Articles in the IMS Bulletin

Starting in 2017, I have contributed a few articles to the Bulletin of the Institute of Mathematic Statistics. A common thread in these articles is an address to fellow statisticians to perceive their discipline as implying a commitment to serving the public by applying thoroughgoing skepticism in order to challenge conventional scientific wisdom, established truths and the scientific and political establishment. To some extent these posts cover ground that I have covered before (e.g., of course, sortition), but the writeup is always new and tailored specifically to an audience of statisticians. I must note that I have been disappointed with the generally muted response (one way or another) to my posts and this disappointment has resulted in reduced motivation and diminished output.

Here is an index to my IMS Bulletin columns over the last 4 years:

Pro Bono Statistics: Statistics in the Public Interest. April, 2017
Statistics, like all of science, is a tool of powerful institutions. In our societies these institutions are widely perceived as not serving the public. Statistics can cut against the grain and promote a habit of skepticism toward power. Such a habit should be part of the statistics curriculum.

Learning as the replication of knowledge. October 2017
The educational system in our society is built upon an implicit model seeing learning as an mechanical, uncreative activity. This model is false and its implicit adoption inflicts great harm on students, teachers and society.

Democracy and statistical sampling. February 2018
Offering sortition – government selected by statistical sampling – to an audience of statisticians.

Statisticians for Democracy: A call to action. November 2019
Allotted bodies are gradually gaining acceptance in the political arena and may have real political impact. It is thus increasingly important to get the associated sampling procedures right. Yet, current sampling practices are quite deficient for multiple reasons. I call on statisticians to create standards by which sampling procedures for allotted bodies can be evaluated to determine how reliable and useful they are.

When experts go wrong…. February 2021
Polling is difficult for various reasons (including low response rates). Ignoring this fact and publicizing the polls as if they are highly accurate is highly irresponsible. It reduces the confidence of the citizenry in the establishments of our society and is thus quite dangerous and must be avoided. (A response to a column by Jeffrey Rosenthal.)

Advertisement

Ontological-epistemological formalizations

It seems that the voluminous discussion going under names such as positivism vs. interpretivism, modernism vs. post-modernism, foundatonalism vs. non-foundationalism, and behaviorism, realism and pluralism could benefit from a formalization of the ontological and epistemological models implied by the various positions staked in the discussion. It may be predictably claimed by some of the discussants that the attempt at formalization stakes a position in the contested landscape rather than maps it. I tend to disagree, but do not address this issue. Below is a generic ontological-epistemological formalization framework which allows the description of a wide variety of ontological-epistemological models by setting different values for its parameters. A few specializations, corresponding to radically different situations on the positivist-interpretivist spectrum, are presented.

Setup

Let the “objective, impersonal” universe (that is, all that is observable other than other observers) be a time series of points in an observation space R – r(t) ∈ R^T.

In addition the world contains a set of observers I, each with an internal state that develops over time s_i(t) ∈ S, i ∈ I, t ∈ T. The observations made by each observer at time t depend upon the state of the objective world at this time and upon the state of the observer at the time: f_i(t) = f(r(t), s_i(t)).

The observers can communicate between themselves. These communications are messages of finite length over a certain finite alphabet A, so the communications of observer i at time t is b_i(t) ∈ A*. The communication of each observer at each time depend on their state at the time: b_i(t) = b(s_i(t)).

Finally, each observer’s internal state, s_i can develop based on the observer’s observations and on the history of communications by other observers (and himself): ∆s_i(t) = s'(s_i(t), f_i(t), B_j(t), j ∈ I), where B_j(t) is the concatenation of all the messages of observer j up to time t: b_j(t’), t’ ≤ t.

Terrorism

The pursuit of the definition of “terrorism” has proven to be a formidable task that is personally useful for some. The standard dictionary definition,

the use of violence and threats to intimidate or coerce, especially for political purposes,

clearly conflicts with accepted usage, for if it were believed then any military activity, or even police activity, would constitute a terrorist activity. Other definitions come closer to the intended meaning:

Terrorism is defined as political violence in an asymmetrical conflict that is designed to induce terror and psychic fear (sometimes indiscriminate) through the violent victimization and destruction of noncombatant targets (sometimes iconic symbols). [Attributed by Wikipedia to Carsten Bockstette.]

