Also Known as: Sharpshooter Fallacy
Description:
This fallacy occurs when it is concluded that a cluster in a set of data must be the result of a cause (typically whatever the cluster is clustered around). This fallacy has the following form:
Premise 1: A cluster L occurs in data set D around C.
Conclusion: Therefore, C is the cause of L.
This causal fallacy occurs because the conclusion is drawn without properly considering alternatives. One ignored alternative is that the cluster might be the result of chance. Another ignored alternative is that the cluster might be the result of a cause, but not the claimed cause.
A cluster can provide grounds for considering a causal hypothesis that can then be properly tested. However, this correlation does not establish causation. Given the role that correlation (in this case, clustering) plays, this fallacy could be considered a variation of the Cum Hoc Ergo Propter Hoc fallacy. However, Texas Sharpshooter has a history of its own that warrants its inclusion under its own name.
The fallacy’s name is derived from a joke about a person (usually a Texan) who shoots away at the broad side of a barn. He then paints a target around the biggest cluster of bullet holes and claims to be a sharpshooter. This creates the illusion that he is a good shot, just as focusing on clusters and ignoring the rest of the data can create the impression of a causal connection. As such, this fallacy can also be seen like Incomplete Evidence in that when a person “draws the target” what is outside the target is conveniently ignored. Since Texas sharpshooter is specifically a causal fallacy, it can be distinguished from the more general fallacy of Incomplete Evidence in this way.
This fallacy can be committed in good faith, out of ignorance of how to engage in good causal reasoning. It can also be used intentionally in bad faith, to try to prove a claim. For example, a person trying to prove that something causes a disease might examine data until they find the clustering that appears to “prove” their claim. As with any fallacy of reasoning, the conclusion could be true. The problem is that the evidence offered fails to support it.
Defense: To avoid being taken in by this fallacy, the defense is to consider whether adequate evidence is offered for the data based causal claim or if the only evidence is the clustering. If you are unsure, the rational thing to do is suspend judgment. It is also important to not fall for applying the fallacy incorrectly. For example, a person who wants to reject a causal claim might wrongly insist that the clustering must be the result of this fallacy.
Example #1
Rich: “Hmm, this data shows that the number of cases of cancer in Old Town is greater than the national average.”
Alice: “Interesting. Do you have any data that is more precise?”
Rich: “Indeed, look at this graphic. As you can see, it shows a significant clustering of cases near the paper mill.”
Alice: “Wow! Those poor people!”
Rich: “You know makes it really bad?”
Alice: “What?”
Rich: “The housing around the mill is for retired senior citizens!”
Alice: “Wait, what?”
Example #2
Michelle: “I was reading through the predictions of Nostradamus. He must have been able to see the future because his predictions came true.”
Hilda: “What did he get right?”
Michelle: “Well, he predicted Hitler. He said ‘Beasts wild with hunger will cross the rivers, The greater part of the battle will be against Hister. He will cause great men to be dragged in a cage of iron, When the son of Germany obeys no law.’”
Hilda: “Wow, that is amazing! ‘Hister’ is close to ‘Hitler’, he was German…well close enough anyway and he did cross rivers.”
Michelle: “Like I said, he made those predictions because he could see the future.”
Hilda: “Did all his predictions come true? That book you have is huge.”
Michelle: “Well, he did write hundreds of predictions and only a few have come true. But he was seeing the future so it will take a while for them all to come true. The important thing is that he got Hitler and some other things right so far!”
Fran: “You know that ‘Hister’ is just the Latin name for the Danube River, right? Also, your translation is a bit off. In any case…”
Michelle: “Shut up!”