f13.net

I'm getting ready for my comp exams which start next week.  My head hurts. I'm tired. And I have to read shit like this:

You just put Lunesta out of business.

The goal of science is to make the complex simple and the goal of social science to make the simple complex.

The purpose of social science is to use specialized vocabulary and a dense, almost impenetrable writing style, plus statistics with lots of pointless data and exhaustive but ultimately trivial "analysis" to hide that they actually have not the slightest fucking idea what they are talking about, and in fact (edit: nearly) everything that's been done in the field for the last 70 years is a complete fucking waste.

If this is essential to your major, save yourself from a lifetime of uselessness and change *now*.  There are fields tangentially related to the "core" social sciences that are actually useful.

--Dave

EDIT: To tell the difference between the useful and the useless: If you pick a paragraph out of the middle and a normal intelligent person cannot firgure out what the fuck they are trying to say, it's almost certainly in the first group.  Physics is complicated and requires special jargon as well, but your typical physics textbook, even at the advanced level, is not complete gibberish.

Heh. Always funny when people make that claim.  Most people have no clue what real political science is because they never took it past undergrad.

This isn't "typical textbook" stuff. That is from a journal article. I know perfectly well what that says, and it provides good information. Reading it for 6 hours every night after work and all weekend is a drag though.

Okay, so if I'm wading through all of that correctly, it's talking about an essay that wrote a bunch of crap about how people learn to sort themselves out into productive groups and how external factors influence that.  And apparently the essay was very influential among people who read the essay and gave a damn.

My question would be, why does that matter?  I mean, if someone manages to unravel the core of human behavior to the point of being able to predict human events ala the Foundation books, then yeah it's useful knowledge, but otherwise it fails to benefit anybody, and I don't see how it becomes useful in getting a job at anything other than teaching sociology.  Understanding human behavior and being able to point and say, 'Gee, that guy bashed in that other guy's head as a result of the following social factors...' is all well and good, but it doesn't actually help the guy with the bashed-in head or actually change anything.

If they quit teaching whatever the fuck it is you're learning, if the entire field of study just evaporated overnight, would anyone outside of it notice?

I'm an academic, so take what I say with a grain of salt.  Things don't have to be immediately applicable to be worthwhile.  There are MANY many examples in science where people did research that had no "point" to it at the time it was done.  Some of that work laid the foundations for current groundbreaking efforts.  The key is to go off in new directions and blaze a trail for others.  Sometimes the trail leads nowhere.  On rare occasions, the trail leads to a goldmine... it's just that the goldmine isn't discovered until decades later when we have more tools at our disposal to flush out the details. 

Just because something doesn't immediately seem worthwhile doesn't mean that it isn't the start of something that may grow to be.   

The piece in the first post sounds like what I expect Tycho to sound like on his blog posts 2 years from now.

Making things inaccessible to everyone except the people that already know and understand it seems.... counterproductive.

Also, pretentious.

Many of these papers are written in journals that only specialists in the field subscribe to.  They are meant to be written with a highly specialized audience in mind.  Often they get converted to more approachable language if the concepts become more widely accepted.  I'll give you an example.  In chemistry, it's far easier and far more to the point to describe things in very technical language not to be pretentious, but to be precise.  It saves space and gets to the point without all of the flowery details that the mainstream publications like to add. 

Of course, there are a lot of arrogant assholes in academia... just keep in mind that there are a few of us that aren't.  We are forced to play along in order to maintain funding to pay our students and techs a salary. 

Or in language us computer geeks would know:

I can write out the algorithm for a novel new sorting mechanism in a few dozen lines of pseudocode, and anyone will be able to figure out what the hell is going on instantly. (At least, among people who code. OR I could take about ten pages to explain it in English, and confuse the hell out of people.

A lot of scientists use dense jargon to get complex ideas and subtle differences across quickly and preciesly. Mathematicians use equations, programmers use pseudo-code (and god knows, our own jargon), social scientists use that, etc.

I can see this sort of shit being useful in math, chemistry, biology, and programming and such - where there's an entire language that's accepted and preferred.

In anything else, it just seems ridiculous.

