Is game theory "useful" for understanding international politics?

According to Robert Powell, the answer is "yes, but know its limits."

[THREAD] Image
Powell's work largely focused on using mathematical formal models (i.e. game theory) to study international conflict.

And, as @jkertzer recently shared, Powell was VERY good at math!
What always stood out to me was his willingness to discuss the limits of using models to understand international politics.

He would do so with short quips buried in his most-influential papers. These quips offered "meta" insights into the role of theory.

Here are 4 examples.
First is his 1991 @apsrjournal paper on "relative gains v absolute gains".…
This paper is his contribution to the great "isms" debate that dominated much of the 1980s in American IR…
He views the debate as boiling down to whether states (a) care about their own gains (i.e. absolute gains), or (b) they care about how their gains compare to the gains of others (i.e relative gains).
His answer, which he supports with a model, is "it depends" (primarily on the destructiveness of fighting) Image
His key "meta" insight comes in the conclusion: the "theoretical" and the "empirical" should not be confused...scholars tend to forget this.

Something could be theoretically useful for gaining insights - say, the unitary actor assumption - even if it is not empirically supported Image
Second is his 2004 @AJPS_Editor paper on how war fighting is a CONTINUATION of bargaining, NOT what happens when bargaining ENDS.…
He acknowledges that treating "war" as the end of bargaining has been useful. Hence, it's not "wrong".

But he then notes that this limits what formal models can say. Hence, to understand more about war, we have to model the wars themselves. Image
Powell thinks progress can be made by treating war as a continuation of that bargaining, not as a different process from pre-war bargaining. Image
But he finds bargaining models to be VERY sensitive to assumptions about (1) the information available to states & (2) the bargaining processes they use (the figure below is a stylized view of such a process). Image
This leads to his key "meta" insight: when theoretical insights are highly contingent on assumptions, make assumptions that match empirical reality. Image
Third is his 2006 @IntOrgJournal paper, which is Powell's most cited paper.…
Powell's paper is a direct response to the 1995 @IntOrgJournal paper by James Fearon, "Rationalist Explanations for War" (which I discussed in a #KeepRealismReal thread)
Powell agrees that war is puzzling because it seems REALLY inefficient: it's costly and results in outcomes that could have been reached without the cost.

His goal in the paper is to focus on inefficiency. This means assuming their on no informational problems. Image
He warns that one can't just treat "commitment problem" as "anarchy". That won't shed any insights because, well, anarchy is a constant Image
Instead, he thinks commitment problems arise and undermine the prospects for peace when there are concerns over future power shifts Image
But he also warns against focusing too much "informational" explanations for war. This leads to his next "meta" insight: theory should not generate "bizarre" readings of history. Image
His example is prolonged conflict: if it's just about information, then long wars must be due to the sides never learning more about each other. But that simply can't been true. Image
Fourth (and my favorite) is from his 1999 @PrincetonUPress book "In the Shadow of Power".…
Why my favorite? Because it's a line I've used many times (in papers, in lectures, and here on Twitter).
When talking about the use of the bargaining model to understand war, he writes:

"The models are very spare, and, indeed, almost certainly too spare to explain any particular outcome in any degree of specificity."
In other words, the models are great for helping one thinking about conditions that make war LIKELY.

In this way, formal models do the same thing as statistical models: they help identify CONDITIONS.…
But causes for particular cases? That's trickier and is really beyond the purpose of ANY model (it is, after all, a MODEL).
In sum, Powell taught us:

- theory assumptions NEED NOT be empirically grounded

- theory assumptions SHOULD be empirically grounded if theory's INSIGHTS are highly sensitive to assumptions

- theory shouldn't lead to bizarre history

- theory is about conditions
Thinking clearly about theory is just one reason the discipline of International Relations is better off thanks to Robert Powell.

And is also why he'll be missed.


• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Paul Poast

Paul Poast Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @ProfPaulPoast

May 21
The world is facing a wheat crisis due to the War in Ukraine. It will be devastating. But war leading to a wheat crisis is not unprecedented. This also happened during World War I.

How the world adapted then could teach us what to do now.

[THREAD] ImageImage
Before talking about World War I, it's important to understand the basics of the current wheat crisis.

Read 37 tweets
May 14
How can the USA and Russia directly fight over the skies of Ukraine without triggering World War III?

Let's consider the lessons from a classic source...Top Gun (1986).
As many folks are aware, debates on NATO installing a no-fly zone over Ukraine centered around concerns that direct US-Russia combat could trigger escalation and a wider (possibly nuclear) war.…
Missing from that debate (and I only say this half joking) are lessons from Top Gun.
Read 33 tweets
May 9
No "Mission Accomplished" statements from Putin during his "Victory Day" speech.

But he did rehash his reasons for the war, namely that Russia "struck back preemptively".

Say what? That simply doesn't make sense.

Obviously, this is part of Putin's long standing complaints about NATO expansion (see the very beginning of his Feb 24 address).…
Read 17 tweets
May 6
Lots of speculation that Putin might declare "victory" on May 9. But with the war in Ukraine far from over (and not going Russia's way), why would he do that?

The truth is that leaders are often eager to declare victory.

Look no further than George W. Bush.

We'll come back to Bush shortly.

With respect to May 9, that is "Victory Day" in Russia, marking the surrender of Nazi Germany (t was on May 9, 1945 when Stalin officially announced the end of the war).
Putin REALLY likes this day, especially giving the annual address.…
Read 33 tweets
Apr 30
"Lend-Lease" -- a World War II program to supply USA allies with arms -- is back.

How did it work then and what does it mean for the War in Ukraine now?

By "lend-lease being back," I'm referring to Congress approving the "Ukraine Democracy Defense Lend-Lease Act of 2022"…
The full text of the act is available here…
Read 27 tweets
Apr 28
@e_sarotte & I are looking forward to our conversation this Friday at @UClubChicago.

Here's a mini-thread previewing the issues we’ll discuss, drawing from our writings & commentary on the subject.

Putin has talked about NATO expansion A LOT, even using the phrase "not one inch". Is he just using enlargement as an excuse, or is he truly worried about NATO moving eastward?…
Related, why is this happening now?…
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!


0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy


3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!