Introduction
In the dominant framework of international politics, state behavior is understood in strictly instrumental terms: governments survive and prosper by directing national agencies to pursue objectives aligned with state interests. This basic premise has been refined in various ways. Constructivist work has emphasized that interests are intersubjectively constituted rather than given. Cultural approaches have shown how national traditions constrain the range of acceptable means. Experimental research has documented how biases, misperception under incomplete information, and plain error distort the link between chosen means and intended ends. Yet a broad consensus remains over a premise that seems all but self-evident: states are most effective when they act strategically, actively pursuing the objectives that serve their interests most directly.
By contrast, a recurrent insight from other fields of study, in particular meta-ethics, holds that there exist ends that are best achieved when they are not directly pursued. Referred to variously as ‘essential byproducts,’Footnote 1 ‘self-effacing ends,’Footnote 2 the ‘paradox of teleology,’Footnote 3 or the ‘paradox of intention,’Footnote 4 these are goals that are best attained by striving after some other, orthogonal objective. The unsettling implication across these various treatments is that a strict instrumental approach would not be strictly instrumental.
While the notion of self-effacing ends has thus been the object of considerable scholarly study, its implications have not been examined for international relations. In this article, I propose that there is much to be gained from doing so. Of course, not all state objectives are self-effacing: yet the conditions that make them so, I argue, happen to be present under the type of interdependence that characterizes the contemporary global economy. At a time of renewed government intervention in global markets, where governments seek to weaponize economic relations, there are limits to what states can achieve through such deliberate moves. Insofar as we do see states wielding effective leverage over others, this is often as a ‘welcome byproduct’ – to borrow a phrase from Albert Hirschman – of past actions pursued for other ends, rather than the result of farsighted policymaking.
The reason why state objectives take on a self-effacing quality under interdependence is straightforward. The arrangements that generate influence – financial systems, trade routes, shared standards and platforms – depend on others voluntarily opting in at the outset. If these prospective participants anticipate that joining today may expose them to coercion or disadvantage tomorrow, they have reason to abstain, even at some immediate cost. The more visible the bid for future leverage, the weaker the incentive to sign on. Plainly strategic designs thus provoke hedging, duplication, and resistance.
Given this requirement for others to willingly participate, we can ask how it is that states nonetheless achieve positions that grant them leverage over those others. As I argue, international arrangements built around ostensibly nonpolitical aims – be it democracy promotion, free trade ideology, or cultural exchange – are more likely to attract broad participation and, over time, generate imbalances that translate into later influence. For analogous reasons, state interests are often best advanced not by direct state intervention, but through the activities of nonstate actors whose goals appear orthogonal to state interests. A third factor that allows for gains to accrue to some states in spite of the self-effacing nature of state objectives is the sheer complexity of the underlying arrangements, which makes their long-term implications opaque to participants themselves, let alone their home governments.
These conditions imply that when states do end up with leverage over other states, it is often as the byproduct of processes no one deliberately designed with that leverage in mind. These gains turn out to be durable precisely because they were not anticipated. I develop this claim by examining two domains of the contemporary international economy: financial infrastructures and global supply chains. There, the pursuit of efficiency by private actors inadvertently created asymmetries in state relations, offering some states considerable leverage over others. This was not by dint of farsighted policymaking, but precisely due to its absence.
The much-discussed weaponization of interdependence thus relies on initial government myopia, in ways that the current calls to weaponize choke points continuously overlook.Footnote 5 In this way, while Great Britain leveraged its position at the center of global financial networks to great effect during World War I, the British Admiralty was actually late in recognizing its unsuspected advantage – one not of its own making, but instead the result of the independent actions of bankers in the City of London.Footnote 6 A century later, the United States is leveraging its position by deploying economic sanctions for geopolitical purposes with increasing frequency.Footnote 7 This coercive ability flows from global financial plumbing that took shape in the 1970s, when private banks created cross-border dollar payment systems, which later gave U.S. authorities unexpected leverage. Tellingly, these private actors were actively seeking to emancipate themselves from state oversight, rather than to contribute to its power.Footnote 8 The coercive potential of dollar-denominated financial networks came as a surprise even to the agencies that eventually weaponized them. The corollary of this surprise is visible in how U.S. partners, including allies, only then began trying – largely unsuccessfully, given the lock-in effects created by early participation – to circumvent this new source of influence.Footnote 9 Time and again, geopolitical advantage is the doing of accidental architects.
As in the financial realm, states have been increasingly leveraging their positions within supply chains by restricting key exports in pursuit of their geopolitical ends.Footnote 10 Yet states have had little input in designing supply chains, and they remain remarkably oblivious to their position within them. Global supply chains emerge from the dealings of a vast number of private firms searching for the least-cost provider for the countless tasks along their production line. Even large multinational corporations – much less would-be governmental regulators – are often unable to trace the full production line of the finished goods they sell.Footnote 11 The proliferation of ‘early warning systems’ across developed economies is a testament to such state myopia. Indeed, in a remarkable turn, governments are now pleading with their own domestic firms for information to map the structure of the global supply chains they are a part of.Footnote 12 Whether it is Chinese dominance of the production of rare earth elements – which were initially low-cost industrial commodities used in menial applications such as glass polishing and ceramics – or global dependence on a handful of countries producing PPE during COVID-19, the relations of dependence between states that result from global supply chains are only rarely the result of state action. Most often, they emerge precisely as a result of state myopia, as governments remain unaware of the benefits and vulnerabilities that result from their firms’ actions.
In this way, states’ dominant positions within trade and financial networks from the 19th to the 21st century were not the product of farsighted policymakers. They resulted from the dealings of nonpolitical actors pursuing nonpolitical interests. In fact, the implication of the argument is that if policymakers had been sufficiently farsighted and had actively set out to achieve the dominant positions they ended up with, they would have had a lesser chance of success. The attainment of leverage over others under interdependence is a self-effacing objective.
The concept of self-effacing objectives also highlights the epistemic, as opposed to strictly material, aspect of economic dominance. Here is a source of leverage that evolves, from its initial emergence to its subsequent dissipation, through shifts in common knowledge. It grows most effectively in latent form, in a state of collective ignorance, unrecognized by bureaucracies and state agencies. It flourishes because both sides are, at first, inattentive to its strategic implications: the eventual beneficiary does not yet recognize the extent of its future leverage, and participating states do not perceive their exposure to it. This mutual opacity allows the underlying structures to develop without triggering nonparticipation, hedging, or duplication. The dynamic shifts only once participants recognize that continued involvement has rendered them vulnerable. At that point, efforts to circumvent, fragment, or replicate existing international arrangements begin, and the advantage is progressively eroded. In this way, states that ultimately benefit may paradoxically gain most when they, too, are initially unaware of the strategic implications of the institutions they are part of.
A theory of self-effacing objectives based on bureaucratic myopia and unforeseen effects might be thought to imply that all statecraft is futile. This need not be the case. As I show in the final section, there remains considerable room for strategic behavior. Happenstance can be designed; states can increase the odds of ‘welcome byproducts’ – although an unavoidable trade-off is involved: if states are to attain self-effacing objectives, they must cede some measure of control. I demonstrate this through a third illustrative case, looking at U.S. foreign information operations in the early period of the Cold War. Indeed, persuasion is a paradigmatic self-effacing goal. To be effective, the would-be persuader should not overtly aim at persuasion, but at some ostensibly unrelated objective, a byproduct of which may be increased persuasion. This insight provided the key to an under-examined overhaul of U.S. info ops from the Truman to the Eisenhower administration. As President Eisenhower put it in 1953, ‘[to be effective] the hand of government must be carefully concealed, and, in some cases, I should say, wholly eliminated.’Footnote 13 I unpack the theoretical reasoning behind this intuition, which reveals the central tradeoff that runs through the article: the more targeted statecraft is, the more likely the counter-response, and the weaker its impact.
