Skip to content
 

Gathering of philosophers and physicists unaware of modern reconciliation of Bayes and Popper

Hiro Minato points us to a news article by physicist Natalie Wolchover entitled “A Fight for the Soul of Science.”

I have no problem with most of the article, which is a report about controversies within physics regarding the purported untestability of physics models such as string theory (as for example discussed by my Columbia colleague Peter Woit). Wolchover writes:

Whether the fault lies with theorists for getting carried away, or with nature, for burying its best secrets, the conclusion is the same: Theory has detached itself from experiment. The objects of theoretical speculation are now too far away, too small, too energetic or too far in the past to reach or rule out with our earthly instruments. . . .

Over three mild winter days, scholars grappled with the meaning of theory, confirmation and truth; how science works; and whether, in this day and age, philosophy should guide research in physics or the other way around. . . .

To social and behavioral scientists, this is all an old old story. Concepts such as personality, political ideology, and social roles are undeniably important but only indirectly related to any measurements. In social science we’ve forever been in the unavoidable position of theorizing without sharp confirmation or falsification, and, indeed, unfalsifiable theories such as Freudian psychology and rational choice theory have been central to our understanding of much of the social world.

But then somewhere along the way the discussion goes astray:

Falsificationism is no longer the reigning philosophy of science. . . . Nowadays, as several philosophers at the workshop said, Popperian falsificationism has been supplanted by Bayesian confirmation theory . . . Bayesianism allows for the fact that modern scientific theories typically make claims far beyond what can be directly observed — no one has ever seen an atom — and so today’s theories often resist a falsified-unfalsified dichotomy. Instead, trust in a theory often falls somewhere along a continuum, sliding up or down between 0 and 100 percent as new information becomes available.

Noooooooooo! As Cosma Shalizi and I argue, “the most successful forms of Bayesian statistics do not actually support that particular philosophy [of inductive inference] but rather accord much better with sophisticated forms of hypothetico-deductivism.”

As reporting, the article is excellent. But it seems that nobody in the room was aware of the “Bayesian Data Analysis” view in which Bayesian models can and should be checked. We have to do more communicating. Naive Popperianism is bad, but so is naive anti-Popperianism. Lakatos would be spinning in his grave.

75 Comments

  1. Anoneuoid says:

    >”Lakatos would be spinning in his grave.”

    Wouldn’t he just say this is a clear sign of a degenerating research programme? By definition, untestable theories will cease to predict novel observations.

    • Andrew says:

      Anon:

      I’m saying Lakatos would be spinning at the notion that “Falsificationism is no longer the reigning philosophy of science . . . Popperian falsificationism has been supplanted by Bayesian confirmation theory.” I don’t think this would make him happy.

      • Anoneuoid says:

        I think he rejected both decades ago in favor of novel predictions, so he would probably facepalm:

        “To sum up. The hallmark of empirical progress is not trivial verifications: Popper is right that there are millions of them. It is no success for Newtonian theory that stones, when dropped, fall towards the earth, no matter how often this is repeated. But so-called ‘refutations’ are not the hallmark of empirical failure, as Popper has preached, since all programmes grow in a permanent ocean of anomalies. What really count are dramatic, unexpected, stunning predictions: a few of them are enough to tilt the balance; where theory lags behind the facts, we are dealing with miserable degenerating research programmes.”

        http://crl.ucsd.edu/~ahorowit/lakatos.pdf

        But then wouldn’t he predict adherents to a degenerating program would adopt a philosophy that justifies their behavior? Did he ever discuss this anywhere?

        • Andrew says:

          Anon:

          I’m not saying that Lakatos would be surprised that adherents of an unfalsifiable theory would choose to move away from falsificationism; I just think it might make him unhappy to see this happening.

          So: spin-in-grave because unhappy, not spin-in-grave because shocked or surprised.

          For example, when, in 2096, Psychological Science runs a special issue entitled “100 Years of Embodied Cognition” featuring an interview with the hologram of John Bargh, several small-N studies on ESP, evolutionary psychology, and correlations among clothing choices and political attitudes among a sample of college sophomores, and sponsored by the Anil Potti Foundation and Marc Hauser Institute, a latter-day blogger can feel free to say “Andrew Gelman would be spinning in his grave.” I wouldn’t be surprised by the appearance of such a special issue but it would still make me unhappy.

  2. David says:

    I don’t see how the quote from your article disputes anything from the cited paragraph. Wolchover said nothing in that paragraph to indicate that Bayesian confirmation is supported by Bayesian statistics. He just said that it is the reigning philosophy of confirmation, which it is.

    Given that, this seems like nothing more than a complaint that ‘Bayesian’ means different things in statistics and in philosophy of science/epistemology. Hardly worth pulling a Darth Vader over.

  3. lewis55 says:

    The real story here is that different disciplines don’t talk to each other enough, such that advances / changes in understanding of terminology in one discipline can take decades to (or never) move over into another discipline, even though we’re working on a related issues. But that’s not anything specific about Bayesian stats as currently practiced.

    Ironically, your realization that these disciplines haven’t been paying enough attention to your discipline only came about because they held an interdisciplinary conference to try to talk better across disciplines (but happened to have left yours out).

    Each corner of the university (that deals with science at least) probably has a slightly different understanding of this same debate, all of which differ in some way probably from the sentiments in the article.

    Alas, the silo’ing of modern universities…

    • Andrew says:

      Lewis:

      Yes, I agree that the conference was a net plus in that it resulted in this news article which made me aware of these discussions, then I blogged it and maybe some physicists will become more aware of the Bayes-Popper synthesis, etc.

      I was upset to see that these conference participants espoused what I consider an outdated view of statistics and the philosophy of science. But, conditional on these attitudes existing, I’m glad to see these outdated attitudes aired, so that there’s a chance to move forward.

      • Keith O'Rourke says:

        > outdated view of statistics and the philosophy of science.

        Where is the vibrant _community of scholars_ working on new views of statistics and the philosophy of science and where do they regularly publish?

        What answer do you think most Phd students in statistics would get from their supervisors?

        • Andrew says:

          Keith:

          Hmmm, where’s the vibrant community of scholars? I’d start with this blog, also Mayo’s: even if I disagree with some of her emphases, there’s some thoughtful discussion. Also lots of discussion on sites such as Scott Alexander’s that seem to be frequented mostly by non-academics. One challenge is for academic researchers such as myself to remain aware of the concerns and interests of the general educated public, and conversely for the members of that public, such as Scott Alexander and his readers, to engage with the relevant academic literature. News articles such as Wolchover’s are a start, and that’s one reason I get upset, that a reader of such an article can get a misleading picture of what Bayes can be.

          • Mayo says:

            Thanks for a shout out. Philosophy of statistics is an important field, it used to be a great one, pursued just as much by philosophers as by statisticians and practitioners. It dissipated around 15 years ago (broad guesstimate), perhaps after Colin Howson and some other subjective Bayesians moved away and settled on some logical work. (A bit after my Error and the Growth of Experimental Knowledge Mayo 1996). Formal epistemology takes place largely along those lines, but there are some pockets here and there of philosophy of statistics, and related areas (e.g., causal modeling).

