Inexplicable knowledge

Is it possible to know something, and yet be unable to convincingly explain how you know it?  I think so.

[Just to be clear, when I say “I know something” I mean that I believe I know it. But then what else could it mean?]

David Henderson recently said:

Like Scott, I doubt that the CIA was behind the JFK assassination, but all I have is doubt. I don’t have the certainty that Scott has and I don’t know what’s behind that certainty.

Just to be clear, I’m not completely certain of anything.  But basically David is right; I claim to “know” that the CIA did not conspire to assassinate Kennedy, with 99.9% certainty.  And yet I cannot explain why I know this.  So do I?  Let’s use an analogy of a picture of Trump’s face, made out of 10,000 dots, or pixels if you prefer.  I might look at the picture and say it’s obviously Trump.  But how do I know that?  None of the individual pixels looks anything like Trump.  Rather it’s the cumulative effect of all those pixels that creates the likeness.

When we go through life we accumulate an enormous amount of information.  Each piece of information is like a dot, and together it gives us a complex worldview that tells us that some ideas are plausible and some are not.  The best I could do is use an analogy, something David would agree with.  I might say that I know that the American Girl Scouts leadership council was not behind the Kennedy assassination. If that didn’t work, pick a conspiracy that was even more far fetched—David’s grandmother.  At some point he’d accept the idea that one might know something because the alternative is too implausible, or at least most people would.  But of course that wouldn’t help at all with the CIA (which really does do nasty things.) The problem is that our life experiences give us each a different set of facts, and a different brain to process those facts.  I see a different CIA from the one David sees.

When I was much younger—like 3 months ago—I used to think it was a productive use of time to try to convince someone that Trump’s a demagogue, because . . . well, because he’s obviously a near perfect dictionary definition of a demagogue. But it’s pointless. For every fact you cite, they’ll point to other non-demagogue politicians who do something similar, at least on occasion.  Trump’s demagoguery is like the picture with 10,000 pixels, you either see it or you don’t.  No single example of the big lie, or of demonizing minorities and foreigners, or of unrealistic promises, or macho posturing, is going to convince anyone, because they’ll always be able to explain it away.  After all, politics is a very messy business.  And each argument is just one dot.

This also relates to monetary policy.  I know that monetary policy was too tight in 2008 and 2009 and that the Fed could have adopted a policy that led to faster NGDP growth.  But if asked to explain how I know this, I’d have trouble explaining my belief.  I lack an elevator pitch.  I could tell people to read my entire blog, from end to end.  But that’s 1000s of pages of argument, and it still wouldn’t even come close to explaining my belief, which also depends on decades of reading economic theory, economic history, and the history of economic thought.  That reading creates the brain architecture or grid that determines where I store all the various facts that I come across, and explains why I often just “know” that a commenter’s facts are wrong, without having actually checked. That doesn’t mean I don’t try to convince people (in the blog I do the best I can), just that it’s very difficult to do.

On the other hand I strongly recommend that people not try to explain their beliefs on the Kennedy assassination, or 9/11, or why Noam Chomsky is wrong about US foreign policy, or why Trump is a demagogue, or why free will doesn’t exist, or why Scott Alexander is brilliant, or what it means to “know something”, unless you enjoy pointless debates.  The odds of convincing anyone are so small that it’s not worth the effort.

PS.  I don’t believe that ‘inexplicable knowledge’ is the right term, but am not sure what is.  What I have in mind is not just tacit knowledge, as it can also involve reading books or articles.

PPS.  I was going to do a post on Trump’s nominees, which so far are mostly lousy. But it’s probably not worth the effort.  So I’ll just do a PS. Confirm them.  In politics, I always try to put principle over expediency.  So although I don’t like many of the nominees, I’ve always felt that Presidents have a right to pick the people who will serve them, unless something truly awful turns up.  I didn’t think it was fair to prevent those women from serving as Attorney General back in 1993, just because of various “nanny-gates”, and I’m not going to change my views just because Trump is President.

PPPS.  Vox recently published this piece by Sherri Underwood:

I remember the precise moment that I realized I regretted voting for Donald Trump.

It was during his 60 Minutes interview after the election. I was, like everyone else, shocked that he had won. It seemed so unlikely based on the polls and the confidence the media had that he would lose. It was a pleasant surprise, and I went to bed on election night thrilled that he would be our president.

But sitting on my couch, sipping coffee as I watched the interview, I saw with my own eyes who Trump really was as a person. He backtracked on one of his signature campaign promises: pursuing an investigation into the Clinton email scandal. It’s not that I want Clinton to be crucified or “locked up” — it’s the nonchalance with which he went back on his word after hammering it repeatedly during the campaign. The ease and quickness with which he reversed his position shook me to my core. I realized in that moment that I had voted for a demagogue. And it was sickening.

