See Text Version of Website


Musings > Tuesday, 30 November 2010 17:02:06 EST

What Is Evidence?

Keywords: epistemology, uncertainty

One thing that I notice all over the place is a lack of understanding about what makes for good evidence.  Often, with conspiracy theories and other forms of nonsense, people tend to present a number of facts, which may be perfectly true statements, that they call evidence but which aren't.

Much of the time, instead of evidence, these are just facts that happen to be in accordance with their theory.  These are facts which you would expect to be true if the theory was true.  The problem is that if the theory is false, that doesn't really much change the likelihood of those facts to be true.

Evidence should point in a very obvious direction.  For good evidence, you would expect it to be much less likely for that evidence to exist if the theory was false than if it were true.  If there are numerous different explanations for a fact, and your theory isn't even the most likely explanation in the first place, then what you have isn't evidence.  it's just a fact that's in accordance with your theory.

For example:

If I went out and got my suit dry cleaned, would you take that as evidence that I'm planning on attending a fancy party soon?  Certainly it's the case that my actions are in accordance with that theory, but there could be several reasons that I got my suit dry cleaned.  I could have gotten a stain on it the other day, or I could be planning to take my wife out to dinner.  It could be that I've just had the suit sitting around for a long time and thought it could use a cleaning just in case I need to wear it at some point.

The theory that I'm planning on attending a party is just one of the many possible explanations for the fact that I had my suit dry cleaned.    It's not evidence, because it doesn't sufficiently narrow down the options so that you can make any kind of informed decision about what my reasoning was.

Some people seem to think, though, that if you pile up a  bunch of these facts together that they add up to actual evidence.  I was even having a discussion with one man who believed this about the ancient aliens theory not too long ago.  He specifically said to me that he acknowledged that none of the facts he was quoting to me were convincing on their own, but that taken together they unquestioningly pointed to the visitation of aliens in ancient times.

Well, the fact of the matter is that he's wrong about that.  You can't take a whole bunch of non evidence and expect it to add up to evidence. 

What you can do is that  if you have some really good evidence for your theory, you can use facts that are in accordance with your theory to bolster and illustrate your evidence.  But those facts are still not evidence on their own, no matter how many of them you have at your disposal.

So if you know that I'm good friends with somebody who's having a party sometime soon, that's good evidence that I'll be invited.  If you then find out that I got my suit dry cleaned, that helps to bolster the theory that I'm planning to attend the party.    But on it's own, it can't be taken as evidence.

And this brings up a point about evidence: it should be the starting point.  The theories that you form should be guided by the evidence available.  What more often happens in these cases, though, is that people start with the theory and then look for facts that fit with it.  When you do that, you can usually find tons of facts that are in accordance with the theory... but you're fooling yourself if you think that's actual evidence.

What it is, is anomaly hunting.  I talked recently about a good illustration of this process.  To show how silly anomaly hunting can be, an anonymous writer started from the theory that the destruction of the Death Star in Star Wars was an inside job.  Using that theory as his guide, he scrutinized the details from the movie and managed to find a number of facts that were in accordance with his theory.  And all his reasoning would sound incredibly convincing to people living in the Star Wars universe who didn't have first hand knowledge of the incident.

The only problem is that none of it is evidence, it's all just facts that are in accordance with the theory.  What you've got to ask yourself about any fact is if the theory turned out not to be true, would it be that much less likely for this fact to exist?

It gets tricky here, though, because if you confront conspiracy theorists they're often so committed to their story that they say "There's no way this would be true if my theory was wrong!".  What can you say to that other than call it a subjective difference of opinion?

So in the end it's a judgment call, and I wish that there were more objective ways to make this kind of evaluation.  But I think that most people will be able to agree on what counts as strong evidence and what's obviously just a fact that's in accordance with a theory.  The problem is the people who see patterns in random noise and can't be convinced that it's just an illusion. 

We need more people who understand about how we can be fooled into believing things that aren't true - people who are willing to question themselves and others even about things that they agree with or have sympathies for.  Perhaps that would help provide a societal inoculation against nonsense.


Latest Comments:
On 6/20/2017
at 5:35am
John Ahmed wrote:

We are Russell Financial Company Ltd located in America and other Europe countries, we offers both personal and investment...

On 6/5/2017
at 11:57pm
Bernie wrote:

This site is crazy :) <a href" ">motrin liquid gels coupon oise</a> The U.S. government was bracing for...

On 6/5/2017
at 1:57pm
Julian wrote:

Very interesting tale <a href" ">methotrexate injections for ra</a> Ana Navarro, a GOP strategist, thankfully decided that she'd...

On 6/4/2017
at 5:20pm
Wilmer wrote:

I sing in a choir <a href" ">cost of crocin advance</a> Falcone admitted wrongdoing and agreed to pay...

On 6/4/2017
at 2:57pm
Darryl wrote:

How many weeks' holiday a year are there? <a href" ">buy cheap clomid uk jsaonline</a> Of late, multiple...

Blog Entries: