Counterspells

Back in the early days of the in­ter­net, when New Athe­ism was king, peo­ple would col­lect and cat­a­logue var­i­ous log­i­cal fal­la­cies. Then, when get­ting into an ar­gu­ment on some fo­rum, they would iden­tify and call out all the log­i­cal fal­la­cies be­ing used by their op­po­nent; Ad Hominem! Ap­peal to Emo­tion! Tu Quoque! Post hoc, ergo propter hoc!

Th­ese days we tend to know bet­ter. Ap­peal­ing to these lists of log­i­cal fal­la­cies and their Latin names doesn’t help you, and it doesn’t help the con­ver­sa­tion. The prob­lem is that to the per­son you’re talk­ing to, throw­ing around the names of these fal­la­cies just sounds like an Ar­gu­ment from Author­ity—

Wait! I mean, it just sounds like you think you’re smarter or bet­ter ed­u­cated than they are, which isn’t rele­vant to the ar­gu­ment. If you re­ally had some­thing to con­tribute, you would ex­plain the prob­lem with their rea­son­ing, rather than just throw­ing out a Latin term they’ve never heard of. Even if they have heard of it, you should still take the time to tell them why you think it ap­plies to what they said.

Log­i­cal fal­la­cies are all sound com­plaints. But when dis­agree­ing with some­one, they de­serve to hear the ar­gu­ments those com­plaints are based on. More than that, you’re not go­ing to con­vince some­one of your po­si­tion un­less you ac­tu­ally try to con­vince them.

For­tu­nately, log­i­cal fal­la­cies are such bad ar­gu­ments that most of them can be re­but­ted in a few sim­ple sen­tences. And the form of each fal­lacy tends to be so con­sis­tent that these re­but­tals can be highly for­mu­laic.


In ac­cor­dance with the Ra­tion­al­ist tra­di­tion that re­quires ev­ery­thing to have a nerdy sci-fi or fan­tasy name, I am call­ing these for­mu­laic re­but­tals Coun­ter­spells.

Like I men­tioned, I know that by now ev­ery­one knows how to avoid us­ing these fal­la­cies them­selves. Most of us can also rec­og­nize when other peo­ple are us­ing these bad ar­gu­ments against us. Some­how the next step doesn’t come as eas­ily.

You might think that an in­tel­li­gent per­son, rec­og­niz­ing a bad ar­gu­ment they’ve seen a hun­dred times be­fore, would be able to provide a knock-down coun­ter­ar­gu­ment. But for me at least, I still reg­u­larly crash and burn in my at­tempts to ex­tem­po­rize re­but­tals. I see a lot of you all make the same prob­lem.

Some writ­ers have pro­posed some great Coun­ter­spells, but no one has col­lected them. To be­gin with, I’m go­ing to re­view some ex­cel­lent Coun­ter­spells by Paul Gra­ham and Scott Alexan­der. I’ve lifted many of their ex­am­ples di­rectly, as these ex­am­ples were care­fully con­sid­ered and strongly writ­ten; check out the links to see the origi­nal pieces.

Gra­ham’s Counterspells

Paul Gra­ham’s How to Disagree is an old fa­vorite. Part of what I find so ap­peal­ing about this es­say is how Gra­ham not only lays out sev­eral forms of un­pro­duc­tive dis­agree­ment, but in­cludes clear ex­am­ples of the sort of thing one should say in re­sponse.

Without fur­ther ado, Gra­ham’s Coun­ter­spells:

Ad Hominem

Forms:

  • If there’s some­thing wrong with [per­son’s] ar­gu­ment, what is it? If there isn’t, what differ­ence does it make that [per­son] is a [mem­ber of cat­e­gory]?

  • The ques­tion is whether [per­son] is cor­rect or not. If [per­son’s lack of au­thor­ity or other is­sue] caused them to make mis­takes, what are the mis­takes? If there aren’t mis­takes, [per­son’s lack of au­thor­ity or other is­sue] isn’t a prob­lem.

Ex­am­ples:

If there’s some­thing wrong with the sen­a­tor’s ar­gu­ment, you should say what it is; and if there isn’t, what differ­ence does it make that he’s a sen­a­tor?

