Many of you may be familiar with the bellingcat organization. Eliot Higgins started looking at and identifying munitions in Syria on a blog called Brown Moses, which he used as a pseudonym for a while. He was profiled in the New Yorker in 2013.
I have been interested in open-source intelligence for a long time. I started with an unclassified problem: how to find trash burial sites at the Los Alamos National Laboratory for potential cleanup. We did a bunch of work with overhead photos and other data, data fusion as it was called at the time. We hired some folks to do infrared photography – the burial pits would collect water and be a lower temperature than surrounding areas.
That was back in the 1990s. My team did some pioneering things.
Earlier, I had a project on a destruction method (supercritical water oxidation) for hazardous wastes. The chemical weapons people, who were just beginning to face the enormous problem of destroying their very hazardous chemicals, asked if that method might be suitable. They eventually decided not to use it, but I had to study chemical agents for a year or so.
Higgins was particularly looking at chemical weapons in Syria and where they were coming from. It’s generally accepted now that the Syrian government has been responsible for the chemical attacks. The information was developed early by the bellingcat consortium (which I’ve contributed to for some time in small ways) and confirmed by the Organization for the Prohibition of Chemical Weapons, the international organization responsible for such things.
That’s a long preface to say that today bellingcat published an article of mine. There are still people who do not believe that the chemical attacks in Syria were carried out by the government. That’s been an argument all along. One of those people is Ted Postol, an emeritus professor at MIT. He and Higgins plan to debate in October. So I went back to Postol’s early arguments and worked through the chemistry. It’s pretty bad.
Cross-posted at Nuclear Diner.
Roger Moore
Unfortunately, no amount of facts are likely to convince these people, because their conviction doesn’t come from factual evidence. They don’t want to believe, so they’ll look for evidence that allows them to continue to disbelieve.
rikyrah
Thanks for the article.
Joey Maloney
Advisory to the jackal community: If you don’t have at least a basic understanding of chemistry, you can go read Cheryl’s article or you could just read her three-word final sentence above. You’ll extract the same amount of meaning from either.
Signed, took chem 101 3 times: Drop, D-, C+
Rms
I liked your statement “no reference needed the answer is four”.
Brachiator
Is there a connection here to this recent story?
Cheryl Rofer
@Brachiator: No direct connection with my article. But not a surprising connection between Assad and chemical weapons manufacture.
Gravenstone
@Joey Maloney: Hey, you were getting better !(assuming you reported grades in chronological order).
sdhays
I’m not sure he’s going to be able to get up after having that bus backed up over him so many times…
Gin & Tonic
@Joey Maloney: Hey, I managed to get my BS without ever taking chemistry at all (satisfied the req by taking Materials Science.) So you’re saying I’ll waste my time with that article?
Gin & Tonic
But really I popped in to say that Bellingcat has/have done outstanding work on the MH17 case as well. Some of us have been following that closely for a little over four years now.
Tom Levenson
Excellent piece over at Bellincat. (And don’t believe that bit about not being able to get it w/out prior chemistry. Might have to look up a couple of things, but the whole piece is clear and easy to follow.)
One thing that strikes me is that Postol has a kind of naive popular physics view of all of science. Get some first principles and a calculator and you can reason out anything. Most sciences don’t work that way (including an awful lot of physics), but I think (and speaking as someone who has written his share of popular physics) that there is a kind of pop-culture view of science that is threaded through with this error. Something good science writing might be able to mitigate, if folks choose to read it.
The Moar You Know
There are still people who think that Donald Trump won the 2016 elections fair and square. I don’t have the time to even address either form of obvious bullshit. The Syrian attacks were either carried out directly by the Syrian government, or by another group with the government’s knowledge and permission, and that really is a distinction without a difference.
Cheryl Rofer
@Tom Levenson: Thanks, Tom! I tried to write it so that anyone could understand it, but you never know until folks without much chemistry try to read it!
