Jeffrey LewisIran, Immunity and Diagrams

By now, readers will have seen this strange little diagram foisted upon the reading public by AP’s George Jahn.

The significant issues are whether the document is genuine, a question I will return to at the end of this post, and if it is, what it means.

As I have written elsewhere, I believe there is very strong evidence that Iran had a covert nuclear weapons program through 2003. I believe that program has, more or less, been in suspension in the meantime. (Iran has continued civil activities which would create an option to build a bomb that Iran could, at a later date, exercise.)

Whether this document is real or not, I would very much expect there to be a number of documents containing calculations relating to the design of nuclear weapons. The release of this document demonstrates an interesting policy problem relating to how Iran might come clean about its nuclear weapons activities as part of any agreement regarding its nuclear activities.

Much of my policy commentary about Iran emphasizes the need to emplacing a verifiable “gap” between Iran’s option to build nuclear weapons and decision to exercise that option. In practice, this means that a sustainable solution would emphasize intrusive access by the IAEA over arbitrary restrictions on the number or capability of Iran’s centrifuges.

A sustainable solution must also provide for Iran to come clean about past weaponization work, a problem that is considerably more difficult to solve than might appear upon first examination. Iranian leaders must surely worry that any disclosure of a past program to deceive the IAEA will be seized upon to undermine any diplomatic settlement, especially given the relatively vocal group of people for whom nonproliferation is merely one of many arguments for regime change. (Recall the quote attributed to Paul Wolfowitz in Vanity Fair about Iraq’s WMD programs being “the one reason everyone could agree on” for attacking Iraq.)

Much of Iran’s early diplomacy during the nuclear standoff arguably reflects this concern. Iran eliminated facilities like the Physics Research Center at Lavizan, but also sought to prevent the IAEA from being able to understand the scope of the program, scraping clean Lavizan and denying access to Mohsen Fakhrizadeh. One interpretation is that Iran was seeking to come into compliance without deepening its predicament. (This is, I think, an implicit theme of Mousavian’s memoir.) This may seem duplicitous, but recall that we are dealing with a bureaucracy. Even in the case of Libya, in which there was a high-level strategic decision to abandon WMD programs, the disarmament was never complete. Jofi Joseph wrote a wonderful article a few years ago on the myth of the strategic decision, arguing that this is a process that involves coordinating various actors over time. Despite what Condi Rice said, we only think we know what disarmament looks like.

The challenge, as Pierre Goldschmidt and others have argued, is to craft an agreement that provides Iran immunity for past weaponization work. One of the more thoughtful observers of such issues on Twitter, Shashank Joshi, wondered why something so evident should be so hard. I found it difficult to compress my answer to 120 characters, but two issues spring to mind.

The first issue is making an offer credible.  One of my friends once pointed out, only half-joking, that the trouble with a deal with Iran is that it leaves the Islamic Republic in place.  There remains a segment of the US political discourse that objects to the regime in Tehran. For these people, the nuclear issue is a useful hammer with which to bang on the Ayatollahs, as it were. A verifiable solution to the nuclear problem takes away that hammer, while leaving unaddressed other concerns — some of which are quite legitimate. There are various accounts of Tehran’s support for overseas terrorist action, including links to the bombing of Khobar Towers and continuing support for Hezbollah. There is also the bizarre, alleged plot to kill the Saudi Ambassador to the United States. The Iranians deny involvement, of course, but the point is that the US and Iran have a larger plate of things to deal with than just nuclear issues. (And, no doubt, Iran has its own list of particulars, starting with those motorcycle bombs and support for the MEK.) Nuclear weapons are not the reason that Washington and Tehran don’t get along, they are merely an enormously dangerous complication. We have to imagine an agreement that can sustain what are likely to be very serious Iranian disclosures about past weapons work with what will seem like limited accountability and in this rather poisonous context.  That will require some bravery on the part of the Obama Administration — a commodity frequently in short supply among politicians and civil servants.

Then there is the challenge of convincing the Iranians. They are aware of the foregoing, starting with the motorbike bombs. The narrative they have constructed about the IAEA is one of cooperation that has not resolved their problems, but mere resulted in moving goalposts. (The Iranians exaggerate, but the claim is not baseless. I supported the very unpopular IAEA workplan for Iran largely because I thought it was important to show daylight at the end of the tunnel. The issue of the weaponization annex arose largely after Iran agreed to more safeguards.) Iran, too, has made missteps. Regardless, the Iranians have convinced themselves that cooperation with the IAEA should be largely tactical, because further concessions will open new fronts of pressure. Copping to a serious effort to break out of the NPT would require an act of political bravery on the part of Iran that I am not sure I can recommend.

So, there is the problem: Iranian officials are more likely than not convinced that coming clean about past activities will worsen, not resolve, the current standoff. What’s worse, I am not sure they are wrong about this. I understand the intellectual value of amnesty or immunity, but I am not sure I can envision the mechanics by which that process would unfold.

Which leaves us locked in the current standoff, marked by a fundamental lack of faith. Too many Obama Administration officials, as I have argued previously, believe that negotiations with Iran are futile. They have crossed the line into a kind of cynicism in which the goal is to “win” each encounter, deepening Iran’s isolation to play for time and set a precedent for the next case. Iranian officials, on the other hand, have convinced themselves that the United States and other Western powers are using nonproliferation as a pretext to undermine them. They, too, are seeking to engage tactically, hoping to blunt the damage from each encounter.

It’s not a very encouraging situation, which is I suppose why I haven’t written much about it.

Crowd-sourcing the AP document

Well, I have to say I am skeptical of this diagram released by AP. As I have said, I think there are such documents. I am just not convinced this is one of them. Among other things, I don’t think it “looks like” real diagrams relating to nuclear weapons effects except in the most general way.

Then there are the numbers — 50 kT is too large, as David Albright has noted, while 2 million kilotons is, well, insane. The back story that has emerged to explain the errors in the document is sufficiently convoluted that I won’t bother to explain it.

I still haven’t seen the whole document, but I wanted to offer another possibility.  The title for the chart does not seem unambiguously nuclear to me — something one might expect if the author “intentionally simplified the diagram to make it comprehensible to Iranian government officials to whom they were presenting it.”  The title is, apparently: Changes in output and in energy released as a function of time through power pulse.

One possibility, which I suppose is unlikely based on the description of the unreleased document, is that it is possible the kT is something else — possibly a measure of energy at the molecular scale.  In such a case, k may be Boltzmann’s constant while T is -time- temperature. The unit is the product of the two — k times T. Now, as it turns out, this is not a sort of physics with which I am well-acquainted. So I am offering that observation in the hope that someone who works in the right field might be enticed to offer a commentary.

Comments

  1. Cheryl Rofer (History)

    A friend who has worked on containment of nuclear tests told me that the standard abbreviation for kiloton is kt, both letters lower case. A second friend confirmed that today. The first suggested thermal kT as a possibility, as you do.

    Also as you note, the caption on the figure does not unambiguously refer to nuclear weapons. In fact, I have seen many figures like this for laser output, giving both peak power and total energy, two numbers that people who work with chemical lasers or laser chemistry are always interested in. Such people may also talk of energy in terms of kT, which comes in a variety of units. It may be more convenient to use different units for energy and power, which would be explained in the text or a missing part of the caption. This could be the explanation for the apparent discrepancy in numbers that Yousaf Butt and Ferenc Dalnoki-Veress describe in the BAS. The way to test this would be to substitute values of k in various units to see if the numbers can be reconciled. I will not be doing that this weekend – something about Burma and Estonia on my mind.

    The time scale of the graph is consistent with that of a chemical laser.

    I also wonder if the power output from a nuclear explosion is in the form of a Gaussian.

    Some more attempts at out of the box thinking here. Scroll down for the discussion.

  2. Amir (History)

    The problem with a long term solution to nuclear issue in middle-east are Israel, Pakistan and US/NATO bases in the region. You can’t expect Iran or any other sane player to abandon the nuclear option while its enemies (or potential enemies like Pakistan) has access to the same thing.
    The solution is to actively (through sanctions and strong political resolution etc) promote the idea of nuclear free region for the whole ME + Pakistan and Central Asia region.
    Unfortunately, US and the west, hypocritically, are against such a move since it is not in Israel favor. (Just see how US can be a hypocrite when it comes to Israel. For example look at its vote against Palestinian people)
    As a result, I think , in the long run Iran will become a nuclear state and there is absolutely nothing (including war) that US can do about it.

  3. Mark Gubrud (History)

    The hypothesis that kT means the thermodynamic quantum does not make sense of the graph, because in that case the energy scale is still wrong by the same factor. And it’s hard to think of a context for which this would be meaningful. If a system is thermodynamic and you pump its energy to 50 kT, you expect it to relax back to ~kT. And it is clear in the graph that the energy starts at 0, not at ~kT.

    On the other hand, the hypothesis that the energy scale is J/10ns makes sense if the graph is the output of integrating a crude model of a nuclear explosion using 10 ns time steps. The axis is labeled kT/s because the author intended to convert it to kilotons/s but either forgot or didn’t bother to do so, or because the graph shown is a an intermediate version that was saved before being converted.

    Do we expect Iranian scientists to care whether the official symbol for kiloton is kt or kT?

    A lot of people seem not to have noticed that the bell-shaped curve is not symmetrical, so this does not come from an elementary calculus text. For an actual fission explosion, the leading edge is defined by exponential neutron multiplication, the trailing edge by disassembly plus the immediate decay of fission products. I don’t know what it should look like but this looks plausible.

    People keep saying you can find this curve on the internet or in textbooks but they don’t say where. I do agree that it could be the work of a student. I don’t think 50 kt is too large, especially if it is a purely theoretical exercise.

    Sure this could easily be a hoax but it isn’t obvious that a hoaxer would be more likely to make the energy scale error than an actual scientist or student would.

    • Allen Thomson (History)

      I hope that Carey or George will weigh in here, but would say that the time scale looks plausible (a few microseconds) and the yield of 50 kilotons, while largish, isn’t outlandish. No idea about the shape of the power vs time curve in an actual bomb.

    • yousaf (History)

      Mark,
      We link to 2 places where such figures can be found in our Bulletin of the Atomic Scientists’ piece — there may be more: tenths of microsec for the final few generations of neutron multiplication would be about right:

      see Figures 2.11 and 7.84 linked to from here, in the 3rd to last para —

      http://thebulletin.org/web-edition/op-eds/diy-graphic-design

      As Mark notes, even if it is a chemical laser (or anything else, kT, KT, kt, whatever…), the error is mathematical: the time-derivative of the energy curve does not yield the power curve. No matter what this diagram has to do with.

      It is a math error, not even a physics error.

      The graph is internally inconsistent.

      BTW, If there are lawyers reading such wonk, they may wish to weigh in on whether computations of nuclear yield, such as may be carried out in a graduate-level physics course for illustrative purposes, are in contravention of the IAEA-Iran Comprehensive Safeguards Agreement?

      Prof. Frank von Hippel has weighed in over at the ACA blog (armscontrolnow) that such a diagram could be made by a grad student or advanced undergrad. We concur.

      Incidentally, the student should fail the class for making such a monumental error as this graph shows.

