Networked Nonproliferation
Making the NPT Permanent
Michal Onderco



Why Indefinite Extension?

THAT THE INDEFINITE EXTENSION of the NPT arose as an issue in the mid-1990s was a result of the terms of the treaty as well as of the broader history of the nuclear nonproliferation regime since the NPT’s entry into force in 1970. Therefore, in an effort to answer the question this chapter addresses, four related questions must be addressed: Why was the treaty of limited duration in the first place? What shape was the regime in twenty-five years after its inception? Were there were any serious alternatives to indefinite extension? and, Why did the US emerge as the leader of indefinite extension?

The extension conference was convened because Article X.2 of the treaty stipulates that

Twenty-five years after the entry into force of the Treaty, a conference shall be convened to decide whether the Treaty shall continue in force indefinitely, or shall be extended for an additional fixed period or periods. This decision shall be taken by a majority of the Parties to the Treaty.1

This original limit to the duration of the NPT was inserted into the treaty to accommodate the European allies of the US, who felt they had been targeted and straightjacketed by the treaty. These countries originally insisted on limiting the duration of the NPT as an insurance against possible adverse future developments. By the time of the conference, however, these countries were not the ones opposed to the extension.

Opponents of the extension questioned more closely whether the NPT had been successful in advancing the state of global nuclear disarmament—one of the chief goals of the treaty—than whether it was successful in preventing the spread of nuclear weapons.2 Although the treaty attracted numerous new signatories—by 1995 it was signed by almost all countries in the world save for a handful of exceptions—critics claimed that it achieved too little to lead to the goal of a world without nuclear weapons, which the nuclear-weapon states had agreed to in Article VI of the treaty. To keep the nuclear states’ feet to the fire, three alternatives had been proposed in line with the wording of Article X. 2 above: a green-light rolling extension, a red-light rolling extension, and a conditional extension.

As mentioned above, by 1995 the European allies of the US had dropped their opposition to indefinite extension, with the result being that the mantle of the opposition was carried by the Non-Aligned Movement (NAM). As the NAM largely acts as a bloc, the fear was real that it would block the NPT’s indefinite extension.3 Thus, from early on, the US emerged as a champion of indefinite extension, in line with its original designs for the treaty.

This chapter will discuss the origins of the limited duration of the NPT, the review conferences (RevCons) until 1995, and the alternatives to indefinite extension, as well as a description of how and why the US became the chief champion of the treaty’s indefinite extension.

Limited Duration

Time-limited duration of international treaties is not unusual. According to Barbara Koremenos’s study, about half of international security treaties are of limited duration, with the average period being five years.4 Why do states put these provisions in the treaties? Because of uncertainty.

Two sources of uncertainty for states predominate in international politics: first, stemming from the international system itself, and second, regarding the distribution of benefits from international treaties. International relations scholars’ starting point for analysis of international cooperation is that there is no higher authority in the international system beyond states, a feature labeled as “anarchy.” Because of the anarchic nature of the international system, states have no assurance of reciprocation or cooperation. Without enforcement of cooperative agreements, uncertainty makes cooperation difficult. Even if states can trust their counterparts’ motivations and willingness today, they have no way of knowing what will happen in the future.5 At the beginning of an agreement, states have a poor understanding of who will benefit from cooperation in the future, therefore they are unwilling to tie their hands for an indefinite term. In the language of Barbara Koremenos, “when uncertainty about the state of the world is high, flexibility is more likely to be incorporated.”6 In the case of the NPT, this is reflected in the fact that the more severe security threat a state faces (or perceives to face), the less likely it has been to sign the treaty.7

At the same time, and relatedly, the uncertainty about the distribution of benefits from international treaties is also tethered to the uncertainty about the state of technology. In treaties such as the NPT, the institutional measures put in place end up governing much of the technology. Thus there is uncertainty about whether states will lose out on the technological end of the bargain, which would be detrimental not only to their security but also to their economic well-being. For this reason, states are unwilling to enter into long-term, inflexible agreements.

Have these concerns been reflected in the NPT discussions as well? It turns out, remarkably, yes. The original limited duration of the treaty—the twenty-five years stipulated in Article X. 2—was included at the express demand of the US’s European allies, which were concerned about the consequences of the treaty on their security and their economies.

