Wednesday, March 2, 2016

Anecdotes and Extreme Examples

NOTE: While looking through my old essays, I found another 15 that struck me as particularly interesting. Some may seem a bit dated, as they discuss current events during the period 2008 to 2010, but the principles they discuss are still relevant.

There is no greater frustration than the feeling one gets, having shown conclusively that a government program is harmful, that its passage would be destructive int he extreme, is then confronted by some anecdotal tale of woe, which somehow carries more weight with the audience than hours of logical proof. No, I am not about to become one of those who denounces the stupidity of the masses ("How the Government Corrupts Relationships",  "Deadly Cynicism", "The Citizen Dichotomy", "In A Nutshell", "Cognitive Dissonance Part 2", "The Right Way", "The Danger Inherent in Banning "Bad Ideas"", "Contradictory World Views" ), that would fly in the face of everything I believe. I truly do believe people are generally capable of understanding political and economic questions. The problem is that recent years have been spent teaching individuals to rely more on their emotions than reason, and as a result tales of woe, extreme cases and anecdotal evidence, carry more weight than they should. So the problem is not that people are stupid, but that they have been taught the wrong methods for evaluating evidence.

These arguments tend to take two forms, both of which are invalid. First, there is the extreme case. This usually comes about when you propose ending a welfare program. Rather than explain the merits of the program, proponents will instead find the most extreme example and act as if it invalidates your entire argument. For example, if you claims that ending disability insurance will cause many who are currently disabled to find they can actually work ("Perverse Incentives"), they will bring up cases of quadraplegics or those suffering from ALS. Second, there is the anecdote. This is usually used when proposing a new law. When asked about the necessity of such a measure, rather than providing any overall numbers, or showing how many would benefit from the law, or what the cost would be per beneficiary, instead they will drum up one or two heart wrenching tales and use them as evidence that the law is needed. Be it old people eating cat food as proof of the need for prescription coverage, or people unable to afford insurance as an argument for ObamaCare, the pattern is the same.

I am sure, having heard such arguments for so long, at least a few of my readers are on the verge of asking "So, what's wrong with those argument?" I know no one will admit to it, but I am sure some think that this sort of anecdotal evidence is a valid argument. And so, for their benefit, allow me to take a look at both types of arguments and show, as simply as possible, what is wrong with each.

Let us look first at the extreme example. In some ways, this is akin to the error I mentioned in "Opposing a General Principle With a Single Example", at least some examples of it, as that mistake takes an exceptional case in one area and compares it to the norm in another. On the other hand, as most of the ways the extreme example is used differ quite a bit, so let us look at it in its own right, rather than rely on my earlier argument.

Let us start with an example unrelated to politics. Let us say that I argue "this specific antibiotic will cure 95% of ear infections". And let us suppose that you disagree, offering as proof the claim that your brother had an ear infection, and the treatment did not cure him. In this case, would you not clearly see the error of the argument? I never said it was a 100% cure, so a single counter example proves nothing.

The problem is that the same logic often fails to work in the case of political arguments. When someone says eliminating welfare or disability would end up making a vast majority find they really could work, we are not saying everyone getting benefits can work, just most of them. There may be some cases where individuals cannot work even without benefits, but that falls more under the logic of the anecdotal argument, so we will deal with it there.

Most likely, this argument works for the same reason the anecdote works, it is a sad tale, which appeals to the emotions, and so, confronted with it, the listener is given a false dichotomy, forcing them to choose between heartlessly ignoring all who have needs, and providing for everyone, regardless of the validity of their needs. In effect, the extreme example equates this one outlying case with all cases, and thus suggests every beneficiary is as legitimately needy as the example*.

The anecdote follows almost the same pattern, equating a single example with the whole. But, in this case, it goes farther, not just making a single case stand in for every case, but also using the need of that single case as a substitute for argument about methodology. In the case of the anecdote, we find those who support a given measure basically saying "Of course this is the best way to solve this problem, look at how needy he or she is!"**

The problem here is that sympathy is not a substitute for logic. The fact that someone is needy does not mean that any proposal which appears to benefit them is a good idea. Nor is the fact that a program seems to benefit a needy individual now an argument that it is the best solution, or even that it is the best solution for them. For example, often the elimination of welfare is opposed by arguments about those "who really can't work." The assumption being that were welfare eliminated, they would starve in the streets. But who says that is the truth? Perhaps their families ignore them because the government "takes care of it". If they were truly needy, perhaps their families would finally step in and help. Perhaps, with more cash available, individuals would donate to private charity which would better serve them than the state***.

It is tempting to wrap this up by arguing that there is no justification for the state supporting anyone, no matter how needy, but that is not my point. For those interested in such arguments, I would recommend "Subsidizing Irresponsibility and Poor Planning". However, as I anticipate that the argument there will raise an exceptional case argument of its own, I am going to address that one argument.

In that article, I argue that welfare, unemployment and the rest all discourage individuals from exercising proper prudence, and encourage them to fail to exercise proper caution. In response, I expect people will raise cases of those who cannot care for themselves, who were harmed "through no fault of their own" and could not have exercised any more caution. In those cases, they will case, how would eliminating benefits help? But that misses three points. First, if we allow support for these few extreme cases, we then encourage irresponsibility in a huge number, effectively making the masses irresponsible to benefit a handful. Second, even if the disabled individual may not have had the ability to prevent or limit the harm that befell him, those around him are more likely to help him if they cannot pass the burden on to the state, state assistance grants his relatives and friends a clear conscience when ignoring him. Third, even ignoring all those arguments, there is nothing in our social contract that says you are exempt from harm. It is sad when blameless people suffer harm, but should we destroy our social order, penalize the productive members of society and encourage irresponsibility among the bulk of humanity, all to avoid any hardship to a handful of people? ("Life Is Not Fair - And Trying To Make It So Makes Things Worse")

Which actually may make this whole argument irrelevant. While I started off trying to dismiss an invalid argument, I think I may have made it pointless. After all, against such claims, be they extreme cases, outliers, or run of the mill examples, even if they represent the majority of individuals, isn't the best response "That's not the government's job?" If so, then maybe we need no other argument.


* Of course, one's need is not a valid justification for paying them, unless you accept some variation of Marxist ethics, but as many who do not think consistently tend to look for "just" arguments, the need of the recipient is regularly used as a shorthand for the legitimacy of a claim. Whether this is valid or not will be discussed later.

** In some ways the extreme example also serves this purpose, by positing the loss of services the extreme case will endure as a reason to keep the entire program. Effectively this tries to equate the extreme case with all cases. On the other hand, one could equally argue that it is simply making the case that a blanket exclusion harms this individual, so, for now, I will treat the two as different.

*** Private charity is often dismissed by big state types, but it is both more efficient and more flexible. As it is not obligated to support everyone who meets certain rules, and obligated to turn away anyone who doe snot, it can focus entirely on those truly in need, and often provides a better level of care. But no one wants to believe that, as they imagine it is the job of the state, and individuals are too selfish and heartless to do anything better than the state.



Of course, this is itself a rational argument, opposed to emotional appeal, and as such unlikely to win over those prone to emotional responses to arguments. However, I like to believe that, given time, people will come to expect more reason and less emotion in arguments. The present appeal of emotional arguments only exists because both parties have pushed the primacy of emotion over reason. As more of us provide reasoned arguments, the public will become used to hearing reason, and will come to expect it, and the appeal of emotional argument will decline. At least that is my hope. Time will tell. (See "Defending Freedom?", "Don't Blame the Politicians", "What We Deserve")

Originally posted in Random Notes on 2010/04/07.

No comments:

Post a Comment