This is a linkpost for https://volpe.envs.net/blog/posts/cia.html
New Comment
There's also the simple possibility that some of this stuff does in fact work.
Due to the environments I move in, I've personally seen extremely weird stuff. Whenever I read research papers, which tend to be old, on experiments that were made on stuff relatively similar to what I observed, they always strike me as weird, because the test designs are almost invariably extremely different from what the people involved in any of that stuff actually do. Basically, the sequence is something like this:
- The people involved use technique P to achieve effect E.
- Researchers become interested in effect E, observed anecdotally.
- Researchers take notice of technique P but discard it as a proper cause for E given that P is almost invariably untestable via double-blind methods, much less triple-blind ones.
- Researchers hypothesize an alternative causal mechanism Q for effect E, this one being double- or triple-blind-able.
- Researchers develop a protocol to test Q and apply it.
- Testing for Q fails at producing E.
- Proper conclusion (not expressed): Q doesn't cause E.
- Improper conclusions (frequently expressed): E doesn't exist and/or P doesn't cause E.
Now, it's possible E in fact doesn't exist, and yes, that's the highest likelihood by far, but so far, in my observation at least, the methods attempted to test the several "Es" I looked at all fell into the above failure mode, which make me unconvinced of much of the almost certainty surrounding that.
For the record, I'm aware of two proposed protocols por testing two Es, developed by practitioners of each, that aim at properly testing their actual respective Ps. Those two protocols would be expensive to put in practice, as they both involve controlling for dozens of variables simultaneously, would require from hundreds to thousands of subjects put into observation for a long period of time, and would have absolutely no double/triple-blinding. Needless to say, obtaining grants for either would be a very tall order.
The important point to note here is the corollary that the CIA very likely believes in AGI as powerful, and probably in alignment risk. And they are terrifyingly effective, or at least they were in the past. I expect them and/or the NSA to ensure that the government controls AGI before it's smart enough to evade their control.
Wrt why they've believed a lot of bunk: Science is hard. You don't have to be dishonest to get wrong answers. You just need Motivated reasoning and confirmation bias. P-hacking happens by accident. It made my fields, cognitive psychology and cognitive neuroscience, wrong about half of the effects we believed in as of about 2010, despite fancying ourselves highly sophisticated on statistics. And an informal equivalent, noticing success more than failure, happens by default, too.
The USSR during late-stage Perestroika and Russia during the 1990s also were affected by similarly pseudoscientific ideas. Additionally, the first chairman of the Russian Commission on Pseudoscience wrote a book trilogy which implied that one of the reasons for such ideas to receive funding is corruption.
I also imagine that if you see your enemies researching something, that makes it feel very important to research it, too. (You wouldn't want to lose a war just because you underestimated the impact of horoscopes.) So if USSR or USA starts to research something for reasons of corruption or stupidity, the other side will do the same, because "if they spend money researching it, there is probably a good reason for that".
This can escalate, like maybe one side just assigns one person to research the horoscopes, but when the information leaks the other side overreacts and assigns the entire department, etc.
Could possibly be done on purpose? Let's start dozen lines of bogus research and let the information leak, so that our enemies get confused and start wasting their energy on useless things. With the possibility that other people on our side, who were not informed of this, will also get serious about the research.
Asymmetric incentives? If I have a few budget and assign a few people to research horoscopes, that is an expense I can justify. If I don't assign anyone to researching horoscopes and they turn out to be crucial, I will get fired for incompetence.
I will add in passing that the CIA requires a polygraph screening for all employees until today despite all the research showing it to be ineffective at telling the truth from lies
The evidence for this is not very good, and is also not assessing the polygraph assessment the modern US government actually conducts. I think it's very plausible that they have worked fairly reliably for the last ~25 years given the dearth of publicly disclosed cases of long-running spying ala Aldrich Ames or Robert Hanssen, and given how much former intelligence officers seem to believe it works & comment on their reliability in passing.
Do you believe the US government has access to better polygraphs than is commercially available and/or studied in the academic literature?
If so do you have non-indirect evidence for this?
