|Poor Man’s Atomic Bomb? Chemical and Biological Weapons Proliferation in an Era of Deep Nuclear Reductions
In his September 10, 2013 address to the nation on Syria, President Barack Obama appealed for the support of the American people in responding to the Assad regime’s use of chemical weapons with a targeted military strike. Over the course of his fifteen-minute speech, the President laid out reasons for why the use of military force against the Syrian government was in the security interests of the United States of America.
If we fail to act, the Assad regime will see no reason to stop using chemical weapons. As the ban against these weapons erodes, other tyrants will have no reason to think twice about acquiring poison gas, and using them. Over time, our troops would again face the prospect of chemical warfare on the battlefield…And a failure to stand against the use of chemical weapons would weaken prohibitions against other weapons of mass destruction, and embolden Assad’s ally, Iran – which must decide whether to ignore international law by building a nuclear weapon, or to take a more peaceful path. This is not a world we should accept. This is what’s at stake.
Among the many reasons to support a strike given by the President, the last one should be particularly interesting to policymakers and scholars who study the causes and consequences of weapons proliferation. The Obama administration asserts that a failure to sanction the use of chemical weapons would weaken prohibitions against other weapons of mass destruction in general, and embolden Iran to ignore international law by building a nuclear weapon. This is a powerful claim, both for what it assumes and for what it implies. But is it true?
In linking the failure to act against the Syrian chemical weapons program with the risk of nuclear proliferation in Iran, the President joined a sizeable group of policymakers and security scholars who have either explicitly or implicitly proposed an underlying relationship between the pursuit or acquisition of one type of “weapon of mass destruction” and the pursuit or acquisition of another. In this case, the President suggests that states may employ a similar calculus to evaluate the utility of pursuing and acquiring different weapons of mass destruction, such that removing the costs imposed on states for pursuing chemical weapons may increase the proliferation risk of nuclear weapons. If true, this has powerful implications for how we understand the causes and consequences of WMD proliferation, particularly if we seek to identify proliferation risk before it materializes. But is there any evidence that political and military leaders treat nuclear, chemical and biological weapons as a similar class of weapons, either in how they evaluate the decision to pursue their acquisition or in the way they use them?
The Obama administration is certainly not alone in grouping – and thereby implying a link – between different weapons of mass destruction. Indeed, more generally, the popular use of the term “weapons of mass destruction” itself can be understood to imply a relationship between nuclear, chemical and biological weapons proliferation in so far as it assumes that the separate weapons technologies can be usefully grouped into a single analytic category. But is this category useful, or does the term “weapon of mass destruction” actually obscure more than it clarifies (Lavoy, Sagan, and Wirtz 2000)?
To further illustrate, consider the following two examples. In his 2002 State of the Union Address, President George W. Bush famously characterized an “Axis of Evil” and noted that, “by seeking weapons of mass destruction, these regimes pose a grave and growing danger”. Roughly five years later, the Commission on the Prevention of Weapons of Mass Destruction Proliferation and Terrorism warned that it was, “more likely than not that a weapon of mass destruction will be detonated before the end of 2013”. Interestingly, in the first case, the Bush Administration implies that there is a useful analytic category of states that can be treated as distinct in their common demand for any weapon of mass destruction. And yet, there is little systematic evidence to suggest that the demand for nuclear, chemical and biological weapons is correlated empirically, such that it might be effective to base policy on mitigating the pursuit of any WMD. Similarly, in the second case, the Graham/Talent WMD Commission suggests that it is analytically useful to understand the latent risk of a nuclear, chemical or biological weapons attack as the likelihood of a single construct. And yet, there is little empirical evidence to suggest that actors utilize nuclear, chemical, and biological weapons for similar reasons, such that a single, common policy response could be implemented to the mitigate the overarching risk of any WMD attack (Koblentz 2009).