Obviously, military activity would often fall within those terms if it were not for the “asymmetrical” condition. Such activity, when considered legitimate by the speaker, is never referred to as “terrorism”.

Bare Facts and Naked Truths

I found George Englebersten’s book Bare Facts and Naked Truths in a used book shop and was attracted to it because of the interesting (if often wrongheaded) quotations it contained. My favorites are:

Most writers regard the truth as their most valuable possession, and therefore are most economical in its use. –Mark Twain

This kind of thing is frightening to me because it gives me the feeling that the very concept of objective truth is fading out of the world … I am willing to believe that history is for the most part inaccurate and biased, but what is peculiar to our age is the abandonment of the idea that history could be truthfully written. In the past people deliberately lied, or they unconsciously colored what they wrote, or they struggled after the truth, well knowing that they must make many mistakes; but in each case they believed that ‘the facts’ existed and were more or less discoverable. -Orwell

We’ve reached a point where, in an orgy of political correctness, everything is true, and nothing is permitted. -Dean Kuipers

Truths are illusions. -Nietzsche

Paul Krugman, mainstream intellectual

Paul Krugman, Princeton economics professor, winner of the Nobel prize in economics for 2008 and New York Times op-ed writer is a person worth paying some attention to. Unlike most other mainstream commentators he does not deal mainly in cliches and often shows respect towards his readers by supporting his positions with hard data. He is also willing to take some risk by, on occasion, pushing against the envelope of mainstream propriety and using impolite words toward people with power.

Of course, he is not without faults. Of course, none of us is. The point is that Krugman’s faults are not too different than those of many of his colleagues. In other words, while Krugman is somewhat of an exquisite specimen of the mainstream intellectual, he is still very much such a specimen.

Comment on Phillip Davis’s “The Market for Scholarly Articles”

As Phillip Davis writes (reprinted in the IMS Bulletin), scholarly authors are driven to publish in journals because that is the way to have their work noticed and readers read journals because these provide some measure of quality assurance. Davis is wrong, however, on multiple counts, when he concludes: “This system is not intended to be fair and democratic, but it saves the time of the reader and functions to help consensus building in science. For those who feel that this perpetuates hegemony, let them eat cake.”

The academic publishing system is intended to be fair; the current system (though better than nothing) performs poorly as a time saving tool for the reader; “consensus forming” (i.e., suppressing non-conventional thought) is not a legitimate function of a scientific communication channel; and, finally, there is no reason to dismiss people who are unhappy with the current system with “let them eat cake”: there is a better way to run the scholarly publishing system.

Steven Levitt misrepresenting his source

Steven Levitt has risen to stardom by riding on the overblown rhetoric resting on the overblown claims of Freakonomics. It is, unfortunately, in the nature of popular books that they oversimplify and over-claim. Academic literature is supposed to be different: rigorous, cautious and, of course, peer-reviewed for accuracy.

Of course, if all those attributes really applied, a career like that of Levitt would have been impossible, since the econometric methodology he employs is far too weak to be able to produce with any credibility the kind of results Levitt is aiming at. It is clear, therefore, that within the community within which Levitt works, some standards of critical thought have been suspended. However, even when credulity is being stretched and poorly supported statements are taken as proven, one may still hope for the superficial ground rules to apply. Specifically, for example, one hopes that when previous research is cited and summarized the findings of the research are fairly represented.

A proposal for democratic media

The term “censorship” describes the act of suppressing certain ideas by those who control some distribution channels. Despite regular attempts by interested parties to limit the term to describe very restricted or extreme cases of suppression of ideas, the term is usually, and very reasonably, understood to cover any attempt at reducing the circulation of an idea, by any person or organization. The negative view, which most of the population, as well as official ideology, take of censorship therefore encompasses any such activity. According to this view, the desirable media system is democratic – i.e., one which allows all people an equal opportunity at presenting their ideas and having them considered by others.