A theory of self-effacing objectives in IR
Thinkers from different disciplines have independently proposed the existence of ends that cannot be attained directly, but only as byproducts of actions taken in pursuit of other ends. In political science, this belief finds its clearest articulation in the work of Jon Elster, who uses it to argue against a common functionalist fallacy, whereby an observer concludes that the incidental benefits of some institutional arrangement were the motive behind the institution’s creation. Instead, Elster argues, such benefits are often ‘essential byproducts’ of the institution: they were not intended in its design, though the institution may be retained once these unintended benefits are observed.
I want to argue that the same may apply to international institutional arrangements that contribute to state objectives. This becomes especially likely in settings characterized by interdependence. I go a step further: it is often the case that such byproducts could not have been achieved if they had been the designers’ original intent. If so, then owing to a process of selection, we should expect that when we do observe structural advantages offering one state leverage over others, it should often be the result of happenstance, rather than design. That is, under interdependence, there should be a systemic bias against functionalist outcomes, and in favor of ‘welcome byproducts.’
Parfit offered an early articulation of the concept of self-effacing ends, and it is in the field of moral philosophy that this notion has been most thoroughly worked out.Footnote 14 He referred to ‘self-effacing’ theories,Footnote 15 offering thought experiments where a strict pursuit of self-interest appeared to run counter to self-interest. In these thought experiments, an actor’s perceived intent affects the response of others in ways that might make a self-interested actor worse off. It would then, paradoxically, be better for that actor to modify their own incentive structure to keep their promises, rather than to extract maximum advantage in every period. Becoming less of an egoist would make them better off, even from the strict egoist’s standpoint. Parfit’s work led to a (still-ongoing) debate in meta-ethics that examines the implications of self-effacing ends for act consequentialism.Footnote 16 Strict consequentialism may thus require departing from a strictly consequentialist motivation: only then may the outcomes judged best under a consequentialist approach be achievable. The result advocates for selecting goals other than the final goal as a means of attaining that final goal.Footnote 17
Though these insights emerge from the study of ethics, the link to international relations is quickly made. Parfit’s thought experiments share much in common with treatments of time-inconsistency in political economy, where actors have an incentive to act in an earlier period that they no longer have in a later period. A state actor may ‘intend’ to keep a promise not to nationalize a foreign investor’s assets, but once that investment is made, it will nonetheless have a dominant incentive to nationalize the assets. Foreseeing this problem, states have reason to tie their hands against their future self-interested moves.Footnote 18 What renders objectives self-effacing across these different treatments is the interaction between two or more agents: an awareness of one’s intent leads others to respond in a way that stifles the realization of that intent.Footnote 19
To be clear, arguing for the existence of self-effacing objectives does not deny the prevalence of rational behavior – it presumes it. As in Parfit’s thought experiments, it is precisely because all actors are rationally pursuing their interests, and their intentions are manifest, that the direct, instrumental pursuit of self-interest is self-defeating. An analogy can be drawn to perfectly efficient markets, where no actor can achieve abnormal profits because any attempt to do so is swiftly negated by competitors. Yet if market actors suddenly ceased trying to outmaneuver one another, the market would also cease being efficient, and opportunities for abnormal gains would reemerge.Footnote 20 As long as all actors tend to their interest, however, beating the market is mostly a product of chance. I argue that these insights also apply to interstate relations under specific settings.
Parallels in international relations theory
While there is no explicit discussion of self-effacing ends in international relations, several established IR theories have similarly highlighted the importance of unforeseen effects and state myopia. Though they did not make the self-effacing aspect of state goals explicit, these theories offer various insights that are useful in building the argument. I thus begin by highlighting some of the similarities and differences of this account with existing approaches.
First, a number of IR scholars have emphasized the role of chance in strategic interactions and its implications for the feasibility of strategy itself. The intuition is not new. Recall Napoleon’s (likely apocryphal) declaration, ‘I like my generals lucky.’ Yet Betts formalized it by drawing on findings about bounded rationality to critique what he termed the ‘illusion of strategy’ in international security.Footnote 21 He argued that the empirical record suggests that intended means and realized ends align no more frequently than dictated by chance: ‘Strategy is an illusion because results do not follow plans. Complexity and contingency preclude controlling causes well enough to produce desired effects. Hindsight reveals little connection between the design and denouement of strategies.’Footnote 22 The result is a type of ‘strategic nihilism’ where the inability to predict the outcome of a given move with sufficient certainty means that any pretense at strategy is illusory.Footnote 23
Betts grounds his account in the limits of human knowledge and information processing. Decision-makers fool themselves if they cling to the pretense that they can know enough about all the intervening factors that will invariably disrupt the best laid plans. My premise is the opposite: if one country can anticipate that a new institutional or commercial arrangement will leave it with leverage over others, those others can anticipate the same. And if they expect that participation will render them vulnerable, they are unlikely to consent to the arrangement in the first place.
Of course, if strategy were impossible because of the inability to see ahead in complex settings, as per Betts, then any relative gains would also be the product of chance. Both Betts’s argument and the one presented here thus challenge the traditional view that states attain their objectives through strategic calculation, yet the reasoning differs markedly.
Second, self-effacing objectives bear relation to the IR literature on the ‘third face of power,’Footnote 24 according to which the most ‘insidious’ form of coercion is not perceived by the coerced as coercion. The link to self-effacing objectives is quickly made: power is most effective precisely when it does not present itself as power. An important ambiguity in this literature exists over whether such power must also be unrecognized by those who benefit from it. Lukes, following Bachrach and Baratz,Footnote 25 leaves open the question of whether dominant actors consciously design and sustain the structures that entrench their advantage, or whether these structures persist through routine behavior that merely reproduces asymmetries without deliberate intent.Footnote 26
These two possibilities – unawareness by the dominated, and unawareness by the dominant – are analytically distinct. In the first case, power is effective because it is misrecognized by its subjects; in the second, it is effective because it is not pursued as power by its beneficiaries. The latter possibility raises profound normative questions about domination, responsibility, and complicity. Since the argument is more narrowly concerned with the conditions under which structural advantage is most likely to endure, I do not attempt to resolve those normative questions. Instead, the insight I take from Lukes is that the more an arrangement appears noninstrumental, the less likely it is to provoke countermeasures. This holds whether the dominant actor is genuinely unaware of its advantage, strategically inattentive to it, or simply refrains from foregrounding it. As Dowding puts it in his discussion of this question, ‘all that is really required is that a person continues a pattern of behavior that brings about desired results.’Footnote 27 Awareness of the benefits of the status quo may be partial or retrospective; what matters for present purposes is that overt bids to consolidate advantage are more likely to undermine the structures that generate it.
The notion of self-effacing ends also bears resemblance to a third foundational concept in IR: the security dilemma. As set out by Jervis, a state that actively seeks its own security prompts neighboring states to respond in kind, reducing the original state’s security, and precipitating an escalation of moves that eventually prove socially self-defeating.Footnote 28 In later work, Jervis explicitly alluded to the futility of states’ attempts to acquire a relative advantage in describing the security dilemma: ‘if a state tries to maximize its power by increasing the resources that would enable it to menace others if everything else remained constant, the others are likely to seek to neutralize or counterbalance the state’s efforts with the result that its power will not, in fact, be maximized.’Footnote 29 Jervis also emphasized the role of surprise, a point to which I return below: ‘As actors seek advantage and try to outstrategize one another, some of them – if not all – must be surprised.’ It is worth insisting on the parenthetical: since there is little reason to assume persistent informational asymmetries,Footnote 30 the target is unlikely to be surprised unless the initiator is as well. To reformulate using the terms of the argument, under the security dilemma, states appear more likely to achieve security by aiming at some other, unrelated end in ways that progressively reassure other states that their own security is not compromised. Yet even here, if reassurance is too plainly the objective, it is unlikely to prove reassuring.
