51 Comments
User's avatar
NickS (WA)'s avatar

The discussion of hypothetical questions reminds me of John Holbo's great posts about thought experiments in Philosophy.

https://crookedtimber.org/2012/02/25/occams-phaser/

https://crookedtimber.org/2013/11/30/laugh-if-you-like-but-death-on-the-tracks-is-funny/

"The genre of the analytic philosophy (Anglo-American, call it what you like) thought-experiment is a mildly humoristic one, in that it tends to Rube Goldbergism. Of course the point is always to solve for variables! You never tie another victim to the tracks, or fatten one up, for any other reason than that he/she is strictly needed in that place or shape. Nevertheless, the more outlandish the set-up gets, the funnier it gets. And I think it’s fair to say that philosophers quietly award themselves style points for (plausibly deniable!) whimsy, above and beyond conceptual substance.

The problem with that, I should think, is that mirth is an emotion that may affect our moral thinking. Specifically, it makes us more utilitarian. See this more recent article as well [sorry, Elsevier paywall]. The trolley scenarios are, or may be, used as intuition pumps for utilitarian purposes. (They may be used for other things, of course.) But it is an underdiscussed fact that they may inherently do so, in part, because trolley tragedies can’t help being a bit funny."

https://crookedtimber.org/2019/12/16/whimsy-analysis-alienation-between-wodehouse-and-brecht/

"The issue is this: whimsy is – well, it’s not an emotion, I don’t suppose. It’s an attitude. More exactly, it’s a mode or manner of being detached. But it’s not a full, nor neutral style of detachment. It’s not the view from nowhere. It’s not action-oriented. But that doesn’t make it pan-observant or unfeeling. It’s perpetually tickled; it’s preferentially attendant to certain things, as opposed to others. (It knows you can’t just tickle yourself. Something else has to do it.)

The concern is that this makes it stupid, not to put too fine a point on it.

..."

Expand full comment
Dan Davies's avatar

I am wholly signed up to John Quiggin's campaign to just tone down the amount of unnecessary violence in thought experiments. I was going to link to Bernard Williams' Critique of Utilitarianism but then looked at it and my God, the two examples he gives have aged even worse than I thought they would. (https://123philosophy.wordpress.com/wp-content/uploads/2018/12/bernard-williams-a-critique-of-utilitarianism.pdf)

Expand full comment
mike harper's avatar

Question? Is Crookedtimber a much less interesting read than in the past when there were some interesting posts. Or, am I just getting old and out of date?

I miss me some Holbo and Berube.

Expand full comment
NickS (WA)'s avatar

I have had the same feeling, though I too am getting older. I have the impression that the comment section has acquired a reputation.

If I recall correctly, in this interview (recommended) BitchPHD calls the CT comments, "the Bellagio poker room of trolling" -- https://www.lawyersgunsmoneyblog.com/2022/01/an-oral-history-of-the-blogosphere-episode-4-bitch-ph-d

Expand full comment
Kevin Hawickhorst's avatar

Reminiscent of an old question that Abraham Lincoln used to pose.

"How many legs does a cow have, if we agree to call the tail a leg?" Answer: four, because calling a tail a leg doesn't make it so.

And unlike most of Lincoln's best "quotes," he really did say this.

https://timpanogos.blog/2007/05/23/lincoln-quote-sourced-calfs-tail-not-dogs-tail/

Expand full comment
TW's avatar

An economist and an anthropologist need to get to a remote area of Melanesia to do some development studies, so they hitch a ride on a cargo plane that flies out supplies once a month. But a storm comes up and the plane crashes on a tiny atoll somewhere in the Pacific. "We're doomed!" says the anthropologist. "There are 2,000 cans of food here, sure, but not one can opener!" The economist thinks for a minute. "Not to worry!" he cries. "I've cracked the problem. We *assume* we have a can opener!"

