Tuesday, August 3, 2010

A Question

How do you define "science"?

One of the problems with the earlier exchange was that people were making claims without even establishing what "science" is. Eventually I came around to offering a rough definition. This was my critique of Razib Khan's article too - he was making critiques of social science with no clear explanation of what he was calling "science".

Anonymous comments are fine if you don't have an account and that's an obstacle to posting. No interacting or critiquing necessary. I'm just curious how readers define "science".


  1. I'd define it (roughly) as reasoned inquiry into an object, which inquiry operates under methodological norms determined independently of the subject and object of inquiry though in a manner that is best suited to obtaining knowledge about the object of inquiry (and so, I suppose, not entirely independent of the object of inquiry insofar as it needs to be applicable).

    I think that experimentation, certain standards like falsifiability, etc. are possible aspects of particular instances of scientific work, but fall within the realm of possible norms and not that of some essence of science. Given all this, it should be obvious that I'm of the "pretty much anything can qualify as science" opinion. When I use the word "science" in day-to-day conversation, it's a convention with some pretty specific connotations, so that the term orbits around the "hard" sciences and, to a lesser extent, the "social" sciences.

  2. Daniel,

    I've read your stuff here and elsewhere for some time now, but this is my first comment (hopefully first of many).

    For me, science is an intellectual pursuit of knowledge whereby one can make observations of the world, form a reasoned hypothesis, and then endeavor to support said hypothesis with evidence, which can be experimental (using control and experimental variables) and/or intuitive (using deductive logic). If a correct hypothesis is far-reaching enough within the intellectual pursuit, it can eventually be deemed a theory or law.

    Personally, I'm onboard with your entire discussion of social science as a science, more especially economics. My background has typically been in the "hard sciences," majoring in undergrad in civil engineering (with several courses in physics, chemistry, and biology). But, my academic interests have grown thru the years to include the science of economics, first through my MBA at the Naval Postgradaute School, which prompted me to apply to GMU where I begin my Econ Ph.D studies in about a month.

    Thanks for the interesting discussion.

    Best regards,
    Ben Bursae

  3. I've always found these definitions of science to be very useful:

    "...science is taken to be a set of activities and habits of mind aimed at contributing to an organized, universally valid, and testable body of knowledge about phenomenon. At any given time, these general characteristics are usually embodied in systems of concepts, rules of procedure, theories and/or model investigations that are generally accepted by groups of practitioners - the scientific specialists." -- Richard Olson, "Science Deified & Science Defied: The Historical Significance of Science in Western Culture," pg. 7-8

    And since science is what scientists do, the scientific method:

    "The give and take between experimental observation, the mathematical formulation of descriptive and predictive theories, and further experimental tests of predictions of these form the scientific method. In the scientific method, human imagination, through intuition and curiosity, is subject to the checks and balances of experiment, thought, and further testing." -- Paul M. Fishbane, et. al., "Physics For Scientists and Engineers," 2nd Ed., pg. 1-2

    Man, I love looking through old textbooks. Ooooh, angular momentum! :)

  4. Ben - you are very welcome to the blog! I'm glad you read it, and I hope you'll comment more.

    I'm going to risk challenging the only commenter sympathetic to my position so far, and ask you about the point on deductive logic and the role that that plays in science. Are you saying that you can hypothesize, deduce, and draw conclusions and that that is a scientific undertaking? I would think that that would open it up to just about anything in the way that Evan has been proposing, so I'm not so sure about that. It seems like that would turn a lot of philosophy, for example, into science.

    I do like the way you've framed science as a problem-solving methodology, though. I think that is important. Science is a process as much as it is anything else, and I think the point that it is a process of theorizing, testing, and re-theorizing is the essential point. I do, however, think that replicability and empirical evidence are essential to this process, and I'm not sure I would personally consider deductive logic sufficient (certainly there are theoreticians that do nothing but think these problems through - but the point is they are a part of a broader scientific process).

