The Power to Demolish Bad Arguments

This is Part I of the Speci­fic­ity Sequence

Speci­fic­ity turns any ar­gu­ment into a game of 3D Chess. Just when it seems like your ar­gu­ment is a clash of two ground armies, you can use your speci­fic­ity pow­ers to take off and fly all over the con­cep­tual land­scape. Fly, I say!

“Uber ex­ploits its drivers!”

Want to see what a 3D Chess ar­gu­ment looks like? Be­hold the con­ver­sa­tion I had the other day with my friend “Steve”:

Steve: Uber ex­ploits its drivers by pay­ing them too lit­tle!

Steve’s state­ment was a generic one, lack­ing spe­cific de­tail. So I shot back with my own generic coun­ter­point:

Liron: No, job cre­ation is a force for good at any wage. Uber cre­ates in­creased de­mand for la­bor, which drives wages up in the econ­omy as a whole.

You can see I was show­ing off my mas­tery of ba­sic eco­nomics. This seemed like a good move to me at the time, but I should have pri­ori­tized which of my skills to bust out. The skill of speci­fic­ity is so badass that it even takes prece­dence over the mighty Econ 101.

When I used my Econ 101 rook to at­tack Steve’s defenses and put his king in check, I was merely play­ing 2D chess. The 3D chess move would have been to dial up the speci­fic­ity.

What hap­pens if I ask Steve to zoom into the sub­stance of his claim? Then the con­ver­sa­tion goes like this:

Steve: Uber ex­ploits its drivers by pay­ing them too lit­tle!
Liron: What do you mean by “ex­ploits its drivers”?
Steve: Come on, you know what “ex­ploit” means… Dic­tionary.com says it means “to use self­ishly for one’s own ends”
Liron: You’re say­ing you have a beef with any com­pany that acts “self­ish”? Doesn’t ev­ery com­pany un­der cap­i­tal­ism aim to max­i­mize re­turns for its share­hold­ers?
Steve: Cap­i­tal­ism can be good some­times, but Uber has gone be­yond the pale with their ex­ploita­tion of work­ers. They’re ba­si­cally ru­in­ing cap­i­tal­ism.

Nooooo, this is not the en­light­en­ing con­ver­sa­tion we were hop­ing for. You can sense that I haven’t made much progress “pin­ning him down”.

But don’t worry… that wasn’t my real demon­stra­tion of 3D Chess. Psy­che!

In the above con­ver­sa­tion, I didn’t em­ploy my speci­fic­ity pow­ers, I just show­cased an­other failure mode. Can you figure out where I went wrong?

It was a mis­take for me to ask Steve for a mere defi­ni­tion of the term “ex­ploit”. I should have asked for a spe­cific ex­am­ple of what he imag­ines “ex­ploit” to mean. How speci­fi­cally does Uber ex­ploit its drivers?

In­stead, I ac­cepted his non-spe­cific re­ply — that “ex­ploit” means “to use self­ishly” — and I tried to coun­ter­ar­gue by toss­ing the big ab­stract con­cept of “cap­i­tal­ism” into the dis­cus­sion. From Steve’s per­spec­tive, “cap­i­tal­ism” is yet one more generic word for him to build his generic re­sponses out of, to­gether with “ex­ploita­tion” and “self­ish­ness”. He loves fling­ing con­cept-words around; it makes him feel like he’s hav­ing a lively in­tel­lec­tual back & forth.

Steve’s mind doesn’t ac­tu­ally con­tain a struc­tured un­der­stand­ing of the sub­ject he’s mak­ing a claim about; it con­tains a ball pit of loosely-as­so­ci­ated con­cepts. He holds up his end of the con­ver­sa­tion by snatch­ing a nearby ball and fling­ing it. And what have I done by men­tion­ing “cap­i­tal­ism”? I’ve gone and tossed in an­other ball.

I hate to ad­mit it, but Steve’s ap­proach in the above di­alogue works for him. By slosh­ing around his men­tal ball pit and fling­ing smart-sound­ing as­ser­tions about “cap­i­tal­ism” and “ex­ploita­tion”, he just might win over a neu­tral au­di­ence of our peers. What a night­mare for us.

