Skip to content

Ed Jaynes outta control!

A commmenter points to a chapter of E. T. Jaynes’s book on probability and inference that contains the following amazing bit:

The information we get from the TV evening news is not that a certain event actually happened in a certain way it is that some news reporter has claimed that it did. Even seeing the event on our screens can no longer convince us after recent revelations that all major US networks had faked some videotapes of alleged news events.

Scientists can reach agreement quickly because we trust our experimental colleagues to have high standards of intellectual honesty and sharp perception to detect possible sources of error. And this belief is justified but only about once in a decade is an experiment reported that turns out later to have been wrong. . . .

In politics we have a very different situation . . . We are convinced that virtually all news reporting is selective and distorted designed not to report the facts but to indoctrinate us in the reporter’s socio-political views. . . .

I have no idea what are those fake videotapes that Jaynes was talking about. But, beyond that, I don’t know what’s more disturbing: that Jaynes thinks that wrong experiments are reported only about once in a decade, or that he thinks “virtually all news reporting” is designed to indoctrinate.

I’m reminded of the Chestertonian position that extreme skepticism is a form of credulity: Once you take the position that virtually all news reporting is propaganda, then you can’t trust anything you read, so you make your way through the world based on whatever things you hear that sound good to you.

P.S. Don’t get me wrong here. Jaynes’s book is an entertaining and thought-provoking read, and I recommend it to all of you. It’s just funny to see someone like Jaynes, who had such strong opinions on technical matters, go off the deep end like this.


  1. Anonymous says:

    It’s hard to date parts of that book since they could have been written anytime during the decades 60’s-90’s. His description of scientist is probably an accurate view of hard sciences (specifically physics and his field of quantum optics) that Jaynes saw during that time. He’s not describing social sciences circa 2017.

    It’s amusing thought that a political science professor things the quote “virtually all news reporting is selective and distorted” is going off the deep end.

  2. Jonathan says:

    I’m not sure what to say about Jaynes’ book but the idea of skepticism isn’t ‘disbelief’ but rather the questioning of perspective. To give a current news example, I live in Boston and am inundated with Roy Moore reporting but almost all of it is about his character and takes the perspective that Republicans are willing to excuse his character for reasons that have to do with party loyalty, hatred of Democrats, disrespect to women, not caring and then a line of more inflammatory allegations and insinuations. There is almost no mention that his opponent is adamantly pro-choice in what may be the most anti-choice state, nor any acknowledgment that to anti-choice people the issue isn’t one of political priorities but one of character and morality because they view abortion as government-sanctioned murder of innocent children. I’m not sure how much of that is intentional skewing to hide the issue versus how much the moment of women coming out about male sexual power abuse is creating a skewing in which the other perspectives are hidden. The latter is interesting because it’s so similar to what happens in so much research: it takes a perspective, constructs the arguments that support that perspective, downgrades or dismisses or doesn’t mention counter-arguments. I’d lump a lot of economics in this category: when every paper has its own model, you’d think they’d realize that all these models are just perspectives. And this gets worse the less reliable and deep the data sources, which means again most economics papers. (Do they understand the limits of controlling for variables?) The other half, that this is intentional, gets more into the softer ‘disciplines’ where absurd ‘creativity’ in service of a perspective becomes closer to the norm. An example is a recent ridiculous argument that Israeli attempts to avoid killing people is actually an intentional policy to maim rather than an attempt to avoid the larger harm of killing. The contrast between the two is that the Moore arguments may be driven by the current perspective cutting out other perspectives so the issue is distorted because of how people see current events, while the latter is the attempt to push a perspective on others. That latter perspective may be genuinely believed but it’s a creed, not a logical construct, of the sort that says Israel only does good things in order to hide bad things, which then feeds into the history of anti-Jewish beliefs about Jews having hidden power and agendas. There really is no difference between arguing ‘intent to maim’ and saying that Israelis only invent medical devices, etc. so they can hide their ‘intent to maim’. I’m taking this long to say this because the Moore case is that of a wave of response dominating perspective while the latter is the belief that all events must be interpreted as though they have this one perspective, as through everything is aimed at harming or hiding harm. The former is not only rational but expected. The latter is irrational if you credit them with acting in good faith belief. That describes much of the ‘soft’ part of academia: if they actually believe this nonsense, then they may be creative but they’re irrational in nearly any setting outside of academia. It reminds me of Galenist views about illness: hanging on despite their obvious failure. That’s an interesting subject, one you touch on with ESP, etc.: the way failure of results makes belief stronger. It’s as if when the results are negative, you focus more on what may be another approach that now is clearer to you, which makes the belief clearer to you, and that seems to make people believe more even though any external observer would say the confidence should be dropping. You see this all the time in politics: the 2 dozen or so extreme right-wingers who travel the country in a belief the nation will turn to them as chaos erupts are deluded and obsessional to any outsider but they believe that each failure is a positive step. They get so locked into their narrow context, they lose the larger threads. A lot of social science research is the same, though I don’t intend to compare them to storm trooper wannabes.

