Monday, January 02, 2017

The Edge question 2017

Every year John Brockman asks his stable of friends an interesting question. Brockman is a literary agent and most of the people who respond are clients of his. (I want to be one.) The question and responses are posted on his website Edge. This year's question is, "What scientific term or concept ought to be more widely known?"

This year, the introduction is more interesting than the responses. Here's part of what Brokman wrote,
Many people, even many scientists, have traditionally had a narrow view of science as controlled, replicated experiments performed in the laboratory—and as consisting quintessentially of physics, chemistry, and molecular biology. The essence of science is conveyed by its Latin etymology: scientia, meaning knowledge. The scientific method is simply that body of practices best suited for obtaining reliable knowledge. The practices vary among fields: the controlled laboratory experiment is possible in molecular biology, physics, and chemistry, but it is either impossible, immoral, or illegal in many other fields customarily considered sciences, including all of the historical sciences: astronomy, epidemiology, evolutionary biology, most of the earth sciences, and paleontology. If the scientific method can be defined as those practices best suited for obtaining knowledge in a particular field, then science itself is simply the body of knowledge obtained by those practices.

Science—that is, reliable methods for obtaining knowledge—is an essential part of psychology and the social sciences, especially economics, geography, history, and political science. Not just the broad observation-based and statistical methods of the historical sciences but also detailed techniques of the conventional sciences (such as genetics and molecular biology and animal behavior) are proving essential for tackling problems in the social sciences. Science is nothing more nor less than the most reliable way of gaining knowledge about anything, whether it be the human spirit, the role of great figures in history, or the structure of DNA.
Many people prefer other definitions of "science" but that's okay. As long as you make the effort to define your terms you can have productive discussion about epistemology and whether something counts as science or not.

Here's some of the answers ....

The Genetic Book of the Dead, Richard Dawkins (adaptation)
Reciprocal Altruism Margaret Levi (more adaptation)
Common Sense Jared Diamond (interesting)
Evolve Victoria Wyatt (an adaptationist perspective)
Affordances Daniel Dennett (???)
Neoteny Brian Christian (starts well, ends badly)
The Neural Code John Horgan (no end in sight for neurobiology)
Scientific Realism Rebecca Newberger Goldstein (interesting)
Life History Alison Gopnik (humans adapt)
Negativity Bias Michael Shermer (I have a negative bias)
Morphogenetic Field Robert Provine (interesting and confusing)
Trolley Problem Daniel Rockmore (I agree, it should be widely known & understood)
Determinism Jerry Coyne (I agree, but good luck convincing the public)
Type I and Type II Errors Phil Rosenzweig (excellent idea!)
Biosociation James Geary (hmmmm ....)
Uncertainty Lawrence Krauss (I agree but the public isn't ready for this)
Epigenetics Leo M. Chalupa (blah!)
DNA George Church (not for everyone)
Non Ergodic Stuart A. Kauffman (as clear as most of Kauffman's writings)
Natural Selection Jonathan B. Losos ("tinkerer" would be a better choice of words)
Gaia Hypothesis Hans Ulrich Obrist (I think public needs to know less about Gaia)
The Scientist Stuart Firestein (confusing)
Sexual Selection Rory Sutherland (sex and adaptation)
Mismatch Conditions Daniel Lieberman (adaptation gone wrong)
Isoplation Mismatch David C. Queller (adaptation makes new species)
The Transcriptome Andrés Roemer (confusing)
Somatic Evolution Itai Yanai (depends on your definition of evolution)
Homeostasis Martin Lercher (I disagree, homeostasis is over-rated)
Phylogeny Richard Prum (I agree with the concept)
Replicator Power Susan Blackmore (adaptation leads to design)
Phenotyic Plasticity Nicolas Baumard (adaptation for plasticity)
Maladaptation Aubrey de Grey (adaptive trade-offs look like maladaptation)
Exaptation W. Tecumseh Fitch (preparing for adaptation)
The Scientific Method Nigel Goldenfeld (ask good questions)

Nobody asked me to contribute an answer to Brockman's question but if they had I would have suggested "Random Genetic Drift." I think everyone needs to know about alternatives to adaptation and the real null hypothesis of evolutionary biology. It would have provided a nice contrast to some of the other answers.


