• Home
• Blog
• Podcasts/RSS
• Store
• Help
• About
• Advertise
• Contact
• Pressroom
• Speaker Directory
• Partner Directory

Raymond Juzaitis: So good morning I guess in responding to a comment that Sam Nunn made yesterday about noise in the audience. I guess I would I am when I was asked to write this paper, I was most pleased and very honored to participate in this kind of an - I guess discussion after 27 years or 28 years in the National Lab's first designing weapons, testing weapons, managing weapons programs, and more recently managing non-proliferation Homeland Security and Intelligence Programs. I am - I am just happy to find a place on the arch you know, among all the Noahs. The task it hand was not it was not a very simple one, my thought it would be simple, but as Sid in John can - can look to, it was not necessarily figuring out what to write, it was figuring about what not to write in a paper like this, because verification complaints issues at the ultimate very heart of getting to the mountain, and and so I was trying to find a threat that would connect not just discussions like we had yesterday morning about bilateral issues between Russia and the United States, where do we live? Where do we start from where we left off in the start? But all the way down to the questions that were being talked about later in the day yesterday, terrorism. And and as an engineer I am also looking for some kind of a threat that would allow me to convert a vision to a set of strategies, to a set of technical problems, to a set of technical programs, that could actually provide and these are not just technology programs, but also programs of diplomacy, programs of policy and approach that would be required you know, to provide the kind of the technologies and and approaches we need to get to the top of the mountain. And and and we have to be something quantifiable, not just describable. And the one I came up with is the one that we talked about yesterday in various forms and that's latency, because that connect us all the way from where we are talking about the alerting to where we are dealing with a a big about this threat that may come out of nowhere from a terrorist. And and in terms of latency that - what does that mean? Well it's a warning time, senator not talked about as a warning time yesterday, for those who are verifying agreements latency is a warning time, because it establishes their period of safety where they can respond, somebody can respond to some violation For the technologist for somebody trying to build bomb latency is how do I get from a particular state of capability, what is the value added I need to to do get an instrument of destruction or political power. Right, so it's looking at from below but at different perspective, its value added to to the country or the terrorist trying to produce this instrument. So it becomes a a very important issue to be quantified, qualified and and it does acclaim providing threat through throughout all our discussions. That said what we can say about verification monitoring that that hooks into latency, but also allows us to bring forward all the lessons that we learned, valuable lessons we have learned in the '80s, the '90s about verification and technology. One is that when you have a choice to verify or not to verify, always choose to verify. Because verification has two benefits, one benefit is to eliminate or to second it builds trust, the very fact that you engage with an advisor or your partner in verification brings a collaborative approach to a problem that allows you to think two things together. Such as what would be done in consultative commissions, it allows for building in - confidence building measures and therefore unilateral approach is may be good and very tempting these days, because they don't take a lot of time to negotiate, and I can do something overnight and I have to worry about convincing getting consensus in the CD, that I have to go some place. You know, so they may be very tempting and Moscow treaties is actually an example of you know, here is something that we can do unilaterally but it doesn't bring the effect that is really required in getting to the top of the mountain, building the kinds of trust levels, that you need and you hopefully would build over a face set of activities which involve verification. So where is the choice verifying, and again trust for the verifying does come into this. Secondly, look to technology to be able to provide transformational capability to enhance the confidence of what you are measuring and what you are assessing. Technology has come to the aid of verification many, many times. In the case we talked about in the early '60s, when we are talking about the CPP the first time you know, the very fact that we did know how to verify underground explosions, kept the nation to limited test boundaries treaties where we out loaded in the places where we can see it better or we still worried about what will happen underground. By the time we get to the mid 7'0s we outlawed you know, certain yields above a 150 years above a 150 kiloton but we have to worry what about those areas where, size may just isn't good enough. And then we decided we had a capability called CORTEX, Correlated Reflectometry for various time experiments, which allows you to have a crushing table in place with with the device sort of intrusive, but nevertheless an object that can be used in a verification process used with onsite inspection and again build, build confidence. And that allows you know, that took 16 years to negotiate. So it does take time, but but technologies are available. The third the third thing is that National Technical Means, if we look at a condition of universal latency, which means I have you bet with this materials all over the place specifically if you are really considering the growth and expansion of nuclear energy. There will be Plutonium in the world you know, the question is can I control it, can in measure it, can I account for every bit of it. You know, so in that kind of a universal latent universally latent regime National Technical Means are going to become absolutely indispensable, because it is only those kinds of capabilities that truly allow you to have the kind of situational awareness that's needed to make sure that somebody is not breaking out of a very arguably unstable regime or everybody knows how to make a thing you know, the materials around, but its just the matter of intent and motivation to get from one point two. So National Technical Means are going to be clearly essential. Am I not doing this right?

Speaker: You trust it but you didn't verify.