            I reblog this post here. http://errorstatistics.com/2015/12/17/gelman-on-gathering-of-philosophers-and-physicists-unaware-of-modern-reconciliation-of-bayes-and-popper/
            The question of the nature and justification of statistical falsification is a crucial issue. It’s also very much an open one in need of clarification and justification. The current confusion about the value and nature of significance levels and other error probabilities, I claim, revolves around a confusion between statistical confirmation (probabilism, whether comparative or not) and the essence of statistical falsification.

            http://errorstatistics.com/2015/12/17/gelman-on-gathering-of-philosophers-and-physicists-unaware-of-modern-reconciliation-of-bayes-and-popper/

          • Keith O'Rourke says:

            First:
            > I get upset, that a reader of such an article can get a misleading picture of what Bayes can be.
            I got upset about clinical reviewers _planning_ to approve approve drugs on the same misconceptions.

            But to understand why it happens (anticipating how to fix it) we need to discern how they come by less than ideal conceptions.

            > this blog, also Mayo’s: even if I disagree with some of her emphases
            OK, but reading blogs requires a fair amount of background knowledge to critically understand whether one is getting ideal conceptions (blogs are about the blogger’s viewpoint as usually are [understandingly] saying “I am smarter than others and you should listen to me”).

            For instance, in your paper discussed on this blog “Beyond subjective and objective in statistics” you [implicitly] identify your view of the philosophy of science with Peirce – “We are therefore “active scientific realists” in the sense of Chang (2012)” [Chang is clear that this is drawn from Peirce and a Peircian scholar]. On Mayo’s site, Peirce is characterized as completely rejecting Bayes outright. Now occasionally, on her website, your perspective mentioned as being a positive positive variation on Bayes. But expecting folks with only a few courses in statistics or the usual Phd to sort this out.

            And taking a blog with this subtitle – DISPATCHES FROM WEIRD PLATONIC SPHERICAL COW PERFECT RATIONALITY OUTSIDE VIEW WORLD as being where one gets ideal conceptions [Scott Alexander]…

            Blogs are helpful but likely more is needed to prevent less than ideal conceptions – widely.

            • Christian Hennig says:

              Keith: Embracing Chang’s active scientific realism doesn’t necessarily mean embracing all of Chang, or all of Peirce that Chang likes, let alone all of Peirce.

              • Keith O'Rourke says:

                Christian:
                By “identify your view with” I did not mean “embrace all of” – perhaps “identify” with was a poor choice of wording.

                But your view is connected through Chang to Peirce’s work which is connected to Peirce inclining to a preference for a frequentist approach. There is ongoing debate whether his arguments would incline more to Bayes with sophisticated forms of hypothetico-deductivism – though on the surface it seems like some of the arguments are miss-aligned – and they may well be.

                Maybe not the best example of why those outside statistics/philosophy have trouble recognizing the most informed view to incline to.

              • Christian Hennig says:

                OK, my previous comment was rather offhand, but can you clarify this a bit? I knew Chang was influenced by Peirce and I knew that Peirce was something of a frequentist, but I don’t know how the latter is connected to Chang’s thoughts about active scientific realism, which may be traced back to Peirce at least in parts. Is there some argument relating these?

              • Keith O'Rourke says:

                > Is there some argument relating these? [active scientific realism to frequentist]
                First, in you paper’s quote from Chang (below), I do not see anything that was not in Peirce.

                Peirce’s logic (philosophy) was normative – how we ought to best conduct inquiry and his inclination to a frequentist approach (early 1900,s) _would_ have reflected those considerations (if he adequately considered it).

                Why would one think these are unrelated or contradictory?

                Chang (2012), who writes: “I take reality as whatever is not subject to
                one’s will, and knowledge as an ability to act without being frustrated by resistance from reality.
                This perspective allows an optimistic rendition of the pessimistic induction, which celebrates the
                fact that we can be successful in science without even knowing the truth. The standard realist
                argument from success to truth is shown to be ill-defined and flawed.”

        • fred says:

          Keith: those are good questions – my guesses at the answers;

          Where is the vibrant community of scholars working on new views of statistics and the philosophy of science?

          There are a few individuals, but it’s not a well-organized field. See e.g. recent JSM session topics for evidence this is not a popular area.

          Where do they regularly publish?

          Mostly, nowhere most statisticians will see their papers. Exceptions exist – e.g. a few papers in Stat Sci – but those papers are written by old-timers, not PhD students.

          Given the premise about a student asking their supervisor, it’s likely answers would also include something like “that’s interesting but you need to graduate in X years and get a job” and “we’re funded to come up with some methods that work for analysis of X’s data, so let’s focus on that”.

    • Martha says:

      The lack of communication between different sciences really came to my attention a few years ago when I was appointed to a committee to develop a science course for prospective elementary school teachers. The committee consisted of one physicist, one chemist, two biologists, two geologists, one person who gave inservice science training to teachers, and one mathematician (me). We had to spend a couple of months of weekly meetings just discussing what science is — different disciplines had quite different views, but generally stemming from the nature of the different sciences (e.g., experimentation is rarely possible in geology.)

  4. Anonymous says:

    Reconciling Bayes and popper is like Russia and Moldavia signing a mutual defense pact.

  5. Greg says:

    Richard Dawid deserves credit for developing what seems to be the best justification of the practices of contemporary theoretical physics that we have. Can “the BDA view” do it better, or does it just suggest that those practices are misguided for the familiar Popperian reasons?

    • Andrew says:

      Greg:

      I do suspect that the BDA approach would be useful in theoretical physics, at least for those parts of theoretical physics that make predictions about observables. I don’t know theoretical physics so I’ll leave that to others to see how that would work.

      • Greg says:

        I agree! But we need something to guide theoretical work on quantum gravity, where we can’t collect relevant data, unless we are to refrain from such theorizing entirely (as distinct from engaging in such theorizing but regarding it as speculative). Dawid has an account that can do this work. It is couched within a framework that may be inferior to yours in cases in which both apply, but in the kinds of cases Dawid is addressing it seems to be the only game in town.

        • ojm says:

          As an outsider, string theory seems to use the same sort of mathematical and physical reasoning that physicists have used for decades. I don’t see why there’s so much controversy about it tbh. The popular hype is a bit tiresome of course.

          BTW the BDA view gives a nice role to external checks when possible, internal self-consistency checks and hierarchical thinking, so it at least doesn’t obviously contradict (to me) what string theorists do.

          • Rahul says:

            But does string theory make any empirical claims that are falsifiable? What other major Physics theory is like that?

            • ojm says:

              Like I said, I’m an outsider. But presumably quantum field theory has been falsified in the sense that it cannot describe gravity consistently. Hence the need for something like string theory…

              • ojm says:

                To stick with the theme – a model elaboration/expansion in the BDA sense. The price of expanding a model to remove an inconsistency is typically greater undetermination. There might be bounds on underdetermination (in both directions) i suppose.

