Essay:Thoughts on libertarianism and ego

Eric S. Raymond annoys me, even more so because I used to respect the guy. Sure, I always thought his absolutist position on gun rights was a little wacky, but he always struck me as being a very reasonable person. Although many of the particulars of his seminal essay The Cathedral and the Bazaar haven't panned out (the most successful open source projects tend to build their own cathedrals, either as a mass collective like Linux or coming from large companies such as IBM, Nokia, and Sun/Oracle), a good number of his arguments for transparency and modifiability rights are obvious on their face. Then again, those are the two major points on which he agrees with Richard Stallman, and it's hard not to look at the rest of his career and wonder what the value of his more original ideas is.

It's not pretty when you dig into it. As the canonical maintainer of the Jargon File, he excised a large portion of the older entries from the MIT and Stanford cultures (particularly the MIT AI Lab crowd) based on the idea that ITSisms and the like were obsolete, relegating them to a file known as "jargon-chaff". He has apparently added numerous of his own coinages that have never become live usages, and many of his cultural notes show a heavy bias towards his own libertarianism. His stewardship of the Fetchmail email utility has come under fire for fuzzy design goals and for abandonment (he put the project into "maintenance mode" to make some kind of point about development cycles, from which it was eventually picked up by different maintainers who chose an entirely different development direction for it. And then there's his views on race and gender, as well as his HIV and global warming denial (the latter apparently based entirely on his politics), his Bircher tendencies, and... yeah, go read his page. We don't need to recount it in detail.

You can't tell a man who knows

It's pretty clear if you read through Raymond's writing that he's a cherrypicker, for example defending the denier side of Climategate by claiming the existence of an "error cascade" and quote mining the CRU emails. The "error cascade" idea is essentially a false analogy from chaos theory and computer science -- it implies that if initial data is wrong (or, as deniers would have us believe, outright falsified), that all future conclusions are tainted by the error. This makes sense in computer science and mathematics, but in the scientific process very few theories are accepted based on a body of evidence with only a single line of corroboration; in this case, the hypothesis of global warming is backed up with the real-world effects of global warming such as glacier melt, increased storm activity and temperature extremes, and changes in the natural range of climate sensitive species, among other things.

Certainly most people think that what they believe is broadly correct. But there's a big difference between believing oneself to be correct because they came to the conclusion based on evidence and believing oneself correct because they cherrypicked evidence to defend a predetermined conclusion. (At least in American society, the Dunning-Kruger effect has a nasty tendency to muddy the waters for people who don't quite get the concept of evidence in the first place.) What's clearly going on here is blatant cherry-picking -- Raymond has clearly stepped outside the framework of scientific thought here. This is not that much different from the thought patterns of the typical "hermit scientist" -- there is a prima facie assumption that those within the field are not to be trusted to know how to do their work, and the hermit scientist will do pretty much anything to avoid the established body of knowledge and processes that they want to overturn. This is a fairly common issue for religious fundamentalists, but the psychology here is slightly different -- hermit scientist types tend to consider themselves explicitly anti-authoritarian and therefore frame their positions as fighting against an inept or malicious "establishment". This has become more noticeable with the rise of the Internet -- people such as Stephen Wolfram (cellular automata as grand unifying theory) and Neal Adams (expanding Earth) are prominent examples alongside Raymond's comparatively uncreative brew of prejudice and bad pattern recognition.

Going back over the history of this makes for some interesting reading. Martin Luther's attempts to democratize the Bible seem to be a big part of it -- though it did encourage mass literacy to an extent, it also created a situation where people could make Biblical interpretations without understanding much of the context in which they were written. On the American frontier, where formal schooling was limited and people lived under a sense of self-reliance (at least until they had to call the US Army in to deal with an Indian "insurrection" or something like that), any random person could set himself up as a preacher and base his teachings on the "literal meaning" of the Bible; it shouldn't come as a surprise that this do-it-yourself attitude towards science and politics should flourish especially well in the United States.

The long tradition of ignorantly second-guessing the experts

Once in a great while, a layperson comes along and sees something no expert has ever seen before, and is lionized for it. The fact that this has happened at all has therefore become a central part of the Great American Inventor -- working alone in a lab, defying the conventional wisdom by thinking outside the box. It is beyond question a very powerful image.

There are a few such cases, but for the most part, such people really don't exist. Take Thomas Edison for example, the personification of the Great American Inventor -- in actual fact, Edison had only one original invention (the phonograph) and one scientific discovery (the Edison effect, which led to Lee DeForest's creation of the triode tube). His work on motion pictures, the telephone, and the light bulb were items he was only one of several participating in research on, and his work on power distribution was ultimately a failure since he refused to acknowledge the superiority of alternating current for long-haul power transmission. On top of all that, Edison was a terrible experimenter, using a brute-force approach to the light bulb problem that may have made it much harder for him to find a working filament. Edison's work was certainly influential, but he couldn't have and didn't do it on his own. In fact, it's not hard to argue that the Menlo Park laboratory itself -- the world's first industrial R&D laboratory -- was perhaps Edison's single greatest creation, and the collaborative effort made possible by such facilities (as well as, of course, university research labs, a much older concept) is a big part of what has created the modern technological culture (see also Bell Labs).

But Americans love them some rugged individualism -- with many people, it's seen as an admission of weakness to be reliant on cooperation (pretty much the entire justification for Randian Objectivism for example), and therefore when a hermit scientist manages to put one over on the Establishment (or is seen to at any rate) it's a major coup for many people. In many ways, this sort of attitude drives the national attitude, not only towards science, but history and numerous other fields. It works in reverse too; in a manner akin to the one single proof fallacy, denialists in particular will focus on the most visible face of whatever area of knowledge they're trying to tear down -- Darwin or Richard Dawkins for evolution, Einstein in physics, Michael Mann or Al Gore on global warming issues, Paul Offit on vaccinations -- and use that person's personal foibles as a cudgel to attack the field. Although such tactics have no bearing on the evidence in such a matter, scapegoating visible figures in the field does tend to be rhetorically effective. There seems to be a paranoid streak in American culture (something historian Richard Hofstadter wrote possibly the seminal work on in 1964) that leads people to go past the evidence for something they don't like directly to the cui bono question; though "who benefits" is a good way to start an investigation, putting the cart before the horse like that is rarely productive.

The Mensa Effect; or, how to be stupid when you're really smart in one easy lesson

There is nothing wrong with being a smart person. The problem is that smart people often have a tendency to assume they know better than others; you see this a lot in the transhumanism community for example. The singularity is coming? Really? Well, Moore's Law of course. But that only covers hardware -- you still have to make the hardware work the way you want it. PZ Myers' 2010 takedown of Ray Kurzweil is a key example -- Kurzweil is, by any meaningful measure, a brilliant man, someone who's managed to do incredible things in music and computer vision. However, it's hard not to view him with a certain degree of pity -- his obsession with life extension and the Singularity borders on the religious, and certainly is far out of whack with the current state of understanding of brain development and intelligence.

Thus the reference to Mensa -- although Mensa is occasionally of interest to puzzle fans, for the most part it's really just a circle jerk for smart people with superiority complexes.