EN World scientists...

[Plane Sailing]

BSc in Geology, Msc in Distributed Multimedia Computing and although I always dreamed of growing up to be a scientist, being a computer programmer (of various sorts) rather took over. Slightly better paid for me too

 

[Jdvn1]

So long as you treat it as real science, yes.

Contrary to popular opinion, "hard" science does not refer to difficulty, but to the physicality. A "hard" science has physical objects and results you can point to. "Soft" sciences are not easy - their targets simply aren't as concrete.

Did you mean to draw a parallel between "real" and "hard," or is that merely a side comment?

If it's a side comment, I think that's an interesting definition, considering string theory wouldn't fall under your definition of a hard science, according to some physicists I know.

If it's not a side comment... I'm sorry I brought it up, considering the discussion it's likely to spark.

I'm working towards a BS in Economics, and was told today that I will be accepted to grad school by the lady who makes that decision--she just hasn't sent the letters out yet.

I might consider Economics a "real" science in that we have mathematical models that describe reality to a reasonable extent (depending on your definition of 'reasonable', I suppose, but we can say some concrete things... and, meteorology is considered a science anyways). I also take it "real"ly seriously, as an academic pursuit, too.

A lot of research, though, doesn't have physical objects and results you can point to, though it does happen at times in specific cases. The research can be statistical, or testing of models to see how closely they vary to real life. My understanding (I have a lot to learn, still) is that tests of basic supply-demand models hold very firm, with relatively concrete results. It may be considered "soft" because results may not be as concrete as most natural science research (but then, what do you call string theory, math, and such?). It's not as "soft" as Psychology or Sociology, I'd argue.

 

[Umbran]

Did you mean to draw a parallel between "real" and "hard," or is that merely a side comment?

Neither, unless you feel that noting that some folks who claim to practice soft science don't actually do so is snide. I have great respect for those who do soft science as science, and a rather strong disdain for anyone claiming to do science when it is clear they aren't. Pet peeve of mine.

I might consider Economics a "real" science in that we have mathematical models that describe reality to a reasonable extent (depending on your definition of 'reasonable', I suppose, but we can say some concrete things... and, meteorology is considered a science anyways). I also take it "real"ly seriously, as an academic pursuit, too.

I'd call that science, so long as you compare your models to reality, are honest in measuring your errors, and modify your models when they don't match reality. You know, scientific method and all that

It may be considered "soft" because results may not be as concrete as most natural science research (but then, what do you call string theory, math, and such?). It's not as "soft" as Psychology or Sociology, I'd argue.

Well, just for the record - math is not an empirical science. It is, perhaps, what you might call a "formal science". Pure math doesn't correlate to real-world phenomena in any way, shape, or form, so there is no such thing as an experiment to test if a mathematical theory is accurate. Math is great stuff, and amazing tool, and I love it. But it isn't testable, so it isn't empirical science.

Until such time as string theory produces testable predictions, it isn't empirical science, either.

 

[hong]

Right now, I manage projects in MIT's IT shop.

So... you're a moderator in real life?

 

[freyar]

Interesting link, Umbran. I can see the distinction you want to make between empirical and formal sciences, but I'm not sure if the distinction between "formal" and "natural" sciences is really as clear as that Wikipedia article implies. At least I can point you to one philosopher of science who disagrees.

Regarding math, I think we may have to agree to disagree. Self-consistency is an extremely important test, though maybe not an empirical one. Math is accurate (ignoring human error); what might be inaccurate is a given mathematical model proposed to describe some phenomenon.

As most-likely the only string theorist in the room, I also think I need to mention a couple of things. First off, self-consistency and consistency with empirically correct models are very difficult to achieve in modeling any fundamental theory of physics. In fact, it's difficult to convey how stringent consistency is. So the fact that string theory passes these tests is in itself an achievement, even if not an empirical one. Another point is that many string theorists (and scientists in bordering fields) take making predictions very seriously. There's been a tremendous amount of work over the last 5 years on building concrete string models of cosmology for comparison to precision tests, like the WMAP and upcoming PLANCK experiments. These are not definitive predictions because, well, nailing down all possible string constructions is not possible right now, but it is some progress. And people are really trying to "compare your models to reality, are honest in measuring your errors, and modify your models when they don't match reality," as you say. I can provide references on request...

 

[Umbran]

Interesting link, Umbran. I can see the distinction you want to make between empirical and formal sciences, but I'm not sure if the distinction between "formal" and "natural" sciences is really as clear as that Wikipedia article implies.

