The Gaza Hospital Tragedy
Thinking in Bets about what caused it
Annie Duke1 draws a contrast between seeking truth and feeling right.2 Duke believes that in order to seek truth in an unbiased way, we have to use probability theory, forecasting, and other tools of decision science.
Consider the uproar over a fire and explosion at a Palestinian hospital in Gaza. All of the decent people among us are outraged and saddened by the innocent loss of life there. But there are deep divisions over who to blame.
Hamas and other Palestinian voices from Gaza claim that Israel deliberately targeted the hospital. Israel claims that the blast was caused by a Palestinian rocket that misfired. Supporters of Israel who are determined to feel right buy the Israeli story. Opponents of Israel who are determined to feel right buy the Palestinian story.
I want to be a truth-seeker. I want to assign probabilities to each of four scenarios.
A The blast as accidental and caused by Palestinian munitions
B The blast was intentional and caused by Israeli munitions
C The blast was intentional and caused by Palestinian munitions
D The blast as accidental and caused by Israeli munitions
As of midnight on Tuesday night, I estimated the probability of A as 88 percent, the probability of B as 10 percent, and the probability of C and D as 1 percent each. These estimates are highly fluid, and I am prepared to make drastic changes as new evidence emerges.
I give zero weight to any “evidence” that comes via Twitter. The self-appointed ballistics experts and timeline creators on both sides are not reliable. Photoshopping is rampant and misleading videos are everywhere. We should not give automatic credence to any individual, relief organization, or government.
Here was my thought process as of last night:
Israel would be highly motivated to make up its version rather than accept responsibility for a strike against the hospital. On the other hand, they also would be highly motivated not to risk everything by putting out a story that could later be shown to be a lie.
If Israel did put out a lie and that is discovered, then it is game over for them. They could make a case for striking the hospital deliberately and owning up to it. They cannot make a case for lying about it.
Note, however, that if Hamas and its friends put out a lie, it will cost them nothing. Jews will complain about “blood libel,” but because no one expects Hamas to behave morally, they have a license to lie.
One factor that suggests that Israel is not lying is that at least one part of its story was not prepared in advance. There is footage from Al Jazeera that Israel interprets as supporting its claims. You can quarrel about whether that footage proves what Israel says it proves, but you cannot claim that the use of that footage is something that Israel anticipated ahead of time.
If it is a lie, the probability that they will at some point be caught is 99 percent. Issuing a lie would require considerable coordination in order to get their story straight. Sustaining a lie would require a large conspiracy. The chances of keeping such a conspiracy under wraps are not high. For example, if the U.S. has satellite images that show an Israeli missile strike, probably someone who sees those photos would talk.
Last night I also estimated the probability that other people will believe Israel’s version of the story if it is true. I speculated that the probability that Israel’s version can be definitively proven to a neutral observer is roughly 30 percent. Any evidence Israel supplies for its story will be discounted. Only evidence that comes from neutral or Gazan sources will convince neutral parties.
I thought that no matter what, the probability that the Arab street will accept the Israeli version is zero. It is like trying to convince a diehard Black Lives Matter supporter that Michael Brown was killed while attacking a police officer, rather than saying “Hands up, don’t shoot.” I see the Arab street as determined to feel right, not to seek truth.
Since last night, we have learned that the United States is inclined to accept the Israeli version, what I call version A. I was prepared to drastically lower my estimate of the probability of A if President Biden had aborted his visit to Israel for any reason.
Assuming that the U.S. continues to accept A, I would raise my probability of A by 1 or 2 percentage points, but not more than that. The U.S. does not qualify as neutral. I would raise my estimate of the probability that neutrals around the world accept A by maybe 5 percentage points, to 35. Again, the U.S. does not qualify as neutral, and neutral countries are not going to jump on board the U.S. bandwagon.
Thinking in Bets was one of the point-scoring categories that I came up with for Fantasy Intellectual Teams. Now that we see what AI’s are capable of, I could make a case for reviving the FITs project.
Another point-scoring category is displaying an open mind.
To score a point, a you must state a position on a question, such as "Vaccine passports will promote better public health" and say something like:
"I have changed my mind about ____, because ___", or
"I will change my mind about ____ if _____", or
"I have not made up my mind about ____, because _____"
I believe that those of us who are inclined to say that “the Gaza hospital tragedy was caused by a Palestinian rocket that went awry” would change our minds about the cause of the hospital explosion if persuasive evidence were provided (such as an American whistleblower saying that satellite photos strongly indicate that Israeli munitions caused the blast).
I doubt that people in the Arab street could give any sort of answer to “What would make you change your mind?” I am willing to change my mind about that if journalists conducting interviews on the street find differently.
I met her last week. A wealthy foundation put on a grueling three-day conference to brainstorm ideas for improving public discourse. I also met Kmele Foster (a 4th-round draft pick in the first edition of Fantasy Intellectual Teams). I’ve been meaning to write an essay about the event, but so far I have not had time.
This reminds me and others of Julia Galef’s distinction between the Scout Mindset (open minded) and Soldier Mindset (closed minded). See my review of Galef’s book.
"Now that we see what AI’s are capable of, I could make a case for reviving the FITs project."
Actually, it's just the opposite.
