This is a theme I’ve considered in numerous previous posts, but it’s worth revisiting in light of recent events. Tyler Cowen directed me to the following tweet:

Consider what would happen if you surveyed 1000 Russians with the following question:

A.  Do you favor using Russian troops to liberate Ukraine from its Nazi-like government?

Then ask another 1000 Russians the following question:

B.  Do you favor invading Ukraine if the locals greet Russian troops with hostility?

I suspect the poll results would differ.  So which poll result reflects actual Russian public opinion?  It depends what you mean by actual opinion.  Do you mean views prior to being well informed of the facts, or views after being well informed on the facts? Views on the invasion they might have imagined, or views on the actual invasion?

Here’s an analogy.  You’ll get one set of answers if you ask Americans if we spend too much on foreign aid, and another if you first tell Americans the relatively small amount we actually spend on foreign aid, and then ask them if that’s too much.  Which one is the actual opinion?  The poorly informed answer or the well-informed answer?  I’d say both, but for different purposes.

People sometimes resist my claim by suggesting that public opinion exists, but that it’s not solid like the trunk of a tree, rather it’s fragile and easily blown about like the leaves on a tree.  But even that isn’t quite right.  We are dealing with something more akin to the Heisenberg uncertainty principle.  Merely asking the question actually changes the answer.  The answers on the foreign aid questions differ because the question can be framed in a way that provides more or less accurate information.  The same is true of the two Ukraine invasion questions shown above.

Because I’m a philosophical pragmatist, for me the bottom line on truth is always usefulness.  If you want to consider public opinion, you also need to consider the purpose for which it will be used.  For instance, are you trying to win an election? 

Putin might be interested in Russian public opinion before launching a war.  But he can also shape public opinion because he controls the Russian news media.  So Putin would make a mistake to rely too much on artificially “manufactured” public opinion.  To employ a term used by economists, it’s not “structural”.   If he’s smart, he’d also be interested in what public opinion in Russia will be once the Russian people learn that Ukrainians view them as aggressors, not liberators. 

So when I say there is no such thing as public opinion, I don’t mean that people don’t have opinions.  Rather I am suggesting that there is no single public opinion that is invariant to the way a question is asked.  Public opinion can be manufactured in many ways, including political propaganda, but also including the framing of survey questions.                   

PS.  In my previous post, I was skeptical of the willingness of Western governments to impose tough sanctions on Ukraine.  (Whether sanctions would be wise is a different question.)    This tweet caught my eye: