Saturday, July 31, 2010

Truth versus Equality

Larry Sanger is interviewed in Slate. Some familiar stuff, in particular from an essay he wrote in 2007, but it's worth being reminded - Larry generally gets Wikipedia theory right. He talks about "the complete disregard for expert opinion among a group of amateurs working on a subject, and in particular because of their tendency to openly express contempt for experts", leading to experts - people who make it their life's work to get things right - being driven away by people who have no respect for that kind of mission. He claims this is a worldview common to many programmers: no one should be regarded as more reliable on a subject than anyone else. He calls this "epistemic egalitarianism."

And he emphases a point that I have made frequently here (for example in this post), that Wikipedia is mostly good when it comes to the hard sciences, and particularly on the sort of facts and figures you might look up in Whittakers' Almanac, or the Guinness Book of Records, or Every Boy's Handbook, but a disaster when it comes to the humanities and social sciences.

And of course it is useless and anything politically contentious. I spotted this gem last week, from the article Modern history of Cyprus.

"Efforts to get Turkey to end its occupation and re-integrate the Northern areas
into a Federal structure continue but to little avail, as Turkey continues to
ignore the UN and the European Parliament and continue it's illegal occupation
of the island. The Cypriot government, in accordance with the opinion and
rhetoric of the UN do not recognize any sovereignty that the puppet Turkish
administration have and therefore do not allow International flights or free
trade in the occupied sections of the island. Discussions are taking place to
try and remove these 'embargoes' as it has been argued they violate the human
rights of the citizens in the Northern areas in a similar manner that the
Israeli imposed embargo on Gaza does."
It has been there for months and seems to be part of an ongoing war between the Greek and Turkish Wikipedian-Cypriots. This edit is curious.

Wednesday, July 28, 2010

Plural quantification and scholasticism

More discussion of plurals at the Maverick Philosopher. Very topical now, but it has been discussed before. Thomas Aquinas:

Unde videtur quod quidquid potest facere unus, possint facere multi, et non e
converso, sicut multi trahunt navem quam unus trahere non posset
- "it is clear
that whatever one can do, many can do, but not conversely. For example, many
persons drag the boat which one person could not drag". Summa Theologiae
III 67 a6 (See also Sententia Metaphysicae, lib. 5 l. 2 n. 11).

Aquinas was writing in the 1260's, but the idea is probably older than that. See e.g. Lambert of Auxerre:

Similiter est suppositio confusa sed immobilis quando signum additur
termino communi in singulari vel in plurali et tenetur collective ut cum
dicitur: 'omnis homo trahit navem', positione facta quod omnes homines trahunt
navem et nullus per se, et similiter cum dicitur: omnes apostoli sunt
duodecim.
"Similarly, confused but immobile supposition is when the
(quantifying) sign is applied to a common term in the singular or plural and is
understood collectively, as in 'every man drags the boat', assuming that all the
men are dragging the boat and none of them by himself. And similarly in 'all the
apostles are twelve'. (Logica, "De suppositionibus et de significationibus")

Later, William of Ockham discussed the 'Apostles' example in his Summa Logicae II. 4.

Sunday, July 18, 2010

How to tell a big lie

"What is most deceptive is not when you lie outright, but when you tell the truth, but not the whole truth. The most important question you can ask of any presentation, not necessarily 'oh is this person lying to me', but 'what isn't this person telling me'. Because if the person lies to you, you can check up on that. But if a person leaves something out, you have no way of checking - you don't know - what has been omitted".

Howard Zinn, interviewed by Truth in Numbers.

Francesco Patrizi

The case of Francesco Patrizi, the Venetian philosopher, is a fine illustration of the nationalistic warfare that infests Wikipedia, and the inaccuracy and distortion and bias that follows as a result*.

The problem is that Francesco has at least three different identities. His place of birth was on the island of Cherso in 1529, which lies off the coast of Dalmatia, now in modern Croatia. The state of his birth was the Venetian Republic, which dominated the area until the early 1800s. His intellectual heritage is the Western philosophical tradition: he spent seven years at Padua studying Aristotelian philosophy, in Latin. He studied Plato, and seems to have had a knowledge of the original Greek texts, even owning a collection of Greek manuscripts [1].

