Taboo Your Words

In the game Ta­boo (by Has­bro), the ob­jec­tive is for a player to have their part­ner guess a word writ­ten on a card, with­out us­ing that word or five ad­di­tional words listed on the card. For ex­am­ple, you might have to get your part­ner to say “base­ball” with­out us­ing the words “sport”, “bat”, “hit”, “pitch”, “base” or of course “base­ball”.

As soon as I see a prob­lem like that, I at once think, “An ar­tifi­cial group con­flict in which you use a long wooden cylin­der to whack a thrown spheroid, and then run be­tween four safe po­si­tions.” It might not be the most effi­cient strat­egy to con­vey the word ‘base­ball’ un­der the stated rules—that might be, “It’s what the Yan­kees play”—but the gen­eral skill of blank­ing a word out of my mind was one I’d prac­ticed for years, albeit with a differ­ent pur­pose.

Yes­ter­day we saw how re­plac­ing terms with defi­ni­tions could re­veal the em­piri­cal un­pro­duc­tivity of the clas­si­cal Aris­totelian syl­l­o­gism. All hu­mans are mor­tal (and also, ap­par­ently, feather­less bipeds); Socrates is hu­man; there­fore Socrates is mor­tal. When we re­place the word ‘hu­man’ by its ap­par­ent defi­ni­tion, the fol­low­ing un­der­ly­ing rea­son­ing is re­vealed:

All [mor­tal, ~feathers, biped] are mor­tal;
Socrates is a [mor­tal, ~feathers, biped];
There­fore Socrates is mor­tal.

But the prin­ci­ple of re­plac­ing words by defi­ni­tions ap­plies much more broadly:

Albert: “A tree fal­ling in a de­serted for­est makes a sound.”
Barry: “A tree fal­ling in a de­serted for­est does not make a sound.”

Clearly, since one says “sound” and one says “not sound”, we must have a con­tra­dic­tion, right? But sup­pose that they both derefer­ence their poin­t­ers be­fore speak­ing:

Albert: “A tree fal­ling in a de­serted for­est matches [mem­ber­ship test: this event gen­er­ates acous­tic vibra­tions].”
Barry: “A tree fal­ling in a de­serted for­est does not match [mem­ber­ship test: this event gen­er­ates au­di­tory ex­pe­riences].”

Now there is no longer an ap­par­ent col­li­sion—all they had to do was pro­hibit them­selves from us­ing the word sound. If “acous­tic vibra­tions” came into dis­pute, we would just play Ta­boo again and say “pres­sure waves in a ma­te­rial medium”; if nec­es­sary we would play Ta­boo again on the word “wave” and re­place it with the wave equa­tion. (Play Ta­boo on “au­di­tory ex­pe­rience” and you get “That form of sen­sory pro­cess­ing, within the hu­man brain, which takes as in­put a lin­ear time se­ries of fre­quency mixes...”)

But sup­pose, on the other hand, that Albert and Barry were to have the ar­gu­ment:

Albert: “Socrates matches the con­cept [mem­ber­ship test: this per­son will die af­ter drink­ing hem­lock].”
Barry: “Socrates matches the con­cept [mem­ber­ship test: this per­son will not die af­ter drink­ing hem­lock].”

Now Albert and Barry have a sub­stan­tive clash of ex­pec­ta­tions; a differ­ence in what they an­ti­ci­pate see­ing af­ter Socrates drinks hem­lock. But they might not no­tice this, if they hap­pened to use the same word “hu­man” for their differ­ent con­cepts.

You get a very differ­ent pic­ture of what peo­ple agree or dis­agree about, de­pend­ing on whether you take a la­bel’s-eye-view (Albert says “sound” and Barry says “not sound”, so they must dis­agree) or tak­ing the test’s-eye-view (Albert’s mem­ber­ship test is acous­tic vibra­tions, Barry’s is au­di­tory ex­pe­rience).

Get to­gether a pack of soi-dis­ant fu­tur­ists and ask them if they be­lieve we’ll have Ar­tifi­cial In­tel­li­gence in thirty years, and I would guess that at least half of them will say yes. If you leave it at that, they’ll shake hands and con­grat­u­late them­selves on their con­sen­sus. But make the term “Ar­tifi­cial In­tel­li­gence” taboo, and ask them to de­scribe what they ex­pect to see, with­out ever us­ing words like “com­put­ers” or “think”, and you might find quite a con­flict of ex­pec­ta­tions hid­ing un­der that fea­ture­less stan­dard word. Like­wise that other term. And see also Shane Legg’s com­pila­tion of 71 defi­ni­tions of “in­tel­li­gence”.

The illu­sion of unity across re­li­gions can be dis­pel­led by mak­ing the term “God” taboo, and ask­ing them to say what it is they be­lieve in; or mak­ing the word “faith” taboo, and ask­ing them why they be­lieve it. Though mostly they won’t be able to an­swer at all, be­cause it is mostly pro­fes­sion in the first place, and you can­not cog­ni­tively zoom in on an au­dio record­ing.

When you find your­self in philo­soph­i­cal difficul­ties, the first line of defense is not to define your prob­le­matic terms, but to see whether you can think with­out us­ing those terms at all. Or any of their short syn­onyms. And be care­ful not to let your­self in­vent a new word to use in­stead. De­scribe out­ward ob­serv­ables and in­te­rior mechanisms; don’t use a sin­gle han­dle, what­ever that han­dle may be.

Albert says that peo­ple have “free will”. Barry says that peo­ple don’t have “free will”. Well, that will cer­tainly gen­er­ate an ap­par­ent con­flict. Most philoso­phers would ad­vise Albert and Barry to try to define ex­actly what they mean by “free will”, on which topic they will cer­tainly be able to dis­course at great length. I would ad­vise Albert and Barry to de­scribe what it is that they think peo­ple do, or do not have, with­out us­ing the phrase “free will” at all. (If you want to try this at home, you should also avoid the words “choose”, “act”, “de­cide”, “de­ter­mined”, “re­spon­si­ble”, or any of their syn­onyms.)

This is one of the non­stan­dard tools in my toolbox, and in my hum­ble opinion, it works way way bet­ter than the stan­dard one. It also re­quires more effort to use; you get what you pay for.