S&T [Science and Technology] can contribute to achieving such progress in several ways: through technical advances that make verifying weapon-reduction agreements easier (and thus make agreeing to them easier); through other technical advances that make nuclear energy technology less likely to be used for nuclear weaponry and/or more likely to be detected if this happens; through applications of science and engineering to the task of reducing the dangers of accidental, erroneous, or unauthorized use of nuclear weapons, as well to the task of obviating any need for nuclear explosive testing of weapons, for as long as these still exist; and through S&T-based integrated assessments clarifying dangers and pitfalls on the path to zero and how to avoid them.
Think of this as a systems analyst offering a 10,000 mile high overview of what Holdren referred to as "those dangers and pitfalls on the path to zero."
I have seven observations to make from that perspective:
(1) In a world in which nobody "officially" has any nuclear weapons, the value of clandestinely having just a few becomes practically infinite. The incentive to cheat becomes irresistible strategic common sense for countries that feel threatened by their neighbors, but are unable to deter them by non-nuclear means. And this cheating could come either through indigenous S&T, or through prices to the likes of an A. Q. Khan network inflating almost without limit.
(2) As the nuclear umbrella of the United States folds up, the thirty-odd countries that are currently under it will take stock of their situation. Some of them will "go nuclear." Thus, although the total number of nuclear weapons may decline, they may be possessed by more and more countries. That is to say, the decline of the arsenals of the great powers may actually increase the risk of nuclear weapons proliferation.
(3) As Holdren notes, proliferation to more countries will mean that more countries have nuclear weapons without having experience in controlling and securing them. Let me amplify this point. Our experience helping Russia with nuclear materials protection, control and accounting (MPC&A) indicates that such activities are strongly influenced by geography and culture. As nuclear weapons are developed by more and more countries in response to the decline of the great power arsenals, there will be a wider variety of MPC&A practices for terrorist organizations (of which al-Qaeda is only one) to probe for weaknesses. The chances for terrorists to obtain nuclear weapons may actually increase as we proceed along the path to zero.
(4) In a world where the great power nuclear arsenals decline below some trigger point, secondary powers may decide to sprint to parity or superiority. (Think of Pakistan realizing that it could surpass the US and Russian arsenals by ramping up production for a year or two.) This could lead to very unpleasant consequences, perhaps the best of which is a new multi-polar nuclear arms race.
(5) Suppose that all the forgoing obstacles are overcome, by almost magical means of surveillance and verification that work even with un-cooperative regimes, by preternaturally skillful diplomacy, etc. The result will be a world made safe for unlimited global warfare. Multi-state hot shooting wars, like World War I and World War II, could again be risked without fear of total annihilation. World War II resulted in the deaths of some 50 million people. Indeed, until the invention and shocking use of nuclear weapons, the number of people killed in wars had been rising exponentially. Eliminating nuclear weapons might cause a reversion to the status quo ante - more people might end up being killed in wars in a nuclear weapon free world than in a world with one or two large, stabilizing arsenals.
(6) From the forgoing it should be obvious that eliminating nuclear weapons will not make peace. It works the other way around. Making peace will eliminate nuclear weapons. (How long do you think North Korea's nuclear arsenal would last if there were genuine peace on the Korean peninsula, for example? I mean real peace, like what happened between the East and West Germany.) But eliminating nuclear weapons is conceptually easy. In essence it requires the ability to count. Making peace, by contrast, is hard. It can't be quantified precisely. It can't be measured. So far, it has even eluded precise definition by sociologists, the very people you'd think would be the experts on peace. Yet, if we do not make peace on earth as we eliminate nuclear weapons, we will simply create a less stable and more dangerous world than the one we have now.
(7) Finally, the whole issue of nuclear weapons may be "overtaken by events." What makes anyone think that the atomic nucleus contains the last word on explosive energy release? Of course, the theoretical physics for such a thing hasn't been discovered yet, but the physical limit on explosive energy release is out there for all astrophysicists to see. It's the Big Bang that began our universe. Now it is true that any explosive event more intense than a nuclear weapon will release the same forms of energy (x-rays, gamma rays, etc.) and it can therefore be treated as a nuclear weapon. That's true, technically. But quantity has a quality all its own. Nuclear weapons are about a million times more powerful by weight than chemical explosives. Can we really deal with another factor of a million? Or a million million? World culture is now such that S&T marches on. Genies will keep popping out of bottles they can't be put back in. Peacemaking is not an option. It is a necessity.
All of this is to say that the path to zero will be perilous, and must be tread with great care. Although the necessity of peacemaking tells us that we must start down that path, we do not yet know how to go very far along it, much less how we will finish.
May God help us.