The implicit universal rejection of censorship notwithstanding, much of the communication patterns that dominate Western society are inherently censoring activities. The members of the elite group that influences (to varying degrees) the content of wide circulation media – publishers, broadcasters, advertisers, editors and reporters – routinely make decisions that amount to suppressing some communications, containing certain ideas, in favor of other communications, containing different ideas. Those decisions, although usually purporting to reflect only objective accepted standards, are in reality almost completely subjective. They therefore reflect the ideas and biases of the very select and atypical group of people who make them.

The structure of scientific revolutions – the 1967 postscript

Below is my page-by-page summary of Kuhn’s 1967 postscript to The Structure of Scientific Revolutions.

The following points seem to me to contain the essence of Kuhn’s thesis:

Exemplars:

1. Exemplars (solved problems) are an important component of any specific scientific world-view. Without exemplars the laws and theories have little empirical content. The exemplars teach practitioners how to attach the relevant abstractions to elements of particular problems and how to see a variety of situations as being alike.
2. The intuitive knowledge of which situations are alike is analyzable but not by specifying rules since it is perceptual rather than interpretive knowledge. It is similar to other knowledge of perception: e.g., identifying certain light patterns as all representing swans. Perceptual knowledge is selected for “success” – success in survival in the primitive case, success in puzzle solving in the scientific case.
3. The situation may be similar in non-scientific schools (e.g., art).

Revolutions:

1. Scientific revolutions – “a special sort of change in scientific thinking involving a certain sort of reconstruction of group commitments” – occur regularly on the smaller scale (within communities of dozens to hundreds).
2. Crisis is the usual mechanism inducing revolutions. Crises supply a self-correcting mechanism which ensures that the rigidity of normal science will not forever go unchallenged.
3. Judgments of scientific values – simplicity, consistency and compatibility – can vary greatly from individual to individual. Therefore, debates on theory-choice cannot be decided in a formal way. Still, those values constrain the dominant group view. This explains why competing world-views are rare in science but are common in other human activities.

Scientific progress:

1. Because the commitment by scientists to puzzle-solving success shapes the long-term structure of scientific development (even if the application of this commitment in any particular case is ambiguous and subjective), scientific development, like biological development, is unidirectional and irreversible. When comparing scientific theories held by various scientific specialties to theories held by the specialty from which the former had their origins, “an uncommitted observer” would be able to consistently distinguish the newer theories from the older. The newer theories will enable more accurate, quantitative, predictions, in a wider variety of situations and they will be more esoteric. Differences in simplicity, scope and compatibility with other specialties are not as telling.
2. Scientific progress is therefore well defined, but, like biological progress, it cannot be said that science comes closer and closer to a certain goal – “reality”. There is no objective reality which stands outside a specific scientific world-view.

Censorship

The Davidson-Kahn Distinguished University Professor and professor of law at Florida International University, in Miami, and dean emeritus of the College of Liberal Arts and Sciences at the University of Illinois at Chicago, Stanley Fish, writes in his New York Times blog about the difference between what he calls the “colloquial” sense of the term “censorship” and what he calls the “philosophical and legal” sense of that word.

According to Fish, according to the colloquial sense,

censorship occurs whenever we don’t say or write something because we fear adverse consequences, or because we feel that what we would like to say is inappropriate in the circumstances, or because we don’t want to hurt someone’s feelings. (This is often called self-censorship. I call it civilized behavior.)

while according to the “philosophical and legal” sense, censorship has the characteristics that

(1) it is the government that is criminalizing expression and (2) that the restrictions are blanket ones.

Fish’s entire basis for the post is the unexplained assumption that the latter sense is the correct one, while the former one is simply a mistake. This, of course, in itself, is a major flaw, since, accepting Fish’s premises, he is criticizing a speaker for using a word in its accepted sense.

Going beyond this point toward more substantial matters, much of Fish’s thinking is resting on false notions.