A great many physicists tend to believe that their physics gives them the power to reason out pretty much anything from first principles. I’ve run into far too many of them in my career…
Matt McIrvin
@Tom Levenson: There are a lot of topics where that approach seems to work, but only because a lot of fiddly details were hidden in the exposition. I’ve heard speculation that this is a contributor to so many engineers becoming scientific cranks: they’ve learned good-enough approximations to the fundamental science involved in their engineering work, and mistake it for a deep understanding exceeding that of the actual practitioners.
As a general rule, when you go through an exercise like this and conclude that all the experts in some field are making the same elementary mistake that any bright high-schooler could find, you’re probably wrong.
Another Scott
@Tom Levenson: Yup, it’s a big problem. Physicists, especially, and Electrical Engineers, also too, are particularly susceptible.
I’m reminded that NDT very confidently, but wrongly, weighed in on DeflateGate. Of course, he did admit his mistake and explain the error, as all good scientists and engineers should…
Cheers,
Scott.
sherparick
@Roger Moore: “A man hears what he wants to hear and disregards the rest.” The Boxer – Simon & Garfunkel, 1968.
Much of the Anti-Anti-Russian Left (a Russia now ruled by a right-wing, racialist, anti-LBGT, anti-semitic, autocrat in alliance with the Russian Orthodox Church) simply won’t accept that Trump or Russia can do anything wrong. It produces a lot of cognitive dissonance where they try to be pro-Trump, pro-Russian, and pro-Iranian at the same time.
Joey Maloney
@Gin & Tonic: Chem was really my bête noire, I think because I’m so bad at spatial visualization. I wasn’t that great at doing force vector diagrams in classical mechanics, either, come to think of it. More abstract stuff, though, where it was all about the math, I ate that up with a spoon.
Anyway, pace Tom Levenson, I won’t project my inadequacies on to others. *I* didn’t get much more out of Cheryl’s explanation than “it’s pretty bad” and I’ll leave it at that.
…and I’m getting bitten by the disappearing name/email bug every time today, which has never happened to me before.
Lee
I am continually amazed at the width & depth of knowledge in the den of rapid jackels.
OT: I had a thought. What if Trump’s latest frantic tweeting isn’t triggered by Mueller closing in but his cabinet mutinied and started taking steps to thwart Russia meddling in our election? It might be against his wishes or he didn’t authorize it but they started.
Lee
Spelling sucks on mobile sorry for the many typos
Major Major Major Major
Thanks and way to go, Cheryl! I’ll (try :) to give it a read this evening.
Geoboy
@Another Scott: I’ll see your physicists and electrical engineers and raise you doctors.
JCJ
From the article:
Seriously? Wow. While I always have liked chemistry I understand that others do not. Nonetheless I would think if one is going to write something even remotely related to chemistry one should have an understanding of hydrogen bonds.
joel hanes
interested in open-source intelligence
Then I can recommend the site Skytruth
https://www.skytruth.org/
(which does NOT require a grasp of chemistry. Usually.)
Steeplejack
@Tom Levenson:
So how does science work? Serious question. What would be the short, snappy alternative to “first principles and a calculator” that would be more accurate? (In other words, not demanding a 1,000-word think piece.)
Doug R
Just like most of the Russia defenders-not actually interested in informing, they just throw enough plausible sounding crap in the air to sow doubt.
Boussinesque
I can’t claim to understand the finer points of the chemistry, but what I do understand makes it clear Postol is talking out of his ass. You mention the likelihood of some “informant” being the one who put him up to this showing of his ass in public—do you have any idea who/what organization might’ve been responsible for pushing this line of attack? Just trying to think who all benefits here in the US from throwing doubt on the Syrian government being the ones using the chemical weapons (I’m assuming that that was the aim?).
