carter’s nuclear security dilemma

Carter’s early approach was characterized by classical liberalism and its pro­mise of peace for peace. However, the exercise of either promising to support a nuclear deterrent counterstrike or effectively engaging in Schelling-style bluffing seemed out of reach for Carter.

Behind-the-scenes conversations offer one insight into what led President Carter to sign Presidential Directive 59, which put the US military on a footing treating nuclear weapons as conventional forces, planned to fight a protracted nuclear war, and incorporated the offensive MX missile system. However, understanding the broader intellectual and political climate is also important. Thomas Schelling and Robert McNamara terminated their active engagement with arms control by the late 1960s at the same time that Albert Wohlstetter, Herman Kahn, Colin Gray, Paul Nitze, and Edward Teller initiated a vocal public campaign to promote a pro-nuclear-use policy.^[285] Jervis defended mutual assured destruction, initially in his 1976, 1978, and 1984 publications, and then more effectively in The Meaning of the Nuclear Revolution in 1989.^[286] Carter was counseled by hawkish Secretary of State Zbigniew Brzezinski; his Secretary of Defense Harold Brown; and his Secretary of Energy James R.

All are agreed that the nuclear arms race was angst ridden.⁵¹ Carter had entered office amenable not only to arms control but also to disarma­ment. He made his intention clear in Presidential Directive 18, which directed that the United States should “‘take advantage of our relative advantages in economic strength, technological superiority and popular political sup­port’ both to seek Soviet cooperation in resolving conflicts, renegotiating arms control agreements, and constructively dealing with global problems and to counterbalance adverse Soviet influence in key areas of the world.”⁵² His apparent lack of adequate concern for defense frightened the pro-nuclear- use contingent of policy analysts. He wrote in his personal diary in August and October of 1977:

Met with the Committee on Present Danger, Paul Nitze, Gene Rostow, and others. It was an unpleasant meeting where they insinuated that we were on the verge of catastrophe, inferior to the Soviets, and that I and previous presidents had betrayed the nation’s interests. I told them I’d like to have constructive advice, balancing all factors with at least the possibility considered that the Soviets did want a permanent peace and not suicidal nuclear war

In Congress, Senator [Scoop] Jackson was the core around whom the most vitriolic anti­Soviet forces coagulated.

weapons to weapons of catastrophic destruction and sought to “control... the level of violence in any conflict” ;see Lawrence Freedman, Evolution ofNuclear Strategy (New York: St. Martin’s Press, 1981), 377-392, quote from 384. There can be no doubt that the Schlesinger Doctrine seeks defense in maintaining the relevance of violence to control outcomes in conflict situations by maintaining the asymmetric policy of escalation dominance.

⁵¹ The Bulletin of American Scientists kept a constant barometer on their members’ estimation of thelikelihoodofnuclearwar; for discussion, seeHerken, Counsels ofWar, 1985, 105, 125, 185, 192, 247, 303.

⁵² Brzezinski to President Carter, subject, “Capitalizing on Our Economic Advantages in U.S.-SU Relations,” undated, NLC-29-11-2-3-3, JCPL, Brzezinski Collection, declassified 2008/04/09, p. 1 of 2.

⁵³ President Jimmy Carter, White House Diary (New York: Farrar, Straus, and Giroux, 2010), August 4 and October 25, 1977, 76, 123-124.

Carter worried some US defense analysts because he seemed to accept that the Soviets had benign intention, and that the United States and the USSR could work together to ensure peaceful coexistence. The Soviet’s 1979 invasion of Afghanistan strengthened their belief that the Soviets intended to extend their empire using military force.^[288] Notwithstanding that MAD was a fact and not a policy, defense rationalists themselves were hard pressed to defend it against the NUTS plan to prepare to fight and win a nuclear war.^[289]

Since analysts conceded that the Prisoner’s Dilemma best characterized nuclear security dilemma and arms race, a policy of mutual assured destruction could no longer be rationally sustained.