Is there a way to reach in and pull him out of this pit? Yes, by ac­ti­vat­ing our speci­fic­ity pow­ers! Here’s how it’s done:

Steve: Uber ex­ploits its drivers by pay­ing them too lit­tle!
Liron: Can you help me paint a spe­cific men­tal pic­ture of a driver be­ing ex­ploited by Uber?
Steve: Ok… A sin­gle dad whose kid gets sick. He works for Uber and he doesn’t even get health in­surance, and he’s maxed out all his credit cards to pay for doc­tor’s vis­its. The next time his car breaks down, he won’t even be able to fix it. Mean­while, Uber skims 25% of ev­ery dol­lar so he barely makes min­i­mum wage. You should try liv­ing on min­i­mum wage so you can see how hard it is!
Liron: You’re say­ing Uber should be blamed for this per­son’s un­pleas­ant life cir­cum­stances, right?
Steve: Yes, be­cause they have mil­lions of drivers un­der these kinds of cir­cum­stances, and mean­while they IPO for $80B.

He doesn’t re­al­ize it yet, but by mak­ing him flesh out a spe­cific ex­am­ple of his claim, I’ve now pul­led him out of his ball pit of loosely-as­so­ci­ated con­cepts. This isn’t your av­er­age 2D ar­gu­ment any­more. We’re now fly­ing like Su­per­man.

Liron: Ok, stick­ing with this one spe­cific per­son’s hy­po­thet­i­cal story — what would they be do­ing if Uber didn’t ex­ist?
Steve: Get­ting a differ­ent job
Liron: Ok, what spe­cific job?
Steve: I don’t know, de­pends what their skills are
Liron: This is your spe­cific story Steve, you get to pick any spe­cific plau­si­ble de­tails you want in or­der to sup­port any point you want!

I have to stop and point out how crazy this is.

You’d think the way smart peo­ple ar­gue is by sup­port­ing their claims with ev­i­dence, right? But here I’m giv­ing Steve a hand­i­cap where he gets to make up fake ev­i­dence (tel­ling me any hy­po­thet­i­cal spe­cific story) just to es­tab­lish that his ar­gu­ment is co­her­ent by check­ing whether em­piri­cal sup­port for it ever could mean­ingfully ex­ist. This is a preschool-level stan­dard that your av­er­age ar­guer can’t pass.

Let’s see how he might re­ply:

Steve: I guess he could in­stead be a cashier at McDon­ald’s. Be­cause then he’d be a W2 em­ployee and get med­i­cal in­surance.
Liron: In a world where Uber ex­ists, couldn’t this spe­cific guy still go get a job as a cashier at McDon­ald’s? Plus, wouldn’t he have less com­pe­ti­tion for that cashier job be­cause some of the other would-be ap­pli­cants got re­cruited to be Uber drivers in­stead? Can we con­clude that the spe­cific per­son who you chose to illus­trate your point is ac­tu­ally be­ing *helped* by the ex­is­tence of Uber?
Steve: No be­cause he’s an Uber driver, not a McDon­ald’s cashier
Liron: So doesn’t that mean Uber offered him a bet­ter deal than McDon­ald’s, thereby im­prov­ing his life?
Steve: No, they just tricked him into think­ing that it’s a bet­ter deal, but it’s ac­tu­ally a worse deal for him.
Liron: So like, McDon­ald’s offered him $13/​hr plus benefits, while Uber gave him an es­ti­mate of mak­ing $20/​hr but it ac­tu­ally works out to $14/​hr once you fac­tor in all his costs like gas and de­pre­ci­a­tion, but Uber gave him no benefits, so his over­all com­pen­sa­tion value is less than mak­ing $13/​hr plus benefits at McDon­ald’s?
Steve: Um, ya, some­thing like that.
Liron: So if Uber did a bet­ter job of ed­u­cat­ing drivers about how much their com­pen­sa­tion plan is re­ally worth, would you stop say­ing that Uber is “ex­ploit­ing its drivers by pay­ing them too lit­tle”?
Steve: No, be­cause Uber preys on drivers who need quick ac­cess to cash, and they also in­tend to au­to­mate away the drivers’ jobs as soon as they can.
Liron: It looks like you’re now mak­ing new claims that weren’t rep­re­sented in the spe­cific story you chose, right?
Steve: Yes, but I can tell other stories
Liron: But for the spe­cific story you chose to tell that was sup­posed to best illus­trate your claim, the “ex­ploita­tion” you’re refer­ring to only “robbed” the driver of the value of a McDon­ald’s cashier’s health in­surance plan, which might be like a $1/​hr loss? And his work sched­ule is so much more flex­ible as an Uber driver… couldn’t that eas­ily be worth $1/​hr to him, so that he wasn’t “tricked” into join­ing Uber but rather made a de­ci­sion in ra­tio­nal self-in­ter­est?
Steve: Yeah maybe, but any­way that’s just one story.
Liron: No wor­ries, we can start over and talk about a spe­cific story that you think would illus­trate your main claim. I’m listen­ing…
Steve thinks for a lit­tle while...
Steve: I don’t know all the ex­ploita­tive shit Uber does ok? I just think Uber is a greedy com­pany.

