A Rough Nuclear Threat Assessment for the United States

by CKR

It must be something in the air. Or maybe it’s that the Subcommittee on Water and Energy Development (of the House Committee on Appropriations) just scratched the funding for the Reliable Replacement Warhead (RRW) program. That’s no guarantee that the funding won’t be put back in at some point on the appropriations bill’s journey through congress, but it got some people’s attention.*

Both Scientific American and a former colleague who calls himself Dr. Strangelove have proposed a debate, preferably on their blogs, on nuclear weapons. I’m not impressed with what they’ve got so far.

If you’re going to discuss the need for nuclear weapons or their possible missions, you should start out with a threat assessment. It’s also possible to begin with your druthers, but not particularly reality-oriented. So I’ll continue from my outline for a threat assessment.

I’ve done my own threat assessment of sorts. It’s developed over a period of years, and I suspect it’s as good as anything that might be done by others with more time and access to more information. I’m open to changing it if better information appears, but that information has to be reliably supported, not just suppositions about terrible things that might happen.

Here is my unofficial nuclear threat assessment for the United States. Note that this is a nuclear threat assessment, not an overall threat assessment. There are still some dangers out there; I am focusing on those that might use nuclear weapons against the United States or in some other way provoke nuclear retaliation by the United States.

Finding 1. No serious immediate threat.
There is no country in the world that seriously threatens a nuclear attack on the United States. Further, the probability that a terrorist organization has usable nuclear weapons is extremely low. The most serious current threat of a nuclear explosion in the United States arises from accidents resulting from the continuing alert status of US and Russian nuclear-tipped missiles.

Finding 2. Threats in the 2-5 year range are extremely low. Most can be managed by US actions.
Relations with Russia are deteriorating. Relations with China are good, except for some friction in the area of trade. An agreement has been reached with North Korea on denuclearization. Iran is unlikely to have nuclear weapons within this time frame. Pakistan’s current instability presents a concern that action against the government might put nuclear weapons in the hands of radical Islamic groups. Russia continues to improve its nuclear weapons security.

There are a number of ways to improve relations with Russia, including delaying construction of antimissile installations in Poland and the Czech Republic. Because Iran is unlikely to have nuclear weapons within this time frame, there is time for negotiation. The instability in Pakistan is the most unpredictable and uncontrollable of the threats in this time frame. We have to hope that rumors are true that the United States has been helping Pakistan to “safe” its nuclear weapons with permissive action links that keep unauthorized people from using them.

Finding 3. Threats in the 5-20 year range are much less predictable, but remain low.
With intelligent diplomacy and some steps back from the more warlike policies of the Bush administration, such as preventive warfare (which step may have already been taken), good relations can be maintained with other nuclear powers. In a similar vein, progress should be possible with North Korea and Iran toward non nuclear weapon status. Instability in Pakistan and friction between Pakistan and India are probably the biggest threats of nuclear war or nuclear weapons becoming available to subnational groups. Regional conflicts could encourage other states (say Brazil and Argentina) to consider a path to nuclear weapons, but the probability of such conflicts seems likely to remain low.

Let’s just stop here for a moment and take a breath. This is a very different threat assessment from anything that might have been done during the Cold War. In fact, it surprised me when I saw it all written down this way. But if we stick to verifiable threats with reasonable probability, I think this is the way it has to come out.

I’m going to make a couple of other assumptions about United States nuclear policy: that conventional weapons are sufficiently powerful that nuclear weapons need not be considered as a response to any but a nuclear attack; likewise, that defending US interests abroad, such as shipping lanes, can be done adequately with conventional weapons.

The bottom line is that there is vanishingly little need for nuclear weapons in the United States as retaliation for a nuclear attack.

But deterrence has a value, demonstrated during the Cold War and in countless small ways then and since. If a nation is strong enough, potential attackers will think twice.

This seems to be largely the reasoning behind the smaller nuclear stockpiles: those of Britain, France, China, India, Israel, and Pakistan. All of those stockpiles, for nations with very different populations, land areas and potential enemies, are similar in numbers.

