CTBT Article XIV Conference

by: Alicia Godsberg

This past Thursday and Friday marked the 6th bi-annual Article XIV Conference, the Conference on Facilitating the Entry Into Force of the Comprehensive Nuclear Test Ban Treaty (CTBT).  This year’s conference was held at the United Nations in New York and was met with a measure of cautious optimism – most states voiced their appreciation of President Obama’s pledge to work toward US ratification of the CTBT, while many states recognized the challenges of obtaining all the necessary ratifications for entry into force of the Treaty and mentioned the challenges to the nonproliferation regime stemming from the lack of the Treaty’s entry into force (despite former commitments to do so) and from the DPRK’s 2006 and 2009 nuclear tests.

Entry into force of the CTBT has been on the international agenda for thirteen years. Because the US, China, UK, France, and Russian Federation have all imposed a voluntary moratorium on national nuclear testing, many question the need for entry into force of the CTBT.  Although the Treaty would bring few new tangible benefits, the political impact of entry into force would be tremendous.  As explained below, the vast majority of sates see entry into force of the CTBT as somewhat of a litmus test for the future viability of the nonproliferation regime. Continue reading

Draft Order Would Set New Limits on Classification

“No information may remain classified indefinitely,” according to a draft of an Obama Administration executive order on national security classification policy.

As a statement of principle, this may seem tame and self-evident.  But until now, no Administration has been willing to make such a categorical statement about the temporal limits of national security secrecy, and it may have significant policy consequences.

An August 4 draft of the executive order (pdf) was prepared by an interagency task force in response to a May 27 memorandum from the President.  The draft is still subject to revision, and has not yet been formally transmitted to the White House for review and approval.  Release of the “highly deliberative draft” executive order was specifically denied by National Security Advisory Gen. James L. Jones in a September 2, 2009 letter (pdf).  But a copy was obtained by Secrecy News.  Some aspects of the draft order were previously reported by Bill Gertz in the Washington Times on September 24.

The draft order, which does not represent anything like a transformation of the existing secrecy system, nevertheless has some valuable and innovative features, as well as some disappointing omissions, and a few retrograde steps.  See this side-by-side comparison (pdf) between the August 2009 draft and the current executive order.

The draft order states (section 3.3g) that all records are to be automatically declassified no more than 50 years from the date of origin, with the sole exception of records that would identify a confidential human intelligence source.  And even such intelligence records must be declassified no more than 75 years from the date of origin, with no exceptions.  This is something new.  An existing requirement for “automatic declassification” at 25 years would remain in place, but at the 25 year point there are still nine expansive exemptions to declassification.  Under the new policy, the exemptions would diminish over time and then disappear altogether.

The draft would require a “Fundamental Classification Guidance Review,” involving a continuing review of all agency classification guides in order “to identify classified information that no longer requires protection and can be declassified” (section 1.9).  This is a version of a proposal advanced by Secrecy News (e.g., here and here), and among all of the potential changes to the executive order, it was our top priority.  If it worked, the fundamental review would introduce a dynamic new element of self-correction into the classification process.

A National Declassification Center would be established to facilitate interagency review of historical records and to resolve quality control issues, presumably leading to more complete and expeditious access to such declassified records (section 3.7).

Other constructive if not bold steps include:  new requirements for training of classification officials in avoiding overclassification;  a requirement to identify by name those who derivatively classify information originally classified by others in order to improve accountability;  a higher threshold for reclassification of declassified information;  provisions for review of previously granted exemptions of file series from 25 year automatic declassification.

On the less constructive side, the draft order affirms that “no agency may declassify information that originated in another agency… without the consent of the originating agency” (section 3.1f).  This reinforces a cherished view that agencies “own” the information they produce, and that they retain control over its release and dissemination.  It is arguably the single most profound conceptual flaw in the classification system, and it immeasurably complicates the declassification and disclosure process.  Even the new National Declassification Center will not possess unilateral authority to declassify information, but will only provide “timely and appropriate processing of referrals” from one agency to another.