The initial draft of the treaty, submitted in 1965 by the US, spoke of its indefinite duration, and the initial Soviet submission in 1967 spoke of an “unlimited duration.”8 The initial US assumption was that the provision for review conferences would appease the opponents of indefinite duration. The US background paper for the February 1967 meeting of the Eighteen-Nation Disarmament Committee (ENDC) stated:

The non-nuclear countries will probably object to signing a non-proliferation treaty of unlimited duration containing no obligations upon the nuclear-weapon powers to halt the nuclear arms race within any specified period. The review clause, however, provides for an automatic conference at the end of five years to review the operation of the treaty “with a view to assuring that the purposes and provisions of the Treaty are being realized.” We have advised some of our allies that at this conference the non-nuclear-weapon states can call upon the nuclear-weapon states to show whether the treaty is living up to its stated purpose of leading toward the easing of tensions and the facilitating of disarmament rather than merely being a step to preserve nuclear monopoly.9

However, opposition to indefinite duration was palpable. In a bilateral meeting with ACDA’s assistant director Samuel de Palma in January 1967, Germany’s foreign ministry’s director of planning Gunther Diehl expressed “serious concern . . . that the treaty would further restrict the capability of Germany industry [sic] to stay abreast of modern technology.”10 Mohamed Shaker wrote in his definitive study of the NPT’s negotiation, citing a Swiss submission to the ENDC, that “such a commitment seemed hardly conceivable in a field where development was as rapid and unpredictable as that of nuclear science and its technical, economic, political and military implications.”11 The Italian representative to the conference Roberto Caracciolo famously said that “[t]o imprison [future generations] in an iron corset, which could not be adjusted to the changing conditions of history, would in our opinion expose that corset to the danger of bursting.”12

Italy, therefore, submitted a formal proposal that would see the NPT as valid for a duration of X number of years, after which the treaty should be automatically extended for the same period unless states withdraw from the agreement subject to a six-month notice.13 That Italy lifted this mantle was not accidental—as Italian historian Leopoldo Nuti writes, the Italian reaction to the initial US draft of the treaty was that of “visceral resentment.”14 As Nuti explains, Italy was concerned both about the consequences of the NPT for European political cooperation and, more importantly, about the regional balance of power and security in Europe. Italian leaders held close discussions with Germany but also with Asian powers such as Japan and India. The Italians demanded that the treaty introduce a measure of flexibility, either by changing the duration or by introducing a possibility for amendments. The possibility for withdrawal included in Italy’s proposal was, in the view of the Italian delegation, the escape clause in case of especially grave circumstances. Italy, however, preferred to have different governments express themselves on this point more explicitly.

Other delegations, particularly the European NATO allies, were similarly not in favor. The Swiss delegation sent an aide-memoire to the ENDC in late 1967, spelling out that “[t]he non-nuclear-weapon states certainly cannot take the responsibility of tying their hands indefinitely if the nuclear-weapon states fail to arrive at positive results in [the direction of limitation of armaments].”15

The Americans tried to engage the Soviets bilaterally on this issue. However, the Soviets rejected the idea that the NPT parties would be able to get out of the treaty easily after the initial period of duration.16 For this reason, the reference to the potential easy withdrawal at the end of the initial period was dropped, and the period of “X years” was replaced in the final joint treaty draft by a period of “twenty-five years.” Shaker argues that this duration reflected the fact that the nuclear era was, at that time, barely twenty-five years old.17

The period was not warmly welcomed. Germany preferred a period of five to ten years, India and Brazil decried legitimation of vertical proliferation, Tanzania lamented that developing countries were left in “comparative technological backwardness for twenty-five years.” Italy tried smuggling the easy-escape clause back in.18 Canada and the US heavily criticized Italy for this step; Dean Rusk himself wrote an agitated note to Italy’s then foreign minister Amintore Fanfani arguing that the US had accommodated all Italian demands.19 As the withdrawal provisions were dealt with in another portion of the treaty, the Italian easy-escape clause was removed. Italy voted in favor of the treaty draft.

Despite the discussions within the NPT setting (outlined in the next section), the situation among US allies changed over the treaty’s first quarter-century. The explanation for this lies partly in the fact that the US “compensated” the allies for their security concerns by providing conventional military assistance.20 Normative change set in, which made nuclear weapons seem undesirable for a number of European countries, particularly Germany.21 European countries were offered credible assurances, including in the form of external deterrence and nuclear sharing.22 By 1995, the gap in the views of the NPT among the European countries decreased markedly, demonstrated by the progress made at the NPT RevCons.


1. “Treaty on the Non-Proliferation of Nuclear Weapons (NPT).” UN Office for Disarmament Affairs (UNODA), New York, July 1, 1968,

2. Miguel Marin-Bosch, “The 1995 NPT Review and Extension Conference,” Irish Studies in International Affairs 6 (1995): 23–31.