In com­plex top­ics such as poli­tics and eco­nomics, most peo­ple who think they’re mak­ing an “ar­gu­ment” are merely mak­ing an in­co­her­ent state­ment. They’re con­fused about their own claim.

Be­fore you think about win­ning the ar­gu­ment, just start by drilling down into whether their point is co­her­ent. You’ll then find that you’re of­ten done ar­gu­ing be­fore you even re­ally start.

In the above con­ver­sa­tion, I hadn’t got­ten to a point where I was try­ing to re­fute Steve’s ar­gu­ment, I was just try­ing to get spe­cific clar­ity on what Steve’s ar­gu­ment is.

As I tried to nail down his point, his point sim­ply col­lapsed down to noth­ing. He didn’t have a sin­gle spe­cific ex­am­ple of what spe­cific world-state could pos­si­bly be a refer­ent of the state­ment “Uber ex­ploits its drivers”.

Zoom­ing Into the Claim

Imag­ine Steve shows you this map and says, “Ore­gon’s coastline is too straight. I wish all coastlines were less straight so that they could all have a bay!”

Re­sist the temp­ta­tion to ar­gue back, “You’re wrong, bays are stupid!” Hope­fully, you’ve built up the habit of nailing down a claim’s spe­cific mean­ing be­fore try­ing to ar­gue against it.

Steve is mak­ing a claim about “Ore­gon’s coastline”, which is a pretty ab­stract con­cept. In or­der to un­pack the claim’s spe­cific mean­ing, we have to zoom into the con­cept of a “coastline” and see it in more de­tail as this spe­cific con­figu­ra­tion of land and wa­ter:

From this per­spec­tive, a good first re­ply would be, “Well, Steve, what about Coos Bay over here? Are you happy with Ore­gon’s coastline as long as Coos Bay is part of it, or do you still think it’s too straight even though it has this bay?”

No­tice that we can’t pre­dict how Steve will an­swer our spe­cific clar­ify­ing ques­tion. So we never knew what Steve’s words meant in the first place, did we? Now you can see why it wasn’t yet pro­duc­tive for us to start ar­gu­ing against him.

When you hear a claim that sounds mean­ingful, but isn’t 100% con­crete and spe­cific, the first thing you want to do is zoom into its speci­fics. In many cases, you’ll then find your­self dis­am­biguat­ing be­tween mul­ti­ple valid spe­cific in­ter­pre­ta­tions, like for Steve’s claim that “Ore­gon’s coastline is too straight”.

In other cases, you’ll dis­cover that there was no spe­cific mean­ing in the mind of the speaker, like in the case of Steve’s claim that “Uber ex­ploits its drivers by pay­ing them too lit­tle” — a stag­ger­ing thing to dis­cover.

TFW a state­ment un­ex­pect­edly turns out to have no spe­cific meaning

“Star­tups should have more im­pact!”

Ac­ti­vat­ing our speci­fic­ity pow­ers let us de­mol­ish Steve’s claim that “Uber ex­ploits its drivers” by show­ing that Steve had no spe­cific mean­ing in mind for it. But Steve is just your av­er­age above-av­er­age IQ guy who went to col­lege and smoked weed if he didn’t have a test com­ing up. Let’s kick it up a notch.