Well, of course not - it is Wikipedia not "The Authoritative Source for Everything". It got the basics down. The subtleties are left as an exercise for the reader

Regarding math, I think we may have to agree to disagree. Self-consistency is an extremely important test, though maybe not an empirical one. Math is accurate (ignoring human error); what might be inaccurate is a given mathematical model proposed to describe some phenomenon.

Um, I think we may be talking past each other here. To me, this is a matter of definition - calling math an empirical science is like calling an orange a poodle.

Empirical science compares a theory to collected real-world data as a check of the accuracy of the model. Mathematics cannot do this, as mathematics itself makes no claims on what real-world data it should be checked against.

A theory in a formal system can be proved or disproved with respect to that formal system, and that's all. There is no experiment, no taking of data, no observation of reality is required, or even called for. It is thus not possible for it to be empirical, by definition of the word "empirical".

As most-likely the only string theorist in the room, I also think I need to mention a couple of things. First off, self-consistency and consistency with empirically correct models are very difficult to achieve in modeling any fundamental theory of physics.

Just so you know - my thesis work was on computer modeling of spin propagation in high-energy jet formation. I know whereof you speak here.

It still remains - if the theory does not make predictions that can be tested, it does not sit in the realm of empirical science. It may sit outside for practical reasons ("Sorry, the technology to make the measurements you want doe snot yet exist"), or it may sit outside for more fundamental reasons. But if it cannot be falsified, it isn't empirical science, because what folks call the scientific method cannot be applied.

That doesn't make it unimportant. Or easy. Or fake. It doesn't make the people who do it any less hard-working. It may be that some version of string theory will, in time, come into the empirical realm. But as far as I'm aware, none of them are there yet. As far as I'm aware, they're all still in the formal realm - a formal realm that's informed by empirical theories, but not yet emerged to the measurable.

That's okay - the same was true for General Relativity once upon a time.

 

[Jdvn1]

I'd call that science, so long as you compare your models to reality, are honest in measuring your errors, and modify your models when they don't match reality. You know, scientific method and all that

Ah, I want to make the distinction that not all economic research deals with models. Sometimes it's more similar to statistical analysis, sprinkled with theory. I think dealing with models is more clearly a science, but statistical analysis is more similar to whatever one may call psychology (my impression is that you may call it a science, but I know people that wouldn't).

Well, just for the record - math is not an empirical science. It is, perhaps, what you might call a "formal science".

Hah! Well, I just mentioned it because some may consider mathematicians scientists. I do agree with what you say, there.

"Formal science," I thought was a term that died off in the 70s! My first encounter with the term was years ago when I picked up a 1969 book entitled Philosophical Essays on Curriculum. Just to provide a definition for the term for everyone else (or, at least, another opinion on what this may refer to), here's how the introduction describes it.

In logic and mathematics, our procedures for arriving at the conclusions are most settled. Moreover, the conclusions are certain, unlike conclusions in every other field. We are sure, for example, that in formal logic, A is A and that in the Euclidean system of geometry the shortest distance between two points on a plane is a straight line. We know this sort of thing as surely as we can know anything.

In the physical and natural sciences, procedures for arriving at conclusions are well established but are not as well established as those of the formal sciences. In the formal sciences no factual claims are made. In contrast, the purpose of the empirical sciences, that is, the physical, natural and social sciences, is to make reliable factual claims about the world. We are not as sure about the truth of factual statements as we are about the truth of formal statements. The method we have for finding the shortest distance between two points is more reliable than the method for finding out whether nails rust in damp air.

I think that's pretty clear, and it also somewhat echoes what Umbran is saying about the applicability of the scientific method.

 

[hong]

Ah, I want to make the distinction that not all economic research deals with models. Sometimes it's more similar to statistical analysis, sprinkled with theory. I think dealing with models is more clearly a science, but statistical analysis is more similar to whatever one may call psychology (my impression is that you may call it a science, but I know people that wouldn't).

Statistics is information engineering. Well, applied statistics anyway.

 

[freyar]

Yeah, I think you're right that we're probably talking past each other. I generally agree with what you're saying, though I do think there's more to the scientific method than falsifiability.

Anyway, don't want to threadjack my own thread , so I'll step back and see if anyone else shows up.

 

[Pbartender]

Eh... don't look at me. I'm a Technician.

As far as I'm concerned even the Engineers are flying backwards and half-blind most of the time.



You'd be surprised how often we have to tell Physicists that they can't do what they want to do, because the real world doesn't work that way.