This post first appeared on Michael Martinez’s blog, SEO Theory, on June 22, 2014.

1

If you search Twitter for word combinations like “Wikipedia credible” you may find people arguing back and forth about how credible the site is as a source of information. I even found a Tweet where someone wrote: “I hate when people say Wikipedia isn’t a credible source. You click on the links on the page which lead to credible sources.” That’s very true. Many (though not all) Wikipedia articles link out to “credible sources” but there are several reasons why providing citations doesn’t make you instantly credible. This is true for everyone; it’s not something that is peculiar to Wikipedia.

2

And then there are reasons why Wikipedia is not and never can be a credible source of information. So let me deal with these two very different explanations separately. But first, let’s talk about credibility.

Credible Information Is Not Always Reliable

3

History is filled with grave mistakes that were committed on the basis of credible information. The interpretations of the credible information may not always be universally supported; and not everyone will agree that information is credible. Nonetheless, information may be presented as credible by a credible (or authoritative) person.

4

For example, during World War II both the Axis and the Allies used extensive networks of spies supported by intelligence agencies with all the latest technology and encryption/decryption skills to study each other’s forces and strategies, report back, and analyze enemy intentions. Knowing that both sides were doing this, both sides in the war resorted to disseminating false information.

5

As it turns out the Allies proved to be more successful during critical operations. The United States’ use of Navajo “Wind-talkers” confounded Japanese intelligence. While preparing for the D-Day invasion of northern France General Eisenhower deployed a false army under the command of General Patton to mislead the Germans about Allied intentions. In at least one airdrop operation dummies on parachutes were dropped out of planes to fool German gunners and inflate the numbers of attacking soldiers being reported to headquarters.

6

The Axis soldiers and spies charged with collecting and reporting this information were generally credible sources of information. They just didn’t pass on reliable information. Credibility doesn’t make you accurate or correct.

7

To be credible, a source is convincing. It may be convincing on the basis of past interaction (such as an undercover police officer who has conducted numerous successful investigations). It may be convincing on the basis of logic (organizing facts in as complete and supportable manner as possible). It may also be convincing due to a lack of alternative or contradictory information.

437

8

In many UFO investigations police officers are often deemed to be highly credible sources of information. However, police witnesses do not always recognize what they are describing. In fact, we don’t investigate Unidentified Flying Objects unless we have reason to believe they are hostile. So just because a police officer credibly reports a UFO doesn’t mean that any claims of extraterrestrial visitation on the basis of that report are valid.

9

Credibility in reporting source does not necessarily confer validity upon any conclusions drawn on the basis of the reported information. Hence, credibility of source in itself is not a persuasive argument for believing something. And that is especially true when you need to establish provable facts. So being credible is no guarantee of being a reliable source of good or useful information.

Why Providing Citations Is Not Enough to Make You Reliable

10

Whether you write a huge long article or a very short quote, if you are relaying information you found elsewhere on the Web it is courteous to provide a citation of your source. Pointing people to your source discredits any allegation that you made up what you are sharing.

11

Nonetheless, even if you publish 1,000,000 articles that all include citations for sources, your credibility depends on more than the credibility of your sources. If your credible sources are wrong, for example, then your use of those sources undermines your credibility. And credible sources can be very, very wrong.

12

Here is an article from 2005 that was published in the Harvard [Men’s Health Watch]: “Help for Your Cholesterol When the Statins Won’t Do.” The article includes the following information:

… 3 percent–4 percent of people […] don’t do well with a statin drug. In a few cases, the drugs simply don’t work, but more often the reason is a side effect. The most common statin toxicity is liver inflammation. Most patients with the problem don’t even know they have it, but some develop abdominal distress, loss of appetite, or other symptoms. Even without these complaints, liver enzyme abnormalities, such as high aminotransferase levels, show up in the blood tests of 1 percent–2 percent of people taking a statin drug. The other major side effect is muscle inflammation, which can be silent or cause cramps, fatigue, or heavy, aching muscles.

At the time this article was published the information was deemed highly credible, even correct, based on the science available at the time. Unfortunately for millions of statin patients worldwide, the research was highly flawed. We now know that 37.5 percent of clinical trial patients could not tolerate statins.

13

Also, based on patient feedback over the past 9 years, most medical resources now show that muscle problems are the most common side effect. When the Harvard [Men’s Health Watch] publishes horribly wrong information about a widely used class of medicines, people should sit up and take notice. To their credit, doctors at Harvard University have come out this year in strong opposition to the use of statins for patients who have not had a cardiovascular event. Recent research shows very convincingly that there is zero benefit for 80 percent of people with high cholesterol from taking statins. Only people who have already suffered a heart attack or other CVE may benefit from use of statins.