The security dilemma thus illustrates how the anticipation of future advantage can undermine present cooperation. The same logic sits at the heart of the relative gains debate that marked IR theory in the late 1980s and early 1990s. At the time, scholars argued that even mutually beneficial cooperation may fail if states are concerned about how gains are distributed, since advantages accruing to one state can translate into future leverage.Footnote 31 Others then responded that institutions, issue linkage, and repeated interaction could dampen distributional conflict and render absolute gains decisive.Footnote 32 The present argument shares with this literature an emphasis on anticipation, but shifts the focus from how distribution shapes cooperation to how durable asymmetries arise. Whereas the relative gains debate asks whether anticipated distributional consequences inhibit cooperation, the logic of self-effacing objectives identifies conditions under which relative gains accumulate precisely because they are not pursued as such.
In sum, IR lacks an explicit treatment of self-effacing ends, a concept that other fields have considered in depth. Yet several related insights nonetheless prove relevant to such a theory: (i) unforeseeable ‘system effects’ undermine strategic behavior; (ii) the most ‘insidious’ form of power is one that the affected parties do not recognize as an exercise of power; and (iii) security itself may be a goal that becomes more elusive as states pursue it, because of other states’ strategic responses.
Self-effacing ends and ‘catalytic’ ends
A self-effacing end is one that is best attained by aiming at some other, orthogonal end. When such orthogonal ends do yield positive byproducts, we can call them ‘catalytic.’ Catalytic ends are those that inadvertently yield structural benefits that could not have been attained had a state’s policymakers aimed at them directly. It follows that these ends will only appear catalytic in hindsight. Yet as I suggest below in discussing implications for statecraft, effective statecraft may largely consist of instinctively discerning which state objectives are likely to prove catalytic. We can distinguish two types of catalytic objectives that meet the requirements of the theory. These are (i) principled, nonutilitarian commitments held by states themselves, and (ii) objectives pursued by nonstate actors, with the tacit support of their home states.
(i) Principled state commitments
In the first instance, a catalytic objective can be a noninstrumental commitment held by states themselves. Governments routinely pursue goals that are articulated in value-driven terms and justified on principled grounds: enhancing transparency, expanding the freedom of trade or information, fostering cultural exchange, or promoting democracy abroad. What distinguishes such commitments is that they are publicly framed – and often internally experienced – as ends in themselves rather than as tactical moves in a competition for relative advantage. Precisely because they are not readily interpreted as bids for dominance, they can attract participation, emulation, or trust from others who might resist more overtly strategic designs. In this way, they may advance state interests, but obliquely. These ostensibly principled objectives may later prove to have rested on mistaken premises, yet this does not diminish their capacity to generate beneficial byproducts. It is thus not uncommon for the unforeseen effects to prove more beneficial than their initially proclaimed intent. The main criterion for a catalytic goal, in this respect, is that it has a credible claim to intrinsic rather than instrumental value.
For instance, the United States has long undertaken a global agenda around democracy promotion, human rights protection, anti-corruption initiatives, and free trade. Generations of American bureaucrats have committed themselves to the pursuit of these principled objectives. Dedicated agencies have been created, and funding has been allocated in proportion to how various initiatives help advance these values. These state objectives shape political rhetoric; they are included in foreign policy documents and national security strategies; some even become part of school curricula.
For a state goal to be catalytic in the sense described above, it must be orthogonal to the pursuit of a relative power advantage and appear to observers as an end that would credibly motivate state behavior.Footnote 33 That is, it must be sufficiently strongly held to have a convincing claim to intrinsic value. In this way, U.S. democracy promotion, which reached its peak following the terrorist attacks of 9/11, may have been either a principled attempt by a democracy to spread its core values for their own sake or a calculated strategic move intended to further U.S. geopolitical interests. Which one of these is the better explanation is an interesting empirical question. Yet it does not bear directly on the theoretical claim that interests us here, which is that the success or failure of democracy promotion, in instrumental terms, should depend on whether it was sufficiently accepted as inherently valuable from the standpoint of the U.S. state apparatus. The stronger this perception was held, the more democracy promotion had a chance of proving catalytic.
There remains a recurrent tension between the promotion of principled beliefs by states and the pursuit of strategic interests. Scholars who subscribe to a realist viewpoint will maintain that states only ever pursue such nominal objectives for self-interested ends, just as they only subscribe to norms of behavior that further their material interests. Yet strategic benefits arise precisely because states do occasionally uphold values in a manner that appears genuinely noninstrumental. Indeed, the study of the many ways by which states take and defend values through socialization, internalization, and mimesis suggests that states often do hold values for their own sake, even if this comes at a cost to their immediate interests.Footnote 34 Principled beliefs may turn out to yield material gains after the fact – yet these gains need not have been the initial reason for the adoption of those beliefs.
There may be no better illustration of this tension between principled versus instrumental reasons than E. H. Carr’s treatment of universal free trade in one of the foundational texts of IR. Carr famously portrays free trade as a ‘utopian,’ ‘wishful vision,’ initially elaborated by French physiocrats and British political economists in the 18th century, then ‘sincerely’ adopted by British statesmen of the likes of Richard Cobden, and only later ‘discovered’ by British elites to provide a relative advantage to hegemonic Britain vis-à-vis the rest of the world.Footnote 35 Carr bemoans the utopian quality of free trade ideology, yet he fails to appreciate how this very quality is what made the notion of universal free trade digestible to foreign elites.Footnote 36 According to Carr, the realist is ‘obliged’ to uncover the hollowness of such convictions, based as it was ‘on a negation of existing reality and on certain artificial and unverified generalizations.’Footnote 37
I share Carr’s reading of the sincerity of 19th-century British elites in their conviction that free trade served the interests of ‘the world as a whole.’ Yet where Carr treats this utopianism as naivety later exposed by his realist lens, I view it instead as the condition of the widespread voluntary buy-in that placed Britain at the core of the liberal trading order. It is precisely because laissez-faire was articulated as a universal principle rather than as an instrument of British advantage that it elicited voluntary emulation rather than defensive resistance. Had it been advanced as a strategy for entrenching British primacy, it would have invited countermeasures and likely failed. If free trade functioned as a ploy, it was one whose success depended on being believed – even by its proponents. Again, the argument here does not rest on whether free trade in fact benefited others at the turn of the 20th century. What matters is that it was embraced, domestically and abroad, as an intrinsically valuable ideal rather than as a geopolitical device. That very sincerity – Carr’s ‘utopianism’ – is what allowed it to become catalytic from Britain’s standpoint.
(ii) The political utility of nonpolitical actors
Britain’s embrace of free trade illustrates how a state’s noninstrumental commitments can, under the right conditions, generate structural advantage precisely because they do not present themselves as strategy. Principled state ideals are one source of objectives that may later prove catalytic. A second source of catalytic objectives lies in the activities of nonpolitical actors whose aims appear unrelated to geopolitical design. I now turn to the political utility of nonpolitical actors, whose autonomy makes voluntary buy-in by other states more likely.
Private firms, religious organizations, and activist NGOs all pursue ends of their own – profit, salvation, reform – with little regard for, and often in direct tension with, state interests. Yet herein lies the paradox: precisely because they are not perceived as instruments of political power, their activities can entrench structures that later serve political ends. When this occurs, the objectives they originally pursued come to appear catalytic from the standpoint of states.
Consider the frequent observation that European merchants in the age of exploration, from the 17th century onwards, acted as unwitting handmaidens to the crown’s interests. Thus, the British East India Company and its Dutch analogue charted the path for subsequent colonial takeover better than royal missions of conquest could have, precisely because commercial profit through trade, rather than conquest, was their proximate goal.Footnote 38 In the New World, the trading outposts and routes initially set up by fur traders and private companies like Hudson’s Bay later became the basis for colonial claims.Footnote 39 Across these cases, sovereign conquest was a byproduct of merchants’ commercial dealings.