  3. Pointeroutguy says:

    I don’t think he’s saying that he personally believes all the tapes are faked. I think he’s stubby that many people distrust the news, and many are willing to go so far as to dismiss video evidence contradicting their political view as faked (the recent post on lack of trust and the ad nauseam religion of “fake news” seem to support this).

    • Bobo says:

      No one is claiming that he said that “all the tapes are faked”. But he did say some of them are faked, without saying which.

      Anyway, Jaynes’s comment about science only being wrong once or twice in a generation is insane.

  4. Anoneuoid says:

    I think Jaynes was using “science” == physics for the “once per decade” comment. Think about projects run at LIGO, or the LHC, his was thinking of stuff like that. I would still love to know what he was referring to regarding the faked videotapes. Can anyone speculate on this?

  5. Terry says:

    I don’t know what’s more disturbing: that Jaynes thinks that wrong experiments are reported only about once in a decade

    I have to agree with this. It is astonishing that even extreme skeptics like Jaynes can have such a naïve view of science.

    I’m not sure exactly what Jaynes means when he refers to “recent revelations that all major US networks had faked some videotapes of alleged news events”, but, if “faking” a video includes “staging” a video, then his claim is pretty clearly true. Here are a couple just off the top of my head. (Note that Jaynes only refers to “some” faked videotapes, not that all or even most are faked.)

  6. a reader says:

    Haha, I thought for sure that the lines you thought was off the deep end was

    “Scientists can reach agreement quickly because we trust our experimental colleagues to have high standards of intellectual honesty and sharp perception to detect possible sources of error. And this belief is justified but only about once in a decade is an experiment reported that turns out later to have been wrong. . . “.

  7. Jonathan (another one) says:

    I’m not sure exactly what he’s referring to either, but there was a lot of scandal in this period relating to Michael Born, a German news producer who produced a bunch of fake news reports. It was covered on 60 Minutes and is discussed in this documentary (which I haven’t seen.) So it definitely was a topic of conversation in the 90’s.

  8. Carlos Ungil says:

    I think by “scientists” he means researchers in “hard” sciences, not in fields where p<0.05 is seen as a positive result.

    • Carlos Ungil says:

      By the way, you’re leaving out part of the quote that suggests that “virtually all reporting” refers to “economic, social, or political topics”. The full quote:

      “In politics, we have a very different situation. Not only do we doubt a politician’s promises, few people believe that news reporters deal truthfully and objectively with economic, social, or political topics. We are convinced that virtually all news reporting is selective and distorted, designed not to report the facts, but to indoctrinate us in the reporter’s socio-political views. And this belief is justified abundantly by the internal evidence in the reporter’s own product – every choice of words and inflection of voice shifting the bias invariably in the same direction.

      “Not only in political speeches and news reporting, but wherever we seek for information on political matters, we run up against this same obstacle; we cannot trust anyone to tell us the truth, because we perceive that everyone who wants to talk about it is motivated either by self-interest or by ideology. In political matters, whatever the source of information, our prior probability for deception is always very high.”

    • Dan F. says:

      Jaynes comment about experiments that turn out to be wrong has to be interpreted – he isn’t calling “experiment” just any old random claim, rather something like an experiment purporting to demonstate the existence of magnetic monopoles, the sort that gets someone made professor at a fancy place like Stanford or Chicago – and he’s talking about physics. He certainly isn’t thinking about claims that people eat more pizza when they sit next to the window, because he would have just discarded that out of hand as nonsense. When people with a certain level of criteria/seriousness speak they apply a certain a priori filter and one has to apply the same filter to understand what they are saying.

  9. Terry says:

    In politics we have a very different situation . . . We are convinced that virtually all news reporting is selective and distorted designed not to report the facts but to indoctrinate us in the reporter’s socio-political views. . . .

    Is this really “going off the deep end”?

    For large chunks of the news, this seems pretty undeniable, especially with regard to the charge of being “selective”. News organizations constantly hype one politically helpful story while ignoring other politically unhelpful stories.