438

14

Credible information can therefore be wrong enough that even citing it is not useful. Basing any article on a citation of a credible source that has been discredited in topical context means you are reporting wrong information, and therefore your article is not credible. So what if the 2005 article was once deemed correct? We now know it to be wrong; therefore using it as a source of information for any current survey of medical practice is inappropriate.

15

You can provide all the citations to credible sources you wish, but if they are wrong then your own credibility may not suffer because you use credible sources but your information is unreliable.

16

The key takeaway here is that just because information is credible does not mean it is accurate, correct, or reliable.

Why Accurate Citations Do Not Make You Reliable

17

So let’s say you go the extra mile and do your research so well that you not only compile credible sources of information you also find independent confirmation of the claims those sources make. Your sources’ credibility is impeccable and the information they present does not appear to be contradicted by any other credible points of view. There are still potential pitfalls you have to overcome.

18

Less Credible Sources May Be More Correct Sometimes people who are hard to believe really do get the facts straight. Maybe they are being intuitive. Maybe they lack any scientific proof. Maybe there is no way to confirm what they are saying. These deficiencies in support context do not make a viewpoint wrong any more than hundreds of supporting contexts make a viewpoint right.

19

The Correct Information May Not Yet Be Known Sometimes everyone latches on to the same plausible explanation for lack of anything better. This even happens in science, especially in theoretical science where independent observations and experiments have not yet shown a theory to be true (or as true as we can confirm it to be given our current state of knowledge). Scientists will tell you that Einstein’s Theory of General Relativity is true because it has been proven so by many experiments. They’ll also point out that we have spent billions of dollars on experiments that attempt to prove the theory false. Although no one expects that to happen we would be practicing poor science by neglecting to attack and challenge the theory from every conceivable angle.

20

We may never fully know if Einstein was wrong.

21

Truth is not democratically determined. It doesn’t matter how many sources report the same information, even if they all arrived at their conclusions independently of each other. They can still be wrong. They may be wrong. In any given topic you have to allow for the possibility that some new verifiable evidence will eventually turn the entire world upside down.

439

22

Credible Sources May Be Corrupt Whether you are dealing with a police officer who has turned to crime, a scientist who is falsifying data, or a reporter who just makes up a false story sometimes your credible source is actively trying to deceive you. You can’t trust anyone. Literally.

23

Credible Sources May Have Been Deceived As noted above, credible sources of information were misled by counter-intelligence operations during World War II. More recently, former U.S. Secretary of State Colin Powell presented the U.S.’s case for invading Iraq to the United Nations. His argument was based on two points: Saddam Hussein refused to allow UN Weapons Inspectors back into Iraq to verify that all his weapons of mass destruction (and the capability to make more) had been destroyed; and the U.S. had intelligence that led government analysts to conclude there was a high probability that Saddam Hussein was hiding something (probably Weapons of Mass Destruction).

24

Secretary Powell was viewed by everyone as a very credible, sincere man who would not intentionally mislead anyone. He was a very credible source. He just didn’t have very good information. After we invaded Iraq our forces did find several hundred chemical warheads that had been falsely reported as destroyed years before. WMD experts dismissed these warheads as being so old their chemicals would be inert. The military destroyed the warheads with all safety protocols just in case.

25

So as it turned out Saddam Hussein did not have a functioning WMD program or even a viable cache of weapons. But no one knew that at the time because some of the intelligence given to Secretary Powell was false. Someone lied and the lies were passed on.

26

The key takeaway here is that lies often get fed into the information chain and they are passed around. Viral propaganda theory tells us that it doesn’t matter if information is true or false: people will accept it if the source is credible, and once they believe a certain point of view it becomes next to impossible to change their minds.

Why Your Beliefs May Make You Unreliable

27

You may have the most credible sources of information to hand. You may have eliminated all relevant doubt. You may have clear-cut scientific proof that shows no one has lied in the chain of information leading up to your own presentation. And yet your own article may be unreliable. Why? Because you believe.

28

As I pointed out above, viral propaganda theory shows us that people are reluctant to change their minds. Just because I come along and point out all the fallacies of the viewpoint that you believe to be true doesn’t mean you’re going to question what you have believed again. You have already questioned that viewpoint and challenged it according to your prior knowledge and beliefs and it has passed all your tests.

440

29

To change your mind now would be to admit you were wrong and most people hate to admit they are wrong. In fact, recent research suggests that people don’t easily stop believing false information. Conviction is most likely a hard-wired trait in our behavior. My guess is that we need it in order to survive challenging times. If you don’t believe you’ll “get through this” then you may give up and die.