Expand full comment
Alexander Harrowell's avatar

the same, but assume we have a new joke?

Expand full comment
gregvp's avatar

"Not to worry!" cries the economist.

He cracks TW's head agains the crate of cans, killing him. He breaks the arms of the corpse and uses the shattered arm bones as knives to carve out the femur. He then uses the femur as a hammer and the arm bones as chisels to open a can.

... Too violent? Sorry, JQ.

Expand full comment
John Quiggin's avatar

Philosophical thought experiments are "a strange and impossible information environment that nobody could ever actually be in, so why should we expect our moral intuitions to give us good or satisfying answers here”." Preach it!

Expand full comment
Alex Tolley's avatar

"reject premise that safe return could be guaranteed"

Maybe I'm missing the point here. While a guarantee of safety is impossible, that doesn't mean that the experience of lunar flights cannot give an estimate of risk. After all, we take a risk of doing anything, from cooking a meal to flying to a destination. Experience and statistics provide a decent estimate of risk.

Rejecting a flight to the Moon because you reject the idea that it could not be guaranteed safe seems an absurd reason to justify what should be the real reason - "not interested/incurious" etc.

If the question was originally "Would you like to fly to Moscow?" and you answered no. Would it make any sense to have an option "reject the premise that you can be guaranteed to return safely" when we know that the vast majority of flights to and from Moscow are safe. Even the POTUS is willing to do that, and he might be far less safe than most civilians taking commercial flights.

OK, as of today, 1/6 of the Apollo lunar flights failed, but the astronauts returned safely. We have not traveled to the Moon, either to orbit it or land, since 1972. I wouldn't want to orbit the Moon, but I would like to land on the surface. But as I have no knowledge of the safety record of a lunar flight due to the lack of flights, I would simply say "no thanks" until at least 100 flights have been made and I can assess the risk as well as the discomfort of the flight.. If I were asked if I wanted to take a submersible dive to the Titanic, I would answer "no", based on lack of interest and the knowledge that a shady company managed to convince a wealthy father and his son to take that ill-fated dive. At least they died extremely quickly. It is the same with suborbital tourist flights. There have been too few to gauge the risk, even if the experience was exceptional.

Expand full comment
John Harvey's avatar

The Space Shuttle flew just 135 times. Twice it came apart in flight, killing all aboard. Yet NASA had claimed that the odds against this happening were one in a hundred-thousand, or almost as safe as flying in a commercial airliner at the time. So safe, you could send a school teacher up on it.

But on January 28, 1984, a cold day in Florida Hell, a "major malfunction" occurred, which was the term of art employed prior to "rapid unscheduled disassembly."

Don't engineers have a wonderfully truthful and emotionally evocative way of describing these unbelievably violent tragedies?

In reality the shuttle was so delicate that the oil from a fingerprint touching the head-resistant tiles could spoil them, and then they could come off and possibly cause the craft to incinerate during re-entry. One of my friends in the aerospace biz told me before a single one went up they'd be lucky to fly a couple dozen before they "lost" one. Challenger was number 25.

Chemical rockets are barely controlled explosions. "Too few to gauge the risk" is absolutely right.

Expand full comment
Dan Davies's avatar

And of course the astronauts were astronauts! A normal British person stopped on a normal British street ... If they are capable of giving an objective assessment of their physical condition and mental aptitude to the challenges of carrying out a lunar mission, you might be as well be asking them "if you could be guaranteed safety, how would you do in an unarmed fight against a chimpanzee?"

Expand full comment
Louise Ankers/Pixel Sisterhood's avatar

This comment thread also makes me happy. I love this survey.

Expand full comment
Alex Tolley's avatar

At some point, humans will travel in space as passengers, just as Arthur C Clarke envisaged in his early SciFi stories, particularly "A Fall of Moondust".