Policymakers and analysts in the non-proliferation community may be quick to argue that very few people actually imbue the term “weapon of mass destruction” with any real substantive meaning. Rather, it is a term of convenience to distinguish a set of non-conventional weapons technologies. But even the most experienced members of the nonproliferation community must be left wondering if there is more at stake than a simple semantic debate after the President’s address to the nation on Syria. As a practical matter, it appears as though leaders are willing to formulate policies based, in part, on the belief that the proliferation and use of these different types of weapons are somehow related.
In this essay, I argue that there is much at stake in correctly characterizing the underlying relationship between different weapons of mass destruction, particularly amid an era of deep nuclear reductions recently outlined by the Obama Administration. To make this case, I begin by providing a very brief history of the term “weapon mass destruction” to describe the increasing use of the term over time, and to highlight the policy implications that may follow from its use. After articulating what is at stake in accurately characterizing this relationship, I then turn to what the historical data can tell us about the relationship between different weapons of mass destruction by summarizing a research design and some empirical results from an ongoing research project involving myself and Michael Horowitz. Next, I focus on what this research cannot definitively tell us, by cautioning academics and policymakers against the temptation to draw premature inferences based on extrapolations from these results to a world with far fewer nuclear weapons. Finally, I conclude with a call for future research to further explore the relationship between “weapons of mass destruction” and the implications that this relationship has for strategic stability.
Are Weapons of Mass Destruction Related?
In contemporary discourse, policymakers and analysts often use the term “weapons of mass destruction” to distinguish a broad class of non-conventional weapons technologies including chemical, biological, radiological and nuclear weapons (CBRN). It is interesting to note, however, that the meaning of the term has evolved dramatically from the time it was first officially used in specific reference to nuclear weapons in the 1946 resolution founding the Atomic Energy Commission, to its current wide-spread usage in US military doctrine to refer to biological, radiological and chemical weapons as well. Perhaps not surprisingly, the increasing scope and application of the term has generated a considerable amount of debate from members of the military and technical communities, who typically differentiate technologies by their strategic purpose and destructive potential. And indeed, the popular use of the term may obscure important differences between the different types of weapons.
Do political and military leaders actually treat nuclear, chemical and biological weapons as substitutes or complements in their overall weapons arsenal? Specifically, is there any evidence that the same factors known to be correlated with the demand for nuclear weapons are also correlated with the underlying demand for chemical and biological weapons? And if so, does possessing (or not possessing) any one type of WMD appear to influence this underlying demand such that states are systematically less (or more) likely to pursue another type of WMD?
The answers to these questions have important implications for nonproliferation and arms control policy. For example, if the evidence suggests that states pursue nuclear, chemical and biological weapons to satisfy the same underlying demand for greater security, it may be the case that international efforts designed to reduce the risk of nuclear proliferation – like the Nuclear Nonproliferation Treaty – could inadvertently shift proliferation incentives towards chemical or biological weapons. In a frequently cited article in Foreign Affairs, Richard Betts (1998) makes a very similar point with respect to the campaign to eradicate chemical weapons, suggesting that the side effects of such a policy may not be entirely benign. Specifically, Betts wonders if international efforts to ban chemical weapons – embodied in agreements like the Chemical Weapons Convention – might only influence the choice of weapon among countries that want some kind of WMD, by simply making one type harder to acquire. “If major reductions in the chemical threat produces even minor increases in the biological threat,” Betts argues, “it will be a bad trade”.