Schelling introduced the initial ambiguity of accepting that a Stag Hunt Assurance Game transforms into the more virulent Prisoner’s Dilemma as a function of uncertainty about others’ intentions. Hence, he gave rise to a characteristic Prisoner’s Dilemma pedagogy that sanctioned the idea that a predatory stance is wholly legitimated by and consistent with benign intent. Schelling’s analysis was accepted by strategists who felt compelled to address the “worst contingency” security dilemma, which everyone seemed to concede resembled the disconcerting PD.^[292] The general acceptance of the widespread applicability of the Prisoner’s Dilemma and PD logic of gain without regard for others is a direct result of the development of rational deterrence theory entangled with evolving nuclear policy.

It was standard to view the high-stakes nuclear superpower standoff in terms of a single-play Prisoner’s Dilemma, which is the default in game theory as a result of emphasis on tangible outcomes, which even in an assurance standoff (Stag Hunt) with sufficient uncertainty necessarily transforms into the intransi­gent PD.⁶1 However, the PD formalization of the security dilemma and wholesale adoption of strategic rationality without a doubt shifted the challenge from assuring the other of one’s benign intent to motivating cooperation through one’s wherewithal to issue credible threats of harm.

Schelling’s strategically rational defense of MAD looks plausible but was found to have three fatal flaws attributable to its PD structure: immorality, incredibility, and irrationality. Were the United States to be hit by a Soviet all- out first strike and the only recourse was to strike back to wreak similar damage on the Russians, not only would this counterattack be immoral, but it stood indicted for lacking any causal power to serve US interests after deterrence had failed and, thus, any credibility to deter in the first place.⁶³ Defense rationalists, consequently, reasoned that MAD rested on an immoral, incredible, and there­fore irrational threat to strike back when such an act can only seal its own doom: “It is perhaps a central tension in deterrence... that its ultimate threat is to engage in a senseless act of total destruction.”⁶⁴ Without any contingency plan in place to fight rather than admit defeat, MAD further seemed effete.⁶⁵ The immorality of the threat of massive retaliation was the undoing of MAD because it signified the incredibility of following through, thus rendering deter­rence inconsequential. Additionally, MAD could be accused of being suicidal if the act of following through on a counter strike would provoke additional Soviet missile strikes on America.⁶⁶

Jervis, “Security Under the Security Dilemma,” 1978; Freedman, Evolution ofNuclear Strategy, 1981. Ofcourse, Thomas Schelling’s Strategy of Conflict (Cambridge, MA: Harvard University Press, i960) had started this practice.

⁶² The question of whether the Soviets had the capability to detect US submarines was raised in the famous Team B Report that concluded that the fact there was no evidence of such technology provides sufficient reasoning that it may exist; Anne Hessing Cahn, “Team B: The Trillion- Dollar Experiment,” Bulletin of the Atomic Scientists (1993), 49:3, 22, 24-27.

⁶³ Jervis addresses this point, following Patrick Morgan in noting that “if people were totally rational, deterrence in a world of mutual assured destruction would not work. To carry out your threat would mean the destruction of your own society; so, if the other side thinks you will retaliate, it assumes you are less than rational Robert Jervis, “Deterrence Theory Revisited,” World Politics (1979) 31:2, 289-324, at 299).

⁶⁴ Quoted from ibid., 300.

⁶⁵ Discussed by Freedman, Evolution ofNuclear Strategy, 1981, 395.

⁶⁶ This was Schlesinger’s stated concern; see his US Senate testimony, “Uses and Abuses of Analysis,” 1968, 340.

Moreover, in the continual contest of wills between the Soviets and Americans all too evident in Carter’s daily log of White House events, MAD suggested a posture of “better Red than Dead” and did not provide a strong position from which to bargain.^[296] The nuclear security dilemma modeled as a Prisoner’s Dilemma transformed into a Chicken game once both sides faced the fear of potential mutual devastation yet still vied for supremacy.^[297] Without continually maneuvering to at least achieve mutual cooperation rather than ignominious submission, it seemed that even if MAD did prevent a nuclear war, it would grant the Soviets a victory in the Cold War.

