After Hegemony, II: Complex Interdependence
Remembering Joe Nye the CI Guy by using his ideas, not just mentioning that he had some.
Joseph Nye recently passed. Drezner hits the high notes. I did not interact with Joe often, but my first interaction with him (at Drezner’s invite) was my first academic roundtable. Memorable (for me).
In his nice remembrance at the Duck of Minerva, Peter Henne encourages us to remember Nye’s work beyond his conceptualization of “soft power”,1 because the intellectual foundations of that thinking ran deeper:
A few decades before his work on soft power [Ed.: i.e., during the Nixon years, as Bretton Woods was collapsing under its own weight], Nye wrote Power and Interdependence with Robert Keohane. This book, and accompanying articles, questioned the realist paradigm that then dominated the study of international relations, arguing that growing complex interdependence among states was transforming their behavior. This led to the emergence of multiple channels connecting states–not all of them official–a lessened hierarchy of issues for policymakers, and decrease in the utility of military force.
Power and Interdependence also insisted that non-state actors (businesses, churches, advocacy networks, nongovernmental organizations) were increasingly relevant, intertwined through a web of mutual obligations, identities, and associations that often stretched across borders. I’m sure you can think of many examples (and it’s worth pausing to do so, to anchor your mind for what follows).
Keohane and Nye positioned “complex interdependence” on the opposite end of an ideal type spectrum from neorealist “anarchy”. Meaning: it’s normatively desirable in the same way that anarchy is normatively regrettable.2 Under anarchy, the world is ordered through the distribution of power: the strong do what they will, the weak suffer what they must. Under complex interdependence, the world is ordered through connectivity: shocks to an individual are absorbed by the collective.
Exploring complex interdependencies to uncover ways to tame power is the priority of this blog right now.
My previous post focused on a Keohanian implementation of the complex interdependence framework, the “regime complex” that governs international economic exchange in the absence of a world government. I introduced the idea that the “regime complex” provides infrastructure for resisting 47’s imperialism and attempts to re-organized the global economy into a personalistic tributary system. I’ll build on that framework, empirically, in the next post.
For now, to honor Nye, I want to dwell on this idea of Henne’s (emph added):
Power and Interdependence helped inspire neoliberalism in international relations [Ed.: probably not the “neoliberalism” you’re thinking of, exactly], but neoliberalism kind of left its rich foundations behind. … Policy-wise, we aren’t living in the utopia promised by early liberal scholars, but part of the reasons current conflicts are so concerning and complicated is that they are occurring in the context of complex interdependence among states. Reviving their insights and applying them to today’s events could be of great use.
A small group of political economy scholars has been working on that intellectual project for the past decade, a.k.a. the “new interdependence approach”. In response to the subprime crisis, Brexit, and Trump, I wrote a historiography referring to “complex interdependence” as the “first notable idea” of international political economy, wondering why it wasn’t being studied more:
Globalization is a concept of interdependence, yet the primary methods used to analyze it render the complex as simple by assumptions that are too seldom examined, much less justified.
It was a claim that our ignorance of complexity — in the literal sense: we ignored it — left us blind, analytically. I believe we are in agreement about that now.
Around the same time, Thomas Oatley published “Toward a Political Economy of Complex Interdependence”, a framework for CI analysis with general application. The tl;dr? Move away from Newtonian partial equilibrium models that assume unit-level independence and well-behaved statistical distributions; move towards evolutionary models that take as fact multi-level interdependence and asshole statistical distributions. Shortly thereafter, the “Weaponized Interdependence” article made waves (and possibly inadvertently inspired some new manifestations of imperialism). Then I published an article linking complex interdependence to the emergence and persistence of structural power in world politics.
So there are some ideas out there. This work (and others in its vein) emphasized that the global political economy was not solely the realm of states, the number of salient actors and issues was large and growing; that economies were no longer national but multinational (so concepts like GDP have less and less descriptive utility); that identities were becoming more multifaceted, less categorical; that macroeconomic data was becoming less reliable for inferential use. It also emphasized that economic models based on representative agents and narrowly-conceptualized rational maximization were unlikely to be useful predictors of outcomes, that political coalitions were shifting (as were value chains, perhaps related), and that global networks were “filling out” or “thickening”, reducing the effectiveness of traditional governance institutions.
These niches have been largely ignored by the core of American IR, and American Econ too.3 The above papers were not published in the premier outlets in the field, despite having a very well-established intellectual heritage (and subsequently receiving large citation counts); instead, they were published in European outlets, special issues, and/or edited volumes, places more open to expansive thinking than the leading American journals.4
Why does this matter? The people who are supposed to be leading us now were educated by the same people who neglected complex interdependence for decades, positing in its place a world that was understandable through simple models and experimental research designs. So they don’t know what to do. They have not intellectually prepared for anything like the moment we are in.
Have you noticed that the opposition to 47 is disorganized? Fragmented? Does not have a unifying message? The Democratic party, the business community, the academic community, the think tank community… all are “stunned” or “silent” or “demoralized” at this moment? World leaders looking at each other, waiting for someone else to do something first? There are individual voices pursuing individual tactics but these are not winning because they are premised on simple (e.g., class-based) coalitional understandings that are out of date.
