Cognitive biases and the rise of China

Harvard’s Alastair Iain Johnston has a must-read article in the Spring issue of International Security in which he dissects the spread of a meme of China’s “new assertiveness” spread among policy analysts, the media, and scholars in the US in 2010. (Available for free as a pdf, at least for the time being.)

As Paul Pillar notes, Johnston not only raises questions about whether China’s foreign policy has become more assertive since the 2008-2009 financial crisis, citing numerous examples of continuity, but he provides an important examination of how foreign-policy narratives form in the twenty-first century. By using memetics — the process by which an idea or belief spreads from mind to mind — Johnston provides a new way of thinking about how conventional wisdom forms. As Johnston notes, the media has played an agenda-setting role before, but with the emergence of the blog and the rise of Twitter as a medium for the exchange of serious ideas, foreign-policy discourse, especially in the US, seems qualitatively and quantitatively different, not just moving faster but also occurring in much greater volume than ever before.

After all, not too long ago to receive the latest conventional wisdom on foreign policy you had to wait for the latest issue of, say, Foreign Affairs to arrive in the mail, with a steady diet of newspaper op-eds and weekly magazine articles to tide you over. Now in the time between issues of Foreign Affairs you can read daily the fifteen blogs at the website of the Council on Foreign Relations (which publishes Foreign Affairs), not to mention the fifteen blogs at Foreign Policy‘s website and numerous other blogs hosted by think-tanks and publications, and the Twitter feeds of the contributors to these blogs and magazines. This all amounts to what Johnston calls a “discursive tidal wave.”

Johnston talks about first-mover advantages and herding when it comes to the formation of prevailing narratives in foreign-policy discourse, but there is another, related problem Johnston doesn’t explicitly mention but more or less illustrates in his essay. Namely, with so much data streaming past our eyes, the dangers from the cognitive biases are surely heightened. As Daniel Kahneman writes in Thinking, Fast and Slow, the human mind is “a machine for jumping to conclusions.” Presumably the more information analysts must sift through, the more they’re likely to fall victim to confirmation biases, the halo effect, the availability heuristic, and other mental shortcuts that can lead to erroneous conclusions. Human beings did not need to invent the Internet to struggle with these biases — and one can easily argue that policymakers have always fallen prey to them — but the Internet is uniquely suited to encourage Kahneman’s “fast thinking” (intuitive thinking in the face of uncertainty).

Johnston’s article, then, is a note of caution to be sensitive to how foreign-policy narratives form today, a warning to analysts to not take shortcuts but instead to use careful scientific reasoning before reaching conclusions and, in the case of China, to be sensitive to history, to avoid making inferences from a small sample size, and to be clear about what they think is driving China’s behavior. There probably isn’t much that can be done about the state of foreign-policy discourse today, but one can hope that the high-speed, high-volume discourse is at least more amenable to self-correction than the older discourse.

(For other comments on this essay see Daniel Drezner and Graham Webster. Drezner’s comment that Johnston falls victim to ahistoricism himself is probably right — I would have liked to see a bit more discussion of how conventional wisdom formed in US foreign policy in the past, beyond a handful of references to media studies articles on the media’s role as agenda setter.)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s