It was a signature belief of the defense rationalists that the threat of violence could be calibrated and applied to either compel or deter actions of the other side.^[298] Both Schelling and Kahn advanced this view. For Schelling, the idea had been to manage risk in mobilizing threats, whereas for Kahn, the plan was to manage military application of force to achieve escalation dominance. In either case, politics and war became indistinguishable.^[299] The recalcitrant Prisoner’s Dilemma game, in which each person in pursuing his best interests mires both in a suboptimal outcome, was accepted in deference to the need to prepare for the worst case in which one’s own defense threatens the security of the other.^[300] According to the PD analysis of the nuclear security dilemma, nuclear weapons signify that defense must take the form of offensive action from which no one can be invulnerable. Even though mutual vulnerability is inescapable, the voices that clamored for proactive preparation to wage war, rather than those coun­seling the acknowledgment of reciprocal vulnerability, prevailed.^[301]

NUTS seemed suited to address each of the signature weaknesses of MAD.^[302] First, it signals the unwavering intention to counterattack if attacked at any rung of engagement.⁷⁴ Second, it has a plan if deterrence fails: to fight for victory no matter what.⁷⁵ Third, it recommends “firing demonstration shots to show resolve.”⁷⁶ Fourth, it accepts the challenge of a nuclear Chicken contest of wills, providing the strongest position from which to bargain.⁷⁷ Nevertheless, NUTS rests on the fallacy that it is possible to meaningfully engage in nuclear conflict, and it ignored the Soviets’ promise to retaliate against any use of nuclear weapons and lost sight of the fact that “the primary objective of nuclear strategy is to avoid wars, not to fight them.”⁷⁸ NUTS openly adopts a one-sided posture on defense in the full knowledge of the fact that achieving strategic dominance is more important than reassuring the other actor of one’s benign intent.

President Carter had entered office exemplifying a classical liberal security posture. The classic liberal resolution of the security dilemma for both interna­tional relations and civil society, articulated in some form by Samuel Pufendorf, Hugo Grotius, Thomas Hobbes, Benedict Spinoza, John Locke, Adam Smith, Immanuel Kant, John Stuart Mill, Isaiah Berlin, and Friedrich Hayek, rests on a several key pillars.⁷⁹ Since self-preservation is basic for every actor, and the

⁷⁴ Escalation control is linked to flexible response under the reasoning that deterring, or preventing further, acts of Soviet aggression depends on having the flexible capability to prevail at any level of conflict. Of course, the debate is arrested on the point of whether introducing nuclear weapons into a conventional conflict would entail “escalation... [that] would still become uncontrolla­ble”; pointed discussion of this debate is in “Senate Foreign Relations Committee Paper on PD 59”; the paper is attached to a memo from Jasper Welch to Brzezinski, September 11, 1980, report is dated September 9, 1980, from San Sienkiewicz, p. 3 of 8, “5/80-1/81,” Box 35, Brzezinski Collection, JCPL.

⁷⁵ The plan is “To assure the survival of the US as a functioning independent nation, capable of political, economic, and military recovery,” stated in “Countervailing Strategy for General War,” attached to a memo from Ermarth to Welch and Utgoff, March 20,1980, “Countervailing Strategy and the Targeting Problem,” memo two pages, report p. ι of 4, “3/80-4/80,” Box 35, Brzezinski Collection, JCPL.

⁷⁶ Quoted from William Odom memorandum to Brzezinski, “Draft PD on Nuclear Targeting,” March 22,1980,p. 4 of 5, “5/80-1/81,” Box 35, Brzezinski Collection,JCPL (note that document is out of temporal sequence in its placement in the file folder).

⁷⁷ The ability to bargain, especially in crisis setting, is mentioned in the memorandum leading up to PD 59, e.g., Special Coordination Committee Meeting, April 4, 1979, direct reference to “crisis bargaining,” as a key topic for discussion, “Summary of Conclusions”; “8/78-4/79,” Box 35, Brzezinski Collection, JCPL. Hollis and Smith provide a helpful discussion of the paradoxes embattling MAD from the perspective ofdefense rationalism, Explainingand Understanding in International Relations, 1990, 173-174. For a thorough analysis of the Schlesinger Doctrine’s contribution to the puzzles of deterrence via MAD, see Freedman, Evolution of Nuclear Strategy, 1981, 374-395; Glaser, “Why Do Strategists Disagree,” 147-148. Note that Schlesinger accepted that the Soviets’ behavior would be based on their perception of the credibility of the US deterrent, which he interpreted as a rationale to further buttress US credibility to engage in nuclear war because he worried that the Soviets perceived the United States as benign; hence, Jervis’s Perceptions and Misperceptions in International Politics, 1976.