  5. Oh by the way - congratulations on GMU. Do you know who you will be working with/what you will be studying there? GMU is a very interesting place. I have a ton of respect for it, but as I'm sure you know if I read this I find a lot of what goes on there frustrating too. Anyway - congratulations and good luck with that.

  6. I think of Science very much as the process, rather than the objective. I also take a pretty narrow view and tend to only want to include Popperian formulating and testing of falsifiable claims as Science. So that would mean that you could definitely do Science in economics, but that many economists aren't doing Science (because they aren't dealing with falsifiable claims) or are doing poor Science (because their tests don't reveal much about the truth of their claims).

    Science generates knowledge about the real world. Most knowledge about the real world can be transformed into technology of one kind or another. As a result, you can usually make reasonable guesses about who is practicing Science and who isn't by whether practical technology is coming out of their work.

    It's useful to note that Science doesn't necessarily produce truth. Newton was wrong. But, Newton was close to the truth. He was much closer than were those who came before him. Einstein got closer to the truth, still. Science produces, not truth, but ever more accurate approximations of the truth.

  7. Robert,

    Science may produce truth, and perhaps it has already, but we'll never be able to confirm it once and for all. Methodologically, the door can never be closed on further advances.

  8. haven't followed your previous post, but please make it clear that i retract the idea that social science isn't science. it was a title written in haste. as for my problems with social science, and specifically economics, you guys made predictions and assertions which were more precise and certain than i think warranted based on what you really knew.

    also, you're critique of sentimentality doesn't apply to any of my readers (or myself). we're invariably atheists or agnostics, so religion isn't a bar (85% are atheists and agnostics). and since i'm interesting biological models and understanding of human nature i don't have an issue with turn man into an animal, or societies into mechanisms.

    some of the same issues of hubris that i see with economists also applies to mathematical evolutionary biology. r. a. fisher famously wanted to construct evolutionary laws analogous to those of thermodynamics. i think he basically didn't do it, that extremely general laws are often trivial.

  9. Razib -
    The link was just a launching point for conversation, along with many other links. We didn't gang up on your position (and actually many of my commenters agreed social science is not a science).

    Could you be a little more specific or provide an example about predictions and assertions which were more precise and certain than warranted? I'm not sure exactly what you're refering to. Usually those predictions are reported with degrees of confidence, like any forecast. I'm concerned that you're listening to something a politician or pundit said and attributing it to social science. Examples? I saw a Youtube video the other day about how in 2012 Nibiru is going to destroy the Earth. I'm not going to attribute that to the physics community, though.

    I'm not sure what religion has to do with sentimentality. Could you explain the point a little more? In fact, I specifically said "An evolutionary biologist won't come out and tell you that we're made in God's image, but the residual hubris of that cosmology remains for a lot of people." I'm making a point that this is a cultural sentimentality, not an explicitly religious one - sorry if I confused the issue. Evan, who commented on these posts, is by no means an atheist or an agnostic and his objections to my post have nothing really to do with sentimentality - they're largely epistemological. So I don't really mean to raise religion as an issue when I mention sentimentality. And certainly that explanation may not apply to you and your readers - but I get the sense it is a common one.

    The modeling I think I can agree on, and that's something that a lot of the economics profession is converging on now - not that modeling is entirely inappropriate, but that its usefulness is limited particularly massive DSGE models built up on tenuous assumptions.

    I enjoy your blog very much - I just started following it recently and I've already linked to it a couple times. My blogroll was too full of economics blogs and I've got a personal interest in the natural sciences so I've added a few.

  10. Us religionists and our sentimentality. Not like those atheists and agnostics, who are Hardened Rational Scientists.

    Really, that is an odd association to make.

  11. Atheists are less sentimental than theists. At least according to the psychological research on this issue. That conclusion shouldn't be that shocking.