      If this graph was made by Iranian scientists they should be fired. (Sadly, according to the AP’s latest article, it appears they were, but by the Mossad.)

    • Mark Gubrud (History)

      Yousaf, I don’t think either of those two figures is the same as the AP figure, although they are related. They display different curves and different variables. I haven’t seen the figure that the AP figure could have been copied from.

      I didn’t know that 10 ns was called a “shake” but it’s the kind of “error” i.e. in putting together a report or presentation which results from neglecting or overlooking a step, not from a lack of technical competence. And again, I don’t know that it’s the kind of mistake a forger is more likely to make than an actual scientist — or student.

      Bottom line, though, is that this could be the work of a student, it could be a forgery, or even if it is the work of an Iranian scientist involved with their nuclear program, it does not show a very high level of work; it could be a very general discussion and it certainly does not prove that Iran has an active weaponization program.

    • Dan Joyner (History)

      Just to respond quickly to Yousaf’s legal question, I can’t see how possession of computations of nuclear yield would be a safeguards issue at all. So no, I don’t think it would constitute safeguards noncompliance.

    • yousaf (History)

      Mark,
      it is not an exact replica, but Fig. 2.11 in The Physics of the Manhattan Project shows the integrated energy release versus time for the fission process. (The link is in the Bulletin piece).

      The time-derivative of that gives the power, as you know.

      The figure is for a 12 kt bomb but otherwise all the information is there publicly available.

  4. Rob Goldston (History)

    If you replace the left hand axis with Joules/shake, classic Manhatten Project speak where 1 shake = 10 nsec, everything makes sense. I digitized the image and the integral of the power curve is spot on the energy curve.

    • yousaf (History)

      Agreed — if the error is corrected, then the error goes away.

      Remaining question: even if it were correct, would such a plot be in contravention of the Iran-IAEA CSA?

      And if the answer is yes, would this Brazilian PhD thesis also be in violation of the Brazil-IAEA CSA? “Numerical Simulation of Thermonuclear Detonations in Fission-Fusion Hybrids Imploded by Radiation”:

      http://www.jb.com.br/pais/noticias/2009/09/05/a-explosiva-descoberta-brasileira/

      Lawyers?

    • yousaf (History)

      PS: In my view, the main problem is not that the plot is actually wrong, but that even if it was right, it is not a big deal — pls read the BAS article linked to above.

      Such a graph is not a difficult thing to make and does not indicate a nuclear weapons program at all.

      Even if it is right.

      Which it isn’t.

    • anon2 (History)

      Jeffrey,

      Do you have a link to Albright’s analysis. This is actually what I suggested yesterday.

      Rob Goldston,

      Do you have a link to a “shake” in Manhattan Project speak.

      Thanks,

      Anon2

    • Jeffrey (History)

      David sent me an email; I’ll have to ask his permission. He will probably publish something soon.

    • anon2 (History)

      Rob Goldston,

      I found shake in Wikipedia. No need to reply to that.

      I also found you — former head of PPPL. Very impressive! This is why I like ACW.

      I’m not so famous,

      -Anon2

    • yousaf (History)

      Albright was quoted in the AP pseudo-retraction.

      Shake is here:

      http://en.wikipedia.org/wiki/Shake_(unit)

  5. Ben Gamari (History)

    Just a minor correction: the Boltzmann constant has dimensions of energy per temperature. Consequently, T would need to be temperature here, not time, to produce the proper units. While this is an interesting hypothesis, it does seem very unlikely that this is the correct interpretation of this diagram. A nice analysis nevertheless.

  6. Mark (History)

    Jeffrey,

    I agree with the vast majority of your analysis and appreciate the points you made about the need to demonstrate to Iran that there’s truly a way out of this mess. But I’m going to mischievously nitpick on this line:

    “A sustainable solution must also provide for Iran to come clean about past weaponization work…”

    To what extent is this actually true? One might argue that the litigation and re-litigation of accusations stemming from evidence of a decade ago is a big part of why we’re going nowhere right now. Ambiguity can be an asset diplomatically, and often is.

    Ambiguity is problematic to the extent that the issue is reduced to a matter of “enforcement” of legally proscribed behaviors. But is this the right framing?

    For the sake of argument, I would make a couple of points: (1) whatever Iran has done in the past, even if by “past” we mean “yesterday,” does not mean necessarily imply path dependence (i.e. that the research must necessarily continue regardless of how the United States – to whom resistance appears to an important part of Khatami’s legacy – handles the issue); and (2) the protectors of the NPT regime in this instance seem to reject the idea (wherever convenient) that treaty compliance is intrinsically and indefinitely beneficial to security, and take their treaty accession and compliance, predictably, a la carte, while taking an absolutist approach towards Iran, as if to say, “you signed the treaty, you sucker: spare us your security concerns and just comply.”

    Without getting into an impossible debate about which side is “wronger,” the question is how much do we really need to know about past activities to move on from them diplomatically? It’s not clear to me that we need a full confessional. After all, proving the ongoing non-existence of “capability” (let alone intent) is a really hard thing to do.

    • yousaf (History)

      Mark,
      great points.

      BTW, on page 291 of El Baradei’s memoir has a relevant tidbit to information we are getting “from a country critical of Iran’s atomic program”:

      “The Age of Deception: Nuclear Diplomacy in Treacherous Times” By Mohamed ElBaradei

      see e.g. google-books:
      http://books.google.com/books?id=OO4C9MHRG_kC&printsec=frontcover&dq=memoir+baradei&hl=en&sa=X&ei=TVG6UKWjBsHViwLeoIGYDQ&ved=0CC0Q6AEwAA#v=onepage&q=%20%22numerous%20questions%20about%20the%20%22&f=false

      p. 291

      El baradei writing about documents provided by Israel in 2009 “purportedly showing that Iran had continued with nuclear weapons studies until at least 2007.”:

      ElBaradei says that the Agency’s “technical experts” had “raised numerous questions about the documents’ authenticity”, and suggested (in the footnote) that US intelligence “did not buy the “evidence” put forward by Israel” in its 2007 National Intelligence Estimate.

      I am not sure if this graph and other evidence are the same things US Intel evidently discounted earlier.

      Maybe Mr. Jahn — or other investigative journalists — can find out?

  7. Cheryl Rofer (History)

    I’ve been thinking along the same lines, Mark. You and Jeffrey makes a number of good points about “coming clean” about past nuclear programs. Not all the members of the NPT have done this. Sweden is one, and I South Africa may be another. (Andreas Persbo had a couple of nice posts here at ACW on Sweden.) So there are precedents for letting sleeping dogs lie.

    • blowback (History)

      Perhaps we should also be enquiring into which countries have supplied nuclear weapon technology and when to Israel. From what I understand gun technology requires no testing but implosion technology does, so who tested Israel’s weapons for them and when? Furthermore, it is rumoured that Israel has enhanced devices which almost certainly requires extensive testing so who provided that or the knowledge to by-pass that testing and when? It is a murky world out there and what the P3+1 countries have supplied to Israel should be made clear. Russia and China, I’m not that bothered about, but I would like to know what the United States, the United Kingdom, France and Germany have supplied and when.

    • yousaf (History)

      Cheryl,
      regarding “coming clean”, there may also be the possibility that some of the evidence that Iran being asked to come clean about is indeed fabricated.

      See my post on El Baradei’s take in his memoirs on some such “evidence” just above.

      So, perhaps sometimes at least, when Iran is saying that they are being confronted with fabrications, that may be true. They may not be able to “come clean” on things that are not true, except by saying what they are already saying.

      Also, we should keep in mind that the IAEA itself recognizes the limits of its legal ability to pursue all such evidence, whether true or fabricated. e.g.

      http://www.iaea.org/Publications/Documents/Board/2006/gov2006-15.pdf

      “…absent some nexus to nuclear material the Agency’s legal authority to pursue the verification of possible nuclear weapons related activity is limited…”

      Thank you for pointing out the case of Sweden — I look forward to checking that out.

      I was also struck that although Egypt, South Korea and Libya were also found non-compliant that only Iran was referred to the UNSC. Seems like more consistency all around would better.

      And, indeed, letting sleeping dogs lie may be a good path out of this quagmire.

    • Andreas Persbo (History)

      Cheryl. I think it’s fair to say that South Africa’s come clean as well. I hope to be able to write that story one day as well. It’s quite pleasing that Sweden and South Africa many years later joined forces in the New Agenda Coalition.

  8. Mohammad (History)

    One of the most balanced pieces on Iran on this blog yet. Thanks Jeffrey.

    I would add another caveat to the “coming clean on the past”, which is Iran’s domestic politics. Admitting to past weaponization work would seriously undermine the legitimacy of Ayatollah Khamenei and by extension, the Islamic Republic, since he has repeatedly said that Iran is opposed to nuclear weapons as a matter of principle (if I’m not mistaken, he has also said that Iran has never been after nuclear weapons). The domestic reputation of Iran’s Supreme Leader is carefully managed; you seldom (if at all) hear grandiose claims or – domestically speaking – politically incorrect talk by him, unlike e.g. what sometimes Ahmadinejad says. The public image the Islamic Republic tries to nurture of the Leader is the ultimately reliable and clean politician.

    This is why I believe that Iran will never come clean on past weaponization work (if it indeed conducted them) and also will never declare itself a nuclear weapons state (in the worst case, it will adopt a nuclear ambiguity policy). Too much reputational capital has been invested – both domestically and internationally, but especially domestically – by the Islamic Republic to voice its fundamental opposition to nuclear weapons.

  9. kme (History)

    2 million kilotons might be “insane”, but Edward Teller pitched a bomb 5 times that size to the GAC.

    • Jeffrey (History)

      I wish my comments had a like function. 🙂

  10. yousaf (History)

    Bob Kelley, a former IAEA inspector and a nuclear engineer has an interesting view on the graphgate situation.

    Not that I agree with him 100% on everything, but his views are certainly worth taking into account — kindly see last sentence of this article, and other quotes interspersed:

    http://truth-out.org/news/item/13087-fake-ap-graph-exposes-israeli-fraud-and-iaea-credulity

    “We can only hope that the claim that the IAEA has relied on this crude hoax is false. Otherwise their credibility has been shattered.”

    =============

    “It’s clear the graph has nothing to do with a nuclear bomb.”

    ===========

    “The pretty, symmetrical bell shaped curve at the bottom is not typical of a nuclear explosion but of some more idealized natural phenomena or mathematical equation,” he said.

    ===========

    “Clearly it is a student example of how to perform integrals to which someone has attached some meaningless numbers.”

    • Cheryl Rofer (History)

      It’s unfortunate that so much of the criticism of the graph relies on something like “I don’t like the way it looks.” That may be a starting point for a scientific statement, but it should never be the end point.

      Mark Gubrud points out upthread that the power curve is not symmetrical – something that many people, including Kelley and Greenwald, have missed.

      The quality of the graphics has nothing to do with the graph’s genuineness, unless someone can find a characteristic that clearly links its production to, say, Mossad. If the graph were produced some years ago, it might have been the best that computer graphics or textbook illustrations could do.