Despite these important implications, academics and policymakers in the nonproliferation and arms control communities still know surprisingly little about the relationship between weapons of mass destruction. Existing research has overwhelmingly focused on the causes and consequences of nuclear, chemical and biological weapons in isolation. For example, we know a great deal about the factors that influence the decision by states to seek nuclear weapons (Solingen, 1994; 1998; 2008; Sagan, 1996; Singh and Way, 2004; Jo and Gartzke, 2007; Hymans, 2006; Fuhrmann 2009; Kroenig, 2009), and what the consequences of nuclear proliferation might be for deterrence and strategic stability in theory (Schelling, 1960 and 1966; Intriligator and Brito, 1981; Berkowitz, 1985; Powell 1990; Sagan 1993; Sagan and Waltz 2012; Brito and Intriligator, 1996) and empirically (Beardsley and Asal, 2009; Fuhrmann, 2009; Gartzke and Jo, 2009; Horowitz, 2009; Kroenig, 2009; Rauchhaus 2009; Narang 2014). However, we know substantially less about the causes and consequences of chemical and biological weapons proliferation. In particular, existing research has yet to determine whether states might seek chemical and biological weapons for the same reasons that they seek nuclear weapons (Koblentz 2009), whether efforts to reduce the supply of one type of WMD could thus increase the demand for another, or if ultimately acquiring one type of WMD appears to influence the risk a state will purse another type.
And yet, understanding the causes, consequences, and underlying relationship between different weapons of mass destruction is just as important – if not more important – today than it has been in the recent past. Consider that in his 2009 Prague Speech, the President vowed to take “concrete steps towards a world without nuclear weapons”, eventually proposing a reduction of strategic nuclear weapons to just over 1000. However, it is not obvious that reducing the availability of nuclear weapons will have benign effects on the risk of chemical and biological weapons proliferation. Meanwhile, as the world celebrates the century long effort to ban chemical weapons by awarding the Organization for the Prohibition of Chemical Weapons (the administering body for the Chemical Weapons Convention) the 2013 Nobel Peace Prize for "defining the use of chemical weapons as a taboo under international law”, it remains to be determined whether these same efforts might shift proliferation to other weapons, either conventional or non-conventional.
What Does the Historical Data Suggest?
In a recent paper published in the Journal of Conflict Resolution (Horowitz and Narang, 2014), Michael Horowitz and I sought to understand the factors that drive the proliferation of weapons of mass destruction across the different types: nuclear, chemical and biological weapons. We asked two straightforward questions: first, are the same factors known to be associated with the proliferation of nuclear weapons also associated with chemical and biological weapons proliferation? And second, do leaders appear to treat nuclear, chemical and biological weapons as substitutes or complements in their overall weapons arsenal? With respect to the second question, we specifically sought to understand what the relationship between different types of WMDs is empirically.
Our approach to answering these questions was decidedly empirical and inductive. That is, rather than engage in a theoretical debate comparing the ease of acquisition and destructive potential across nuclear, chemical, and biological weapons, we chose, instead, to observe historical patterns in the pursuit and acquisition of different WMDs across countries over time to determine whether states appeared to behave as if these weapons were substitutes or compliments. To do this, we estimated something akin to the cross elasticity of demand across WMDs by measuring the impact of pursuing and possessing any one type of WMD on the risk a state will eventually pursue another type, holding that state’s underlying ‘‘willingness’’ to pursue a WMD (demand) constant. In other words, at any given level of demand – which we approximate using a set of control variables that previous research has shown to be correlated with states’ willingness to pursue a nuclear weapon – we tried to estimate the independent effect of acquiring one weapon on the risk a state will pursue another.
To begin, this approach required accurate historical data on nuclear, chemical and biological weapons pursuit and acquisition across time and space. And although there is some emerging consensus around which countries pursued and possessed nuclear weapons over time (Gartzke and Kroenig 2009), there was no previously established data on chemical and biological weapons proliferation. To compile this data, we relied on six different sources: (1) the Stockholm International Peace Research Institute, (2) the Center for Nonproliferation Studies, (3) Arms Control Today, the (4) Carnegie Endowment for International Peace, (5) the Chemical and Biological Arms Control Institute, and (6) the Stimson Center. Fortunately for us, the codlings in these six sources were highly correlated. However, they did not always agree on which exact countries pursed or acquired chemical and biological weapons in any give year. Nevertheless, we were able to confirm the robustness of our results to different sampling rules that required either unanimity across sources, agreement across a majority of sources, or any single source reporting pursuit or possession of a chemical or biological weapon by a country in any year.