Consequently, there is a war on Wikipedia about Francesco's identity. Is it defined by the Western intellectual heritage with its Latin and Greek origin? All that Francesco learned in his years at Padua and afterwards are derived from it. Is it the Republic of Venice, now in modern Italy? Francesco could not have been taught without the universities in Italy whose staff and administration were not paid for by Croatians. Or is it the modern state of Croatia, where his birthplace now lies?

Wikipedia, or at least its current version, defers to the first, calling him by the Latinised name of Franciscus Patricius. Adopting the second, the Stanford Encyclopedia of Philosophy article [2] uses his Italian name. But in Croatia, of course, he is known as Frane Petrić, although (I am told) they made a mistake here: they used the Croatian diacritic sign on the consonant, invented only in the middle of the 1800s.

The whole issue presents problems when you are doing real history. How do we separate the contribution of an individual from the contribution of the tradition they are writing in? No intellectual contribution made by an individual, even by an original thinker such as Aristotle, is uniquely due to their special talent, although ability is a necessary condition. An equally important influence is individual teachers, who in Francesco's case would have been Italian. So in what sense does a country - a modern country like Croatia, say - own a persons's work?

The problem becomes particularly acute in a place like Wikipedia, where the only intellectual interest - that is to say, no intellectual interest at all - lies simply in a nationalistic dispute, in this case between Italians and Croatians. The talk page shows it beautifully. "He was Italian for culture and birth" says Giovanni Giove (presumably an Italian). "It is sad to see that User:Factanista has started an further edit war to imposte [sic] his nationalistic POV" says Giovanni. "It is you who is edit-warring " replies Factanista (presumably a Croatian). "I'm going to report the behaviour of User:Factanista to an administrator" says Giovanni. "Please do, you will be reported shortly yourself" objects Factanista.

The history of the article itself [3] is instructive. Hundreds of reverts and unreverts, as the intellectual battle unfolds. An Italian editor removes the word 'Croatian' [4] with the comment "oh please, lets avoid stupid national bickering". It is immediately reverted back [5] with the reply "then stop to deny his roots and ethnicity". Which of course makes it all the more absurd. What does ethnicity have to do with the work of a man who wrote, in Latin, about the Greek philosophers Aristotle and Plato? It is different in the case of someone like Chopin who, though he studied in France and was influenced by the French tradition, wrote specifically Polish nationalistic pieces. This case, by contrast, is more like the case of Joseph Conrad, who was Polish but whose works are all in English. Or Wittgenstein, who was Austrian and who actually wrote in German, but who was taught by an Englishman** (Russell), and whose work was most influential in England where he lived and worked for most of this life. Ethnicity has very little to do with it.

What is really sad is how the article suffers as a result. In its current version (permanent link above) all but the first three paragraphs are copied verbatim from the 1911 Encyclopedia Britannica article [6]. The introduction is mostly taken from the SEP article. I have commented on this irony before - that "an ultra modern high-tech medium like the Internet, when it reflects anything of importance, reflects the obselete views of 19th century scholars". None of the people who have edit-warred for at least four years over the ethnicity of this man, seem to have any interest in his thinking.

At least I found something valuable in my research today: When ethnicity did not matter in the Balkans by John Fine. About "The back-projection of twentieth-century forms of identity into the pre-modern past by patriotic and nationalist historians". Possible holiday reading.

*Many thanks to Peter Zuvela, who brought this to my attention.
** Wikipedia has Russell down as a Welsh philosopher, of course.

Saturday, July 17, 2010

Ochlocracy

Ochlocracy. It means 'mob rule' or perhaps 'crowdsourcing'. I'm going to write more about this later, having recently revisited Gustave le Bon's wonderful The Crowd.

Meanwhile here is what one philosopher said about the wisdom of the crowd.