      The evidence that this graph is being used by the IAEA is all from Jahn’s anonymous sources at this point. So we don’t know whether or how the IAEA has used it, which makes it a bit early to be pointing fingers at them.

  11. hass (History)

    Someone somewhere drew a graph of something, and not even a technically correct one, and we have one of the longest posts on ACW demanding that Iran must “come clean” about weaponization studies that no one has verified even occurred. tsk tsk.

    Oh and nevermind whether they constituted any sort of NPT violatio since the IAEA even conceeds the allegations involve no diversion of nuclear material.

    • Cthippo (History)

      Only the first three paragraphs and the end deal with the graph, the rest is a general discussion of the state of play of the Iran – US – Rest of the interested world discussion.

  12. michael dittmar (History)

    Hi,

    actually i am wondering what is supposed to be interesting about
    such a trivial diagram physics students do very often.

    in case you want to be worried about knowledge about bombs
    think about what you find on the internet easily
    or plots like that one:
    http://lollitop.blogspot.fr/2009/12/how-to-make-atomic-bomb-merry-christmas_25.html
    or just google for “nuclear bomb”
    http://www.google.com/search?q=nuclear+bomb&hl=en&client=safari&tbo=u&rls=en&tbm=isch&source=univ&sa=X&ei=hgS7UKzIGMKm0QXrzIGwCA&ved=0CFQQsAQ&biw=1225&bih=627

    The only thing I find interesting is that it creates such discussions
    which are far far away from the real questions.

    Meaning that people blindly assume “strong evidence”
    without a single hard fact. After years of media manipulations

    “As I have written elsewhere, I believe there is very strong evidence that Iran had a covert nuclear weapons program through 2003.”

    People really worried should think about the violations of the
    NPT treaty by the 5 official nuclear weapon states
    since they signed it 40 years ago.

    They all signed to do everything they can to eliminate nuclear weapons

    they and all others by the way also signed that

    a) no nuclear related research should be shared with countries that
    have not signed the treaty (like Israel, India, Pakistan ..).
    Thus the non nuclear knowledge keepers for so called peaceful
    studies should not have relations with the list of countries above
    b) share of knowledge about with “peaceful” applications of nuclear energy.. including obviously enrichments ..
    thus all actions towards Iran these days are violations of the NPT treaty..

    so much for the legal side ..

    c) for those who want to be worried about nuclear weapons
    i suggest to watch the movie Dr. Strangelove

    or for people with little time only the final scene
    http://www.youtube.com/watch?v=s4VlruVG81w

    regards

    Michael

  13. BiBiJon (History)

    George Jahn’s mea culpa is worse than his original sin
    ======================================================

    “The U.N. agency reported on Nov. 8, 2011, that it had obtained diagrams it suspects shows Iran doing studies in nuclear yields, adding: “The application of such studies to anything other than a nuclear explosive is unclear to the agency.” And the senior diplomat on Tuesday confirmed that the graph seen by the AP was indeed one of those cited by the IAEA.”

    Read more: http://www.foxnews.com/world/2012/11/30/purported-iranian-nuke-graph-off-scientifically-but-diplomats-say-iaea-still/#ixzz2DtM6dLSQ

    Recall DG Amano’s claim that the “evidence” supplied by “two member states” were “overall credible.” Well, now George Jahn is reporting that among the ”overall credible” pile of evidence which the Agency received and showed to IAEA BoG, was this unsigned, undated, untitled, unmarked with security classification, amateurishly produced graph with humongous errors in physics and arithmetic.

    US should be very unhappy with Jahn’s sources, “officials from a country critical of Iran’s atomic program.” Having gone to great lengths not to share the documents with Iran or anybody else for appraisal and verification, now some desperado goes and spills the beans right in front of a scientific community who notice at a glance the graph is a graft in both senses of the word.

    The explanations from “officials from a country critical of Iran’s atomic program” is on (ridiculousness) par with the graph itself. I.e. well, if you correct the mistakes, then the graph will be correct, duh! Or that, for some inexplicable reason Iranian officialdom understand kilotons better than kilojoules which led to the “scientist” revising the axis labels incorrectly. Ya, right. While the attribution to an assassinated scientist who cannot answer the charge of sloppiness, may briefly shield the ‘sources’ from further questions, one hopes Dr. Shahriari’s other publications will soon show there’s no way that he would make such reputation-shattering obvious colossal mistake; certainly not as part of a presentation to Iranian policy makers.

    The phrase, “the graph seen by the AP was indeed one of those cited by the IAEA” opens another can of worms. Was this graph deemed the most persuasive, average, or the easiest to understand by lay people among all the other documents cited by IAEA? It sure spells trouble for the rest of the documents if this particular choice for a leak were justified according to any of those categories. It renders everything else as utterly unpersuasive, overall implausible, and/or incomprehensible by the warmongering punditocracy whose pontifications hitherto appears to have been baseless.

    Amano is going to have one hell of an interesting time at the next BoG meeting.

  14. Cthippo (History)

    It does raise a question which is critical for Iran and important for the Burma discussions, namely what constitutes a nuclear weapons program?

    I’ve heard a story from a number of people (it may even be true), that back in the day one of the common class projects in college level organic chemistry was synthesizing LSD because students were interested and it demonstrated some particular process. One could say that because this was being taught as part of the curriculum in a publicly funded university that it constituted a government program to produce drugs. On one level that would be technically true, but on another no reasonable person would call a gram or two made once a year anything more than a lab experiment.*

    Likewise, does doing the basic math on material consumed and yield mean you have a bomb program? What about doing design work? I would hazard a guess that the combined readership of ACW could probably design a functional, if over-conservative device. Does that mean that Jeffrey is next in line for regime changing? My understanding of the NPT, and feel free to correct me, is that until you actually start messing with weapons grade fissionable fuel, something Iran has never possessed, then you are not in violation of the treaty. By this reading, Iran has nothing to come clean about since they do not have and have never had any weapons-usable material, and only now might be able to make some if they decided to do so.

    In the Burma case, while they may be trying to make a handful of non-nuclear components, they are so far from anything even remotely resembling a functional device that the concern seems unwarranted except as a source of wonkpron. Keeping an eye on them is probably a good idea, but we’re a long long ways from anything that should reasonably trigger sanctions.

    * In an unrelated aside, there was an article in the paper today about the state of Washington searching for people who know how to set up a state-run marijuana production and distribution program since the voters (me included) approved exactly that 😀

    • hass (History)

      Legally, the line is clear: when the IAEA cannot verify that there has been no diversion of nuclear material for non-peaceful uses. That’s the standard in the NPT, which is why the model comprehensive safeguards agreements state that their “exclusive purpose” is to ensure non-diversion. In the case of Iran, the IAEA has verified non-diversion in every single report. Legally until then, countries are perfectly free to make whatever diagrams or studies that they damn well please.

      In the eyes of the US and Israel, the “capability” to make a nuke equated with a nuclear weapons program, or the “intent” to do so with a civilian nuclear program at some indefinite point in the future. Note that by some estimates, there are 40 nations (1 out of 5 countries) that are today capable of making nukes.

    • George William Herbert (History)

      Cthippo writes:
      Likewise, does doing the basic math on material consumed and yield mean you have a bomb program? What about doing design work? I would hazard a guess that the combined readership of ACW could probably design a functional, if over-conservative device. Does that mean that Jeffrey is next in line for regime changing

      Jeffrey, at risk? …

      The Nuclear Weapons Archive was started by Gary Au in Australia, passed to Carey Sublette when Gary got an Australian MOD security clearance job, moved from University of Melbourne to a Finland site to Italy before landing on US hosts. As far as I know it’s never had any negative IAEA scrutiny of that nature, in any of those locations.

      You can but Serber’s book on Amazon.

      An engineering test program by a private individual might garner more attention, but calculations seem to raise few hackles.

  15. Seb Tallents (History)

    It’s impossible to know without more context, as the graph looks utterly generic and could be just about anything, but I would not be surprised to find a graph like this in an undergraduate text book on lasers.

    On the other hand, 50kT is equivalent to about 1.25 eV. For reference the chemical bond between two hydrogen atoms is about 4.5eV, and it take 13.4 eV to fully ionise a hydrogen atom, so this looks odd for a laser pulse.

  16. George William Herbert (History)

    The graph is not perfectly symmetrical, but does appear to be “wrong”.

    Nuclear bomb cores during detonation are a race between energy buildup and disassembly. The buildup is exponential, not a gentle smooth function up. When inertia fails and core expansion begins it ‘s very rapid, a handful of doubling times (say … Generalizing horribly… 0.05 us ?). Roughly 85% of fission energy is prompt and available, mostly fission fragment kinetic, then gamma rays and neutrons. Neutrinos are prompt but unavailable. Fragment decays contain too little energy – perhaps 7%.

    The graph’s shape seems wrong on buildup and seriously wrong on decay. There should be a tail, but not like that.

    I don’t know what this is, but unlike R265 this does not strike me as “true”. Whether it’s a naive / wrong Iranian simulation, a simplification for nontechnical readers not intended to be technically correct, a bad forgery, or I have something grossly wrong I don’t know. I am betting on the first two, cannot rule out bad forgery, don’t think I am wrong.

    • George William Herbert (History)

      Looking back through all the notes – Yousaf and Ferenic in the Bulletin piece pointed to Reed’s book “The Physics of the Manhattan Project”.

      The fission / Time graph there – Fig 2.12 in http://www.springer.com/cda/content/document/cda_downloaddocument/9783642147081-c1.pdf?SGWID=0-0-45-1001458-p174027365. – is log base 10 in time. It takes tens of nanoseconds for fissions to drop a factor of 100, and decays about a factor of 10 every 8 ns after reaching the flat part of the decay ( on the log(10) graph ).

      R265 will have a somewhat different curve, but not that different…

    • Jeffrey (History)

      Yeah, this has been bothering me quite a bit.

    • Mark Gubrud (History)

      It looks to me like the time scales in Reed’s Fig. 2.12 and the AP figure agree, as far as you can tell by eyeballing it. Where they don’t agree is that the asymmetry is different; whereas the trailing tail in the AP figure is longer than the leading tail, the opposite is true for Fig. 2.12. However, Fig. 2.12 is for Little Boy, whereas the AP graph would presumably be for an implosion bomb. I don’t know if this would explain the difference. Also, Fig. 2.12 is for the fission rate, whereas the total energy yield must also include the decay of the most unstable fission products.

      I have not seen the convincing evidence that the AP graph is inaccurate. The fact that the power curve matches the energy curve if it is interpreted as Joules/shake should set that issue to rest.

      Even so, the fact remains that even if this comes straight from the desk of a high-level scientist in the Iranian nuclear program, it proves nothing about an ongoing Iranian weapons program. It looks like something that might have been prepared for a Wednesday afternoon seminar. Or a student exercise, as Yousaf suggested.

    • George William Herbert (History)

      Mark – the fig 2.12 is log(10) energy – linear time; the AP / Iran one is linear – linear. The shapes do not agree if you convert both to log-linear or linear-linear; for immediate energy release the drop from peak to 1% of peak should be a small fraction of 100 ns (0.1 us) – probably order of 20 ns. Area in under the power output graph from then to t=infinity should be about 7% of the prior total.