The results of our analyses were telling. Specifically, we found that the underlying demand for nuclear, chemical and biological weapons appears to be correlated. That is, many of the same factors that cause states to “go nuclear” also appear to systematically influence the risk that states will seek chemical and biological weapons. With respect to the relationship between different weapons of mass destruction, we found that nuclear, chemical and biological weapons generally appear to function as complements at the pursuit stage, as simply initiating pursuit of any one WMD appears to independently increase the risk that a state will seek all three simultaneously, controlling for other factors. Finally, and perhaps most interesting, we found some evidence that WMDs do function as substitutes in one important fashion: once countries acquire nuclear weapons, they appear far less likely to pursue or possess chemical and biological weapons. That is, the data appears to support the popular notion that chemical and biological weapons function as a “poor man’s atomic bomb”, since acquiring a nuclear weapon appears to satisfy demand and reduce the risk of chemical and biological weapons pursuit, but not vice-versa. This last finding is also remarkably consistent with the idea that nuclear weapons acquisition may uniquely entail some prestige.
Of course, these results are not without their limitations. First, these are systematic empirical regularities estimated across states in the international system over time. There certainly are, however, important historical cases that do not fit these general patterns well. For example, both the United States and the Soviet Union maintained chemical weapons programs for decades after they acquired nuclear weapons. Second, the pursuit and acquisition of weapons of mass destruction are relatively rare events, particularly with respect to nuclear weapons. For this reason, some of our findings may be driven by the behavior of only a handful of states, which could limit the applicability of the findings. Finally, our results are only instructive if the historical data under analysis are accurate. However, because WMD programs are notoriously secret, determining which states actively pursue or possess a nuclear, chemical or biological weapon in any given year is a non-trivial measurement challenge. We were careful to check the robustness of our findings to different datasets and different sampling rules, but this still assumes some independence across measurements. In the end, we emphasized these limitations and encouraged caution in making strong policy inferences based on our results.
What are the Implications of these Results for Strategic Stability amid Deep Nuclear Reductions?
On April 5 2009, President Obama traveled to Prague and delivered a speech outlining America’s commitment to stopping the spread of nuclear weapons. The President’s speech made it clear that the ultimate goal was a world without nuclear weapons. The President also made it clear that this was a long-term objective and one that may not be achievable in his lifetime. However, in the near term, the President laid out a trajectory that the United States should be on in order to accomplish this goal. Among other things, the speech called for reducing the role of nuclear weapons in US national security strategy – and urging others to do the same – by reducing the size of the US nuclear arsenal. Specifically, the President called for the negotiation of a New Strategic Arms Reduction Treaty (START) to reduce the number of missile launchers by half, limit the number of deployed strategic nuclear warheads to 1550, and limit the number of deployed and non-deployed inter-continental ballistic missiles launchers, submarines ballistic missile launchers, and heavy bombers equipped to deploy nuclear weapons to 800. Alongside these reductions, the President also proposed to strengthen the Nuclear Non-Proliferation Treaty as a basis for cooperation, calling for more resources for monitoring and sanctioning violators of the NPT. Four years later in Berlin, the President announced a series of concrete steps towards disarmament that went well beyond even the 2010 New START treaty with Russia, including a reduction of strategic nuclear weapons to just over 1000.
The expected benefits from nuclear reductions were mentioned in the Prague Speech and more clearly articulated in the 2010 Nuclear Posture Review (NPR). According to the 2010 NPR, the “changed – and changing – international security environment” since the end of the Cold War is one in which the “threat of global nuclear war has become remote, but the risk of nuclear attack has increased”. More precisely, the Obama Administration estimated that the greatest risk of nuclear weapons use came from nuclear terrorism and from nuclear proliferation to countries at odds with the United States, its allies and partners, and the international community. In so far as the opportunity for nuclear proliferation and the risk of theft by a terrorist group are likely to increase with the supply of nuclear weapons globally, it is easy to see how even unilateral reductions in nuclear stockpiles by the United States and Russia might lower these risks.
