The two books under review tackle the relationships of the revival of nuclear power in the world, nuclear assistance, and nuclear proliferation. The books summarize, prolong, or introduce debates about the relations between two or three of those variables in the field of security studies in interesting ways, but they remain based on a nonpolitical understanding of history, they are overconfident in the findings they can derive from the use of statistical methods, and their understanding of nuclear vulnerability would benefit from being widened.

Matthew Fuhrmann’s Atomic Assistance and Part II of The Nuclear Renaissance and International Security (including chapters by Alexander Montgomery, Matthew Kroenig, and Justin Hastings and a comment by Erik Gartzke) examine the causes and consequences of nuclear assistance in a context of a putative nuclear renaissance after the Fukushima-Daiitchi nuclear disaster in Japan. Part I of the edited volume investigates the conditions of validity of the subsequent analysis by taking a closer look at the drivers of the nuclear renaissance (it includes chapters by Bernard Gourley and Adam Stulberg, Allison Macfarlane, Fuhrmann, Adam Stulberg, and Joshua Busby and an excellent comment by Christopher Way), and Part III engages in an investigation of the effects of the nuclear renaissance on international crises and conflict (including chapters by Kyle Beardsley and Victor Asal and by Michael Horowitz and a comment by Dan Reiter). Following from those analyses, the authors draw recommendations on several pressing policy issues: Will this nuclear renaissance take place? If so, to what extent is the reliance on nuclear power for energy supply going to increase and where? Should we welcome it in the name of climate change mitigation or equal access to energy worldwide? How should nuclear-capable states balance between their commitment not to deprive any state of its “inalienable right to peaceful uses of nuclear power” under Article IV of the Nuclear Non-Proliferation Treaty with their efforts to prevent the further spread of nuclear weapons?

The two books can be read as a broad discussion of those important issues, and the edited volume is deliberately constructed as such, with a comment concluding every part and a valuable effort to identify points of disagreement. (Fuhrmann and Todd Sechser’s article on “Crisis Bargaining and Nuclear Blackmail,” published in the Winter 2013 issue of International Organization, already appears as a rejoinder to the chapter by Beardsley and Asal in Part III of the edited volume). As a consequence, a lot of the criticism I would like to address is already touched upon in the book in at least one chapter, if not by the commentators on each section. (For example, Reiter already identifies missing cases in the new data set created by Horowitz for his chapter on p. 316; on the issue of the connection between nuclear cooperation and nuclear proliferation, Montgomery and Kroenig outline the clearest alternatives to Fuhrmann’s perspective).

One should also welcome Fuhrmann’s notion of nuclear assistance for not leaving aside the U.S. aid to foreign nuclear programs (p. 15 of his book), contrary to a common trend in the field, and Busby’s solid argument that for nuclear power to mitigate climate change by limiting green gas emissions, it would need to expand on an unprecedented and very unlikely scale, which would in turn cause a nuclear waste problem of a magnitude no one is prepared to deal with. For all those reasons, the edited volume will prove a particularly good support for teaching as well as further research.

In spite of all the merits of this debate, the two books use statistical analysis as a way to explain history and reach policy recommendations, which creates multiple problems. Even if some authors are perfectly aware of the limits of the exercise (see Gourley and Stulberg, p. 36), it leads to a deterministic understanding of history that does not speak its name, depoliticizes the historical record, overestimates the validity of its findings, and does not fit policy needs. As suggested earlier, the excellent commentary by Way touches upon those issues, insisting that “we need to know more” (pp. 161–62) and that “the future may not resemble the past” (p. 157), showing how politics matter in those decisions. However, the problem is deeper, even without contesting the ontological and epistemological premises of the works under review.

It is true that Fuhrmann addresses the issue and explains that his work is probabilistic and not deterministic (pp. 12–13 and 51 of his book). The problem is that this claim is not sustainable if recommendations are extrapolated into the future: Such an extrapolation requires turning a probabilistic assessment of the past into a deterministic and depoliticized juncture between the past and the future. As the philosopher Hannah Arendt argued, the “breach between past and future” cannot be filled with knowledge only; it is the space where human responsibility is exercised and where something unprecedented can appear in the world. Identifying a segment of the past and stretching it into the future denies the role of this political responsibility and falsely suggests that the analyst can know what will continue and what will not. It is historically deterministic, even if denied as probabilism, and politically problematic precisely because it depoliticizes history.