Three months ago I would have mercilessly mocked her stupidity.  Now I respect her much more than I respect myself.  Writing that article took courage.  Not surprisingly, she’s a Midwesterner.

PPPPS.  Speaking of the Midwest; Minnesota, Iowa and Wisconsin were three of the most liberal states in the country back in 1988, going for Dukakis while Illinois and California went for Bush.  This fascinating factoid from the National Review suggests they are about to turn red:

In the Upper Midwest, demographic trends have lent a hand: In 2004, Iowa, Wisconsin, and Minnesota were among the few states in which the oldest white voters were the most liberal, and the generation born of the Great Depression has been dying off.

Those old hippies are the Dukakis voters.  The Wisconsin I grew up in is gone—just faded memories.  And one more dot to slightly rewire the political map in my brain, which has both spatial and temporal dimensions.

BTW, Politico has a piece on Pepin County, Wisconsin that is the single best article on the election that I have read.  I will do a post.

Update:  Regarding Trump’s alleged demagoguery:

Screen Shot 2017-01-20 at 5.10.56 PM



53 Responses to “Inexplicable knowledge”

  1. Gravatar of AL AL
    20. January 2017 at 15:09

    Is it too early for a glass of wine?

  2. Gravatar of Christian List Christian List
    20. January 2017 at 15:29

    I don’t believe that ‘inexplicable knowledge’ is the right term, but am not sure what is.

    Maybe the word you are looking for is simply intuition or gut feeling. Or pattern recognition.

    I saw with my own eyes who Trump really was as a person.

    So what did she do with her eyes during the campaign?! This story is ridiculous. She’s either pretty stupid or the story is made up, fake news so to speak.

  3. Gravatar of Doug M Doug M
    20. January 2017 at 15:30

    What you speak of is “epistemology.” The theory of knowledge. Which is all about how do you know what you know.

    Not to get sidetracked on philosophy, as an academic, you really should be know how you know those facts that are directly in your area of expertise, such as how tight money caused the 2009 recession. It is important. If you can’t articulate your assumptions, and the evidence that took you to that conclusion, then you are just a talking head.

  4. Gravatar of Lorenzo from Oz Lorenzo from Oz
    20. January 2017 at 15:33

    That Politico piece is indeed excellent. And what’s alleged about The Donald’s demagoguery?

  5. Gravatar of Philo Philo
    20. January 2017 at 15:40

    If the utterance of any English sentence whatever were equivalent to the utterance of “I believe that” followed by the sentence, it would follow that “I believe that” was empty, adding nothing to the sentence in which it occurred. But you know that is untrue–in fact, that it is ridiculous.

    Another thought: if all your assertions really amounted to saying that YOU (Scott Sumner) believed such-and-such, and all my assertions amounted to saying that I (Philo) believed such-and-such, then we could never contradict each other.

    You are hopeless in philosophy. Fortunately that hardly matters.

  6. Gravatar of ssumner ssumner
    20. January 2017 at 16:12

    Al, Yes.

    Christian, No, not gut feeling, not if you have 1000s of pages of technical economic research buttressing your opinion.

    Yes, maybe fake news. And perhaps your comments are fake comments. And maybe my posts are fake posts.

    Doug, Yes, I know what epistemology is.

    I certainly know why tight money caused the recession, but what if the explanation takes 20 years?

    Philo, You said:

    “Another thought: if all your assertions really amounted to saying that YOU (Scott Sumner) believed such-and-such, and all my assertions amounted to saying that I (Philo) believed such-and-such, then we could never contradict each other.”

    Not at all, I could bring new information to your attention that would cause you to change you mind. It would prove that what you believe is not so. (Or what I believe is not so, in the unlikely event I were wrong.)

    And if someone asserts something, I think the listener usually assumes they believe it to be true. At least I assume they mean that. If you tell me that Paris is the capital of France, then I assume you mean “I, Philo, believe that Paris is the capital of France.”

    And if you say, “no, I mean Paris really, really is the capital of France” I take that as ” I, Philo, believe that Paris really, really is the capital of France.”

  7. Gravatar of B Cole B Cole
    20. January 2017 at 16:49

    When does “inexplicable knowledge” morph into “bias confirmation” and then hubris?

    Egads, millions of highly intelligent people supported Hillary Clinton, and millions of highly intelligent people supported Don Trump.

    Why are asset bubbles specious but labor shortages real?

    Because not everyone thinks alike.

  8. Gravatar of Art Deco Art Deco
    20. January 2017 at 17:26

    You can explain passably how you know that the CIA did not conspire to kill Kennedy:

    1. No discernable motives.

    2. No credible evidence in favor of the proposition.

    3. It is provably true that Lee Harvey Oswald killed Kennedy.

    4. No credible evidence Oswald was ever employed by the CIA.

    5. No discernable reason they ever would employ him even as a janitor. He was bizarre and erratic and had failed at every blessed thing he’d ever done with his life.