However, there were also potential costs from reductions to the nuclear arsenal laid out in the 2010 NPR. First, reductions might make it difficult to maintain strategic deterrence and stability at reduced nuclear force levels. Second, reductions might weaken U.S. assurances to its allies and partners and thereby strain regional deterrence relationships. And third, a reduction in the number of nuclear weapons may reduce the overall effectiveness of the US nuclear deterrent by reducing redundancy in a post-nuclear testing era. In sum, reducing the size of the US nuclear arsenal has the potential benefits of reducing the seemingly urgent threats of nuclear terrorism and nuclear proliferation, but at the potential cost of strategic stability with existing nuclear powers – most notably Russia and China – and the failure of regional deterrence if assurances to US allies and partners are suddenly incredible.
It is interesting to note, however, that in looking ahead to a world without nuclear weapons, the 2010 NPR was relatively silent about the potential impact that nuclear reductions and a strengthened Nuclear Non-Proliferation Treaty might have on the proliferation risk of chemical and biological weapons. This, despite invoking the umbrella term “weapons of mass destruction” or “WMD” over a dozen times in its roughly 50-pages. The 2010 NPR did list the elimination of chemical and biological weapons as an important objective towards which the United States should direct future efforts, but only “so that over time all states possessing nuclear weapons can be secure in making deterrence of nuclear attack the sole purpose of nuclear weapons”. It did not explicitly consider the possibility that further constraints on the global supply of nuclear weapons might inadvertently increase the demand for chemical and biological weapons among hostile countries and terrorist groups that would suddenly face fewer opportunities to secure nuclear weapons.
So what inferences – if any – can we make from the research described above to the likely impact of deep nuclear reductions on the risk of chemical and biological weapons proliferation? Might policies that limit the supply of nuclear weapons simply shift proliferation risk elsewhere? Even more to the point, could chemical and biological weapons acquire the status of a “poor man’s atomic bomb” and increase in inverse relationship to global nuclear stockpiles?
The short answer to these questions is that we cannot yet know what the likely impact of deep nuclear reductions will be on the risk of chemical and biological weapons proliferation. This is because existing research – including our own study – does not provide the type of empirical evidence needed to forecast these outcomes with any real confidence. To illustrate this, I anticipate four mechanisms through which restrictions in the global supply of nuclear weapons might be posited to increase the risk of chemical and biological weapons proliferation. I then show that each of these inferences is nevertheless unsustainable based on the findings described above.
The first inference that one may be tempted to draw from past findings is that a policy focused on achieving reductions in the global nuclear stockpile could cause chemical and biological weapons to acquire the status of a “poor man’s atomic bomb” and increase as a result. As noted above, our findings suggested that states appear to seek chemical and biological weapons for many of the same reasons as nuclear weapons. Furthermore, our findings also indicated that states that do not possess nuclear weapons appear to be systematically more likely to pursue chemical and biological weapons than states that do possess nuclear weapons. When combined, it may seem reasonable to suppose that, conditional on some level of demand for one of these weapons, reductions in the global supply of nuclear weapons could cause some countries to seek the deterrence and compellence benefits of nuclear weapons elsewhere by shifting to chemical and biological weapons as imperfect substitutes.