The ques­tion is whether I’m cor­rect or not. If the fact that I’m just a col­lege stu­dent caused me to make mis­takes, what mis­takes did I make? If there aren’t mis­takes, then my age isn’t a prob­lem.

Re­sponse to Tone

Forms:

  • It mat­ters much more whether [per­son] is wrong or right than what their tone is.

  • I agree about [is­sue with tone], but I care much more about if [per­son] is in­cor­rect some­where. Could you point out where you think the ar­gu­ment is wrong?

Ex­am­ples:

It’s clear that OP is pissed off about this is­sue, but it mat­ters more whether OP is wrong or right than what their tone is. Do you think they’re wrong, or just an ass­hole?

Ok, maybe I was too flip­pant, but I still think my points are cor­rect. If you think I’m ac­tu­ally wrong, could you point out where you dis­agree?

Contradiction

Forms:

  • I think [per­son] is just [stat­ing the op­pos­ing case], with­out ex­plain­ing why [per­son] thinks it’s cor­rect. What ev­i­dence that con­vinced [per­son]?

  • It’s good to know [other per­son’s po­si­tion], but I ac­tu­ally still dis­agree. But I need to un­der­stand what ev­i­dence makes you think [po­si­tion].

Ex­am­ples:

I think you’re just stat­ing that you op­pose gun con­trol, with­out ex­plain­ing why you think it should be op­posed. What is it ex­actly about gun con­trol that makes you so strongly against these poli­cies?

Ok, it’s good to know that you’re con­cerned about ge­net­i­cally en­g­ineered crops speci­fi­cally, but I ac­tu­ally still dis­agree. What ev­i­dence makes you think that these crops aren’t safe?

Alexan­der’s Counterspells

Non­cen­tral Fallacy

One of my fa­vorite pieces by Alexan­der is his de­scrip­tion of his can­di­date for the Worst Ar­gu­ment In The World, what he calls the Non­cen­tral Fal­lacy.

One thing that I like in par­tic­u­lar about this piece is that he clearly ex­plains how one should re­spond to this type of ar­gu­ment. This sets us up nicely for Coun­ter­spells.

Forms:

  • Yes, so what? [Ob­ject of dis­cus­sion] is [mem­ber of a cat­e­gory]. But there are all sorts of other things we know about [ob­ject of dis­cus­sion], and I think they’re also rele­vant to the ques­tion at hand. [Give ex­am­ple of other rele­vant in­for­ma­tion if pos­si­ble.]

  • Nor­mally when we think about [is­sue] we mean some­thing like [typ­i­cal/​cen­tral ex­am­ple of is­sue]. [This case] differs in that [ways in which this case is differ­ent].

  • Ob­vi­ously [what we’re dis­cussing] is an ex­am­ple of [is­sue]. The typ­i­cal ex­am­ple of [is­sue] — [give typ­i­cal ex­am­ple of is­sue] — is pretty [bad/​good/​etc.]. But a lot of the rea­sons [typ­i­cal ex­am­ple of is­sue] — [list of 2-3 ways typ­i­cal ex­am­ple of is­sue is bad/​good/​what­ever] — don’t ap­ply to the wildly atyp­i­cal case of [what we’re dis­cussing].

Ex­am­ples:

The typ­i­cal case of mur­der is Charles Man­son break­ing into a house and shoot­ing some­one. Abor­tion differs in that the vic­tim is an em­bryo or fe­tus with less biolog­i­cal com­plex­ity and in­tel­li­gence than the av­er­age rab­bit.

The typ­i­cal ex­am­ple of theft is some­one mug­ging you in a dark alley and tak­ing your pock­et­book. Tax­a­tion tech­ni­cally qual­ifies as theft if you define the lat­ter as “tak­ing some­one’s money through im­plied threat of force”, but it also differs from dark-alley-mug­ging in sev­eral im­por­tant ways, like that it’s lev­ied by a demo­crat­i­cally elected gov­ern­ment, that it’s sup­posed to be spent on use­ful pro­grams, and that it’s col­lected in an or­derly and pre­dictable fash­ion. Th­ese differ­ences seem to be im­por­tant enough that most peo­ple sup­port tax­a­tion even though they don’t sup­port dark-alley-mug­gings.