⁷⁸ For discussion, see Freedman, Evolution of Nuclear Strategy, 1981, 379, 391, 385.

⁷⁹ Richard Tuck’s The Rights of War and Peace: Political Thought and the International Order from Grotius to Kant (Oxford: Oxford University Press, 1999) explains how liberal motive of self-preservation pertains to all actors, actors can peaceably coexist only if they concede to each other the right to exist and therefore voluntarily desist from harming others. This way of interpreting amicable relations among nations or individuals reduces to the pledge of “peace for peace” and the threat of “war for war.”⁸⁰ Where liberalism views a state of nature as the return to the unconstrained natural right to all things, and civilization as a rarefied state organized by self-adopted rules and commitments, neoliberalism views the achievement of order as a function of equilibria arising from actors’ uncon­strained aspirations. Liberalism views the recourse to violence as a breakdown of social order; neoliberalism views social order as derived not from promises, but from credible threats of violence.⁸¹

To understand the transformed approach to mutual security, we must grasp how the Prisoner’s Dilemma was used to motivate MAD, and identified as an inescapable logical puzzle miring MAD in the inevitable outcome that deter­rence relies on an irrational threat.⁸² The incredibility of the MAD deterrent threat was subject to ongoing attention by defense rationalists.⁸³ The strategic analyst Lawrence Freedman captures the dilemma of nuclear deterrence: “Yet the question of how nuclear weapons could be used in war remained and continued to nag at responsible officials as well as academic strategists. Once one openly admitted that the nuclear arsenal was unlikely ever to be activated then the deterrent lost all credibility.”⁸⁴ MAD seemed arrested by paradox: if

international relations theory predicated on the no-harm principle anteceded the civil model for liberal governance. Michael Doyle is particularly insightful on the classic liberal tradition in international relations, Ways of War and Peace, 205-314. Obviously, classic liberalism would come under attack on many fronts in the late twentieth century; see, e.g., Samuel Moyn, The Last Utopia (Cambridge, MA: HarvardUniversity Press, 2010), but its eclipse did not need to result in Prisoner’s Dilemma logic of aggressive self-defense regardless of others.

⁸⁰ For a succinct discussion of this tradition as it was initially articulated by Grotius and Hobbes, see Richard Tuck, Hobbes: A Very Short Introduction (Oxford: Oxford University Press, 1989), 26­29; note how far neoliberalism is from liberalism given that in its day the latter was considered “illiberal” for condoning voluntary slavery and absolute monarchy; at least it established a normative order by uniting might with right instead of permitting might to establish right; on the latter, see Russell Hardin, “Does Might Make Right,” in J. Roland Pennock and John William Chapman, eds., Authority Revisited (New York: New York University Press, 1987), 201-217.

⁸¹ This is a central theme ofThomas Schelling’s research, see Strategyand Conflict, 1960; the idea is that both threats and promises are levied to achieve an end that would rather be obtained without needing to take the act as either threatened or promised.

⁸² For an emphatic statement of this, see Jervis, “Deterrence Theory Revisited,” 1979, 300.

⁸³ This concern had been articulated by Brennan and the pro-nuclear-use strategists as early as the late 1960s as assured destruction was renamed mutual assured destruction, or MAD; James R. Schlesinger refers to the “suicidal implications” of assured destruction; “Uses and Abuses of Analysis,” 1968, 340; Harold Brown admits assured destruction’s incredibility deriving from the fact that “it is at least conceivable that the mission of assured destruction would not have to be executed at all in the event that deterrence failed,” although it is important that any “potential enemy” not be led to believe this possible; Harold Brown, “Report to Congress 1979 Budget, FY 1980 Authorization Request, and FY 1979-1983 Defense Programs,” January 23, 1978, 57.

