Foreign Policy Research Institute A Nation Must Think Before it Acts Donald Trump Folds the Nuclear Umbrella: Is American Extended Deterrence Passé?
Donald Trump Folds the Nuclear Umbrella: Is American Extended Deterrence Passé?

Donald Trump Folds the Nuclear Umbrella: Is American Extended Deterrence Passé?


The road to wisdom?
Well, it’s plain and simple to express:
Err, and err, and err again
but less, and less, and less.
                    -Piet Hein, Grooks (1966)


It might be said that among foreign policy and national security watchers, Donald Trump is a gift that keeps giving. The New York Times reported recently that Mr. Trump said:

“[H]e would be open to allowing Japan and South Korea to build their own nuclear arsenals rather than depend on the American nuclear umbrella for their protection against North Korea and China. If the United States ‘keeps on its path, its current path of weakness, they’re going to want to have that anyway, with or without me discussing it,’ Mr. Trump said.”[1]

Mr. Trump implicitly acknowledged Kenneth Waltz’s c.1980s observation that new nuclear states will likely “come in hostile pairs and share a common border.”[2] Thus his response begs a question: if a nuclear-armed North Korean and China justify South Korean and Japanese nuclear weapons, what would Mr. Trump say about Iran’s neighbors? What Mr. Trump omitted (or perhaps the question did not allow for it) was that, again to paraphrase Dr. Waltz, the nuclear challenge does not neatly reduce to a simple menace. The ability to fashion a crude nuclear device is one thing; the ability to deliver a nuclear warhead accurately to its intended target is quite another. While a survivable counterstrike capability arguably mutually deterred antagonists during the Cold War, deterrence may mean something entirely different in the face of a cross border threat from, say, North Korea. There, an indigenous nuclear counterstrike capability may not deter an adversary nearly as well as, say, a robust anti-ballistic missile defense.[3]

Mr. Trump’s timing was not propitious. On the same day The New York Times published his remarks, the North Korean propaganda website DPRK Today posted a new video[4] titled “Last Chance,” in which a submarine-launched nuclear missile is shown laying waste to Washington, D.C.[5] Five days later, published reports citing Pentagon sources claimed North Korea has developed a new long-range mobile intercontinental ballistic missile, the KN-14, which is a range-extending variant of an earlier North Korean ICBM.[6] And on that same day, the South Korean Ministry of National Defense announced a USD28 Billion investment in anti-missile defense. This includes acquiring an unspecified number of Taurus KEPD 350 missiles, a long-range air-to-surface missile designed to penetrate dense air defense systems, and destroy hard and underground targets, including air base facilities, bunkers, port facilities, command, control & communication stations, ammunition storage facilities, bridges, ships in ports, and runways.[7]

The disconnect between Mr. Trump entertaining a nuclear armed South Korea and Japan, and actual events on the Korean peninsula in the span of just several days is troubling. It is nonetheless worth pondering Mr. Trump’s suggested withdrawal from East Asia. For brevity’s sake, the author focuses mostly on the Korean peninsula though many points regarding South Korea apply as well to the suggestion of a nuclear-armed Japan.

The first question that comes to mind is “withdraw what?” It has been some time since extended deterrence — the American “nuclear umbrella” that covers regional allies — meant forward deployment of American strategic systems and tactical nuclear weapons in East Asia. Moreover, the number of United States troops stationed on the Korean peninsula is at or near its lowest level since 1951. Is there any serious suggestion that removing the remaining American force — thereby eliminating a tripwire to deter overt North Korean aggression — would positively influence the actions of that rogue state, which wraps its nuclear weapon ambitions in the language of a counter-deterrent to the American nuclear umbrella. Is Mr. Trump aware that wartime operational control (known by the acronym OPCON) of South Korean armed forces transfers to the United States?[8] In mid 2014, the South Korean national security adviser Kim Kwan-jin vowed “to relentlessly retaliate against the enemy and completely make it surrender in case of the enemy’s provocation.”[9] Is Mr. Trump really suggesting the economic burden of extended deterrence in East Asia is so onerous that the United States ought to relinquish the OPCON and accede to an autonomous South Korean nuclear counterstrike force, thereby removing a critical escalation barrier? 