I don't know how old either of you is, but air travel was far more risky than today. Piston-engined aircraft flying in weather. But it got steadily better. The Comet 1 was involved in 25 hull-loss accidents, including 13 fatal crashes, which resulted in 492 fatalities, before it was grounded. I recall the BEA Trident that went down after takeoff from Heathrow back in 1972. Five years later, there was the Teneriffe disaster of 2 747s colliding on takeoff. Airline operations have improved safety, and you can view the statistics. It was this high level of safety that made the 2 Ba 737-Max crashes so standout. The only time I felt even remotely nervous was flying in a small aircraft from Guernsey to Gatwick in bad weather. I was also apparently in a near accident in an airliner that nearly overran the end of the runway while landing at Toronto in winter.

Airline historic safety stats (# incidents, but not expansion of air travel also occurred.

https://en.wikipedia.org/wiki/Aviation_accidents_and_incidents#Statistics

Airline safety today:

https://www.airlineratings.com/articles/the-worlds-safest-airlines-for-2025. I guess choose Air New Zealand for the highest safety ;-)

Expand full comment
John Harvey's avatar

Been around long enough to know four people who got killed flying: 2 in gliders, one in a turboprop, one in a military jet. My brother once flew a charter jet with a TV crew in it to the site of an airliner crash in Canada, and could barely control the plane in the fierce winds. Myself, I was at the scene of a helicopter crash before the bodies were covered; the worst thing I ever saw.

Airliners are incredibly safe compared to small aircraft. Best: big planes flying into big airports with long runways. I cringe when I go into Midway in Chicago, with short runways and a fence with a road on the other side. This is the only place where Southwest Airlines ever killed anybody; the guy botched the landing and ran through the fence, killing a boy in a car on the road just outside the fence.

You are very well-informed! Have you read Ernest K Gann's "Fate is the Hunter," about how flying was when it was really, really dangerous? And of course St. Ex....

Good sources:

Juan Browne on Youtube, James Fallows on Substack, the Professional Pilots Rumor Network on the web. The flying Bible: "Stick and Rudder" by Wolfgang Langewiesche, whose son was the recently-departed William Langewiesche, a tremendous writer on flying and much more.

I am a real stickler for focusing on what is real and true partly because of the flying. No patience for BSers. The first thing you learn about flying is you can't screw around; you can get killed. That concentrates the mind.

You can joke and laugh again after you land...

Expand full comment
Alex Tolley's avatar

JFK Jr. certainly paid the price for not taking flying seriously. Test pilots did take flying very seriously, but they still died "pushing the flight envelope". The risks they once took are no longer allowed, AFAIK.

I suspect commercial Moon trips will be more like being on cruise liners, which stay in space, with some transportation mode to reach the ship and the same at the Moon. Mature transport systems with excellent safety records.

Expand full comment
John Harvey's avatar

If you think about it, in the US we've had cars around for over a century but still kill 40,000 people a year with them.

We don't kill anything like that number with any other mode of transportation.

Individual drivers must make many decisions and perform many actions correctly for the system to work, and they often don't.

In a way, getting the big jets perfected was a lot easier. There are fewer of them, the sky is bigger than the highway, they have autopilots, they are flown by professionals, each accident is carefully investigated...vs a self-absorbed driver with a phone in an SUV that isolates you from your environment.

But it is orders of magnitude harder to accelerate a vehicle to 17,000 miles per hour than to 60.

On the other hand, who knows what powerful computers and AI or beyond will enable? Already there are rockets that can go back to the pad they just launched from. On the other other hand, they had so many contingency plans for what could go wrong on the Shuttle, and still they lost two out of a small number.

Let's be real: flying that Shuttle was playing Russian Roulette. Barely doable IF you are willing to consider killing some people an acceptable cost. It never made space flight routine.

If meaningful space travel is ever achieved, it won't be with chemical rockets. It may be completely inconceivable to us now. Maybe these UFO sightings are in facf examples of it? In our vast universe, isn't it likely that some civilization has figured this out?