⁸⁴ Freedman, The Evolution of Nuclear Strategy, 1984, 392-393. nuclear armaments would only be used to seal the end of civilization, then there could be no conceivable plan for their use unless one embraced mass destruction and reciprocal suicide. However, if one held the nuclear arsenal with no inten­tion of ever deploying it, then it could not stand as a credible deterrent threat.

Freedman thus goes on to explain, “If weapons had to be designed for operational use then some sort of guidance was necessary, which required stating a preference for one form of nuclear employment against another.”^[303] Freedman identifies a puzzle over what makes deterrence work, capturing the standard application of the Prisoner’s Dilemma model to represent the puzzle of deterrence.^[304] Without second-strike ability, each side was vulnerable to the other’s initiation of a first strike; the introduction of second-strike capability neutralized the other’s unilateral advantage, but only if one would actually follow through on a massive counterattack, or at least was believed that it would do so. Insofar as the strategic policy of MAD kept weapons in their silos until devastation was already certain, American nuclear arms would serve no function. To strategic planners, it seemed necessary to stipulate an opera­tional use for nuclear weapons so that they could serve a constructive causal purpose furthering national goals.^[305] If one started with the premise of striving for strategic dominance, even if ultimately the fact of MAD results in a game of nuclear Chicken, at least one clearly signals the intent to prevail rather than settle for submission.

Whereas MAD took seriously the inability to constructively wage nuclear war, and the Soviets’ continual assertion that any use of nuclear force would lead them to counter with massive retaliation, NUTS was wholly dedicated to developing the meaningfulness and possibility of waging nuclear war and to acquiring weapons accordingly.^[306] The difference between the two perspectives is clear in a brief exchange between Brzezinski and Brown. Brzezinski points to three major points of discussion in moving forward with PD 59:( 1) the require­ments of stable deterrence, (2) “requirements of stable crisis bargaining,” and (3) “requirements of effective war management.” Brown, following the logic that escalation control and war management are extremely unlikely, especially in prolonged conflict, said that “it is important to have our planning for all out nuclear war well in hand because all out spasm war is the most likely possibility, given the unlikely possibility of nuclear war in the first place.” In other words, nuclear deterrence is sufficient to prevent the escalation into nuclear conflict, and the emphasis should thus be on preventing conflict in the first place. Once nuclear conflict is initiated, he reasons, fighting meaningfully misses the main point of deterrence to avoid war altogether. However, holding out the hope of being able to successfully prevail in prolonged nuclear combat, Brzezinski offers the counter­ing thought that “the very likelihood of all out nuclear war is increased if all out spasm war is the only kind of nuclear war we can fight.”^[307]

Additional discussion makes clear that the Carter administration abandoned the MAD footing both as an acquisition policy and as an employment policy, notwithstanding the overall recognition, as Jervis repeated throughout his career, that “MAD as a condition with which we and the Soviets are stuck, has obtained at least since the late 1960s.”^[308] Thus, it is impossible to exit the reality of mutual assured destruction. Nevertheless, the flexible response, coun­tervailing policy was gradually and continually introduced both as the guideline for purchasing weapons systems and for their employment. The MX missile system controverts MAD, which is based on accepting mutual vulnerability. Flexible response plans to employ nuclear weapons as a natural escalation from conventional warfare, with the plan of capping escalation; however, in reality, it cannot guarantee escalation control any more than MAD can guarantee deter­rence. In 1980, as PD 59 was moving through the approval process, US govern­ment defense analysts observed that “MAD as an employment doctrine has never really been in force, thus PD 59, which would be a dramatic departure had that been so, is rather just another step in a gradual and long-run policy evolution.”^[309] This is because the so-called Schlesinger Doctrine had been inher­ent in strategic rationality since the late 1960s. From McNamara onward, the SIOP had targeted almost every Soviet concern worth targeting. Still, of course, Carter’s endorsement of the policy to procure and deploy powerful first-strike weapons and his commitment to having the power to engage in lengthy nuclear war was wholly unprecedented.^[310]