  9. Gravatar of Jerry Brown Jerry Brown
    20. January 2017 at 17:51

    I don’t know professor, but it is common knowledge that the Girl Scouts are pawns of the Radical Islamic Terrorists. It could be that they (the girl scouts) were involved with the JFK thing too.

    That I cannot come up with any evidence to support these things does not mean they aren’t true. So there!

  10. Gravatar of Dan W. Dan W.
    20. January 2017 at 18:00

    I thought Scott expressed this idea so well I rewrote with very slight modification:

    “I used to think it was a productive use of time to try to convince someone that Obama’s a demagogue, because . . . well, because he’s obviously a near perfect dictionary definition of a demagogue. But it’s pointless. For every fact you cite, they’ll point to other non-demagogue politicians who do something similar, at least on occasion. Obama’s demagoguery is like the picture with 10,000 pixels, you either see it or you don’t.”

    Obama’s demagoguery was to attack the political right – aka the “Tea Party” – as having racist motivation and being the blame for why he was unable to work with Republicans. Trump made the “Tea Party” his ally. He also invited anyone who did not feel represented by the Democratic agenda to be represented by him – and many took him up on the offer.

    And then there is Hillary’s demagoguery as she invited ridicule on everyone who would not support her. Well, she and her supporters can ridicule all they want. But they will do it standing outside the White House fence looking in.

  11. Gravatar of Benjamin Cole Benjamin Cole
    20. January 2017 at 18:41

    From Trump’s inauguration speech:

    “For too long, a small group in our nation’s capital has reaped the rewards of government while the people have bore the cost,” Trump said. “Washington flourished, but the people did not share in its wealth.”

    “The establishment protected itself, but not the citizens of our country. Their victories have not been your victories. Their triumphs have not been your triumphs,” Trump noted. “And while they celebrated in our nation’s capital, there was little to celebrate for struggling families all across our land.”


    Not sure I can disagree with Trump’s sentiments. In fact, this is what many people mean by regulatory capture. And really, who writes the tax code?

    In addition, Trump has suggested paring back intelligence agencies, the F-35 fighter plane, and the silly $4 billion fancy Air Force one boondoggle from Boeing.

    Trump’s style is bombastic, blustery. His past personal life and commentary that of a bad-boy celebrity, not an august statesman.

    But I find anti-Trump commentary a bit overwrought, even hysterical. Let’s see what Trump does.

  12. Gravatar of Jim Glass Jim Glass
    20. January 2017 at 18:41

    Coincidence, or hidden forces exerting their hidden powers…?

    Japanese tuna prices and GDP growth.

  13. Gravatar of Major.Freedom Major.Freedom
    20. January 2017 at 18:53

    The Huffington Post is FAKE NEWS.

    Summer you wrote:

    “Is it possible to know something, and yet be unable to convincingly explain how you know it?  I think so.”

    “[Just to be clear, when I say “I know something” I mean that I believe I know it. But then what else could it mean?]”

    What do you mean by “know” that differs from “believe” such that you use both words in the same sentence in a way that distinguishes them from one another?

    One of the most common misconceptions is the notion, derived from religious creeds, that because our thoughts are physically nothing more and nothing less than “mere” brain patterns, or “mere” states, rather than all powerful infinite pure energy with omniscience and omnipotence, that this somehow prevents us from having actual true objective knowledge about anything in the universe. That everything we could ever think, is not true knowledge, but “mere” beliefs.

    Oh contraire.

    For one thing, EVEN THINKING THAT THOUGHT presupposes your having the ability to know objective reality, namely, the objective reality of you yourself and your place in the world around you. To say that you ARE such and such a being that can “only” have such and such constitutions that you call “having beliefs”, that is you transcending those very beliefs and understanding yourself to be a being that has more than mere beliefs. This is a true objective knowledge that you actually alluded to in your opening statements of knowing something but not knowing how or why you know them.

    Even if you tried to understand THAT as also “merely a belief”, guess what, you again transcending those beliefs and again understanding yourself as a being that is more than the first belief PLUS that next belief. No matter what, there is a you that remains no matter how hard you try to think of your thoughts as “mere” beliefs. You could try to write an infinitely length sized book on how your beliefs are just predicated on other beliefs. Beliefs all the way down, so to speak. But the entire time there is an understander of the beliefs. There is an activity there. The aggregator of concepts. The being which not only has beliefs, but understands itself as having beliefs.