Doug R
@Boussinesque:
Russia?
cliosfanboy
@Matt McIrvin: there is a great XKCD on this. I’ll try to find it when I am back at my desk and not on my phone. Or you can use the Google machine to find “President of physics”
sukabi
Straitjacket and thorazine have been deployed…no press appearances or travel. White House is trumps bunker.
germy
Tom Levenson
@Matt McIrvin: Yes. The smartest person in the room delusion has led many astray.
joel hanes
@Steeplejack:
Everything matters at some scale, and the devil is in the details.
Example: if only the earth and sun existed in the universe, the earth’s orbit would be a perfect mathematical ellipse. But the moon tugs us toward it, so the earth actually “slaloms” back and forth, and other planets, esp. Jupiter, tug the ellipse out of shape, and then gravity changes the shape of space-time subtly, which must be accounted for.
GPS doesn’t work if you engineer from first principles: they made it accurate by compensating for the relativity difference in the rate of clocks between satellite orbit and the Earth’s surface.
Good engineers and scientists are aware of this bias toward over-simplified models, and often mock themselves with cracks like “First, assume a spherical cow”.
Cheryl Rofer
@joel hanes: Skytruth looks interesting. With satellite photos now so available, there are many, many things that can be done in open-source intelligence. Geo4Nonpro is one that I frequent, with emphasis on finding nuclear and missile proliferators.
Major Major Major Major
@cliosfanboy: also “Struggle no more! I’m here to solve it with ALGORITHMS!”
https://xkcd.com/1831/
Steeplejack
@cliosfanboy:
XKCD “President of Physics.”
Cheryl Rofer
@Steeplejack: Science is a collection of facts, organized by numerous principles and theories. You need both the facts and the organization.
Cheryl Rofer
@Boussinesque: The informant is a person who calls herself “Syrian Sister.” She is an apologist for the Assad regime. I believe that others have figured out her identity, but I have wanted to stay away from that particular troll farm.
Tom Levenson
@Steeplejack: I’m not sure there’s a simple answer, or rather that I have one, but I’ve been looking a lot at the period immediately before Newton publishes the Principia and the critical elements of that birth-of-moder-science period are empiricism and quantification (really, mapping experience onto math, not just numbers): Find a way to turn description into numbers that upon which you can perform calculations; and draw those systematic descriptions from experience in ways that others can repeat. (really the other way round.)
So science is the creation of rigorous tools for making measurements and the analysis of those measurements using the formal apparatus of mathematics, which can be as simple as turning raw data into percentages and as complicated as multi-dimensional manifolds doing all kinds of stuff that make my brain hurt.
Make any sense?
Steeplejack
@sukabi:
But Trump already is traveling. Isn’t he at Bedminster to start his 11-day “vacation”?
CliosFanboy
@Steeplejack: thanks! Make sure you read the mouse-over text.
@Major Major Major Major: another perfect one!
Tom Levenson
@Tom Levenson: @Cheryl Rofer: alternatively, what Cheryl says.
The Moar You Know
@sukabi: Straitjacket and thorazine indeed. Well, he did it to himself, can’t help stupid.
Major Major Major Major
@Tom Levenson: speaking of science writing, is Gleick’s time travel book good? Thinking about picking it up.
YetAnotherJay formerly (Jay S)
One brief criticism of the article from a non chemist: You use the chemical formula HF without spelling it out as hydrogen fluoride. I took it as an undefined acronym instead at first read.
Tom Levenson
@Major Major Major Major: Yes. My review of it here.
schrodingers_cat
@Steeplejack: First principles approach works for many problems. But a scientist’s skill lies in figuring out which details are important and which details to leave, that depends on the problem you are trying to tackle. For example, the approximation of earth as a point object works when we are discussing planetary motion across the sun but not when you want to study ocean currents or wind patterns.
ETA: And any mathematically elegant theory has to explain observed facts, otherwise it doesn’t matter.
sukabi
@Steeplejack: pretty sure he’s been in Bedlam for a while…the next couple of hours will be interesting to see if he’s been “contained” or if he’s out golfing.
Hard to tell with the mountains of Bullshit his crew produces on an hourly basis.