Expand full comment
John Harvey's avatar

People have no idea, but think they do. I got my glider license from the test pilot of the WWII German rocket fighter, Rudy Opitz, and if the fuel that thing used spilled on you, it would dissolve your body. Somehow, he survived and made it to almost 100. Hard to believe; but hard to believe stuff happens.

I think a "guarantee of safety" is only offered in cases where something genuinely dangerous is proposed. If you get one, run the other way.

"I promise." "Trust me." "Perfect safety record."

Run the other way!

More like: "Perfect safety record...so far!" (ha-ha)

I also have some Florida swampland to sell you. Free alligators!

Expand full comment
Alex Tolley's avatar

The Me 163 Komet - using high-test H2O2. Very brave pilots to fly those devices. I saw one at a British museum once (Science Museum?). A slightly higher survival rate than the Japanese suicide pilots?

Expand full comment
John Harvey's avatar

That's the one.

https://en.wikipedia.org/wiki/Messerschmitt_Me_163_Komet

https://www.todayifoundout.com/index.php/2020/10/the-german-rocket-fighter-that-dissolved-its-pilots-alive/

https://avweb.com/news/komet-test-pilot-dies/

The book "Rocket Fighter" gives even more details.

These weren't suicide flights. They were just extremely dangerous, and ultimately, ineffective.

Expand full comment
Alex Tolley's avatar

The Challenger disaster was caused by brittle, frozen O-rings on the SRBs, and was precipitated by NASA managers pushing for the launch to please the government. Columbia was caused by the loss of tiles on the wing that exposed the metal airframe to reentry heating. NASA is now more cautious, which is why the Boeing Starliner was not allowed to carry the 2 astronauts back from the ISS until this year, because of uncertainty about the Starliner's fuel pumps. In the event it got back safely.

But yet, when Christa McAuliffe went up as a teacher, it was stated that the Shuttle was very safe for civilians to be part of the crew.

But experience is important, which is why the Space Shuttle had a less than 2% failure rate for the flight over its lifetime. Nowhere near airliner safety, but a reasonable bet for survival. If those odds were demonstrated for a Moon flight with a landing and return, I think I would take it if offered.

If McAuliffe had been told there was a 2% chance she might die, would she have gone on the flight, or declined?

Expand full comment
John Harvey's avatar

How you frame it matters. Also, if she knew how she would die...and for no good reason!

Haven't got the stats, but both the Russian rockets and Musk's rockets (except for his big one) seem to have a better safety record than the shuttle. The thing we should really pissed off about is that the shuttle was supposed to be cheaper and safer than single-use rockets, and it was neither. It was created so NASA and its contractors would have something to do after Apollo. It was way too complex and fragile. Before it ever flew, the Washington Monthly had a cover story: "Beam Me Out of This Death Trap, Scottie!"

At least the Soviets had the sense to send their copycat Buran shuttle into space for its only mission by remote control. It successfully landed itself; no need for a human pilot.

And what we all learned from the commission that investigated the Challenger explosion about the unwillingness to face reality...and then how the self deception continued and ended up killing a second crew...!!! Don't send humans up to do a dangerous job that robots can do.

Also, Christa McAuliffe's mom grew up about 25 minutes away from here, in Waterbury, CT, where I worked for the local daily newspaper.

Expand full comment
Alex Tolley's avatar

"Don't send humans up to do a dangerous job that robots can do."

I absolutely agree. The USAF is doing just that. Every robotic space probe and lander we send affirms that. The good news is that robots and their agency are improving by leaps and bounds and will eventually exceed any human capability.

Expand full comment
John Harvey's avatar

Yes.

Of course, we saved a lot of troops by dropping the atomic bombs, but also got the power to destroy civilization.

You must have seen "Dr. Strangelove." which was quite eloquent on this point. We have't destroyed ourselves yet, but it's only been 80 years, give us time!

