www.blogherenow.net

In my previous post, “The Republican Modus Tollens”, I pointed out that arguments apparently having the valid form

1. If P then Q
2. Not-Q
3. So Not-P

allow for serious irrationality when P represents a matter of well-confirmed scientific theory and Q represents a prescriptive policy preference or a tenet of religious faith. So, to use one of the examples from my previous post, we have arguments about climate change that seem to be guided by the following pattern of thought:

1. If climate change is occurring, then we should regulate CO2.
2. We shouldn’t regulate CO2.
3. So climate change isn’t actually occurring.

Of course, probably no opponent of climate change has ever explicitly made just this argument. My point is that the arguments they do make are at least partly motivated by some such pattern of thought. Representing the pattern this way makes clear that they are often reasonable enough to recognize that if climate change were occurring, then we (perhaps) should regulate CO2. The problem is that they also very strongly desire not to regulate CO2 (perhaps for quite defensible reasons, such as worrying about the economic effects of such regulation), and this very strong desire against a possible policy choice, along with the normally valid modus tollens pattern of thought, leads them irrationally to deny a well-confirmed theory. In order to do so, they must massively over-weigh evidence contrary to climate change, sometimes fantasize about global conspiracies of scientists, and so on. It is this last move - the irrational denial of a scientific theory - that indicates they are being guided, at bottom, by strongly held policy positions and this modus tollens pattern of thought, or something similar to it.

Psychological explanations for why people argue like this aside, I suggested that the main logical problem with such arguments is located in their conditional (‘If P, then Q’) premises. This problem arises whenever P describes a putative matter of fact and Q expresses a prescription of some sort (often signaled by the inclusion of ‘should’ or ‘ought’). In such cases, the conditional statement can be viewed in one of two ways. If we view it as the sort of statement that actually belongs in a modus tollens argument (what logicians call a ‘material conditional’), then it can be criticized as being false on purely logical grounds. Famously, ‘is’ does not materially imply ‘ought’ - descriptive language does not materially imply prescriptive language. That does not mean that facts are irrelevant to policy choices, of course. As I put it in my previous post, facts can certainly bear on policies; it’s just that they never logically necessitate policies. On the other hand, if we view the conditional premises as mere recommendations, then they don’t belong in a modus tollens form of argument at all, the argument form is only superficially similar to modus tollens, and the conclusion does not validly follow from the premises.

Now, almost as famous as “‘is’ does not imply ‘ought’” is another philosophical saying: “One person’s modus tollens is another person’s modus ponens”. Modus ponens is, like modus tollens, a valid form of argument that starts from a conditional premise. But in its second premise, instead of denying the conditional’s consequent (the statement that follows ‘then’), it affirms its antecedent (the statement that follows the ‘if’), and instead of deducing the conditional’s denied antecedent, it deduces its affirmed consequent. This sounds a lot more complicated than it is, as this sketch of the form shows:

1. If P then Q
2. P
3. So Q

Starting from the conditional premise in the previous argument, we arrive at an argument that is commonly asserted by liberals:

1. If climate change is occurring, then we should regulate CO2.
2. Climate change is occurring.
3. So we should regulate CO2.

“When conservatives deny the antecedent because they deny the consequent, they are allowing their policy preferences to influence their view of the facts… when liberals affirm the consequent because they affirm the antecedent, they are merely allowing their view of the facts to influence their policy choice.”A conservative upset with my prior post might try to turn the tables, observing that if the conditional premise is a problem in the modus tollens argument, then it is equally a problem in this modus ponens argument. I would reply that yes indeed, it is a problem, but not quite equally one. This is because in the cases that concern us, the antecedent is descriptive and the consequent is prescriptive, rather than the other way around. It is still true that the conditional statement is either false or not a material conditional (because it is a mere recommendation). But the difference is that when conservatives deny the antecedent because they deny the consequent, they are allowing their policy preferences to influence their view of the facts. On the other hand, when liberals affirm the consequent because they affirm the antecedent, they are merely allowing their view of the facts to influence their policy choice: precisely the rational thing to do. Even if both arguments are unsound or invalid from a purely logical viewpoint, in these cases the modus tollens sort of argument is irrational in a way that the modus ponens sort of argument is not.

But couldn’t the Republican arguments be expressed in a modus ponens form? Certainly. For instance:

1. If climate change is not occurring, then we should not regulate CO2.
2. Climate change is not occurring.
3. So we should not regulate CO2.

The problem is that this way of representing their pattern of thought leaves the conspiracy theories and evidential biases that they rely upon to justify premise (2) totally unexplained. The same can be said for the other two arguments I discussed in the last post, in which premise (2) (of the modus ponens versions) would deny evolution and the safety of the HPV vaccine, respectively. Admittedly, some who deny evolution sometimes do so by promoting the notion of intelligent design, but the arguments for intelligent design are at least as weak as the arguments for the denial of climate change. The modus tollens representations help to explain why Republicans make such arguments (namely, because they are passionately against certain policies that they think might follow from acceptance of the facts); the modus ponens ones don’t.

By the way, some readers might have noticed an asymmetry in the title of this post versus that of the previous post: here I used ‘liberal’ instead of ‘Democratic’, whereas there I used ‘Republican’ instead of ‘conservative’. That is because, if the behavior of the House of Representatives and the current crop of Republican presidential candidates is any indication, the Republican party really has morphed into a purely conservative party. For better or for worse, thanks at least to the “blue dog” contingent, the Democratic party has not yet made a similar transformation into a purely liberal party.

Finally, it’s perhaps worth emphasizing that although the sort of irrationality I’ve sketched out above currently seems more common on the Right than on the Left, liberals are certainly not immune to it. The more passionately one holds a policy position, the more likely one is to fall into this style of thinking, and liberals can be just as passionate as conservatives. The answer, of course, isn’t to be less passionate. It’s simply to be more mindful of how those passions might influence one’s thinking.