  12. That may be the case; I'd be interested to see the literature. I'm not sure why this should be seen as corroboration for the association of sentimentality with religion, though, unless you're falling into the sort of Platonism that you (rightly) set aside earlier in this conversation.

  13. Ya, I also think its plausible theists are more sentimental than atheists, but all this sort of misses my point.

    The kind of sentimentality I think plays into the odd readings of the social sciences is a sentimentality about how special mankind is. Insofar as most atheists are probably humanists of some variety, I think they're just as susceptible to that sort of sentimentality. And as I pointed out to Khan, I'm specifically thinking of residual sentimentality about man. You don't have to believe man was created from dust and breathed into 6,000 years ago to have that residual self-esteem boost.

  14. ...in fact, many (myself included) who believe that humanity was created in the image of God also have no trouble being decidedly un-sentimental about the state of humanity or what to expect from human reason and action. Other doctrines of various faiths make the idea of a religious glorification of human snowflake-ness less convincing than non-religious humanisms that don't enjoy the realist benefits of ideas like human sin or ignorance (which are funded primarily by theological considerations). One might argue that something like modern capitalism relies on such metaphysical foundations, and it's unwise to dismiss one aspect of the metaphysics as sentiment without realizing that one's own (perhaps now non-religious) lack of sentiment probably descended from the same source.

  15. I wasn't trying to be insulting. I do think atheists and theists are "wired" somewhat differently though.

  16. Oh, no doubt. In any case, I wouldn't find it insulting for someone to call me sentimental. I do hope I am, to a significant extent.

    I think we found the point more irrelevant than insulting.

  17. ..."no doubt" about the lack of intention to insult, that is. I would doubt that atheists and theists are "wired" differently, although 1) I'm no neuroscientist and 2) we'd probably want to clarify what we do and don't mean by "wired" differently, first.

  18. Well, supposedly the brains of theists and atheists act differently to certain stimuli in an MRI. Whether that's because atheists and theists are "born that way" or due to life experience I can't say.

  19. The only thing I've read about that is some of Sam Harris's stuff on belief, and it's fascinating work. Now is an exciting time for experimental & empirical work on religion; I think the hard sciences and religious thought have both reached a place where this sort of work is becoming really a practical possibility and a fruitful endeavor. Boiling it all down to "sentiment" or "reason", though, seems to bring us back to the unhelpful labeling like "enligthened" or "unenlightened" in that article Daniel critiqued on liberals/conservatives and economic thought.

  20. This was the sort of thing I was thinking about: http://secularright.org/SR/wordpress/?p=4324

  21. xenophon, that's my post you linked to :-) the short of it is that atheists have a tendency toward social retardation and lack normal agency imputation. there's a whole area of cognitive psychology focused on this.

    need to go record a podcast, may comment later.

  22. The kind of sentimentality I think plays into the odd readings of the social sciences is a sentimentality about how special mankind is. Insofar as most atheists are probably humanists of some variety, I think they're just as susceptible to that sort of sentimentality. And as I pointed out to Khan, I'm specifically thinking of residual sentimentality about man. You don't have to believe man was created from dust and breathed into 6,000 years ago to have that residual self-esteem boost.

    i see where you're getting at here. and i know that it's part of the criticism of some anti-social "science" types. i'm in particular thinking of cultural anthropologists who "interpret", and don't do science. but not relevant for myself or more readers. we're *sad* that humans aren't mechanistic and robotic, to make them analytically tractable :-)

  23. we're *sad* that humans aren't mechanistic and robotic, to make them analytically tractable :-)

    Social science would be awfully boring and far too easy if they were :)

  24. xeno, yes.


  25. Lee,

    I think Robin Hanson nicely sums up my view on the truth:

    "Since our minds are smaller than the world, the world tends to be more complicated than our mental models of it. Yes, sometimes we think things are more complex than they really are, but far more often reality is more complex than we appreciate."


All anonymous comments will be deleted. Consistent pseudonyms are fine.