# Tag Archives | probability

## The inability to find any record of Harrison J. Bounel means he must be a real person!

The American Thinker published an article, “Barack Hussein Obama and Harrison J. Bounel,” by criminologist Jason Kissner today that argues that the association between the name Harrison J. Bounel and Barack Obama in some unspecified transaction in a public database is much more than coincidence or an attempt at identity fraud. While the argument is couched in the language of probability, no actual math was harmed in the writing of the article.1

Detractors of the birther Social Security fraud theory point out that no one has been able to locate this Harrison J. Bounel who birthers claim is a real person to whom the social security number used by Barack Obama was originally assigned. If such a person existed, debunkers say, then one ought to be able to find some independent record of him.

Kissner turns the tables and says that the inability to find other records of Bounel is proof that he is a real person (are you confused yet?).  Kissner’s argument is that the database record could not be fraudulent because it would have been too difficult to actually find someone with no other record in order to perpetrate the fraud.

Kissner’s claim is inserted in a straw man argument refuting the idea that some anti-Obama person created the fake record for Bounel. I suppose someone might have speculated on the possibility that the public record for Bounel with Obama’s SSN was created for the purpose of creating an anomaly in Obama’s record (I might have even done it), but I think it is more likely to be an error or an attempt at identity theft. The straw man context is not important because Kissner’s argument fits the real argument of random error or fraud as well as it does the straw man.

Kissner’s fallacy, however, is the ad hoc assignment of probability to something that’s already happened. It’s like looking at the winning lottery number and saying “what are the chances this number would come up?” and then arguing that the lottery must be rigged. In order to make the probability argument, one has to set the criteria in advance, or they have to be necessary. The fact that the name “Harrison J Bounel” doesn’t belong to anybody is not necessary to the hypothesis and so it’s not significant, no matter how improbable it is.

But the probability of coming up with a name belonging to nobody isn’t that low; in fact, it’s easy. I took the names “Kissner” and “Bounel” and used them with the first and middle names of the members of my immediate family (6 total). I got only one hit on Google for the 12 names I tried.

And finally Kissner’s argument works equally well against Obama fraudulently using an SSN belonging to Bounel: How could Obama have found someone so totally devoid of any record?

What Kissner won’t address is: assuming Harrison J Bounel is a real person (birther hypothesis), what are the chances that there is no record of him anywhere, not birth announcement, obituary, immigration record, census, city directory, grave, genealogy nor criminal record?