Mr. Trump and his critics alike overlook an important fact. The East Asia nuclear umbrella has been provided solely by American strategic forces — either deployed in the United States or at sea aboard Trident ballistic missile submarines — since the United States withdrew nuclear weapons from the region in September 1991.[10] As an aspiring CINC, surely Mr. Trump is aware the United States’ mutual defense treaty with South Korea is no longer bolstered by the presence of tactical nuclear weapons on the peninsula, as it was at the time when the North Korean threat was largely based on conventional weapons? Moreover, Mr. Trump must be aware of sharp opposition to deploying Terminal High Altitude Area Defense (THAAD) batteries inside South Korea? China asserts it would make South Korea a forward base in the United States Missile Defense Agency’s Ballistic Missile Defense System[11] and explicitly threaten China’s asymmetric anti-access/area denial (A2/AD) strategy, one aim of which is to exclude American forces from the so-called “First Island Chain”[12] off China’s coast.[13]

Mr. Trump makes much of how the economic burden to sustain NATO is distributed unevenly, to America’s detriment. But in East Asia, the United States has allies but no NATO-like multilateral alliance. An October 1953 mutual security treaty is the legal foundation of its bilateral security relationship with South Korea.

One thing is certain: nuclear weapons have not altered the anarchic structure of the international political system.[14] Another near certainty is that by airily discounting extended deterrence on the Korean peninsula, Mr. Trump risks exacerbating the security climate there. Some might argue his desultory comments are disturbingly simplistic. No North Korean armed attack can be separated, wrote Anthony H. Cordesman and Ashley Hess, “from the role US forces would play in a conflict, from Japan’s willingness to support US basing and staging into Korea, and from the role China would play in trying to limit any threat to the DPRK as a buffer state.”[15] Moreover, a strategic assessment by the United States Forces Korea indicates that North Korea has built a surprise attack capability specifically designed for affecting the economic and political stability of South Korea “with little or no warning”. The risks associated with actively encouraging South Korea to deploy an indigenous nuclear force should be patently obvious: factors such as short geographic distances and North Korea’s use of mobile launch platforms amplify the threat of a preemptive attack, likely forcing South Korea into a dangerously reactive posture. This would return us to the dangerous days of the late 1950s, when the necessity of “launch on warning”[16] was taken for granted.[17]

A wiser statement by Mr. Trump might perhaps have suggested extending the Obama Administration’s concept of “regionally tailored deterrence architecture” to the Korean peninsula, something which might have demonstrated awareness both of the concept itself as well as the nature of the military threat posed by North Korea, especially its important asymmetric and nuclear aspects.

Like so many other things about Mr. Trump, it is difficult to assess whether he lacks fundamental knowledge about foreign policy and national security — his failure, for example, to take account of the fact that South Korea and Japan are non-nuclear weapon states under the 1970 Nuclear Nonproliferation Treaty and have permanently foresworn the development and/or acquisition of nuclear weapons (and the United States’ pledge not to help them do so) — or whether he simply opts for provocative statements to press home larger points. It admittedly requires generosity to grant that it is the latter. Mr. Trump habitually takes complex subjects — extended deterrence’s continued relevance in East Asia, given the geopolitical changes there over the past thirty years — and distills them to a simplistic rhetorical frame. He uses this frame to justify preconceived beliefs — like, for example, the one underlying his rhetorical question, “how long will we go on defending South Korea from North Korea without payment?”[18]

No concept including extended deterrence should be sacrosanct nor treated as immutable. After all, extended deterrence has changed almost constantly in composition — responding to the ever-evolving political and security environments — and application — it was not, for example, United States declared policy in Japan until 1968 — since first adopted.[19] The dilemma is that Mr. Trump so often fails to find the optical center. He is ineluctably drawn to the simplicity of what he no doubt sees as “realist” positions, which for him translate easily into compelling political rhetoric. A parallax view and depth of perception are essential to new knowledge about anything.[20] His pronouncements about South Korean and/or Japanese indigenous nuclear arsenals come across as obtuse and one-dimensional.