No one is sure what to do now because they were wrong. Not wrong normatively, like the “populist” thieves claim. Wrong analytically. Frustratingly, in a world of complexity it’s really hard to be right for very long.
Democratic institutions were supposed to create (or at least enshrine) prosperity, prosperity was supposed to generate peace, peace and prosperity would be protected by overlapping and self-reinforcing “governance” institutions, and this entire arrangement was supposed to be a stable equilibrium with a gently upwards slope.5
This was intellectually appealing to elites (IMO) because it subdivided complex interdependence into governable subsets (“issue-areas”), which could be treated as distinct and thus solvable with domain-specific knowledge. This knowledge could be delivered through elite educational systems and administered via bureaucratic systems with built-in accountability mechanisms that provided some semblance of meritocracy. As long as you assume that all knowledge relevant to that domain is contained within that domain — i.e., that complexity does not describe the world — then the equilibrium should be stable. And so, like economists with can-openers, that is what they did assume.
Because complexity wasn’t being considered it wasn’t being managed. Voting publics began attacking complexity — from the reactionary right and radical left — because they didn’t understand it and how much of it was too their advantage. The semblance of meritocracy eroded. They didn’t understand complexity because they were told they didn’t need to understand complexity, and they were told that because elites believed it didn’t matter (and/or would be solved by “markets”, which is the same thing as believing it doesn’t matter).
It does matter. To understand why we don’t need to re-invent the wheel. Many of the concepts that we need have already been imagined. There is intellectual infrastructure available to us.
Trump backed down on trade, because he didn’t have enough leverage or power to destroy the system of complex interdependence. How can you divide-and-conquer complexity? You cannot.
He’s not done trying. He will do more damage.
Keohane and Nye gave us an enormous amount we can use to understand how, why, and what to do about it, even though that was perhaps not their intention in writing. In their original “Power and Interdependence” article they wrote:
We are now in a period in which other states still look to the United States for leadership, because of the inchoate state of policy coordination in the European Community and the lack of self-confidence on the part of the Japanese, who have moved to the fore-front as an economic power so recently. This American leadership is currently directed toward changing the present structure of monetary and trade relations.
Other states no longer look to the US for leadership. Policy coordination in Europe is much greater. Japan is more self-confident, and there are many more relevant players now besides that group. Against all that, and more, this American “leadership” is currently directed toward changing the present structure of monetary and trade relations. They want to make things more simple, less complex, fewer interdependences, more independence. They’re calling it the “Mar-A-Lago Accord”.
I increasingly think the system has overgrown its ability to be “led”, and certainly it has overgrown its ability to be controlled. Many will view this as bad. I will attempt to argue that it is good. A difficult step, but necessary for human society to reach its fullest potential. Free people, fully-grown people, should not wish to be led, they choose their own associations and through that embed themselves into overlapping systems of interdependence that provide reciprocal protection and mutual benefit.6
Keohane and Nye gave us a lot to work with, but we can’t just nod to it on the shelf. We need to actually use the infrastructure, adapt and apply it to our times and situations, reinforce webs of interdependence that are durable against attacks from barbarians.
Getting to that outcome is the project of civilization, and it’s our project too. During his life Joe Nye built as much of that infrastructure as he could, from inside and outside the government. Let that be an example to us of what is possible. We haven’t had to do this for awhile, we’ve been coasting, it’s no surprise that these muscles are out of shape. Large-scale social transitions are difficult, especially when they occur in the context of rapid technological change. We will see if we’ve learned enough lessons from the past to avoid some of its calamities.
Next post will be empirical and pragmatic: how do we cut the US out of the trading system (so they can no longer abuse it)? Eventually we’ll get to finance, and other “issue-areas”.
The cynical could describe “soft power” as Gramscian hegemonic analysis watered down for normative liberals. Not that there’s anything wrong with that.
This is why “neoliberals” (in the IR sense) are sometimes called “idealists”.
Like Henne, I was discouraged from reading about complex interdependence in graduate school by senior IR profs. Keohane was in the audience at APSA as several of the papers linked above were first presented… at one point he exclaimed “it’s like being back in the ‘70s!”
Which emphasize theoretical parsimony and assembly-line production of formulaic research designs, and thus reduced-form models containing strong assumptions of independence that did not match the process through which the data those models were tested on was generated. For many, concepts like “interdependence” brought to mind Cold War structuralisms that were very much out of vogue during the behavioral turn in the “end of history” period: remember Freakonomics? Very individualistic, very microeconomic, very tractable, lent itself to “gold standard” experimental research designs in which “complexity” is strictly controlled away and well-designed markets can fix all problems. (How’d that intellectual project end up again? Oh, right. Right. Right.)
We’re building to discussion of the “Kantian tripod”. There are some, um, reasons to be concerned. But also clues for avoiding the worst outcomes.
Thatcher had it 100% backwards: there is an observable society, there are no observably-independent people.