In philosophy I've got experience of people getting into long fucking pointless arguments because someone uses a term that isn't precisely defined. Likewise in some of the more obscure and jargonish philosophers (especially the continental style ones) they will use jargon sounding words to convey ideas that would take a paragraph to explain. When you're writing for an audience you would expect to be familiar with such terms it's far easier to just use them than attempt to explain them all right at the beginning of an article every single time.

In other words, yes there are plenty of people who use jargon just to put off others and sound cleverer but using technical terms can often be a precise and efficient way of communicating complex concepts to people who are going to be familiar with that jargon.

Yeah, except that all the above statement says is that people in large structured bureaucracies tend to develop a preference for, and reward the following of, abiding by the standards of the large bureaucracy.  BUT, it doesn't make that claim, it's just reiterating what a couple of other people said.

It is, actually, entirely worthless because I am pretty sure Asimov noted the same thing in Book 1 of the Foundation Trilogy in 1950.

The point being that, in social sciences (political science, sociology, psychology, and all of that) -- there IS an entire language that's accepted and preferred, complete with a lot of words that have very specific meanings and connotations that are different than their usage in everyday English.

You just don't know that jargon, and you probably know less of those fields than you do of things like chemistry, biology, and programming.

The shit in that paragraph, however, is still someone using big words to confuse the issue. So says my roommate anyway, I can't even parse it.

Probably, but I've read CS papers that suffer the same damn flaw. Difference is that in computer science, I can tell bullshit from interesting shit because I know the jargon and I know the material. I'm not qualified to judge, say, political science stuff like that. (Although I can generally spot the more egregious games of silly buggers with statistics).

Heh. I should post a formal rational choice model in here and see what you guys think.

This kind of stuff is why MBA programs are teaching business communication/writing classes as a requirement now. Undergrad academia forces you into a style of writing that tries to be confusing/longwinded in the hopes of sounding educated on the topic. Some people naturally just assume if they can't understand all of the details, it must be intelligently over their head. People tried to do this in the business world and get blasted for it because the audience doesn't want to hear flowery details with fancy language. Business people want the information and they want it quickly. Thus you get "Executive Summaries" on basically everything that's more than 10 pages long.

Warren Buffet said: "There seems to be some perverse human characteristic that likes to make easy things difficult." I would agree, and it's something I think the business community is actively trying to eliminate in newer business students.

I can see the point of this.  I work with an engineer who is VERY long winded and goes down to the molecular level explaining how the spin of the electron is causing a slight increase in dI/dt which in turn is causing the voltage isolation boundary to be stressed on a sub-micron scale which explains why a module burns up after 20 hours of hard run time and thus is causing this one IC on the other side of the board to occasionally trip false positive errors before completely ceasing operation with no bangs or warnings.  It's great to have the technical details for the record, but the suits in charge of "do I escalate this to the CEO?" over customer failures don't give a damn.  They want to know a) that you know why it failed, b) what you are going to do about it, and c) how long it is going to cost\take.  If it can't be summarized in 3 easy bullet points you have failed your mission.  If the CTO wants to know, you give him the gory details.  If the Project Manager or the Director of Customer Relations wants to know, you give them the short and sweet bullet points (but make sure you aren't lying or BS'ing).

The point of jargon in most fields is to *reduce* the amount of words required to explain something (to someone who knows the jargon and what it is built on).  In the social sciences, it usually acts in the exact opposite fashion.

Social science work doesn't have to be impenetrably dense and encoded.  Only when it is trying to hide the fact that what it has to say is actually fairly trivial.

--Dave

I would love to read the thesaurusized version of a study entitled "Why men like boobies."

I personally think that understanding how Congress, the Presidency, the Judiciary, the Bureaucracy work, how voters process information, what informs vote choice, how public policy is made, etc. etc. are not "trivial," but whatever.

But, why does it have to be cryptic? I read that and, I don't know, it just doesn't make sense. I'd like to know how politics work, why do I need to pay money to find out?

I can rephrase it without losing significant meaning:

Organizations in similar sectors structurally organize similarly regardless of the effectiveness of that organizational structure.

