The Argument from Common Usage

Part of the Stan­dard Defi­ni­tional Dis­pute runs as fol­lows:

Albert: “Look, sup­pose that I left a micro­phone in the for­est and recorded the pat­tern of the acous­tic vibra­tions of the tree fal­ling. If I played that back to some­one, they’d call it a ‘sound’! That’s the com­mon us­age! Don’t go around mak­ing up your own wacky defi­ni­tions!”

Barry: “One, I can define a word any way I like so long as I use it con­sis­tently. Two, the mean­ing I gave was in the dic­tio­nary. Three, who gave you the right to de­cide what is or isn’t com­mon us­age?”

Not all defi­ni­tional dis­putes progress as far as rec­og­niz­ing the no­tion of com­mon us­age. More of­ten, I think, some­one picks up a dic­tio­nary be­cause they be­lieve that words have mean­ings, and the dic­tio­nary faith­fully records what this mean­ing is. Some peo­ple even seem to be­lieve that the dic­tio­nary de­ter­mines the mean­ing—that the dic­tio­nary ed­i­tors are the Leg­is­la­tors of Lan­guage. Maybe be­cause back in el­e­men­tary school, their au­thor­ity-teacher said that they had to obey the dic­tio­nary, that it was a manda­tory rule rather than an op­tional one?

Dic­tionary ed­i­tors read what other peo­ple write, and record what the words seem to mean; they are his­to­ri­ans. The Oxford English Dic­tionary may be com­pre­hen­sive, but never au­thor­i­ta­tive.

But surely there is a so­cial im­per­a­tive to use words in a com­monly un­der­stood way? Does not our hu­man telepa­thy, our valuable power of lan­guage, rely on mu­tual co­or­di­na­tion to work? Per­haps we should vol­un­tar­ily treat dic­tio­nary ed­i­tors as supreme ar­biters—even if they pre­fer to think of them­selves as his­to­ri­ans—in or­der to main­tain the quiet co­op­er­a­tion on which all speech de­pends.

The phrase “au­thor­i­ta­tive dic­tio­nary” is al­most never used cor­rectly, an ex­am­ple of proper us­age be­ing the Author­i­ta­tive Dic­tionary of IEEE Stan­dards. The IEEE is a body of vot­ing mem­bers who have a pro­fes­sional need for ex­act agree­ment on terms and defi­ni­tions, and so the Author­i­ta­tive Dic­tionary of IEEE Stan­dards is ac­tual, ne­go­ti­ated leg­is­la­tion, which ex­erts what­ever au­thor­ity one re­gards as re­sid­ing in the IEEE.

In ev­ery­day life, shared lan­guage usu­ally does not arise from a de­liber­ate agree­ment, as of the IEEE. It’s more a mat­ter of in­fec­tion, as words are in­vented and diffuse through the cul­ture. (A “meme”, one might say, fol­low­ing Richard Dawk­ins thirty years ago—but you already know what I mean, and if not, you can look it up on Google, and then you too will have been in­fected.)

Yet as the ex­am­ple of the IEEE shows, agree­ment on lan­guage can also be a co­op­er­a­tively es­tab­lished pub­lic good. If you and I wish to un­dergo an ex­change of thoughts via lan­guage, the hu­man telepa­thy, then it is in our mu­tual in­ter­est that we use the same word for similar con­cepts—prefer­ably, con­cepts similar to the limit of re­s­olu­tion in our brain’s rep­re­sen­ta­tion thereof—even though we have no ob­vi­ous mu­tual in­ter­est in us­ing any par­tic­u­lar word for a con­cept.

We have no ob­vi­ous mu­tual in­ter­est in us­ing the word “oto” to mean sound, or “sound” to mean oto; but we have a mu­tual in­ter­est in us­ing the same word, whichever word it hap­pens to be. (Prefer­ably, words we use fre­quently should be short, but let’s not get into in­for­ma­tion the­ory just yet.)

But, while we have a mu­tual in­ter­est, it is not strictly nec­es­sary that you and I use the similar la­bels in­ter­nally; it is only con­ve­nient. If I know that, to you, “oto” means sound—that is, you as­so­ci­ate “oto” to a con­cept very similar to the one I as­so­ci­ate to “sound”—then I can say “Paper crum­pling makes a crack­ling oto.” It re­quires ex­tra thought, but I can do it if I want.

