Q&A: Nuclear attribution

Rob Quayle wrote in concerning Q&A: Why Iran's not crazy (again)

One conclusion from the material below is that nuclear attribution techniques are so sophisticated and so highly classified that they cannot be evaluated, or even acknowledged. However, the Stanford undergrad has a good point: If nuclear attribution techniques are robust, but not acknowledged, then they would seem to be less effective as deterrents than if they were acknowledged.

Care to comment?

***

Tom Barnett says in the Dec 19,2009 blog:
"Iran cannot pass a nuke to terrorists and not have it tracked back"

This source (foreignpolicy.com) seems to agree:

http://www.foreignpolicy.com/story/cms.php?story_id=3653
By Debra Decker Page 1 of 1
Posted December 2006

Using sophisticated techniques such as radiochemical analysis, scientists can determine the isotopic fingerprint of elements like uranium and plutonium, even after they have been used in a detonated bomb. Then, by isolating telltale trace contaminants and using traditional law-enforcement methods, analysts can narrow down the range of original sources, providing in the process a kind of "birth certificate" for nuclear material. They could then know whether the material used in a bomb might have come from, say, North Korean or Iranian facilities.

But this source does not:

http://cisac.stanford.edu/publications/nuclear_attribution_as_deterrence/

Nuclear Attribution as Deterrence
Journal Article
Author
Michael Miller - 2006 CISAC Honors Graduate

Published by
Nonproliferation Review, Vol. 14 no. 1
March 2007

Recently, post-explosion nuclear forensics, or nuclear attribution, has gained a new spotlight within the scientific and policymaking community working on nuclear weapons. Academics are beginning to ask whether post-explosion forensics might create a replacement for an international nonproliferation regime or at least offer a fallback option to deter states and individuals from selling nuclear materials. This paper examines current attribution technology from unclassified literature and finds the technology to be well developed but not foolproof, such that nuclear attribution currently provides little deterrent value. If current capabilities were publicized more thoroughly, and if the post-explosion process of assessing the evidence were internationalized, states and intermediate actors might be deterred more effectively. This paper also discusses the development of a nuclear fingerprint database; while useful, its impact on deterrence would be minimal.

This article is based on the author's undergraduate honors thesis, completed during 2005-2006 in CISAC's Interschool Honors Program in International Security.

Nor does wikipedia, which says:
Nuclear attribution is the process of tracing the origin of nuclear material that has been used in a nuclear explosion.[1] The problem is not necessarily a straightforward one, for it may be possible to obtain nuclear precursors through the black market, and therefore relatively anonymously.

Tom replied:

Without sounding coy, because I honestly don't know enough to play coy here (that's my defense and I'm sticking to it!), my guess is that far more can be done to ID the ultimate source than is commonly acknowledged in the unclassified world (been a while for me on this subject). I would also opine that, to people in this business, this is fairly well known, so I'm not sure the Stanford grad student's point isn't already well-taken.

In general, the powers-that-be don't care to discuss the underlying reality of deterrence, because their preferences are--go figure--that as few people as possible "get" the bomb, either for real or just in understanding. The mystery of MAD is part of its appeal--don't go there!

But again, read the first cited piece above and you get at the crux of my thinking: some nuke goes off and somebody will have to claim credit, as these things always come out in the end due to serious investigation. And when that understanding is reached (either voluntarily or by force), people with fixed assets (as in, countries) will be severely punished, and quite popularly so--if for no other reason than because it will feel great and send the right kind of signals to the system. And yeah, the system will stand by and let us do it at that point, for all the same motivations.

So, yeah, if Iran really wants to commit suicide, it can be arranged--believe me.

But no, I don't, on that basis, feel our entire national security policy and grand strategic vision should be casually held hostage to whatever that jackass in Tehran happens to say, day-in or day-out. We make it clear to him, like every other clown that's come along, that if he really wants to play with fire, we will be more than happy to immolate his entire civilization.

Why? Because we're Americans, and our entire DNA is about going all the way--when pushed.

Comments

[Andrew in DC]

Rob wrote: "If nuclear attribution techniques are robust, but not acknowledged, then they would seem to be less effective as deterrents than if they were acknowledged."

On this, I'd think - Not necessarily. The techniques themselves may be kept secret so as not to tip our hand. If the techniques were known, publicized, etc then those seeking to produce an un-attributable explosion would employ counter-techniques. This would just start the whole cycle all over again. IMHO, deterrence works better (in this case) when the enemy doesn't know what you have.

[~FR]

Does this resemble the issue around fail-safes? Wasn't one of the reasons we were confident in our highly classified fail-safe tech was that we gave it to independent actors, especially the USSR?

If nuclear attribution techniques are 'double-plus-top-sekret', but shared with responsible parties (the perfidious UK, the impossible French, and the hostile Russians), then it would seem that we could achieve deterrence via those independent back-channels. We would also have independent scientific peer-review.

[RobQ]

I thought it over and believe the following catches the essence of it. Strategically: Never say anything about anything technical. Never make an open claim, & you'll never make a falsifiable claim by mistake. That way the strategist won't ever get caught in a bluff. And the "enemy" we be kept guessing. Which is good when MAD is at stake.