      It should look like a nearly immediate cliff down to less than 5% energy per unit time and longer exponential decay from that.

      I am using grossly rough numbers on purpose, but the tailoff is just wrong.

    • yousaf (History)

      George is correct. The only other thing I can think of to give any remaining benefit of the doubt to the AP report is that some significant simplifications were made as part of a class project by, e.g., a student.

      Another explanation is that it was merely illustrative: the existence of a spreadsheet does not mean a model was run.

      The spread-sheet could have been constructed, e.g., by taking datapoints off of the integrated energy curve in Fig 2.11, and also plotting the time-derivative (power).

      Another explanation is that it another physical phenomenon being plotted.

      Another is a forgery.

      And Mark is correct that this even in the very worst case “might have been prepared for a Wednesday afternoon seminar” and is not indicative of a weapons program.

    • Mark Gubrud (History)

      George – If you look at Fig. 2.11 the yield curve there is linear scale. As you can see, it has an appreciable trailing tail width, although here the trailing tail is markedly narrower than the leading tail, the reverse of the AP figure. But remember that this is a Little Boy model. Are the physics of disassembly substantially the same as for implosion? Intuition fails me here.

      Peak fission rate for the Little Boy model occurs at around 870 ns. The integrated yield at that point is only about 70% of the final total. The tail is about 30 ns wide and contains about 30% of the yield.

      I wouldn’t describe this as “a nearly immediate cliff” but I agree it is faster than the trailing tail shown in the AP graph.

      Apart from any differences between Little Boy and implosion, I’m also wondering if the AP graph could be showing a boosted device, and what difference that would make. Anybody here know?

    • yousaf (History)

      ” Are the physics of disassembly substantially the same as for implosion? ”

      Implosion compresses the material directly reducing the neutron mean free path. So time between neutron multiplication generations also goes down: i.e. the reaction occurs faster and goes further towards completion before disasembly occurs.

      “Anybody here know?”

      Yes, there are people who know the hairy details about this business (e.g. Bob Kelley) and others, but understandably one may not want to get into specifics about this publicly.

      Bob Kelley has 35 years of weapons experience at Livermore and Los Alamos. This is what he says —

      http://truth-out.org/news/item/13087-fake-ap-graph-exposes-israeli-fraud-and-iaea-credulity

      “We can only hope that the claim that the IAEA has relied on this crude hoax is false. Otherwise their credibility has been shattered.”

      =============

      “It’s clear the graph has nothing to do with a nuclear bomb.”

      ===========

      “The pretty, symmetrical bell shaped curve at the bottom is not typical of a nuclear explosion but of some more idealized natural phenomena or mathematical equation,” he said.

      ===========

      “Clearly it is a student example of how to perform integrals to which someone has attached some meaningless numbers.”

    • yousaf (History)

      PS: Mark, if you compare the time needed for the integrated yield to go 10%-90% for the AP “Iran” graph, it is substantially larger than the time needed for the same excursion in Fig 2.11 (Little Boy design)

      Is there a reason to think the Iran graph corresponds to a implosion device?

      If so, it’s 10-90% excursion should be faster.

    • George William Herbert (History)

      It’s not showing a boosted fission weapon; those are even more “peaky” – You get the normal exponential ramp-up then at around 250 – 350 tons yield, the fusion reaction goes “bang” far faster than the fission did, and however much boost material you had you get that many neutrons out at once in a wave that gives you something like 10-20 more kilotons yield in something on the order of another neutron doubling interval. It’s gentle exponential up and then WHAM in a completely nonlinear manner.

      With Plutonium boosted fission weapons, they’re not that difficult to “tune” so that the immediate aftermath of the boost is near reaction completion due to fisson product poisoning at the peak 35-50% efficiency you can see in reality. WIth Uranium and the naturally larger critical mass (probably 4-5 times bigger pit mass-wise) either you have that much more boost gas and yield target, or for the same boost gas you have another doubling or two after that before fission product poisoning takes over.

      With those reactions, it’s extinction due to that effect and not necessarily core expansion, and that’s much more abrupt as the molar fraction of products rises. 1 mole of fissile material generally gives you 2.something moles of product over the reaction time, which at least thermalize if not absorb neutrons, and either thermalization or absorbtion will poison the fast prompt reaction quite quickly.

      In terms of implosion disassembly vs Little Boy… Tampers slow down the disassembly. Reflectors not so much. Implosion weapons usually optimize to maximize compression and reflection to increase alpha / reaction rate. The rate of expansion is roughly proportional to the energy generated divided by total pit mass, including tamper and reflector. Alpha goes as density squared. Implosion system mass (and cube root of dimensions) goes as total pit mass times imploded density to a low power, less than 2.

      The disassembly dynamics are complicated, but keep in mind that by the time we’re talking about this all the materials are for all intents and purposes dense gases (energy levels far exceed any material strength) and for a gun weapon, it’s not total tamper mass that matters, but how fast the interface between the denser hotter core and tamper pushes out into the bulk surround. The LIttle Boy WC tamper matters; the heavier steel around it really ultimately doesn’t much. And later guns, such as W33, demonstrated that the tamper? Not so much. W33 is “reputedly” Be reflected and for all intents and purposes untamped; disassembly wise they went the max-alpha-they-can-get, untamped route.

    • Mark Gubrud (History)

      George — If what you say about the fusion reaction going “BANG” relative to the fission, and its termination being due to poisoning at 35-50% efficiency, is correct — and I don’t know if it is, but it sounds reasonable and you obviously have done some prior homework on this, then I agree this can’t be a model of a boosted weapon.

      Yousaf — The rise in the yield curve for the Little Boy model is about 3 times faster than the AP figure, and the asymmetry of the power curve in the AP figure is the reverse of what it is for the Little Boy model. I don’t have an explanation of why these should be the case.

      Bottom line — These observations do cast doubt on whether the AP curve is plausibly a model of a fission explosion produced by a competent scientist. More careful analysis based on open source science might be able to show that conclusively.

      However, the fact that the power curve is mislabeled does not prove the case. The fact that the correct symbol for kiloton is kt not kT does not prove the case. Both of these are easily discounted as trivial oversights.

      If it can be shown conclusively that the time scale is wrong or that the shape of the curves is wrong, and can’t be right under any reasonable set of assumptions about the design of a bomb, that would prove the case.

    • yousaf (History)

      Mark: “The rise in the yield curve for the Little Boy model is about 3 times faster than the AP figure”

      Yes, and under the hypothesis that the AP figure was an implosion device, the AP figure would have to be faster due to the shorter n mfp.

      It is not. According to George, the AP figure is also not a boosted device. So then you are left w/ a Little Boy type, and it does not fit that model either.

      Therefore it fits no known model of nuclear weapons.

    • George William Herbert (History)

      I think too much is being made of kT vs kt – my personal usage varies wildly depending on what time of day I started writing something and what device I’m using. I see sources with both, though there is a preferred right way. What you get at 1am on my iPhone with some developing news is very different from scholarly work I have time to edit significantly. I just don’t care enough to be consistently “kt”…

      That is the only point I’d make here; the diagram’s other factors are unrelated to preferred abbreviation.

    • yousaf (History)

      I agree with George on that — that’s why for our Bulletin piece we gave the AP the benefit of the doubt that there is some reason to think it may be nuclear related, and treated kT like kt. (Also, kT is not really a unit, unless as someone suggested in a joke, its kilo Tesla!)

    • anon2 (History)

      George et al,

      What if this graph is just a talking point for a powerpoint presentation to politicians or funders by someone working on a bomb.

      The graph designer would not need to be technically accurate. They would only need to convey an idea of what a bomb explosion might be to give them sufficient credibility to achieve their goal — continued funding of their work.

      What would the fastest way be to create a graph. Have Excel draw a normal distribution in a chart table and then differentiate it by taking the differences in values between cells. Then, put up the chart with whatever scale you like for the energy — kilotons sounds right as politicians have heard of kilotons, and add this differentiated chart with whatever scale it is that you would like as the politicians have no idea what it is or how you created it.

      Of course the same goes for a purposeful leak to an AP reporter. But my point it is the idea that someone wanted to convey this idea to someone else is the main idea. What is more important is to find out who created the chart, when did they make the presentation, to whom was the target audience, and for what purpose. If the purpose was to get the clandestine funding committee of the Supreme leader to dump cash in a Parchin experiment, we have one case. If we have Israel creating a chart for propaganda purposes we have another.

      I would bet that this chart source, where-ever it is has fingerprints all over it in the form of metadata (hidden digital fingerprints) that have been analyzed. If it has been proven that the chart was part of a presentation for funding by an assassinated nuclear scientist and that it was generated in 2009, we have evidence that at least this scientist was asking for funding in 2009. Right now, the chart is so far out of context that we have no idea what its purpose really is.

    • George William Herbert (History)

      Anon2 –

      That was the second speculative explanation in my top-of-this-thread post; a nontechnical simplification not intended to be precise or accurate.

      That’s possible. It’s somewhat credible, I see viewgraph / PowerPoint engineering a lot. But it has a fair degree of detail, and the detail is outright wrong in key areas. Usually you get simplifications or omissions, not material inaccuracies.

      So, I don’t know. It could be that, but in a weird way. That’s not “no”. It’s just not what I’d expect.

      Any number of precise explanations seem true, but for it to be a smoking gun about a weapons program I would want the little details to be right and match what we think they are evidently doing with R265, or tip another, newer program. None of that seems true.

      Lacking access to the rest of the document at the moment, the credibility is hard to assess. There are likely more details that would suggest what explanation is right. But we don’t have that to look at.

    • yousaf (History)

      Agree with George,

      Also let’s keep in mind that if anyone is going to get overly concerned about this graph (if true, if nuclear, if Iranian) then they should probably investigate this PhD thesis written at a *military* institute in Brazil:

      “Numerical Simulation of Thermonuclear Detonations in Fission-Fusion Hybrids Imploded by Radiation”:

      http://www.jb.com.br/pais/noticias/2009/09/05/a-explosiva-descoberta-brasileira/

      If someone in Iran was making a case for funding using a crap graph then that is not yet a national program: it is a scientist making a theoretical case. How does IRI have to answer for that?

      Is it in violation of the CSA?

    • Mark Gubrud (History)

      anon2- Problems with your scenario:

      Both the probability density of the normal distribution and its integral (erf) are available in Excel, so no need for any numerical calculus.

      However, the power curve is NOT a normal distribution, and the energy curve shown IS its correct integral, if the power unit is taken as Joules/shake (see comment by Rob Goldston).

      The factor 1 kt = 4.18E12 J was not inserted accidentally.

      If you assume the faker started with the yield curve scaled to give an impressive-sounding 50 kt, why did he convert the power to J/shake and then mislabel the axis as “kT/sec”?

    • Mark Gubrud (History)

      George and Yousaf–

      I read the peak power for the Little Boy model as ~2.5 kt/shake, and the peak power for the AP graph as ~4 kt/shake. So AP actually ends up going “faster” before it is done.

      Overall, it takes AP a few times longer to build up to peak power, and the whole power pulse (defined by disassembly rate) is a few times longer. The first would be explained by a lower initial alpha (and/or a more dilute mass), the second by more total effective (core + tamper) mass.