  6. Olav says:

    Andrew, I think there is a fundamental difference between how you view Bayesianism and how many philosophers view it. For you, a Bayesian probability distribution (i.e. a joint distribution over parameters and evidence) is first and foremost a pragmatic device that allows you to handle measurement uncertainty and take into account background information in a principled way. Your beliefs are relevant when you are constructing a distribution, but the resulting distribution does not literally represent your degrees of belief. To you, Bayesianism is justified because of its success in statistical problems, and any worthwhile Bayesian philosophy ought to make sense of and be informed by Bayesian statistical practice. Bayesian falsificationism, as I understand it, is an attempt at this.

    Many (probably most) philosophers, on the other hand, start from the other end. To them, degrees of belief are the primary objects of interest, and the main project is to have a theory of rational belief change. Bayesian probability distributions are only of interest because there are arguments that purport to show that degrees of belief ought to obey the probability axioms. But if there were some other mathematical framework that allowed for a better theory of rational belief change, then many philosophers would ditch Bayesianism immediately, regardless of how successful Bayesian statistics is (and indeed, there are plenty of philosophers who do work in other frameworks than Bayesianism, e.g. Ranking Theory). They don’t see their project as making sense of Bayesian statistical practice. In that sense, Bayesian falsificationism isn’t really a genuine alternative to Bayesian confirmation theory, because Bayesian falsificationism dispenses with the (to many philosophers) vital and all-important identification of probabilities with degrees of belief.

    • Andrew says:

      Olav:

      I see what you’re saying, but on the philosophical level I don’t think it makes sense to talk about things like Pr(Hypothesis A is true). So to the extent that Bayesian confirmation theory is about watching the probabilities of different hypotheses go up and down, I think it’s a fundamentally mistaken approach (even though, of course, it can be useful in many settings, I don’t deny that). And I’d urge the physicists who are thinking in those ways to move toward a falsificationist Bayesian perspective.

    • Rahul says:

      +1

      That’s a very key point.

      I think thousands of pages of redundant arguments are posted because people use the term “Bayesianism” loosely & the two sides really mean totally different senses of it when they argue.

    • Keith O'Rourke says:

      Olav:

      With regard to the primary objects of interest regarding questions of being rational – is this not just a major division in philosophy itself as to whether being rational needs to be informed by and responsive to practice?

      • Olav says:

        Keith, that is one major division in philosophy, indeed, though I hasten to add that many Bayesian philosophers of science are interested in scientific practice, just not necessarily statistical practice. For example, here is a paper that tries to give a Bayesian justification of a form of argument (supposedly) found in theoretical physics: http://www.laeuferpaar.de/Papers/NAA-10.pdf Indeed, I think this paper is a paradigm example of what the people in Andrew’s original post probably mean by “Bayesian confirmation theory.”

  7. Why were Einstein, Feynman, Newton, Galileo, Copernicus et al. banned from the abstracts, papers, presentations, and coverage of the Munich conference titled, Why Trust a Theory? Reconsidering Scientific Methodology in Light of Modern Physics. I couldn’t find one single quote from any of the Greats concerning the nature of science. I have included many great quotes below, so that you might be able to share them with your students and at future conferences.

    It is almost as if the fake pseudoscience of String Theory/Multiverse Mania needs fake experiments (BICEP2 dust observations) and fake philosophers and philosophies so as to keep the billions of taxpayer dollars flowing into the massive anti-science hoax. It is almost as if the multimillionaire multiverse maniacs are denying the bright light of Einstein, Feynman, Newton, Galileo, and Copernicus so that the anti-Philosophy, anti-Physics, talentless televangelist hacks can create the largest, darkest version of Plato’s cave ever known to mankind–a dark, sordid cave where belligerent bloggers shall rule the multi-billion-dollar industrial hype complex for all of eternity via phony press releases and legions of useful idiots, exiling anyone who actually ventures beyond the cave into the Light (dx4/dt=ic).

    When it comes to defining science and scientific theories, should not the words and thoughts of Feynman, Newton, Galileo and Einstein be included?

    Should not the words, thoughts, and ideas of the likes of Einstein, Feynman, Maxwell, Faraday, Newton, Galileo, and Copernicus be included front and center in books and at conferences contemplating the character of physics and science?

    Might anyone know exactly when Brian Green’s lie(s)–”Superstring theory successfully merges general relativity and quantum mechanics”–will be removed from Einstein’s noble book The Meaning of Relativity?
    Please be sure include the spirit and soul of Einstein, Feynman, Maxwell, Faraday, Newton, Galileo, and Copernicus in future conferences, books, and publications!

    The words of the Greats allow us to see String Theory for what it is–a failed hoax by which tens of thousands of pseudoscientists, televangelists, and faux philosophers have profited massively at the expense of the taxpayer, student, physics, and physicists, writing blurbs for one-another’s hype-based books while occupying the university and transforming it into the darkest version of Plato’s Cave ever known to mankind, during their ruthless reign of darkness and deceit.

    • Anoneuoid says:

      The best you can do is point to specifically what questions you have regarding string theory (or string theories). Everyone already accepts it does not appear to lead to any predictions, which is surely a black mark against it. I mean, 28 billion/year is claimed to be wasted on medical research just due to failure of researchers to accurately describe what they did.[1] Is string theory a big deal in the face of that, the consequences of which you are much more likely to come in contact with?

      Also, arguments from authority don’t really help your cause, on the other hand those people often do yield insightful quotes so they are worth reading.

      [1] http://www.nature.com/news/irreproducible-biology-research-costs-put-at-28-billion-per-year-1.17711

      • Rahul says:

        I look at the 28 billion/year as an incidental cost of doing business.

        String theory, I see as a business we shouldn’t be into, in the first place.

        • Chris G says:

          Wolchover described some laughably bad reasoning – paraphrasing – “We’re really smart and can’t think of anything else so it must be true.” Seriously?

          > String theory, I see as a business we shouldn’t be into, in the first place.

          I did molecular physics for my Ph.D. I loved the science. I never got the appeal of string theory. (It seemed to the physical sciences what postmodern critical theory was to the humanities.) For me the whole appeal of science is: Write an equation and make a prediction. Collect data to test your prediction. If model and data are in agreement then great, now make a more daring prediction. If the model is inconsistent with the data then use the observed deviations to improve your understanding of how the world works, refine your model and go back to Step 1.

      • I recently began a cordial discussion with Niel Turrock, Paul Steindhart, and the Princeton University Press:

        Thanks for the wonderful conversation all!!

        I was wondering if Niel Turok, Paul Steindhart, Brian Green, and the Princeton University Press had had a chance yet to remove Brian Green’s lies from the introduction to Einstein’s exalted The Meaning of Relativity (http://www.amazon.com/Meaning-Relativity-Including-Relativistic-Non-Symmetric/dp/0691164088/), where Brian Green writes:

        “Superstring theory successfully merges general relativity and quantum mechanics.”

        No it doesn’t. :) This statement is a mistake/lie/hype and the time has arrived for such statements to be removed from the introduction to Einstein’s epic, exalted work.