A second inference that one may be tempted to draw is that a proposed strengthening of the Nuclear Non-Proliferation Treaty may increase the risk of chemical and biological weapons proliferation. Understood in the terms of our study, policies and institutions designed to monitor and sanction the unilateral pursuit or dissemination of nuclear weapons material and technical expertise – like the NPT or the Nuclear Suppliers Group – might be understood as supply constraints that effectively increase the transaction costs of nuclear weapons acquisition. Furthermore, previous research has shown that the supply of sensitive nuclear assistance (Kroenig 2009) and civilian nuclear assistance (Fuhrmann, 2009; Brown and Kaplow, 2014) are both positively associated with the risk of nuclear weapons pursuit and acquisition across countries and over time. When combined, it may seem reasonable to suppose that – given some demand for a “weapon of mass destruction” – chemical and biological weapons could seem like relatively cheaper pursuits under a more robust global nuclear nonproliferation regime that further regulates the supply of nuclear weapons.
A third inference that one may be tempted to draw is that reductions in the global supply of nuclear weapons and a strengthening of the nuclear nonproliferation regime could increase the risk of chemical and biological weapons pursuit among terrorist groups. If one is willing to assume terrorist groups aim to influence countries like the United States by threatening to impose costs in order to achieve concessions – whether this be through strategies like coercion, provocation, spoiling or outbidding (Kydd and Walter, 2006) – then it may seem reasonable to suppose that limiting the availability of nuclear weapons might shift the demand to other coercive instruments such as chemical or biological weapons.
A final inference that one might be tempted to draw is that unilateral reductions in global nuclear stockpiles could have a second order effect of weakening extended deterrence commitments to nuclear client-states, and thereby increase the risk of chemical and biological weapons pursuit among them. Today, it is widely believed that the US commitment to extended deterrence over South Korea and Japan is partly responsible for their decision to forgo nuclear weapons. However, one explanation for why Great Britain and France chose to develop their own nuclear capabilities 50 years ago – despite their inclusion under the US “nuclear umbrella” – is that the two countries ultimately doubted the credibility of the US commitment to defend Western Europe in the event of a Soviet attack (Schelling 1966). If accurate, it may seem reasonable to suppose that unilateral reductions in the nuclear weapons stockpile of a patron-state may influence the perceived credibility of its commitments to extended nuclear deterrence, and thereby increase the risk that client-states will pursue some kind of WMD to fill an unmet demand for security.
Although each of these fours mechanisms seem like plausible inferences, it is important to note that none of them are sustainable given the available evidence. This is because all four mechanisms are based on empirical findings that are highly contingent. For example, the first and second inferences are based on research designs in which the marginal effects of individual factors are estimated while holding everything else constant across countries over time. As a result, one may only infer that reductions in the overall supply of nuclear weapons could shift proliferation risk to chemical and biological weapons, if – and only if – other factors do not change in anticipation of this effect. If, for example, policy makers are careful to simultaneously lower the availability of chemical and biological weapons (perhaps by strengthening the CWC and BWC as a basis for cooperation), one should not expect to observe substitution. Similarly, although the inferences here are based on analyses that attempted to control for the effect of the CWC and BWC – along with any emerging norm against the use of these weapons – historical patterns in the pursuit and acquisition of WMDs may be less diagnostic today if chemical and biological weapons are perceived to be more taboo. Finally, the analyses behind these inferences do not explicitly consider the role of conventional forces, or how the pursuit of acquisition of different WMDs is related to the pursuit of conventional weapons superiority. As a result, it is not clear whether the risk of chemical and biological weapons pursuit will be mediated by the opportunity to pursue conventional weapons.
The third and fourth inferences are even harder to sustain. In addition to the being plagued by all of the problems listed above, these inferences require additional assumptions that have little to no empirical support. For example, to sustain the third inference, one would need to further assume that the demand for nuclear, chemical and biological weapons is similarly correlated for terrorist groups, something neither our study nor any other research has the data to support. Without an explicit analysis of terrorist groups, it is simply impossible to know whether the demand for different weapons of mass destruction is more or less elastic among groups compared to states in the international system. Similarly, to sustain the fourth inference, one would need to assume that any unilateral reductions in the nuclear weapons stockpile of a patron-state would lower the perceived credibility of an extended nuclear deterrence commitment. And yet, there is no systematic evidence to date that such reductions leave allies any less assured, nor is there any evidence that having some unmet demand for security would cause client-states to pursue any WMD without distinction.