Think Mueller’s phone has probably been ringing non-stop since Trump’s “yes we colluded” tweet. How many of Trump’s staff would be willing to go to jail for him? It’s deal cutting time.
Don
@Cheryl Rofer: as Lord Rutherford is purported to have said, “All science is either Physics, or stamp collecting.” And then we try to figure out what String Theory is….?
schrodingers_cat
@YetAnotherJay formerly (Jay S): H is the chemical symbol for hydrogen and F for fluorine, so the chemical formula for hydrogen fluoride is HF.
Brachiator
@Cheryl Rofer:
I originally heard this story as a BBC news report, which also included interviews with people “in the know” who claimed that Syria was not using chemical weapons. But if someone is killed who supposedly was a chemical weapons scientist, one obvious question would be why was this person targeted.
Reading your article was very enlightening. A couple of things stood out.
In context, I would wonder why anyone would bother with Postol in the first place. Maybe I’m naive, but I would think that you would want to go to a person who had knowledge of weapons production, chemical engineering, etc., not someone who is trying to logically deduce conclusions about whether chemical weapons were used based on his general knowledge. This reminds me a bit of local TV weather presenters who make grand pronouncements about climate change.
I ain’t no rocket scientist, but your article made a lot of sense to me. I think I at least have a grasp of the type of questions that should be asked and how the evidence has to be evaluated.
ETA: I originally posted from my mobile device, but for those interested a story about the car bomb attack on the alleged head of the chemical weapons facility.
Roger Moore
@Tom Levenson:
I think an even more critical misunderstanding is one Cheryl points out in the article: confusion between science and engineering. Engineering uses science, but there’s plenty of engineering work built around getting a result that works well in practice even if the details of how exactly it works are not fully (or even partially) worked out. As you probably know, a lot of science is built around explaining phenomena people have taken advantage of without understanding.
Major Major Major Major
@Tom Levenson: oh thanks!
schrodingers_cat
@Don: I think the problem is testosterone not physics. Physics is dominated by men (mostly old) compared to even chemistry. Hubris comes naturally to that group no matter what profession.
ETA: I have to input my nym for every comment. Tiresome. Windows 10 and Firefox.
Steeplejack
@sukabi:
Yes, I know (both Bedminster and Bedlam). I was reacting to the “White House as bunker” metaphor. Kind of pointless if the principal is not actually in the bunker.
Steeplejack
@Don:
Ahem.
Matt McIrvin
@Steeplejack: In addition to what everyone else has said, science is a social process. One of the most important ways to keep from going off the rails is just to keep track of what others are doing, and certainly not blindly follow authority, but don’t assume they’re all idiots either.
schrodingers_cat
Question for the hive mind: The best way to deal with the salary requirement question at the beginning of a job interview process.
YetAnotherJay formerly (Jay S)
@schrodingers_cat: I had figured that out, but as a non chemist reading the article introducing the formula HF without the name threw me for a bit, especially since the typography didn’t appear to be different between formulas and text. Other chemical formulas were introduced by name before their first use. This was not.
Cheryl Rofer
@Don: The typical physicist pejorative for everything else.
Major Major Major Major
@schrodingers_cat: “I’m looking for a competitive offer.”
If they absolutely demand a number go on Glassdoor and see what comparable people make at the same company.
Roger Moore
@Steeplejack:
Real science has to include a lot of experimentation. Working out from first principles is neat when you can, but you have to test your theory against the real world. When your theory conflicts with the real world, you have to scrap the theory and stick with the real world*. It’s great when you can explain a higher level phenomenon in terms of lower level entities (e.g. explaining genes in terms of the chemistry of nucleic acids) but this is almost always after-the-fact explaining of real world experience rather than predicting new phenomena from first principles.
*This can be somewhat overstated. Understanding the result of an experiment requires a theory to explain what you’re seeing, and it’s possible there are problems with that explanation. So real-world science involves a lot of extra tests to see if you’re understanding your experiments right.