Maybe once a generation everybody should be required to see that movie, and read John Hersey's "Hiroshima," to get sobered up.

We never solved the problems they revealed, nor did we outgrow them.

Didn't Einstein say you can't solve problems from the same level you were at when you created them? We are still at that level, or below!

We have gotten very good at living in unreality.

Expand full comment
OblivionNecroninja's avatar

Nah, more nukes = less wars. Nobody has even tried to invade a nuclear power.

Also, Nuclear Winter is super fake (the paper on which the entire concept is based was basically all assumptions).

Expand full comment
Alex Tolley's avatar

I also recommend watching the British docudrama "The War Game"(1966) about a nuclear attack on London. When I first saw it, I couldn't eat for over a day I was so sickened by the events. https://www.imdb.com/title/tt0059894/

Then, for added misery with no happy ending that makes Mad Max look like a picnic, "Threads" (1984) https://www.imdb.com/title/tt0090163

Expand full comment
Deep Learning State Machine's avatar

Seemed a bit funny to me that the St Petersburg Lottery seemed to be the one case where the one where economists would reject the premise, "nobody would offer such a lottery" yes sure but is that really the point?

Expand full comment
John Harvey's avatar

There is one horse you should never bet against, and that is Motivated Reasoning.

The point of the famous streetcar example is not to resolve an actual moral dilemma involving streetcars, but to make a meta-argument in favor of ulititarianism.

But a funny thing happened on the way to the victory lap:

Two utilitarians got into an argument about how to prove how wonderful utilitarianism was. They decided to go to a bar to settle it. But first they got into another argument about which bar to go to...the various pros and cons...what should they have to drink...what the winner of the bet should receive... some related questions about marginal costs and opportunity costs...what would Ayn Rand do....

Meanwhile, the earth was racing around the sun at 67,000 miles per hour.

Babies were being born.

People were living their lives.

There were tears, and laughter.

So...how do you settle an argument between two utilitarians? You don't. But, both may die.

How?

The ulititarians cried for help when the bar they finally made it to caught fire. There was only time to rescue one. Who should the fire department rescue?

Fire chief: "I have to stop and think about this. This is literally the situation they asked for a practical demonstration of."

The philosopher Jack Benny understood the problem:

https://www.youtube.com/watch?v=-tVzdUczMT0

Expand full comment
Kaleberg's avatar

In Jack Benny's era they were called lifeboat problems as in whom do you throw out of the lifeboat.

Expand full comment
John Harvey's avatar

Lifeboat jokes? Now I gotta look this up.

So, if you are the only occupant of a lifeboat, do you throw yourself out of it? Can you be a "whom" to yourself? And how do you solve that perennial "Either/Oar" problem on a boat with THREE oars?

Maybe these lifeboats are causing such trouble we should just get rid of them? We could re-brand them as "Troubleboats."

My other problem today: having driven into it, how do you drive out of a one-way cul-de-sac?

Human reason has its limitations, but do we blame the human, or the reasoning?

Q. How do you say "Hi!" to a hypothetical? :-)

Expand full comment
Ray Davis's avatar

https://en.wikipedia.org/wiki/Lifeboat_(1944_film)

(FWIW, I wouldn't want to leave either Hitchcock's movie or Manny Farber's essay on the trolley tracks.)

Expand full comment
John Harvey's avatar

Is this the Manny Farber who was the film critic? Could you point me toward his essay on trolley tracks? I did check out the Hitchcock link. I minored in cinema so I am always interested in film links.

Expand full comment
John Harvey's avatar

Duh, was it simple-minded me who thought the Farber essay was about trolley tracks, and arguments involving them....?

I think I finally grokked your meaning!

I think you refer to his essays as something not to be left on a set of trolley tracks...although I suppose that could be the setup for a suspense thriller...LOL

What did Farber say about "Bullitt?"