As in the discussion of the third face of power above, how aware sovereigns were of this dynamic remains an open question. Early chartered trading companies were granted extensive powers and privileges by their sovereign sponsors, likely with some understanding that their commercial ventures could pave the way for the projection of state power into new regions.Footnote 40 The financing of commercial expeditions and the hybrid nature of state trading companies suggest that the Crown was tacitly aware of the benefits of delegating to actors driven by private profits, and that it saw their objectives as sufficiently aligned to sponsor these expeditions. Yet the fact remains that the agents on the ground were representatives of the trading company, not its sponsoring government.
Just as merchants pursued profit without intending to further state power, missionaries’ activities appear to have generated comparable unintended effects. They sought religious conversion and humanitarian reform, often with little concern for imperial strategy – and at times in explicit tension with it.Footnote 41 Yet the language and practices of the ‘civilizing mission’ helped normalize imperial hierarchies and legitimize narratives of progress brought to allegedly ‘backward’ societies.Footnote 42 Doing so, they rendered colonial presence more palatable to local populations. If so, missionaries’ primary objectives proved catalytic from the standpoint of states. Because they were not widely perceived as agents of geopolitical design, missionaries’ efforts could soften resistance and entrench influence in ways that overt political intervention could not. Their autonomy – real and perceived – was what made their unintended political effects possible.
By contrast, when global infrastructure is the creation of state-actors, the expectation is that it should be less likely, owing to a process of selection, to produce lopsided state benefits. Other states, wary of the threat of potential exploitation, should condition their entry on credible assurances that it will not eventually cost them. The go-to illustration in institutional theory is the postwar institutional system, which featured hegemonic restraint rather than domination, aimed at ‘reassuring’ partners.Footnote 43 The system’s ‘constitutional character’ has thus been associated with decreasing returns to power rather than an attempt to extract gains through exploitation. The implication of the theory is that if it had been otherwise, the arrangement would never have seen the broad buy-in from weaker states that underpinned its success. Indeed, for all the scholarly interest in the distributional effects of multilateral institutions, these appear relatively muted. To put it bluntly, the global dollar system benefits the United States more than its position within the International Monetary Fund (IMF), although as a state, it had far more direct influence in the design of the latter than the former.Footnote 44 Insofar as postwar institutions have advantaged their original designers, these effects have been nuanced rather than overt, achieved through veiled, often discursive means, rather than plain coercion.Footnote 45 One observable implication is that institutional proposals that are too plainly in the service of the initiating states are unlikely to be adopted; I return to such failed attempts below.
Self-effacing objectives under interdependence
Not all state goals are self-effacing. The paradox arises under specific conditions. State objectives take on a self-effacing quality when the accomplishment of the goal requires voluntary buy-in from others, which would be withheld if the strategic intent were obvious. These conditions are most fully present under interdependence, which explains why, when global trade and financial infrastructures do yield structural advantages, it is most often as an unintended byproduct. Where these features are absent – where goals can be achieved unilaterally because they do not require buy-in by others – direct strategic pursuit remains entirely feasible. This is the case for domestic investments in education, scientific research, or infrastructure, where gains accrue internally and do not hinge on foreign buy-in. Even in domains such as the build-up of military capability, where foreign countermeasures may follow, an advantage can be attained without securing others’ consent; success depends on accumulated capacity, rather than opacity or indirection.
The following discussion develops these scope conditions by showing how they operate in practice. The lens of network analysis proves useful for this purpose. Rather than treating power as reducible to material capabilities, network analysis conceptualizes it relationally, as a function of an actor’s position within the web of ties that bind it to others. It moves beyond aggregate measures of wealth or military strength and instead examines structural properties such as centrality, clustering, and vulnerability. It is these positional features that produce the form of leverage at the heart of the argument – advantages that emerge from structure, and that are often recognized as such only in hindsight.
I first trace how trade and financial relations generate dependence as an unintended byproduct of voluntary buy-in, then I examine how the technical complexity of these networks obscures their implications from policymakers; finally, I turn to the role of nonstate actors in shaping these networks.
The specter of trade dependence has recently become salient in both policy and scholarly circles.Footnote 46 It is often presented as a contemporary corrective to the sanguine belief that interdependence necessarily promotes peace and stability.Footnote 47 Yet the recognition that mutually beneficial transactions can have distributional effects by creating relations of dependence is a longstanding insight of trade theory. A recurrent aspect of these accounts is the central role played by state myopia: trade dependence emerges without either state being aware of it. It is not the product of far-sighted policymakers, but a byproduct of dealings between efficiency-driven commercial actors.
In an early treatment of trade dependence, Hirschman explicitly describes the relative gains that can result from nominally mutually beneficial trade relations as a ‘byproduct,’ rather than the result of design.Footnote 48 Referring to the British trade routes established by British merchants passing through Gibraltar, Suez, and Singapore, Hirschman called it a ‘welcome byproduct’ that enabled Britain to ‘cut off the trade of other countries passing through these points,’ which in turn ‘gave her considerable direct power over, and influence in, other countries, in that they were always exposed to the potential threat of a sudden stoppage of their trade at Britain’s will.’Footnote 49 In today’s parlance, Hirschman was describing British-controlled ‘choke points.’Footnote 50 With respect to the foreign influence that Germany was able to amass through the conduct of its own trade relations, Hirschman added: ‘foreign trade brings about a maximum degree of dependence for certain countries which is by no means always the result of conscious policy on the part of other countries.’ It was nonpolitical actors who constructed the world that political actors found themselves operating in, only rarely understanding its origins.Footnote 51
The implication is that the gain to Britain of its trade relations was in no way intended, and that although it was later greatly beneficial to Britain’s strategic interests, this effect was not initially foreseen. An analogous claim can be made about the global financial network over the same period. As Harold James put it in describing the market for bills of exchange that placed London’s credit markets at the center of the global financial world, ‘the networks are hub-dependent, but the hub needs to exist in a certain veiled discretion.’Footnote 52 In other words, becoming a central hub may turn out to be highly valuable, but this value can only be attained under a general lack of awareness. The more recent literature on weaponized interdependence repeatedly makes a similar point: ‘[a] recurring theme […] is that embedded networks existed for some time before actors comprehended how to weaponize them.’Footnote 53 Separately, Farrell and Newman note, ‘America’s underground empire wasn’t the product of any grand master plan. Instead, it emerged half by accident.’Footnote 54
This article’s central claim follows from this observation. Namely, it is precisely because these benefits are the unintended byproducts of actions taken for other ends, often by nonpolitical actors, that they contribute to state interests to the extent they do. That is, if Britain’s capture of choke points in the 19th-century trade network – or the United States’ influence over contemporary global financial networks – had been designed according to a strategic plan by these state actors, it is unlikely that they would have yielded the relative advantage they eventually did. Interdependence becomes weaponizable as a result of state myopia. Three aspects of interdependence account for how state objectives become self-effacing. Next, I outline these in turn, by using global trade and financial networks as my illustrative cases.
(i) Networks emerge through voluntary buy-in
The level of interdependence between states is measured by the military alliances, digital data flows, economic ties, or diplomatic relations that link states to one another.Footnote 55 One of the key concepts of network analysis is preferential attachment, whereby new entrants within a network form connections with entities, or nodes, that already have a high degree of connectivity.Footnote 56 These central nodes offer tangible benefits of access and efficiency. They connect entrants to the network’s core, and the lowest-cost paths across the network typically pass through central nodes. The result is sometimes described as a ‘rich get richer’ phenomenon, whereby entities that already have many connections accrue ever more connections over time, resulting in the scale-free aspect of most social networks. Most importantly, new entrants join voluntarily, and out of a self-interested bid for maximum reach at the lowest cost. Unlike traditional material power, which can be accumulated without the consent of rivals, infrastructural or network power thus depends on securing others’ participation before its strategic consequences are fully visible.