Raymond Juzaitis: I did verify this morning, I did. Okay, so the paper outlines, four strategies in the context of verification that actually link what nuclear weapon states need to do, what non-nuclear weapon states need to do? And all those things you know and these are the two bottom points that verification actually does not address but are extremely important, and we are eluded too yesterday. First of all we - as we all talked yesterday nuclear weapon states must return to negotiate the reductions and specifically verification regimes, but the job will be a little bit harder, we never had to verify warheads as warheads or to discriminate them necessarily from spurious radiation sources or some other spoofing kind of a source of radiation. And so the focus will be much more it will be a much more difficult technical problem but not insurmountable. We can start and particularly INF had already started long certain areas in that regard. For example we developed ways to discriminate between three warheads and one warhead on the SS-20, SS-25 when that was an issue at a facility that was producing both of them So so and the trilateral initiative at the end of the 90's and all actually it was set up to address the issue that were going to look at warheads and dismantlement next. So how do we design the process that we can be more intrusive, but yet still maintain security and having to do with the point designs, the nature of the designs themselves as well as vulnerability of facilities. So so a lot of these steps have already been taken and start, but with the new focus on warheads and dismantlement we have to take it to the next level. For the non-nuclear weapon states, there is also an issue of latency, and that is how do they stay and centralize the state the high states of latency, high is of course good in latency terms and low is bad. And and clearly strengthening the safe guards regime I think this is that mentioned yesterday particularly with emphasis on the additional protocol. And using technology in the additional protocol to expand the scope or the intrusiveness if you will of various measurement technologies that allow you to infer not just the correctness of the declarations but also as I said is the completeness then I with some sense of accuracy or confidence tell you that there are no non declared activities going on in a region that can be more important possibility that even the things that I am measuring so so again there is a whole set of technologies that could be given to the hand of international inspectors to transform their capability to monitor that kind of a regime, so the AP is clearly structuring the the NPT for strengthen the future. In the last two things I am unfortunately not really verifiable at this point as we talked yesterday when we talked about the zero option yes there is a fissile material control treaty that allows you to go the verification path and it can be in fact be verified. What about all other material out there that's in its in the noise its in the uncertainties of production processes just given the fact that you have any kind of material in a process of production, you are going to have production losses you know, little known divergence of a highly noisy background in which you have to look for needles in the hay stab diversion against the very noisy background. So so these or may be not verifiable at this point the way we talked about verification but nevertheless all subject to monitoring and clearly not national technical means And finally in terms of compliance there have to be credible response options as difficult as it is if you think about the holistic missile defense in today's context because it has has this product yesterday it actually provides a barrier to getting to lower arms and if you look at it from the top of the mountain you may want to have that to establish the defense dominance, so that you can control or deter any kind of break out from a universal rated regime you know the problem is how do you how do you implement in the way that's not threatening and get it the way you needed and clearly credible response options that have to be included and we think about them now as well is what happens if I actually introduce a nuclear device out of the move some place because we are in fact establishing these rings - security rings of factors around the country and the around the globe we actually do and we need to be able to with confidence be able to disable any first of all infer you know what the nature of that device is disable it and make sure that the consequences are as low as practically achievable because of somebody mentioned yesterday consequence management has to be considered in this kind of regime at some point. So so those are the four strategies my obligatory I am sorry wrong way. So so I have taken this from the National Academy of Sciences latest latest report and I just superimpose some bullets on it this is basically the justification chart of a nuclear weapon from nuclear materials all the way to an operational deployed organ and if and if you know yesterday we were really focused on this level which is we are really talking about things that are of very low latency you know on the other hand if you convert everything through a methodical process into material the highest states of latency you can have you know you can look at break the problem down into material balances at each of these states and there is even a way of of calculating you know the latency state of a stock pile by simply assigning latency factors to materials that are in various boxes integrating them and dividing them by the total sum material that's in the box So so there are ways of of approaching latency in a quantified sense so six pointing to the clock and I am just wanted to be just in the obligatory technology slide I typically went in, this is this is what I would regard and this is under development as a possible transformational technology its the ability to use a very penetrating possibly the most penetrating radiation known to man neutrinos which are extremely perhaps and carries a lot of energy of the universe which is the subject of a lot of Nobel quality research these days you know, can we use those to actually peek inside a reactor through the pressure vessels through the containment structure and be able to infer the quantity of plutonium in a in a operating reactor without without interfering with the process of reactor operation. And so the national labs have demonstrated this kind of a detecting technology in fact one has been operating at San Onofre Nuclear Power Plant in South South California for one and a half years and it attending, and as actually inferred the fuel loadings that you know, when you swap out fuel at the end of a refilling period there is a build up in plutonium and a drop in uranium and the technology has been able to to predict what that fuel loading will be, because of with successfully discriminated between Plutonium fission and Uranium fission So there are technologies that can be used and I will leave with at that - I just leave you with a set of questions just to start off the discussion. These are the issues that I think, there are still nagging issues is it realistic to think we can negotiate the kind of Invasive Verification Protocol to insure the degree of vigilance that were require? Can we in fact engage a coordinated mission driven technology development program, something like you know, went to the stewardship program that will allow the laboratory still look at the technical aspects of the zero option? Can we in fact maintain proper firewalls between intelligence community which we know is going to be an important part of the solution to this problem as well as an international inspectorate which is also part of the solution and finally as we know this thing in Homeland Security now you know, the ability to snoop around you know, sometime bucks up against time warning principles of personal privacy. And - there is just at least questions can a degree of vigilance and awareness needed to verify this kind of a regime in harmonize with the foundation of a free society and the sovereignty of independent nations. So those are, that's it.

George Shultz: Now I am wondering lets see, I actually had a question, you had a point?

Speaker 2: I had a I had a comment which if is the you all comment that goes back to the previous discussion and trying to connect a lot of what we have been talking about in the last couple of days and you know exactly the right time to find Mr. Jeanloz, this is a right time that you if it comes that it was a caging by the question that if Kenyan or JCS on a CBT which I think take all and raised, and I I just want to describe where these issues fit in the defense department today for those who don't spend as much time in that department as I do and and all the issues we are discussing mostly are not only not only not on the backburner or they are not even on the stove today in the defense community, the nuclear issue, Robert Gates that when he came in to office assessments to priority equality he said Iraq, Iraq and Iraq. I mean it was a very sensible, you may limit that let us find the answer that any of us would give under his name. Circumstances then you go down the chain underneath then Gordon England the Under Secretary for Acquisition Technology Logistics is next person in the chain and then the assistant to the Secretary of Defense, all those jobs are empty and I know for effect that those people anything with except below the undersecretary ATNL of level all most never received or have time to receive briefing about this issue. This too many other things, so to do on the uniform side pretty much the same, the important voice today Dick would not be the Joint Chief of Staff corporately it would be the relevant Cocom who is the Stratcom commander. If those of you who have been kept up with the changes in the unified command plan and so forth today's Stratcom is not our mothers Stratcom. It has a lot more to do there nuclear weapons the Stratcom commander has been giving responsibility for space, and our space programs on both the NRO side and the air force side or ____ you know, and that takes a lot of managerial time and attention. He is responsible for cyber the DOD role, Cyber Defense and Cyber Offense and that's a big responsibility takes a lot of monitoring time and attention going to be granted. New thing he is responsible for the new trial which includes missile defense which takes a lot of time and attention. And when it comes to strategic strike, global strategic strike there is a non nuclear part and a nuclear part and he spends most of his time on the nuclear part. So if you getting all that you see that the man of even the relevant code comes time and attention that you spend on this matters is is very small. I am I am involved in the investigation of the minor departs they afraid and when the story is told there I think that's kind of the that what happened there is symptomatic and the fact that nuclear weapons matters are just not a big deal today. In the Defense Department they are not the answer to any of the problems that people are feeling most hugely be at Iraq, Afghanistan and we set of the ground forces from pairing the air and maritme time forces for possible competition with China prosecuting the DOD role in the global war on terror you know, these with the big items way down here is is nuclear forces, and the conclusion of all that its - for this enterprise that seems to me its is is too far. The first is the that the relative - if this is ever is brought up and this administration or future administration for action any of these items we are going to find that lot of a relevant decision makers in the defense community don't have a lot of background so the people who would now the four stars and the three stars who you know, they were 06's when the Cold War ended and this issue went away and it just hasn't been part of their lives and so they are just not a similar one. And more broadly when that you first wrote the Gang a four op-ed and I was talking to Bill Perry about it, and I remember - so one thing I said to you which was I said. Who do you think you are listening? If you think about nuclear issues are sort of distribution, the only people listening were the tales of the distribution. That sort of antinuclear or sort of proud and pronuclear proud, the great middle which you want to appeal to that cares about National Security Affairs in a broad sense isn't listening, isn't tuned in, isn't even really very knowledgeable. And I think that's the challenge of this group that a you four gentlemen command the attention of people who think broadly about defense and there are sort of some hope in there, but I just want to make point that in defense this is not a big deal and its going to be difficult Dick to you know, really get people to pay attention in and reengage

George Shultz: That's been on the other hand remarkable from the the response to that editorial you know the countries in around this country until so its been as you said the issues being minor and that people not paying much attention and normally said the picture describes the core -

Speaker 2: You you can get the attention of that in the middle of the distribution and so forth about the course

George Shultz: You know so that's part of the problem. How do you how do you get something done and focused that obviously need attention and you know, hope it doesn't take a catastrophic event to do that that's -

Raymond Juzaitis: Actually and let me speak to that you know, a small part the Domestic Nuclear Detection Office was in fact set up you know, specifically to address the concerns you know, across many communities that weren't doing enough against the nuclear threat, and so the vise president mostly the wise president you know, was involved in setting that up. So DNDO has a quite a big you know, so if you look at who is spending money on these problems, but without a clear plan forward nevertheless you know, the DNDO clearly office of not proliferation, DOE is spending moneys on things like this the deal DEO distract defense type reduction agencies spending huge amounts of money. The problem is that it is not coordinated, and there is not an overarching plan of nuclear security apart from NNSA's weapons program.