Do you believe the US government has access to better polygraphs than is commercially available and/or studied in the academic literature? If so do you have non-indirect evidence for this?
Yeah; my evidence for this is strewn across public interviews with 'former' CIA officers, books about the U.S. intelligence community, and personal discussions with family members who are/were officers in the U.S. military. It would take time I don't have to go and gather it all but for the basic "polygraphs they are using seem to work" I think you can just look at what people who have recently left have said, and the (non-)history of long term employed defectors over the last ~15y.
1After experimenting with him for eight days the CIA concluded
That seems like a strange reading of what happened. The CIA gave the former laser physicist and later paranormal researcher Russell Targ (the person who coined the phrase "remote viewing") money to run a study on Uri Geller and Russel Targ concluded in his report on the study that Uri Geller demonstrated his abilities.
This does not mean that the CIA overall believed that Uri Geller had abilities.
You could just as well say that Nature concluded that Uri Geller has abilities given that Targ published in it as well a year later. The fact that the CIA let Targ publish the experiment they funded in a journal suggests to me that they didn't thought it was useful enough to classify the experiments.
I will add in passing that the CIA requires a polygraph screening for all employees until today despite all the research showing it to be ineffective at telling the truth from lies[1]
That's not a claim that's true. Polygraph screening isn't perfect at telling truth from lies but it's better than chance. Just because they are used does not automatically mean that the people who use them are overvaluing it. They might very well be overvaluing the results of the test, but that's not something that's shown.
Hmm. My understanding is that a "polygraph screening" is an interrogation; conducted and judged by human interrogators; using a polygraph machine as a sort of interactive stage prop to put the subject under more stress. The output of the polygraph machine is not dispositive; the judgment of the human interrogator is. That is, whether you pass or fail is not a fact about the machine's output; it's about what you say, how you act, and whether the interrogator believes you.
Many stories of people "failing polygraph screenings" involve the subject confessing. If you do that, it is not going to matter what the machine output says!
So the question "Does polygraph screening work?" does not mean "Does the polygraph machine itself detect lies?" bet rather something more like "Does adding a polygraph performance to a human-conducted interrogation make the interrogation more effective?"
The Wikipedia page on James Vicary (The "inventor" of Subliminal advertising) mentions that the CIA investigated subliminal advertising and found it to be somewhat useful. After reading the source I found that the report is more skeptical than presented but it did tickle something in my brain; the CIA was (is?) interested in so much bunk! I've compiled an incomplete list of some pseudo-science the CIA was involved in. Lest you think they were just doing their duty and investigating all possible ways to get an edge, I will quote some of their findings where they seemingly 'fell for it'
List of involvements
Uri Geller
After experimenting with him for eight days the CIA concluded
source
Dowsing
Maybe prof. Calculus was right after all! From the Discussions and conclusions section of a study on dowsing.
"Interesting" theories of consciousness (Do your own research!)
In a 30 page document investigating the "Gateway experience" and hemi-sync, one can find all kinds of great paragraphs. Here's a choice quote.
Robert Monroe was the inventor of Hemi-Sync. If you want a more complete picture of his worldview, this is from his Wikipedia page
Misc.
I will not get drawn into MK Ultra, but there's a lot more thataway. I will add in passing that the CIA requires a polygraph screening for all employees until today despite all the research showing it to be ineffective at telling the truth from lies[1]
What to make of it
On the one hand you'd like your government to research anything that might help them defend you. It wouldn't be as interesting if they investigated various forms of pseudo-science and concluded that it will not serve to advance the interests of the state. But they didn't do that, they found a lot of this useful.
What I think may be happening is that the Agency wanted to study a specific topic to find if there's any there there. There aren't many scientists capable and willing to do these experiments. The few that are, are people on the fringes, people who have risked their reputation and possibly destroyed it. You have to assume the reason they did is because they do believe in the theories being tested. Maybe a combination of belief plus the potential for more money for future experiments caused them to design faulty experiments.
If you have any better theories, let me know.
It's possible they just want to test the performance of a prospective employee under pressure