The necessity of buy-in from others even characterizes the relationship of a hegemonic power to the smaller states within its orbit. Under Pax Britannica, the hegemon provided a public good of financial coordination, which other states voluntarily joined in search of efficiency. As per James, ‘no country thought that it was losing by participating in the new international monetary order.’ In the wake of the 1871 Franco-Prussian War, a growing number of – though by no means all – European states joined the Gold Standard. Foreign merchants sought the predictability provided by a common financial medium and the low borrowing costs provided by the London credit market. By the end of the 19th century, over half of global seaborne commerce was transacted through London-based financial institutions, denominated in sterling and financed by British banks.Footnote 57 And although the Bank of England certainly played a stabilizing role that proved crucial to making London the global hub starting in the late 19th century, ‘at the beginning, it is not clear that the Bank of England knew what it was doing.’Footnote 58 A similar lack of awareness was widespread among those who would later become vulnerable to British leverage as a result of joining the British-led monetary regime: ‘members [of the gold standard system] did not, on the whole, think that they were laying themselves open to any particular pressure from the core country, the United Kingdom.’Footnote 59 In this light, the hegemon’s privileged position appears more fragile: as opposed to dominance by military budgets and the direct projection of power, this preponderance relies on continued buy-in from others. The rich only get richer if the nonrich dutifully continue to join under the rule of preferential attachment. If the implications of centrality can be foreseen, preferential attachment breaks down. Entrants grow more willing to incur an inefficiency cost; they tolerate redundancy in exchange for a lesser risk of future vulnerability.Footnote 60
A similar pattern emerged in the postwar period, but at the micro-level of payments infrastructure rather than at the macro-level of reserve currency arrangements. The dollar had already been established as the key reserve currency under Bretton Woods, and by the 1970s, its global role was secure. What was new was the emergence of private cross-border payment systems and clearing houses built by U.S. banks, which became the plumbing of global finance. At the time, joining these networks was a choice: firms and financial institutions were attracted by the efficiency and liquidity of dollar-denominated transactions, not by any expectation of geopolitical leverage. As Farrell and Newman (Reference Farrell and Newman2023) note, U.S. officials themselves paid surprisingly little attention to the system as it was being built. Yet the concentration of transactions in dollar-clearing channels eventually created the chokepoints that U.S. authorities set about exploiting 30 years later. Here too, the privileged position of the hegemon depended on continued voluntary buy-in: the architecture emerged because participants sought efficiency gains. State actors on both sides remained largely oblivious.
Most ties do not involve a hegemon, yet they nonetheless reflect some form of dependence. Whenever a given country specializes in the production of a good, a service, or a production task, its trade partners grow dependent on it. In a global supply chain economy, relations of dependence are thus the norm. These result from choices by firms that agree to delegate tasks to the producers that perform them most efficiently. This is even true of specialization that appears to result from immutable natural resource endowments. Rare earths are a case in point. While China holds large deposits, other states – including Canada, Brazil, Vietnam, Russia, and India – also possess significant reserves and are now racing to expand output, though facilities take years to develop.Footnote 61
Although China now exerts considerable leverage through its ability to restrict exports of gallium, germanium, and other rare earths, this edge owes little to strategic foresight. Until the early 2000s, rare earths were mainly used in low-tech applications such as glass polishing, ceramics, and metallurgy. Their sudden revealed importance for semiconductors, solar panels, and defense technologies reclassified them as ‘strategic,’ triggering government interventions to control supply.Footnote 62 The case of rare earths underscores the recurrent paradox. Once a good is labeled ‘strategic,’ only materially dominant states can seek to outspend rivals, as others seek to blunt their efforts through countermeasures and collective resistance.Footnote 63
Most of the time, trade dependence has no meaningful implication. Country A’s dependence on wooden toys from Country B does not affect geopolitics.Footnote 64 Yet this can quickly change in the wake of unexpected demand shocks. A salient example is the sudden rise in demand for cotton masks, face shields, and ventilators in the wake of the COVID-19 pandemic. When states became aware that these were produced in a handful of countries, the latter suddenly gained a source of leverage.Footnote 65 The case of COVID-19 also speaks to how difficult it can be to foresee vulnerabilities arising from interdependence ahead of time. If cotton masks can acquire the status of strategic goods and become a source of leverage overnight, what other goods might have similar potential? In all such cases, initially voluntary investments into existing institutions have run-on consequences that rejigger power relations between states – yet these states rarely have a say in the matter, because they have little awareness, and even less control, over the choices of nonstate actors.
(ii) Network complexity
The interstate networks that make up trade supply chains and global financial transactions are vast, highly complex, and of a technical nature that often makes them opaque to observers. Such network complexity can result in state myopia, which in turn makes the accumulation of latent structural advantage possible. Under bounded rationality assumptions, complexity makes oversight difficult. Even when states have well-defined preferences, and decision-makers act strategically to maximize their conception of the national interest given the constraints they face, they have limited computational resources, time, and information, which constrains their ability to make optimal decisions.Footnote 66 As a result, they rely on heuristics and strategically allocate their attentional resources in view of their limitations.
This aspect of contemporary trade and financial networks may seem improbable, given markets’ ability to process large amounts of data. Should markets not produce a reliable sense of every node’s position within commercial networks? The short answer is no. In the case of trade, network complexity stems from the pursuit of efficiency by the distinct actors along a supply chain, all seeking to minimize their costs at every node – from raw material suppliers to retailers – which creates a dense web of interdependencies. A supplier finds a lower-cost sub-supplier, who farms out various tasks of the production process to further suppliers, and so on down the chain. This is compounded by just-in-time manufacturing, extensive outsourcing across borders, and competitive pressures to rapidly adapt to changing consumer demands. The multitude of roles played by various actors along the chain, including logistics and IT support that run parallel to the main production and distribution lines, further contributes to complexity. Ultimately, the extensive, multi-tiered structure of supply chains, combined with numerous intermediaries and a general lack of transparency, makes it exceedingly difficult for MNCs to trace the provenance of their own inputs, even in the face of growing demand for supply chain accountability.Footnote 67 It becomes all but impossible for any node to have a full view of the network.
The management of networks like those created by global supply chains often adopts a modular approach by breaking the system into discrete, manageable units. Modularity contributes to further opacity; ‘[the pursuit of] transparency [is] impractical, time-consuming, and distracting.’Footnote 68 This practice is reflected in the supply chain management software used by MNCs, which is split into several products, called ‘components,’ which are further split into function-specific ‘transactions.’Footnote 69 Supply chain professionals will specialize in a single one of these components, and ‘black box’ the rest. Some scholars have suggested that firms have an incentive in perpetuating the network’s illegibility, and ‘strategically obscur[ing] knowledge,’ so as to exploit lower labor, environmental, and governance standards.Footnote 70 An analogous incentive may exist in financial networks, where opacity allows financial agents to engage in price discrimination and preferential access.Footnote 71 When entrants choose what nodes to link to, they may themselves remain unaware of their second-, third-, and nth-degree ties, and the home-state dependencies that result.