    This subtle reality of the consciousness is what escapes most people. True objective knowledge is extremely modest, limited, but absolute. True objective knowledge is better understood as meta knowledge. It is more than a knowledge of things. It is knowledge of what knowledge is. The true objective knowledge you can have is where you yourself, and the rest of the universe, coincide. True objective knowledge is that reality which is present in everything you could possibly think or do.

  14. Gravatar of Ray Lopez Ray Lopez
    20. January 2017 at 18:57

    An honest piece by Sumner. The problem with Sumner’s beliefs is that they are grounded in ephemeral effects: monetarism is a weak force. So Sumner must rely on counterfactuals in his eye’s mind, or confirming examples from history, perhaps the most famous being the USA devaluing off gold in 1933/34. I do the same thing when it comes to raising Total Factor Productivity by advocating a better patent policy. The trouble is, while my rhetoric is sound (inventors do respond to incentives; not every inventor is a geek who cares little about money and more about fame, nor a noble Nobelist, who invents out of pure love of science), there’s little evidence for my position. Most people don’t invent, steal from others, the actual inventors just shrug off the theft as ‘gee whiz, that’s life’ and go on with it, and… life & invention goes on. Same in monetarism. Having a whiz-bang great monetary policy like NGDPLT and simply practicing a feed-forward policy like the real-bills doctrine, or the Greenspan/Bernanke/Yellen put, is usually hard to quantify as to what real effects they have on the economy. I think for this reason Shiller simply threw up his hands and espouses ‘animal spirits’ now.

    I just read the above, and it sounds unusually charitable to Scott. Must be getting mellow in my old age. Will change that in the next post.

  15. Gravatar of Ray Lopez Ray Lopez
    20. January 2017 at 19:06

    @Jim Glass – I saw that clever graphic from the Economist and Japan’s tuna shaped graph on GDP, and I think what they are doing is drawing curved lines between years to give a “tuna fish” shape, rather than straight lines. So it’s fake news, except that indeed there’s no correlation, just like the Phillips curve in the 1970s.

  16. Gravatar of E. Harding E. Harding
    20. January 2017 at 19:09

    The only place in Pepin county that swung Dem last year was the place there that got the largest Bernie %age in the Dem primary in Pepin.

    Yup, Twin Cities Hipsters.

    But that town (Stockholm) still went for Trump by five points.

    Fact is, you could see Pepin County would swing GOP already from the primary results. Trump got more votes there in the primary than any other candidate. Trump got 58.2% of the combined Trump+Clinton vote in the primary, and 59.06% of the vote in the general. Even Lyin’ Ted Cruz would have easily won Pepin County against HRC (though Bernie might have pulled out a narrow win over Lyin’ Ted, he would still have lost to DJT in Pepin).

    This is one of your better posts.

    I only think there’s an 80% chance the CIA did not assassinate Kennedy. I don’t know why you think the CIA is so moralistic regarding domestic politics. They probably took down Nixon.

    “That reading creates the brain architecture or grid that determines where I store all the various facts that I come across, and explains why I often just “know” that a commenter’s facts are wrong, without having actually checked.”

    -Be careful. Some of your “facts” are just propaganda and completely nonsensical assumptions. Most of the things you think about foreign policy are wrong, for instance. You should check much more often.

    “I was going to do a post on Trump’s nominees, which so far are mostly lousy. But it’s probably not worth the effort. So I’ll just do a PS. Confirm them. In politics, I always try to put principle over expediency. So although I don’t like many of the nominees, I’ve always felt that Presidents have a right to pick the people who will serve them, unless something truly awful turns up.”

    -Many of Trump’s nominees (Puzder, his Goldman appointees, Ben Carson, Rick Perry) really are lousy to the extreme.

    “Writing that article took courage. Not surprisingly, she’s a Midwesterner.”

    -Locking up Clinton would be fun, but knowing Trump was lying about it would not have changed my vote in the least. I’m much more policy-focused than that. And, yes, Trump will repeal and replace Obamacare. Hillary would never do that. So the woman should still be criticized for her nuttiness.

    “Regarding Trump’s alleged demagoguery:”

    -The Norks do not strike me as demagogues.

  17. Gravatar of Jerry Brown Jerry Brown
    20. January 2017 at 19:54

    I don’t like Trump and his inauguration speech did not make me any more happy about him being our President. But I would have to agree with Tim Taylor here

    If I am wrong about Trump I hope to be as fair as Tim Taylor.

    And why is it so difficult to comment at your Econlog site?

  18. Gravatar of Engineer Engineer
    20. January 2017 at 20:04

    I asked Siri the other day how many economists it took to screw in a light bulb. She said…And I quote…One hundred..One to screw it in and 99 to blog about it… Ahh I think she could be our first woman president

  19. Gravatar of Jerry Brown Jerry Brown
    20. January 2017 at 20:06

    Damn. Screwed up again. Hopefully this is the link

  20. Gravatar of mbka mbka
    20. January 2017 at 20:45

    Beautiful post on knowledge, Scott.