The draft order does not make any provision for a “declassification database” that would offer easily accessible electronic versions of declassified records, or at least bibliographic data on exactly what has been declassified.

The draft does not provide enhanced oversight or declassification authority to the Information Security Oversight Office.

The draft would perpetuate the veto authority that was granted to the CIA by the Bush Administration over declassification decisions made by the Interagency Security Classification Appeals Panel, but it would transfer that authority from the CIA to the Director of National Intelligence.  CIA’s seat on the interagency Panel would be reassigned to the DNI.

Beyond the ongoing battles over parochial agency interests that are at stake in the new draft, there seems to be a growing sense that the existing secrecy system, even if it is to be buffed and polished one more time, has finally reached obsolescence.

“As soon as we complete our revision of the existing Order,” wrote Gen. Jones on September 2, “I plan to begin discussions… about a more fundamental transformation of the security classification system.”

The Qom Uranium Enrichment Facility – What and How Do We Know?

On Friday, President Obama, President Sarkozy, and Prime Minister Brown revealed  a covert Iranian uranium enrichment facility near Qom. Obama announced that “the size and configuration of this facility is inconsistent with a peaceful nuclear program.” In a briefing , Senior White House Administration Officials clarified that the facility is designed to hold about 3,000 centrifuges. Although, this number is not large enough to “make sense from any commercial standpoint, […] enough for a bomb or two a year, it’s the right size.”
It is too early to independently verify the US statement that Iran is planning on setting up 3,000 centrifuges at Qom. The IAEA has confirmed  that it has received a letter from the Iranian government announcing the facility. The letter affirms that Iran will provide more information as appropriate. Iran claims that it is not bound by the revised Code 3.1 of its Subsidiary Arrangement with the IAEA and, therefore, they need to announce new facilities only 180 days before nuclear material enters the site and material will not be introduced for at least 6 months as of last Monday, when Iran sent the letter to the IAEA. In an interview with CNN’s Larry King Live, President Ahmadinejad explained  that they have informed the agency a year before they were obligated to and Iran’s Atomic Energy head, Ali Akbar Salehi has said  that no nuclear material has entered the facility yet. Iran claims  that no machines have been installed.
However, some press reports  state that the facility is “within a few months of being completed”. The good news here is that Iran has publically announced the facility (although it is unclear whether it decided to do so only because the cat was already out of the bag) and has said  that it would comply fully with the IAEA (although Iran and the IAEA do not agree what those obligations are).
ISIS recently published satellite images of possible locations of the Qom enrichment plant. Unfortunately, as cool as satellite photos are, they only show tunnel entrances in a mountain. We have found many of those around that area, playing around on Google Earth. Moreover, the locations are simply guesses based on information that has been disseminated by the media. We cannot tell much about the number and type of centrifuges that will be installed at Qom from the ISIS satellite imagery. White House Administration Officials have admitted  that “we’ll have to wait for the IAEA to get inside there and to report back.”
3,000 Centrifuges at Qom – How Do We Know?
How could the Administration know that Iran is installing 3,000 machines? One way would be to compare the area of the Qom facility to that of the enrichment plant at Natanz. Centrifuges take a certain amount of floor space and if we knew the average area per cascade, we could approximate how many machines can fit in a given space. As FAS’ Acting President, Ivan Oelrich points out, you can come up with an estimate for the size of the facility based on the amount of rock that the Iranians are throwing out (if they are digging a hole in a mountain, they have to dispose of the material somewhere). Geoffrey Forden has an example  of what such an analysis could look like. He reached the conclusion that the amount of rock is consistent with the Administration’s statement. Unfortunately, this involves a lot of assumptions and as Forden puts it, “doesn’t prove anything”.
You can also tell something about the size of the tunnel if you knew how much explosive was used to blast the hole. We can also consider the power lines that are going inside the facility and estimate the energy consumption that they are meant for. Perhaps the US has someone working on the inside or has intercepted communications saying, “Send 3,000 centrifuges to Qom.” However, there is no way to know that a particular tunnel will be used to house centrifuges until we have more information provided by other sources. The White House admitted that at early stages of construction, such a facility can have multiple uses and this is in partly why they chose to wait until they had enough evidence to make a compelling argument to the IAEA. Still, outsiders cannot independently verify this information.
What Type of Machines?
If we accept the 3,000 number as true, we also have no way of knowing what type of centrifuge Iran will install at Qom. Other than the IR-1 currently operational at Natanz, Iran has been testing 4 other types
of machines: IR-2, IR-2m, IR-3 and IR-4. It is foreseeable that Iran could wait until one of the more advanced machines is ready for mass production and install those instead. Since carbon fiber models are known to have at least twice the separative capacity relative to aluminum alloy ones, newer models are expected to have a much better performance that the current IR-1 setup at Natanz. The type of machine used would greatly change what they can be done with a set up of 3,000 machines.
