Good afternoon, and thanks for giving me the chance to talk to your team.

This is a good time to be talking and to be thinking about deterrence.  Our country faces unprecedented challenges in multiple theaters, across multiple battlespace domains, and at multiple levels of confrontation, competition, and conflict. 

At the top of the list, of course, is the so-called “two-peer” problem of nuclear deterrence – that is, the challenge of how to deter two nuclear weapons possessor near-peers at the same time with only one force posture of our own, and to be able to do so not just at the strategic level but also at the theater level where Russian and Chinese forces present acute threats to U.S. allies and friends alike.  Having picked a fight with Iran that we apparently have little idea how to win, moreover, we’ve also now got a great deal of trouble on our hands when it comes to intra-conflict deterrence – by which I mean the challenges of both vertical and horizontal escalation management once hostilities have commenced.  So our challenges are rife.

Many of the basic challenges of deterrence are as old as the possibility of human conflict itself, of course.  Nevertheless, each era has to confront them in a new context. 

The nuclear era led to a particular blossoming of writing and thinking about such problems, but deterrence theory and practice became distinctly unfashionable after the end of the Cold War, which led us rather rashly to assume that all that ugly, expensive, dangerous, and scary work of strategic competition was now forever behind us.  As a result – the mental capacities we use for strategy being a bit like muscles, in that they atrophy if you don’t use them – the U.S. national security community has been struggling to wake itself up and get back to the mental gym again.

It’s a pleasure to speak to your team now, because you’re engaged in what sounds like some really important work in developing some of the physical capabilities our nation will need to meet these challenges. 

Policy, Strategy, Operations, and Capabilities

I’ve been asked to say a few words about some of the policy and strategy issues related to the work you’re engaged in, and I’m happy to try.  Naturally, you’re only going to hear my personal opinions, and these won’t necessarily reflect the views of anybody else.  But I do think it’s really important for all components of our national security community to pull together as we try to discern a way forward in the face of these myriad challenges.

The reason it’s important to try to look at these issues in a comprehensive way that involves both capability-developers like you and policymakers and strategists like me isn’t too surprising.  Strategy without the tools with which to implement it is pretty useless, of course, but developing powerful tools without a sound strategy for employing them will also be inadequate.  So we need to get thing right on both accounts.

In his classic work On War, the great Prussian military strategist Carl von Clausewitz stressed the importance of what he called “politics” or “policy” – which word one uses in English depends which translation you prefer from the German, I guess – in providing the guiding rationale for any use of military force.  War, he famously put it, is “politics by other means,” from which it followed that it was the job of the military commander to try to ensure to make sure that operational activities and military engagements always serve the purposes for which one is fighting in the first place, neither exceeding what is needed to serve that purpose nor falling short.  And it’s the challenge of military leadership to do this notwithstanding warfare’s frequent tendency toward escalation and the pervasive impact of what he called the “friction” created by chance, uncertainty, errors, blunders, and adversary action.

That may seem pretty commonsensical.  Nevertheless, even such a simple insight – about the need to ensure that military activity supports a clear strategy – can be very hard to live up to in practice.  And this is also true with the related challenge of making sure that our development of weapons systems is done in ways that create usable military capacity in support of a clear vision of overall strategic utility. 

Military history, of course, offers many illustrations of capabilities that kept being pursued long after their ability actually to contribute to higher purposes had expired.  Nor is it short of examples of leaders pursuing seemingly brilliant strategies, but doing so in ways unmoored from the harsh realities of what capabilities and resources were available to execute them.  Keeping strategy and policy and military capability development from becoming unmoored from each other is thus critical

The graduate school I run – the School of Defense and Strategic Studies – has specialized for more than half a century in helping prepare national security professionals to struggle with these sorts of issues.  We did this first at the University of Southern California, then at Missouri State University (MSU) in Springfield, Missouri, and most recently as an MSU outpost in the Washington, DC, area.  So let me take a stab at contextualizing the work you’re do.

In some sense, the problem we in the United States face today is twofold.  We lack both the full range of diverse, numerous, and effective tools we need either for deterrence or for warfighting and escalation management if deterrence fails, and we lack clear concepts through which to make whatever tools we do possess actually serve our national interest in challenging times. 

Deterrence Threats 

Let me start by talking about nuclear weapons.  As far as these tools are concerned, over the years we’ve allowed a very problematic and dangerous gap to develop between certain aspects of our capabilities and those of our adversaries.  We’ve not fallen dramatically behind the Russians at the strategic level, thankfully.  But with China now engaged in expanding its nuclear arsenal in what appears to be a breathless race toward strategic parity, we now face the emergence of a “two-peer” problem for the first time – and while we’re working on this challenge pretty hard now, we’re still not entirely sure what to do about it. 