Expand full comment
Ray Davis's avatar

Oh, & Farber clumped "Bullitt" with "Coogan's Bluff", "Madigan", & "The Detective" as "a crowd-pleasing movie that has to do with a disenchanted cop, a city in which no corner is untainted, and an artichoke plot. Wrapped around a heart that is just a procedural cop story... a shrubwork of Daily News stories, the whole newspaper from beginning to end: the sensationalism, sentimentality, human interest, plus some liberal editorials. Each film has its mini-version of the drug scene, investigating committees, philandering wives...."

(Man, it's hard to stop transcribing that prose once I start. The Library of America collection is recommended.)

Expand full comment
Ray Davis's avatar

An understandable mistake, but yes, you grokked! :)

I wish Farber HAD tackled those dopey trolley games. He occasionally had a chance to write about topics other than movies -- I particularly remember a nice essay on comics.

Expand full comment
Indy Neogy's avatar

“the reason that this seems like a dilemma, or that the utilitarian answer seems weird or repulsive or underjustified, is that the weirdness has been built into the example, because it is describing a strange and impossible information environment that nobody could ever actually be in, so why should we expect our moral intuitions to give us good or satisfying answers here”

This statement is why even though I'd guess we disagree on plenty of things, I'm a dedicated reader. It's always amazing to me how rarely people poke at the construction of various questions. ("Cognitive biases" is another area (alongside economics and philosophy) where I frequently get heartburn reading the survey questions.)

Expand full comment
mike harper's avatar

I get requests to participate in polls that have a political bias. I try to evaluate for who and for what purpose the poll is being done. It usually is obvious. My fun is then producing answers to the questions that will fuck with the minds of pollsters.

Expand full comment
Kalen's avatar

I think that this hip moment for lumpen utilitarianism among a certain clade is motivated by it being a sort of puzzle solution rather than a universally productive moral framework. You're wandering around the Ethics Online MMO and find three people tied to train tracks- what gives you the most GoodBoy points? ('one weird trick to win'!) Then that outcome is treated as the result of a sort of physics experiment with universally applicability instead of an insight with *almost no applicability* because it was constructed to explicitly not resemble the world. Virtue ethics, deontology, and utilitarianism mostly give you the same answers as to what you should do with your day- if you're in a place where they are giving you wildly divergent answers you should probably back away slowly because something is very wrong.

Expand full comment
Samuel Williams's avatar

For a book I liked that locates exactly this kind of objection in academic philosophy – Avner Baz’s (2012) ‘When Words Are Called For: A Defence of Ordinary Language Philosophy’. Baz argues that something called ‘the theorist’s question’ is always being snuck in here, in a way that makes it look like it arises organically from the situation and matters to the people in it. The theorist’s question has the form, ‘But is this a real instance of x or not?’ Baz is not convinced that this question ever actually arises outside a philosophy department where the academics want it to arise, and you might get a kick out of his very thorough demolition of every possible version of these kinds of thought experiments.

Expand full comment
Kief's avatar

And most economic-political ideologies are based on the premise that you can organise an economy and society around a set of rules that will produce a lovely result for everyone - assuming humans don't pervert the rules for selfish purposes.

Expand full comment
Kaleberg's avatar

Wasn't there a trolley problem in which one had to choose whether to throw some guy off a bridge so his body mass would stop the trolley and save three other people? That guy had to be seriously massive. If I could throw him off a bridge, I'd probably do better by leaping off the bridge and stopping the trolley myself. Jack Kirby could do the illustrations.

Expand full comment
Manoel Galdino's avatar

I don't know much about separation theorems. Are you saying that they can't be “approximately true”? Either the conditions hold, or you’re misleading yourself by pretending separability exists? Am I interpreting your critique right?

Expand full comment
Dan Davies's avatar

A separation theorem is a theorem, so it's true if it exists. The question is how well the assumptions and axioms required to deduce the theorem approximate reality.

Expand full comment