I suppose I lost some of the names mentioned...

Also "isomorphism" doesn't mean what the author thinks it means, in fact it means basically the opposite of what the author intended. (Isomorphic things look different but are structurally the same, whereas the author says that the similarity is surface-level) . Jargon is fine when it is well-defined, as someone else pointed out a big problem in some disciplines is poorly-defined jargon. Richard Feynman wrote about that sort of stuff, where a philosophy class would be discussing some concept and it turns out everyone was talking about something different.

Well, it was a lot better than some of the stuff Women's Studies / English Departments put out, I'll give it that much. It did at least say *something.* It was just overly verbose for what it said.

That's from a sociologist I believe, reviewing a book taking a sociological look at organizational structure. Sociologists are wankers.

Theodor Adorno wrote a wonderful essay called "Resignation" that touches on this.  It reads more than a bit like a justification for academic inaction in the face of injustice...  And it is, now that I think about it.  The argument is that there is a major role to be played for just thinking - not acting, but just thinking.  When action is expected, thinking becomes reactive and prescriptive, and loses its uniquely insightful character.  That's not at all to say that instrumental thought isn't worthwhile, but that there needs to be room for people to just think.  Not everything needs to be productive.  Not everything should be.

I'm so hopelessly over on the continental philosophy side of the argument though.  Give me Heidegger and Deleuze and Nietzsche and Agamben and Foucault and Badiou and Derrida and totally impenetrable prose and man, I'm happy.

Edit: And good luck on the comps.  I'll be in your shoes in a year's time.

But it didn't bother to ask "Why".  If it was parallel evolution, common origin, or perhaps a local optimum in the organizational possibility space (with proposed dimensionality and fitness criteria)?  Did they find cases where organizations in similar sectors organized *differently*, and examine them for comparative value and possible causes?

Lots of words to support a trivial observation, with not the slightest *clue* that something deeper even might be at work.  The investigational and analytical tools they needed developed 70 years too late, and now they compete in the jargon equivalent of the Obfuscated C contest.

--Dave

Holy crap. You are making way too many conclusions based upon one paragraph out of a book review.  Did a social scientist touch you in a bad place or something?

No, just wasted 6 months of my reading time and $2000 or so of my money before I realized that the reason I wasn't finding anything deeper than trivial observations of human behaviour not much better than "Everything I needed to know I learned in kindergarten", but at far lower value density, was because there wasn't anything *there*.  Everything worthwhile the field had tools to find had been established 70 years ago or more, and everything since then was just political poll analysis and mental masturbation to justify grant money and satisfy "publish or perish".

Eventually I moved on to some of the more rigorous portions of Anthropology, and the *much* more productive fields of Evolutionary Psychology and Social Simulation.  But that 6 months trying to sift tiny nuggets of base metal out of thousands upon thousands of words of obscurantist *shit* still bugs me.

--Dave

Dave, what the hell are you doing out of the politics forum??  :-D

You apparently didn't read very much then or went to a shitty program.

Then is there something you would recommend I read?

--Dave

Depends on what you are interested in. I suggested one book about how Congress works that is particularly good in the other thread.

Unfortunately, the books I am most interested in don't seem to have been written yet.  As you might have taken from my "Government as a problem in information processing" thread, my current interest is looking at social environments as dynamic processes, and applying the tools of emergent systems and information processing to them.  I had some interest in "Social Systems" theory, but eventually realized that in the absence of "emergence" research when it was developed, it had degenerated into a formalism for bridging narrative-based sociology with statistics-based sociology.

So my greatest interest would be in work that focuses on the *procedural* analysis of organizations, how they behave in a stepwise fashion, your pointer in the other forum looks very interesting.

--Dave

There are actually multiple books and articles on information processing in political systems and subsystems in government. It's one of the hot areas right now. I'm too short on time right now to dig a list up, but if you remind me after November 9th I'll get some for you.

Ew Evolutionary Psychology. Talk about scraping the bottom of the barrel. Evo Psych is nothing more than a bunch of "just so stories" that sound plausible but have no actual evidence in support of them, no predictive power and no falsifiability.