    There is some pattern recognition in this. And when you use small amounts of facts for sweeping conclusions, using prior experience as a guide, I’d call it thin slicing. The problem is that people recognize different patterns when viewing the same data, see various notorius flip-flopping optical illusions. Not only this, but you even may be able to see some patterns better if you have _less_ information. Example, the smile of the Mona Lisa changes if you defocus your eyes. You decrease your optical spatial frequency resolution and pick up on not less, but different, frequencies, i.e. information. This goes a nice way to show how the “less well” informed sometimes _may_ see some things clearer than the “better” informed. Seeing the forest for the trees is a related concept.

    But there are definitely aspects of tacit knowledge here too. When I teach tacit knowledge, I include a wide array of cases: knowledge that can be taught, but only through practice (e.g. piano playing), knowledge that can be experienced but only shown, not taught (example, smells), and knowledge that could be taught easily, but is not and remains hidden because no one knows it’s important. For the last case, I propose an example of an old factory worker ho gets fired for only working 10 min a day. Trouble is, he was the only one who knew how to oil an old machine whose manuals have been long lost. No one knew this was important, not even himself. Without his 10 min worth of daily oiling, the machine now breaks down, and no one knows why. Hayek addressed this latter type of knowledge in his papers on knowledge in society, and in “Why I am not a conservative”, where he warned that we should not arbitrarily change things in society whose function we may not fully understand. He used a kind of evolutionary logic: If something’s been around for a long time, and worked, it is probably well adapted and implicitly includes information picked up over time from many sources. This btw is THE major argument against would-be revolutionaries, including the current calamity-in-chief of the US. Do not push over things you don’t fully understand.

    But the best way to describe your kind of knowledge is in the book I recommended to you before, ET Jaynes’ “The logic of science”. Once you have 9,999 pixels, you have a vast amount of prior knowledge in the Bayesian sense. This gives you a very high confidence, or probability, to be right about the 10,000th pixel. But the 10,000th pixel’s meaning in the picture is impossible to explain without knowing about the priors, i.e. the rest of the pixels and their interrelationships. The logic of science is to fit all knowledge in theories that form a relational consistent whole, always including all prior knowledge. This is also why no real life scientist rejects theories just because of a single Popperian falsification event. That’s because the large amount of previous corroborations of a theory demand very strong evidence of falsification to be pushed over.

    So, you “know” why you are right on any new emerging factoid because of accumulated prior knowledge fitted into coherent systems of theories tha have been corroborated. This “knowing” could even be quantified probabilistically. You really are very probably right if you have a lot of prior knowledge on a subject. Single new facts either fit theory, or remain temporarily suspended if they don’t fit, until sufficient evidence is accumulated to change your mind and form a new subset of theory.

  21. Gravatar of Matthew Waters Matthew Waters
    20. January 2017 at 21:23

    The CDC did a study with anti-vaxxors. Laying out evidence didn’t work. What’s more surprising is graphic evidence of the diseases didn’t work either.

    IIRC, showing the diseases actually strengthened the anti-vaxxors sentiment. The study authors theorized that more emotional evidence triggered the more reptilian, fearful parts of the brain to protect one’s child.

    I’ve seen this pattern many times. The more an issue matters, the less open people are to logic and evidence. I don’t get it.

  22. Gravatar of Jerry Brown Jerry Brown
    20. January 2017 at 21:41

    Matthew you don’t get it because you have a 10,000 pixel brain.

    If you had a 100 pixel brain like me you would see this clearly.

    oops, my pixels are full- time to empty them again. Be right back..

    Much better now. What was I talking about?

  23. Gravatar of Steve Steve
    20. January 2017 at 22:46

    The PPPPS is good. I have a lot of political anecdotes that I could share.

    I’ve definitely noticed, on social media and elsewhere, several of the most vocal anti-Trump Republicans are from Texas, while some of the strongest Trumpers are from the Driftless area. There’s definitely a realignment going on.

    I think a big factor is that Texas has always been libertarian, right-to-work. Immigrants come to work, not to collect from non-existent social services. Free trade is good for Texas, too.

    In contrast,the Scandinavian model is being screwed by global arbitrage. Unions, social services, public goods, all require a community mindset among residents; they can’t survive with free-riders. That community mindset doesn’t exist with multi-national corporations, nor with third-world immigrants.

    Places that still offer strong social safety nets like California and Massachusetts are able to fund themselves by taxing resident monopolist corporations. That’s not a nationally transferable model.