The casual, bordering on offhanded way in which Mr. Trump challenges orthodoxies infuriates his many critics and endears him to supporters. When he avoids the incendiary, it sometimes serves a larger purpose, whether intended or not. He is criticized for challenging such bedrock principles of the alliance system as interdependence, but its weak effects were recognized long before Mr. Trump came on the scene.[21] Moreover, it is hard to dispute that the United States’ postwar alliances were built on a pattern of asymmetric interdependence[22] that is in many places today outmoded. The formerly dependent nations of post-war Western Europe today are mere free riders, and the leverage that once came from the United States bearing a disproportionate burden of the common defense has gone away or been greatly diminished.

Deferring judgment on whether the resultant policy pronouncements are sound, Mr. Trump does a service by challenging cherished shibboleths during this, an election year. For this Mr. Trump’s critics relentlessly portray him as several degrees removed from mainstream thinking on foreign policy and national security matters. The New York Times wrote “Mr. Trump explained his thoughts in concrete and easily digestible terms, but they appeared to reflect little consideration for potential consequences.” So consider the following statement:

“For Europe today it is not unfair to pose the matter thus: If NATO is the answer, what is the question? For East Asia a parallel point can be made: If the U.S.-Japanese Mutual Security Treaty is the answer, what is the question?”[23]

Had it come from the mouth (or keyboard) of Mr. Trump, it would have earned the opprobrium of many self-proclaimed members of the “Republican national security establishment”.[24] Instead, the highly respected scholar of national security policy and strategic theory, Dr. Colin S. Gray, penned it two decades ago.

So we will endeavor here to distill the message from the messenger.

This essay addresses (albeit briefly) the main arguments of the protracted debate over alliances and extended deterrence to test a larger point: whether defenders of the nuclear status quo — that oddly-paired animadversion to nuclear aspirations[25] and accedence (if reluctant) to the stabilizing effect of existing nuclear states (at least before North Korea became one) — have, like Hans Christian Andersen’s metaphorical Emperor, no clothes. This is not Anderson’s tale of skillful deceit but instead, one about the seductiveness of a rarely challenged orthodoxy. It is the paradoxical notion that while an emergent nuclear force destabilizes, once emerged, a nuclear state is apposite to virtuous “stability” and “deterrence”. Kenneth Waltz put it this way:

“Much of the writing about the spread of nuclear weapons has this unusual trait: It tells us that what did not, happen in the past is likely to happen in the future, that tomorrow’s nuclear states are likely to do to one another what today’s nuclear states have not done. A happy nuclear past leads many to expect an unhappy nuclear future.”[26]

So, would Mr. Trump’s openness to a South Korean and/or Japanese nuclear counterstrike force serve the interest of deterrence and stability, or undercut it?

The hoped-for steady-state system — one whose properties remain unchanged over time — is illusory. As Hermann Joseph Muller wrote (and is quoted here with unintended irony), “There is no permanent status quo in nature; all is the process of adjustment and readjustment, or else eventual failure.”[27] Put another way, “the nuclear challenge to our security does not reduce to a simple menace.”[28]

That it is an illusion reflects a fundamental truth. Since Hiroshima, nuclear weapons have, to a far greater degree, proliferated vertically as existing nuclear states added to their arsenals over time than they spread horizontally with the emergence of new nuclear-armed states.[29] Nine nations today admit to possessing, or are thought to possess, nuclear weapons: the United States, Russia, the United Kingdom, France, Israel, Pakistan, India, China, and North Korea. Collectively they possess some 16,000 nuclear warheads, more than 90 percent of which belong to Russia (7700) and the United States (7100).[30]

Staying with the system model, a pathology arises “when theorization becomes a closed system, with no connection through which insights can be applied to the outside world.”[31] Examples of this abound in the realm of nuclear nonproliferation, for example this from David Santoro:

“In the face of Japan’s and South Korea’s nuclearization [sic], the United States should cut them adrift because endorsing their decision (through a UK-like arrangement) or acquiescing to it (à la France) would be untenable. […] Washington should remind its allies that if they broke out of the NPT, they would break up their alliance.”[32]

Even the case for “tolerating” proliferation suffers from this self-same limitation, witness Elbridge Colby’s counterpoint:

“[P]roliferation can be tolerable, in the sense of something that one might not like but can be endured. If proliferation inexorably led to disaster, then Santoro would be right. But it need not. Rather, it can, at least in certain circumstances, be managed.”[33]