According to the Carnegie Endowment for International Peace, those numbers are
Britain 200
France 350
China 410
India 70 – 110
Israel 100
Pakistan 50 – 110

Others believe that the number for China is closer to 100. The Federation of American Scientists estimates 145 warheads deployed, which could be consistent with a total of 410, but also with lower numbers. Jeffrey Lewis and the US Department of Defense estimate that the total is closer to 100, perhaps as few as 80. I tend to agree with this lower estimate.

In terms of strategy, and again I have not spent as much concentrated study on this as I might, but I have considered it over a long time, there appears to be almost a power law for nuclear weapons:
1 – 10 Demonstrates basic capability; useful for intimidating non-nuclear neighbors and deterring larger adversaries.
10 – 100 Capability for warfighting ranging from nuclear versus conventional to nuclear exchange.
100 – up Capability for nuclear exchange. Close to complete deterrence of rational adversaries.

Except for the United States and Russia, the numbers of nuclear weapons are on the borderline between deterrent and warfighting. This suggests that the motivation is largely deterrent, and it also suggests that numbers of a few hundred each should suffice for the United States and Russia as well.

_________________________________
* If you’ve clicked that second link, you’ll see it goes to the third incarnation of the LANL Blog. I hesitate to get wrapped up with that uncivil lot, but part of the discussion is going on there. So I’ll just put a warning here that various sorts of garbage will be deleted from the comments thread. The WhirledView bloggers are pretty broadminded about comments, but we have our limits. Complaints about our policy will also be deleted. It’s our blog.

Comments

[Andy]

Interesting discussion.

One thing I would quibble with however is the applicability of a "nuclear threat assessment" in relation to how many weapons we should have. Nuclear weapons, at least among the primary nuclear weapons States, are solely weapons of deterrence, and they deter not only nuclear, but also other kinds of attack from nations. Therefore they cannot simply be compared to perceived risk or threat of nuclear attack or nuclear war. The US, for example, used an implicit threat of first-use against the Soviets to deter a conventional attack by the Soviet's much larger conventional force in Europe. Similarly, the US has an intentionally ambiguous position WRT other so-called WMD and will not rule out the use of nuclear weapons in response to other mass-casualty munitions. So US deterrence strategy at least is, and has been for many years, based on much more than a solely nuclear threat.

Additionally, you mention the nuclear terrorist threat (which I agree is quite low), but including them in your threat assessment would seem to indicate that you may believe they can be deterred through our own arsenal. I would argue that it would make little difference how many warheads we have since we could not, and would not, use them against such a group and therefore they are unlikely to be deterred like a nation-state would.

Overall, though, I think you make a good case in your "findings." What's important in my view is that the US maintain a parity of capability vs the other nuclear weapons states. By capability I mean more than simply the number of deployed warheads, and would include other factors such as immunity to a nuclear or conventional first strike. As an example, if the US based all of it's nuclear deterrent on submarines it could arguably maintain a deterrent with fewer weapons since they would be immune to a nuclear first strike and practically immune to a conventional first strike.

[CKR]

I tried to include all the potential nuclear threats against the United States. That includes terrorists with nuclear weapons.

I agree that US nuclear weapons are unlikely to deter terrorists (although they may deter nations from transferring nuclear weapons to terrorists). That means that additional nuclear weapons in the US arsenal will not add to deterring terrorists.

I believe I've also dealt with the question of national deterrence by looking at nuclear arsenals around the world. A couple of hundred should be more than enough for this purpose.

[Pinky and The Brain]

Thanks for the plug, Cheryl. The comments on the LANL blog can be downright brutal, but we won't delete them. Send all the uncivil types our way. The civil ones are welcome too.
Cheers,
Pinky

[CKR]

You're welcome, Pinky and The Brain.

We're basically open to comments, but we draw some lines. Personal attacks, for example, don't add anything useful to the discussion.

We welcome substantive comments like Andy's.

[A.E.]

That's a great, refreshingly hysteria-free analysis. Much talk in the media about nuclear strategy tends to dip into Jack Bauer "if Los Angeles was nuked, who would we glass" kind of territory.