Matthew Bunn: Right. Well, I had a sort of supplementary paper that's in your book about how to we actually sort of monitor the process of dismantling the warheads and getting one of the fissile materials. I don't want to say too much about it except say I think that's a sensuality moving down this pass toward very to the levels of zero in a confident way. And there is - there are now technologies that where some further refinement and negotiation can do reasonably well and conforming dismantle and without revealing sensitive information. The US and Russian scientists have been working together to develop and demonstrate technologies like that for a long time, so that there will be tools that are familiar to the technical community on both sides available when politicians decide they want to take up those tools. I think in more to not focus too much just on you know, can we confront that a particular number of warheads in this manner, I think that's in a certain sense although its challenging that sort of the easy part, I think you really need to think about how to why understand what's the total picture of how many warheads are and how much fissile materials there is and right now we have a couple of little what I call islands of transparency like for the high uranium purchase agreements in a sea of secrecy. And what we need to move toward overtime step-by-step is a situation where you have a sea of transparency with some islands of secrecy in it, and if you have declarations and tax and seals to maintain chain of custody and monitoring the sampling at many different points in the life cycle overtime you can build up an increasing level of of trust and confidence that the declarations that have been made are more or less accurate and complete. You can certainly do a good job confirming that the clearance.com. The issue which - I think it's a great paper but I think one thing one thing I would add to the paper for you is a more explicit treatment of the question everybody asks about zero which is head into another 10 decades somewhere and the answer is that you know there is there is no technical way you will ever be able to confirm that you know down to that level of warheads or fissile material. But overtime you can build more and more confidents at the other guy and totally the rug profiler over on there. I think you do need to move fairly quickly to take warheads that you at a certain stage consider access put them in to secure storage that's jointly monitoring that committed to enter this mantel, therefore many were and typed you can actually verify the disable of tending the eventual this mantel because of some limited rate at which you can dismantle. Then dismantle the warheads with some model of verification put the fissile material in secure and monitor storage. If we don't - if we don't secure the fissile material properly we are going to be worse of from a preparation point of view than may be would have been a we thought about warheads. And then ultimately reduce the fissile material stocks you want to get to allow where you have a low agreed number of war heads at a particular stage and fissile material stock that are no larger than those needed to support that agree or liable a warhead. One issue that's going to come up about that as we go to greater levels is naval fuel, because we have well over a hundred tons of HEU on this country that is a sign as a reserve for eventually we use as naval fuel. At a level of thousands of war heads around right now who cares If there is a hundred tons of HEU set aside for naval fuel, but if you add a situation where each side is only allowed to have say a couple of hundred warheads a hundred tons of HEU is allowed at potential capability that settle there and so as we were to doing toward the future I think we do need to think as Bob is saying about what we are going to do about the naval fuel issue -

Raymond Juzaitis: Can I say something you oh you want a -.

Matthew Bunn: Yeah, let me just say one last thing and that is there are things what's going to take a lot of time to dismantle 25000 warheads obviously, there are things you can do in a hurry, and I think that as I was saying yesterday there are initiatives that you can imagine where by where is the Untied States and Russia would designate some thousands of warheads and I hope including the ones that are have the least impressive controls against on authorized users if there are still and put them in to jointly monitored storage sort of that essentially being under jointly monitored lock and key and that could be done in a period of months and then you will work out the details of how they will eventually be dismantled on that on a more relaxed timescale

Raymond Juzaitis: There is another aspect that I didn't bring in is that if you know you are going to the top of the mountain and then nuclear material accountancy will be the lengua franca. Then there is a way of back and saying how do I there remain involving nuclear energy program across the world in a way that gives me a chance to monitor it better. And - do proliferation resistance in a technology way instead of just a talk lets talk about it way, there are ways that we can and it was lead to you yesterday new new reactors, I mean naval reactors will tell you what the GV you know, we need that HEU. Because that's what allows us to spend those years you know without refilling. But, can we move to a different kind of a reactor there with in hung you and I I don't think its going to LEU. You know, it may be in some other way or you know, it may be a fast reactor. You know, we could be burning plutonium up in in naval reactors, you know. So so there are technologies, and and that is the sort of the tragedy of GENAP. GENAP was you know, I I won't get into the political statements about GENAP, because I was a fan of GENAP as a Nuclear Engineer. When I saw GENAP on the that was like they said was oh my god, they finally you know, we can join the rest of the World you know, in a in a sustainable approach to nuclear energy which deals with the backend of the fuel cycle. The problem is that it was commercialized immediately. You know, they lets lets start building plants. You know, as opposed to what fuel forms do we really want to go after? You know, what is the degree of proliferation resistance it's built in? You know, can we design a safeguards regime with reactors like this that that has a system, you know, monitors. So, designing in from the beginning when you see a nuclear renaissance on the horizon is another way of controlling your materials as supposed to trying to catch everybody with some plutonium at the end of the day.

Speaker: Well, I disagree with. But what you just said is I think we by all the way to the top of the mountain But we are only half way up-

Raymond Juzaitis: It's just its more it seems more compelling when you have to go to the top.

Speaker: I think you were whole discovery has one is there anything that you can do to how to enrich Uranium in terms of chemical additions to make it more radioactive in the sense of terrorist?

Matthew Bunn: No. It's been pretty easy to tag it. But, in order to tag it, you can put for example a a small amounts of various isotopes that, you know emit you know penetrating careless. In fact, a lot of the HEU in the world already is a sort of a little bit tagged because it was made especially the Russian issue form reprocess to Uranium and it has a tiny little bit of U2-32 which decays to something without much more penetrating gamma than U2-35. You know, what I really had in order to do that, you basically have to process the HEU. And my view is that if you are processing the HEU anyway lets find it out unless you actually need it for something. In those cases where we are processing HEU fabricating and the fuel is something, I think we are to be moving toward you know tagging that in the course of of doing that, to make it easier to detect as its

Raymond Juzaitis: Right, but we have we have challenged our weapon designers and out reactor designers to sit down together and look at fuel forms and can we design new fuel forms that you know, in in a reactor design, there is this thing called the burnable poison; because you want the reactor to to last for lot longer. You you put in some poison into the fuel, it suppresses its reactivity and it burns out with time

Matthew Bunn: Poison not for humans but for neutrons

Raymond Juzaitis: No no, I mean, for for neutrons I am sorry. Again the process you know, so can you design something into it that when you take something out of a full out of a reactor flux, it builds in you know, a radioactive signature

Matthew Bunn: Correct.

Raymond Juzaitis: You know, and which is usually burned up if if it's in a reactor. But then, you take it out and it starts building up. So so that could you know, so there are but those kinds of challenges have to be looked at and and I think I am, you know getting our nuclear professionals together to look at problems like that in a common common mission, you know is something that I don't think we want to lose that opportunity, you know at this point.

George Shultz: Rose, I have been trying to get you into this

Rose Gottemoeller: Thank you, thank you very much. First of all, this is very interesting paper in this discussion but very interesting. I am on the same page with you where you, with regard the GENAP and and its phase. So, but that's a if I comment, I actually had a question to try to draw together the threads of our, - two discussions this morning Raymond's presentation made a lot of reference to what's going on kind of in the world outside, so that we keep monitoring that and enhances, compliments, supplements CTBT monitoring. Can you say the same in your words? Are there you know, University students looking at things that are you know, commercially available that helped. And how would you draw that that that into this?