If, as the President argued in his Prague Speech, the threat of global nuclear war has gone down but the risk of a nuclear attack has gone up due to a fundamental changes in the international security environment, then a policy intended to significantly lower nuclear force levels may enable the US to better address the pressing challenges of nuclear terrorism and nuclear proliferation. Furthermore, as the 2010 Nuclear Posture Review argued, nuclear reductions may also put the US “in a much stronger position to persuade our NPT partners to join with us in adopting the measures needed to reinvigorate the non-proliferation regime and secure nuclear materials worldwide” while serving to “decrease incentives for additional countries to hedge against an uncertain future by pursuing nuclear options of their own.” When combined, these efforts have the potential to significantly reduce the availability of nuclear weapons to hostile states and terrorist groups by restricting the overall supply. But what becomes of the strategic salience of other “weapons of mass destruction” amid these deep nuclear reductions? Do they acquire the status of "poor man's nuclear bomb" in this altered state? And under what conditions might this occur?
This essay did not provide definitive answers to these questions. Instead, it chose to position these specific questions within a broader research agenda aimed at understanding the relationship between different weapons of mass destruction in general. It then summarized one recent to see what inferences – if any – can be drawn from past findings to the likely impact of deep nuclear reductions on the risk of chemical and biological weapons proliferation. I argued that – despite the importance of previous findings – much more research must be done before scholars can make reliable inferences about the likely consequences deep nuclear reductions. To illustrate this, I anticipated four mechanisms through which restrictions in the global supply of nuclear weapons might be posited to inadvertently increase the risk of chemical and biological weapons proliferation, and I concluded that each of these inferences are unsustainable given the available evidence.
The limitations of these inferences not withstanding, this essay sought more generally to renew a debate about the expected benefits, direct costs, and inadvertent side effects of deep nuclear weapons reductions for strategic stability. Clearly, security analysts and policymakers seek to reduce the availability of nuclear weapons because they believe this will make the world a more secure place. However, if increasing security is the ultimate goal of nuclear weapons reductions, then any benefits from limiting the supply of nuclear weapons must be weighed against the direct costs and unintended consequences.
At present, any unintended consequences of deep reductions in the global supply of nuclear weapons are highly speculative. However, this does not mean that the assumptions necessary to sustain these inferences should not be clearly articulated so that they can subjected to scientific scrutiny and potentially falsified. Rather, this essay sought to highlight the temptation to make premature inferences about the likely outcome of a specific policy based solely on general tendencies we observe in the historical data. Many of the empirical findings that researchers identify only follow under very specific measurement and modeling assumptions, and thus we cannot be certain that they will accurately forecast future outcomes in the real world. It is thus critical to the collective credibility of any research program – including research on the causes and consequences of WMD proliferation – to resist the oftentimes strong private incentive to influence policy prematurely.
Finally, and more generally, this essay sought to call the attention of researchers to provide a better understanding of the causes and consequences of chemical and biological weapons proliferation. In particular, there is a tremendous amount of analytic leverage to be gained from determining whether nuclear, chemical and biological weapons can be usefully understood as belonging to a single analytic category of “weapons of mass destruction”. If the pursuit of these weapons generally share some defined and predictable relationship, we can begin to derive empirically testable hypotheses about how various policies and structural changes are likely to influence supply and demand across WMDs. If, on the other hand, there appears to be no relationship in the pursuit of nuclear, chemical and biological weapons, then policies premised on identifying and mitigating the overall risk of WMD proliferation or of a WMD attack might be confusing and dangerously misguided.
Beardsley, Kyle, and Victor Asal. 2009. “Winning with the Bomb.” Journal of Conflict Resolution 53(2): 278-301.
Berkowitz, Bruce D. 1985. “Proliferation, Deterrence, and the Likelihood of Nuclear War.” Journal of Conflict Resolution 29 (March 1): 112-136.