    Also my grandma is a PA Hillary voter…but listening to her talk politics is insane; she thinks Hillary will protect factory workers from fat cat bosses who create physically dangerous factory floors. It’s like the world hasn’t changed since FDR! So PA gets redder when the old guard passes.

    I also can recount lots of instances of city people being extremely rude and condescending to rural folks. I’ve seen it quite a bit. I see it regularly with tourists from NY and Boston chewing out NH service workers.


    I don’t fully trust the Politico piece. You find one confederate flag, one homophobe, and one person willing to use the n word, and you’ve got a narrative! It’s hard to know how much deference to grant the author. Maybe it’s an accurate picture, maybe it isn’t.

  24. Gravatar of Alice Alice
    20. January 2017 at 23:14

    The Problem of Knowledge by A. J. Ayer

  25. Gravatar of Steve Steve
    20. January 2017 at 23:56

    Also the Politico narrative made a deal out of Minnesotans moving to Pepin County, as if there is a political difference between people from MN and WI. There isn’t. The only difference is that Minneapolis is bigger than Madison, and therefore able to swing the state.

  26. Gravatar of Postkey Postkey
    21. January 2017 at 01:09

    “On the other hand I strongly recommend that people not try to explain . . . why Noam Chomsky is wrong about US foreign policy, . . . ”

    Quite right. Don’t think. Just accept that he is wrong!

  27. Gravatar of Anand Anand
    21. January 2017 at 02:15

    The 10,000 pixel argument is same as a Bayesian argument. You evaluate all the evidence and see how much of it matches a given hypothesis (or a given set of hypotheses) – updating the probabilities along the way. This also allows for different priors – namely that you’ll need a lot of evidence to believe that the CIA was behind Kennedy’s assassination, because the hypothesis is not very plausible a priori (it’s not a 50/50 prior, for sure).

    Another complication with political or religious arguments is that people are not really rational in these domains, so they don’t update their probabilities easily. Still, one can try.

    I don’t really debate people on political matters to convince them; I try for an exchange of views. That is often (not always) beneficial even if nobody is convinced. Who knows, I might even be convinced by them, or be forced to examine my own priors. I cut off the discussion when diminishing returns sets in. Seems to work for me.

  28. Gravatar of ssumner ssumner
    21. January 2017 at 06:43

    Art, Yes, that’s part of it. But the “motive” thing requires a lot more information in the background, as some might not agree with you and I. And there’s many other factors, such as the difficulty of keeping that sort of thing secret, the fact that the CIA would not pick such a clumsy technique for killing Kennedy (it was a lucky shot), etc. etc.

    Dan, Obama has many flaws but he is one of the most professorial and least demagogic politicians I have ever seen. The fact that you can’t see that speaks volumes. It’s obvious that you don’t even know what the word ‘demagogue’ means.

    mbka, Excellent comment.

    Steve, You said:

    “You find one confederate flag,”

    That’s not what it said, read it again. You also missed the point about the Twin Cities migrants.

  29. Gravatar of Ray Lopez Ray Lopez
    21. January 2017 at 08:07

    @Alice- both you and Sumner are over your head here. The mind-body duality and the ‘problem of knowledge’ are philosophical problems studied by generations of scholars and has no neat resolution.

  30. Gravatar of Michael Rulle Michael Rulle
    21. January 2017 at 08:55

    I like your metaphor of inexplicable knowledge. It is an “explanation”, for example, of how anti Trumpers cannot understand how anyone could vote for Trump. Oddly, my inexplicable knowledge, can understand why people did not vote for Trump.

    I did not like Trumps inaugural speech, yet it was the same thing he has said all along. My sense now is that the political class fears him because they believe he has the populace on his side.

    Someone wrote they doubled the probability of a nuclear war from 1 percent to 2 percent under Trump. Thats what I call pure inexplicable knowledge.

    The big fear really is that he could be Hitlerian in style, by appealing to those who seek to blame others for their self perceived victumhood caused by (elites, politicians?). I would rather believe he is the Wolman (sp?) Rink poltician rather than a demagogue.

    But I found that my inexplicable knowledge said we needed to stop the leftward march at least for one or two terms. If the economy goes well, and wars are limited, everything will be fine regardless of the face of bluster. If not, he will be out in 4 years.

  31. Gravatar of Art Deco Art Deco
    21. January 2017 at 09:23

    Art, Yes, that’s part of it. But the “motive” thing requires a lot more information in the background, as some might not agree with you and I.

    Not a whole lot. Kennedy was a roue, but he was at work a perfectly mainstream pol of the sort the spooks worked for throughout the post-war period. Fanciful mass entertainment (e.g the film Executive Action) and Kennedy family press agents (eg Arthur Schlesinger) attempted to peddle the idea that the Johnson Administration discarded and disrupted the Kennedy Administration’s prudent plans in VietNam, which is a hoot. See Henry Kissinger’s critique of the overthrow of Ngo Dinh Diem, which occurred just three weeks ‘ere Kennedy’s death; see Marguerite Higgins’ critique of Diem’s opposition.