  1. As a creationist i say this guys definition makes every thing of knowledge to be scientific. the science of war, hunting, repairing the roof etc.
    Thats why i conclude there is no such thing as science. There just is conclusions and a methodology before the conclusions. Any subject can be that and so any subject can be scientific. YET the world says no. only the complicated things are scientific.

    Why not the bible as a reliable method for knowledge? who says its not reliable? if omne thinks it is then to that person its reliable and so scientific.

    If science actually can be segregated from any human conclusions THEN it must be on its methodology.
    So the scientific method has occurred even if it didn't add any new knowledge.
    This guys definition would not allow that.
    i think there is just people figuring things out and figuring out its been figured out by how they figured it out.

    1. Why not the bible as a reliable method for knowledge? who says its not reliable? if omne thinks it is then to that person its reliable and so scientific.

      Interesting viewpoint. A lot of your fellow theistic travelers see things the same way, but don't express it quite as clumsily, and hence not as honestly, as you do, Robert.

  2. Being nitpicky: "Phenotyic" --> phenotypic?

  3. The only chapters that seem to touch on or be related to physics are those of Kaufmann, Krauss and Coyne (!). Surely there are many concepts in physics that the general public doesn't understand, and should.

    The general public thinks Einstein proved "everything is relative" and thinks Heisenberg's uncertainty principle (if they've heard of it) is due to the observer altering the observed state by observing it.

    Other physics problems widely misunderstood or unknown:

    1. "Fine tuning." The Disco Tute now use "Fine Tuning" as a synonym for Intelligently Designed. The human genome is "fine tuned", blah blah blah. 99% of what's written about the cosmological fine tuning problem is horse$&!+. But there are real fine tuning problems in physics, and the public should be able to at least know there's a distinction between real fine tuning problems and "God did it."

    2. "Entropy and Second Law." The general public thinks that entropy (if they've even heard of it) is a measure of how inconvenient the universe is for them, and that the Second Law (if they've heard of it) means that the universe by nature just keeps getting more inconvenient for them, unless they vacuum their room or some shit.

    3. Eigenvectors. Yeah, sure, it has a German name, but it underlies all of quantum and classical mechanics. Try teaching any quantum concepts without understanding eigenvectors. Yeah, I wish high schools taught more linear algebra. But seriously, what do you think the musical note made by a plucked string is? And its higher harmonics?

    4. Wouldn't it be nice if the general public knew that Laplace's equation is still true in the quantum and relativistic realms, while Newton's shitty equations aren't?

  4. And with the possible exception of "uncertainty", none of these chapters seem to focus on the public's gross ignorance of statistics. Consider how the public misunderstands, or is ignorant of:

    1. Classical stats. Bayesian is one thing, but the general public doesn't even get the basic ideas of classical stats, the distinction between say probability density and probability mass, or the importance of defining a rejection region, so it makes them vulnerable to the Bridge Hand fallacy. "Why have people who were very remotely connected to Hillary Clinton, or who knew somebody who knew someone who was associated with Hillary Clinton, died over a 30 year period? I computed a very small probability of that happening. That proves Hillary Clinton murdered hundreds of people!" And of course, the probability of evolving such-and-such is tiny, therefore God did it.

    2. Bonferroni correction and the danger of hypothesis fishing. a.k.a. p-hacking. Obviously a problem for medical studies, epidemiological studies, etc. I've talked to grad students in biostatistics who didn't know about the Bonferroni correction for multiple tests.

    1. ""Why have people who were very remotely connected to Hillary Clinton, or who knew somebody who knew someone who was associated with Hillary Clinton, died over a 30 year period?"

      I will occasionally have undergraduate students make a statement along those lines. I've been responding by giving a rough count of the number of people I've known from high school, college, family, and jobs, who have died over the last 45 years.
      While the count isn't as high as the numbers (real or false I don't know) tossed around for the Clintons, it is uncomfortably large. That number, with the comment that I have not come into contact with as many people as the Clintons, due to their professions, has resulted in a few students realizing the foolishness of the "death conspiracy".

  5. This comment has been removed by the author.