3. William C. Potter and Gaukhar Mukhatzhanova, Nuclear Politics and the Non-Aligned Movement: Principles Vs. Pragmatism (London: Routledge, 2012).

4. Barbara Koremenos, “Contracting around International Uncertainty,” American Political Science Review 99, no. 4 (2005): 549–65.

5. For a paradigmatic realist statement about anarchy and uncertainty, see John J. Mearsheimer, The Tragedy of Great Power Politics (New York: W.W. Norton, 2001). For the most advanced institutionalist theorizing about how institutions accommodate and bind uncertainty, see Barbara Koremenos: “Loosening the Ties that Bind”; “Contracting around International Uncertainty”; and The Continent of International Law: Explaining Agreement Design (Cambridge: Cambridge University Press, 2016).

6. Koremenos, The Continent of International Law, 47.

7. Christopher Way and Karthika Sasikumar, “Leaders and Laggards: When and Why Do Countries Sign the NPT?” (working paper, Research Group in International Security [REGIS], Montreal, 2004); Scott D. Sagan, “Why Do States Build Nuclear Weapons? Three Models in Search of a Bomb,” International Security 21, no. 3 (1996): 54–86.

8. Mohamed Ibrahim Shaker, The Nuclear Non-Proliferation Treaty: Origin and Implementation, 1959–1979, vol. 2 (London & New York: Oceana, 1980), 859. Shaker also engages in some serious exegesis about the difference between the two drafts. The difference, however, is not the central point here.

9. “ENDC Meeting; Geneva, February 21, 1967; Background Paper; Significant Legal Questions Affecting the Non-Proliferation Treaty” (February 11, 1967). US Department of State, FOIA Virtual Reading Room, Case no. M-2020-00304–00304.

10. The formulation is written in de Palma’s memorandum from the meeting; see ACDA, “FRG Views on Non-Proliferation Treaty” (January 26, 1967), US Department of State, FOIA Virtual Reading Room, Case no. M-2020-00304.

11. Shaker, The Nuclear Non-Proliferation Treaty, 859.

12. Ibid., 860.

13. George Bunn and Charles Van Doren: Options for Extension of the NPT: The Intention of the Drafters of Article X.2, PPNN study no. 2 (Southampton, UK: Programme for Promoting Nuclear Nonproliferation, 1991) and Two Options for the 1995 NPT Extension Conference Revisited (Washington, DC: Lawyers Alliance for World Security, 1992); Shaker, The Nuclear Non-Proliferation Treaty.

14. Leopoldo Nuti, “‘A Turning Point in Postwar Foreign Policy’: Italy and the NPT Negotiations, 1967–1969,” in Negotiating the Nuclear Non-Proliferation Treaty: Origins of the Nuclear Order, ed. Roland Popp, Liviu Horovitz, and Andreas Wenger (Abingdon, UK: Routledge, 2017), 77–97.

15. Shaker, The Nuclear Non-Proliferation Treaty, 861.

16. Bunn and Van Doren, Two Options.

17. Shaker, The Nuclear Non-Proliferation Treaty, p. 862 (vol. 2).

18. Ibid.

19. Nuti, “‘A Turning Point in Postwar Foreign Policy,’” discusses the note written by Rusk to Fanfani; Shaker, The Nuclear Non-Proliferation Treaty, discusses the US and Canadian reactions within the ENDC.

20. Jennifer Erickson and Christopher Way, “Membership Has Its Privileges: Conventional Arms and Influence within the Nuclear Nonproliferation Treaty,” in The Causes and Consequences of Nuclear Proliferation, ed. Matthew Kroenig, Erik Gartzke, and Robert Rauchraus (New York: Routledge, 2011), 32–60; Verdier, “Multilateralism, Bilateralism, and Exclusion.”

21. Harald Müller and Wolfgang Kötter, Germany, Europe & Nuclear Non-Proliferation, PPNN study no. 1 (Southampton, UK: Mountbatten Centre for International Studies, 1991); Harald Müller, “The Internalization of Principles, Norms and Rules by Governments: The Case of Security Regimes,” in Regime Theory and International Relations, ed. Volker Rittberger (Oxford: Clarendon, 1993), 361–90.

22. This applied not only to NATO allies but also to other countries. See, for example, on Sweden, the work of Thomas Jonter in his The Key to Nuclear Restraint: The Swedish Plans to Acquire Nuclear Weapons During the Cold War (London: Palgrave, 2016).