Assessing Denuclearization in the Post-Cold War Era

Initial Findings and Implications

Story Stream
recent articles

Counter-proliferation and denuclearization of aspiring and especially adversarial, nuclear states have long been a top U.S. national security priority.[i]  However, the long-term consequences of denuclearization is something that is rarely studied, although volumes have been written about the strategies that U.S. leaders employ to prevent the spread of, and occasionally rollback, nuclear weapons.  While the benefits of denuclearization are obvious for the U.S., it is not evident that this policy benefits the security of denuclearized states.  In fact, realist theory predicts that in a self-help world, losing a means of self-defense, such as nuclear weapons, will increase the vulnerability of the denuclearized state, at least when the underlying security threat that made proliferation attractive in the first place is not mitigated.  This paper quantitatively analyzes the cases of denuclearization in the post-Cold War era and demonstrates that denuclearized states are more likely to be involved in a war than states that did not denuclearize.

Defining Denuclearization

This research project relied primarily on the Uppsala Conflict Data Project (UCDP) datasets for state-based conflict to test the hypothesis that denuclearized states were more likely to be the victim of war.  The UCDP dataset contains hundreds of conflicts, but only 18 new, discrete conflicts between 1991-2015 reached an intensity (the commonly accepted definition of > 1,000 combat fatalities in a year) sufficient to be coded by UCDP as a “war.”[ii]  There is no existing definition of denuclearization, despite this being a stated goal of U.S. policy towards several states.  I, therefore, defined denuclearization as a state losing, either voluntarily or by force, its nuclear weapons program or the entirety of its nuclear inventory.  I generated a new variable to account for denuclearization status within the UCDP dataset.

Leveraging existing research into non-proliferation shows that from 1991 to 2015 nine states met this definition of denuclearization, shown graphically below.  If they were subsequently involved in a war they are coded orange, and if they were not attacked they are coded blue.  At first glance, four of nine denuclearized states were involved in wars during this period.  This suggests denuclearized states did not experience improvements in their security.[iii]

Four of these states gave up their nuclear weapons inventories.  South Africa is the most commonly cited successful case of denuclearization as following the end of the Cold War and the collapse of the South African apartheid regime, they voluntarily dismantled their nuclear weapons stockpile and entered the non-proliferation regime.[iv]  The other three states to surrender their nuclear weapons inventories, Kazakhstan, Ukraine, and Belarus were grandfathered into nuclear status when the Soviet Union collapsed and voluntarily surrendered their nuclear weapons in the immediate post-Cold War era.[v]

The remaining five states gave up their nuclear weapons developmental programs as a result of international pressure or preventive attacks.  The most prominent of these was Iraq, whose Weapons of Mass Destruction (WMD) programs were a basis for the 2003 U.S.-led invasion.  However, post-invasion evidence revealed that while Iraq was actively trying to build a nuclear weapon during Desert Storm, the 1991 attacks crippled its program and it effectively ceased at that point.[vi]  The U.S. invasion of Iraq in 2003, in conjunction with international sanctions, contributed to Libya’s 2003 decision to cease its nuclear weapons program and surrender its Uranium enrichment equipment.[vii]  The last three denuclearizers (Syria, Algeria, and Romania) had very nascent programs and, in the cases of Romania and Algeria, lacked any obvious reason to pursue nuclear weapons.  All three ceased nuclear weapons activities after a short developmental effort and were relatively “weak” denuclearizers based on the lack of serious programs.

Quantitative Analysis

Quantitative analysis of this data shows a statistically significant relationship between denuclearization and incidence of war.  No such relationship was identified between other national military or material capabilities and the incidence of war.[viii]  As a matter of odds, denuclearized states were much more likely to be involved in a war in any given year than were non-denuclearized states (as shown below).  In the end, this analysis statistically demonstrates that in the post-Cold War era denuclearized states are more vulnerable than non-denuclearized states. 

Implications of historical denuclearization

The results of this analysis are troubling for the future of the U.S.-led nonproliferation regime.  Actual and potential regional nuclear powers likely understand the vulnerability associated with denuclearization.  While Iran has temporarily suspended its nuclear program under the Joint Cooperative Plan of Action (JCPOA), it has not yet denuclearized and can resume its weapons program when the JCPOA expires (or is otherwise terminated).  Most recently, North Korea appears insensitive to Western pleas and threats for denuclearization.

