Welcome! Log In Create A New Profile

Advanced

steve horwitz on false accusations

October 19, 2018 11:53AM
Steve Horwitz
October 17 at 12:37 PM ·

It might be the case that the historical data on false accusations aren't relevant when you change the default rule about how we treat accusations of a crime.

Suppose there are some number of people who want to harm others with false accusations but are deterred by the current rules of evidence and the presumption of innocence. Now we change the rule to “believe all accusers.” This will increase the payoff to a false accusation by reducing the burden of proof on the accuser. This will lead to more false accusations on the margin. Suppose we are able to detect even a minority of those false accusations. The increased absolute number of proven false accusations will cast doubt on all accusers, including legitimate ones, undermining the goal of “believe all accusers” and reducing the number of guilty people actually apprehended.

But note that because the rule “believe all accusers” changes the payoffs to those inclined to engage in false accusations, the claim that false accusations have historically been rare might not be very relevant once we change the rule to “believe all accusers.”

Change the rules, change the payoffs, change people's behavior. Econ 101. Believing all accusers may well lead to more false accusations, which harms both the falsely accused and genuine victims making accusations. And the fact that such false accusations have historically been rare is not a reason to think they would be rare, and thereby dismiss the possibility of false accusations, under a new set of rules.
Subject Author Views Posted

steve horwitz on false accusations

chimi 251 October 19, 2018 11:53AM

Re: steve horwitz facebook public figures page

chimi 112 October 19, 2018 11:55AM



We're sorry, but you must register to post in this forum. This is the only way we can keep spammers out. You can remain totally anonymous, and we will keep your email address completely confidential.

Click here to login