Cheryl Rofer
@Brachiator:
That is the most important thing for anyone outside any realm of expertise. It’s what I try to get across in much of what I write.
Postol came in early and hard on the Ghouta chemical weapons attack in 2013. His field is missiles and their trajectories. He got some things right earlier, so he seems to feel that allows him to be right about anything else he may care to opine on. It’s not clear to me whether Syrian Sister got to him first and encouraged his opinionating or whether he started talking and then she showed up. Young attractive woman pays attention to older man; compare Maria Butina. Works every time.
rikyrah
@Lee:
WHY EITHER/OR?
Why can’t it be BOTH/AND?
otmar
@YetAnotherJay formerly (Jay S): that puzzled me as well.
Roger Moore
@Brachiator:
He gives them the answers they want to hear.
sukabi
@Steeplejack: has he actually been spotted outside the WH?
schrodingers_cat
@Major Major Major Major: That’s pretty much what I said too. The recruiter came up with a number, which sounded reasonable to me.
Tom Levenson
@Roger Moore: Yes. That’s a confusion that is likely to continue to get worse, IMHO.
Steeplejack
Thanks to all for the responses to the “short definition of science” question. All helpful, but none succinct enough to rebut “You can figure out anything with some first principles and a calculator!” Which is probably what makes those reductive one-liners so maddening. “No, you can’t!” never seems to work as a comeback, for some reason, and the next thing you know you’re in a long, tedious argument with someone who can’t be convinced of anything.
I don’t consider myself a scientist, but I spent many years in a science-y field, software development. A lot of similarities in the issues—with both principles and personalities—with what I have read about science. There is such a thing as “computer science,” but it’s at the rarified, Rutherfordian end of the spectrum. Most of what I did was down at the engineering level, and when I say “engineering” I mean more like machine-shop engineering than classical, tech-school engineering.
But, like I said, there are many similarities to the “scientific method” in the search to understand systems and solve problems, and many similarities in the personalities you run into. Always found it interesting to sit in a meeting where probably half the guys (always guys) thought they were the smartest one in the room. Good times. I imagine physics is like that.
schrodingers_cat
@Tom Levenson: His excellency, the magnificent Tikka has not graced the FP in a long time. How is the summer heat treating his floofiness.
Ruckus
@Tom Levenson:
A lot of people don’t want or know how to think through a process while remembering where they started or some would say see the process. Some are capable of this naturally, but most have to be taught. A lot of science requires this ability. Actually a fair amount of all work does as well but not to the same degree.
Lee Hartmann
@Tom Levenson:
It appears that Postol got his undergrad degree in physics and PhD in nuclear engineering, so I wonder if he has the physicist hubris that he can understand everything pretty much from first principles. I liked his criticisms of missile defense, but that may just reflect my own bias.
YetAnotherJay formerly (Jay S)
@Steeplejack: I was leaning to a concise scientific method description, but the problem you describe is akin to the “but my calculator says” problem. Someone doesn’t understand the problem well enough to see the errors they are making, and are refusing to check their results. If that someone has credentials in another area, they are used as a voice of authority for something they are not an expert in.
RSA
@Matt McIrvin:
This is a nice summary of a position I’ve taken in online discussions with climate deniers and the like. It still surprises me, a little, that people don’t see what they’re doing.
WhatsMyNym
@Steeplejack: How many times have you’ve been able to write to the specs of a problem and didn’t have to change the software again?
Science is the same. Might work in the lab; in practice, somethings just never make it out of the lab.
schrodingers_cat
@Tom Levenson:
The calculator is unnecessary. Beyond intro physics no calculator is needed in physics classes. You derive stuff, rarely if ever, does one calculate an actual number.