We’re also a pretty much maximum conceivable capacity in terms of our production capability – both for delivery systems and for the nuclear warheads themselves – even while still struggling to simply to replace and modernize legacy systems.  In effect, we basically hard-wired our production throughput years ago, and in both respects, to a post-Cold War level that is low enough to give our adversaries little reason to fear an arms race with us, and indeed one that may tempt them to think they can win such an arms race.  Obviously, that’s not good!

And the problem is even worse at the theater level, where we face a terrible situation of adversary overmatch in nuclear forces, including in dual-capable (i.e., optionally nuclear or conventional) delivery systems.  Thanks to their continued building while we dismantled most of our capabilities after the Cold War – and, in Russia’s case, also as a result of arms control cheating, for the Kremlin didn’t bother to fulfil the promises it made under the “Presidential Nuclear Initiatives” (PNIs) of the early 1990s and then violated the Intermediate-range Nuclear Forces (INF) Treaty to boot – Russia and China now field vastly more theater-class nuclear systems that we do. 

In terms of deterrence theory and the potential demands of escalation management in a crisis or conflict, this has left us with a stunted “escalation ladder” with only one rather small and limited “rung” on it.  Specifically, in between no nuclear use and the use of strategic nuclear systems, we today have no more than a relative handful of mere gravity bombs deployable on U.S. or NATO partner aircraft. (And they are gravity bombs, for goodness’ sake; you generally actually have to overfly the target, or at least come close enough to “lob” the device through the air in an arc.  What pilot really wants to do that against a seriously defended target these days?  And what Russian or Chinese target worth using up one of our few B61 bombs to engage wouldn’t be seriously defended?)

By contrast, the Russians (and China too)  have buit themselves an escalation ladder against us that has many rungs.  This makes it far easier – and, I fear, rather tempting – for them to confront us with situations in which, faced with their limited nuclear use or other varieties of nuclear-weapons-facilitated coercive bargaining, we would lack more than the most paltry nuclear response options short of a catastrophic escalation to the strategic level.  This theater imbalance undermines deterrence in peacetime, just as it also undermines our chances for successful escalation management – or effective warfighting – if deterrence were to fail.  This, I fear, will be slow to fix.

Deterrent Impact of Conventional Tools

But note that I said this state of affairs derives from our relative lack of flexible nuclear responses.  That’s where yourwork on theater strike capabilities may come in, because nuclear responses are not the only ones that could be useful, and building out our toolkit in the conventional arena is surely also part of the right answer to these deterrence problems.

Having first been used to win a conventional war, and having been deployed by NATO for generations to deter an invasion by Warsaw Pact conventional forces, nuclear weapons have always been tools of what is called “cross-domain deterrence” – that is, they have utility in deterring not just nuclear attack, but also the use of force beyond the nuclear arena.  It’s a pernicious fallacy, therefore, to argue that nuclear weapons are, or even could be, instruments the “sole purpose” of which is simply to deter the use of other nuclear weapons.  As illustrated, for example, by NATO’s unwillingness to engage directly against Russian forces in Ukraine, as well as Russia’s unwillingness to strike directly and openly against NATO countries subject to the collective security promises of Article V of the NATO Treaty, nuclear weapons clearly do have utility in deterring high-end non-nuclear violence.  (“Sole purpose” thinking was a fashionable aspiration for a while in anti-nuclear circles, but it represented a deeply ahistorical confusion and was always a bad, deterrence-undermining idea.) 

But this cross-domain utility works both ways.  I don’t think conventional weaponry is capable – either in theory or in practice – of entirely substituting for nuclear weaponry.  (To hold certain assets at risk, nothing other than a nuke will really do.)  Yet because they are clearly able to hold credibly at risk some things that our adversary greatly prizes, it is certainly the case that conventional capabilities such as the systems on which you’re working can also have real deterrent value (not to mention warfighting value) vis-à-vis a nuclear-armed aggressor. 

For that reason, effective conventional strike tools can help carry some of the burden of deterrence in the nuclear arena, making our conventional and nuclear toolkits into complementary and mutually-reinforcing ones.  From the perspective not just of peacetime deterrence but also of escalation management in crisis or conflict – not to mention that of full-up warfighting – it is thus critical to have more, more flexible, and more effective non-nuclear options as well as nuclear ones.

