A Little Bit on Anecdotes

I came across a plea today on Facebook, asking for people to send to a particular individual stories about the negative impact of the Affordable Care Act (ACA), colloquially called “Obamacare,” for the president who signed it into law.  I pointed out that anecdotes are not good evidence, and I had objections to their use as such, since I feel it is emotionally manipulative and therefore not totally truthful.  I was met by a lot of opinions counter to my own. The arguments were fairly varied, there were some ad hominem attacks, but most seemed to take my use of the word “anecdote” as some sort of insult, so I thought I’d write a short piece explaining anecdotal evidence, appeals to emotion, and the part I often see them play in politics.  Here is the post:
I saw it because it was shared by Julie Borowski, who does some great videos on libertarian issues on youtube, check her out:
I. Anecdotal Evidence.
Anecdotal evidence isn’t really evidence, at least in the clinical or scientific sense.  It is a story, or an example, usually involving a real person or a particular case. Moreover, it is often used in the place of evidence, supporting or intending to lead to a particular conclusion. Facts only become evidence in the context of a conclusion, like saying “smokers die of heart disease at a higher rate than non-smokers” supporting “People should not smoke.”  Telling a story about a particular negative impact of the ACA, like saying, “This is Dan and Rachel. Obamacare cost them $10,000,” is an anecdote; a story.  It might make you feel bad, but it doesn’t represent the negative impact of the ACA for anyone beyond them. I could tell you a story about my father being killed by a truck. Would you be persuaded to ban trucks?  It’s an anecdote, and isn’t sufficient for the support of most arguments.
II.  Confirmation Bias.
            Anecdotal evidence tends to contain elements of confirmation bias, which is a logical fallacy where one ignores evidence that is contrary to one’s argument or conclusion. The woman in the link was essentially asking for such.  She did not ask for positive stories about the ACA, she asked for specifically negative ones. If the stories were truly evidence, they would be neutral to the collector, who would then form a conclusion from it, not the other way around.
III.  Data and Presentation.
            Somebody asked, “What is the amount of stories of people being affected [that would] drive us over the hump of “anecdotal”? 10,000 stories?” (brackets added by the author of this article). The short answer is that stories cannot be evidence until they are put in context of an argument, and then, who would read them? When dealing with statistics it is the commonalities that matter and must be presented; the facts of the story are superfluous beyond the commonality they have with one another.  Here is a fact:
Under Obamacare, 30-year-old men face average premium hikes of 260%
This is data that would be common to all the people affected, including myself.  They may live in New York or Florida, be gay or straight, be black of white.  What matters about the data is the commonality that allows us to put it in context.  If I said, “I’m seeing a 260% increase in my health insurance premiums,” you could not consider it evidence that the AVERAGE 30 year-old man would see a jump in his premiums.  Only data collection can support that.
II.  Emotional Manipulation.
            Why do we tell so many stories?  Why are there so many human interest pieces on the news?  In short, they are effective, or the news providers wouldn’t use them. Politicians would abandon all appeals to emotion if voters did not respond to them well. That’s really what anecdotes are; They aren’t evidence.
            Is emotional manipulation truth?  The story might be true, on some level, but is the decision that is a reaction to that story based on truth.  Essentially it is not, or it would be made with reason and evidence instead of just emotion. The story does not create an honest relationship between the speaker and the listener. In fact, the speaker has made an assumption on some level that the listener will not respond to evidence alone.  At this point the rationalization for engaging in emotional manipulation is the ends: convincing someone of the correct position for the wrong reasons. 
IV.  Moral Arguments.
Whether or not the end justifies the means has been a topic in morality and ethics discussions since the dawn of time.  Those of you who know me know my opinion on it: action is always the locus of discretion in morality.  When you make an ends-justifies-means argument, or imply it through your actions, you are representing an entire moral system that moves the locus of discretion away from the actions themselves to the consequences (or intent, if you are a Kantian).
What is interesting about the above the appeal for stories is that they desire to show the negative impact (the consequences of means) of the implementation of the ACA when the originators consider such consequences as being justified by their perceived ends. They are essentially making a moral argument that the end does not justify the means while simultaneously utilizing actions that support the opposite conclusion. The moral argument is self-contradictory if you yourself are unwilling to apply it.
A true moral argument against the ACA is actually quite easy to make, especially if you use libertarian morals:
Killing people is wrong.
If you don’t get health insurance, the government will tax/fine you (for the purposes of this argument those actions are interchangeable).
If you don’t pay the fine, the government will eventually arrest you.
If you feel the arrest is unjust, and refuse to be imprisoned, you will be killed.
The ACA is immoral/evil.
V. A Fight With Mike Tyson.
            The left has been doing the emotional manipulation game a very long time. The image of the right as being heartless automatons is mostly a product of how good the left is at evoking emotion in people. Emotional appeal is their world, and you better be prepared if you want to step into the ring with them, because it will be like jumping into a fight with Mike Tyson in his prime. If you were wanting to knock them out, you would have a very hard time of it, especially in the health care debate.
This is ultimately the problem with exchanging stories: there is no end to them, and the listener is left to decide the merits of them on a one-to-one basis, rather than as a whole field of evidence. You can tell a very touching story about the financial hardships brought on by the ACA on a particular family, but the other side can merely tell their own story about someone who died because they didn’t have health insurance (I used Chuck Schuldiner of Death as an example).  If you weigh these two stories on their own, and believe that the needs of the many outweigh the needs of the few, or that greater needs outweigh lesser ones, than even logically you will lose the argument.
The real argument attempting to be made is that the amount of pain inflicted by the law is greater than that it prevents. One can only make this argument beyond anecdotes, in the world of real, large-scale evidence.  You could certainly add anecdotes to real evidence, but what does that accomplish?  Either someone is persuaded by reason or they aren’t; I don’t believe you can add reason and emotion together and reach some threshold that tips the scales in your favor.
VI.  Summation

Define your position. Make real arguments with meaningful evidence and avoid biasing it. Follow through with consistent moral ideology. Make pragmatic and trade-off arguments with evidence, not appeals to emotion. Don’t resort to tactics that are intellectually and morally inferior to reason. 
The Mike Tyson reference is one I see used a lot by Stephan Molyneux, a brilliant libertarian philosopher, and it is usually quite a good analogy.  Check him out at: http://www.freedomainradio.com/

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.