Doug R
Just how many nyms does everyone have? Depending on my computer, two left clicks gets me 1 to 4 nyms which I can left click on. Then 2 -4 emails and post. This in a house with 3 adults and 2 laptops. Firefox or Chrome on Windows 10
Matt McIrvin
@RSA: The Emperor’s New Clothes–all the pros blinkered by threats, flattery and groupthink, which a mere child can puncture–is a powerful story. Sometimes it’s even true. In most cases it isn’t, though. At the very least it’s an extraordinary claim that needs extraordinary support.
Jay
@Brachiator:
“But if someone is killed who supposedly was a chemical weapons scientist, one obvious question would be why was this person targeted.”
Orxy Blog’s twitter feed notes that the scientist in question has spent the last couple years working at a Hezboallah/Iranian missile assembly site in Syria.
Roger Moore
@Steeplejack:
If I wanted a single sentence rebuttal, it would be that it isn’t science unless you compare your theory to the real world. Maybe you can figure out anything from first principles and maybe you can’t, but you can never know for sure until you’ve compared your first principles calculation to experiment.
Alternatively, if somebody really thinks they can solve any problem from first principles, tell them there’s a Nobel Prize waiting for the first person who can accurately predict protein folding based on quantum mechanical calculations.
mapaghimagsik
wow, that leaves a mark. Well done.
RSA
@Roger Moore:
Great summary.
J R in WV
@schrodingers_cat:
Easy way is to ask for a substantial increase over your current salary based upon the additional things you have learned to do well since your current salary was established.
Harder and better way is to investigate the salary structure in the hiring organization, or similar organizations in general, for positions like yours and use that information in comparison with your desired increase in salary to make a move, to inform your ask in your interview.
Always a tricky thing. Probably should always ask for a little more than you must have.
When Joe Manchin was CEO (governor) for WV he introduced a terrible salary hiring policy. We were required to always ask a job applicant for their minimum salary requirement, and then the personnel office required us to make an offer substantially lower than the applicant’s requirement. Thereby insuring that well qualified and capable applicants would be very reluctant to accept any offer you could make.
Humiliating for all of us, esp for Mr Manchin as the incoming quality of state workers declined abruptly on his watch.
Another Scott
@Steeplejack: Bertrand Russell, Ludwig Wittgenstein, and Kurt Gödel are nice rebuttals to the “first principles and a calculator” argument.
If one can’t come up with a perfect, first-principles way to do simple mathematics, and one can’t, then it’s obvious (or should be) that first principles and a calculator aren’t enough for (nearly any) real problem.
Russell and Whitehead thought they could come up with a perfect, first-principles mathematics – The Principia Mathematica. They failed.
Cheers,
Scott.
J R in WV
@schrodingers_cat:
When you discuss salary in an interview with the actual hiring people, go higher than the recruiter suggested.
It is to his benefit for you to be placed regardless of the salary you make, which he doesn’t care about as he doesn’t need to live on it. He cares about the volume of hires he can place, so under-bidding on money is to his advantage. Recruiters have a different set of motives than people being hired and the people hiring them.
You didn’t mention a recruiter in your first post.
RSA
@Another Scott: Nice, subtle argument.
J R in WV
@YetAnotherJay formerly (Jay S):
Yes, but calculators don’t always provide a correct answer. Today they can approximate the correct answer very well. In the 1980s I worked on a business tax system, which computed past due taxes plus interest and penalties. Every business tax had its own method for determining a due date, a truly bewildering variety of picking the middle of a month, etc, etc.
Plus each person who was manually processing each kind of tax (14 different ones IIRC) had a different model of desk-top calculator, and used it in a slightly different manner. So their answer was always, always a little different from the computer system’s computation. We finally sat them down in a conference room, with their individual calculators, and went through computations together. They all got different for the same set of data.
Enuff said, trust the computer, we’re talking about fractions of a cent per day in differences, and they were both bigger (good) and smaller (bad), so it was a wash at the end of the fiscal year. Which was different for different taxes… so glad I quit that job!!
EthylEster
I liked the part where Postel identified his collaborator as a quantum chemist.
The topic is an extremely practical one. Why consult a theoretical guy?