Iran may be preparing for the set up of one of the newer centrifuge models. After the last IAEA report  on Iran came out in August 2009, there were statements  in the press that Iran was slowing its expansion of uranium enrichment at Natanz. As it turned out, Iran had decreased the number of operational centrifuges but continued to install new machines and run centrifuges in vacuum. Although some speculated that this may mean that Iran is running out of UF6 or centrifuge parts, a slowdown in the rate of set up of new machines may mean that Iran is preparing for a new centrifuge model. If Iran is close to developing a reliable, higher performance machine, it may prove more economic to wait or slow down setting up IR-1s. So, it is definitely possible that by the time Qom is fully piped and electrified, a new type of centrifuge will be ready for installation.
What Can Be Done with 3,000 Machines?
The size of a facility does not determine whether it can or cannot produce weapons-grade, or highly-enriched, uranium (HEU). Both enrichment to a low degree for a nuclear reactor and to a high degree
for a nuclear weapon are done by gas centrifuges, in fact, potentially exactly the same machines.
One way to tell whether a cascade of centrifuges is used for LEU or HEU production is to look at the  configuration of the machines, or how they are piped together. The set up and piping of the cascade will be different if they are enriching natural uranium to low-enriched uranium (LEU) when compared to natural uranium to HEU. However, they always have the option of using a LEU production set up and simply running the material through several times until they get HEU.
Aside from what is possible in theory, certain things make economic sense and others don’t. To enrich enough LEU for an average 1000 MWe reactor, you need 135,600 kg-SWU/yr. If the 3,000 machines are IR-1s with a separative capacity of 0.5 kg SWU/yr, it would take them about 90 years to get one year’s fuel load. This of course makes no sense. However, if they want to get one bomb’s worth of HEU (from natural uranium), they need 6,320 kg SWU/yr and this would take you a little over 4 years. All of these examples can be worked through with FAS’ new and improved uranium enrichment calculator.
The third option is to take LEU from Natanz and enrich it to a bomb’s worth of HEU. This would take about a year, depending on how much material they are willing to waste. So, if they are trying to divert LEU from an existing facility such as the one at Natanz, the numbers add up perfectly (almost too perfectly). However, diversion of nuclear material from the enrichment plant at Natanz or the conversion plant at Isfahan is near impossible to go undetected if the facilities are under IAEA safeguards. Although uranium mines and mills are not under safeguards, so far there is no sign of a clandestine conversion plant in Iran. There is always the option that the Iranians could just kick the inspectors out and have breakout in one year or less.
A Pilot Plant
On the other hand, Iran hasn’t claimed that the centrifuge plant at Qom is an industrial facility, but a “semi-industrial-scale plant ” or a “pilot plant”. If they are planning on testing a handful of new machines (like at the Pilot Fuel Enrichment Plant at Natanz) or having a set up of centrifuges someplace where an Israeli air raid will not have much effect, to retain enrichment capability and rebuild their industry, this may make more sense. They would not need huge amounts of machines to do this. Currently at PFEP, Iran tests its new centrifuge models by running several cascades of 10 or 20 machines at a time.
Recently , Iran proposed to buy 19.75 percent enriched uranium from the US for medical purposes. According to the IAEA, uranium with about 20 percent enrichment is considered HEU, although it is not of weapons-grade. If the US declines the offer (which it most probably will), Iran could use this as an excuse to make its own medical grade material at the new facility.
According to unclassified US document s released by ISIS, although the Qom plant is reportedly located on an Islamic Revolutionary Guard Corps Base, it is managed by the Atomic Energy Agency of Iran.
So, is the “size and configuration” of the plant inconsistent with a peaceful nuclear facility? Not entirely.  While the circumstantial evidence raises suspicions, based on available evidence, we cannot currently prove it is a military facility.   First, we have no way to confirming the Administration’s statement that Iran will set up 3,000 centrifuges at Qom until the IAEA receives and verifies design information of the facility.  Even if the intelligence were correct, Iran could have changed its plans since the existence of the facility became public, especially if no machines have been set up yet. The 3,000 announced centrifuges by the US are definitely not enough for industrial-scale production of LEU for nuclear reactor fuel. This doesn’t automatically mean that the facility was meant for bomb production, especially if there are no machines installed yet. We don’t know how the plant is configured since, again, no machines have been installed. And, again, this will not be known until inspectors are on the ground.
The location of the facility in a protected and heavily disguised location certainly isn’t helping Iran’s peaceful nuclear program claim. Although repeated Israeli threats of an attack may have developed circumstances for Iranian nuclear safety concerns, this does add to Iran’s track record of ambiguous behavior.
Since the technology to enrich uranium to a small degree for nuclear fuel and to a large degree for nuclear bombs is the same, ultimately the question falls on proving Iran’s intent. Senator Feinsten, the Chairman of the Senate Intelligence Committee said  that Iran’s “intention to produce weapons-grade uranium in the Qom facility has not yet been proven.” If Iran is developing a peaceful program, then it should assuage concerns by adopting further transparency measures, like implementing the revised Code 3.1 of the Subsidiary Arrangements and ratifying the Additional Protocol. On the bright side, US intelligence was good enough to be able to detect a covert nuclear facility. And Iran’s letting inspectors in at Qom is good news.