Betts, R. K. 1998. The New Threat of Mass Destruction. Foreign Affairs, 26-41.
Brito, Dagobert L., and Michael D. Intriligator. 1996. “Proliferation and the Probability of War: A Cardinality Theorem.” Journal of Conflict Resolution 40 (March): 206-214.
Fuhrmann, Matthew. 2009. “Taking a Walk on the Supply Side: The Determinants of Civilian Nuclear Cooperation.” Journal of Conflict Resolution 53(2): 181-208.
Gartzke, Erik, and Dong-Joon Jo. 2009. “Bargaining, Nuclear Proliferation, and Interstate Disputes.” Journal of Conflict Resolution 53(2): 209-233.
Intriligator, Michael D., and Dagobert L. Brito. 1981. “Nuclear Proliferation and the Probability of Nuclear War.” Public Choice 37: 247-260.
Horowitz, Michael. 2009. “The Spread of Nuclear Weapons and International Conflict: Does Experience Matter?” Journal of Conflict Resolution 53(2): 234-257.
Horowitz, M & Narang, N. 2014. Are Nuclear, Biological, & Chemical Weapons Substitutes? Comparing the Determinants of WMD Proliferation. Journal of Conflict Resolution
Hymans, J. E. 2006. The Psychology of Nuclear Proliferation: Identity, Emotions and Foreign Policy. Cambridge University Press.
Jo, Dong Joon, and Erik Gartzke. 2007. “Determinants of Nuclear Proliferation: A Quantitative Model.” Journal of Conflict Resolution 51(1): 167-194.
Kroenig, Matthew. 2009. “Importing the Bomb: Sensitive Nuclear Assistance and Nuclear Proliferation.” Journal of Conflict Resolution 53(2): 161-180.
Koblentz, Gregory D. 2009. Living Weapons: Biological Warfare and International Security. Cornell University Press.
Kydd, A., & Walter, B. 2006. The Strategies of Terrorism. International Security, 31(1),49-80.
Lavoy, P. R., Sagan, Scott D., & Wirtz, James J. (Eds.). (2000). Planning the Unthinkable: How New Powers will use Nuclear, Biological, and Chemical Weapons. Cornell University Press.
Narang, V. 2013. What Does It Take to Deter? Regional Power Nuclear Postures and International Conflict. Journal of Conflict Resolution, 57(3), 478-508.
Powell, Robert. 1990. The Nuclear Revolution and the Problem of Credibility. Cambridge, MA: Cambridge University Press.
Rauchhaus, Robert. 2009. “Evaluating the Nuclear Peace Hypothesis: A Quantitative Approach.” Journal of Conflict Resolution 53(2): 258-277.
Sagan, Scott. 1993. The Limits of Safety: Organizations, Accidents, and Nuclear Weapons. Princeton, NJ: Princeton University Press.
Sagan, Scott D. 1996. Why Do States Build Nuclear Bombs? Three Models in Search of a Bomb. International Security, 21(3).
Sagan, Scott D., & Waltz, Kenneth N. 2012. The Spread of Nuclear Weapons: An Enduring Debate (pp. 108-109). New York: WW Norton.
Schelling, Thomas. 1960. The Strategy of Conflict. Cambridge, MA: Harvard University Press.
———. 1966. Arms and Influence. New Haven, CN: Yale University Press.
Singh, S. and Way, C.R. 2004. “The Correlates of Nuclear Proliferation: A Quantitative Test.” .” Journal of Conflict Resolution 48(6): 859-885.
Solingen, Etel. 1994. The Political Economy of Nuclear Restraint. International Security 19 (2): 126-69.
———. 1998. Regional orders at century’s dawn: Global and domestic influences on grand strategy. Princeton, NJ: Princeton University Press.
———. 2007. Nuclear logics: Contrasting paths in East Asia and the Middle East. Princeton, NJ:Princeton University Press.