Well, ob­vi­ously. That’s kind of the point. And the typ­i­cal ex­am­ple of racial dis­crim­i­na­tion—the Ku Klux Klan burn­ing your house down or some­thing—is pretty bad. But a lot of the rea­sons KKK-house-burn­ing is bad—liv­ing in fear, lock­ing down­trod­den groups into a cy­cle of poverty, to­tally lock­ing qual­ified peo­ple out of any job—don’t ap­ply to the wildly atyp­i­cal case of af­fir­ma­tive ac­tion.


Alexan­der has an es­say called Va­ri­eties of Ar­gu­men­ta­tive Ex­pe­rience, writ­ten as some­thing of a re­sponse to Gra­ham’s es­say. This is ex­cel­lent for my pur­poses, since it means that it has a very similar form. Here are some Coun­ter­spells drawn from this source.

Gotchas

Gotchas don’t share as much struc­ture as other things on this list, so the form of their Coun­ter­spells are not as re­li­able as some oth­ers. I think that Alexan­der would also tell me that Gotchas aren’t worth en­gag­ing with, and maybe he’s right. But I hold out hope for peo­ple who are still play­ing gotcha, and a calm re­sponse can help to re­fo­cus the con­ver­sa­tion on the is­sues, so here is an at­tempt.

You’ll no­tice that my ap­proach here is to dead­pan ac­cept the fic­tion that they are mak­ing an hon­est con­tri­bu­tion. Other ap­proaches might also be pos­si­ble.

Forms:

  • Well, [things] can vary along di­men­sions other than [thing be­ing pointed out].

  • Yeah, but I still sup­port [thing] be­cause it has [these other con­se­quences].

Ex­am­ples:

Well, I don’t want to move to Cuba, but that’s be­cause gov­ern­ments can vary along di­men­sions other than how big they are, and even though Cuba has a big gov­ern­ment there are a lot of things about it that I don’t like.

Yeah, but I still sup­port se­ri­ous gun con­trol mea­sures be­cause I think there are big­ger is­sues than crim­i­nals hav­ing guns, like suicide rates and mur­ders by pri­vate cit­i­zens.

Sin­gle Facts

Forms:

  • [Fact] ad­mit­tedly does sup­port your ar­gu­ment, but things with some bad fea­tures can be good over­all. [Rele­vant ex­am­ple, if pos­si­ble.]

  • [Fact] is a good point, but the is­sue is larger than that, and I think [fact] is out­weighed by [facts that out­weigh it].

  • Even if we all agree that [fact] is true, you’re only point­ing out one [bad/​good] qual­ity of [thing]. I don’t think it’s such a big deal. Why is [fact] re­ally im­por­tant?

Ex­am­ples:

Trump ly­ing about his grades ad­mit­tedly does sup­port your ar­gu­ment that he’s some­times dishon­est, but things with some bad fea­tures can be good over­all. Trump can be dishon­est as a busi­ness­man some­times, but he has other good qual­ities.

Ob­vi­ously for­eign in­ter­ven­tion does cost Amer­i­can lives, and that’s some­thing I worry about too. But the is­sue is larger than that, and I think we should be pre­pared to make these sac­ri­fices some­times — things like geno­cide and re­li­gious dic­ta­tor­ships are much worse.

Even if we all agree that Hillary is crap at email se­cu­rity, you’re only point­ing out one bad qual­ity of hers. I don’t think it’s such a big deal. Why is her email se­cu­rity cred so im­por­tant?

Sin­gle Studies

There are a lot of ways in which shar­ing a sin­gle study can be a non­pro­duc­tive con­tri­bu­tion to a dis­cus­sion, so here are a few va­ri­eties of Coun­ter­spell.

Forms:

  • On any con­tro­ver­sial is­sue, there are usu­ally many peer-re­viewed stud­ies sup­port­ing each side. In just [how­ever much time you spent search­ing] I was able to find [X num­ber] ar­ti­cles with the ex­act op­po­site find­ing. If we want to an­swer this ques­tion, we need to take a look at the liter­a­ture as a whole.

  • There are a lot of ways in which a study can be wrong, and it’s not always ob­vi­ous when they are. Some­times hon­est re­searchers make a mis­take, or just get un­lucky. Is there any other ev­i­dence out there for [thing] be­sides this study?