        A problem with including such lies in the introduction to Einstein’s seminal work is that ambitious third-generation-failed multiverse maniacs see the lies and then reason that they too can score millions of dollars, tenure, and TV shows by exalting lies over truth and pseudoscience over science. And they’re right. The unscientific, dishonest multiverse maniacs are rewarded most generously by the decades-old, billion-dollar regime of failure, while true science (dx4/dt=ic) is exiled alongside the very soul of science–the honorable yearning for simple Truth and Beauty.

        It is always difficult to determine that which the failed multiverse maniac hates more–science, god, truth, beauty, or philosophy. But make no mistake–their massive ego must destroy them all so that they might rule supreme via youtube videos and the billion-dollar industrial pseudoscience hype machine.

        The relentless stringy/multiverse/inflation hype and lies over the past several decades have been rewarded with billions of taxpayer/foundation dollars and billions more in overhead. True physicists and scientists with a soul have been exiled and disappeared from the academy. The foundationally dishonest, failed regime founded by Green, Witten, Guth, Linde, et al. has transmogrified the physics landscape so that the only way to rise in theoretical “physics” is to hype the failed programs of senior pseudoscientists. And thus true Scientists, Noble Philosophers, and Physicists are exiled.

        A great thing about youtube/videos is that their extreme hubris and idiociy are being recorded for future generations to marvel at. Here we see one of the regime’s useful idiot drones claiming that BICEP2 “is the most exciting example” of inflation theory being tested, while also claiming that inflation theory actually predicts parallel universes, whereas, in reality, inflation theory is not a theory and makes no predictions:

        http://teachingexcellence.mit.edu/must-see/the-bicep2-results-and-what-they-mean-part4
        https://www.youtube.com/watch?v=bJpIclDmi2M
        https://www.youtube.com/watch?v=w0uyR6JPkz4
        (looking for “bruises” in the CMB is not a test of anything other than “bruises” in the CMB)
        https://www.youtube.com/watch?v=ZlfIVEy_YOA
        (celebrating fake results for a fake theory–this is what science has come to)

        These videos need to be replaced with the fact that inflation is not science as it cannot be tested. Or at least MIT could open up the comments and allow the Truth be shared. But instead, tens of millions of more dollars will be wired to the pseudoscientists for serving decades-old failed regimes of darkness, lies, hype, and passionate deceit. And things fall apart as The best lack all conviction, while the worst Are full of passionate intensity.

        Turning and turning in the widening gyre
        The falcon cannot hear the falconer;
        Things fall apart; the centre cannot hold;
        Mere anarchy is loosed upon the world,
        The blood-dimmed tide is loosed, and everywhere
        The ceremony of innocence is drowned;
        The best lack all conviction, while the worst
        Are full of passionate intensity.

        Surely some revelation is at hand;
        Surely the Second Coming is at hand.
        The Second Coming! Hardly are those words out
        When a vast image out of Spiritus Mundi
        Troubles my sight: a waste of desert sand;
        A shape with lion body and the head of a man,
        A gaze blank and pitiless as the sun,
        Is moving its slow thighs, while all about it
        Wind shadows of the indignant desert birds.

        The darkness drops again but now I know
        That twenty centuries of stony sleep
        Were vexed to nightmare by a rocking cradle,
        And what rough beast, its hour come round at last,
        Slouches towards Bethlehem to be born?

        Two chapters in Nobel Laureate F.A. Hayek’s book The Road to Serfdom are “When the truth ends,” and “Why the worst rise to the top.” Well, many of us went into science because, like Newton, Faraday, Maxwell, and Einstein before us, we loved Truth and Beauty. When the failed string regimists disappeared Truth and Beauty from the academy, they had to disappear many of us, while wiring tens of millions of dollars to their useful idiots.

        The regime has thus spawned legions of third-generation failed physicists, egomaniacs, and multimillionaire multiverse maniacs who are now raging against science and the scientific method itself, all so that they can keep their masters’ perpetual-money machine going as they advance in their own careers as televangelists of pseudoscience, lies, hype, and deceit.

        https://edge.org/response-detail/25322 (Yes–in order to rise in today’s world of pseudoscience, one must rage against Falsifiability, and thus experimental tests, empiricism, physical reality, and physics.

        This is the world which Niel, Paul, et al. have created. They could change it if they wished, and remover Brian’s lies from the introduction to Einstein’s The Meaning of Relativity, as they have the power and position. But will they?

        Please remove Brian Green’s lies from the introduction to Einstein’s Meaning of Relativity. Please replace the lies with the wisdom of Copernicus, Newton, Einstein, and Galileo, which I include below. A great place to start would be by presenting Einstein’s thoughts on Copernicus, Newton, and Galileo.

        Albert Einstein: Once it was recognised that the earth was not the center of the world, but only one of the smaller planets, the illusion of the central significance of man himself became untenable. Hence, Nicolaus Copernicus, through his work and the greatness of his personality, taught man to be honest. (Albert Einstein, Message on the 410th Anniversary of the Death of Copernicus, 1953)

        The failed Stringy/Inflation regimists have taught man to be dishonest, with their prize students/useful idiots receiving tens of millions of dollars to further the hype/lies/deciet.

        Albert Einstein: But before mankind could be ripe for a science which takes in the whole of reality, a second fundamental truth was needed, which only became common property among philosophers with the advent of Kepler and Galileo. Pure logical thinking cannot yield us any knowledge of the empirical world; all knowledge of reality starts form experience and ends in it. Propositions arrived at by purely logical means are completely empty as regards reality. Because Galileo saw this, and particularly because he drummed it into the scientific world, he is the father of modern physics — indeed, of modern science altogether.

        And so you see why the failed Stringy/Inflation regime had to get rid of the Soul of Einstein, Galieo, and Copernicus–the Soul of Science.

        Please remove Brian Green’s lies from the introduction to Einstein’s Meaning of Relativity. Please replace the lies with the wisdom of Copernicus, Newton, Einstein, and Galileo.