The limits of this argument are (or ought to be) self-evident. Richard Betts wrote four decades ago, “As long as anti-proliferation strategy goes no further than schemes to keep the genie in a few bottles, we risk doing both more and less than necessary.”[34] A corollary is that forms of conflict we deem most attractive — in other words, those that play overwhelmingly to our comparative strategic advantage — are unlikely to appeal to our foes.[35] The converse is also true: as Waltz wrote, “the less strategically attractive nuclear weapons appear to the United States, the greater the attraction of those weapons and other WMD to possible foes and other ‘rogues’.”[36]

A more penetrating study of this question is beyond the scope of this essay, so instead the author concentrates on a set of four observations, which he commends to Mr. Trump and his fellow aspirants. The first is what Waltz called the “permanence of the nuclear fact for strategy,” prosaically described as “the clock of strategic history cannot be wound back to an era of pre-nuclear innocence.”[37] The second is what Betts called “the moralist fallacy”:

“The superpowers’ possession of nuclear weapons makes their argument against acquisition by other sovereign states morally hollow, but strategic realities make the have-nots’ arguments against the superpowers irrelevant.”[38] 

The third is what the author dubs the nuclear force reduction paradox. Extended deterrence distorts the design of the American nuclear force relative to what is minimally required to defend the continental United States. Were the United States to contract its extended deterrent and reduce the span of its nuclear umbrella, the argument goes, it could substantially reduce its nuclear force. However, one argument for extended deterrence is that it makes it more dangerous for aspiring nuclear states to cross over rather than to remain just below the nuclear threshold, since crossing triggers the possibility of preemption. Contracting the extended deterrent risks a set of perverse effect. One is that while the United States’ capacity to preempt a nuclear threat is likely undiminished were it to reduce the size of its nuclear force, a smaller American nuclear force perforce increases the political status of small nuclear forces, e.g., North Korea’s (or as the author speculates elsewhere, Iran’s). Another is that while North Korea admittedly was imperfectly deterred by extended deterrence, contracting the extended deterrent sends entirely the wrong signal to the Pyongyang regime.

Betts points out another perverse effect: “Security incentives for getting a bomb are less likely to flow from fear of superpower intervention than from fear of non-intervention.”[39]. Two points are worth emphasizing. First, the effect of American extended deterrence is to deter the use of nuclear weapons against an ally. It is not so much whether a determined adversary, e.g., North Korea, develops a nuclear strike force. Second, the proliferation effect where an ally fears the United States will not intervene in its defense is arguably greater than extended deterrence’s nonproliferation effect on a determined adversary. This of course begs the question whether South Korea would denounce its nonproliferation status and develop an indigenous nuclear force  (or attempt to do so covertly).

The fourth point is borrowed from Colin Grey. He writes regarding the future relevance of extended deterrence:

“The military, strategic, and hence even political effect that once could flow only from nuclear capabilities in the future will be provided by non-nuclear military means. Precise conventional destruction or disablement will, indeed already can, replace some of the desired effects of nuclear use.”[40]

He continues in a paragraph heavy with implications for its usefulness in conflict avoidance in the future:

“[T]he future of warfare will not register conflicts of kinds to which nuclear (or other) weapons of mass destruction will be militarily relevant. On this claim, the great wars of the twenty- first century will see widespread and hugely savage combat, but that combat will occur at very close quarters in ever more urban geography. The future of warfare, in this view, though savage indeed, is unlikely to have the character of the great struggles of the Westphalian period of modern history, which is to say of state against state.”[41]

Problems worthy of attack
prove their worth by hitting back.
                    -Piet Hein

Mr. Trump’s musings to The New York Times on the future of extended deterrence in East Asia and its implications for non-proliferation are important if lacking full and proper context, and despite being spun down to an apothegm.

Deterrence requires the United States to maintain an element of ambiguity with respect to the nuclear umbrella, one result of which is that adversaries like North Korea continuously probe to ascertain (and some might argue, to extend) the limits of American inaction. Another effect is that ambiguity leaves nations that rely on the American nuclear umbrella uncertain as to the specific conditions under which the United States may be willing to defend their country if attacked. In both instances the presumed necessities of American nuclear ambiguity are for some a starting point to argue for an indigenous nuclear capability, since South Korea (and Japan) would most likely embrace an unambiguous, declarative policy in which each stated conditions under which they would use nuclear weapons in self-defense. Ambiguity for them would serve no purpose other than to obscure the clarity of their warning to potential adversaries.