  • I think this study in­ves­ti­gates [a much weaker sub­prob­lem; be as spe­cific as pos­si­ble] rather than [the larger prob­lem].

Ex­am­ples:

On any con­tro­ver­sial is­sue, there are usu­ally many peer-re­viewed stud­ies sup­port­ing each side. In just 15 min­utes, I was able to find 7 ar­ti­cles (links here) that con­clude that rais­ing the min­i­mum wage de­creases un­em­ploy­ment. If we want to an­swer this ques­tion, we need to take a look at the liter­a­ture as a whole.

I think this study in­ves­ti­gates whether trig­ger warn­ings af­fect cer­tain be­liefs about trauma in the very short term rather than whether trig­ger warn­ings are helpful or harm­ful in col­lege courses, which is the thing most peo­ple ac­tu­ally care about.

Iso­lated De­mand for Rigor

This ac­tu­ally can be coun­tered in a va­ri­ety of in­ter­est­ing ways; see this es­say for a more com­plete treat­ment. Most ex­am­ples lifted from there.

Forms:

  • This is wrong be­cause [clear coun­terex­am­ple to the de­mand for rigor]; [pro­posed rule] is a fake rule we never ap­ply to any­thing else.

  • Pre­sum­ably you think [thing that doesn’t pass the de­mand for rigor]. So why does [is­sue] have to jus­tify it­self ac­cord­ing to rigor­ous crite­ria that noth­ing could pos­si­bly pass?

  • So you think that [rule stated ex­plic­itly]? Wouldn’t that im­ply [weird con­se­quence they prob­a­bly don’t sup­port]?

  • You seem to be happy to [do thing that breaks the de­manded rigor], but you switch to a po­si­tion that [other thing isn’t ok be­cause of de­mand for rigor]. Why the differ­ence in opinion?

Ex­am­ples:

Repub­li­cans have also been against lead­ers who presided over good economies and pre­sum­ably thought this was a rea­son­able thing to do; “it’s im­pos­si­ble to hon­estly op­pose some­one even when there’s a good econ­omy” is a fake rule we never ap­ply to any­thing else.

Pre­sum­ably there are lots of gov­ern­ment pro­grams you sup­port (maybe PBS?), but you would never dream of de­mand­ing that we de­fund them in the hopes of donat­ing the money to malaria pre­ven­tion. But since you don’t sup­port air strikes, sud­denly that plan has to jus­tify it­self ac­cord­ing to rigor­ous crite­ria that no gov­ern­ment pro­gram that ex­ists could pos­si­bly pass.

So you think that no one should ever be forced to pay for some­thing they don’t like?Wouldn’t that im­ply liber­als shouldn’t have to pay for wars? That seems like a weird con­se­quence you prob­a­bly don’t sup­port.

You’re prob­a­bly happy to talk about speed and blood pres­sure and co­mas and the crime rate, but some­how you switch to a po­si­tion that we can’t talk about IQ at all un­less we have a perfect fac­tor-an­a­lyt­i­cal proof of its obey­ing cer­tain statis­ti­cal rules. Th­ese seem equally rigor­ous, so why are you ok with one and not the other?

Disput­ing Definitions

Alexan­der cor­rectly notes that this is one of the hard­est is­sues to re­but; both be­cause it’s easy to be se­duced by this sort of ar­gu­ment, and be­cause think­ing crit­i­cally about words and defi­ni­tions is a very difficult skill to learn, a point not read­ily com­mu­ni­cated in a few sen­tences.

Even so, here are my own at­tempts at a Coun­ter­spell for this is­sue:

Forms:

  • I’m not sure it mat­ters whether or not [thing] is a mem­ber of [cat­e­gory]. Peo­ple might even dis­agree with the defi­ni­tion of [cat­e­gory]. [Give ex­am­ple of defi­ni­tion drift if pos­si­ble.] I think we both care a lot more about things like [ac­tual is­sues with ma­te­rial con­se­quences].

  • You can define [cat­e­gory] as [your defi­ni­tion] if you want, but that’s not usu­ally what peo­ple mean by [cat­e­gory]. I don’t think fac­tual or moral ques­tions de­pend on how we use words.