"Philosophy: I'm a philosopher; why don't I edit the article on my subject? Because it's hopeless. I've tried at various times, and each time have given up in depressed disgust. Philosophy seems to attract aggressive zealots who know a little (often a very little), who lack understanding of key concepts, terms, etc., and who attempt to take over the article (and its Talk page) with rambling, ground-shifting, often barely comprehensible rants against those who disagree with them. Life's too short. I just tell my students and anyone else I know not to read the Wikipedia article except for a laugh. It's one of those areas where the ochlocratic nature of Wikipedia really comes a cropper".

By Wikipedia editor Mel Etitis, who is a well-known philosopher in real life. He left Wikipedia shortly after this comment in 2007.

Friday, July 16, 2010

Truth in numbers?

In my previous posts on truth [1], [2], I explored the idea that truth has no special benefit over falsity. "It is a piece of idle sentimentality that truth, merely as truth, has any inherent power denied to error". Here, I shall suggest that it is worse than that. Truth has powerful enemies, and there are forces systematically favouring error. As befits a scientific investigation, I shall present the theory, followed by evidence (using the example of Wikipedia once more) supporting it.

First, the theory: Those who support the truth are in a large majority. Of 100,000 people, probably all but ten would like to see the truth. But their interest is only feeble. The remaining ten support false beliefs of various kinds. They probably think these beliefs are true, but they are false for all that. And their false belief is passionate and determined. It follows that, if anyone is allowed to publicise their belief, and if there is a moderate cost to publicising it, such as arguing about it or being involved in an 'edit war'*, the proponents of error will always be victorious. For they are passionate in their error, whereas the others are only feeble in their truth.

Now, the evidence. No sane or normal or reasonable person contributes to Wikipedia, and so all contributors fall into the following broad classes: deviant, aficionado, quack, activist, cultist, crank. The reasons for their persistent interest, together with Wikipedia article examples, are as follows. I list them in approximate order of the power that belongs to them.

1. First are the deviants, usually the sexual deviants. The pedophile lobby has long been an active force on Wikipedia. They are powerful because, like everyone else, they are passionate about their sexuality, seeing it as in some sense normal, and also because no reasonable person is likely to edit articles about pedophilia, 'pederasty' and so on. Hence monstrosities like this, which is indistinguishable from monstrosities like (WARNING: pedophile website - not safe for work) this. See also the article about the PPA 'Wikipedia Campaign' at Wikisposure.

2. Aficionados ('fans') are a large and diverse group on Wikipedia. They are mostly harmless devotees of obscure subjects like Japanese comic books or American science fiction TV. The error lies not so much in the uncritical approach to the subject as the undue weight given to a subject which is essentially ephemeral and unimportant and unencyclopedic. This is probably harmless, although not in the case of Ayn Rand, who I have discussed before. The coverage of her ideas is extensive in Wikipedia, and out of all proportion to her real importance in philosophy, which is practically nil, particularly outside the United States. Aficionados are powerful in Wikipedia because they are mostly viewed as harmless, and because there is no 'weighting' policy in the encyclopedia. Quite reverse: they even have a policy: Wikipedia is not paper, and so there is no practical limit to the number of topics it can cover, or the total amount of content. The practice of a normal reference work, which is to assign pages in rough proportion to the received importance of a subject, does not apply here. Thus the academically marginal Ayn Rand receives more coverage than Aristotle, the father of Western philosophy and easily the most important figure in the Western intellectual tradition. The article on his Sophistical Refutations, for example, is no more than a list of contents. Compare this in size and scope with any article on the nonsensical and philosophically illiterate work of Rand, e.g. this.

3. Quacks are peddlers of fake cures, bogus medicine and psychological theory. There is plenty of this on Wikipedia. Their interest is commercial rather than idealistic. The Wikipedia administration does attempt to weed out blatant commercial advertising, but it is also corrupt. The subject that touches me the most is the rubbishy and fraudulent Neurolinguistic Programming. These articles had support at high levels of the Wikipedia administration, and so quite a few more neutrally-minded editors (including myself) were banned for attempting to trim them. See also EMDR, or Ken Wilber, or this lot, gulp.