      I am wondering if the difference could be explained by either a heavier tamper, or, since the returns from a tamper are diminishing, could we be looking at a theoretical study of a bomb using lower enriched uranium, or some other alloy with an inert heavy material?

      Perhaps it would be a gun design, so no compression.

    • yousaf (History)

      The tail is the main problem with the AP graph.

      And if you cross-ref w/ IAEA report to what it could be — if it was not wrong — it is supposedly (but not really) an implosion device:

      http://www.iaea.org/Publications/Documents/Board/2011/gov2011-65.pdf

      para 52:

      “52. Information provided to the Agency by two Member States relating to modelling studies alleged to have been conducted in 2008 and 2009 by Iran is of particular concern to the Agency. According to that information, the studies involved the modelling of spherical geometries, consisting of components of the core of an HEU nuclear device subjected to shock compression, for their neutronic behaviour at high density, and a determination of the subsequent nuclear explosive yield. The information also identifies models said to have been used in those studies and the results of these calculations, which the Agency has seen. The application of such studies to anything other than a nuclear explosive is unclear to the Agency. It is therefore essential that Iran engage with the Agency and provide an explanation. ”

      =====

      shape of curves wrong. I welcome input from George.

    • Mark Gubrud (History)

      To clarify: “AP” refers here to the hypothetical device whose power and energy curves would correspond to those shown in the Associated Press graph.

      Regardless of whether AP is an implosion or gun design its neutron population growth and decay is controlled by alpha/tau where alpha increases with mass size relative to critical as well as with density, and tau decreases with density.

      It is true that if all else is equal alpha/tau will be greater in implosion than for the same mass of the same material not imploded, but alpha can be arbitrarily small depending on the mass size relative to critical and tau will be larger if the mass is less dense, e.g. if it is diluted with inert mass.

      Thus it is not correct that implosion necessarily implies a faster initial buildup than with a gun design.

      George was saying even for gun designs max alpha is preferred but maybe there is another regime to consider.

      For a weapon using lower-enriched U the critical mass will be larger, and tau will be somewhat larger. Initial alpha can be tuned to match the rate of disassembly, which will be slower due to the larger mass.

      I’m guessing that AP might be a lower-enriched U bomb, whether it is an implosion or gun design.

      Tell me why that’s obviously wrong.

    • George William Herbert (History)

      I agree that it’s mostly shape of curve that’s wrong, no matter what other assumptions were made.

      Mark, a larger Uranium implosion, say with DU tamper, or a gun type weapon would not change the tail off shape.

      http://www.nuclearweaponarchive.org/Nwfaq/Nfaq4-1.html#Nfaq4.1.5.1

      Expansion happens due to supersonic expansion at the pit edge (for untamped, the fissile core edge; for tamped, the effective tamper edge) into the effective surrounding vacuum. Also the bulk of the material expands internally, due to the pressure. So you have an internally expanding sphere whose edge is boiling off rapidly. The internal expansion is the primary focus of the Serber equation (and earlier Bethe – Feynman equation).

      The efficiency equation with mass density and tamper effects included doesn’t change the tail-off shape significantly. Once you reach a sufficient energy level per unit core mass, the expansion is running on timescales like the timescale of doubling the fission reaction. At best you can hope for most of the tamper to be involved in retarding expansion, though in practice thick tampers aren’t terribly useful, the shockwave will be faster than the temperature transfer and though density may be the same on both sides, pressure on the inside will be immense while very low on the outside due only to inertial pressure.

      If the full thickness of the tamper is taken into effect, that’s say 16 kg HEU plus 16 kg DU tamper vs a bare 4 kg untamped Pu core; about a mass factor of 8 difference, and about a pit thickness of 2 difference for equivalent compressions. The energy level per unit pit mass at which expansion velocity becomes significant is roughly the same; so a say 10 kt untamped Pu weapon would, with equal compression and alpha, give you about 80 kt of energy in the HEU + DU pit, and about twice as long from there to disassemble. Roughly. Very roughly. Using the equations or modeling and specific configurations is necessary at some point.

      A gun bomb may be much bigger – 64 kg or more of material, 100s of kg of tamper – but alpha will be far lower due to lack of compression, so expansion will be significant earlier.

      A gun bomb with boosting is nearly ideal; it would have the low-alpha problem, but at 250-350 kt the fusion burn lights off and you get that rapid massive neutrons spike and their secondary and tertiary fissions…

    • Mark Gubrud (History)

      George, I appreciate the discussion, and the pointer to Carey’s site, but I’m not seeing any clear answer to my question (what’s obviously wrong with the lower-enriched U scenario).

      If critical mass is a few times larger, disassembly would be a few times slower. That could account for the total pulse time of AP being a few (2-3) times longer than for the Little Boy model.

      Initial alpha being a bit smaller and tau a bit bigger easily accounts for slower buildup.

      You say the shape should not change but it’s not clear why not. Even the Little Boy model has an appreciable tail. It’s just a few times longer in AP.

      The shape of the (scaled) curve will depend on the time history of alpha/tau. There are a number of variables here which could affect this. So I am not seeing the argument that no reasonable configuration comes up with a shape like the AP graph.

    • Mark Gubrud (History)

      So, on the subject of errors in graphs, Reed’s Fig. 2.11 can’t be correct, because it shows fission stopping while alpha (Reed’s alpha/tau is Sublette’s alpha) is still close to its initial value. I conclude that alpha is plotted on a different time scale than the yield, and Reed forgot to mention this.

      Sublette says Little Boy was only 80% enriched while I’ve read other sources who claim it was 99%.

      And basically all George’s comments about the supersonic blowoff and the internal expansion are taken from Sublette’s discussion which he admits are just two limiting simplified models which don’t give very accurate predictions for the yield, and which don’t really tell me anything about the shape of the power curve under any given set of assumptions.

      Bottom line is I’m skeptical about all of this now. I don’t think anybody here knows if the AP graph might or might not represent a valid model of a possible device.

      We do know it is not a simple illustration of the relationship between the gaussian distribution and the error function.

      Nobody has shown a graph in the open literature from which it could possibly have been directly copied.

      We do know it does not represent a laser pulse experiment.

      We do know it involves a tnt ton equivalent scaling factor.

      Maybe it’s the work of an Iranian equivalent of a Reed or Sublette.

      Maybe it’s the work of a forger.

      Maybe it’s a model of a low-enriched U weapon as I suggested.

      I think we are in maximum entropy territory here.

    • yousaf (History)

      Fig 2.1 and 2.3 in this declass Glasstone ref are helpful for understanding power output (power will more or less scale w/ alpha) in realistic devices:

      http://www.alternatewars.com/WW3/Glasstone_Intro/Introduction_JUN-1972-REV.pdf

      For devices built from lower enriched U (below 80% or so) one could get some yield but to substantially alter the parameters such as to get different timescales you’d prob. have to go below 50% and then we are looking at fizzles and not smth that could get to 50kt with.

      I’d welcome any solid references on any feasible devices built from lower (20-50)% U.

      e.g. Super Kukla was a five ton 20% enriched prompt critical assembly but it was basically an elaborate dirty fizzle.

    • yousaf (History)

      PS: The pdf on here seems dead — if anyone can get it that would be great:

      http://www.osti.gov/energycitations/product.biblio.jsp?osti_id=15011793

      I hope we can avoid FOIA’ing it….

    • Rob Goldston (History)

      I plotted the power curve digitized (by hand) from the AP plot on a semilog scale and it starts out pretty much exponential with alpha roughly 0.4 e-folds/shake. I also measured that roughly 53% of the yield comes after second critical, which is by definition the peak of the power curve.

    • Mark Gubrud (History)

      Thanks, Rob.

      It seems to me that those numbers are at least roughly consistent with a large critical mass lower-enriched (probably below 50%, possibly as low as 20%) uranium device which might be implosion-assembled but with low compression, giving the low initial alpha and slow disassembly due to low yield/mass. Without a good quantitative model (or experience in nuclear weapons design) I can’t say if the numbers really fit a plausible model but that seems to be where they are pointing if they make any sense at all.

      This would not be a very practical weapon although it might serve as a deterrent. More likely, if this is real and comes from Iran, I would guess it was just a thought experiment to consider the possibility of doing something with lower-enriched material, or maybe just a pedagogical illustration of the physics.

  17. yousaf (History)

    Cheryl,

    Thank you for your input above.

    One must note that the graph itself is from Jahn’s anonymous sources too. Either we throw the whole story out or we consider what Jahn says about the graph AND the rest of what he says too.

    I am happy to throw this whole story out.

    Here is what Jahn says in the original AP story:

    “A senior diplomat who is considered neutral on the issue confirmed that the graph obtained by the AP was indeed one of those cited by the IAEA in that report. He spoke only on condition of anonymity because he was not authorized to discuss the issue.”

    I had been giving the quoted diplomat associated with the IAEA the benefit of the doubt that there was other evidence associating this graph with nuclear matters (besides the length of the pulse) and the possibly wrong “kT” instead of “kt”. The diagram has a caption in Farsi which merely says: “Changes in output and in energy released as a function of time through power pulse.”

    However, if there is no other evidence linking this graph to anything nuclear, it may well not be — there are many prompt pulses in various physics phenomena, as you noted earlier also. This is also why we said in our Bulletin piece “*presumably* from a fission device…”

    In any case, the error still remains: the time-derivative of the energy does not give the power, as plotted.

    Rob Goldstone and David Albright are certainly correct that if the error is corrected (~10^5 factor going from kt/sec to Joules/shake!) then there is no longer any error, but this is almost tautological.

    The graph as distributed — of whatever it is about — is still wrong mathematically.

    Regarding your comment “If the graph were produced some years ago, it might have been the best that computer graphics or textbook illustrations could do.”

    That is not what Jahn says — he says:

    “The models were allegedly created in 2008 and 2009”

    And if this really was a scientist preparing something to senior government officials [Jahn says, “…intentionally simplified the diagram to make it comprehensible to Iranian government officials to whom they were presenting it.” ] then they would likely have used higher quality plotting software ca. 2008. e.g. They may have used Mathmatica or Powerpoint, instead of what looks to me like an Atari or Commodore64 plot.

    Jahn says in the pseudo-retraction:

    =====
    “Shahriari — the scientist suspected by the IAEA to have made and modified the diagram and provided the spread-sheet with the right information — was assassinated two years ago. Iran said it believes Israeli agents were responsible for the killing.

    “Nobody would have understood the original, so he modified it into an artificial unit to make his case,” the first diplomat said.

    Albright said Friday that the explanation “made sense.”

    ====

    I think whole story makes no sense, and tend to agree more with Bob Kelley on this.

    Again, it is important to keep this in context: Even if the graph were technically correct (it is not), and if the graph was made by Iranian scientists, and the graph is related to nuclear issues, it still is not obvious to me that computations violate the IAEA-Iran safeguards agreement.

    Iranian scientists and students are allowed to compute what they like. It is a sovereign nation. Yes, it is bad form and perhaps against the spirit of the NPT (if any of this story is true), but not a legal breach. See Prof Joyner’s comment above on that.