Raymond Juzaitis: Well, there are and I am I am trying to figure out how to you know, are there the the one that comes most recently to mind and and talking to President [0:34:43] ____, if you are right, we we join the common forces that will we had we had about just students who who, a the University you know, who who met up with Russian Students at the you know, from the [0:34:57] ____ reactor design. And they traveled through Switzerland looking at reactors and they went to Zug, and their topic of discussion was fuel cycles of the future, you know. So so there is a a lab lab, University University kind of of activities that's going on. That's a little bit more sociological and maybe you are after

Rose Gottemoeller: Yeah. I am talking about verification applications. And transparency.

Matthew Bunn: Yeah, is its an obvious one; that is just exploding in the commercial world. The use of these interrogable tags, you know in a Wal-mart warehouse, you will have a bunch of boxes, that each has a - inventory and literally just walks down out with the little thing that just read is reading off all the stuff on the tags. He doesn't have to actually look at the boxes. We don't have that on the continuous for our nuclear weapons. I have no idea why we don't. But, since I got sort of obvious,

Raymond Jeanloz: Can I add a can I have your two fingers? We turn to think in terms of hardware and tags and seals, there have been huge events in the national labs of contributed also as I as I think you know there is a complementary versioning of a capability in handling the information and so, the database and so on. And in my view that's a quite revolution that's incredibly powerful, but all of the internet commerce all of the e-commerce has tied to the fact that you can securely transmit information. And quite frankly, I think there is a huge opportunity for applying those developments of the past decades to the arms control

Raymond Juzaitis: In fact in fact, to that point and and sectors shows to alluded to you yesterday; the information aid that is actually what makes this vision of an infinitely verifiable future where the credits will ____ even possible to think about it. And in fact that every every national lab I know, has now established its grand challenge programs in integrating called you know, called ____ is detection with information technology. And then with huge amounts of but but to do that, you have to integrate multiple modes of information sensors textual information, photograph you know, photo reconnaissance information in fact you know, electromagnetic signatures you know and and to basically capture it, look for discover the knowledge, and they are doing that sign kind of thing; and Google does it. There there are other companies that do it. So there are commercial technologies out there. I was I was amazed that a a few weeks ago, there was an article in 'Time Magazine' about Burma, that they were using commercial overhead imagery to locate what was happening to the monks? You know, - you know looking at patterns of traffic and activity to see you know, how how things revolving there. So so those technologies are really I think that is ultimately the transformational technology's linkages of censored development with with the information. You know, that's going to give you the high confidence to have actionable information you know, in a very dispersed way

George Shultz: Okay. Thank you.

Roald Sagdeev: I would like to make a very brief remark but in a larger context. I think what we are evidencing now after this - very interesting contributions by Raymond that actually we have now established to the new field of science and technology in support of non-proliferation causes. It embraces you know, not this it started which states make matters which are much analysis which is a neutron spectra knowing through these if its anti-neutrino and I think same [0:38:46] ____ plan. The very intimate contacts between the American National Labs and Scientists and Russian Labs on these specific issues. That initiated to that and support too financially by grants from ISTC the International Science and Technology Center. So and now and for example as a one byproduct, [0:39:15] ____ our audient; who is a Chief Fraction Nuclear Weapons Scientist and he a coach here he is a Rose Gottemoeller of working group. He was appointed a couple of years ago by a Russian Government as a Chief Commissar to certify that every new design of nuclear reactant reactor would be certified as a proliferation resistant. Its it's very important its went beyond I just wanted to give one another example. The [0:39:48] ____ of this new field of science and technology are or either they they are used by other agencies. NASA used to discover huge reserves of ice on Martian Poles, specifically using instruments whether it was instrument from Los Alamos Lab measuring neutron spectra coming from us and similar instrument from Russian side. And the latest example is, NASA now is landing project upon matching two years from now. And they wanted to do it in SETO. But they needed active neutron immune to penetrate to the surface. So, that was confutations or ration came up, these are formerly completely classified to neutron pulse neutrons. So, that's which is a little gadget going with very warhead, it could provide the cheap neutrons so, and they says, okay, and that's a finally before accepting it as an it's an instrument of light. They came to destruction company and said, "Do you did you have a track racket for expert of this device." So the Russians said, "You know, we actually re-fabricated something, it's like 30,000 for expert, but you and we both very lucky because it was not expert

George Shultz: All right I am putting the end. I want to go back to to Mark.

Mark Fitzpatrick: Thank you. I thought the the theme Ray has started off in the presentation is latency was a very useful construct to look at this and that most of the discussions was in the context of the nuclear weapons in the world Nuclear Weapon States, drawing down to zero. How the how latency it can be increased. In some of the discussions seem to bit futuristic, but I think there is a very immediate issue with with late and see that and has to do with Iran's capability. Iran is rapidly getting to the point of of having 3000 module of gas case they can spin full time. Yeah. But they are similar in question will be that they will be viewed as a latent nuclear weapons capability and for many of us most of us probably. If they if they have that capability, we have to assume that they will have a have a weapons program. And how you know, the concepts that you have discussed can be applied to increase the latency you immediately with the Iran program might be something if you were to do any revision of the paper might be given a little more emphasis

Raymond Juzaitis: Well, I think that that gets you in they immediately get you quickly into intelligence and National technical means. I think yes, some of these futuristic technologies on the outside are going to take some time to develop. But, - Clearly when you talk about latency that also says that, "I can make a latency assessment." Right, which means what's the time to a nuclear weapon? And I well, we have shown that we haven't been very good at that lately. You know, as a community even in the intelligence community. So so I think that's all source assessments of of latency in these countries should become I would think, you know, a significant I mean, we do National Intelligence estimates all the time. But are they focused enough, you know to give you some confidence about those statements and I don't think that confidence is rigorously assessed. I mean, in from from my, you know, I I managed some intelligence programs at the laboratory and you know, that was always a struggle you know, we we talk a lot about may be that some of that John can can talk about. You know, the trade craft and how do you protect it from politicization, you know. And you know, can you establish a rigor not unlike what the designers use to certify nuclear weapons. Can you certify that kind of a an assessment, with that kind of rigor that allows you to predict just that kind of a number would certainly to a National Policy Community. So