For the United States, extended deterrence — for which nonproliferation remains a primary motive[42] — drives the design of its nuclear posture far more than its own defense requirements. Moreover, especially in the case of America’s NATO allies and Japan — but increasingly, South Korea, too — extended deterrence no longer has the effect of avoiding the situation in which development of an indigenous nuclear capability would divert resources from the development and maintenance of conventional capabilities. In the case of many NATO allies, that diversion occurred regardless. In the case of South Korea and Japan, it is implausible that development of an indigenous nuclear force would meaningfully crowd out investment in conventional defense.

Time has eroded many once salient arguments used to bolster the case for extended deterrence. Does that mean Mr. Trump is correct? No, the author believes, it does not. Few East Asian nations are capable of developing an indigenous nuclear weapons program, and none save North Korea (which denounced its nonproliferation pledge in 2003) can do so without exiting the international nonproliferation regime, something that would likely prove highly consequential. This nevertheless makes for a less easy scenario than if only superpowers possessed nuclear weapons, or for that matter, if every nation did. We return to Kenneth Waltz:

“So long as the system is one of fairly small numbers, the actions of any of them may threaten the security of others. There are too many to enable anyone to see for sure what is happening, and too few to make what is happening a matter of indifference.”[43]

Uncertainty as to what is happening means that fear and worst-case scenarios rule the day. It is true that nuclear weapons are now joined in the arsenal of the United States (and to a lesser degree, China and Russia) by conventional systems with strategic effect. That is not to say, however, that nuclear and conventional weapons with comparable strategic effect necessarily have equal salience when it comes to deterrence. The author claims otherwise: the destructive near-equivalency of nuclear and conventional weapons does not translate into equal deterrent effect, at least not yet.

Nuclear weapons states have long claimed their franchise confers stability. If that argument ever made sense, it stopped doing so when North Korea acquired its nuclear weapon capability. Mr. Trump’s receptiveness to setting aside extended deterrence is puzzling: while it would no doubt have some effect on the configuration of American strategic forces, it is unlikely to be a disfiguring or greatly economizing one. Thus contracting the nuclear umbrella so that it no longer covers East Asian allies is unlikely to carry with it great savings. Mr. Trump seems to believe that in the face of an already-nuclear North Korea, an indigenous South Korean or Japanese nuclear force would simply substitute an in-region deterrent for an out-of-region one. What, however, about China’s reaction? Or Russia’s?

The idea of a nuclear armed South Korea or Japan is (or again, ought to be) intuitively abhorrent to those among us who value nonproliferation. It would effectively accede to an understanding of non-proliferation status as a temporary flag of convenience, valid only until such time as a nation opted to denounce it, as North Korea did a decade ago.

It is also important to recognize that while history does not repeat itself, it does rhyme, as Mark Twain is reputed to have observed. So consider this: a June 2014 analysis by the International Crisis Group observed that “younger Koreans view the issue [of the OPCON transfer] through the lens of sovereignty and the ROK’s modernization”[44] It brings to mind the words of a young Bismarck, who wrote in 1856 that Germany “has regularly adjusted relationships in a radical fashion by warfare, and there are no other means in this present century by which the clock of development can be made to show the correct time.”[45] Let us hope that other means exist today.

While his choice of targets may exasperate critics, Mr. Trump’s penchant for challenging orthodoxies animates an often-stultifying political conversation. It undoubtedly is key to his popular appeal. As the early 20th century German military strategist Joachim von Stülpnagel wrote,  “It will not be enough to content ourselves with time-honored rules based on false assumptions; we must rather seek to think through [these] new conditions.”[46] Whether or not one believes Mr. Trump has thought them through, he is right not to blithely accept rules that may not make sense in the current condition. 

As with India and Pakistan, regional conflicts and local security dilemmas drive small and regional state nuclear proliferation. Given an unpredictable North Korean regime, any suggestion that increased reliance on nuclear threats and bluster will have any effect other than to erode the nuclear taboo — and increase the chance nuclear weapons are used in East Asia — is fanciful. So it is not troubling for Mr. Trump to ask whether extended deterrence has continued relevance there. It is, however, when he answers hastily that it does not.