    Lastly, let’s keep in mind that if anyone is going to get overly concerned about this graph (if true, if nuclear) then they should probably investigate this PhD thesis written at a *military* institute in Brazil:

    “Numerical Simulation of Thermonuclear Detonations in Fission-Fusion Hybrids Imploded by Radiation”:

    http://www.jb.com.br/pais/noticias/2009/09/05/a-explosiva-descoberta-brasileira/

    • Cheryl Rofer (History)

      Yousaf –

      I’ve been thinking about working up a logic diagram about what one can believe under various assumptions about truth-telling by various actors in Jahn’s all-star anonymous cast.

      But I decided instead to think about Estonia, one of my favorite places in the entire world. I will post some interesting material on that in a few more days, with lots of pictures.

      I maintain that we know nothing from Jahn’s story about what the IAEA has in terms of evidence on Iran’s calculations, or about what the IAEA inspectors may believe about that evidence beyond that they think there is enough there to continue their investigations into it.

      Thank you for digging out the 2008 and 2009 dates. I agree that computer graphics were better by then. But if the graph is in fact from before 2003, then it could be part of the weapons program that intelligence services seem to agree was ended then.

      I would say that the quotes from Bob Kelley by Gareth Porter sound like instant reactions that I wouldn’t hold Bob to for the longer run. I know the power curve looked symmetrical to me at first; the integral curve seems to provide enough distraction to provide a mild optical illusion in this direction.

      Throwing the whole of Jahn’s story out is tempting to me, too, but sometimes there is something to be learned from misinformation.

    • Mark Gubrud (History)

      The graph definitely looks to me like it was made using Excel. That’s M$’s crappy dashed line there. I’ve struggled with it many times.

  18. Cthippo (History)

    Actually, this graph is clear and convincing proof of NK / Iran collaboration.

    In proper SI units, a kT is the abbreviation for kilo-tesla, a measurement of magnetic flux, and no living person could have a magnetic personality of fifty Kiloteslas! No one except supreme leader Kim Jong Un that is!

  19. yousaf (History)

    I believe this is where the graph-of-doom appears in IAEA doc’s:

    http://www.iaea.org/Publications/Documents/Board/2011/gov2011-65.pdf

    para 52:

    “52. Information provided to the Agency by two Member States relating to modelling studies alleged to have been conducted in 2008 and 2009 by Iran is of particular concern to the Agency. According to that information, the studies involved the modelling of spherical geometries, consisting of components of the core of an HEU nuclear device subjected to shock compression, for their neutronic behaviour at high density, and a determination of the subsequent nuclear explosive yield. The information also identifies models said to have been used in those studies and the results of these calculations, which the Agency has seen. The application of such studies to anything other than a nuclear explosive is unclear to the Agency. It is therefore essential that Iran engage with the Agency and provide an explanation. ”

    ============================

    If, in response to this, IRI said that they were being presented with fabrications, then it is possible that the IRI is right.

    • Cheryl Rofer (History)

      Seems to me that this surmise goes too far. The graph could fall under this paragraph and could be what Jahn intends when he says the November 2011 report says that the IAEA says “it had obtained diagrams”. Note the plural. But this paragraph also seems to include more than the graph. But it’s a very tenuous connection.

      Paragraphs 53 and 54 describe a great deal more.

    • yousaf (History)

      Para 53 refers to stuff in 1997.

      Both 53 and 54 refer to numerical work — not necessarily weapons related — and not nuclear material. Much more serious numerical work on weapons was done in Brazil.

      Keep in mind that the IAEA itself recognizes the limits of its legal ability to pursue all such evidence, whether true or fabricated. e.g.

      http://www.iaea.org/Publications/Documents/Board/2006/gov2006-15.pdf

      “…absent some nexus to nuclear material the Agency’s legal authority to pursue the verification of possible nuclear weapons related activity is limited…”

  20. Andy (History)

    Great post and a lot of great comments!

    With respect to this graph, it seems to me that many are jumping to conclusions one way or another. In my opinion we don’t have enough information to make any judgements about this graph. We don’t even know what, exactly, this graph is, or what it represents. It’s presented in a manner divorced of context as well as the data and calculations used to create it. Whether or not it’s an important indicator of a current or historical Iranian weapons program depends completely on the context in which this document was produced, assuming it’s even related to nuclear weapons. Speculation is fun, but I think we all need to be careful before passing definitive judgments.

    On the topic of Iran’s need to “come clean” regarding past covert nuclear activities, I generally agree with Jeffrey’s analysis plus the added Iranian domestic political context provided by Mohammad. Personally, I don’t think such an agreement is likely with the current Supreme Leader in power without some significant face-saving measures for Iran, even if the West was interested in such a deal, which I don’t think is the case. Plus, as Jeffrey notes, the nuclear issue is still tied up with all the other issues that divide Iran and much of the Western world.

    As a practical matter, however, I don’t think “coming clean” is strictly necessary. While I agree with Jeffrey and the US IC that Iran had a covert weapons program, I also think that program was a strategic necessity in the face of the threat posed by Iraq under Saddam Hussein. I don’t think it’s mere coincidence that Iran’s covert program ended in 2003. Iran’s strategic position fundamentally changed when Iraq’s regime was overthrown and the strategic rationale for an Iranian nuclear weapon pretty much evaporated. Consequently, I think it’s more important to focus on the future by convincing Iran to adopt the additional protocol. With the AP in place, Iran’s ability to successfully and secretly restart a weapons program would be greatly diminished. With the AP in place, I don’t think it’s a big issue to let sleeping dogs lie. Of course, Iran would require a quid pro quo and the price would likely be high, but probably worth it in the long term. Unfortunately, this result doesn’t look very likely either at present because, again, there is much more than the nuclear issue at play.

    • hass (History)

      Iran doesn’t need any “convincing” to adopt the AP and has already — repeatedly — offered to ratify it. In fact Iran has offered several compromise offers that far exceed the requirements of the AP. The problem is the US which refuses to meet Iran’s condition that its right to enrichment should also be recognized. Further, note that the IAEA doesn’t agree with your assumption that IRan had a nuclear weapons program prior to 2003: “With respect to a recent media report, the IAEA reiterates that it has no concrete proof that there is or has been a nuclear weapon programme in Iran.” http://www.iaea.org/newscenter/mediaadvisory/2009/ma200919.html

    • Dan Joyner (History)

      I agree with Andy, here, on the prudence of looking forward and not backward with regard to whatever nuclear weapons related work Iran may have done in the past (as well as with Hass’s corrections). Getting too caught up in what may have happened in the past in this regard, will only comprise a distraction and a hinderance to efforts to find a negotiated, diplomatic way through the current crisis. That’s why I think that Albright and ISIS are not helping with their continuous speculation about what Iran “may” have been doing in the past at sites like Parchin, or what the PHRC “might” have been up to 15-20 years ago, as in their most recent such report:
      http://isis-online.org/uploads/isis-reports/documents/Applied_Science_Co_nov_14_2012-final1.pdf

  21. Thomas (History)

    I haven’t had time to read all of the comments, but the diagram seems looks exactly like a laser pulse experiment. Then kT is simply the thermodynamic energy in whatever units the experimenters are working in.

    • Mark Gubrud (History)

      If you’d taken the time to read the comments, you would have seen several reasons why this is not a reasonable interpretation. These are the strongest:

      * Interpreting kT as the thermodynamic energy scale does not resolve the discrepancy between the power curve and the energy curve, whereas interpreting the power curve as J/(10ns) does. This can’t be an accident since it depends on the unit kiloton = 4.18E12 J.

      * The starting energy is clearly closer to zero than to “kT”.

    • Thomas (History)

      Hi Mark,

      I think you misunderstood me; I meant kT as stand-in (for whatever reason) for units of energy. That is, substitue kT = J and then having a starting energy of zero would be what one expects.

      On relating the energy and power axes, there’s clearly some unit shenanignans going, as you pointed out. Instead of resolving it by saying kT = kiloton, with the left axis labeled in kT/s but numbered in J/10ns (while the x-axis remains in micro-seconds), I would suggest that the left axis is numbered in ergs, while the right axis is numbered in joules/100, which would work with the time being in micro-seconds.

      The ergs -> joules conversion seems a little less arbitrary to me than your suggestion about the different labeling / numbering / time units.

      10^13 ergs would also be the peak power of several of the exisitng or planned laser pulses in IC fusion, though those are running for only 10ns or so.

    • Mark Gubrud (History)

      Yes, Thomas, kT in standard thermodynamics nomenclature means Boltzmann’s constant times absolute temperature, and is an appropriate energy scale for systems in or near thermodynamic equilibrium (for which a temperature is defined) because it is the average energy that any two degree of freedom (e.g. kinetic + potential energy of a 1D harmonic oscillator) subsystem will have at that temperature.

      If the graph shows the energy of some subsystem of a thermodynamic system, such as an atom or molecule being pumped by a laser, then the initial energy should be 0.5 kT per degree of freedom; an atom or molecule will have at least 3 and potentially many dof.

      However, the graph clearly shows that the initial energy is zero (or anyway, not 1.5 or ~1 kT).

      The fact that the power axis is given in J/shake is confirmed by the fact that its integral then corresponds to the energy curve using the appropriate conversion factor between Joules and kilotons.

      10^13 ergs is an energy, not a power, and even ergs/sec does not measure the power of a laser field which would have dimensions of power/area.

    • seb tallents (History)

      Its not a laser pulse… 50kT is a very small amount of energy, more like a single photon.

    • seb tallents (History)

      To make the point clear, if kT means Kb T then the energies involved are so small we are talking about processes well into the realm of the discrete and such a smooth graph is gibberish.

  22. seb tallents (History)

    Scrub that, i’m talking nonsense based on a huge implicit assumption. Do not drink and derive,

  23. David Albright (History)

    I have been following this thread with interest. We at ISIS are trying to understand the graphs better and their relationship to the information gathered by the IAEA. With that in mind, I would like to ask for some clarification of a comment by Yousaf Butt. Yousaf, you wrote that there are problems in the shapes of the graphs. I assume you are not talking about elementary calculus issues, since the energy graph appears that it can be differentiated to yield the shape of power graph. With a relabeling of the vertical axis to the units of joules/shake, the units are consistent and the area under the curve gives the predicted total explosive yield. (I only approximated an area calculation but I got close to the 50 kilotons.) In addition, the shapes of the graphs of the yield rate and yield are very similar to what one would expect to obtain from certain, well-known coupled hydrodynamic-neutronics computer codes. Would you kindly explain what are the problems in the shapes of the graphs you have identified? Thanks in advance.

    • yousaf (History)

      I am fascinated by your logic.

      “In addition, the shapes of the graphs of the yield rate and yield are very similar to what one would expect to obtain from certain, well-known coupled hydrodynamic-neutronics computer codes. ”

      Really which well-known ones?

      And if they are well-known what exactly is the IAEA’s issue with Iran using these well-known codes?

      I hope you are not referring to the open-source 2009 toy-model published in a Western journal from a German prof.

      ====

      The answers to your questions can be found in comments above by George and myself and in figures 2.12 in Reed and the Glasstone ref I linked to above. And you can talk to actual engineers who have worked with weapons such as Bob Kelley, and others.