Pakistan and India though, is a real problem. Assuming you had the ear of the next president, what immediate steps would you advise him/her to take to tamp down the instability level on the subcontinent?

[CKR]

Thanks, A.E.

The analysis came out less threatening than I expected. It's almost as if we're experiencing a reverse-Herman-Kahn reaction. Where Kahn was scorned for speaking the horrors of nuclear war, it's a much less dangerous world that people seemingly don't want to hear about now.

I don't have an easy answer to your question about Pakistan and India. But I intend to write at least one more post in this series that will have some ideas on how to defuse nuclear tensions.

[Doctor Strangelove]

CKR -

We agree with your findings, especially 1 and 2. There is very little if any credible *nuclear* threat against the US in the near to medium term. Beyond that (finding 3) things are a lot less easily judged.

It is amazing how little response We (and Scientific American) got on this topic. It is a hugely sad statement that there is so little real discourse on the topic. We hope that you do better here at generating a conversation.

We also agree with your implications that the US can have a *LOT* of deterrence from 10-100 warheads and that above 100 we probably intimidate "any rational adversary" and by extension or implication, nothing will deter an irrational adversary.

9/11/2001 proved to us by example that terrorists cannot be intimidated (mainly because they cannot be targeted, and those who can are probably already suicidal).

We don't know how many folks we had to talk to on 9/12/2001 who insisted that a "retaliatory strike" was necessary, even though they could not clearly identify an adversary. A few though nuking Kabul made sense... the rest were just sure "someobody" needed to pay. But who, where?

Had we nuked Kabul, and then maybe Baghdad and then maybe Tehran, we would probably have then had our hands forced to nuke Karachi and God (or Allah) knows how many other major cities in the muslim world. And to what effect?

As a child many of us gave into the temptation to hit a hornets' or bees' nest with a rock, or pour a little gasoline into a anthill and light it off. If we were lucky we didn't get stung/bit too badly... but I doubt anyone ever wiped out the "scary hive".

We don't mean to compare Islamic peoples or cultures to hive insects so much as to compare humans in any culture/group has hive insects. Using the same analogy, the folks who planned and flew the planes into our buildings were doing little more than hitting the hornets' nest with rocks. Thousands of people were killed and significant disruption was achieved but the biggest effect is that we swarmed to the middle east and "stung" everyone we could find.

At this point in history, our nuclear arsenal is a deterrent only against others living in "glass hives", bu t little else.

It is a sad time.

- Doc

[CKR]

Doc -

I have tried to get some dialog going before on this subject, with poor results. This post has the best response so far.

I think it's partly the yuck factor (nobody likes to think about nuclear war) and partly the revenge factor you cite, along with far too much fear being whipped up by our national leadership and the media.

Popular opinion, usually in the 60-70% range according to polls I've seen cited, seems to favor getting the nuke numbers down, and we see that one subcommittee of Congress has just agreed with that. Not to mention the Nuclear Nonproliferation Treaty.

I'm doing some research on that and hope to have more to say in a future post, hopefully next week.

[Big D]

I thought the RRW had more to do with the fact that our current weapons are getting a little old (read: unreliable) than with increasing their numbers. Having 2,000 warheads means little if there's a good chance that they wouldn't go boom if we wanted them to.

Also, I would note that having "10-100" will only allow you to threaten (or deter) 1 decent-sized nation. Beyond that, you get into the same logic that led us to declare "2.5MRC" on the conventional side during the 90s, when we weren't even sure if we could handle the "2". Note that England and France are under our nuclear umbrella, and the rest are limited by expense, resources, and/or technology.

[CKR]

I plan to write more on the RRW, hopefully in the next week or so. The logical followon to a threat assessment is the question of what needs to be done in response, so I'll continue in that vein.

If other nuclear arsenals are limited in number, it doesn't matter what the reason is. And that "nuclear umbrella" for England and France makes sense only under Cold War conditions.

The point I'm making is that those conditions no longer pertain. Your entire last paragraph, Big D, is based on Cold War assumptions.