    Perhaps it is an exaggeration to say that “virtually all” reporting is selective and distorted, but I don’t see there is any denying that a large amount of reporting is selective and distorted.

    Indeed, if a news organization simply doubles the space it gives to stories it likes and halves the space it doesn’t like, then in a very real sense, it is literally true that the organization’s news is selective and distorted in its totality.

    • phayes says:

      It’s not all distortion by simple (partisan) design, either.

      • Terry says:

        Thanks for the link.

        The same phenomenon seems to come up whenever numbers are involved – not advanced math or statistics, but simple concepts like A is bigger than B, percentages, or whether a trend is upward or downward. Numbers seem to be a foreign language to most reporters and they seem to just repeat whatever people tell them with some sort of good/bad spin on it.

        Dean Baker has a worthwhile blog about this at

      • Martha (Smith) says:


        Thanks for the link. I’m not knowledgeable enough about economics to follow all that the Wren-Lewis paper says, but it seems to make a convincing case that at the very least the press is too oriented toward certainty (which does not surprise me, since so many people are).

        Also interesting/ironic: There is so much complaining from anti-evolution people about the press not giving “equal time to the other side,” but it sounds like the press has definitely not given “equal time to both sides” when it comes to economics (at least the situation described in the link).

        • Kyle C says:

          There is an entire, well developed, mathematically rigorous point of view in academic economics that government budget deficits (essentially) do not matter in a country like the U.S., which prints a world reserve currency and has extensive foreign trade, but you will not find a single newspaper, radio, or tv journalist who even admits that this position exists, much less who will take it seriously.

          • Martha (Smith) says:

            Wow. Sounds like the university PR departments need to spend more time with folks from economics (and probably less with folks from psychology) — but maybe they just want the “sexy” stuff. Hmm, maybe economics folks should try to get into TED talks?

  10. ojm says:

    Jaynes’ book is an entertaining read and has some good stuff, but it’s also pretty off the deep end on the whole…

    • Hernan Bruno says:

      Jaynes seems like a huge intellect. But the book is a bit of a mess. I don’t think people can learn from it if they don’t know the material already.

      • Hernan Bruno says:

        Ok. That comment was hard. The book has many interesting and entertaining points. I just find the exposition a little too idiosyncratic. I would recommend anyone to learn statistics and probability from a standard textbook before trying to re-learn it a la Jaynes.

    • Dan F. says:

      One should keep in mind that the book was published posthumously. There exist(ed) other versions of the book, some of them quite different. In any case its author never completed the book himself.

      Not sure why it should be qualified as “off the deep end”. It is far more coherent than most of what is written about probability by statisticians. The central idea is to view probability as rules for logical inference. His point of view may be hard for statisticians to understand if they understand little physics, since the principal context for applications he has in mind is physics (moreover a certain sort of physics – he is not thinking much of astronomy).

  11. Don’t know why you should find it disturbing or surprising that Jaynes should think that “‘virtually all news reporting’ is designed to indoctrinate,” or perhaps just think that many people believe this. I’ve seen a number of comments from journalists in which they take it for granted that their job is to tell people what to think. Take, for example, this quote from Gregory Ferenstein:

    “In the golden era of the gatekeeper model, journalists could just scrub the news of ideas they found offensive and quickly marginalize “abnormal” candidates. But the internet eradicated the mainstream media’s gatekeeper power. Today, anyone can publish an idea and, if it’s popular, will garner an overwhelmingly powerful following. In this new media world, I believe we’re going to have to practice journalism through persuasion rather than censorship.”

    Note that he *equates* journalism with opinion molding here.

    Then we have MSNBC’s Mika Brzezinski complaining that Trump “could have undermined the messaging so much that he can actually control exactly what people think. AND THAT, THAT IS OUR JOB.” (Emphasis added.)

  12. Roger says:

    TV news does have a lot of fake news, and attempts to indoctrinate us in the reporter’s socio-political views. There are many current news story where is it not even clear that the reporter believes the truth of the events in the story. For example, there are stories that report that someone is alleging some misbehavior many years ago, and there is no way to determine whether the story is true or false. Siding with the accusers has become a political statement, whether any of it is true or not.

  13. Jim Quirk says:

    The link in the article should probably go to the completed version of the book (on Amazon or similar).

    Here are the ISBN numbers for the curious/interested:

    ISBN-13: 978-0521592710
    ISBN-10: 0521592712

Leave a Reply