  32. Gravatar of Saturos Saturos
    21. January 2017 at 09:26

    With this attitude Eliezer Yudkowsky would never have wrote the Sequences.

  33. Gravatar of Major.Freedom Major.Freedom
    21. January 2017 at 09:27

    Ray, where did Alice say that there is a “neat resolution” to the mind body problem? On what grounds can you say she is “in over her head”?

    There was just a book reference.

  34. Gravatar of Saturos Saturos
    21. January 2017 at 09:29

    *written. Dammit, I’m slightly inebriated.

  35. Gravatar of Rick G Rick G
    21. January 2017 at 09:36


    “See, I have the advantage of having found out how hard it is to get to really know something, how careful you have to be about checking the experiments, how easy it is to make mistakes and fool yourself. I know what it really means to know something. And therefore, I see how it is that they get their information and I can’t believe that they know it—they haven’t done the work necessary, they haven’t done the checks necessary, they haven’t done the care necessary. I have a great suspicion that they don’t know how this stuff is done and they are intimidating people by it.”

  36. Gravatar of Bob Bob
    21. January 2017 at 10:28

    It’s not quite inexplicable: It’s teachable, but it could take far longer to teach than it’s worth, especially when there is a big abstract thinking differential. I see this kind of example all the time at work, in areas that are far less politically charged. I might have 100 charts in front of me, describing how a large computer system is working. I look at those charts, and see a problem, while someone with less experience or less aptitude might not see said problem, and the information comes from the correlations in 20 of the charts. Part of the job of someone that is good at seeing problems in the charts before they are critical is to make it easier for other people to see similar problems in the first place. In the happiest of situations, we manage to make a computer realize that there is a problem and alert us.

    This is not limited to humans or simple computer programs: If you look at what we call deep learning in AI, you’ll see that we have computers doing the same thing. Our advancements in computer vision, self driving cars and fraud prediction work in the exact same way.

    Now, the computers are also teaching us how our systems can fail. Sometimes the machines catch on to the wrong insights, and end up making terrible mistakes. Since we can see exactly how the AI is making its decisions, we can change the data that we give it to confuse it: My friend Julia Evans. wrote am article on this, titled How to teach a teach a neural network that a panda is a vulture. It’s out there on the internet.

    If you ever want to look deeper into this, Tyler has plenty of contacts: He’s getting interviewed next week in San Francisco by the CEO of the company where Julia works.

  37. Gravatar of Student Student
    21. January 2017 at 12:44

    Anyone catch this little nugget at the 3:31 mark of Trumps speech at the CIA?

    “.., the rain should have scared them away but god looked down and said we’re not gonna let it rain on your speech…”. Followed quickly by “… it stopped immediately and when I walked off and it poured…”

    The dude is both a demagogue and narcicist. Scary.

    What makes it even funnier is that his presumes all this while he has to look at the text on his papers to recite the Our Father at the national prayer service at the Bascilica this morning.


  38. Gravatar of Student Student
    21. January 2017 at 12:45


  39. Gravatar of Dan W. Dan W.
    21. January 2017 at 13:41


    Obama is a partisans ideologue and when agitated he blamed his political enemies and characterized them as Neanderthals.

    But don’t believe me. Just look at the evidence! The more the Democratic party was defined by Obama the worse the party did! I think a real professional president would have had much superior outcomes then did Obama.

  40. Gravatar of Christian List Christian List
    21. January 2017 at 14:11

    he blamed his political enemies and characterized them as Neanderthals.

    Pretty accurate. I remember an interview by him with one of his buddies from the New Yorker on why he lost the midterm elections in 2010. They finally agreed that it must have been because of racism. The same people that elected him in 2008, turned into racists in 2010, only to elect him again in 2012.

    The same with people from states like Wisconsin in 2016. They elected Obama two times, only to turn to Trump in 2016. What’s the “reason” for this according to the media? Racism and bigotry of course.

  41. Gravatar of dtoh dtoh
    21. January 2017 at 15:42

    I think we all have different models we build in our brain to explain things. The models are simplified, i.e. involve short cuts, that make it easier and quicker to figure things out. We do this in part to arrive at factual conclusions more quickly, but a lot of it is also to square the real world with our own egos and prejudices and to justify the contradictions and hypocrisy of our own actions and beliefs.

    It’s easy to convince people if you can do it in a way which is consistent with their existing models, but convincing people of things which requires a modification of their models is very hard, because a) changing the model requires a LOT of cognitive work, and b) it often requires a painful self-examination.

