All Academic Fields of Study, Ranked by Realness

[Jack]

The whole field of philosophy is currently engaged in a self-referential argument over whether or not it is a "science," entirely because philosophers believe that being a "science" would get them more respect. Much of academia suffers from the same competitive affliction.

We are solving this entire problem, by revealing the exact order of all academic fields of study, ranked by how real they are. (Completely fake fields of study have been left off the list.) What makes one field more "real" than another? Don't act like you don't know. Come on.


1. Physics
2. Astronomy or other Space Science
3. Philosophy
4. Engineering
5. Math
6. History
7. Chemistry
8. Biology or other Life Science
9. Foreign language (Useful type)
10. Computer Science
11. Agriculture
12. Geology or other Earth Science
13. Architecture
14. Literature
15. Law
16. Geography
17. Music
18. Economics
19. Study of Some Foreign Place or Culture
20. Archaeology
21. Anthropology
22. Religion or Theology
23. Art
24. Education
25. Foreign Language (Useless type)
26. Political Science
27. Drama or Film
28. Phys Ed, Sports Management or other Major Designed For Athletes
29. Journalism or "Communications"
30. Business
31. Psychology
32. Sociology


[Image: cb-agulto/ Flickr]

http://gawker.com/5899756/all-academ...ed-by-realness

I can't take this article seriously but it does raise an interesting question: what are the fields of study that you consider dealing primarily with physical reality and which deal more with less concrete subjects? Which deal primarily with objective reality and which focus mostly on subjective interpretations of reality?

Depending on the criteria I'd definitely reorder that list.

[Angry Citizen]

I think engineering is without a doubt the most "real". Engineering is solely about applications, and applications are intrinsically real.

[barts]

In my view, there is some merit in classifying academic fields as "simple" and "complex". Physics, chemistry, engineering, etc. are simple because, in most cases a right answer (with some exceptions) to a question or problem can be found. Moreover, once the right answer is found it can be used to predict or applied to what might happen in the future. The complex subjects like biology, literature, law, economics, sociology, history (it's a long list) often don't have a "right" answer. The subjects are complex because often the variables can't be known or, if known, be predicted how they might influence anything, and the variables inevitably keep changing unpredictably. And what might be a right answer today, may not be in five years.

Too often, it seems, those in the complex fields harbor an envy for what is called the "hard" sciences--some call it "physics envy". They're hard because they lend themselves to math which too often unduly impresses people. Numbers don't lie, they believe, when, in fact, they lie all the time. So enamored are people with mathematics that models are created for predicting things like wildlife population dynamics or economic behavior, which while impressive looking and have some useful but minor research value and a lot of propaganda value, are essentially meaningless when applied to the real world.

Just an observation based on limited data points, but it seems to me that those involved in the simple sciences and academic fields while often good at their specialties are not infrequently hopeless when it comes to the complex fields, while those in the complex fields find the simple fields (apart from the sometimes very esoteric math) conceptually very easy. It's very easy to grasp how to build a bridge. It's enormously difficult to grasp the complexities of politics, finance, social demands, and so on that determine if a bridge will be built at all, by whom, and where, among other things. Building a bridge is simple. Deciding about building a bridge is complex.

[Senor Hoint]

http://xkcd.com/435/

[Angry Citizen]

So enamored are people with mathematics that models are created for predicting things like wildlife population dynamics or economic behavior, which while impressive looking and have some useful but minor research value and a lot of propaganda value, are essentially meaningless when applied to the real world.

Are you aware of how these models are constructed? Their mathematical basis is far more rigorous than you might imagine.

It's very easy to grasp how to build a bridge.

Oh yeah?

http://web.mit.edu/emech/dontindex-b.../emechbk_7.pdf

^ That part is necessary just to build up enough equations to solve for these:

http://ocw.nthu.edu.tw/ocw/upload/8/...pter_10-98.pdf

Using methods like these:

http://en.wikipedia.org/wiki/Gauss%E...an_elimination

Using code like this:

[m,n]=size(a);
c=a;
iter=0;
err=inf;
for i=1:n
c(i,i)=0;
x(i)=0;
end
x=x';
for i=1:n
c(i,1:n)=c(i,1:n)/a(i,i);
d(i)=b(i)/a(i,i);
end
while err>error
x_old=x;
iter=iter+1;
for i=1:n
x(i)=d(i)-c(i,:)*x;
if x(i)~=0
err=abs((x(i)-x_old(i))/x(i));
end
end
x=x*lambda+(1-lambda)*x_old;
end

[I am the author]

So please, don't tell me "simple" sciences are easy. Designing programs to solve for the force in a column under a given (and often variable) load requires more thought than the "complexities of politics and finance" - which we also must often consider as a matter of course. Who do you think writes up the budget requests, anyway?

[barts]

Are you aware of how these models are constructed?

Yes.

So please, don't tell me "simple" sciences are easy. Designing programs to solve for the force in a column under a given (and often variable) load requires more thought than the "complexities of politics and finance" - which we also must often consider as a matter of course. Who do you think writes up the budget requests, anyway?

As I say, simple sciences are easy to understand in principle. Sure, there are complexities, but complexities are in the simple sciences, usually, simply a large number of simple operations. The whole is complex, but the individual parts rarely are. I find it amusing that you used a string of arcane code to make your point. You know that every line of code is, of necessity, simple. It's simple, too, because it doesn't change. In the complex fields the code doesn't work because the variables and premises are often labile, unpredictable, and chaotic.