The implications of network complexity become apparent when states have a sudden reason to determine their actual position in a given network, revealing how they had been in the dark all along. It often takes large-scale disruptions, such as a natural disaster, an inter-state conflict, or a pandemic, for firms and their home states to invest in ‘network mapping,’ which even in the best of cases remains ‘resource intensive and difficult.’Footnote 72 The COVID-19 pandemic made governments newly aware of their dependency on a handful of foreign providers for essential goods. Yet to this day, according to supply chain experts, most North American MNCs are unaware of how a disruption in Asia would affect their operations.Footnote 73 This is even true for high-value-added, geopolitically strategic industries such as semiconductors. Following the Tohoku earthquake and tsunami that hit Japan in March 2011, which disrupted supply chains across the globe, it took a team of 100 executives at a Japanese semiconductor manufacturer more than a year to fully map the firm’s position in its own supply chain.Footnote 74 If network participants lack a full view of the system they are a part of, it follows that their home governments have an even harder time seeing the network that has emerged from these private actors’ dealings. Observing the opportunities and vulnerabilities in real time is next to impossible; these only become apparent after the fact. Such myopia makes strategic calculation unlikely, because the network is not seen as a source of leverage until the different actors’ positions have been reified.
In recent years, following the application of trade policy to national security ends, a growing number of states have hurriedly set about creating government agencies devoted to detecting supply chain vulnerabilities before they are exploited. Japan was first in creating a centralized agency tasked with detecting supply chain ‘choke points’ before they can be exploited by trade partners.Footnote 75 Yet given the complexity and opacity of the supply chains, government agencies have limited ways of mapping these networks by themselves. In the United States, the White House has turned to private sector firms to provide this information voluntarily, opening requests for information (RFIs) ‘asking all parts of the supply chain – producers, consumers, and intermediaries – to voluntarily share information about inventories, demand, and delivery dynamics.’Footnote 76 The result has been an attempt to engage with firms to come up with ‘node-specific problem-solving.’Footnote 77 The United States and the EU have banded together to create an ‘early warning system to better predict and address potential semiconductor supply chain disruptions.’Footnote 78 Increasingly, these early warning systems are being inserted into regional trade agreements.Footnote 79
This large investment in network mapping is a first step in unwinding the implicit consent that both financial and trade networks have been premised on. In industrialized countries, state consent to commercial ties by private actors has been the default position since World War II; this has begun to change. The rise in export controls has imposed constraints on sellers of goods that are labeled as strategic resources;Footnote 80 states have been making increasing use of investment-screening mechanisms to limit foreign countries’ influence in their domestic markets.Footnote 81 States are exercising similarly unprecedented oversight over the placement of digital networks through which much of the world’s trade in services takes place. Thus, in 2020, the U.S. Department of Justice prevented Google from laying a subsea fiber optic cable in Hong Kong.Footnote 82 This was an obstruction of preferential attachment, from fear that the low-cost path offered by the proposed Google cable would ‘would expose U.S. communications traffic to collection by the PRC.’Footnote 83 States have thus begun not only paying attention to the global subsea fiber optic cable system, but also directly intervening to shape it. When these networks initially form, governments typically lack the foresight to anticipate their future distributional consequences. This is due to the complexity and technical nature of these networks, which in turn results from the fact that these networks often emerge from the activities of nonstate actors, the point I turn to next.
(iii) Networks as constructs of nonstate actors
The networks that make up the global economy have significant implications for state interests, yet policymakers have rarely had much say in their design. Most often, the actors at the origin of international networks are merchants, multinational firms, or nongovernmental organizations, all with their own interests.
Thus, the ability to monitor and interrupt global financial transactions was not the product of far-thinking U.S. Treasury officials. It was largely the doing of private American banks, most notably Citibank and J.P. Morgan. By the late 1970s, these banks had established an international clearing house system that facilitated global payments. By creating vast networks that ran freely across borders, these libertarian-minded private actors envisioned themselves as evading government influence, rather than contributing to it, by creating a borderless global payments system.Footnote 84 Rather than celebrate the resulting concentration of operations in a given state, the bankers worried about it. To them, high centrality represented vulnerability because of single points of failure. Yet because the financial piping they relied on all converged on U.S. territory, and because it was denominated in dollars, it inadvertently ended up contributing to the geographic centralization of the global financial sector. Banks kept prioritizing high-capacity links to ensure rapid, reliable data transfer and low-latency connections. They negotiated peering agreements and chose server locations based on cost-effectiveness, positioning themselves so as to allow for shortest path connections, in a way that minimized cost, and allowed for just enough redundancy to ensure reliability. As the network grew, and new entrants to the global economy decided how to conduct their business, the rule of preferential attachment kept favoring one country over all others.
The resulting arrangements have afforded the United States an edge over rival states. It is because the world’s financial transactions run through U.S. banks that the U.S. Treasury can project its domestic power globally, by blocking specific actors from the system, or harvesting data about their financial dealings. It can intercept, surveil, and exert pressure on allies and foes alike. Yet state agencies were long oblivious to the emergence of the modern trade and financial networks they have recently set about exploiting: ‘U.S. officials at the Federal Reserve and elsewhere paid surprisingly little official attention to what was happening.’Footnote 85
Much the same was true a century earlier, when Britain was at the center of the global trade and finance networks. The British Admiralty was initially uninterested in the political benefits of having global trade transactions run through London credit markets and all the world’s maritime insurance underwritten by a single London-based insurance company.Footnote 86 Yet this insight later became central to the waging of World War I, as it offered the British Admiralty an unprecedented glimpse into the dependence of rival countries on strategic commodities, which proved instrumental in the design of a blockade of goods headed for Germany.Footnote 87
To be sure, governments can intervene directly in the formation of these networks, impeding some connections and encouraging others. The U.S. government’s recent interventions in digital networks – from banning the Chinese company Huawei from U.S. markets to blocking the Google cable landing in Hong Kong – are illustrative. In practice, however, governments have rarely anticipated the distributional implications of these arrangements, in part because of their technical complexity and in part because the networks themselves emerged incrementally from the activities of nonstate actors.
To summarize, state objectives are likely to be self-effacing under interdependence, where the emergence of networks relies on willful participation from new entrants. Such buy-in becomes unlikely if other states foresee how it may lead to future vulnerability. Where leverage depends on continued participation, overtly strategic bids for primacy can undermine the very networks on which leverage depends. By a process of selection, when structural gains do nonetheless accrue to some states, it is most often as a result of happenstance, rather than design. Gains are achieved not because governments are farsighted, but because they are myopic. Such myopia is common for two reasons. The sheer complexity of these systems makes it difficult to assess the balance of opportunities and vulnerabilities they will generate. And given how these networks are typically constructed by nonstate actors pursuing efficiency or profit, states often exert remarkably little influence over their original design.
The same reasoning explains why state-led efforts at structural dominance tend to fail. Attempts to design institutions or networks to secure future leverage rarely succeed precisely because their strategic intent is transparent to potential participants. China’s alternative 2018 ‘new IP’ initiative, although technically ingenious, saw limited buy-in and was eventually abandoned.Footnote 88 The Free Trade Area of the Americas collapsed under suspicion of entrenched dominance of U.S. interests,Footnote 89 just as Venezuela’s counterweight to that initiative, the Bolivarian Alliance, quickly faltered under symmetric concerns. Japan’s proposed Asian Monetary Fund never took off, as others balked at its implications for Japanese primacy.Footnote 90 China’s Belt and Road Initiative, lavishly funded and heavily promoted, has encountered the limits of what state-led international infrastructural initiatives can deliver.Footnote 91 Most recently, the U.S.-led Board of Peace, a proposed peacekeeping organization that fashions itself as an alternative to the UN in conflict governance, has encountered similarly low buy-in from allies despite – or precisely because of – strong promotion by the Trump administration.Footnote 92 In other words, where cooperation hinges on willing entrants, openly instrumental bids for dominance invite resistance, whereas advantages that endure tend to arise precisely when they are not pursued as such.
‘Welcome byproducts’ and state surprise
What additional observable implications follow from a theory of self-effacing state objectives? How would one assess whether policy outcomes are a byproduct of orthogonal goals or the result of policy foresight? After all, the functionalist fallacy consists precisely in wrongly assigning intent to final outcomes: upon seeing an institution’s effects, observers mistakenly conclude that these were the purpose behind its formation.Footnote 93 Moreover, the political actors who gain from such byproducts have an even greater incentive to perpetuate such a fallacy, taking credit ex post for outcomes that were not foreseen, or foreseeable, ex ante. The beneficiaries of unintended byproducts seldom have any reason to highlight their lack of foresight.