Similarly, if you say “What is the walk­ing-stick of a bowl­ing ball drop­ping on the floor?” and I know which con­cept you as­so­ci­ate with the syl­la­bles “walk­ing-stick”, then I can figure out what you mean. It may re­quire some thought, and give me pause, be­cause I or­di­nar­ily as­so­ci­ate “walk­ing-stick” with a differ­ent con­cept. But I can do it just fine.

When hu­mans re­ally want to com­mu­ni­cate with each other, we’re hard to stop! If we’re stuck on a de­serted is­land with no com­mon lan­guage, we’ll take up sticks and draw pic­tures in sand.

Albert’s ap­peal to the Ar­gu­ment from Com­mon Usage as­sumes that agree­ment on lan­guage is a co­op­er­a­tively es­tab­lished pub­lic good. Yet Albert as­sumes this for the sole pur­pose of rhetor­i­cally ac­cus­ing Barry of break­ing the agree­ment, and en­dan­ger­ing the pub­lic good. Now the fal­ling-tree ar­gu­ment has gone all the way from botany to se­man­tics to poli­tics; and so Barry re­sponds by challeng­ing Albert for the au­thor­ity to define the word.

A ra­tio­nal­ist, with the dis­ci­pline of hug­ging the query ac­tive, would no­tice that the con­ver­sa­tion had gone rather far astray.

Oh, dear reader, is it all re­ally nec­es­sary? Albert knows what Barry means by “sound”. Barry knows what Albert means by “sound”. Both Albert and Barry have ac­cess to words, such as “acous­tic vibra­tions” or “au­di­tory ex­pe­rience”, which they already as­so­ci­ate to the same con­cepts, and which can de­scribe events in the for­est with­out am­bi­guity. If they were stuck on a de­serted is­land, try­ing to com­mu­ni­cate with each other, their work would be done.

When both sides know what the other side wants to say, and both sides ac­cuse the other side of defect­ing from “com­mon us­age”, then what­ever it is they are about, it is clearly not work­ing out a way to com­mu­ni­cate with each other. But this is the whole benefit that com­mon us­age pro­vides in the first place.

Why would you ar­gue about the mean­ing of a word, two sides try­ing to wrest it back and forth? If it’s just a names­pace con­flict that has got­ten blown out of pro­por­tion, and noth­ing more is at stake, then the two sides need merely gen­er­ate two new words and use them con­sis­tently.

Yet of­ten cat­e­go­riza­tions func­tion as hid­den in­fer­ences and dis­guised queries. Is athe­ism a “re­li­gion”? If some­one is ar­gu­ing that the rea­son­ing meth­ods used in athe­ism are on a par with the rea­son­ing meth­ods used in Ju­daism, or that athe­ism is on a par with Is­lam in terms of causally en­gen­der­ing vi­o­lence, then they have a clear ar­gu­men­ta­tive stake in lump­ing it all to­gether into an in­dis­tinct gray blur of “faith”.

Or con­sider the fight to blend to­gether blacks and whites as “peo­ple”. This would not be a time to gen­er­ate two words—what’s at stake is ex­actly the idea that you shouldn’t draw a moral dis­tinc­tion.

But once any em­piri­cal propo­si­tion is at stake, or any moral propo­si­tion, you can no longer ap­peal to com­mon us­age.

If the ques­tion is how to cluster to­gether similar things for pur­poses of in­fer­ence, em­piri­cal pre­dic­tions will de­pend on the an­swer; which means that defi­ni­tions can be wrong. A con­flict of pre­dic­tions can­not be set­tled by an opinion poll.

If you want to know whether athe­ism should be clus­tered with su­per­nat­u­ral­ist re­li­gions for pur­poses of some par­tic­u­lar em­piri­cal in­fer­ence, the dic­tio­nary can’t an­swer you.

If you want to know whether blacks are peo­ple, the dic­tio­nary can’t an­swer you.

If ev­ery­one be­lieves that the red light in the sky is Mars the God of War, the dic­tio­nary will define “Mars” as the God of War. If ev­ery­one be­lieves that fire is the re­lease of phlo­gis­ton, the dic­tio­nary will define “fire” as the re­lease of phlo­gis­ton.

There is an art to us­ing words; even when defi­ni­tions are not liter­ally true or false, they are of­ten wiser or more fool­ish. Dic­tionar­ies are mere his­to­ries of past us­age; if you treat them as supreme ar­biters of mean­ing, it binds you to the wis­dom of the past, for­bid­ding you to do bet­ter.

Though do take care to en­sure (if you must de­part from the wis­dom of the past) that peo­ple can figure out what you’re try­ing to swim.