Con­sider this ex­cerpt from a re­cent se­ries of tweets by Michael Seibel, CEO of the Y Com­bi­na­tor startup ac­cel­er­a­tor pro­gram:

Suc­cess­ful tech founders would have far bet­ter lives and lega­cies if they com­peted for hap­piness and im­pact in­stead of wealth and users/​rev­enue.
We need to change [the] model from build a big com­pany, get rich, and then start­ing a foun­da­tion...
To build a big com­pany, get rich, and use the com­pany’s reach and power to make the world a bet­ter place.

When I first read these tweets, my im­pres­sion was that Michael was pro­vid­ing use­ful sug­ges­tions that any founder could act on to make their startup more of a force for good. But then I ac­ti­vated my speci­fic­ity pow­ers…

Be­fore elab­o­rat­ing on what I think is the failure of speci­fic­ity on Michael’s part, I want to say that I re­ally ap­pre­ci­ate Michael and Y Com­bi­na­tor en­gag­ing with this topic in the first place. It would be easy for them to keep their head down and stick to their origi­nal wheelhouse of fund­ing suc­cess­ful star­tups and mak­ing huge fi­nan­cial re­turns, but in­stead, YC re­peat­edly pushes the en­velope into new ar­eas such as found­ing OpenAI and cre­at­ing their Re­quest for Car­bon Re­moval Tech­nolo­gies. The Y Com­bi­na­tor com­mu­nity is an amaz­ing group of smart and morally good peo­ple, and I’m proud to call my­self a YC founder (my com­pany Re­la­tion­ship Hero was in the YC Sum­mer 2017 batch). Michael’s heart is in the right place to sug­gest that startup founders may have cer­tain un­der­used mechanisms by which to make the world a bet­ter place.

That said… is there any co­her­ent take­away from this se­ries of tweets, or not?

The key phrases seem to be that startup founders should “com­pete for hap­piness and im­pact” and “use the com­pany’s reach and power to make the world a bet­ter place”.

It sounds mean­ingful, doesn’t it? But no­tice that it’s gener­i­cally-worded and lacks any spe­cific ex­am­ples. This is a red flag.

Re­mem­ber when you first heard Steve’s claim that “Uber ex­ploits its drivers by pay­ing them too lit­tle”? At first, it sounded like a mean­ingful claim. But as we tried to nail down what it meant, it col­lapsed into noth­ing. Will the same thing hap­pen here?

Speci­fic­ity pow­ers, ac­ti­vate! Form of: Tweet reply

What’s a spe­cific ex­am­ple, real or hy­po­thet­i­cal, of a $1B+ founder trad­ing off less rev­enue for more im­pact?
Cuz at the $1B+ level, com­pet­ing for im­pact may look in­dis­t­in­guish­able from com­pet­ing for rev­enue.
E.g. Elon Musk com­pa­nies have huge im­pact and huge val­u­a­tions.

Let’s con­sider a spe­cific ex­am­ple of a startup founder who is highly suc­cess­ful: Elon Musk and his com­pany SpaceX, cur­rently val­ued at $33B. The com­pany’s mis­sion state­ment is proudly dis­played at the top of their about page:

SpaceX de­signs, man­u­fac­tures and launches ad­vanced rock­ets and space­craft. The com­pany was founded in 2002 to rev­olu­tionize space tech­nol­ogy, with the ul­ti­mate goal of en­abling peo­ple to live on other planets.

What I love about SpaceX is that ev­ery­thing they do fol­lows from Elon Musk’s origi­nal goal of mak­ing hu­man life mul­ti­plane­tary. Check out this in­cred­ible post by Tim Ur­ban to un­der­stand Elon’s plan in de­tail. Elon’s 20-year play­book is breath­tak­ing:

  1. Iden­tify a ma­jor prob­lem in the wor­ld
    A sin­gle catas­trophic event on Earth can per­ma­nently wipe out the hu­man species

  2. Pro­pose a method of fix­ing it
    Colonize other planets, start­ing with Mars

  3. De­sign a self-sus­tain­ing com­pany or or­ga­ni­za­tion to get it done
    In­vent reusable rock­ets to drop the price per launch, then dom­i­nate the $27B/​yr mar­ket for space launches

I would en­thu­si­as­ti­cally ad­vise any founder to fol­low Elon’s play­book, as long as they have the stom­ach to com­mit to it for 20+ years.