      Have a good day.

  24. David Albright (History)

    The discussion above does not answer my question. And Mark makes very good points in his comments. The comments that the graphs published by AP cannot represent the calculation of a nuclear explosive seem particularly unconvincing. In terms of the code I mentioned it is a Los Alamos Scientific Laboratory coupled hydrodynamic-neutronic computer code called HENRE, and its use is illustrated in an article by Sandmeier, Dupree, and Hansen, titled “Electromagnetic Pulse and Time Dependent Escape of Neutrons and Gamma Rays from a Nuclear Explosion,” published in Nuclear Science and Engineering, in 1972. It models a 24 kiloton plutonium implosion device and includes graphs of the calculated yield rate and yield, which are similar to those presented by the AP. (The yield rate graph has to start at zero, which is why Reed’s figure 2.12 seems irrelevant; his figure 2.11 by the way has a yield curve in the expected shape.) One of ISIS’s consultants, Mark Gorwitz, knew about this article. The article was also listed by the Iraqis in their September 1996 declaration about their theoretical nuclear weapons work that lasted until January 1991. In any case, we at ISIS are trying to better understand the graphs, their origin and significance, and how the IAEA treats this type of evidence. I would welcome comments and discussion relevant to that goal. We will address Yousaf’s legal arguments made above in a different forum.

    • yousaf (History)

      Yes, it is possible that an Iranian scientist or student — or whomever generated the AP graph — ran open-source models from 1972 — it is also possible they ran the model from 2009 mentioned above. This only shows that such research can be done for academic or illustrative purposes. Congratulations on digging up an even older source!

      “The yield rate graph has to start at zero, which is why Reed’s figure 2.12 seems irrelevant; his figure 2.11 by the way has a yield curve in the expected shape.”

      Really? Why would you expect the yield to be exactly zero during the first generations of neutron multiplication?

      Yes, I urge you to continue “trying to better understand the graphs.”

      2.12 is a log-lin graph and if you took the time to plot it on a lin-lin scale you will begin to see what a (somewhat) realistic explosion power curve looks like.

      BTW, Is your argument that someone running an open source model from 1972, or copying a graph from a journal is illicit activity? I trust your legal arguments will hold up better than ISIS’s previous stabs at law:

      http://armscontrollaw.com/2012/11/27/a-whole-isis-report-devoted-to-little-ol-me/

  25. David Albright (History)

    At least Yousaf has stated that the AP graph can represent a sophisticated calculation of the yield of a nuclear explosion and one done by an Iranian scientist. I think it is too early to conclude who did the calculations or for what purpose, and to the calculations’ relationship to a wide body of evidence collected by the IAEA about the possible military dimensions of Iran’s nuclear program. The graphs are only one page of a set of information in the hands of the IAEA. Before forming a conclusion, it is better to learn the actual content of this information, its timing, which Iranians are mentioned, and the scope of the work. To that end, it is essential to urge the Iranians to cooperate with the IAEA, recognize that the information in the IAEA’s hands is sufficient to raise such questions, and support the IAEA in asking Iran for clarifications. Reed’s excellent model CriticalityNumerical.xlw is in his words, ”an order-of-magnitude pedagogical model” where he has had to make simplifying assumptions. So, I would not call it more realistic than the HENRE code in modeling an implosion system. For example, Sandmeier, Dupree, and Hansen show a more rapidly decreasing alpha after the onset of the explosion than shown in Reed’s figure 2.11. The implication is that Reed’s estimated alpha falls too slowly for an implosion design, which appears to me to mean the yield rate or his fission rate curve also decreases too slowly. Perhaps, someone could ask Reed for his opinion. With regard to the zero (or better near zero) for the yield rate (in joules per shake or kilotons per shake) in an implosion system, the reason is that the yield rate curve reflects the onset of the explosion. In any case, wanting to learn more is hardly to be criticized. For those interested in our legal reasoning why Yousaf is wrong about the IAEA’s mandate, I would refer you to “Understanding the IAEA’s Mandate in Iran: Avoiding Misinterpretations” by Olli Heinonen, Orde Kittrie, and me and endorsed by John Carlson, Trevor Findley, Pierre Goldschmidt, and Andreas Persbo (available at http://www.isis-online.org/uploads/isis-reports/documents/misinterpreting_the_IAEA_27nov2012.pdf). By the way, it is not a work of ISIS but a collective effort that incorporates many decades of experience and expertise in safeguards law and practice.

  26. David Albright (History)

    At least Yousaf has stated that the AP graph can represent a sophisticated calculation of the yield of a nuclear explosion and one done by an Iranian scientist. I think it is too early to conclude who did the calculations or for what purpose, and to the calculations’ relationship to a wide body of evidence collected by the IAEA about the possible military dimensions of Iran’s nuclear program. The graphs are only one page of a set of information in the hands of the IAEA. Before forming a conclusion, it is better to learn the actual content of this information, its timing, which Iranians are mentioned, and the scope of the work. To that end, it is essential to urge the Iranians to cooperate with the IAEA, recognize that the information in the IAEA’s hands is sufficient to raise such questions, and support the IAEA in asking Iran for clarifications. Reed’s excellent model CriticalityNumerical.xlw is in his words, ”an order-of-magnitude pedagogical model” where he has had to make simplifying assumptions. So, I would not call it more realistic than the HENRE code in modeling an implosion system. For example, Sandmeier, Dupree, and Hansen show a more rapidly decreasing alpha after the onset of the explosion than shown in Reed’s figure 2.11. The implication is that Reed’s estimated alpha falls too slowly for an implosion design, which appears to me to mean the yield rate or his fission rate curve also decreases too slowly. Perhaps, someone could ask Reed for his opinion. With regard to the zero (or better near zero) for the yield rate (in joules per shake or kilotons per shake) in an implosion system, the reason is that the yield rate curve reflects the onset of the explosion. In any case, wanting to learn more is hardly to be criticized. For those interested in our legal reasoning why Yousaf is wrong about the IAEA’s mandate, I would refer you to “Understanding the IAEA’s Mandate in Iran: Avoiding Misinterpretations” by Olli Heinonen, Orde Kittrie, and me and endorsed by John Carlson, Trevor Findley, Pierre Goldschmidt, and Andreas Persbo (available at http://www.isis-online.org/uploads/isis-reports/documents/misinterpreting_the_IAEA_27nov2012.pdf). By the way, it is not a work of ISIS but a collective effort that incorporates many decades of experience and expertise in safeguards law and practice.

    • yousaf (History)

      Yes, as we mentioned in our Bulletin piece such graphs and academic papers are freely available online — this does not mean they are “sophisticated”: they are toy models, and get many details wrong. In the words of one weapons scientist they are “childish and facile”.

      Yes, even Reed’s model is not a real model: it is also a toy model but captures the tail behavior better. Talk to some people with weapons experience, e.g. George above:

      “It should look like a nearly immediate cliff down to less than 5% energy per unit time and longer exponential decay from that.

      I am using grossly rough numbers on purpose, but the tailoff is just wrong.”

      ============

      I hope the IAEA has some better evidence than that Iran ran a model from 1972 or 1981 or 2009 that was freely available in libraries or the internet!

      The armscontrollaw article I mentioned was a rebuttal to that ISIS document, by a Prof. of law who has written at least 2 books on the NPT.

      I reiterate what I mention above:

      “Again, it is important to keep this in context: Even if the graph were technically correct (it is not), and if the graph was made by Iranian scientists, and the graph is related to nuclear issues, it still is not obvious to me that computations violate the IAEA-Iran safeguards agreement.

      Iranian scientists and students are allowed to compute what they like. It is a sovereign nation. Yes, it is bad form and perhaps against the spirit of the NPT (if any of this story is true), but not a legal breach. See Prof Joyner’s comment above on that.

      Lastly, let’s keep in mind that if anyone is going to get overly concerned about this graph (if true, if nuclear) then they should probably investigate this PhD thesis written at a *military* institute in Brazil:

      “Numerical Simulation of Thermonuclear Detonations in Fission-Fusion Hybrids Imploded by Radiation”:

      http://www.jb.com.br/pais/noticias/2009/09/05/a-explosiva-descoberta-brasileira/

    • Dan Joyner (History)

      And you can see my response to David and his crack legal team’s analysis of this complex international legal question – you know, the report written by two non-lawyers and one non-international-law expert – at this link:

      http://armscontrollaw.com/2012/11/27/a-whole-isis-report-devoted-to-little-ol-me/

    • Jeffrey (History)

      I am sorry, but I find it very inadequate to simply complain that two people are not laywers, while the third is not the “right” type of lawyer. (How a lawyer who worked in the Office of the Legal Adviser at the U.S. Department of State on precisely these topics is unqualified to participate in a discussion with the great Dan Joyner eludes me.) This is precisely the same thing you did with the review of your book, despite the fact that the lawyer who reviewed it spent 14 years at ACDA and more than three years as Special Representative of the President for Nuclear Nonproliferation.

      This is to say nothing of a person, like Olli, who served as a Deputy Director-General for Safeguards and is certainly qualified to describe what the IAEA OLC thinks about the scope of safeguards he applied.

      Their point is a simple one: Your view is not the one held by the vast majority of lawyers, including those at the IAEA. I don’t understand why you won’t acknowledge that your view is the minority view. Why not simply admit you have a revisionist opinion and argue that we ought to accept your different interpretation on the merits?

    • yousaf (History)

      Thanks Dan.

      One has to enjoy the way Albright presents his opinion as a declarative “expert” finding: ” To that end, it is essential to urge the Iranians to cooperate with the IAEA, recognize that the information in the IAEA’s hands is sufficient to raise such questions, and support the IAEA in asking Iran for clarifications. ”

      No, not at all. That is only an opinion. Urge away.

      It is not at all clear that the information is authentic and in any case according to the AP report the person who allegedly did it was assassinated. The Iranians may already have cooperated by saying that such evidence is fabricated.

      Does Albright know it is not fabricated?

      We should keep in mind that the IAEA itself recognizes the limits of its legal ability to pursue all such evidence, whether true or fabricated. e.g.

      http://www.iaea.org/Publications/Documents/Board/2006/gov2006-15.pdf

      “…absent some nexus to nuclear material the Agency’s legal authority to pursue the verification of possible nuclear weapons related activity is limited…”

      What needs to be done — instead of hounding Iran about possibly fabricated evidence possibly from an assassinated scientist that has nothing to do with nuclear material — in my view, is to make a deal to suspend 20% enrichment in exchange for removing some sanctions.

    • yousaf (History)

      Jeffrey makes an interesting point though still a West-centric one.

      Since the NPT extends to many developing nations it is essential also to solicit their opinions and the legal arguments of their lawyers. e.g. Amb. Mohamed Shaker.

      If one does so, I think one will likely find that Prof. Joyner’s opinion would align with the majority’s.

      Getting a few people together who are not lawyers but instead have worked for the IAEA and the USG is hardly a way to obtain an unbiased legal finding on whether they think the IAEA and the USG is doing the right thing.

      But, hey, I am no lawyer.