Extrapolating from a global past onto a specific country’s future for the sake of reaching policy recommendations is overconfident. Indeed, the quantity and reliability of the available data on nuclear assistance do not suit large-n statistical analysis (see Harald Mueller and Andreas Schmidt’s systematic critique of two data sets used in the two books under review in William Potter and Gaukhar Mukhatzhanova’s edited volume on Forecasting Nuclear Proliferation: The Role of Theory, 2010). This is all the more surprising as many strong methodological critiques of Fuhrmann’s work and statistical analysis of the proliferation puzzle had previously been published by Alexander Montgomery and Scott Sagan in the Journal of Conflict Resolution (“The Perils of Predicting Proliferation” [April 2009]: 302–28) and by Sagan alone in the Annual Review of Political Science in 2011 (“The Causes of Nuclear Proliferation,” p. 232). They seem to be unaddressed and remain valid. Such a method can only indicate patterns when policymakers are interested in case-specific pathways.

Moreover, Fuhrmann’s main claim that even nuclear assistance that would not be considered sensitive is conducive to proliferation is more debatable than he makes it sound. In the policy community, the proponents of an interpretation of the Nonproliferation Treaty willing to subordinate peaceful uses of nuclear energy and disarmament to the nonproliferation obligation have been making the connection between nuclear assistance and nuclear proliferation for a long time. As they welcomed Matthew Kroenig’s argument linking sensitive nuclear assistance to proliferation in his earlier book (see J. Peter Scoblic’s very favorable review in the Nonproliferation Review [March 2011]: 315–19), they predictably applaud Fuhrmann’s radicalized version of this argument (see Henry D. Sokolski’s review in the Nonproliferation Review [March 2013]: 179–84, quoting the precedents of this argument).

On top of preaching to the choir, Fuhrmann’s main contention is debatable. As a matter of fact, he mentions (Atomic Assistance, p. 152) but then neglects the well-established facts that 1) most states have never had any interest in nuclear weapons acquisition in the first place and 2) that most of those who did contemplate nuclear weapons capabilities ultimately gave up (see, among others, this author’s “Oracles of Proliferation” in Nonproliferation Review 18 [March 2011]: 297–314). For example, he overlooks cases like Sweden, which has received a significant amount of nuclear assistance but decided to give up its nuclear weapons plans at the end of the 1960s. It is now well documented and primary sources are being declassified (see Thomas Jonter’s chapter in Jeffrey Knopf, ed., Security Assurances and Nuclear Nonproliferation, 2012).

This double neglect has massive implications since, as Alexander Montgomery notes in the edited volume (p.178), “when the pool is limited to countries that have actively been seeking nuclear weapons, the effect [of nuclear assistance on the likelihood of nuclear proliferation] is reversed.” This is just another example of the way in which the role of political choice and political change is underestimated by such a probabilistic approach, which seems to assume that probabilistic risk assessment can replace political judgment. For example, Fuhrmann finds that providing nuclear assistance increases the probability of nuclear proliferation by 0.6% per year (p.152 in his book). He needs to stretch this over a 25-year period to reach a number that sounds unacceptable to him. What exactly does that mean, given that renunciation of nuclear weapons has existed at every stage of the development of a program? How relevant is this average number about the part of the global past that we know about to predict future decisions in a specific country? More importantly, a policymaker might be more risk prone than Fuhrmann or any other nonproliferation analyst.

Finally, while both books, and in particular the edited volume, offer valuable insights into very timely policy issues, they seem to ignore the elephant in the room: global nuclear vulnerability. This is all the more surprising given Fuhrmann’s decision to address the issue of nuclear insecurity in the subtitle of his book. The understanding of nuclear insecurity in both books is rather narrow. It only includes the limits of nuclear power stations and weapons’ safety, nuclear terrorism, and the impact of the nuclear renaissance on climate change and on the spread of the weapons to other countries. It disregards how the persisting practices of nuclear deterrence and nuclear security might encourage or provide justifications for modernizing the existing arsenals as well as acquiring nuclear weapons, and it seems to overlook the global condition of nuclear vulnerability, which has been ours since the coupling of intercontinental ballistic missiles and thermonuclear weapons against which no protection is or has been credible. The issue of nuclear terrorism suggests defenselessness, but all of the important and well-tackled issues in those two books are only instances of a broader problem that would have needed thorough consideration: global nuclear vulnerability.