    And BTW – If you can’t explain something succinctly, it’s usually because you’re using a defective model.

  42. Gravatar of Lawrence D’Anna Lawrence D'Anna
    22. January 2017 at 09:09

    I think the word you’re looking for is “ineffable”

  43. Gravatar of ssumner ssumner
    22. January 2017 at 11:07

    Art, Here’s one case where you are speaking to the converted. But as you know a lot of people have powerful conspiracy complexes when it comes to things like the CIA (or the Rothchilds, or other such bogeymen.)

    Dan, You are pretty hopeless. I really don’t know why you comment here.

    dtoh, The problem isn’t constructing simple explanations (I can do that), the problem is constructing simple explanations that convince others. Especially other well informed people.

    Ley me know when you’ve found someone able to do that in macroeconomics. 🙂

  44. Gravatar of ssumner ssumner
    22. January 2017 at 11:07

    Lawrence, Close, but that’s not quite it. I would emphasize the complexity aspect, not the “difficult to put into words” aspect.

  45. Gravatar of ssumner ssumner
    22. January 2017 at 11:15

    Saturos, Those were great, but even I didn’t quite finish. I think I read about 100. Some were over my head.

    Rick and Bob, Interesting comments.

    Student, Some of that stuff just leaves you shaking your head.

  46. Gravatar of Dan W. Dan W.
    22. January 2017 at 12:54


    How ideological must a person be to commute the sentence of a treasonous soldier? When Obama did that whose interest was he representing? It wasn’t the nation’s interest nor the interest of our allies. It was done for selfish reasons and to placate a small but vocal minority.

    And that is how Obama governed for 8 years and why he has left the Democratic party in ruins. Obama devoted great energy to issues of great importance to a minority of people and otherwise dismissed the concerns of the majority.

    Trump told those dismissed by Obama that he would represent their interests. They gave Trump their vote and Trump won the election.

  47. Gravatar of dtoh dtoh
    22. January 2017 at 15:50

    You won’t convince well informed people because well informed people have already invested lots of cognitive effort into developing their own mental models that already work pretty well. As I said, it takes a LOT of work to change your model. Also it means repudiating your previously held beliefs (i.e. admitting that you were wrong), which most people don’t like to do.

    These are exactly the reasons you are unwilling to consider an alternative transmission mechanism for monetary policy.

    When you’re unwilling to re-examine your own beliefs, why would you expect anyone else to do so?

    You should focus your arguments are smart people who are not yet well informed. Being right doesn’t mean shit if it doesn’t accomplish anything useful. You might as well use your cognitive abilities to do crossword puzzles.

  48. Gravatar of dtoh dtoh
    22. January 2017 at 16:07

    That may have sounded harsher than intended. I actually think you’ve been incredibly effective at getting a lot of people to rethink monetary policy.

  49. Gravatar of Scott Sumner Scott Sumner
    23. January 2017 at 08:42

    Dan, You said:

    “How ideological must a person be to commute the sentence of a treasonous soldier?”

    It’s hard to imagine a less ideological action, unless you think Obama supports treason. Oh wait, you are Dan!

    dtoh, I was going to respond, but you second comment did it for me. Name one economic blogger who has been more successful in getting pundits to rethink their views of monetary policy. Maybe Krugman, because he has a vastly larger audience, but I’ve been successful beyond my wildest dreams. So why change to an approach that has not worked for other bloggers?

    You’ve occasionally offered alternative ways of selling my message, I’ve looked at your proposals, and determined they would not be persuasive to the people I am trying to convert.

  50. Gravatar of dtoh dtoh
    23. January 2017 at 10:05

    I stand by my first comment. Does the Fed have an NGDPLT? Then you have NOT been successful.

    I actually think you have a good but sub-optimal model. If you had a better model, you might succeed.

    You also tend to get sidetracked.

    On the off chance that someone in the White House reads your blog, why antagonize and possibly bias the person who picks the Fed Chair against NGDP targeting.

    BTW – Who are you trying to convert?

  51. Gravatar of Scott Sumner Scott Sumner
    25. January 2017 at 18:42

    dtoh, You said:

    “BTW – Who are you trying to convert?”

    People who influence monetary policy (a group that does not include Trump and his advisers.)

  52. Gravatar of Dtoh Dtoh
    26. January 2017 at 00:27

    Trump doesn’t influence monetary policy? Come on be serious.

  53. Gravatar of Lorne Smith Lorne Smith
    30. May 2017 at 11:42

    Scott writes: “when I say “I know something” I mean that I believe I know it. But then what else could it mean?”

    So: Person A believes that s/he knows that economics is not a science; person B believes that s/he knows that economics is a science. Hence: here are two items of knowledge: both that economics is and that it is not a science. From a contradiction, anything follows.

Leave a Reply