  • You can define [cat­e­gory] as [your defi­ni­tion] if you want, but I’m much more con­cerned with [prac­ti­cal is­sue un­re­lated to the defi­ni­tion of a word].

(Note that this has a lot of over­lap with the Non­cen­tral Fal­lacy.)

Ex­am­ples:

I’m not sure it mat­ters whether or not be­ing trans­gen­der is a men­tal ill­ness. Peo­ple might even dis­agree how to define what is and isn’t a men­tal ill­ness. Some peo­ple still don’t think de­pres­sion would count! I think we both care a lot more about things like whether or not trans­gen­der peo­ple should be al­lowed to change their le­gal gen­der, and whether they face dis­crim­i­na­tion.

You can define com­mu­nism so that the USSR doesn’t count if you want, but that’s not usu­ally what peo­ple mean by com­mu­nism. I think the fac­tual and moral is­sues stand, re­gard­less of what words you use.

You can define war as “a for­mal state of armed con­flict be­tween state gov­ern­ments” if you want, but I’m much more con­cerned with the fact that peo­ple are dy­ing in this con­flict, what­ever we call it.

Why can’t the chef call a tomato a veg­etable? He just wants to think about whether or not to put it in a salad or on a burger. What’s wrong with that?

Other Counterspells

And here are a few of my own de­sign, for some com­mon fal­la­cies:

Ar­gu­ment from Authority

Forms:

  • Just be­cause [source] is [bet­ter ed­u­cated/​more fa­mous/​more ex­pert/​smarter] than me, isn’t rele­vant to [the is­sue]. Some­times ex­perts are wrong and make mis­takes. What ar­gu­ments does [source] make that con­vince you of that po­si­tion?

Ex­am­ples:

Pinker be­ing a Har­vard pro­fes­sor and gen­er­ally a smart guy isn’t rele­vant to his po­si­tion on AI threat. Some­times ex­perts are wrong and make mis­takes. What part of Pinker’s ar­gu­ments do you think re­fute my po­si­tion?

Straw Man

Forms:

  • Well I can’t speak for ev­ery­one else, but [straw man] isn’t what I be­lieve about this case. Are you sure you un­der­stand what my po­si­tion is? [If char­i­ta­ble, at­tempt to clar­ify your po­si­tion.]

  • (If you are very con­fi­dent) I don’t think any­one re­ally be­lieves [straw man]. Per­son­ally I be­lieve [state your po­si­tion]. Do you dis­agree with that?

Ex­am­ples:

Well I can’t speak for ev­ery­one else, but I don’t ac­tu­ally want to ban all pri­vate own­er­ship of guns. Are you sure you un­der­stand what my po­si­tion is? All I’m sug­gest­ing is that we en­force the laws we have, and close some loop­holes re­lated to pur­chas­ing firearms at gun shows.

Tu Quoque

Forms:

  • It’s pos­si­ble be against some­thing and still some­times par­ti­ci­pate in it. I may be a hyp­ocrite about [what­ever], but that doesn’t make me wrong.

Ex­am­ples:

It’s pos­si­ble be against some­thing and still some­times par­ti­ci­pate in it. Sure, it’s hyp­o­crit­i­cal for me to eat meat, but that doesn’t make me wrong about an­i­mal rights.

Sure, it’s hyp­o­crit­i­cal for me to smoke and tell you not to, but I’m still right that it would be bad for your health, and I can still want you not to start!

Ap­peal to Popularity

Forms:

  • The ma­jor­ity has been wrong all the time. [Pick one of the many ex­am­ples, one that your au­di­ence will agree with.] I don’t care who [agrees/​dis­agrees with premise], I want to know whether [premise is cor­rect/​in­cor­rect].

Ex­am­ples:

The ma­jor­ity has been wrong all the time. Peo­ple used to think that the sun goes around the earth! I don’t care how many sci­en­tists be­lieve in global warm­ing, I want to know whether the ev­i­dence is strong enough to think it’s a real threat.

I don’t care if ev­ery­one on that sub­red­dit thinks it’s right, I want to know why YOU think it’s right.

Come on, if all your friends jumped off a bridge, would you jump too?


There are, of course, many other things that can go wrong in the course of a con­ver­sa­tion. There will be other con­cise, effec­tive re­but­tals, which can be added to this list.