My first take was that the funny thing is that Friedman will probably not notice when war does not break out between the McD US and non-McD Iran. That confirmation/observation bias would mean that a non-event that would contradict his claims would go unnoticed.
Laurenist pondered whether there was a McD knock-off in Iran that would suffice. And that gets to the heart of the problem: just as there are multiple causal logics to explain the democratic peace (democracies don't fight with each other because of norms, domestic structures of accountability, transparency, etc.), we do not have a clear idea why dyads (pairs of countries) which both have McDs within them do not war with each other.*
* Yes, there are exceptions but social science is a probabilistic thing--few things are true 100% of the time (no law of gravity here).Why don't two countries having McD's fight each other? Are their citizens too obese to form armies? Hardly, given American obesity and yet willingness to war. Is it that consumer-oriented cultures are less interested in expansion and more interested in video game destruction? Is it that globalization as represented by McD's means that countries have joint interests?
One way to test all of this is to actually compare franchises: see if Burger King (with its monarchical symbology H/T to Laurenist) produces correlations akin to that by McD's. See if knock-off chains do better or worse? Their reckless defiance of rule of law by ripping off intellectual property rights suggests worse.
Hmm, there is much pop social science to be done: to the Bat-computer we must go.
Actually, the fun thing here (for me, as the poli sci geek) is that this example is so very fruitful for thinking about the limits of correlation and concepts. A correlation is not a theory. What is the causal logic? Is it a fast food peace or is there something about McD's that matters--which means that our theory tells us how far we can apply the concept.