[1] David E. Sanger & Maggie Haberman (2016). “In Donald Trump’s Worldview, America Comes First, and Everybody Else Pays.” The New York Times [published online 26 March 2016]. Last accessed 27 March 2016.

[2] Kenneth Waltz (1981). “The Spread of Nuclear Weapons: More May Be Better.” Adelphi Papers number 171 (London: International Institute for Strategic Studies).

[3] Mr. Trump’s suggestion he is amenable to an indigenous South Korean nuclear force seems out of step with South Korea’s military deterrence policy. It is built around the “Kill Chain” and a hoped for indigenous preemptive strike system known as the “Korea Air and Missile Defense” (KAMD). An integrated information, surveillance and strike system, the Kill Chain is central to South Korea’s deterrence strategy of “Launch on Warning” to strike preventively the source of a nuclear and/or missile launch if it is thought to be imminent. For an assessment of the Kill Chain, see: “The “Myth” of the Kill, Kill, Kill Chain” available at KAMD is seen by many analysts as a political move to keep all sides happy and forestall a difficult decision. For now, it remains at the conceptual stage, and at best would replicate off-the-shelf United States systems. On 30 March 2016, the South Korean Ministry of National Defense announced a USD28 Billion investment in new weapon systems, stating ” “We will prioritize focusing all of our resources on countering North Korea’s threats (against South Korea) by deploying the Kill Chain and Korea Air and Missile Defense.” See: Seoul to develop new weapons to counter North Korea.” [published online 30 March 2016]. Last accessed 31 March 2016. 

[4] The video can be viewed at

[5] “North Korean propaganda video depicts Washington DC under a nuclear attack and warns ‘US imperialists’ not to ‘budge an inch toward’ the country.” The Daily Mail [published online 26 March 2016]. Last accessed 31 March 2016.

[6] “Pentagon Confirms New North Korean ICBM: KN-14 missile increases nuclear threat to United States.” The Washington Free Beacon [published online 31 March 2016]. Last accessed 31 March 2016.

[7] Last accessed 31 March 2016.

[8] Peacetime OPCON, which had been under United States control since July 1950, was transferred back to the ROK in 1994. It is worth noting that the South Korean government has postponed OPCON’s planned termination three times since 2007, hoping to keep the United States engaged directly on the Korean peninsula as a deterrent to North Korean offensive actions.

[9] ” Seoul vows relentless retaliation against N.K. if provoked.” The Korean Herald [published online in English 8 June 2014]. Last accessed 30 March 2016.

[10] Richard C. Bush III, Vanda Felbab-Brown, Martin S. Indyk, Michael E. O’Hanlon, Steven Pifer & Kenneth M. Pollack (2010). U.S. Nuclear and Extended Deterrence: Considerations and Challenges (Washington, D.C.: Brookings Institute) 29-30.

[11] For an overview of the BMDS, see: Last accessed 29 March 2016.

[12] The First Island Chain is described by an imaginary line running throughj the Kurile Islands, Japan and the Ryukyu Islands, the Philippines, and the Indonesian archipelago.

[13] Fore a detailed discussion, see: Sukjoon Yoon (2015). “Are China’s THAAD Fears Justified?” The Diplomat [published online 20 February 2015]. Last accessed 29 March 2016.

[14] The final observation is credited to Waltz (2000). “Structural Realism after the Cold War.” International Security. 25:1, 5.

[15] Anthony H. Cordesman & Ashley Hess (2013). The Evolving Military Balance in the Korean Peninsula and Northeast Asia. Volume II: Conventional Balance, Asymmetric Forces, and US Forces (Lanham, MD: Rowman & Littlefield) xii. Last accessed 29 March 2016.

[16] The two terms used to describe the firing of missiles in response to an anticipated or actual initiation of attack by an adversary are launch on warning (LOW) and launch under attack (LUA). Technically, LOW could require an attack once indications of an impending attack were received whether or not the missiles had actually been fired, whereas LUA describes only the situation in which it was confirmed that missiles had actually been fired. The United States has always viewed LOW as high risk given that warning system errors could lead to the accidental start of a nuclear war.

[17] For example, a May 1958 assessment by a United States State Department official concluded that only if the Soviet Union launched a non-nuclear attack, or if Western intelligence had advance warning of a Soviet nuclear strike would there be time for the United States to consult with allies ahead of using nuclear weapons. See: Last accessed 29 March 2016.

