Tuesday, May 19, 2015

Wikipedia Absurdities

NOTE: I reproduced this post, as I realized I had never copied over any of the posts I wrote about the absurd claims made in Wikipedia, and elsewhere, that Rush Limbaugh had said some amazingly loopy things. Since this is good evidence of the weaknesses in Wikipedia's reliance on being in print as proof of validity, I needed to repost it.

I was originally going to write about an absurd Wikipedia problem I discovered while reading the discussion page for the article on MacGuffins, but I then found an even more perfect (and more political) example on the wikiquote page about Rush Limbaugh, and in the related talk pages.

However, since I mentioned the MacGuffin page, I will use it to point out the problem with Wikipedia's approach.

The problem simply put is that Wikipedia rules result in utterly absurd outcomes. Specifically, the absolutely slavish reliance on citations of external sources, coupled with the prohibition against "original research", as well as how the writers understand the concept of "original research", results in bizarre situations.

The MacGuffin page, as I said, provides a perfect example, being the page which called this to my attention. On the page in question, there is a quote form George Lucas, which pretty clearly misuses the word. Several writers and readers pointed this out and asked that the quote be annotated to mention that the word was being misused. This was opposed by a large group which argued that while there was a citation for Lucas' quote, there was no citation saying Lucas was wrong, nor that the definition traditionally sued had not changed by the time Lucas sued the word. Those asking for the comment countered with the obvious argument that a word means what it means, and using it contrary to the definition should be mentioned in an encyclopedia. To which those opposing any mention of Lucas' mistake countered by arguing that using a definition without a citation would be "original research" and thus impermissible.

Admittedly, my description is a bit convoluted, but so is the argument. So allow me to provide a better illustration using Wiki-Logic.

In an article on gravity, someone finds a published book which says "objects fall up at 9.8 meters per second squared", a fact which he places in the article, complete with citation.

Later a sensible fellow comes along, and removes that quote as it is absolute rubbish. However, the original poster returns, puts up the quote, and asks an editor to protect his quote, as it is cited and thus permissible, while the man "vandalizing" the page by removing it has no citations to back him up. And, the editor does, as that is the rule.

So, our sensible man comes back and finds he can't edit the offending line. So instead he writes in the talk section, arguing that anyone can plainly see objects falls down, not up. However, the other writers rebuke him for using "original research", and tell him to bring a citation to a valid source which says objects fall down and he can post it.

And so he does, bringing ten citations, ranging from physics texts to Isaac Newton to Aristotle. He thinks that eh has won, he has clearly shown that objects fall down, and that the original post is absurd.

But not so fast! He has citations, true, but so did the original poster.

And thus, the final page reads "There is some disagreement among experts, some arguing that gravity causes objects on Earth to fall down, others arguing that they fall up."

And that is, quite simply, the logical outcome of the methodology employed by Wikipedia. It does not matter how contradictory or nonsensical what you say might be, it matters not how much ti might offend against common sense, if you can cite it, it is true. And if there is a conflict between cited sources, provided they meet the rather scant Wikipedia guidelines, then both positions get in, no matter how absurd.

Oh, there is one remedy. You could always call for a vote. Which means, rather than common sense, or even the neo-compurgation method of dueling citations, truth has now been reduced to majority rule. In other words, the choices are: accept that any citation, no matter how absurd, carries equal weight, or else surrender the definition of what is true tot he random chance of which internet rovers happen by while voting is taking place. THAT is the way Wikipedia defines truth.

Which brings me to my example form yesterday' post about Rush Limbaugh.

On the Wikiquote page about Limbaugh there is an extremely inflammatory quote:
You know who deserves a posthumous Medal of Honor? James Earl Ray [the confessed assassin of Martin Luther King]. We miss you, James. Godspeed.
Please note that second bullet point, it points out the entire problem.

On the discussion page you can see a perfect example of the problem with Wikipedia. An individual, confronted with this quote, argues quite reasonably that, had Limbaugh said anything as shocking as is here attributed to him,it would have been all over the media. The relatively inoffensive quote about McNabb was the talk of the airwaves for days, so wouldn't a quote praising James Earl Ray make it into the press? Yet the only citation is a single book which appears to be rabidly opposed to Rush Limbaugh. (I am surprised the writers even allowed those who objected to post that second bullet point. Even that small concession to common sense seems extremely moderate for Wikipedia.)

Which, as you can guess by now,  was dismissed as "original research". Since there was a citation in favor fo the quote, and no citation proving it wrong, it gets to remain, no matter how unlikely that it is true. Common sense is irrelevant, as "original research" it is outweighed by scurrilous yellow journalism.

Let us think about this for a moment. The logic is disturbing. By the rules of Wikipedia,. if someone publishes an outright lie, it can be cited as often as one desires, and the only way to have it removed is to find an explicit refutation in print. And even then, it is not removed, the refutation is simply printed along side it.

This is absurd. There is little cost to print a lie, even the worst slanders usually go unchallenged, provided the audience is small enough. However, as they go unchallenged, they often go unrefuted as well. For instance, if someone set out to "prove" a celebrity were gay, he could print tons of small web pages or self published books arguing his case, and as long as only a handful of people saw them, it is unlikely the celebrity would ever hear of them, much less sue for slander or libel, or print a formal refutation. However, by the logic of Wikipedia, that means those slanderous charges can be cited in Wikipedia, while the lack of refutations means that no counter arguments can be mentioned.

And that is where, once again, Wikipedia is far inferior to a traditional encyclopedia. A real encyclopedia would not print outright lies or patent absurdities, they would not accept silly slanders simply because there was no explicit refutation to which they could point. But that is almost guaranteed by Wikipedia's procedures and rules.

Which means that, far from being an alternative to an encyclopedia, Wikipedia is in danger of becoming something akin to a gossip rag, except less reliable with fewer quality controls.


My earlier criticisms of Wikipedia's theories or practices can be found in the following articles:
Stop Confusing Me With The Facts!
Mystery Quotes
The Failure of Wikipedia
Opinion Masquerading as Fact
Funny Numbers
What is Wrong with Wikipedia?
Endangered Species
Sterility of Formal Economics
Deceiving Themselves?
A Question About Language
Why People Don't Take Academics Seriously
There may be one or two comments about Wikipedia which were omitted, but I believe this list is as comprehensive as possible.


I will grant that sometimes an editor will allow the patent absurdity of an assertion guide him and remove bizarre statements despite supporting citations, but that is both unusual and a violation of Wikipedia rules. Technically, Wikipedia is bound to the two methods of compurgation by citation and truth by vote, and common sense has no place, as being an example of the dreaded "original research". So, though sometimes an editor breaks the rules and shows some initiative, that does not excuse Wikipedia, especially as such displays of sense are technically violations fo explicit Wikipedia rules.

NOTE:  For those who aren't obsessed with medieval legal procedure in England (and other Germanic and Nordic nations), compurgation was a legal procedure whereby a man established innocence by gathering the vows of a set number of individuals. If he could get enough individuals of the right social ranks to swear a solemn oath that he was innocent, then he was judged to be free of guilt, or liability in a civil case. I use it to describe Wikipedia, as often the senseless gathering of citations, with no thought as to their relative worth, or even validity, strikes me as being as unrelated to the quest for truth as compurgation was. 

Originally posted in Random Notes on 2008/10/22.

No comments:

Post a Comment