George Shultz: Actually John was

John McLaughlin: Actually the regent is something I wanted to say, are three quicker points about the intelligence dimensions of monitoring and verification more at the conceptual level and the technical level. First as I said to one or two to you yesterday, I think if we visualize this process, that we can work down here as a long term process, which I think we would to get to zero. I think you can see the intelligence and monitoring aspects of this as a curve, as in the sense of that descents again. And at the the beginning of the curve if you are taking about the first steps that we discussed yesterday, with increasing a warning time for example through origin between us and the Russians. This is something that our intelligence community understands while has the president for doing there is a body of people who understand it, and I think we could have high confidence that they would integrate well and quickly under this, in a way that we give the people the kind of confidence that Ray is asking for about monitoring and verification. There is a history here As you go further along this curve and you get to try to understand how we can deal with non-deployed weapons with nuclear explosive materials and such then we are into an area where intelligence will struggle. In apart because we don't have declarations, monitoring and verification doesn't work well with all these declarations, that has distinct for detection with a real pretty good at detecting and shielding things like that but you have declarations in those areas, you are starting to climb a steeper hill and when you get away from the traditional nuclear states and traditional arms control wings then, your are into an area were intelligence will struggle, it can do the job. But that the classic intention between monitoring and verification will start to become more account. By which I mean monitoring is the way I have always started with is literally reporting that you see and determine and verification has a huge political component too. And in that middle stage, I would think that the political component of monitoring and verification that is the capacity of political leaders to accept a higher degree of uncertainty or deal with it that will be more important and more acute than in the earlier stages of the classes where we are simply in that we are talking about de-alerting and and reducing existing start plans. Then as you come down the curve and we get closer to what Ray defines as universal latency we succeed. We were to build up the body with data, and practice that will begin to increase our confidence even though one or two undetected things would have greater ceilings, but intelligence would start to perform more confidently and better as you come down toward universal latency I think. So that's the curve And the second point I was making is that points that lie in that curve, conflict and our political leaders would have to wrestle harder than they typically have had to wrestle with verification decisions than in the past. You know, the third point I wanted to make is a longer line of what Ash was saying about the defense department. If we look at the Intelligence Community today it is well postured and highly focused on things like weapons programs, terrorism and the possible conjunction between them. That's their highest priority. It is not focused on arms control as we typically have known it for some of the same reasons that Ash indicated you went back to the period of the historic bench control trees are of large, this was a religion in the intelligence business. There are large numbers of people who did this all the time and thought about it every waking moment. The numbers of people who do that and think about it every waking moment has declined and what that tells me is as we go along here we will have to engage senior members of the intelligence community leadership fairly, early odd to say, there is a train coming down the tracks here. That's going to affect your priorities. In fact, that you have future meetings to this I would recommend inviting in addition to well, I would recommend inviting someone from director of national intelligences, collection leadership, so that they begin hitting in their heads that they have to collect some new tasks and priorities may be increasing if if we move down this this path. Now when you get to that point too, it's very important in my judgment because of the role increasingly that organizations like the IAEA will have that we have a very strong good relationship between intelligence and those organizations. And it it fluctuates with various believe it or not it wasn't the it wasn't Iranian decedent groups that the point of the IAEA towards in the times and as the hard, it was actually American intelligence. And and so there have been times when intelligence set the relations, but it did not have a good relationship say in the run up to the Iraq war, where there was detention between those two activities and when we did not get feedback that we were delighted to have IAEA and the what they would not find in finding. So the relationship there has to be very good between the unsighted specters of the intelligence community, particularly as you are in that so high point of the curve, those are the thoughts about the intelligence community.

John Abizaid: Okay that was it. You know, to that to what Jean saying and I think when you look at the nuclear battlefield, out there when you think about the future, you have to say in order to get the zero you got to have a great degree of situational awareness. And while our national, technical means may be good or intelligence capabilities on the asymmetric battlefield are actually, woefully lagging. And so there needs to be a demand placed on the system through the intelligence community and the military communities to increase our ability to technologically understand better what's happening in place like Iran, what's happening with a weapon that might go loose that travels around the globe etcetera and these gaps are actually very, very concerning and they haven't been filled properly. If we had a very difficult time even coming to groups technologically with the improvised explosive device problem in Iraq not alone trying to deal with the nuclear problem. But unless we put very, very serious demands on the intelligence in the military communities, to make this a national priority, I think it's very difficult to gain the situation awareness they you need to to get to the zero.

Richard Perle: And Ray said some thing, if I understood correctly in passing that the I don't want to allow to go on that more. I think you said right, that ballistic missile defense is a barrier to lowering numbers of nuclear weapons. Did I understand that correctly

Raymond Juzaitis: It could be constituted as a barrier to lowering number of weapons, -

Richard Perle: It could also be constituted

Raymond Juzaitis: It's something that we talked about yesterday, so I am just trying to repeat some of the may be what's before you hear that

Richard Perle: Yeah it had been in the morning. That certainly wasn't Ronald Reagan's view. Well, he regarded ballistic missile defense as the insurance that permitted you to take variable steps in reducing the numbers. And I think one - quite apart from that could argue that ballistic missile defense can be a powerful discouragement to proliferate. If you only have to get a single weapon in order to have confidence which you can deliver a single weapon, - that was all discussed yesterday.

Raymond Juzaitis: No, but in fact I wanted to bring up the fact that in that kind of a situation absolutely it would be.

George Shultz: Absolutely yeah.

Raymond Juzaitis: But its how do I get from now to there

George Shultz: It's no question of cooperation in the process I remember exchange where at Reykjavik where we had not only discussed eliminating nuclear weapons but also as you remember it should getting Ballistic Missiles down to zero as well. And Gorbechov said to President Reagan, "Mr. President, if we eliminate ballistic missiles, why do you need a defense against them". And President Reagan said well people know how to do Ballistic Missiles and there is always some roads straight out there that by getting a new and it will be as happy as we are and have a defense against that state. And we will share our missile technology with you and okay Mikhail Gorbachev well Mr. President, you won't even share milk technology with us Context is -- but anyway, that I think that this is what he says by that time was

Speaker: We will take that milky

George Shultz: Milk milk In other words willingness for transfer technology was very - but I think however it was but some roads today is that is going to produce certainly 3000 warheads that are going to match you which you might have to defend against but we are into this something and for you are - we have got out of order here Bob, I think you are next.

Robert Einhorn: Thank you. You know, that the the title of the section is that it the challenges that verification compliance, I think we have been a lot better at overcoming the challenges in monitoring and detection then we have been at overcoming the challenge of enforcing compliance. I mean look look at some of the main cases you know, the IAEI detected that North Korea had more plutonium and they declared and that began the that started crisis in the early 90's George has mentioned we we were able to detect the targets of then you know previously contested in Richmond facility. We had a - Libya we found a quick come dealing with the Libya. In each of these important cases detection monitoring investigation by an International Organization -. And the problem is yet you know its it's a question credit claim when supposed after detection what? And if the Iran case is a critical test plays. You know Iran is applying in nuclear weapons capability in plain side. Everybody knows it - everybody every one sees it happening but what do we doing about it I think that's a huge challenge and as you go down towards zero it becomes a major challenge unless you had because the security counsel is an important vehicle of imposing complaints the less you would have union amenity among the P5 you paralyze and we find ourselves you know in paralysis in the case in the case of Iran. There is a possibility of trying to put pressures on outside security counsel and we see that happening now the US taking leading putting financial economic pressures on Iran that it seems to be working but this can take a long time and it may not have a real impact if before Iran is managed to master to be a technology - centrifuge investment so it think for this group in particular I think its important that we look at where I think is the bigger challenge in technology is unlikely to deal with the help you there that's enforcing compliance.

George Shultz: That's next point you are next.

