Evolving Thoughts

Skewed views of science

Posted by John S. Wilkins on January 16, 2009
(15)
More »

Larry Moran points us to the following video on what science is and why pseudoscience is not to be taught or accepted without serious evidence (which makes it science). My only comment to add is that emotional appeals are information and evidence, but they are information and evidence about the speaker, and not about the things that are being spoken of.

There are a number of people trying to give short definitions of science on the blogs right now (see here). I have a one line definition that I think captures everything I want it to, and nothing else:

Science is the process of saying as much as one can about measurable (empirical) phenomena in as simple a model as possible. Or more simply, science says as much as it can in as short a sentence as possible.

(15)
More »

Comments

  1. #1 John Pieret
    January 16, 2009

    Or more simply, science says as much as it can in as short a sentence as possible.

    Science: minimum words, maximum knowledge.

    There, that’s better.

  2. #2 pixelsnake
    January 16, 2009

    Great video, thanks for sharing!
    I’m going to pass this around to my non-sciencey friends.

  3. #3 Chris Nedin
    January 16, 2009

    Science densely measurable

  4. #4 Ben Breuer
    January 16, 2009

    Science: observed logic.

    As opposed to “just” derivative forms of logic. But I don’t actually know what I’m talking about there (musicologist speaking) and it’s late, so: \lurk \snorfle

    Like the blog and the blogger, btw. Please carry on.

  5. #5 Cecil
    January 16, 2009

    Science: black hole.

    We think information can be retrieved from it, but we’re not really sure how.

  6. #6 Tristram Brelstaff
    January 16, 2009

    The person who declares evolution to be nonsense because no one’s ever seen an ape mutate into a man, is as conspicuously uninformed as the person who shouts “Snap!” at a poker game,…

    Nice.

  7. #7 Cannonball Jones
    January 16, 2009

    “minimum words, maximum knowledge”

    Science also has to create new words all the time so I can go one better – Science: minwords, maxknowledge. Do I get a prize? :)

    Excellent video though, there seems to be more and more clips in this vein appearing on YouTube in the past few months. That can only be a good thing.

  8. #8 John S. Wilkins
    January 16, 2009

    Not unless you can get it down to one word (not “science”).

  9. #9 jeff
    January 17, 2009

    When you specify something as simply or in as few words as possible, you leave it more open to interpretation. More words will constrain more. That’s one reason legal language is so verbose.

  10. #10 Robert E. Harris
    January 17, 2009

    No, no. Science is HARD! See

    http://www.theonion.com/content/node/38575?utm_source=onion_rss_daily

  11. #11 abb3w
    January 17, 2009

    Science Minimizes Description Length to Maximize Probable Correctness.

  12. #12 Alan Kellogg
    January 17, 2009

    1.Life on Earth did not evolve over millions of years, life on Earth evolved over billions of years, and is evolving even as we speak.

    2. All things being equal, that explanation using the fewest words is usually the best. (Yes, it is an alternate application of Ockham’s Razor.)

  13. #13 James Mattingly
    January 18, 2009

    “Science is the process of saying as much as one can about measurable (empirical) phenomena in as simple a model as possible. Or more simply, science says as much as it can in as short a sentence as possible.”

    Well these aren’t equivalent, so it’s hard to say how the latter is a simpler expression of the former. Probably building models (mathematical, conceptual, etc) is closer to what one really wants done in the sciences.

    Isn’t it clear anyway that the latter is not quite what one wants. For, is the idea that we ignore any constraints on the background language that gives the sentence meaning? If so then “gleb = blarg”is a pretty short, maximally informative sentence on the assumption that we define “gleb” and “blarg” properly. So science is now complete and all we have left is the trivial task of constructing the right language—by Wilkins’ stipulation, not part of science. If we don’t ignore such constraints then we seem to be in the position of saying that “science says as much as it can in as short a sentence as possible” with the proviso that we do so in some specified language. Or maybe the idea will be to construct, using various tools (math, logic, etc) languages that do allow such maximally informative short sentences. (Perhaps like Lavoisier’s interpretation of Condillac?) Isn’t stuff like this what derailed certain of the logical empiricists?

    Perhaps we’d be better off concentrating more on science’s active search for and responsiveness to data in its attempt to construct its models. And it wouldn’t hurt to keep in mind that sometimes the right thing to do is to allow the models to be more complicated than is immediately necessary, in order to facilitate future theory construction.

  14. #14 bob koepp
    January 19, 2009

    I wouldn’t hazard a formal definition, but just to get the flavor right, I’ll sugggest that (empirical) science is the art of framing questions about phenomena in a way that permits experiment/experience to limit the range of acceptable answers.

  15. #15 abb3w
    January 26, 2009

    James Mattingly: Or maybe the idea will be to construct, using various tools (math, logic, etc) languages that do allow such maximally informative short sentences.

    Mathematics gives rise to Formal Languages; Formal Languages give rise to Computability via Church-Turing Automata.

    Which yields what I noted earlier: Science Minimizes Description Length to Maximize Probable Correctness.

    Read the nice paper (doi:10.1109/18.825807), mon.