Mark the calendar, my graduate studies have just been made useful on the boards.
This is a fascinating question. Thomas Kuhn is the big guy here–lots of the debate and study on this question relies on his theory about sciences.
He essentially keyed it to paradigm development and consensus. Does the field agree on what problems are in its bailiwick? Do they agree on which are the important problems? Are there carefully-delineated methods of finding answers? Do people readily agree on how to judge the results of scientific study, and how much truth they hold? The more the answer is “yes,” the more apt the title of “hard science” is.
So, which are soft? Depends on who you ask, I guess. Physical sciences and mathematical sciences generally are considered to be hard. Some Life sciences and most if not all social sciences are soft.
Perhaps, rather than there actually being hard and soft sciences we have hard(nosed) and soft(headed) scientists. The less wiggle room a science has for simply “supporting” a theory by means of dogmatic repetition, the less tolerance it has for softheads.
Perhaps a good yardstick of scientist classification would be response to the Sokal incident.
Depends on the science in question. One problem with the “soft” sciences is the inability to control the environment in question. The economist’s motto, “All other things equal,” is an idealization as much as the ideal gas law is, but in physics scientists often have the ability to isolate the phenomenon in question a lot easier than, say, an economist can play with the economy willy-nilly just to test his hypothesis. It isn’t that the variables are necessarily unknown.
In real “soft”-sciences research, yeah, the variables are unknown; but as per Ed Lorenz and the rest of “chaos theory”, there’s plenty of reason to think that even if you had the data, the sheer volume of it would swamp any attempt to analyze it.
Can you imagine coding the data for temperature, humidity, and wind velocity for each cubic centimeter of air in the entire atmosphere? Analyzing the effects of every single one of the others on every single one of them to yield it’s next-millisecond state and then plugging in the new data and repeating until you get to next Wednesday? Then toss in terrain variables, surface elevation, abdelo, temperature, whether or not it has rough edges that might affect dew formation, absence or presence of moss with or without moss blight, chemically reactive elements or compounds in the surface rock and how they might or might not accept or release oxygen, and when you have all that data for every square cm of surface calculate for every cc of air the effects of every sq cm of surface area along with their effects on each other (both the surface and the air, of course) and of course vice versa?
And that would just be a very oversimplified model, leaving out the exact position of every living creature and all the biological and psychological variables necessary to predict what each of them is going to do next, including of course their effect on each other, all of which would indeed have an effect on the weather. And vice versa, of course, don’t forget the reciprocal effects. It is generally conceded that the life-form behavior (including that of the humans) would be significantly more complicated to model than the weather, and yet to model the weather with complete specificity you’d have to incorporate all the life-form behavioral model data.
Don’t forget yourself and your activities as researcher and how they are affecting the very things you’re studying. You, and your research, would be part of the total picture, of course.
See what I mean? Even if you somehow had all the data, courtesy of God’s Own Computer or something, you would not live long enough to lay out the equation for one single variable!
A brute-force computational problem. I’ll assume in ‘soft’ sciences, the schema is itself fuzzy. That might change, in psychology, if they ever manage to successfully correlate brain function to mind behaviour.
I knew someone would bring ‘chaos’ up.
I’ve always drawn the line between hard and soft sciences based on falsifiability.
To illustrate the point, I’ll use two ideas. One from physics and one from psychology.
From physics we have the idea of the Luminiferous Aether. In the late 19th century, when light was discovered to have wave-like properties, the aether was hypothesized as the medium through which it propagated.
From psychology we have Freud’s analysis of dreams. I’ll quote the man himself:
Now, the aether hypothesis is falsifiable. That is, the existance of the aether would have certain observable effects on measurable phenomenon. If you set up an experiment and make the measurement, and you don’t see these effects, then the aether (as you have hypothesized it) doesn’t exist.
Freud’s hypothesis, on the other hand, is not falsifiable. Every dream can transformed into a “senseful psychological structure”. You cannot, even in theory, devise an experiment that will test whether the interpreted dream accurately represents any particular psychological structure.
Note that it doesn’t matter whether these ideas are correct or incorrect. The aether hypothesis has the quality that you can devise an experiment that could prove it wrong. Dream analysis lacks this quality.
(I do believe that psychology is becoming a harder science. Using an example from the 19th century maybe a somewhat unfair characterization. On the other hand, I think it has a long way to go before it really sheds it’s heritage as a kind of erudite guessing game.)
Maybe this discussion would have been better suited to IMHO.
Just because one applies scientific method doesn’t mean what follows is science. Maybe part of the debate centers around the differing definitions of “science”. It seems some are arguing the first definition “the state of knowing : knowledge as distinguished from ignorance or misunderstanding” while others are arguing the third “knowledge or a system of knowledge covering general truths or the operation of general laws especially as obtained and tested through scientific method” http://www.m-w.com/cgi-bin/dictionary?book=Dictionary&va=science
To those who posit that the soft sciences are too complex to be defined in absolutes. That position seems to beg the question. Why is it too complex?
While an animal behaviouralist cannot take 5 chimps and blend them into a slurry to obtain meaninful results, two obervers witnessing the activity of 5 chimps might report the activity differently. What one sees as play, the other might see as foreplay.
I’m inclined to agree with AHunter3’s assessment that many fields of study "have a major envy-of-physics thing going. They’d do themselves a favor by going under the category of art rather than science. There’s nothing ignoble about art is there?
In a broad sense, I think that anything that deals with value (as a measure of worth and as opposed to amounts and such), should be considered an art not a science.
Sorry that this post is so disjointed. I don’t have time to collect my thoughts, yet I don’t want to appear disinterested in the thread.