Foreign Policy Blogs

Establishing and Disproving Causality

I’ve noticed a couple arguments recently that use the same logic to disprove a causal relationship.  The first one comes from Alan Dershowitz:

[J Street’s] Executive Director, Jeremy Ben-Ami, has joined the off key chorus of those who falsely claim that Israel, by refusing to make peace with the Palestinians, is placing the lives of American soldiers at risk… [This] argument is totally false as a matter of fact. At the same time that Israel was seeking to make peace in 2000-2001 by creating a Palestinian state on the West Bank and in Gaza with a capital in East Jerusalem, Al Qaeda was planning the 9/11 attack.

The second one comes from John Bolton:

[Obama] has repeatedly said he believes lowering U.S. nuclear-warhead levels will encourage support for the Non-Proliferation Treaty’s weapons prohibitions on non-nuclear-weapons states. This is the purest form of theology, since the empirical evidence is entirely to the contrary. As the Cold War ended, Moscow and Washington made dramatic reductions in warhead levels, huge in percentage and absolute terms. Nonetheless, nuclear proliferation continued, and the pace is quickening.

Both men try to refute a causal argument:

If A, then B.

By attempting to demonstrate:

If no A, then still B.

This isn’t the best way to disprove a causal relationship, though.  The best way is to argue that something else entirely is causing the second thing to happen:

If C, then B.

Bolton doesn’t do this.  Dershowitz, however, does:

…[T]he basic complaint that Muslim extremists have against Israel is not what the Jewish state does, but what it is: a secular, non-Muslim, democracy that promotes equal rights for women, gays, Christians and others. Regardless of what Israel does or doesn’t do, its very existence will be anathema to Muslim extremists.

But one could still argue that Israel’s behavior functions as an “antecedent condition.”  To quote Stephen Van Evera, from his book, Guide to Methods for Students of Political Science, an antecedent condition is:

A phenomenon whose presence activates or magnifies the action of a causal law or hypothesis.  Without it causation operates more weakly (“A causes some B if C if absent, more B if C is present” – e.g., “Sunshine makes grass grow, but causes large growth only in fertilized soil”)…

The “fertilizer” argument probably best explains the situation.  After all, Osama bin Laden kicked off his 1996 fatwa by noting, “The horrifying pictures of the massacre of Qana, in Lebanon are still fresh in our memory.”  It’s difficult to argue that such propagandistic proclamations about Israeli behavior play no role in inciting extremism.

Similarly, we can understand the holes in the Bolton argument by considering the effect of an antecedent condition.  Will U.S. strategic nuclear arms reductions lead to increased support for the NPT?  They didn’t before, why would they now?  Maybe before there was an antecedent condition lacking.  As Van Evera notes, sometimes the lack of an antecedent condition can make causation operate “not at all,” as in:

“A causes B if C is present, otherwise not” – e.g., “Sunshine makes grass grow, but only if we also get some rainfall.”

Perhaps there’s a factor that exists now that was lacking in the early post-Cold War days.  Perhaps this factor, whatever it may be, could lead to the causal relationship whose existence Bolton hopes to disprove.