        Dr. Elliot McGucken

  8. If we simply replaced the entire content of the Munich conference with the words, thoughts, and philosophies of the likes of Max Born, Richard Feynman, Einstein and other Nobel Laureates and Luminaries such as Sir Isaac Newton and Galileo, would not physics and its future be better off?
    R.P. Feynman: I don’t like that they’re not calculating anything. I don’t like that they don’t check their ideas. I don’t like that for anything that disagrees with an experiment, they cook up an explanation-a fix-up to say, “Well, it might be true.” For example, the theory requires ten dimensions. Well, maybe there’s a way of wrapping up six of the dimensions. Yes, that’s all possible mathematically, but why not seven? . . . So the fact that it might disagree with experience is very tenuous, it doesn’t produce anything; it has to be excused most of the time. It doesn’t look right.
    Max Born: All great discoveries in experimental physics have been made due to the intuition of men who made free use of models which for them were not products of the imagination but representations of real things.
    Einstein: Time and again the passion for understanding has led to the illusion that man is able to comprehend the objective world rationally by pure thought without any empirical foundations—in short, by metaphysics.
    Gerard ‘t’Hooft: Actually, I would not even be prepared to call string theory a “theory” rather a “model” or not even that: just a hunch. After all, a theory should come together with instructions on how to deal with it to identify the things one wishes to describe, in our case the elementary particles, and one should, at least in principle, be able to formulate the rules for calculating the properties of these particles, and how to make new predictions for them. Imagine that I give you a chair, while explaining that the legs are still missing, and that the seat, back and armrest will perhaps be delivered soon; whatever I did give you, can I still call it a chair?
    Einstein: Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius—and a lot of courage—to move in the opposite direction (dx4/dt=ic).
    Sheldon Glashow: It is tragic, but now, we have the string theorists, thousands of them, that also dream of explaining all the features of nature. They just celebrated the 20th anniversary of superstring theory. So when one person spends 30 years, it’s a waste, but when thousands waste 20 years in modern day, they celebrate with champagne. I find that curious.
    Einstein: The theory must not contradict empirical facts. . . The second point of view is not concerned with the relation to the material of observation but with the premises of the theory itself, with what may briefly but vaguely be characterized as the “naturalness” or “logical simplicity” of the premises of the basic concepts and of the relations between these which are taken as a basis.
    Sheldon Glashow: But superstring physicists have not yet shown that theory really works. They cannot demonstrate that the standard theory is a logical outcome of string theory. They cannot even be sure that their formalism includes a description of such things as protons and electrons. And they have not yet made even one teeny-tiny experimental prediction. Worst of all, superstring theory does not follow as a logical consequence of some appealing set of hypotheses about nature.
    Philip W. Anderson Physicist and Nobel laureate, Princeton: “Is string theory a futile exercise as physics, as I believe it to be? It is an interesting mathematical specialty and has produced and will produce mathematics useful in other contexts, but it seems no more vital as mathematics than other areas of very abstract or specialized math, and doesn’t on that basis justify the incredible amount of effort expended on it.
    My belief is based on the fact that string theory is the first science in hundreds of years to be pursued in pre-Baconian fashion, without any adequate experimental guidance. It proposes that Nature is the way we would like it to be rather than the way we see it to be; and it is improbable that Nature thinks the same way we do. The sad thing is that, as several young would-be theorists have explained to me, it is so highly developed that it is a full-time job just to keep up with it. That means that other avenues are not being explored by the bright, imaginative young people, and that alternative career paths are blocked.”
    Would not young physicists be better off hearing these words of wisdom, as opposed to the words of Bee, Kane, Dawid, et al.?
    How many undergraduates, graduate students, or professors could recite Newton’s or Einstein’s rules for physics?
    Einstein’s Three Rules of Work: 1. Out of clutter find simplicity (dx4/dt=ic). 2. From discord find harmony. 3. In the middle of difficulty lies opportunity.
    Newton’s Four Rules of Science and Natural Philosophy Rule 1: We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances.
    Rule 2: Therefore to the same natural effects we must, as far as possible, assign the same causes (dx4/dt=ic).
    Rule 3: The qualities of bodies, which admit neither intensification nor remission of degrees, and which are found to belong to all bodies within the reach of our experiments, are to be esteemed the universal qualities of all bodies whatsoever.
    Rule 4: In experimental philosophy we are to look upon propositions inferred by general induction from phenomena as accurately or very nearly true, not withstanding any contrary hypothesis that may be imagined, till such time as other phenomena occur, by which they may either be made more accurate, or liable to exceptions.
    I contend that if the brilliant light of this wisdom were allowed to shine forth, the murky string theory and multiverse fog would dissipate like an early morning sea mist kissed by the rising sun.
    What we need, more than ever, is to turn towards the wisdom of those who actually advanced physics. We should look to apply their diverse array of mathematical, philosophical, and physical methods in our own context!
    Einstein: The theory must not contradict empirical facts. . . (there are no empirical facts supporting 11 nor 42 dimensions; nor tiny, vibrating strings) The second point of view is not concerned with the relation to the material of observation but with the premises of the theory itself, with what may briefly but vaguely be characterized as the “naturalness” or “logical simplicity” of the premises of the basic concepts and of the relations between these which are taken as a basis.
    Schrodinger: The world is given but once. . . (No multiverse!) The world extended in space and time is but our representation. Experience does not give us the slightest clue of its being anything besides that.
    Nobel Laureate Robert Laughlin: [String Theory] has no practical utility, however, other than to sustain the myth of the ultimate theory. There is no experimental evidence for the existence of strings in nature, nor does the special mathematics of string theory enable known experimental behavior to be calculated or predicted more easily. . . String theory is, in fact, a textbook case of Deceitful Turkey, a beautiful set of ideas that will always remain just barely out of reach. Far from a wonderful technological hope for a greater tomorrow, it is instead the tragic consequence of an obsolete belief system-in which emergence plays no role and dark law does not exist.—A Different Universe, Laughlin
    Planck: That we do not construct the external world to suit our own ends in the pursuit of science, but that vice versa the external world forces itself upon our recognition with its own elemental power, is a point which ought to be categorically asserted again and again . . . From the fact that in studying the happenings of nature . . . it is clear that we always look for the basic thing behind the dependent thing, for what is absolute behind what is relative, for the reality behind the appearance and for what abides behind what is transitory. . this is characteristic not only of physical science but of all science.
    Einstein: Mathematics are well and good but nature keeps dragging us around by the nose.
    Poincare: Geometry is not true, it is advantageous.
    Sir Francis Bacon: And all depends on keeping the eye steadily fixed upon the facts of nature and so receiving their images simply as they are. For God forbid that we should give out a dream of our own imagination for a pattern of the world; rather may he graciously grant to us to write an apocalypse or true vision of the footsteps of the Creator imprinted on his creatures.
    In Einstein’s Mistakes, Dr. Hans Ohanian reports on how physics advances via the emphasis not on math, but on physical reality, “(Max) Born described the weak point in Einstein’s work in those final years: “. . . now he tried to do without any empirical facts, by pure thinking. He believed in the power of reason to guess the laws according to which God built the world.”
    In Disturbing the Universe, Freeman Dyson writes, “Dick [Feynman] fought back against my skepticism, arguing that Einstein had failed because he stopped thinking in concrete physical images (as MDT does!) and became a manipulator of equations. I had to admit that was true. The great discoveries of Einstein’s earlier years were all based on direct physical intuition. Einstein’s later unified theories failed because they were only sets of equations without physical meaning. Dick’s sum-over-histories theory was in the spirit of the young Einstein, not of the old Einstein. It was solidly rooted in physical reality.”[xxxiii] In The Trouble With Physics, Lee Smolin writes that Bohr was not a Feynman “shut up and calculate” physicist, and from the above Dyson quote, it appears that Feynman wasn’t either. Lee writes, “Mara Beller, a historian who has studied his [Bohr’s] work in detail, points out that there was not a single calculation in his research notebooks, which were all verbal arguments and pictures.”[xxxiv] Please see MDT’s Fig. 1, presenting a physical model, at the end of this document. (Many more to come!)
    In Dark Matters, Dr. Percy Seymour writes, “Albert Einstein was a great admirer of Newton, Faraday, and Maxwell. In his office he had framed copies of portraits of these scientists. He had this to say about Faraday and Maxwell: “The greatest change in the axiomatic basis of physics—in other words, of our conception of the structure—since Newton laid the foundation of theoretical physics was brought about by Faraday’s and Maxwell’s work on electromagnetic phenomena.”
    Yes! And in his book “Einstein,” Banesh Hoffman and the great Michael Faraday exalt physical reality over mere math:
    “Meanwhile, however, the English experimenter Michael Farady was making outstanding experimental discoveries in electricity and magnetism. Being largely self-taught and lacking mathematical facility, he could not interpret his results in the manner of Ampere. And this was fortunate, since it led to a revolution in science. . . most physicists adept at mathematics thought his concepts mathematically naïve.”
    “We don’t know what we are talking about.” –Nobel Laureate David Gross on string theory
    “It is anomalous to replace the four-dimensional continuum by a five-dimensional one and then subsequently to tie up artificially one of those five dimensions in order to account for the fact that it does not manifest itself.” -Einstein to Ehrenfest (Imagine doing this for 10-30+ dimensions!)
    “String theorists don’t make predictions, they make excuses.” – Feynman, Nobel Laureate
    “String theory is like a 50 year old woman wearing too much lipstick.” -Robert Laughlin, Nobel Laureate
    “Books on physics are full of complicated mathematical formulae. But thought and ideas (the fourth dimension is expanding relative to the three spatial dimensions at c), not formulae (dx4/dt=ic), are the beginning of every physical theory.” —Einstein/Infeld, The Evolution of Physics
    Einstein: “As far as the laws of mathematics refer to reality, they are not certain, and as far as they are certain, they do not refer to reality.”