Try modelling even simple, real ecosystem, and then using the model to predict anything useful. That's complex. In fact, so complex it's proven so far impossible.

Lastly, try coming up with a model to predict whether or not your budget request will be approved.

[Angry Citizen]

Yes.

Elaborate. I'd love to clear up a misunderstanding, since I believe you're afflicted by one.

Try modelling even simple, real ecosystem, and then using the model to predict anything useful. That's complex. In fact, so complex it's proven so far impossible.

Okay. Try modeling turbulent fluid flow through a pipe. It's simple to arrive at the necessary equation, known as the Navier-Stokes equation, through conservation of energy - one of the easiest physics problems in existence. However, it often becomes impossible to model the flow in all but the most trivially simple situations. Damned if I can remember who it was, but a smart fellow once said that he'd ask God to explain two things: turbulent flow and quantum mechanics. He did not anticipate an answer for turbulent flow.

Take another, similar example. You can model the motion of an individual atom with ease. But try modeling the behavior of a mole of atoms contained in a small box. Without statistical thermodynamic assumptions, it would take literally life-ages of the universe for the fastest computers in the world to calculate the behavior exactly.

Now we return to the point. What makes "complex" fields any different? Take predator-prey analysis. Rabbit eats grass, fox eats rabbit at certain rate. These variables interact with one another, and their rates of population growth interact with one another. It's not fundamentally different from a large series of atoms interacting with one another, affecting both position and velocity. The only real difference is that population models, ecosystems, etc are fundamentally discrete events, and thus introducing calculus is in fact an assumption not grounded in reality. This is where errors creep in. However, the general trend modeled by the calculus is valid and grounded in mathematical rigor, and I defy you to show me otherwise.

[Chris the Chees]

Try modeling turbulent fluid flow through a pipe. It's simple to arrive at the necessary equation, known as the Navier-Stokes equation, through conservation of energy - one of the easiest physics problems in existence.

Try learning latin and reading secretary hand. This one is pretty simple and coms with a handy 'translation':



All academic subjects are difficult until you know what you are doing.

[barts]

Elaborate. I'd love to clear up a misunderstanding, since I believe you're afflicted by one.

What do you presume to be my misunderstanding?

What makes "complex" fields any different? Take predator-prey analysis. Rabbit eats grass, fox eats rabbit at certain rate. These variables interact with one another, and their rates of population growth interact with one another.

You've made the classic mistake. There is no discrete predator-prey analysis. Except in the computer. You've listed three variables: rabbit, grass, fox. There are no ecosystems with only three participants. Moreover, the variables are not discrete. Change even one thing like the age of the rabbits, and all predictions become meaningless.

Even an appeal to large numbers to get a statistically right answer, as you do with modelling in the quantum world or with turbulence or thermodynamics, is meaningless in ecosystems.

As I say engineering is simple, ecosystems, or writing a play that people will visit the theater to see is complex. Some simple things can be difficult, no complex things are easy.

[Angry Citizen]

Try learning latin and reading secretary hand.

I'm learning French. It's Latin-lite. Does that count?

[Angry Citizen]

What do you presume to be my misunderstanding?

Nice dodge. I'm not certain, but I'm sure there is one somewhere, especially given your reticence.

You've made the classic mistake. There is no discrete predator-prey analysis.

It would be a classic mistake if I hadn't made it clear that no such analysis exists. Sometimes I wonder if you bother reading my posts at all, barts:

The only real difference is that population models, ecosystems, etc are fundamentally discrete events, and thus introducing calculus is in fact an assumption not grounded in reality. This is where errors creep in.

[barts]

Nice dodge. I'm not certain, but I'm sure there is one somewhere, especially given your reticence.

Dodge? You asked an unanswerable question; perhaps you didn't notice. Were you expecting a 10,000 word treatise on my understanding of theories about mathematical modelling? The fact is you said I was afflicted by a misunderstanding. To say that you must have some notion that misunderstanding might be, but it seems you don't and are now, as you say, dodging. Your statement about any misunderstandings I might have is so vague as to defy response. Misunderstanding about what? Confidence intervals? Data collection? Assumptions? What?

You wrote,

Take predator-prey analysis. Rabbit eats grass, fox eats rabbit at certain rate. These variables interact with one another, and their rates of population growth interact with one another. It's not fundamentally different from a large series of atoms interacting with one another, affecting both position and velocity.

As I say, nothing could be further from reality. There is a profound difference between the interaction of atoms and the interaction of rabbits and other species, energies, and chemicals within an ecosystem. The latter is "simple", the former "complex". While you might not be able to predict where an individual atom might be, you can with some reliability predict what might happen in aggregate to the atoms as energy is applied to them. Increase the energy in an ecosystem, and changes can be expected to the ecosystem but what those changes might be are unpredictable, except at the extremes, because of the complexity.

Indeed, calculus is, of course, used in modelling ecosystems. Why? Because it looks good in scientific publications, and gives the spurious illusion of validity offered by mathematics.

It seems odd to me that those involved in the simple sciences are so defensive when the term simple is applied. The arguments become not rational but rather religious-like. Engineers can build bridges that can last a thousand years. There isn't a human being on the planet who can build an ecosystem that would last that long because ecosystems are complex and dynamic, bridges aren't. Indeed, there's not a human being on the planet who can fully describe the ecosystem in single drop of sea water, but there is a good understanding about the chemistry and physics concerning sea water.