It's always great fun reading about an Evo Psych theory then concocting your own plausible but opposite theory.

Until you start mixing it with Cognitive Neuro-Psychology, then it starts becoming a much more solid science.  Not that anything in that direction is past the very earliest stages, yet.  It would be more accurate to say that it is not *yet* falsifiable, much like most of the physics of the 1930's was beyond their ability to verify (hell, we *still* haven't fully verified Quantum Mechanics).  It's certainly a hell of a lot better in that department than any *other* school of psychology.

--Dave

EvoPsych will never be falsifiable because it has no predictive power. I suppose you could come up with some EvoPsych theory and then discover some lost tribe someplace that proved or disproved something about it somehow...who knows...

There is nothing *to* verify for the most part. What does EvoPsych say about what will happen tomorrow? Nothing really.

Let's play the game. Point me to an EvoPsych theory and I'll come up with a plausible but exactly opposite theory -- it's really great fun.

First point me to *any* non-Behaviorist theory of psychology that satisfies the same criteria of prediction and falsifiability.  :-D

The core tenet of evolutionary psych, that many parts of our mental processes are genetically derived and therefore the result of a selection process, is pretty damned hard to argue with.  Where (or if) you transit from derived to learned behaviors controlling actions, and what role (if any) volition and consciousness play, is at least potentially a solvable problem using evolutionary psych tools.  It has at least the potential of being a *true* "science of the mind", something no other school of psychology has offered since Behaviorism.

When you compare Evolutionary Psych's "Just So" stories with Freudian theory, or even so-called "Cognitive" theory, it's a marvel of scientific rigor.  It's easier to attack *because* it aspires to rigor and precision, rather than just being a flat-out confabulation worth no more than the reputation of the adherents.  It's "Germ Theory" vs. the "Humors".

--Dave

That's true, but that says nothing about individual theories and how to derive them. Also EvoPsych to my knowledge does not rely much on actual biology or genetics. It isn't grounded in "this gene causes aggression and as we can see over time this gene has become more prevalent due to the following reasons..."

That would be the scientific approach, if it is based on genetics start with the genetics and then somehow trace the genetic traits through time, and only then explain why the traits propagated the way they did. In the end *why* they propagated the way they did probably doesn't even matter very much vs. the fact that they did at all.

True enough, and there is some work being done there in fruit-flies.  In fact, one of the *first* phenomena that argued for EvoPsych was that a genetic change (resistance to a pesticide) had a behavioral effect (putting their pupae at the edges of the petri dish, when they were normally random).

That's also why the work in Neuro-Cognitive Psychology is so critical.  As we find out more about how brains actually *work*, we open doors for tracing the influence of genes on behavior.

Beyond that, the progress in a-life and social simulation ties in as well, if we can show a *step-by-step* simulation of how a behavior can be selected for, it greatly increases the strength of the argument that it was.  Of course, this amounts to an extremely detailed "Just So" story.  But as the models get more sophisticated, it's very probable that they'll start showing predictive power.  In fact, they can do a very *good* job of post-dicting insect behaviour, already.

--Dave

EDIT: The point is that it's not just about a "Just So" story.  There's an implicit assumption that a selection process is a selection process, if you can define and run a simulation of behavioral interactions that shows how a behavior can be selected for, then you've got a good argument that it *was* selected for in a way very similar to that.  Since other forms of psychology use incredibly baroque and frankly arbitrary frameworks to explain behavior, it seems odd to attack EvoPsych for *that* particular sin.  Can any modern psychologist *really* predict a psychotic break or suicidal depression on the basis of anything but their own experience with the patient and the resulting "gut feeling"?

That the change in your simulation was a numeric weighting and the change in the real world was a gene that produced a different protein which acted through a possibly *extremely* complicated causal chain could be significant, but only once you've gone to the *next* step, which is trying to predict how a desirable behavior may have indesirable side effects (that presumably don't outweigh the advantages from the main effect), or how a mutation can pre-adapt for other mutations. 

That would be Maxwell or Einstein to the current Gallileo or Newton stage.  Newton couldn't explain *where* the force of gravity came from, only describe how if something like it existed, it would have certain implications.