Furthermore, in all four cases of denuclearized states being attacked, nuclear powers were either the aggressor or at least heavily involved.  The U.S. itself attacked two denuclearized states, Iraq in 2003 and Libya in 2011, in both cases ending in regime change.  Russia seized Ukrainian Crimea in 2014 and invaded eastern Ukraine, while the West stood by and did not honor the 1994 Budapest Memorandum security guarantees provided when Ukraine surrendered its nuclear weapons.  Finally, both Russia and the U.S. have become involved in the Syrian conflict.  These examples of negative post-denuclearization consequences, particularly the case of Libya, are likely to disincentivize voluntary denuclearization by additional countries in the future.

The results of this study should not be taken to endorse the “proliferation optimist” school that argues nuclear weapons for everyone results in a more stable situation.  What it does show is that nuclear weapons are effective in mitigating the incidence of wars and denuclearization without resolving the underlying security threat is not productive.  Ultimately, the increased vulnerability of denuclearized states threatens to discredit non-proliferation activities or additional denuclearization.  Additionally, while the results of this study were statistically conclusive, given this dataset’s small sample size more qualitative research is required in individual cases.  In particular, the states that were attacked appear to have never resolved their underlying security threats while states whose security threat evaporated (South Africa) or gained the protection of a major power (Belarus) remained whole and safe.  Further research could help identify the factors that promote denuclearized states’ security over the long-term.


In the end, U.S. foreign policy appears to be working at cross-purposes, and the goals of non-proliferation and denuclearization are undercut by other, perhaps well-meaning, foreign policy initiatives.  As a result of our past foreign policy actions against denuclearized states, it seems increasingly unlikely that pleas, vague threats, non-binding security guarantees, and appealing to the greater good will continue to work as they did in the immediate post-Cold War era – at least in the case of states that face perceived security threats.  If the U.S. is serious about containing and rolling back nuclear weapons, it will find itself making hard choices to either forcibly denuclearize potential proliferators or offer more concrete security guarantees that we have been willing to in the past.  Given these dynamics, it seems likely that in certain cases the U.S. will choose a third option of accepting nuclear proliferation in these limited cases and embrace deterrence over non-proliferation.

Joe Petrucelli is a Ph.D. student at George Mason University, a Senior Analyst at Systems Planning and Analysis, Inc., and a Naval Reserve officer.  The opinions expressed here are his own and do not reflect the positions of his employer, the U.S. Navy or the Department of Defense.


[i] The National Security Strategy of the United States of America, March 2006,, 18; National Security Strategy, February 2015,, 11.

[ii] UCDP/PRIO Armed Conflict Dataset v17.1, Allansson, Marie, Erik Melander & Lotta Themnér (2017) Organized violence, 1989-2016. Journal of Peace Research 54(4).
Gleditsch, Nils Petter, Peter Wallensteen, Mikael Eriksson, Margareta Sollenberg, and Håvard Strand (2002) Armed Conflict 1946-2001: A New Dataset. Journal of Peace Research 39(5), (accessed on 17 June 2017).

[iii] Scott Sagan, “The Causes of Nuclear Weapons Proliferation,” Annual Review of Political Science (2011: 14) C-2; Alexandra Debs and Nuno P. Monteiro, Nuclear Politics:  The Strategic Causes of Proliferation (New York:  Cambridge University Press, 2017), 71-73.

[iv] Peter Liberman, “The Rise and Fall of the South African Bomb,” International Security 26: no. 2 (2001): 74-75.

[v] Rebecca Hersman, “Strategic Challenges to WMD elimination,” The Nonproliferation Review 23: no. 1-2 (2016): 32.

[vi] Charles Dulfer, “WMD elimination in Iraq, 2003,” The Nonproliferation Review 23: no. 1-2 (2016): 177-178.

[vii] Norman Cigar, Libya’s Nuclear Disarmament:  Lessons and Implications for Nuclear Proliferation, Middle East Studies at Marine Corp University Monograph No. 2 (January 2012), 2.

[viii] National Material Capabilities dataset v5.0, Singer, J. David, Stuart Bremer, and John Stuckey. (1972). "Capability Distribution, Uncertainty, and Major Power War, 1820-1965." in Bruce Russett (ed) Peace, War, and Numbers, Beverly Hills: Sage, 19-48, (accessed on 16 June 2017); Correlates of War Project, 2017, "State System Membership List, v2016," (accessed on 16 June 2017).

Show comments Hide Comments