One means of assessing whether an outcome is an intended product or a welcome byproduct is by looking for signs of surprise. Recall Jervis’s observation that given the automatic counterbalancing inherent in a security dilemma, for one actor to outstrategize another, some, ‘if not all, must be surprised.’ While the premise is that all actors are acting strategically, the expectation under an assumption of common knowledge is that when we do observe a state ‘outstrategizing’ another, no one, including the state in question, should have foreseen its eventual advantage. Odd though it may seem, there does, in fact, appear to be considerable evidence of surprise within state agencies as they realize the extent of their leverage over other states.
In the first modern case of weaponized interdependence, the historical record suggests that Britain realized Germany’s own vulnerability to a trade blockade in the process of measuring its own dependence on imports. Both came as a revelation – in fact, until that point, the data that would have been required to measure either state’s position within the trade network simply did not exist.Footnote 94 There is even reason to think that Germany only became aware of its own trade vulnerability after intercepting findings by the British Admiralty, which led to its efforts to ensure the reliability of neutral ports in Rotterdam and Antwerp.Footnote 95 Eventually, it became clear that in joining a global trade regime operated on British trade routes and financed by the London-based trade credit market, states were providing Britain with tremendous leverage. Yet this was something that neither Britain nor the other participants in the early 20th-century trading system realized. There is evidence of surprise on all sides.
Today, states caught off guard by the implications of their commercial entanglements might be thought a remnant of a distant past, when reliable trade figures were in short supply. Yet while data quality has improved, the complexity of these networks has increased in tandem. The result is that states today appear as blind as they were in the 19th century. Thus, in 2021, South Korea is said to have been ‘caught … by surprise’ when its economy was nearly paralyzed as a result of a shortage of urea following Chinese export measures.Footnote 96 To have ex ante knowledge that a trade partner holds leverage over a strategic commodity like urea requires the type of real-time centralized information processing around supply chains that even highly developed economies lack.
The recurrence of surprise is also apparent from the flurry of activity that countries take when their vulnerability is suddenly made apparent. Within weeks of the Chinese measures, Korea responded by creating its own centralized government body designed to detect vulnerabilities before they are leveraged by trade partners.Footnote 97 The recent creation of such ‘early warning systems’ strongly suggests that states like Korea would not have allowed their firms’ commercial decisions to result in a similar vulnerability, had they had sufficient foresight to prevent it. In other words, China was able to build up this source of leverage over Korea because no one was watching.
This focus on complexity, information, and surprise highlights the way in which epistemic shifts can be as important as material ones in relations between states. Latent advantage can accrue as long as state actors remain unaware. As states become aware of their position in the network and its strategic implications, they seek to exploit their advantage or counter their vulnerability. What separates these two epistemic moments is often some form of crisis – a pandemic, a natural disaster, or an inter-state conflict that brings latent advantages to the fore. While this account does imply a type of ‘epistemic power,’ it is worth distinguishing from existing concepts of ‘knowledge power,’ often associated with Susan Strange, who subsumed it under structural power.Footnote 98 There, some know more than others, and they wield power by controlling access to that knowledge – religious authorities are a commonly cited example. Yet here I assume the exact opposite: as one state grows aware of its leverage over another, we should expect that this other state will become aware of it too. There may be a brief informational advantage, but it is likely to be short-lived, and no theory should be premised on it.
Implications for statecraft
It might seem that if beneficial outcomes are a ‘welcome byproduct’ rather than the result of foresight, any attempt at ‘outstrategizing’ others would be futile, as unlikely as gaining abnormal returns in a perfect market. Strategic behavior on the part of all parties in an interdependent setting would leave only the possibility of accidental advantage. This need not be the case; happenstance can be designed. Yet this process is not costless. If states are to attain self-effacing objectives, they must cede some measure of control. To show how states manage the paradox of self-effacing ends, I look to a third issue area, distinct from the trade and financial networks I have used as my illustrative cases thus far. My purpose here is twofold: (i) to show that although self-effacing state objectives are especially likely under commercial interdependence, they are not limited to such networks; and (ii) to show that when states recognize the self-effacing aspect of their objectives, they can act to create conditions that render ‘welcome byproducts’ more likely.
Foreign influence in U.S. Cold War info ops
Persuasion is a paradigmatic self-effacing objective: to be effective, the would-be persuader would not overtly aim at persuasion, but at some ostensibly unrelated objective, which might inadvertently yield persuasion as a byproduct.Footnote 99 To tie to the main argument, persuasion requires a form of voluntary buy-in analogous to the emergence of commercial networks. The target of persuasion must willfully concede to the persuader’s message. In both cases, acceptance cannot be coerced: cooperation in networks ceases to function if participation is forced, and persuasion ceases to persuade if the audience suspects manipulation. Suspicion of the persuader’s underlying incentives thus makes assent less likely, just as suspicion of a state’s strategic design makes others reluctant to join common networks.
These insights are vividly illustrated by a sudden shift in the conduct of American foreign influence campaigns. Following Stalin’s death in 1953, rising neutralist and anti-American sentiment across Europe led to the realization that U.S. foreign influence programs under President Truman, such as the overtly anti-communist ‘Campaign of Truth,’ had been ineffective, ‘if not counterproductive.’Footnote 100 The stakes were high: psychological warfare had become the principal battleground of the Cold War. As Edward Barrett, the assistant secretary of state for Public Affairs, put it in 1950, the goal was ‘to penetrate the Iron Curtain with our ideas.’Footnote 101 Officials within the U.S. administration feared that the United States was falling short on this key front.
In response, President Eisenhower created two presidential committees to review U.S. overseas propaganda during his first year in office: the President’s Committee on Governmental Organization and the President’s Committee on International Information Activities. The latter, known as the Jackson Committee, produced what arguably remains the most important government study of the mechanisms of state information operations. It concluded that U.S. propaganda efforts up to that point had suffered from a ‘severe credibility problem,’Footnote 102 since ‘foreign audiences’ were ‘quick to take offense at advice and exhortation received from abroad.’Footnote 103 The report warned that all the American resources spent on foreign influence could actually have net negative results: ‘anti-communist propaganda may antagonize more foreigners than it convinces.’
The Jackson Committee had identified the self-effacing aspect of their objective: persuaders who aim squarely at persuasion generate resistance and fall short of their goal. The most effective propagandists would not be propagandists. Tellingly, these insights did not originate from policy circles, but from then-emergent fields of advertising and public relations. Edward Bernays, the putative inventor of the field of public relations, came from the advertising industry to work as counsel for U.S. propaganda efforts before and after World War II. In advocating for a different approach to foreign influence, Bernays offered an especially apt metaphor for what a strategy for securing ends-as-byproducts would consist of: ‘It’s like shooting billiards, where you bounce the ball off cushions, as opposed to pool, where you aim directly for the pockets.’Footnote 104 Foreign influence was likely to be ineffective if it took the shortest, most direct path. Faced with a self-effacing objective, strategic actors would instead aim at apparently unrelated catalytic goals.