  9. Chris G says:

    > Massimo Pigliucci, a philosopher at the Graduate Center of the City University of New York, pointed out that falsifiability is woefully inadequate as a separator of science and nonscience, as Popper himself recognized. Astrology, for instance, is falsifiable — indeed, it has been falsified ad nauseam — and yet it isn’t science.

    Perhaps all science is falsifiable but falsifiable things aren’t necessarily science?

    > Physicists’ preoccupation with Popper “is really something that needs to stop,” Pigliucci said. “We need to talk about current philosophy of science. We don’t talk about something that was current 50 years ago.”

    Do philosophical concepts have a shelf life? Does the passage of time guarantee progress? Are there ideas out there which can’t be improved upon?

    > … modern scientific theories typically make claims far beyond what can be directly observed — no one has ever seen an atom …

    What does she mean by “seen”? Visible to the naked eye? If that’s your criterion then most types of cells haven’t been “seen”. No one has “seen” a virus. Do we believe cells and viruses are just convenient hypotheses which follow from inductive reasoning or do we believe that we have actually seen them via the use of scientific instruments? With respect to seeing an atom, how is an atomic force microscope so different from an optical microscope in its ability to enable visualization of things too small to observe with the naked eye?

    For what it’s worth, I’m with Ellis and Silk.

    • Rahul says:

      +1 These criticisms by Massimo Pigliucci etc. sound weak.

      Falsifiability is a necessary condition but not sufficient. I could claim E = m c^8. That’s eminently falsifiable & indeed false. So is astrology. You are allowed to make stupid conjectures but must be willing to accept them proven false.

      The iffy territory is when theories (e.g. string theory, Van Raamsdonk’s quantum gravity etc.) build elaborate, science-ey theories but these theories lack any way to empirically test them. To be proven false, a theory must expose some potential vulnerability by which it may at least be proven false.

      • Andrew says:

        Rahul:

        In the usual scale-free parameterization, c=1, so E = mc^8 is fine.

        • Rahul says:

          Sigh. It is so hard to be right. How about “Gravity is repulsive”?

          Must I fall back upon “2 + 2 = 5”? Though I’m sure some mathematician somewhere has developed a ZFCTRQ axiom system in which that too is true. :)

        • Ha, this made me laugh out loud literally, because I’m always going on about dimensionless models.

          We have 3 free dimensions: length, time, mass. So, we can always choose a parameterization in which this equation becomes 1=1. However, a property of mathematical models of the universe is that they need to be *invariant* to the choice of parameterization, since the universe doesn’t know what you mean by “a meter” or “a second” or “a kg” or “a slug” or “a foot” etc.

          In any dimensional parameterization, E=mc^8 is dimensionally inhomogenous, energy on the left, and energy * velocity^6 on the right. The fact that it’s not invariant to parameterizations means its wrong even before testing it. In fact, we are stuck with models of the form E = K * m c^2 for dimensionless constants K by this scale invariance principle.

          • ojm says:

            How many fundamental dimensions are there? Why?

            • Rahul says:

              Seven, I thought. Could be my naive view. Not sure if it was a leading question.

              Why: Because none of these can be expressed in terms of the others. You can choose more than one such set as fundamental, but I thought the set size always stays the same.

              • I think “why” is actually a really deep question, and may have something to do with symmetries/conserved quantities in physics. Also, for things that are quantized, like charge, it’s not really true that things are invariant to rescaling, but they are approximately invariant asymptotically for large charges (so for example, when dealing with “static electricity” on macroscopic objects for example, but not when dealing with say droplets of oil in the Millikan oil drop experiment). There is a fundamental unit of charge.

                And, there are invented dimensions, such as whatever it is that “dollars” is the unit of.

                I think in most problems it’s really just three fundamental dimensions, which I usually think of as “length” “mass” and “time”, it’s possible that it’s really appropriate to consider “entropy” in some settings.

                The SI people talk about the fundamental dimensions being

                length, mass, time, charge, temperature, amount, and luminous intensity, but charge is quantized, and so is “amount”, and luminous intensity is of interest only to humans as it’s a measurement that combines quantity of light with sensitivity of the eye… sensitivity of the eye isn’t exactly “fundamental”.

              • Rahul says:

                @Daniel:

                Why should quantization exclude units from being fundamental? Even if quantized, “charge” or amount cannot be expressed in terms of a more parsimonious set of units, correct?

                My naive view of fundamental units was as a sort of minimal basis set of independent vectors that can then be used to span the rest of dimension space of interest to us.

                Isn’t that property what matters in applications like dimensional analysis, Buckingham Pi theorem etc?

              • @Rahul, when there is a distinguished unit, such as the unit of charge, then the symmetry property doesn’t work. The description of the world isn’t invariant to your choice of the unit, there is really only ONE unit, and anything else we choose logically has to be an integer multiple of that unit.

                However, when you’re talking about a LOT of charge, like 10^23 or whatever, then the fineness of the integer grid is sufficiently fine to have “near invariance” to the choice of a unit. that’s what I meant by asymptotically for large charge.

                So it’s not that charge isn’t a fundamental thing to measure, it’s just that it’s always an integer so it doesn’t have the symmetry property of interest.