4. Cults are groups with strange beliefs who have an interest in publicising their existence, recruiting new members, and usually suppressing the more unpalatable facts about their financial statements and other irregularities. Too many of these to mention, but some of the more amusing include the Brahma Kumaris, who seem to enjoy some support among Wikipedians, at least judging from the way that those who opposed them are so regularly blocked, simply for saying stuff like this ("You and your other adherents have wasted too much time of too many people's lives ... never mind mentioning the broken families and suicides that litter your religion's history"). The Scientologists did not fare so well, as is well known (their IP is currently blocked), but that is only because a group of prominent Wikipedians dislike scientology - it has nothing to do with any self-governing mechanism that prevents cults from promoting their views on the global electronic reference work. See for example Prem Rawat, defended for years by a prominent Wikipedia administrator, although he eventually came to grief.

5. Activists are the supporters of extreme political movements. These are also numerous, but the force can be less strong with them because there are often equally extreme political movements who are bitterly opposed, and so they 'edit war' the talk pages of the articles concerned. See the archives of the articles on The Troubles, the thirty-year conflict in Northern Ireland, which was intellectually as violent as the actual troubles were physically bloody. Or Islamic Republic is the same as Arab Republic. However, the articles on Yugoslav communism, e.g. on Tito seem to have got by with their strange point of view unscathed. They do not reflect the views of more recent historians who view Tito as essentially a Stalinist, but rather adhere to and reflect the propaganda of the former Communist Party of Yugoslavia. The article about Tito is written in a child-like manner, reminiscent of Yugoslav primary school textbooks from the 1970s. This may be the result of the 'partisan' group of editors who control those articles, and the relative lack of interest from anyone else.

6. Cranks. These are individuals who have a passionate belief in some idea, theory or system that they developed on their own, and which has been rejected by the academic establishment. Naturally they turn to the encyclopedia which anyone can edit. Cranks have little power on Wikipedia, because unlike the rest they are not part of a larger group and can easily be picked off, and also there is no fraudulent secondary literature they can tap. Nonetheless they flourish in dark corners - I am glad to see that Boolean logic, which has nothing to do with George Boole, still survives, and that the Ancient Egyptian Race theory seems to be thriving, at least between periods of illness or coma. And how about this whole category of stuff which is completely deranged (" the Mental Plane is located between, and hence is intermediate between, the astral plane below and the higher spiritual realms of existence above").

Add to this the eccentric and perverse features of Wikipedia governance and software. First, the ability of any IP addresses to edit, which results in a tidal wave of crude vandalism every day. This in itself is not a problem, since there are about 500 administrators who instantly clear it up. The problem is rather the administrators, many of whom are retired ex-military types or police, and who are not interested in encyclopedias at all. They just enjoy whacking vandals. The problem is that they don't really distinguish between vandals and ordinary editors, and don't really understand the disputes at all. They have no theoretical interest in a dispute between a scientist and a homeopathist or chiropractor. But if they see either of them getting out of line, i.e. infringing Wikipedia's strict 'civility' rules, they whack them anyway.

Second is the fact that accounts are anonymous, and there is not even elementary identity checking. Thus a problem editor who has been 'blocked' can instantly register again. Furthermore, they can create multiple accounts ('sockpuppets' or 'socks') to give the illusion of strong support for error. This frustrates the supporters of truth, many of whom stoop to the same tactics. Worse, Wikipedia has evolved an elaborate secret police ('checkusers') whose job is to spy on accounts for 'sock'-like behaviour and block them if necessary. This deflects the administration from the real job of building a comprehensive and accurate reference work, moreover it encourages types who are intellectually unsuited for such work.

Which is, of course, why Wikipedia is nothing like a comprehensive and accurate reference work. How would we change this? Well, the theory (that small numbers of passionate devotees of error will always defeat an army of those with a feeble and weak interest in the truth) suggests abandoning the idea that 'anyone can edit'. It's fine if only 'disinterested' persons can edit, but as the different but connected meanings of 'interest' suggest, it is difficult to get disinterested people to do this. They simply aren't, er, interested. Another idea would be this. Levy a small 'falsity tax' on everyone. A majority of people would vote for this, if the tax were proportionate to their feeble dedication to the truth. The money from the tax would then pay for experts with a proven neutrality and lack of 'interest' to write articles. This is essentially the economic model of a university, a system invented in the Middle Ages, and which therefore existed long before Web 2.0.

