Category Qualifications (w/​ exercises)

This is the sec­ond post in the Ar­gu­ing Well se­quence. This post is in­fluenced by A Hu­man’s Guide to Words and Against Lie In­fla­tion.


In the last post, we dis­cussed a com­mon prob­lem in ar­gu­ments that Prove Too Much. In this post, we’ll gen­er­al­ize that prob­lem to help de­ter­mine use­ful cat­e­gories. But be­fore we go on, what’s wrong with these ar­gu­ments?

Ex. 1 [Stolen from slat­estar­codex]

“A few months ago, a friend con­fessed that she had abused her boyfriend. I was shocked, be­cause this friend is one of the kind­est and gen­tlest peo­ple I know. I probed for de­tails. She told me that some­times she needed her boyfriend to do some fa­vor for her, and he wouldn’t, so she would cry – not as an at­tempt to ma­nipu­late him, just be­cause she was sad. She counted this as abuse, be­cause her defi­ni­tion of “abuse” is “some­thing that makes your part­ner feel bad about set­ting bound­aries”. And when she cried, that made her boyfriend feel guilty about his bound­ary that he wasn’t go­ing to do the fa­vor.”

By this defi­ni­tion of “abuse”, a ma­jor­ity of peo­ple are “abu­sive”. It would be bet­ter to re­serve that word for a smaller group of peo­ple who are in­ten­tion­ally ma­nipu­lat­ing peo­ple.

Ex. 2: I also had a friend that was “one of the nicest peo­ple ev­ery­one knows, wow she listens so well!”. She ad­mit­ted that she was ac­tu­ally “self­ish and ma­nipu­la­tive” be­cause she did nice things to peo­ple so they’d like her.

I hon­estly wish ev­ery­one was as “self­ish and ma­nipu­la­tive” as this girl; how­ever, it makes those two words nearly use­less. It would be bet­ter to re­serve those words for peo­ple who cre­ate win-lose situ­a­tions (You give, I take) as op­posed to win-win situ­a­tions (Oh wow, you make me feel im­por­tant. I want to be your friend).

Gen­eral Frame

What is the gen­eral frame of the prob­lem in the two sce­nar­ios? You have 2 min­utes.

The claim re­lies on a “bad defi­ni­tion” of a word. It’s “bad” be­cause my ex­pec­ta­tions weren’t met. If you say you’re self­ish, then I ex­pect that you cre­ate win-lose situ­a­tions, but if all you’re ac­tu­ally a re­ally nice per­son, then you mis­led me.

In an­other way, I can say if you meet the qual­ifi­ca­tions a,b,c, then you are a mem­ber of that cat­e­gory. The prob­lem arises when you have the “wrong” qual­ifi­ca­tions, as in, the per­son you were talk­ing to was ex­pect­ing differ­ent qual­ifi­ca­tions.

X meets qual­ifi­ca­tions a,b,c for [word]: → X is [word]

Now that you have a new frame to fit ev­ery­thing in, let’s dive in to a cou­ple curve­balls.

Ex. 3: If God was all-pow­er­ful, could he make a rock so big that he couldn’t lift it?

This is about the qual­ifi­ca­tions of the word “all-pow­er­ful”, and it’s im­ply­ing that one of those qual­ifi­ca­tions is “can cre­ate a situ­a­tion that un­qual­ifies it­self as all-pow­er­ful”. You could define all-pow­er­ful that way; how­ever, I (and most other peo­ple) are ex­pect­ing a defi­ni­tion that means “has a lot of power/​abil­ities” like mir­a­cles, time travel, etc and not some­thing para­dox­i­cal.

Ex. 4: Is a hot­dog a sand­wich?

Con­sid­er­ing just pre­serv­ing ex­pec­ta­tions, if some­one asked me to make them a sand­wich, and I went to hand them ei­ther (A) Ham sand­wich, (B) To­mato w/​ mayo sand­wich, or (C ) Hot­dog, which one would most sur­prise them?

Ex. 5: If a tree falls in the woods, and no one is around to hear it, would it make a sound?

What qual­ifies as a sound? If we agree that it’s vibra­tional waves be­tween 20Hz-20kHz, then it made a sound. If we agree that there has to be some­one to hear it, then it didn’t make a sound. Since the pur­pose is com­mu­ni­ca­tion/​ ex­pec­ta­tion preser­va­tion, we can just agree on a set of qual­ifi­ca­tions, solve the philo­soph­i­cal prob­lem, and move on.

Ideal Algorithm

What al­gorithm were you run­ning? What’s an ideal al­gorithm for cor­rect­ing these types of ar­gu­ments? You have 3 min­utes (The pre­vi­ous ex­am­ples should fit in your al­gorithm)

1. What is the key word?

2. What are their qual­ifi­ca­tions for that word?

3. What are the de­sired qual­ifi­ca­tions given the con­text?

4. If (2) and (3) dis­agree, then ar­gue about which set of qual­ifi­ca­tions provide clearer com­mu­ni­ca­tion given the con­text.