Until that point, U.S. propaganda efforts had tended toward an overt, heavy-handed approach not much different from those of the totalitarian regimes against which they pit themselves against.Footnote 105 Then, largely in response to the Jackson Committee’s recommendations, the Eisenhower administration stepped away from the overt ‘Campaign of Truth’ to a set of more nuanced tactics. The main change consisted of reducing direct action, relying instead on various surrogates and front organizations, and removing attribution to U.S. sources wherever possible, so that messaging would appear to arise organically from independent voices pursuing their own interests. The Jackson Committee explicitly listed ‘American business firms,’ ‘universities and foundations,’ ‘American missionaries,’ and ‘labor groups’ as under-utilized surrogates that could effectively carry pro-U.S. messaging without overt government involvement. Another identified source of proxies was refugees from the Soviet Union and its satellite states: ‘politicians, cabinet members, writers, and academics,’ whose origins credibly set them apart from official U.S. government messaging.Footnote 106
Concurrently, U.S. government agencies invested heavily in supporting sympathetic media outlets. They subsidized their activities, facilitated their foreign correspondents’ work, and provided them with confidential briefings and administration-scripted information. Yet in employing American media as an ‘honest surrogate,’ U.S. government officials understood they needed to allow these media actors sufficient autonomy, so that they would remain convinced of their own objectivity and continued devotion to the ideals of a free press.Footnote 107
Such reliance on surrogates notably differs from the more conventional form of delegation seen in principal–agent relationships. There, agents are explicitly empowered to implement predefined policies aligned with the state’s objectives. By contrast, in the pursuit of self-effacing ends, surrogates pursue orthogonal goals, whose alignment with state interests appears incidental. Crucially, surrogates can only be effective if they perceive themselves as genuinely serving their own cause, rather than that of the state. The state thus relinquishes more control at the outset than in traditional delegation by a principal to an agent.
Eisenhower himself was highly aware of the tradeoff involved. His oft-quoted claim about how the ‘the hand of government [in propaganda campaigns] must be carefully concealed, and, in some cases, I should say, wholly eliminated’ comes from a 1953 letter about the Voice of America radio program – one of the main, and most overt, organs of U.S. propaganda efforts.Footnote 108 The last clause of that sentence is worth insisting on: Eisenhower recognized that the state’s interest may be best served by removing itself entirely from the equation.Footnote 109 A loss of direct control might ultimately translate into greater influence. This insight permeated the newly transformed U.S. propaganda machine. To yield beneficial byproducts, surrogate agents needed to pursue their own goals, which would necessarily be distinct from official U.S. government goals. Yet the calculation was that the tradeoff would pay off. As the Jackson Committee report put it:
Far greater effort should be made to utilize private American organizations for the advancement of United States objectives. The gain in dissemination and credibility through the use of such channels will more than offset the loss by the Government of some control over content. Footnote 110
In relying on surrogates with distinct interests, U.S. officials were willfully giving up control; it was a costly, but calculated move. Today, this insight continues to be applied by democracies and autocracies alike. A recent treatment of contemporary Russian information operations shows how Russian state media sources reach ‘unwitting audiences’ in Estonia by relying on ‘relay stations,’ outlets that disseminate Russian content while ‘benefit[ing] from local credibility.’Footnote 111 Similarly, Chinese state officials speak of ‘borrowing a boat out to sea’ (jiechuan chuhai) to refer to the practice of using foreign media to deliver Chinese messaging.Footnote 112 Then, as now, propagandists maximize their influence by relinquishing direct control and acting through surrogates, proxies, or ‘cut-outs’ that avoid the defensive reflex of wary audiences.Footnote 113 When strategic actors recognize the self-effacing nature of their objectives, they increase their odds of attaining these by relying on indirect means, bouncing the ball off cushions, rather than aiming for the pockets. What they lose in direct control, they gain in overall effectiveness.
Conclusion
This article has argued that international relations includes a class of objectives that are self-effacing: goals that cannot be attained through direct pursuit, but only as byproducts of actions aimed elsewhere. Self-effacing objectives are especially likely under the type of interdependence that characterizes the global economy. In interdependent settings, the pursuit of dominance invites countermeasures, while advantages that emerge indirectly – through the dealings of nonpolitical actors, or as byproducts of unrelated goals – are less likely to provoke resistance.
The mechanisms behind this pattern are threefold. First, interdependent networks require voluntary buy-in, and other states are unlikely to join an arrangement if they can foresee that it will leave them strategically disadvantaged. Second, the formative stages of such networks are usually driven by nonstate actors – firms, banks, NGOs – pursuing their own ends. Finally, the technical complexity of these systems makes it difficult for governments to perceive emerging asymmetries until much later. The combination of these features means that durable sources of leverage are more likely to emerge as byproducts, rather than the result of farsighted policymaking. Network power under interdependence is a self-effacing objective.
Historical experience bears out this paradoxical conclusion. Britain’s ability to monitor and cut off German trade at the outbreak of World War I rested on financial and insurance networks whose strategic value policymakers only recognized on the eve of war. The United States’ current leverage over global financial flows likewise grew from dollar-clearing systems created by private banks seeking to escape regulation, not from government design. Even China’s recent dominance in rare earths reflects a similar pattern, where resources once treated as marginal industrial inputs became strategic chokepoints, owing to no foresight on the part of policymakers. In each case, centrality was attained precisely because it was not deliberately pursued.
To be sure, not all state objectives are self-effacing. Investments in scientific research, economic productivity, or infrastructure all benefit from farsighted planning and strategic action. Yet when these goals take on a relational dimension, when states become intent on achieving scientific dominance in a given area, outcompeting their trade partners in a strategic industry, or deploying their infrastructural designs abroad, these objectives invite countermeasures and take on an increasingly self-effacing aspect. In this way, the argument assumes strategic state action rather than denying its possibility. It also assumes a high level of common knowledge: if Home can see ahead to how its actions will offer it an edge, theory should assume that Foreign can do so as well. The one sure way for Foreign not to become aware of its potential vulnerability is if Home is equally oblivious to the sources of its own advantage.
Yet statecraft is not futile, even in the face of self-effacing objectives. I examine a pivotal shift in U.S. information operations during the Cold War to illustrate the point. The Eisenhower Administration recognized the self-defeating aspect of overt propaganda campaigns and adjusted its approach by relying on ‘honest surrogates.’ While this strategy involved ceding some control, it ultimately enhanced the United States’ foreign influence. Contrast this indirect approach to today’s prevailing approach to statecraft. States are increasingly taking direct action in markets to gain an edge in sectors deemed critical for national security or economic competitiveness, through policies ranging from export controls and local content subsidies to import barriers and investment screening mechanisms. The objective is not increased productivity per se, so much as the attainment of an edge vis-à-vis foreign competitors. In response, firms appeal to the state for support and protection, presenting themselves as indispensable to national interests by framing their activities as critical to securing this relative advantage.Footnote 114
The reason such a direct approach is unlikely to yield its desired outcomes is not because of the common concern over governments’ inability to ‘pick winners,’ but because of the transparency of the strategic intent. Under the high degree of interdependence that characterizes the global economy, gaining dominance is a self-effacing goal. Major powers have been mimicking one another’s policies, focusing on the same strategic industries, and promoting domestic firms at the expense of foreign firms; the net effect is that few state-actors have succeeded in gaining an edge. The implication of the analysis is that states are more likely to gain a lasting advantage by promoting ecosystems in which nonstate actors are able to pursue their own interests, in ways that are broadly aligned with, but not directly tied to, state interests. By relinquishing some control, states paradoxically stand to gain the most enduring forms of leverage. Broader, less targeted investments, such as in basic research, may be better geared at achieving self-effacing objectives by cultivating conditions for ‘welcome byproducts.’ Ultimately, in an interdependent global economy, it is not the calculated moves of state planners, but the quiet contributions of accidental architects that stand to redefine inter-state relations.
Acknowledgments
I thank Jeff Colgan, Christina Davis, Noel Forster, Eric Haney, Harold James, Dominic Johnson, Edward Keene, Alexander Kuo, Jan Eijking, and the participants of seminars at McGill University and Oxford University for helpful comments. This publication was made possible in part by a grant from the Carnegie Corporation of New York. The statements made and views expressed are solely the responsibility of the author.
Competing interests
The author declares none.
Open access