*Edit war: long protracted dispute on Wikipedia. Sometimes between the forces of truth and error, more often between propopents of different kinds of error (such as environmentalists and oil company employees).

Thursday, July 15, 2010

The Cult of the Amateur: Lessig vs Keen

Someone here told me to look at here, where Lawrence Lessig replies to some of the criticisms of Internet 'culture' made by Andrew Keen in his book The Cult of the Amateur ("How Today's Internet Is Killing Our Culture and Assaulting Our Economy ").

Many of Lessig's points are spot on, for example his criticism of Keen's sloppy grammar - "what is puzzling about this book is that it purports to be a book attacking the sloppiness, error and ignorance of the Internet, yet it itself is shot through with sloppiness, error and ignorance". He is probably right to say that policy makers should not be concerned (as Keen apparently is, p.115) when the costs of an activity drop because society has found a way to do the same activity more efficiently. There is no point in preserving the living of those who farm trees to make printed books if (I say if) exactly the same social benefit can accrue from disseminating that information through the internet.

However, I take issue with Lessig about the ghastly Wikipedia. Keen claims

Since Wikipedia's birth, more than fifteen thousand contributors have created nearly three million entries in over a hundred different languages—none of them edited or vetted for accuracy (p4).
Spot on. Perhaps an exaggeration: no doubt some of Wikipedia's entries have been vetted for accuracy, and so (as we logicians know) it is literally false to claim that none have been vetted. But most of them haven't, and that will do. Keen disagrees:

this is absurdly false. Wikipedia is constantly edited, and attributions
constantly vetted for accuracy. Indeed, for many of the articles, the level of
editing and vetting is vastly greater than any article published in any
encyclopedia ever.
Not at all. There are superficial attempts at vetting for accuracy, and the level of attempted vetting for accuracy is certainly vastly greater than ever before tried by humans. But trying is not the same as achieving. There have been 911 edits to the Wikipedia article Existence since the first version begun on 1 June 2001, 9 years ago. It is still much worse than the original version (which was not particularly good either). Regular readers of this blog will appreciate this is a subject I have some acquaintance with. The kind of 'vetting' that goes on in Wikipedia is merely superficial. And so it is (in my view) for pretty much all the articles on philosophy, and most of those in the humanities generally.

Lessig suggests that Keen means Wikipedia is not "vetted" by experts.

Or exclusively by experts (for again, experts certainly participate in
Wikipedia). This is related to Keen's obsession (indeed, I'm sure if he has one,
his shrink must have a field day with this obsession) with "experts" and makers
of "taste." So central is this to Keen's argument, it deserves its own
heading
.
But having read on, I didn't find any argument against this central claim - namely the claim that Wikipedia suffers from not being vetted by experts. In any case, I think the articles speak for themselves. Some parts of Wikipedia are good, much of it is written by amateurs (meaning: non-experts). But the parts that are written by amateurs are generally not good, and the parts that are good are mostly not written by amateurs.

Monday, July 12, 2010

Monday, July 05, 2010

Truth in ten minutes

Well, Wikipedia now has a child protection policy, largely cobbled together as a reaction to the Fox News article about pedophile activism. The best part is the policy talk page where Wikipedians have been ardently discussing the rights of minorities to edit Wikipedia. “We should consider that such persons openly avowing pedophilia provide the world with a very rare perspective and a potentially useful resource.”

The idea that the truth will emerge if you open the means of promoting it to absolutely everyone is fundamental to how Wikipedia works, and (as the remark above proves) is an article of faith to most ‘Wikipedians’. But as I suggested here, truth is more fragile. There is nothing magical or complicated or difficult about recognising it yourself. The problem is to get it others to recognise it. Although it takes in general about 10 minutes to discover the truth about most matters, only one person in every thousand people is likely to spend that time. Of the rest, 9 people will find it so painful that they can get only as far as 3 minutes, and will spend the rest of the time and much more in opposing every attempt at publicising it. The remaining 990 people have only a minute to read about it, and thus are forced rely on the other 10. This is the best model for understanding Wikipedia.