By Ivanka Barzashka

On Friday, President Obama, President Sarkozy, and Prime Minister Brown revealed a covert Iranian uranium enrichment facility near Qom. Obama announced that “the size and configuration of this facility is inconsistent with a peaceful nuclear program.” In a briefing , Senior White House Administration Officials clarified that the facility is designed to hold about 3,000 centrifuges. Although, this number is not large enough to “make sense from any commercial standpoint, […] enough for a bomb or two a year, it’s the right size.”

It is too early to independently verify the US statement that Iran is planning on setting up 3,000 centrifuges at Qom until the IAEA receives and confirms design plans of the facility. Although the circumstantial evidence certainly isn’t helping Iran’s peaceful nuclear energy claim, we cannot definitively conclude that the enrichment plant has a military function. Senator Feinsten, the Chairman of the Senate Intelligence Committee said that Iran’s “intention to produce weapons-grade uranium in the Qom facility has not yet been proven,” although there are strong indications.

Continue reading

Govt Defends Use of State Secrets Privilege

The government’s assertion of the state secrets privilege in a pending lawsuit brought by a former Drug Enforcement Administration agent will not be affected by the new Attorney General policy limiting the use of the privilege, the Justice Department said last week, because it is already in compliance with the new policy.

In a September 24 appellate brief (pdf) in the case of Horn v. Huddle, Justice Department attorneys urged an appeals court to overturn a lower court ruling that would authorize the parties in the lawsuit to disclose classified information to their attorneys. The Department also defended its use of the state secrets privilege.

An August 26 ruling in the case held that the parties’ counsel had a “need to know” the classified information possessed by their clients, and the court therefore directed the government to authorize the sharing of that information.

The government immediately objected. “The district court’s extraordinary order — compelling the government to grant security clearances and to authorize disclosure of classified national security information to private counsel… — unnecessarily usurps the Executive Branch’s authority and responsibility to protect from disclosure classified national security information as to which the state secrets privilege has been invoked,” the government argued in its September 24 brief.

The government also declared that the Attorney General’s new policy limiting the use of the state secrets privilege, which takes effect on October 1, would have no impact on the present case.