30

But if that is correct then conviction (or self-imposed belief) is irrational and irrational behavior cannot easily be changed. The same factors that contribute to stubbornness when you’re trying to survive also contribute to your pig-headedness when you’re flat out wrong.

31

So when you prepare an article on any topic (and this is equally true of me), you will write to support your belief and to undermine any doubts or challenges to your belief. You may have persuasive arguments to shoot down all objections to the points you make but it is easy to find examples of compelling arguments that rely in part on errors of omission. Intentionally (or even subconsciously) omitting relevant information changes the degree of completeness of your argument. In fact, errors of omission are often used to strengthen arguments.

32

Omitting inconvenient facts and points of view makes you an unreliable source of information, even if the omissions make your arguments more persuasive and compelling. You may omit something merely because you couldn’t think of everything you should have mentioned, or because you sincerely believe what you are omitting is not worthy of inclusion. Nonetheless, you can easily paint an incomplete (and therefore inaccurate) picture.

33

But many people turn to errors of omission as a way of winning arguments. If they present only the facts that support their own points of view they can score points and change minds. It’s easier to win people to your cause by omitting relevant information than by pointing out that the other guy left something out.

34

You may also lend more credence to sources you trust than to sources you trust less. Hence, you introduce a bias into your presentation by saying things like “Dubious Dave has shown conclusively that frogs leap 15 feet backwards” and “Scientific Sam argues with Dubious Dave but has yet to provide a compelling argument” and “Silly Sally has challenged Scientific Sam several times.”

35

The key takeaway here is that people favor the viewpoint they believe and tend to treat opposing viewpoints unfairly. To do otherwise might force them to admit they were wrong.

And So Wikipedia Introduces More Problems

36

In addition to all the problems listed above, Wikipedia has its own peculiar set of issues that make its well-cited articles unreliable.

37

Poor writing is the worst aspect of Wikipedia. The English-language Website’s content is a spaghetti-weave of mismatched idiom, much of it written by non-native English writers. But even among the native English writers there are regional variations in word choices and colloquial expressions.

441

38

Poor internal annotation is another grave problem with Wikipedia. It provides no mechanism for explaining idiomatic expressions, unless those expressions are notable enough to have their own articles (but even then many marked up articles do not exist and so explanations are lacking). Use of unexplained idiom makes it hard for the average reader to understand what a given contributor is trying to say or whether other contributors agree with him.

39

Wikipedia’s conflict resolution system always favors the biased reverter. Whenever someone tries to correct a Wikipedia article, the first person to revert that change will win any reversion war because Wikipedia rules don’t allow the person who made the change to make the last reversion. All disputes must be left up to the community to decide, and there you are asking incompetent people to decide between two opposing points of view, one of which almost certainly has more experience in Wikipedia’s in-site debates than the other.

40

Wikipedia’s rules for citation also don’t require that a proper context be provided. For example, you could write “Michael Martinez says that SEO is not all about links” or you could write “Michael Martinez opposes link-based SEO strategies” and then provide a numbered citation that links to an SEO Theory article like “How Best to Use Links for SEO” (where I do in fact say “SEO is not all about links” but go on to show you how to use links in all sorts of SEO).

41

In this example you can write something completely misleading simply by omitting proper context: “Martinez disagrees with SEOs who rely on links.” Most readers will not check the citation so they will never know that I actually show people how to use links in SEO.

42

Wikipedia further instructs contributors not to use blogs as credible sources. Many articles use blogs anyway but sometimes blogs are the only sources of information on a topic. Worse, many news articles cite very unreliable people as sources of information (for many of the reasons given above). In the world of ignorant people the self-confident man is king, no matter how wrong he is. We see that in Internet marketing all the time, not to mention politics and government.

43

Finally, Wikipedia’s content may change at any given time. How credible would you think your neighbor if this morning he tells you it’s June 2 and in the afternoon says June 3 and in the evening changes his mind back to June 2? Would you want to be treated by a doctor who says today, “You have high cholesterol,” tomorrow “your cholesterol is fine,” and two weeks from today says “dubious studies suggest that statistically your cholesterol will fall somewhere inside the marginal zone because you eat butter”?

44

Wikipedia, other Wiki sites, and indeed all crowd-sourced Websites are fundamentally and inherently UNcredible. They lack credibility because they can be changed, not because of whatever they say today.

45

A credible source of information is consistent. An incredible source of information is unpredictable.

442

46

A credible source of information may be reliable or it may be unreliable.

47

In fact, it doesn’t matter how credible Wikipedia seems to you or me, it is truly an unreliable source of information simply because there is no way that any of us can ensure that its information is accurate, reliable, correct, or complete.

48

All the citations in the world won’t change that.

AT ISSUE: SOURCES FOR DEVELOPING A DEFINITION ARGUMENT