quires three times your life sav­ings, and you must pro­duce it or fail.

So try­ing to have “max­im­ized your prob­ab­il­ity of suc­cess”—as op­posed to try­ing to suc­ceed—is a far lesser bar­rier. You can have “max­im­ized your prob­ab­il­ity of suc­cess” us­ing only the money in your pocket, so long as you don’t de­mand ac­tu­ally win­ning.

Want to try to make a mil­lion dol­lars? Buy a lot­tery ticket. Your odds of win­ning may not be very good, but you did try, and try­ing was what you wanted. In fact, you tried your best, since you only had one dol­lar left after buy­ing lunch. Max­im­iz­ing the odds of goal achieve­ment us­ing avail­able re­sources: is this not in­tel­li­gence?

It’s only when you want, above all else, to ac­tu­ally flip the switch—without quo­ta­tion and without con­sol­a­tion prizes just for try­ing—that you will ac­tu­ally put in the ef­fort to ac­tu­ally max­im­ize the prob­ab­il­ity.

But if all you want is to “max­im­ize the prob­ab­il­ity of suc­cess us­ing avail­able re­sources”, then that’s the easi­est thing in the world to con­vince your­self you’ve done. The very first plan you hit upon, will serve quite well as “max­im­iz­ing”—if ne­ces­sary, you can gen­er­ate an in­ferior al­tern­at­ive to prove its op­tim­al­ity. And any tiny re­source that you care to put in, will be what is “avail­able”. Re­mem­ber to con­grat­u­late your­self on put­ting in 100% of it!

Don’t try your best. Win, or fail. There is no best.