“The assertion of the privilege in this case satisfies the standards in the new policy concerning the applicable legal standards, narrow tailoring, and limitations on the assertion of the privilege. Moreover, the privilege as invoked in this case has been carefully reviewed by senior Department of Justice officials, who have determined that invocation of the privilege in this litigation is warranted,” the government brief stated.

Calculating Output of the New Iranian Uranium Enrichment Plant

On Friday, President Obama announced that the United States knows of a new, undeclared, and hidden underground gas-centrifuge uranium enrichment facility in Iran, near the city of Qom.  Some news reports suggest that 3000 centrifuges will be housed there.  How significant is this discovery?  Well, just in time, our crack FAS researcher, Ivanka Barzashka, has posted on the FAS website a calculator to help you answer questions just like that.


Continue reading

Next Obama Speech: The Pentagon

President Obama has once again pushed nuclear weapons, and his vision for a world free of nuclear weapons, to the center of the world’s stage with his speech yesterday before the United Nations’ General Assembly and his chairing of the United Nations’ Security Council meeting this morning. He reiterated his goal of ratifying the Comprehensive Test Ban Treaty (CTBT), of negotiating a Fissile Material Cutoff Treaty (FMCT) that would end production of bomb-grade nuclear material (something the Bush administration supported in theory but without any verification procedures), of negotiating a treaty with Russia that will “substantially reduce” strategic nuclear warheads, and of strengthening the Non-Proliferation Treaty (NPT). The President also said “We will complete a Nuclear Posture Review that opens the door to deeper cuts, and reduces the role of nuclear weapons.” This morning, as chair of the UN Security Council, the President got unanimous consent to Council resolution that endorsed all the points made before the General Assembly.

The President’s remarks are powerful and plain and were overwhelming well received by all of us who have long hoped that the world might someday be free of nuclear weapons. Still, I am worried that the message has been clearer at the UN, and in Prague, than it is here in Washington. If we look at the direction the bureaucracy and politics are taking here, there is reason to worry that the President’s vision will be dangerously diluted. Continue reading

A Problematic New Policy on State Secrets

The Department of Justice yesterday released its long-awaited new policy on the state secrets privilege, which the government uses in litigation to withhold evidence when it believes that disclosure would harm national security.  The new policy, presented in a memorandum from the Attorney General, includes procedural and substantive changes to current practice.  But it reserves decisions over the exercise of the privilege to the executive branch, and it appears to have garbled its treatment of judicial review.

See “Policies and Procedures Governing Invocation of the State Secrets Privilege” (pdf), memorandum from the Attorney General, September 23.

The new policy specifies that the use of the state secrets privilege must be supported by an evidentiary record that justifies its use and demonstrates that it is necessary in order to avoid “significant harm” to the national security.  A recommendation to invoke the privilege must be reviewed by senior Justice Department officials, and approved by the Attorney General.  The policy also provides for Inspector General review of claims of government wrongdoing when adjudication of those claims is prevented by the privilege.

Collectively, these measures “will provide greater accountability and ensure the state secrets privilege is invoked only when necessary and in the narrowest way possible,” said Attorney General Eric Holder.

Perhaps unsurprisingly, the new policy, which will take effect on October 1, would preserve executive branch discretion over the use of the state secrets privilege.

More surprisingly, the policy seems to have fumbled the question of judicial review.  A Justice Department news release about the Attorney General’s memorandum declared promisingly that “in order to facilitate meaningful judicial scrutiny of the privilege assertions, the Department will submit evidence [justifying the privilege] to the court for review.”

But strangely, the memorandum itself says no such thing (as noted by Bill Leonard).  Questioned about the discrepancy, a Justice Department official said yesterday that the intent to submit the evidentiary record to the court for review, though left unstated by the Attorney General, was “a necessary inference” and he said that it would be done “in every case.”  Maybe so.