And that’s where programs such as your precision theater-strike work may come into play.  You’re working on tools that have can help narrow the dangerous, deterrence-erosive gap that has developed in the theater-class nuclear arena, because the kinds of thing you’re building have the potential to let our commanders engage high-value and perhaps fleeting targets with great precision and great dispatch. 

Such conventional capabilities would useful in conventional warfighting, of course – and hence itself supportive of overall deterrence, by the way, since even a nuclear-armed enemy will surely be less likely to pick a fight he knows will be harder for him to win without nukes.  But such capabilities would also add both to the deterrent effect of our nuclear toolkit, and to our nuclear warfighting capability if deterrence were to fail, by somewhat reducing the range and number of critical targets that our thinly-stocked theater nuclear arsenal might have to engage in a conflict that began to escalate.  Such conventional systems would also give us more answers to efforts by Russia or China to engage in coercive bargaining with nuclear weaponry, since it would help give us more responsive options short of a strategic nuclear exchange.

Nor is that necessarily all, for that’s just assuming that such new U.S. delivery systems remained only conventionally armed.  If you ask me, work on conventional prompt strike capabilities could also be leveraged directly into nuclear deterrence as well. 

I wrote an article for War on the Rocks not long ago, for example, outlining a possible approach to a “Europeanized” system of “extended” nuclear deterrence that would involve producing small nuclear warheads that could be optionally swapped into the existing range of long-range conventionally-armed precision-strike missiles that America’s European allies are already purchasing and deploying in considerable numbers.  If I had a magic wand, I’d like to see the United States acquire such dual-capability options itself – such as for systems like the ones on which you are working – and I hold out hope that the Trump Administration will finally choose to boost our theater nuclear capabilities, which indeed some reports suggest may now actually be contemplated.  (Our adversaries have had optionally nuclear-armed hypersonic precision-strike systems for years; why cannot we deploy our own to help deter the threats they present?)

Even without that step, however, we need a more flexible range of options in the face of the range of threats and challenges with which our adversaries have equipped themselves to confront us.  So I’m very glad of the kind of work you’re pursuing, and hope that we can expand the American responsive “toolkit” quite a bit – in both non-nuclear and nuclear ways – soon.

Integrating Theory and Practice for National Security 

But I should also stress – for the reasons I indicated a moment ago – that this is only part of the answer.  In order to be able to use those tools effectively and make them serve the Clausewitzian “policy” purposes of deterrence, escalation management, and warfighting efficacy in extremis, such capabilities need to be both deployed and employed pursuant to a clear and thoughtful strategy that is informed by the careful study and evaluation of the adversary’s capabilities and motivations, by sound operational analysis, and by a realistic eye to the availability of resources and the sustainability of our approaches.

And that’s where technical wizards like you and the “policy and strategy” elements of the national security community can, and indeed must, work together. 

To facilitate such cooperation, I hope that non-technical policy people such as myself will become ever more conversant in the technological and engineering aspects of capability development.  (If full “fluency” in the language of technology and engineering is perhaps too much to ask, at least we need to avoid being simply lost in discussing such matters, for one cannot figure out how  to use something effectively in support of strategy unless one understands what it can do, what it can’t, and why.)  And I hope that the technical community can likewise increase its own ability to “speak” the language of policy and strategy, so as to help national leaders grasp out the various implications and possibilities that may be inherent in our present or potential future technological toolkit.   

In some sense, in fact, I’m trying to model this kind of conceptual conversation with you right now.  You’ll notice, for instance, that I did two things just a moment ago. 

First, I used a historical and game-theoretical analysis of the implications of our relative nuclear capabilities in Europe and the Western Pacific to argue that we need new and improved theater-class nuclear tools.  Second, I argued that based upon what I understand to be the general technical capabilities of the type of precision-strike systems you’re building, the tools you’re developing will help do concrete things that would indeed help bolster deterrence and improve our ability to handle escalation management challenges in those theaters.

My first point thus moved from deterrent logic to desired technical capability, while my second moved in the opposite direction: from specific technical capability to the identification of ways in which such capability could meet an existing national need that can be seen through the prism of deterrence theory.  I think both of these directional arrows are critical, and I would love to see ever more of a lively, deep, and continuous “conversation” develop between the policy and strategy elements of our national security policy community, on the one hand, and its technical and developmental ones on the other.

Many times in the past, these elements have shown that when they work together and understand each other well enough to draw out the best in each community for the common good, really great things can happen.  It’s time to do that again. 

Thanks for listening, and thanks for what you’re doing. 

— Christopher Ford