27 February 2013

On Scientism, Data, and Scientists

Scientism

Steve Sailer discussed priming a couple of weeks ago, and linked to an article that noted some of the problems facing the science of psychology:

At the same time, psychology has been beset with scandal and doubt. Formerly high-flying researchers like Diederik Stapel, Marc Hauser, and Dirk Smeesters saw their careers implode after allegations that they had cooked their results and managed to slip them past the supposedly watchful eyes of peer reviewers. Psychology isn't the only field with fakers, but it has its share. Plus there's the so-called file-drawer problem, that is, the tendency for researchers to publish their singular successes and ignore their multiple failures, making a fluke look like a breakthrough. Fairly or not, social psychologists are perceived to be less rigorous in their methods, generally not replicating their own or one another's work, instead pressing on toward the next headline-making outcome. [Emphasis added.]

Some scientific disciplines are certainly more reliable than others (eg. physics vs. psychology), and some disciplines can’t even be considered reliable at all because they cannot falsified (eg. evolutionary psychology or “theoretical” anything).  In light of this, it’s incredibly helpful to view any claim that, “science shows X to be true” with a suspicious eye.

In the first place, a lot of press releases made by scientists don’t actually follow the scientific method (archaeology seems to be especially guilty of this, because discovering an artifact is not anywhere close to being the same as forming and testing a hypothesis); they simply post a theory (not the same as a hypothesis, by the way), and don’t bother to test it.

In the second place, a lot of scientists don’t bother to replicate their tests.  This seems prevalent in psychology, but in the health and medicine field as well (particularly in regards to the health benefits of certain foods and supplements).

This isn’t to say that scientists never test and replicate their theories, only that there are a decent number of occasions when they do not.  The lack of rigor, though, makes Scientism is a foolish belief system, because not all scientific disciplines, or even scientists, are equal, and it is therefore ludicrous to place the complete trust in all the various scientific disciplines.  It is even ludicrous to place as much trust in psychology as in neurology, or as much trust in biology as in physics.

Furthermore, bad science can be quite profitable.  When a science release uses lots of scary imagery to sell its findings, or when a scientific press release seems to have findings convenient to the needs of a political advocacy group, or if a scientific press release sounds suspiciously similar to a marketing campaign, you can be pretty sure that whatever research was conducted was done so for profit, and not necessarily the pursuit of truth or better understanding the world in which we leave.  Like religion, science has its own cult of believers and adherents.  And, like religion, science has its share of phonies and fakes whose only interest in science is exploiting the naiveté and ignorance of believers for personal gain.

Scientists

Perhaps the reason why adherents to scientism are so susceptible to being duped by disingenuous researchers masquerading as scientists is because most adherents are likely a) disingenuous as well and b) not as intelligent as they would like to think.

Regarding the former, it’s intriguing to note that, in the US especially, there are a large number of scientists and, tellingly, research institutes.  These cost money.  Most scientists and RIs have to justify their existence and need for funds, and so every last one of them must come up with cutting-edge or novel research to justify further funding.  As such, science by press release becomes almost a necessity since RIs need people to be aware of their existence and necessity.  Thus, research can be rushed or not double-checked (or theoretical), and thus a lot of bad science goes mainstream because a lot of scientists would rather have cushy jobs at a research institute instead of teach school, work at a pharmacy, or other more menial, degrading labor.

Regarding the latter, there seems to be an over-supply of scientists, as evidenced by their high unemployment rate.  This would suggest, as Vox points out, that most scientists don’t have a basic grasp of the very simple economic law of supply and demand.  If they can’t figure this out, then perhaps it would be a bit much to expect them to be able to tell whether a press release from a research institute is bullshit or not.

At any rate, most scientists do not appear to be as intelligent as they perhaps believe themselves to be.  In fact, adherents to scientism are likely quite stupid because their simplistic faith in science, which in turns causes them to be quite arrogant, is indicative of a foolish and ignorant mind.  Those who have ever bothered to deeply study a specific scientific discipline, or even epistemology, are inclined to acknowledge their ignorance far more than their knowledge.  It is the truly knowledgeable who are most aware of their ignorance and the limits of their knowledge, and only fools place their faith in knowledge.

Data

The fundamental matter of science is data.  Where adherents of scientism go wrong is believing that data is somehow objective and intrinsically meaningful.  It is not.

In the first place, while the content of data may be objective, its selection and definition is inherently arbitrary (see here, here, and here for more).  Or, as mathbabe put it:  “Don’t be fooled by the mathematical imprimatur: behind every model and every data set is a political process that chose that data and built that model and defined success for that model.”

In the second place, there are limits to data.  As David Brooks noted, data struggles with social, it struggles with context, creates bigger haystacks, can’t handle big problems well, has a tendency to favor the memetic, and obscures values.  It also has a hard time handling small problems (especially when the data appears to have no pattern to it). Further, data cannot provide a value judgment since value judgments are intrinsically subjective.

When all is said and done, data is not nearly as useful or as meaningful as most science fetishists imagine.  It appears that data is sometimes used to rationalize predetermined policies or, worse yet, obscure reality.  Sometimes it’s just irrelevant (like CPI) to its intended use.  Trusting in science because it relies on objective data is shallow and misguided, and, frankly, false.

In Conclusion

Science, as a method of acquiring knowledge, is a wonderful tool for understanding the world.  The knowledge that the scientific method provides is often useful and certain.  However, most of what is passed off as science is actually just “science.”  There are many charlatans who pay lip service to the scientific method, and then pass their decidedly less-than-rigorous research off as science, leaving science fetishists, many of whom are shallow and foolish, to believe press releases that are often later proven to be false, misleading, irrelevant, or highly contextual.

As a result, many people have rightly come to distrust that which is passed off as science, while others have doubled-down in their beliefs.  This has caused quite a divide, and prevents people from see the world as it is.  When science is completely riddled with lies and irrelevancies, real science gets lumped in with and likewise tuned out.  Alternatively, those who are dedicated to the cause of science get duped into believing nonsense spouted by charlatans, feeding into a tribalist mentality.  The result is that you have one group of people ignoring legitimate science because they view all science as bullshit, while another group of people constantly mocks and belittles the doubters while not realizing that most (not all) of that which is proclaimed in the name of science is actually bullshit.

Truth be told, though, scientists have only themselves to blame.  Scientists were supposed to be the gatekeepers of knowledge, but they were corrupted by power and money.  This is why anthropogenic global warming still has credence in a lot places while theories of gene-based human biodiversity remains ignored (and studying the latter would likely have more measurable benefits, particularly in the field of health and medicine).  Because of the corruption of scientists, Science now has a bad name, and fetishists are busy having a pissing match with agnostics.

The irony in all of this how that which is passed off as science has become increasingly useless while its fetishists have become more bold and shrill in proclaiming science’s near-universal ubiquity.  Consequently, this means that science was far more useful when everyone understood its limits.  By trying to make science explain everything, it is no longer capable of explaining anything.  In order for science to be useful again, its adherents need to accept and understand its limits.  Once that happens, agnostics will trust science more, and the debates over science will become more meaningful and less rancorous.  In the meantime, though, the self-abasement of scientists and fetishists remains amusing.  Too bad it won’t end well.