Internal executive branch procedures to limit official secrecy are not inherently futile or self-serving.  The Interagency Security Classification Appeals Panel, an executive branch body which reviews appeals of mandatory declassification review requests that were denied, has actually been more effective than any court in combating overclassification.  To the surprise of everyone involved, it has overturned the classification of information in a majority of the cases presented to it since 1996.

More often, however, independent review from outside the executive branch plays an essential role in identifying and reconciling competing interests in secrecy and disclosure.

In a practice that is closely analogous to the new state secrets policy, the Justice Department is supposed to conduct its own evaluation of agency denials of Freedom of Information Act requests and to defend agencies in court only when the denied information is clearly exempt from disclosure under FOIA.  If such evaluations were reliably performed, and if only proper agency denials of FOIA requests were ever defended, then the government should never lose a FOIA case.  Yet we know that that is not what happens.  Courts rule against the government in FOIA cases with some regularity, despite the fact that the Justice Department says it only supports cases where the government position is the legally “correct” one.

In the same way, and for the same reason, the executive branch cannot reasonably be expected to serve as the sole and final arbiter of the proper use of the state secrets privilege.

“While I am pleased that the Obama administration recognizes that the Bush approach was a mistake, its new policy is disappointing because it still amounts to an approach of ‘just trust us’,” said Sen. Russ Feingold (D-WI)  “Independent court review of the government’s use of the state secrets privilege is essential.  I urge the administration to work with Congress to develop legislation that sets reasonable limits on the privilege and will not be subject to change under each successive president.”

Iraqi Civilian Casualties, and More from CRS

Noteworthy new reports from the Congressional Research Service include the following (all pdf).

“Iraqi Civilian, Police, and Security Forces Casualty Statistics,” September 17, 2009.

“Iran’s Nuclear Program: Status,” September 18, 2009.

“Military Recruitment on High School and College Campuses: A Policy and Legal Analysis,” September 22, 2009.

“The Second Amendment and Incorporation: An Overview of Recent Appellate Cases,” September 21, 2009.

“Legal Standing Under the First Amendment’s Establishment Clause,” September 15, 2009.

Brazil Book on Nuclear Weapons Draws Scrutiny

A book published this year in Brazil on “The Physics of Nuclear Explosives” prompted concerns at the International Atomic Energy Agency (IAEA) that it revealed classified nuclear weapons design information and that it might signify a renewed interest by Brazil in developing such weapons.  The U.S. Government also requested further details on the matter, the Brazilian press reported.

According to the Jornal do Brasil, which first disclosed the controversy on September 6, the IAEA “wanted the book to be recalled” and demanded more information on the author’s work.  The government of Brazil refused to censor the book and rejected what it described as IAEA interference.

“The Physics of Nuclear Explosives” (“A Física dos Explosivos Nucleares”) by Dalton E.G. Barroso provides a rather comprehensive account of the physical principles involved in nuclear detonations, including both fission and fusion weapons.  There are chapters on reactor physics, radiation hydrodynamics, dynamic compression of solids, inertial confinement fusion, and more.  Most of this information is already freely available to those who care to search for it.  But the author has broken new ground in presenting the results of his numerical simulations of nuclear detonations and characterizations of particular weapons, such as the W-87 warhead.

“One presumes that many of the specific results presented here have never been published in the open scientific literature,” he wrote in the Preface to the book.  “However, such results are based on well-known physical and mathematical models.”

Far from implying that Brazil may have a clandestine nuclear weapons program, Brazilian Defense Minister Nelson Jobim said that the absence of official secrecy surrounding the book demonstrated that the opposite was the case.  “The mere possibility of publishing this work in Brazil, and the material’s free circulation, serve as eloquent proof of the non-existence of an unauthorized nuclear program in the country.”

Dr. Barroso made the same point to Secrecy News.  “My research is academic and has only scientific interest, for if it is not so, how could my book be published?”

“I think the IAEA reaction was premature and exaggerated,” he said by email.  “There are no data in the book that were obtained in any kind of experiments or that were passed to me in secret.”

“A Física dos Explosivos Nucleares” by Dalton E.G. Barroso, now in its second edition, was published by Livraria da Física (439 pages, 2009, in Portuguese).