              • Rahul says:

                Also, using Bekenstein’s bound etc., very deep down, isn’t is likely that length itself is quantized too? Or is that still too speculative?

              • I’m not very sure about what’s currently theorized about quantized length or quantized time, but I suspect that whatever those quanta might be, the asymptotic approximate invariance is already in play at scales of the diameter of a proton or the duration of one cycle of a massively high energy cosmic gamma ray or whatever, so we can call it all good for even outlandish scales…

                mass also, you might think mass had a fundamental unit, like the mass of whatever the lightest quark is or something… but the relativistic equations say that mass varies with velocity, and if velocity is length/time and is not quantized (or hyper-finely quantized or whatever) then mass, and maybe more fundamentally, energy has to be hyperfine-quantized as well.

              • Rahul says:

                @Daniel

                My point is, if you really want to go deep down the same quantization critiques can be applied to them all the fundamental units. It’s only a matter of how deep down you go.

              • Yes, but the scale invariance symmetry assumption for charge works approximately when you’re dealing with a lot of charges, but we sometimes are dealing with a small number of charges like in a silicon fabrication problem or something…

                similarly, maybe the scale invariance symmetry assumption for length only works for lengths larger than say 10^-25 times the diameter of a proton… well, OK fundamentally it’s maybe quantized but for every single actual application the scale invariance holds in asymptotic approximation to at least 1 part in 10^20 or so…

                so, the question is more like “is it safe to make this assumption?” and for charge the answer is “sometimes no” but for length the answer is “essentially always yes”.

              • xiaokj says:

                Actually, there is an obvious answer to this question. And it is zero.
                http://arxiv.org/abs/physics/0110060
                Follow Professor Michael Duff’s argument there. There is actually a follow-up, and the opposing comments borderline on silliness.

                That is, the only constants that matter to the operation of the universe, are dimensionless ones. Instead of c, hbar, epsilon naught and so forth, the universe only cares about the fine structure constant, say. Indeed, c and hbar are routinely set to unity, dimensionless, and in turn those fix the unit measurement system in use.

                As such, velocity is itself dimensionless, so the people saying that E = mc^8 is equivalent to E = mc^2 are correct.

                But Daniel is correct that quantisation removes the symmetry. So charge of electron is the integer unit, and independent of these other considerations.

                Maybe one can argue that fine structure constant and friends could be seen as the fundamental dimensions of sorts, but do note that they are dimensionless. They do indeed correspond to the best attempts at defining the thing we naively mean, though.

                As for M,L,T as the fundamental dimensions as Daniel is so aggressively pushing, that is straight out. First of all, mass is really not part of anything fundamental, as anybody that actually understands modern physics understands that all appearances of mass in old classical physics actually are measuring absolute values of energies. Rest mass is but just energy in the zeroed momentum frame. Learning a bit of GR puts that extremely squarely to rest. Also, by setting c=1, length and time are one unit, not two. Again, following Professor Duff’s argument is best.

    • Rahul says:

      Pigliucci, a philosopher by profession, should be the last person to complain about a pre-occupation with things-50-years-old. His profession leads in the obsessive reverence of ancient thoughts.

      Philosophers, even today, build entire careers out of studying Plato’s & Aristotle’s philosophies. Studied in minute, brain-numbing, redundant detail.

      There’s hardly any other academic discipline more obsessed with ancient thoughts than Philosophers.

      • Ram says:

        The subject you’re referring to is “history of philosophy”, rather than “philosophy”. There is a substantial subdued dedicated to the former, but you won’t find contemporary philosophers working in epistemology, philosophy of science/math/mind/language etc. devoting any substantial effort to interpreting the ancients.

        To be sure, there is a lot more textual interpretation in philosophy than in the sciences, since it is easy to mischaracterize philosophers’ positions and arguments, but this is what motivates the (often excessive) formalism used in much of the field these days, to move to a place where there are simply theories and arguments for and against those theories which can be evaluated independently of their source.

        The difficulty with expecting philosophy to look like science is once it really starts to, it stops being philosophy and starts being science. Nevertheless, the questions philosophers discuss are often interesting and important, even if they don’t yet admit of scientific treatment, and while it is easy to dismiss philosophy, the fact is we all have views on many of these questions, often unarticulated, which we take for granted in our reasonings in science and in everyday life. Surely there is at least some value in getting those views right.

    • Keith O'Rourke says:

      > Do philosophical concepts have a shelf life? Does the passage of time guarantee progress? Are there ideas out there which can’t be improved upon?

      Good questions – No, no and maybe.

  10. Andrew, I was glad to see your emphasis on model checking being a major part of Bayesian statistics, both here on this blog and of course it has a prominent place in your book also. I don’t understand why this view is hard for some to accept. I was recently on a panel to evaluate the standard Earthquake forecasts for California, called UCERF, which are used by the government for building codes, insurance, and other purposes, and my main argument was that they need to be included in the earthquake model evaluation modules (which exist and are excellent). One member of the panel said that because UCERF is a sort of Bayesian model that averages other models together, it would be “inappropriate” to assess its goodness-of-fit. He actually used the word inappropriate. I found this very frustrating, and ultimately told the panel that I would sign off my approval of UCERF if and only if they included the model in the evaluation module. In other words, even if the model does poorly, it’s fine with me as long as they assess it. To me, proposing a model and fitting it to data without evaluating the fit … is that even science? It seems very common unfortunately. If you’d ever like to collaborate on something involving model evaluation for point process models, Bayesian or otherwise, please let me know. I’d definitely be interested.
    Rick

  11. How about some more contemporary philosophy of science? Start with Quine’s “Two Dogmas of Empiricism” and work your way up to Rorty’s Philosophy and the Mirror of Nature if you want to see how philosophers think about these debates. And that’s just when I got off the philosophy bus about 20 years ago. Of course, this’ll make Keith O’Rourke happy, because it’ll send you straight back to Peirce, the father of pragmatism. In my experience, scientists (like Einstein) tend to be straight-up Platonists. Contemporary philsophy of science is much more nuanced in its acceptance of the human angle in theorizing. Gleick’s bio of Feynman does a great job of bringing this out in early quantum theorizing.

    For example, we didn’t discover that Pluto isn’t a planet, we defined it through semantic social convention, presumably due to pragmatic concerns. Rather than redefining planethood to exclude Pluto, we could’ve just as easily moved to include it (and many other Solar System bodies). There’s no experiment we can do that answers the question “is Pluto a planet?” without taking the human step of defining what the English word “planet” means. Nothing is forcing us to define “planet” to mean a body that’s massive enough to be roughly spherical, orbits in the plane of other planets, and clears its orbit. It just seemed more useful to do that than to include hundreds of other Solar bodies to the list of planets. This is the same thing chemistry taught us about water—it turns out there’s quite a bit more to the water we interact with than H20. For example, we could’ve defined “water” in such a way to exclude salt water; if you think about it, it’s not even clear where the line is between water and mud.

