Lately I’ve been writing essays for a course I’m taking at Uni called “Source Reliability”. Readers of this blog will know that I’m rather keen on this subject. We get our essays accepted or not accepted – they aren’t graded. But the professor comments on them, and he liked my latest essay. It’s about Wikipedia and has a debacle between the science journal Nature and Encyclopaedia Britannica as its starting point. If you haven’t heard about the debacle, here’s what it says in Wikipedia (and it’s in fact quite a correct description):
On 14 December 2005, the scientific journal Nature reported that, within 42 randomly selected general science articles, there were 162 mistakes in Wikipedia versus 123 in Britannica. In its detailed 20-page rebuttal, Encyclopædia Britannica, Inc. characterized Nature’s study as flawed and misleading and called for a “prompt” retraction. It noted that two of the articles in the study were taken from a Britannica year book, and not the encyclopedia; another two were from Compton’s Encyclopedia (called the Britannica Student Encyclopedia on the company’s web site). The rebuttal went on to mention that some of the articles presented to reviewers were combinations of several articles, and that other articles were merely excerpts but were penalized for factual omissions. The company also noted that several facts classified as errors by Nature were minor spelling variations, and that several of its alleged errors were matters of interpretation. Nature defended its story and declined to retract, stating that, as it was comparing Wikipedia with the web version of Britannica, it used whatever relevant material was available on Britannica’s website.
Below find my essay – only edited slightly for use here (no footnotes etc.). If you can’t be bothered to read the whole thing – about a 1000 words – then scroll down to the bottom. There’s my tips for what to think about before you delve into a Wikipedia article.
The battle between Encyclopedia Britannica (hereafter EB) and Nature was intriguing – not least because it, in my view, is somewhat beside the point. Nature’s intentions were honourable, I believe, in letting their very informed readers know if it can be considered worthwhile – not safe – to use Wikipedia for anything. And they seemed to be rather baffled themselves at the result, that yes, it is worthwhile, also for the informed user, to consult Wikipedia. In my view the article did not try to put EB down.
One of the more interesting facts the investigation revealed was that the learned test persons were more sceptical towards the random articles than towards the articles within their fields of expertise. For reasons that I can’t quite understand, many teachers at all levels of the schooling system tell their pupils to NEVER use Wikipedia. Many times I’ve heard well educated and academically trained people say that they never use Wikipedia, because it’s completely untrustworthy. But upon inspection, they have never used it, so how is it that they know? Probably this is why the test persons were so sceptical towards the articles about subjects outside their intellectual comfort zone.
It is also interesting to notice the aggression and fervour with which EB responded to the article. A lot of their response may be correct in a narrow sense, but entirely beside the point, because the Wikipedia articles had had the exact same treatment. And the Nature article is actually quite critical about some things in Wikipedia – like the occasional rather poorly constructed articles and poor readability. This fervour may be related to the sad fact that academia frowns upon academics who choose to put their skills to use for the general public. Nature surveyed 1000 scientists, of which only 10% had ever helped updating Wikipedia. It probably doesn’t improve your academic career to invest time enlightening the public on your speciality.
And then there are all the things you can get from Wikipedia, which EB doesn’t give you. There are articles about every little town or village in the Western World, every politician, every pop group, every artist, every historical person, every technical term or gadget known to man – almost. And then there’s the freshness – the articles updated at the speed of light when events develop. Apart from the way they are created, these two factors are what really separates Wikipedia from EB. And why to some extent comparing them is a bit like comparing apples and pears. And access to EB is on subscription basis. In Denmark and here in the UK you can gain free access to EB via your local library. But unfortunately, most people don’t know this – or just can’t be bothered. In EB you cannot see when an article has been created or updated – or at least I can’t find it. And there are very few outside links and no references.
When I was a child we had two encyclopedias in the house: Lademanns and Gyldendals. I quickly discovered that Lademanns was best for looking up things to do with nature, science and geography because of the many, good colour photographs and illustrations. Whereas Gyldendal was best on history and literature, because the entries were better and longer. But, and this is the point, it never occurred to me to doubt the authenticity of any of the articles. And I wasn’t taught that at school either. I didn’t hear about source criticism (kildekritik) before high school (gymnasiet), where I had a history teacher (an elderly gentleman) who made it an issue. It was the first time I had ever heard of anyone questioning a source. Every time he gave us something to read, he asked us to consider who had written it, why he had written it and who we thought were the intended audience. This simple wisdom has stayed with me always and I try to remember to apply it to all things I read or hear.
The thing about Wikipedia, which could maybe teach many more Internet users source criticism, is exactly the knowledge of how it is written and (not) edited. One must always consider the fact that the article one’s looking at might just have been tampered with by some idiot or a person with malicious intent. Or that it’s written by somebody who has an overblown perception of her own knowledge. This is not a thought that automatically comes to mind when looking up something in EB or another “trusted source”. So I believe that the way Wikipedia is constructed actually encourages its users to be source critical. And that scepticism could even follow the user when she ventures outside Wikipedia and looks at other sources.
Quite often Wikipedia is an excellent starting point for research on a subject. Usually it becomes clear very quickly what kind of person or persons are responsible for a Wikipedia article. Some of them are clearly written by scholars or by extremely knowledgeable amateurs and their sources are often gold, when the goal is to move on to primary sources. Other articles are not so well written or edited and one instantly gets wary. That very often reflects on the sources, which will be few and erratic. And I believe this wariness and alertness to be very healthy for the users.
Setting aside the times I use Wikipedia to look up the full name of a pop star or the use of a technical gadget, I try to ask myself these questions while reading a Wikipedia article:
What kind of person wrote this? Syntax, writing style, approach to subject. Is the faulty English because the writer doesn’t have English as her mother tongue or is it a warning sign?
Why did the person write this? Out of pride, to boast, for political/religious reasons or because the person honestly feels it is her duty to share her knowledge?
Does the article have the feel of having been worked over many times? If so, I check the history and debate pages.
What are the sources like? Are there many? Are they online, off line or a mix? How many of them are readily accessible (not necessarily online, but from a library)?
How sensitive is the subject? Can I maybe believe some parts of the article, but not other parts? This may be the case for quite a few historical articles, where basic facts are agreed on by everybody, but where historians disagree on the interpretation of certain incidents or documents. This is also the case for articles on pharmaceutical compounds.
Am I looking at a subject where recent events have led the article to be expanded or changed? The article about Sarah Palin is an obvious example. One can go back to the version of the article a couple of weeks before she was chosen as running mate for McCain and get an impression from that.
The above rules of thumb could very well be applied to most other sources as well. But with most other sources you can’t check the previous versions…