So how does this re­late to Michael’s tweets? I be­lieve my ad­vice to “fol­low Elon’s play­book” con­sti­tutes a spe­cific ex­am­ple of Michael’s sug­ges­tion to “use the com­pany’s reach and power to make the world a bet­ter place”.

But here’s the thing: Elon’s play­book is some­thing you have to do be­fore you found the com­pany. First you have to iden­tify a ma­jor prob­lem in the world, then you come up with a plan to start a cer­tain type of com­pany. How do you ap­ply Michael’s ad­vice once you’ve already got a com­pany?

To see what I mean, let’s pick an­other spe­cific ex­am­ple of a suc­cess­ful founder: Drew Hous­ton and Drop­box ($11B mar­ket cap). We know that Michael wants Drew to “com­pete for hap­piness and im­pact” and to “use the com­pany’s reach and power to make the world a bet­ter place”. But what does that mean here? What spe­cific ad­vice would Michael have for Drew?

Let’s brain­storm some pos­si­ble ideas for spe­cific ac­tions that Michael might want Drew to take:

  • Change Drop­box’s mis­sion to some­thing that has more im­pact on happiness

  • Donate 10% of Drop­box’s prof­its to efforts to re­duce world hunger

  • Give all Drop­box em­ploy­ees two months of paid va­ca­tion each year

I know, these are just stabs in the dark, be­cause we need to talk about speci­fics some­how. Did Michael re­ally mean any of these? The ones about char­ity and em­ployee benefits seem too ob­vi­ous. Let’s ex­plore the pos­si­bil­ity that Michael might be recom­mend­ing that Drop­box change its mis­sion.

Here’s Drop­box’s cur­rent mis­sion from their about page:

We’re here to un­leash the world’s cre­ative en­ergy by de­sign­ing a more en­light­ened way of work­ing.

Seems like a nice mis­sion that helps the world, right? I use Drop­box my­self and can con­firm that the product makes my life a lit­tle bet­ter. So would Michael say that Drop­box is an ex­am­ple of “com­pet­ing for hap­piness and im­pact”?

If so, then it would have been re­ally helpful if Michael had writ­ten in one of his tweets, “I mean like how Drop­box is un­leash­ing the world’s cre­ative en­ergy”. Men­tion­ing Drop­box, or any other spe­cific ex­am­ple, would have re­ally clar­ified what Michael is talk­ing about.

And if Drop­box’s cur­rent mis­sion isn’t what Michael is call­ing for, then how would Drop­box need to change it in or­der to bet­ter “com­pete for hap­piness and im­pact”? For in­stance, would it help if they tack on “and we guaran­tee that any­one can have ac­cess to cloud stor­age re­gard­less of their abil­ity to pay for it”, or not?

No­tice how this par­allels my con­ver­sa­tion with Steve about Uber. We be­gin with what sounds like a mean­ingful ex­hor­ta­tion: Com­pa­nies should com­pete for hap­piness and im­pact in­stead of wealth and users/​rev­enue! Uber shouldn’t ex­ploit its drivers! But when we reach for speci­fics, we sud­denly find our­selves grasp­ing at straws. I showed three spe­cific guesses of what Michael’s ad­vice could mean for Drew, but we have no idea what it does mean, if any­thing.

Imag­ine that Dara Kos­row­shahi, CEO of Uber, wanted to take Steve’s ad­vice about how not to ex­ploit drivers. He’d be in the same situ­a­tion as Drew from Drop­box: con­fused about the speci­fics of what his com­pany was sup­pos­edly do­ing wrong, to be­gin with.

Once you’ve mas­tered the power of speci­fic­ity, you’ll see this kind of thing ev­ery­where: a state­ment that at first sounds full of sub­stance, but then turns out to ac­tu­ally be empty. And the clear­est warn­ing sign is the ab­sence of spe­cific ex­am­ples.

Next post: How Speci­fic­ity Works