    The key is that once we make the definition, we can start to answer the question of whether Pluto is a planet or not. But even then, we have to be careful for all the reasons Quine outlined in his seminal paper “Two Dogmas of Empiricism”, most notably that definitions depend on other definitions and thus create a web of definitions that are very hard to ground out in the way the logical positivists were trying to do in the empirical philosophy tradition before Quine. At this point, I should give a shout-out to Teddy Seidenfeld, a Bayesian philosopher of science with whom I used to teach philosophy of language at Carnegie Mellon—he loved that Quine paper and taught me just what a critical turning point it was in philosophy; of course, he also knew all the predecessors—nothing happens in a vacuum.

    The key step is to understand that all of our scientific concepts have this character, including atoms, gravity, and all the other fun stuff from physics, chemistry, and biology. For instance, the whole notion of “gene” is now up for debate given the epigenetic and regulatory turn and the notion of splice variants. It used to be a location on the genome, but now it’s not even clear how “gene” should be defined most usefully. This is sometimes referred to as the “linguistic turn” in philosophy—realizing that we create definitions and concepts rather than discover them.

    And that brings us straight back to Susan Gelman’s work on how such concepts are formed and used by people (and that includes scientists!). To me, this is the most fascinating problem in cognitive psychology (and cognitive anthropology, if you think Sapir and Whorf and the relation between language and reasoning belong in a different field). If you want to see the essentialist thread in philosophy, read Hilary Putnam; for an opposing point of view, dig out Wittgenstein.

    • jrc says:

      “In my experience, scientists (like Einstein) tend to be straight-up Platonists.”

      …and yet engineers, who actually use physics to make stuff happen, tend to be pragmatists philosophically (as well as in the colloquial sense). I remember my electrical engineer dad one day saying “oh no, those equations aren’t ‘real’ in the sense of exactly describing the world, but you can do a lot with them.” I believe we had that conversation about the time I was reading Quine in a Philosophy of Science undergrad course.

      To add just a touch on the history of thought here – to me, Quine’s ideas about “wholism” and the inter-connected web of ideas and definitions that constitute “Science” made more sense after I read Kuhn and thought about the problem in historical context. It was like “look, here are other fully-formed theories that explained a ton of things we could see and gave us a sense of how the world ‘is’, but which were totally, insanely wrong” (I’m looking at you, Ptolemy, and you too, Newton, and Einstein, and Next Guy). Once I realized that throughout history many, many people much, much smarter than me believed whole-heartedly and unquestioningly in things about the world that were 100% completely wrong, it became much easier to accept the limits of my own (and my own era’s) actual knowledge of the world.

      And finally, in terms of importance to social science: Sometimes I think that, tucked away in our little corners of the world, social scientists get so caught up in our web of definitions that we forget we created that web to illuminate and understand things in the real world, not to sustain itself in some sort of perpetual self-reinforcing, where the concepts we created to understand particular features of the world become the object of interest themselves, and the features of the world we were interested in retreat from view (or maybe better said: we end up retreating from the real world and ensconce ourselves in our own web, despite the fact that the only reason we created the web was to describe the world).

      • Martha says:

        “… in terms of importance to social science: Sometimes I think that, tucked away in our little corners of the world, social scientists get so caught up in our web of definitions that we forget we created that web to illuminate and understand things in the real world, not to sustain itself in some sort of perpetual self-reinforcing, where the concepts we created to understand particular features of the world become the object of interest themselves, and the features of the world we were interested in retreat from view (or maybe better said: we end up retreating from the real world and ensconce ourselves in our own web, despite the fact that the only reason we created the web was to describe the world).”

        +1

    • Martha says:

      “The key step is to understand that all of our scientific concepts have this character, including atoms, gravity, and all the other fun stuff from physics, chemistry, and biology. For instance, the whole notion of “gene” is now up for debate given the epigenetic and regulatory turn and the notion of splice variants. It used to be a location on the genome, but now it’s not even clear how “gene” should be defined most usefully. This is sometimes referred to as the “linguistic turn” in philosophy—realizing that we create definitions and concepts rather than discover them.”

      +1

      And the word “probability” also has this character. In particular, everyday uses of the word may mislead understanding the technical definitions.

    • Keith O'Rourke says:

      The challenge is how to bring “enabling of good inquiry” perspectives of philosophy into the practice of statistics without losing everyone in the swamp of philosophical debate.

      “It’s rational not to check your priors” is definitely not a “enabling of good inquiry” perspective but it is widely out there even in the statistical discipline. So is p.value > .05 means nothing is going on or as a statisticians recently retorted to my question on their apparent use of that _logic_ – “no, of course not, but its a strong indication that nothing is going on” [power was very low].

      I think I was fortunate to essentially have started with Peirce and have not found a better source of “enabling of good inquiry” insight – so far.

    • Chris G says:

      > This is the same thing chemistry taught us about water—it turns out there’s quite a bit more to the water we interact with than H20. For example, we could’ve defined “water” in such a way to exclude salt water; if you think about it, it’s not even clear where the line is between water and mud.

      Labeling is an act of dimensionality reduction. There isn’t a hard boundary which defines the class “water” but I can measure the optical transmission of a sample, measure turbidity, measure viscosity, do a chemical analysis and countless other tests and define a properties-based “water” class which is highly distinguishable from the “mud” class. That a sample may possess properties which aren’t unambigously “water” or “mud” doesn’t mean that you can’t reasonably define them. The labels are useful to the extent that (to borrow Keith’s phrase) they enable good inquiry. As with any form of dimensionality reduction defining a class by a name only runs the risk of blowing away useful information – but it doesn’t necessarily mean that you have done so.

    • Chris G says:

      > Start with Quine’s “Two Dogmas of Empiricism” and work your way up to Rorty’s Philosophy and the Mirror of Nature if you want to see how philosophers think about these debates.

      I read “Take Care of Freedom and Truth Will Take Care of Itself” about a decade ago. Actually, I started reading and then quit about halfway through because it didn’t resonate with me. Worth noting perhaps that I picked it up not because I expected to like it but that it seemed important to grapple with it. Two questions:

      1. Given a bad reaction to “Take Care of Freedom…” how would you recommend approaching “Philosophy and the Mirror of Nature”?

      2. Given a bad reaction to “Take Care of Freedom…” a decade ago better to take another crack at it or try a different one of his works?

  12. Laplace says:

    To get an idea of how successful this is likely to be, just examine the sizzling intellect of the most successful philosopher of statistics out there.

    The philosopher in question looked upon the mass of failed of replications and presumably noticed (1) the papers were always frequentist, (2) usually written by people who only studied frequentist methods, (3) learn from Frequentist text books which often characterized bayesians as borderline insane, and (4) published in areas that practically requiring frequentist methods; and concluded based on (1)-(4) that the mass of failed replications is due to Bayesians.

    https://twitter.com/learnfromerror/status/677901059910705153

    This reminds me of the time when the majority of planes were crashing due to pilot error. To fix the problem, pilots held a conference with philosophers of flying. These philosophers had never flown a plane (or done anything else for that matter), but they brought special perspective to the discussion because they had read previous philosophers of flying who also hadn’t ever flown a plane. The number of flying accidents dropped considerably as a result.

Leave a Reply