Sig Hecker: I have that my hand up to commit on Rose's question we later through the international aspects of working together and and what I wanted to say is in terms of the detection technologies we are not only how to go in the interest reporting with the lots of event as you well know a lot of things together with the Russian. So there is not a problem in working together in those areas internationally, however in terms of having that is to have a significant impact the bigger problem in Russia again though as you all know is the issue of transparency particularly as it comes to the nuclear material and in the in the early and then mid 1990's when secretary [0:58:03] ____ had requested the plutonium study the first 50 years, as part of that process she is specifically out of these classify the various important aspects related through this file and that is particularly the Isotopics of the nuclear materials involved and then including actually the chemistry you know, which starts to get down the recipe is that how you make this stuff. Although she left and she stopped shorted of the classifying what specific material are in specific weapon systems, and the reason that she did that when she was suddenly criticized for that but the reason that she did that is becoming, if you are going to do the verification aspect, you keep going to do this position and if you are going to work together with somebody else then keeping all of that secret is the significant impediment to making progress and so she thought it was more important that we classified that we would loose that centrally nothing in stand point of military capability I think that was the right decision and she initially have a deal to [0:59:14] ____ that the Russians will do the same and and he was not able to deliver and so the Russians that they still keep the Isotopic secret, they still keep the chemistry secret. You know we don't exactly what Isotopics they use, we know what chemistry they use, you know that's not a trade secret or such, however because of that it makes actually works together and know that feels problematic and so we need progress not only here in the technical cooperation, but we actually need the political help and there is being no movement in that direction since -

Rose Gottemoeller: Chairman could I just make a point quickly to follow up on this we have had some conversation within Moscow just in a last week with [1:00:00] ____ concerning the question of whether the Russians would be able to engage on nuclear forensic cooperation. And still he was very clear in saying isotopes issues still sensitively, we cannot talk about them, but I was very interested seeking that he began inching down the road saying that well if you want to study nuclear reference is together, let's use reactive plutonium. So it was a tiny - I agree you know they are not stepping into their sensitive area, but it was a tiny movement forward that I thought could begin to set down some momentum for working together in these low sensitive areas nuclear forensics being one of them.

George Shultz: Matt do you had a comment to -.

Matthew Bunn: Yeah a couple of things first of all on this issue of Iran and latency and and proliferation verification, more generally I just wanted to make people aware of their huge intelligences in monitoring challenge opposed one particular technology and that is the centrifuge. Centrifuge is a Centrifuge facility for making you know, bonds work with Uranium very year is going to be smaller and useless power than your typical Safeway, and its just - if if a country really goes out of its way its just extremely hard to find and - and as we move toward a larger nuclear energy establishment in the world there will be I think incentives to for people to try to figure out that should be even super to do enrichment, and provide might even turn out to be a smaller way and there there have been experiments already that involve only at the molecular scale, but essentially desktop size devices that are able to achieve very substantial levels of enrichment. And you could imagine that some day somebody might get on that and they could do that it you know significant quantities of material. And so I think we we need to realize that while the technology of verification is improving the technology returned we are trying to catch out are getting harder to find On the issue of the international inspection I think that's something we need to grab it with as it relates to the stages that we have been talking about and in particular at what point of for what parts of this system is bilateral US Russian verification adequate and at what point it really means international verification. Its being a lot of criticism from other countries saying how should - why should we believe that when you say that you dismantle thousands of nuclear weapons. Why should we believe when you know UN and the Russians are monitoring each other should we believe that that that demonstrates anything to us? Now I think there is a certain amount of this in generous and some of that, but I think for some of these it would be politically very important to get international inspection involved early on when we configure out ways to do that without compromise sensitive information. The United States and Russia and IEA launch in the late 90s and trilateral initiative to work out ways to point excess nuclear material under international monitoring without compromising sensitive information. And they essentially demonstrated the approaches that would doubt on legal regime that would allow that to happen but neither the US government or the Russian government is actually interested in doing it and so it does and what is sitting their ready to be picked up -

Raymond Juzaitis: Information information barriers.

Matthew Bunn: Right its ready to be picked up but I think I think its something we probably ought to think or about we report on in the context of trying to to establish our bonafide as we move towards the 2010 NPT.

William Perry: I would like to ask John and Ray end by the comment specifically on the question of detecting centrifugal because its an important issue in the map itself. Understanding that can be small and don't it much power only the other way we have in intelligence point of view of learning with recent book of him so we cant might be pursuing centrifuges

Speaker: Go ahead -

Raymond Juzaitis: While without there intelligence approaches you know, that there is something called Vibrometry that can be used. You know, and but again it requires some access you know -.

Matthew Bunn: You got to know where you have to look.

Raymond Juzaitis: You got to know where to look in these, there are electromagnetic techniques that that -

Speaker: About what monitoring where the countries are buying?

Raymond Juzaitis: Well that seeks - that's the whole, again that situation awareness and if you going to have this information based safeguards or verification regime you could be monitoring all these networks or and that is actually starting in the intelligent community, they are - they are putting a lot of effort now in to trafficking that you know, and in watching those -

Matthew Bunn: And it starting at the as John was saying more interaction between IEAE for tracking these networks to some people will talk to who wont talk to the US intelligence, and US intelligence who which has some information the IEAE doesn't have, but its problematic because as everyone in this room does the IEAE leaks like I said and so that working out the information sharing is we could just where one of

John McLaughlin: One of the things there are technical approaches that are promising but as you know, the problem is and not in a tactical means classically photographs, you can see a building which can see inside of it and the buildings getting smaller. So as with all of these things you need a synergy among all of the methods of collection. If you if you have second which allows you to hear what people are talking about, and you have some imaginary which allows you look develop some candidates and you have human which allows you to good human which allows you to understand what's going on inside the building or what people intend to have going on inside the building, you can develop a picture but you need all these things working together.

Sidney Drell: And a very simple think that to keep in mind is that the technology for give centrifuges just itself without being started is advancing so the efficiency really matter that what going to achieve get out of the each one of these columns, its going up I am really reminded too downlink the size goes down by motive mind without already is the most immediate technique - technical problems within the way of centrifuge technologies.

John McLaughlin: But it is but is thus take you to humite a lot I mean the problem with humite is good agents take a long time to develop and you can swarm the amount of project like you can swarm say get out of project with swarm imaginary out of project. They take a long time to develop, you don't have a lot of good ones who are absolutely reliable you like that more and you get them targeted on and with had good luck with humite on some aspects of centrifuges.

Sidney Drell: I think that's the answer by the way and so.

Raymond Juzaitis: But there is one of the peace you need something else, but the technology approach is to say okay at any given stage I may not detect one thing with a lot of confidence, but there is a whole string of events. Its very specific to a particular country frankly, because it has to do with their trade agreement you know, who is trading with them what they are doing you know, but then you can actually time correlate you know various signals. You know what do I need to see if my model of how they would do this you know its happening, you know do I see traffic in a certain you know you know do I see so its not- its not specific to it, can I see that particular centrifuge operating at that many RPMs but can I see other signatures at activity indicative of that kind of a process that I could correlate

John McLaughlin: Well the other thing is and correct me of the technology is changing that the last time I understood this at some level of the hundreds of pieces that have to go in to centrifuge, most of them can be machined by average machine to flat there about 13 or 15 that require special machining that AQ content. So you can focus on the things that you know can only be produced uniquely which is how we got in the [1:08:40] ____ network could contact work among other but there again human was able to penetrate is factories and is is supply chain and his financial chain. Whether this is in the book you read but

Matthew Bunn: And I think one of the things you need to do also is look at the people, because in any given country like this you know its not going to be a lot of say vacuum engineers for example and if you keep an eye on it you know, the people who go off to a foreign country to study one of the key topics involved in centrifuge technology, how to keep an eye on those people when they go back to their country and see you know what time they are going to and so on.