[18] “From the Desk of Donald Trump: South Korea.” YouTube [posted online 10 April 2013]. Last accessed 29 March 2016.

[19] Waltz (2000), op cit., 34. In 1982, the Japanese government stated publicly its view that the United States extended deterrent included the possibility of first use of nuclear weapons in response to an attack on Japan involving only conventional forces

[20] The author credits this observation to Bruce Cumings (2002). Parallax Visions. Making Sense of American-East Asian Relations (Durham, NC: Duke university Press) 4 .

[21] For example, see Waltz (2000), op cit., especially pp. 14-16.

[22] The term “asymmetric interdependence” was coined by Robert O. Keohane and Joseph S. Nye. See Keohane & Nye (1989). Power and Interdependence, 2d ed. (New York: Harper-Collins).

[23] Colin S. Gray (1999). The Second Nuclear Age (Boulder: Lynne Rienner Publishers, Inc.) viii.

[24] The term belongs to the signatories of the “Open Letter on Donald Trump from GOP National Security Leaders” dated 2 March 2016, each of whom committed themselves ” to working energetically to prevent the election of someone so utterly unfitted to the office [POTUS].” See: Last accessed 29 March 2016.

[25] Keith B. Payne calls this “the ‘nuclear taboo’,” which he characterizes this as “the emerging global ‘norm’ against nuclear weapons.” See: Keith B. Payne foreword to Gray (1999), op cit., viii.

[26] Kenneth Waltz. See Waltz (1981). “The Spread of Nuclear Weapons: More May Be Better.” Adelphi Papers number 171 (London: International Institute for Strategic Studies).

[27] Quoted in Tracy M. Sonneborn (1968). “H.J. Muller, crusader for human betterment.” Science. 162:3855 (15 November 1968) 772-776.

[28] Waltz (1981), op cit.

[29] The juxtaposition of vertical and horizontal proliferation is credited to Waltz (1981), op cit.

[30] These figures include deployed and stockpiled nuclear warheads, as well as retired warheads that remain intact. Figures for the other nuclear-armed states are as follows: France (300); China (260); United Kingdom (225); Pakistan (120); India (120); Israel (80); and North Korea (8). Source: United States State Department (2015) cited in “Nuclear Weapons: Who Has What at a Glance.” Arms Control Agency [published online 23 June 2014]. Last accessed 27 March 2016.

[31] Richard K. Betts (1997). “Should Strategy Survive?” World Politics. 50:1 (October 1997) 31.

[32] David Santoro (2014). “Will America’s Allies Go Nuclear?” The National Interest [published online 30 January 2014]. Last accessed 28 March 2016.

[33] Elbridge Colby (2014). “Choose Geopolitics Over Nonproliferation.” The National Interest [published online 28 February 2014]. Last accessed 28 Marcxh 2016.

[34] Richard K. Betts  (1977). “Paranoids, Pygmies, Pariahs & Nonproliferation.” Foreign Policy. 26 (Spring, 1977), 157

[35] Gray (1999), op cit., 13.

[36] Waltz (1981), op cit.

[37] Ibid.

[38] Betts  (1977), op cit., 157.

[39] Ibid., 158.

[40] Gray (1999), op cit., 3.

[41] Ibid.

[42] James Schlesinger made this case in May 2009 when he reported the recommendations of the Commission on Strategic Posture to the House Armed Services Committee. See:[1].pdf. Last accessed 29 March 2016.

[43] Waltz. (1981), op cit.

[44] Daniel Pinkston (2014). “U.S.-ROK Alliance Management: OPCON Transition and ISR.” Published online by the International Crisis Group 18 June 2014. Last accessed 29 March 2016.

[45] Bismarck to Otto von Mantauffel, 26 April 1856. In Wolfgang Windelband & Werner Frauendienst, eds. Otto von Bismarck: Die Gesammelten Werke (Berlin: Deutscheveralgsanstalt). Quoted in William Carr & Harry Hearder (1991). The Wars of German Unification 1864-1871 (London: Routledge) 58.

[46] Quoted in Michael Geyer (1980). Aufrüstung oder Sicherheit: Die Reichswehr in der Krise der Machtpolitik 1924-1936 (Wiesbaden: Steiner) 81.