Run­ning this al­gorithm on the pre­vi­ous ex­am­ples is easy, ex­cept for the “tree fal­ling in the woods”. The key word is “sound”, but there is no “(3) de­sired qual­ifi­ca­tion” for sound in this case. If there is no agreed qual­ifi­ca­tion, then what’s needed is to agree on a qual­ifi­ca­tion. Updating

  1. What is the key word?

  2. What are their qual­ifi­ca­tions for that word?

  3. What are the de­sired qual­ifi­ca­tions given the con­text?

  4. If 3 doesn’t ex­ist, then agree on a set of qual­ifi­ca­tions.

  5. Else if (2) and (3) dis­agree, then ar­gue about which set of qual­ifi­ca­tions provide clearer com­mu­ni­ca­tion given the con­text.

What are the differ­ences/​similar­i­ties/​re­la­tion­ships be­tween this and Prov­ing Too Much? You have 3 minutes

My fram­ing of Prov­ing Too Much is a sub­set of this (I did kind of gave that away in the in­tro). It’s about the cat­e­gory of 100% Truth/​ac­cu­rate pre­dic­tions/​ map-ter­ri­tory cor­re­spon­dances in the con­text of want­ing to find ac­tu­ally true things that re­flect re­al­ity. I ex­pect a qual­ifi­ca­tion to lead to only true claims; how­ever, if it im­plies false or in­con­sis­tent claims, then my ex­pec­ta­tions are vi­o­lated and that qual­ifi­ca­tion is wrong.

Hard ex­er­cise: The cat­e­gory of 100% true rea­sons to be­lieve things should have no mem­bers. How would you con­struct a cat­e­gory for 0-100% be­liefs that is ac­tu­ally more use­ful?

With this al­gorithm down, let’s tackle a few more prob­lems.

Fi­nal Prob­lem Set

Ex. 6: That salad with cu­cum­bers on it should be called a fruit salad, be­cause cu­cum­bers are botan­i­cally a fruit

The key word is “fruit salad”. Most peo­ple who or­der a fruit salad at a restau­rant and get cu­cum­bers as their “fruit” would not be happy be­cause their ex­pec­ta­tions were vi­o­lated. As the say­ing goes, the cus­tomer is always right.

Ex. 7: Chris­ti­an­ity is true to me like Is­lam is true to some­one else.

If I told some­one that I made them soup and put a bowl of ce­real in front of them, then they might laugh or be dis­ap­pointed. Both im­plies a vi­o­la­tion of ex­pec­ta­tion.

Oh, I don’t mean true in that way, but I am talk­ing about some­thing close, let’s call it “re­flects re­al­ity” in­stead of truth. Like if Chris­ti­an­ity re­flects re­al­ity, then if I built a time ma­chine, I should be able to go back in time, see Je­sus die, be buried, and rise again. Then the bible re­flects re­al­ity. If Is­lam re­flects re­al­ity, then I should be able to go back in time and see Je­sus as­cend to heaven, but not die or re­s­ur­rect. Then the Qu­ran re­flects re­al­ity. So the Bible and the Qu­ran can’t both re­flect re­al­ity since they’re pre­dict­ing I would see two differ­ent things. Does what I’m try­ing to say make sense?

Ex. 8:

“Let’s run fur­ther than we ran yes­ter­day”
“You mean *farther”

Us­ing the more gram­mat­i­cally cor­rect word doesn’t change our ex­pec­ta­tions. In ei­ther case, I’d ex­pect to run a greater dis­tance than yes­ter­day.

This is in­ter­est­ing be­cause it gen­er­al­izes to all gram­mar cor­rec­tions that don’t change ex­pec­ta­tions when the con­text is com­mu­ni­ca­tion. If the con­text is in­stead sig­nal­ling com­pe­tence (like a re­sume), then it would be im­por­tant.

Ex. 9: Is ce­real soup?

This one is in­ter­est­ing be­cause it’s say­ing the qual­ifi­ca­tions for the word “true” are differ­ent than in Prov­ing Too Much. In­stead of be­ing abra­sive and claiming the word “true” as mine and can only mean one thing, I can in­stead just Ta­boo the word and re­place it with its mean­ing.

Ex. 10: Is wa­ter wet?

If some­one told me the pool wa­ter is wet, I’d think they were say­ing some­thing triv­ially true to be silly. If they told me, in all se­ri­ous­ness, the wa­ter isn’t wet, then I might think the wa­ter is fake/​an illu­sion or that gen­er­ally some­thing is wrong with the wa­ter. So in the con­text of two nor­mal peo­ple com­mu­ni­cat­ing, wa­ter is ex­pected to be de­scribed as wet. (If the con­text is chem­istry re­search pa­pers, there may be a differ­ent an­swer)


One of the pur­poses of ar­gu­ing well is clear com­mu­ni­ca­tion. When talk­ing to some­one else (or your­self!), know­ing what key words means to each per­son aids in un­der­stand­ing each other and helps in avoid­ing con­fu­sion.

In the next post, we’ll be dis­cussing false dilem­mas, how they arise, and how to deal with them. What’s wrong with the clas­sic ex­am­ple:

“You’re ei­ther with me, or against me!”

[Feel free to com­ment your an­swer to the hard ques­tion and if you got differ­ent an­swers/​ gen­er­al­iza­tions/​ al­gorithms than I did. Same if you feel like you hit on some­thing in­ter­est­ing or that there’s a con­cept I missed. Ad­ding your own ex­am­ples with the Spoiler tag >! is en­couraged]