Raymond Juzaitis: In fact that in any of the counter terrorism program, that was actually starting to happen you know, it gets you in trouble sometime because you tap and tell how they use computers and means talking to who where you know on the other hand that easily translates and not in the environments well

Henry Kissinger: No I I want to ask a question because I find all of this extremely fascinating and enlightening I am learning a lot I am asking myself how does all this information get translated to other who make us will acts on it, and also how do you establish category that what you are going to do about it once you the once you have that information and indeed what can you do about it. And it may be that the political systems will produce enormous reluctance to accept the sophisticated information you are going to talk about like precisely because it would have consequence. So unless you have some discussion, I am trying to link what I am learning here to what presidents and secretaries instead has to do about it. And what happens to system that apparently expressions by what you discover that they may be rigid about how do it burned and when you take essentials for example, you get other countries involved in the US soviet or international inspection. They don't want that crisis so sometimes they may want nuclear weapons but as a general rule they they likely to keep things at a low level of tension. And I we can solve this here, but the other thing about it is as we elaborate these papers, -

Speaker: It's like the aluminum tube problem.

Henry Kissinger: Then to avoid all the examples some of the departments that it is renowned I mean alone.

Speaker: This you will get verifying intensions

Henry Kissinger: Beg your pardon

Speaker: Basically verifying it's intensions.

Henry Kissinger: Right

Speaker: Rather than the advice

Henry Kissinger: And I we tell you about marginal changes.

Speaker: This discussion happened in 1966 of the NPT in the

Henry Kissinger: It would be better at the - get information, they more marginal it will appear that when it first appeared.

Speaker: Harry but of the quality that could be used for centrifuges.

Harry Rowen: I am sorry what was the question.

Speaker: How many how many companies manufactured troops for centrifuges?

Matthew Bunn: Not a big number and in that five that five countries. 10 I don't know - it's like 20 year or less probably

Richard Perle: That's the question less than a 100 lets see you have good idea to start with the ones who do which is where I gather all the centrifuges that are now employed under few weapons have originated. We can - we can't even closely monitor 20 companies. I had been worried

John McLaughlin: Well, the other the other aspect you can hear is here to narrow the range of uncertainty and that's part of what the answer and this question is how does how do you translate to housing makers, secretaries of state in congress, your levels of confidence in what you are saying and because ultimately it's like you know, [1:13:09] ____ said, the role of intelligence is to narrow the range of uncertainty may not be the determinant precisely but the narrow the range of uncertainty, you can look at measurements, you can look at specifications. Otherwise insight into what they are trying to acquire and the specifications, measurements match something like the p2 centrifuges unknown series of specifications which then you you have a sense of respire surprise to get. But and they may be trying to make the tubes themselves. May we not may not know, we don't know how many people do make tubes as far as I

Sidney Drell: I can't resist to remark more and that is if you don't have process in the government involving the national security council, or a size advisory board, these things don't get through and I will go back in the history how Eisenhower because he had people to talk to like Kelly and [1:14:03] ____ got us moving with in a year to solve this price attack problem with with satellites some with U2s, and we go back to Henry's time when the whole idea of real time intelligence came up and you may remember in your office, it would be it would be mostly it was a channel that went around the system that wasn't operating and now you don't have any event. I can tell you from 50 of the experience over the last eight years up till 2000, up until the first year of this administration. There were a lot of things we saw that there were very [1:14:40] ____ brought up, technology changing rapidly, I think get in I think get in and then a lot of times I worried about, I don't see a system working that sort. We can talk all around, but at the administration anyone doesn't organize itself to listen when things like this we are going to we are going to lose

Raymond Juzaitis: And the community itself needs to enhance the rigor of the assessment. It's like the nuclear weapon designers after testing went away. Yeah I think this will go off, well how do you know. Well, I know, I am an expert, you know, so so what's stewardship basically forces to do is take an expert community. And start quantifying their confidence. So how do you know?

Sidney Drell: At every expert community somebody outside -

Raymond Juzaitis: Exactly, and you are seeing that use that same thing in the intelligence community, you demand latency assessments within certain standards of rigor and you have it reviewed. I think you would see that I have an obligation to people who patiently head to head and I will start with Harry, Tom and Joe, the next - .And

Harry Rowen: They want to raise the question of people who are not not part of the core, lets say Washington establishment. It would not be some very politically government but not admitted then certainly concluded private citizen's academics. You know, how in some domains, example economic matters is a lot more expertise outside of the government than inside. No question, political announces, what we are speaking, same thing. When we get to this sort of topic, you have been at this warning, it's not so clear, but I was struck by what Ray had to say about people who were I guess were not primarily interested in detecting they explained for their all reasons

Speaker: Actually they are studying earthquakes, the structure of earth and totally academic topics

Harry Rowen: - are the reasons they keep it but it's like a very substantial contribution. Now, there are other people out here, these people in this room rooms for example, it works and really important stuff and I think an impact. So that is a category, people not part of the government or not part of the court, contracted relationships is not sometimes what they have to say gets lost since in the case of aluminum tubes but this case I will give you other example, which is on this subject but I love it. John Perry had a student student a year some ago who addressed under your direction I presume what has North Korea making their way - illicit activities. It's a fantastic piece of - I mean this is somebody in this academy really used to contribute. So let this the category and what could be done it's not clear to make better use or encouraging more. This kind of activity, I mean I very conscious of the limitations of the establishment in the lot of domain it may be may be more technical it is hard that could people make and say can really work on. But it's still a very important category, and I just want to raise it

Tom Graham: I would like to ask Matt a question, Matt you said that enrichment systems, centrifuged systems are growing smaller along with intelligence, getting better. What's your your sort of sense of you know, the zero world where you obviously things have have changed had to change dramatically, to permit or to happen international cooperation that would be required we really vast increase in the intelligence and verification, access and so forth. How bigger risk do you think it really would be if someone to get aware with a program all the way to being closed to a weapon and and no one know about it. Then what is what is what is that risk in your sense

Matthew Bunn: In my own view, I am more worried about people who used to have weapons still having 20 or 100 squirreled away somewhere and that seems to me of much more substantial risk. But I do think as we move toward that world, we need to you know, it's a it's a world that always out and so you have to invigilate being different in a variety of ways not just taking the existing world that's objecting nuclear weapons and one of the things I think we need to work on is creating a norm of whistle blowing you know, certain sense around the world that you know, that it is an active patriotism, not an active betrayal, if your country is doing something that is contrary to the loss that your country has entered into to tell somebody about it. and it's going to be a lot easier in some countries and there in other countries where again you are again talking about a world that's always out, so you are I think you are going to have to have a world that has a lot more democracy and openness in more places than the world as they exist now, if you are going to get to that world with zero nuclear weapons. But I think I think in general there are things we can do now to start promoting the notion essentially proliferation was applying, almost all of the really important victories that we have had on whether it's interdicting it on still an [1:20:39] ____ your plutonium, whether it's finding the tons or whatever. It was much more often that there was somebody inside that operation who ratted on every body else and so many is out ways to encourage that I think it's a very exact

George Shultz: And those causes comes in with Joe, -

Josef Joffe: Well, there is some questions a brief description I got that intelligence getting better and better as we overlapped and synergized all that so if that's the case, I would come everybody believes that the Iranians have vast redundancies in their system, which means that is not just a few you know, half dozen targets where we have bomb. By the way, it's a question the question of proactive not anti proliferation has not appeared in this soon which I find well depositing, but how do we why do we why do we think they have so much regardless, obviously, it's our is our intelligence not as good or are they particularly clever in deceptive or are we forming according to the clever propaganda.