    • Dan Joyner (History)

      Jeffrey,
      I have made my substantive arguments in response to Albright and his report at the link I provided in my comment. I’ve also made those substantive arguments in other places on my blog, and in the currently ongoing roundtable at the BAS, which I cite to on my blog. So I have made substantive arguments on all of these points ad nauseum. I don’t make those substantive arguments on your blog because you don’t like it when I do that. Youve made that perfectly plain before. Thats why I now limit myself to very short comments here.
      But since you have raised the issue, I think this is a point that people in the nonproliferation issue community still refuse to understand – the legal issues involved in this area are complex issues of international law. And yet, everyone who works in this area in any capacity, and with any experience and education, seems to think that they can do correct international legal interpretation. This baffles me. International law is an incredibly complex discipline that many of us have spent many years becoming highly educated in and mastering through research, publication and teaching. But the problem is that there have been far too few real, trained international legal scholars who have worked on arms control law issues in the past. And so in their absence, everyone else in the area has just assumed they could do legal analysis just as well. This is incorrect.
      Its not an ad hominem attack to question someone’s professional credentials to perform a specific kind of analysis. Its simply one way – in addition of course to substantive argumentation which I’ve also done – to point out that there are complications of doing legal analysis in this field that the authors of this report cannot understand because they aren’t trained to understand them.
      ALbright and Heinonen, whatever their other qualifications, are not lawyers. How can they possibly understand how to do complex international legal analysis properly? Orde Kittrie is a law professor. What I’m saying is that, in my opinion, and anyone who looks at what there is of Orde’s published work can judge for themselves, he has written very little actual international legal scholarship. Its mostly domestic law and arms control specific stuff like on sanctions. But not real international legal scholarship. And therefore in my opinion he has not demonstrated the ability to understand how to do complex international legal analysis.
      I will admit that my legal analysis is in the minority among lawyers in the nonproliferation epistemic community. And thats basically who is represnted by this report. But I maintain that among international legal scholars – i.e. members of faculty of law schools who specialize in international legal scholarship – my arguments are considered correct.
      This means that among the nonproliferation epistemic crowd I am constantly criticized. I can live with that. But riddle me this. If I’m so wrong about everything, why did David feel he had to muster 8 people to write and endorse something attempting to explain that I’m wrong? If I was so obviously wrong, wouldnt everyone be able to see that for themselves? On the contrary, its because my arguments are persuasively correct that David and others feel the need to commit so much in the way of reources and time to try and convince people I’m wrong.

    • Jeffrey (History)

      “But riddle me this. If I’m so wrong about everything, why did David feel he had to muster 8 people to write and endorse something attempting to explain that I’m wrong? If I was so obviously wrong, wouldnt everyone be able to see that for themselves?”

      Actually, there are plenty of terrible ideas that persist largely because certain parties find them convenient. China’s alleged 3,000 nuclear weapons hidden in tunnels strikes me as one example of something evidently false assertion that nonetheless pollutes our discourse because it is convenient for the preexisting policy preferences of certain persons. Without speaking for David, I would observe that, whatever the merits of your arguments, they will be very popular with certain people who would like to prevent the IAEA from being able to administer safeguards effectively. You may not feel responsible for how your arguments are used — perhaps you feel the chips will fall where they may — but you can hardly fault David, Olli and others who disagree with your factual premise from worrying about the consequences.

      And they may have a point. It would be easier to accept your arguments if your analysis of the narrow mandate was accompanied by any desire at all to see the IAEA given enough tools to do its job. The fact that you appear to prefer, as a policy matter, a narrow mandate for the IAEA naturally raises skepticism about your assertion about the situation as a legal matter.

  27. yousaf (History)

    BTW, folks may be interested in my take on what lessons the IAEA can draw from the this disaster:

    http://www.csmonitor.com/layout/set/print/Commentary/Opinion/2012/1205/Flawed-graph-weakens-case-against-Iran-nuclear-program-video

    excerpt:

    “The image also does not imply that computer simulations were necessarily even run: Similar diagrams can be found in nuclear science textbooks and on the Internet, and it could even be a sketch copy for illustrative purposes, for instance for a class project. Odd as it may sound, it is possible that the IAEA is confronting Iran with shoddy homework gleaned from some Iranian college.

    There are a number of other problems in the IAEA reports on Iran: For example, the agency keeps saying that it cannot “provide credible assurance about the absence of undeclared nuclear material and activities in Iran” or that “all nuclear material in Iran is in peaceful activities.” But the agency cannot be expected to do this – that is not its job. Pierre Goldschmidt, the former deputy director of Safeguards at the IAEA summed it up well: “The Department of Safeguards doesn’t have the legal authority it needs to fulfill its mandate and to provide the assurances the international community is expecting.”

    In fact, not only is it legally problematic to fulfill such a verification, it is a logical impossibility: The agency cannot prove the absence of something. There can always be somewhere in Iran where the IAEA has not looked. In fact, no one can reasonably task the IAEA to prove a negative in any country, whether it be in Brazil, Argentina, or the 49 other nations for which it is evaluating the absence of undeclared nuclear activity.

    The most sensible way to wind down the impasse with Iran now is to recognize that although Iran may have been non-compliant with the IAEA in the past, it is in full compliance with its safeguards agreement now: The nation is not diverting any declared nuclear material to any weapons program. The IAEA has verified this every year since it began monitoring Iran’s program. Hounding Iran about possible activities it may or may not have done years or even decades ago – especially if some of the allegations are possible hoaxes – is not going to solve anything.

    A smart move would be to start to roll back sanctions on Iran in exchange for Iran suspending its uranium enrichment to 20 percent, as both Henry Kissinger and I have suggested.

    …..If the world community wants the non-proliferation regime to survive it will have to insist on greater consistency in the application of international laws.

    In another instance, although Iran, Egpyt, South Korea, and Libya were all found non-compliant in the past, only Iran was referred to the UN Security Council. Again, Pierre Goldschmidt captured it well: “The actions taken by the [IAEA] board in each case were inconsistent and, if they go uncorrected, will create unfortunate precedents.”

    The leaked “Iranian” graph doesn’t bolster the IAEA’s case against Iran – it undermines it. The IAEA is rapidly losing credibility. It should stick to its technical mission of nuclear materials accountancy and call off the wild goose chase in Iran.”

  28. David Albright (History)

    I wanted to correct the link I posted that refutes the positions of Yousaf and Dan Joyner on the IAEA’a mandate. It should be http://www.isis-online.org/uploads/isis-reports/documents/Misinterpreting_the_IAEA_27Nov2012.pdf . I do not think many serious people would agree with Yousaf’s mischaracterization of the Los Alamos HENRE code as a toy model. And he now states that his evidence for his statement that the shape of the AP graphs have problems is what he calls a toy model. He demonstrates why it is hard to take his assessments seriously. I would also like to partially correct and reiterate a sentence from my earlier post. Despite the limitations of the public information, it is essential to urge the Iranians to cooperate with the IAEA, recognize that the information in the IAEA’s hands is sufficient to raise questions about the possible military dimensions of Iran’s nuclear program, and support the IAEA in asking Iran for clarifications.

  29. yousaf (History)

    Albright fails to grasp the difference between a code and a model. A code can be completely correct but be used in a toy model: any published model of nuclear explosions will be little more than a toy model. The real people who can peer-review such things cannot — for obvious reasons — comment substantially on such models.

    Albright also ignores the opinions of actual weapons scientists.

    Albright thinks that because some Iranian student or professor may have taken a graph from a 1972 publication (or a 1981 one or 2009 one) that we need to be concerned.

    In fact, it is fine to be concerned. It is simply not a legal problem, nor does it indicate a nuclear weapons program. Absent a nexus to nuclear materials it is irrelevant as far as the IAEA is concerned. In any case, even if we believe the alleged story, the person who made the graph was assassinated.

    Albright also does not provide any evidence that the graph is actually genuine — does he have any evidence that it is not a hoax?

    Instead of trying to resolve the Iran nuclear issue and working towards a deal to suspend 20% enrichment, Albright would like to be worried about a possibly hoaxed graph that can be obtained from the internet.

    Please.

  30. David Albright (History)

    Yousaf’s statements about a code versus what he calls a toy model merely confirm that he knows little about the types of such codes that exist in the public domain and their uses. And we are after all talking about the yield rate and yield graphs generated by those codes, not the impact of equations of state data or neutron production rates from a neutron initiator. Robert Kelley’s dismissive remarks about the AP graphs, which Yousaf hangs on, were made in the heat of the moment and were not accurate, as posts on this thread have demonstrated. I would be surprised if Bob would defend those early comments today in the same way– at least when I contacted him he did not want to discuss them. Yousaf also seems to be avoiding discussing the shapes of the graphs, which codes predict these shapes, what this could say about the AP graphs, and a host of other technical questions this media report has generated. Moreover, Yousaf and Dan make interpretations of the IAEA mandate that are widely seen as wrong and launch into tiresome personal attacks and mischaracterizations of people’s positions. Do others want to weigh in on the technical discussion?

    • yousaf (History)

      I did not see any tiresome personal attacks — can you clarify?

      As I mentioned before, it is important to keep this in context: Even if the graph were technically correct (it is not), and if the graph was made by Iranian scientists, and the graph is related to nuclear issues, it still is not obvious that computations violate the IAEA-Iran safeguards agreement.

      Iranian scientists and students are allowed to compute what they like. It is a sovereign nation. Yes, it is bad form and perhaps against the spirit of the NPT (if any of this story is true), but not a legal breach.

      Does Albright have any evidence that the graph was not taken from a journal and planted in Iran? i.e. a hoax?

      If so, Iranian comments that they are being confronted with fabrications may well be correct. That would be them “coming clean” on this issue. That would be Iranian cooperation. How are they supposed to cooperate with possible forgeries?

      How/Why is David Albright so adamant this is not a forgery?

      All that the several publicly available papers with toy-models of nuclear explosions show is that it is OK to do such academic research and that, e.g., anyone in Tehran or Tel Aviv could have gotten ahold of such toy model graphs very easily, even over the internet.

      We mentioned clearly in the Bulletin piece and several times in the comments that such graphs are no secret and are easy to obtain. They also do not indicate that models were actually run.

      Several news organizations including the Fairness and Accuracy In Reporting have commented on the similarity between the run-up to the Iraq war and vague “could be” “maybe” “possible” innuenedo now being used in discussions of Iran:

      http://consortiumnews.com/2011/11/08/an-iraq-wmd-replay-on-iran/

      The record speaks for itself.

      It would be useful now to reach a deal on suspending Iranian 20% enrichment instead of launching a wild goose chase into how a possibly hoaxed toy-model graph from the internet got into Iran.

  31. David Albright (History)

    I leave it to the reader to judge the nature of Yousaf Butt based on what he said at the start of his post and the link at the end. The article that Butt posted is untrue about me and many other things but in line with what Yousaf calls “factual” sources. Yousaf seems unable to carry on a civil discourse and I will not be drawn into his nastiness.

    • Jeffrey (History)

      Dear David:

      I apologize. I’ve been in Paris and not moderating this thread. I see things are evidently out of hand with the attacks on you. I will adopt a firmer policy in moderating posts and will not permit this to continue. I’ll send you a note separately to discuss remedies.