Now that we've seen what the AI's are capable of, you could make a much stronger case that we have already tumbled into the irreversible pitfall of a beyond-Black-Mirror world that makes anything like a FIT's project utterly hopeless.
The Gaza Hospital Incident is indeed a tragedy. But in the scheme of things, it is a minor tragedy, one of countless other such tragedies that swirl as mere specks in the ember-glowed smoking vortices of violence and warfare and death and destruction that is the destiny and fate forever for all who dwell in such lands until the very end of days.
But the PROBLEM of the Gaza Hospital Incident, of trying to figure out the true story, of who and what to believe, is THE REAL TRAGEDY and one of -monumental significance- that dwarfs all other concerns or any of the actual details of our squabbles, even though we may flatter ourselves into promoting the value of a few hundred or thousand human lives, as if they weigh -anything- in comparison to what is really at stake for us all.
They say that, "In war, truth is the first casualty." That's always been true, since the dawn of us.
But this time is different! Now, the casualty is not just some little truth, but "THE TRUTH"! Not just some facts or claims in dispute, but the very conceit that we have some reliable way of discerning the real state affairs from ... merely well-crafted illusions! Because the truth we now know is that we can't know the truth!
That is because no intellectual can now judge from a distance who or what can be trusted -as even real-.
Consider, "Photoshopping is rampant and misleading videos are everywhere."
Photoshopping?! There are AI-generated videos that look more real than any real video - already. Today. Millions of them. They can look so much better than real, that we start to doubt that something real could look that good, so they actually have to introduce the perfect amount of perfectly applied imperfections to perfectly fool us into thinking it's just imperfect enough to be 'real'.
Plato's cave, but the cave wall seems more realistic than reality, even to those looking in from outside the cave. "Damn, the progress in your cave displays has taken us out here in reality by surprise. I mean, that looks, like, more real than it actually is out here, even though, I'm like, looking at reality with my own eyes and stuff, but, compared to what's on your wall, hey, we're coming in, ok? Scoot over, make room, this is amazing!"
What makes conspiracy theories implausible is that the costs are implausible, that is, the costs of fabrication, coordination, etc. are implausibly high. But those costs just fell to the floor, and -everybody knows they are on the floor- so now what? Convincing fakes is not something for states or big companies or even experts, but something your cousin can do, and in fact does do, all the time.
What's the "health of intellectual discourse" value when you solve for the new equilibrium?
Reality is both material, and the information about those materials, so empirical evidence about facts involves the famously economically distinct atoms and bits. But if your atoms were not in close proximity to the atoms in a 'current event', everything between them and the intellectual's conception is just bits, and those bits are now -all- plausibly perfectly-realistic and incredibly-convincing lies.
To give you an idea of the trouble we're in, consider that there is no difference in principle between "socially-constructed" and "artificially generated". Content that has ideological or multimedial verisimilitude close enough to real to be trusted, will be able to convey just enough untruth as is calculated to effectively mislead.
Or in the alternative, to the extent a 'denier' of certain obviously true things was once a well-deserved epithet, it is now the case that literally everyone can plausibly deny the reality of any fact they find uncomfortable or inconvenient, simply because everyone knows it is now so easy, cheap, and fast to create realistic fabrications of any non-physical, bits-world, immaterial 'evidence' of anything.
Which is to say, "Evidence Is Over". And if evidence is over, there is no intellectualism possible beyond the pure airs of the peaks of considerations of mathematical and logical abstractions of greater and greater beauty, depth, elegance, subtlety, and sophistication.
Which is nice and all - I happen to enjoy what little I can grasp of it - but it's not "trying to discover the truth about the state of the real world of human affairs", which now seems to be in the process of flying beyond all our grasps. Instead of trying hard to keep firm hold of a few trustworthy tethers and anchors, we rushed to set them all aloft in the winds as lost feathers or worse, "Ashes to the Flames We Ourselves Lit!"
We are not ready for a world in which existentially radical skepticism about everything one is being shown and told goes from being something associated with the mentally ill to being -completely reasonable- as the most rational course of action in a world capable of making fabrications with infinite speed, costlessly, perfectly. What's a "whistleblower" in that world? Anything? What's Brin's "Transparency" in that world, a transparent lens to a kaleidoscope of illusions?
Now, ok, I get it, there is so much crazy """AI""" hype out there, that strikes as similar to all the other pipedream vaporware bubbles, that it is easy to adopt a kind of world-weary eye-rolling snarky attitude about it all, like it's really not even 1% of the big deal that it's made out to be.
Wrong. If anything, if you can even believe it, the hype-meisters are -under- selling it. They are probably not even trying to under-sell it, it's just getting better so fast that their attempts to over-sell are too slow for the reality to surpass their puffery!
Whatever twitter's faults, WSJ, NYT and everyone else ran with the unlikely, if not false, story that "Israel did it." (They haven't really corrected anything either . . . they're now just saying that the US agrees with Israel that it was the Palestinian's fault, but the framing is largely unchanged.) That twitter contains false and unverified information is a strawman. So does the MSM. Twitter is self-correcting, however, if you're a reasonably capable "truth-seeker." The MSM is not. If anything, it's twitter that keeps the MSM in line.