Speaker: Yes, I didn't realize that it's the redundancy again I can fear that sure What?

Josef Joffe: I think the - have discussions about you know, discussion about a military -- has not survived. There military people talk up dozens and dozens of 100s of targets, and we still would miss some critical

Speaker: I didn't hear in this -

Josef Joffe: No but I mean this part of the discussion which unfortunately is part of the - even if it's outside this world.

Speaker: That's because it's going against it's going against missiles and forces they are going to respond get started

Josef Joffe: That I understand, but I mean

Speaker: He is telling about the possibly it's poor, elegant

Josef Joffe: - enrichment, mailing, publication that's passed of

Speaker: Yeah yeah that's not like that's not like - our own word.

John McLaughlin: I think the answer is when you have a lot of there is a lot we don't know. And I have been doing I have been reading this for two years but the last time, I had an idea I would say that you can't establish the outer boundaries of their program and this is the leaf with their our redundancies is held where I don't know if you going to establish that do you want to know no Even knowing but knowing even what we know gets you to a very difficult military problem, but that's not in our discussion, how do you take it out or how did you delay it, you take it out or so forth, and even based on what we know, it's a difficult discussions.

George Shultz: In that you have to come back to the point that Bob made a while ago namely, you know, something that what you are going to do about and then how do you organize yourself so you could take some kind of action.

Speaker 3: I just thought that in terms of what Henry was asking, what are the policy implications of this it remind me what senator Nunn said yesterday that we don't know what we are going to know later on and we shouldn't be in a rush to adopt things and plans and steps that a lot of people going to be skeptical about and I think that that really is going to mean and it's a bottom line lesson I was just discussing.

Speaker 2: Now, that's exactly the same point. I I you just have his recommendation because you go in the next phase, but I don't think you can persuasively discuss verification and compliance without discussing courses of action that can be taken at different levels of information and for this circumstance where we imagine ourselves is real universal latency this phrase. There are about four different circumstances, one is that that it was pretended was today that China has scored away 20; well everybody else is nominal again to none. What you do about that Second would be not a big power but North Korea getting eight, doesn't change the balance you know, we still think we can defeat and that's creating [1:24:57] ____ but still change it's makes it longer and more difficult to that but that's a second case. A third case so those are two government cases. The third case is Al Qaeda getting the hold of a bomb. In this circumstance where we don't have nuclear weapons, now in fact our nuclear weapons don't really help us much in that circumstance today so that that's perhaps in case that doesn't really change materially in this new world. You got to worry about it, but you worry about it anywhere where you have the nuclear weapons And then there is the fourth case in which it would Iran and Hezbollah which our country uses a cut out this some of the attacks. But it seems to me those are the four circumstances, in terms of perpetrators and you have to ask yourself at the stage of suspicion, at the stage of conviction, at the stage of materialization in each of those threats. What are you going to do, and there are national courses of action, and there are international courses of action. And we have been talking kind of vaguely as though there would be some international course of action in each of those circumstances but I may be we can engineer one - some sort of collective arsenal, collective response for that new thing that it doesn't help to talk about the technology. It doesn't help to persuade people outside of the room talking about technologies and the [1:26:18] ____ and so forth without leaping through through courses of action. There was no paper on that but may be in the next phase, paper in that engineers have those courses of action would be would be useful

Sam Nunn: Okay. I think all of these interesting and relevant questions that have been talked about and opposed and they have to be given serious thought but I think we all have to keep in mind that at least from our point of view in hitting them out, we are talking about a number of trials leaving up the mountain, we are talking about hard ground and to get to that hard ground, then you are going to start thinking about the 20 weapons, the materials and so forth. The world is going to be far different, if we have taken all of these steps because others at on that half ground. That's not going to be the same world we are now. So to try to get the answers from what we do at the hard ground so that to get to the top of the mountain where you know, we can count the wings on angels, I mean it's an interesting intellectual exercise but some probably, some other generation is going to have that question with a hell of lot more technology and a totally different political plan. But if you assume we have taken these steps, and if you are not going to take this steps to get to the hard ground, you are not going to face those questions anyway, because you don't want to get to zero

George Shultz: I was thinking as Joe was discussing the various things that are around that we know and not know and so on and what to do because remembering may be not precisely as with respect to what we do. I remember that the Iranians were acting like they might mine waters in the gulf back in the late 80s. And the president of Iran went to the United Nations and he was giving a speech at the United Nations saying that they under no circumstance would consider mining putting mines in the gulf after all, that's where they live. At the same time our navy was taking moving pictures of the doing it. and so we publicized that boarded the ship, you remember this Richard, took the sailors off the ship, sank the ship, told the Iranians come and collect your sailors in Doha and that that stopped and so you don't have to throw nuclear weapons around and make a point. And sometimes, some thing is discrete as for and that would help a lot now, see in what I have some

Sidney Drell: No we have this talk about the CTBT and my secretary handed me this morning the statements and the congressional records yesterday turns out that the defense authorization bills for 2008 said would put the senate on directly to support of CTBT ratification. This angered that much of senators, mainly republican Senator Kyle made speeches to the I thought I would just point out five reasons that they are not going to get a CTBT One US deterrent cannot be maintained without testing. Two set the reasons with senate, rejected the treaty, you do so again today, it's that the treaty is not verifiable. Third CTBT is unverified ability needs a ban will not have uniform affects and thought by behaved, forth. a CTBT would damage the struggle against proliferation on the one hand the inherent unverifiable of CTBT can be expected to encourage road stake regimes to believe they could presume nuclear weapons and finally descended, rejected the CTBT because it realize the stewardship program is a corruption according to our way we can't maintain our whole logic and 40 senators signed that letter. That's what happened in the senate yesterday.

Sam Nunn: Forty senators signed that letter - what you just said?

Sidney Drell: Yes, but you got to the implication again that that this is a letter of senator Kyle, he calls the call the task finally in managing the defense authorization bill, there is something and something called Power grid 3122 section and 3122, which is as I said in the beginning, it's interpreted by these people as putting the senate on directly and supporting CTBT ratification without debate - why anybody would do that, I don't know. But that's what he is quoting as a reason where this letter with 40 name on it

Sam Nunn: Well, I think I had points out on the problem and I think you would you would have put talent in doubtful category, -

Sidney Drell: On the other hand - Henry [1:31:08] ____ and me saying -

Matthew Bunn: But this is typically troubling because what's people have their names on something like that it's facing so much more difficult and you know, 40 is enough to prevent that

Sam Nunn: Well, you know, it looks already - excuse me just give out your point

George Shultz: I just was going to say that going to - this way you got 60 already or you are looking for a seven more.

Speaker 3: That's right the lets have full. But the thing is that there is never any attempt in the Clinton years to educate and you need to to brief in you know, these people. You know, lot of them were turned, there was a letter inside that I found right former senators, a lot more turned but - who didn't followed that closely. At least 20 of them by that, they never got any briefings

Sidney Drell: That's up they had - but now today forty people put their names, this is to be a

Sam Nunn: Well, to Tom's point, it's a mistake to put something like an inability you have had hearings and you got testimonies from people like Raymond And laid the foundation because you get to the other side all rolled up and then they jump off and give every reason throw everything but you keep thinking at it when there is been no hearings on either side and this debate of was basically ignorant.

Speaker: Absolutely correct

George Shultz: Yeah okay. Well lets take a not too long a break and we will come back and turn to Jack