In July 2010, the U.S. State Department issued the most recent edition of its Congressionally-mandated report, Adherence to and Compliance with Arms Control, Nonproliferation, and Disarmament Agreements and Commitments. This report, which is required by statute, is prepared by the Department's Bureau of Verification, Compliance, and Implementation (a.k.a. "VCI" in Department jargon), but it is exhaustively cleared by a long list of interagency offices and organizations within the Department, across the policy community, and in the intelligence bureaucracy. Commonly known as the "Noncompliance Report" (NCR), this document articulates the official position of the U.S. Government, as a whole, on which countries are in compliance with their obligations, and (usually more interestingly) which are not.
It is a document worth studying, not only for what it tells the reader about the subjects it addresses, but sometimes also for what it says about the U.S. Government that produces it. In this case, to pick one example – and I shall likely deal with more NCR issues in future postings – the treatment of nuclear testing in the 2010 NCR is of special interest. It may tell the reader more about the Obama Administration itself than it actually does about nuclear testing. But first, some history.
I. NCR History and Evolution
The NCR is today a product of the U.S. State Department, but it was previously drafted by the Arms Control and Disarmament Agency (ACDA), at least until that organization was folded into State at the end of the 1990s. In ACDA days, NCRs were spare documents, containing carefully-written and sometimes pointed conclusions, but seldom offering much detail. Notably, however, the NCRs seem to have had an impressive record at flagging emerging compliance issues.
Today, when even the best and most accurate U.S. Intelligence Community (IC) assessments on weapons of mass destruction (WMD) have to labor in the shadow of the gravely flawed IC analyses done on Iraqi WMD from the mid-1990s through early 2003, it is perhaps reassuring to remember that history shows that the NCR – which is not actually an intelligence product but rather the output of compliance analysts in the policy community, informed by intelligence analysts and international lawyers – got some key issues right.
ACDA played a prominent role in calling attention to the Soviets' violations of their arms control obligations, pursuing with vigor and persistence during the 1980s the issue of Moscow's construction of an anti-ballistic missile (ABM) radar at Krasnoyarsk, in violation of the ABM Treaty of 1972. At the beginning of the 1990s, just before the USSR collapsed, ACDA also raised concerns about whether the Soviets had complied with the Intermediate-range Nuclear Forces (INF) Treaty of 1987 with regard to SS-23 ballistic missiles transferred to Warsaw Pact allies. Information received in mid-1991 by U.S. authorities subsequently indicated that East Germany, Czechoslovakia, and Bulgaria had indeed fielded a total of 24 such missiles, and that the Soviets had apparently had warhead-transfer arrangements with these governments. Moscow, therefore, had indeed "probably violated the Elimination Protocol" of the INF Treaty with regard to how it declared and handled re-entry vehicles associated with and released from these cooperative programs.
More dramatically, in January 1993, the ACDA-drafted NCR prepared at the end of the George H.W. Bush Administration assessed that Iran had demonstrated a continuing interest in nuclear weapons and related technology, with a special emphasis upon developing centrifuge technology, and that Tehran was in the early stages of developing a nuclear weapons program. (There may in fact have been earlier public warnings about Iran's nuclear program – which we now know began in the mid-1980s – but I have not myself seen anything dating from before 1993.) These alarms about Iran were repeated in NCRs published from 1994 through 1999, and into 2003, by which point the world was indeed beginning to learn a great deal about Iran's nuclear ambitions, and the previously secret centrifuge facility at Natanz was being visited by International Atomic Energy Agency (IAEA) inspectors. (By 2005, the NCR was able to offer a detailed analysis of Iran's activity, in large part based upon information publicly available from the IAEA, and to offer a formal U.S. Government compliance finding that Tehran had violated Articles II and III of the NPT, as well as its safeguards agreement.)
The January 1993 NCR also got Libya right, raising concerns about that country's compliance with its Nuclear Nonproliferation Treaty (NPT) and IAEA safeguards obligations. This, mind you, was fully a decade before Libya admitted having a nuclear weapons program. (Muammar Qaddafi determined at the time of the U.S.-led ouster of the Saddam regime in Iraq that it would be prudent to allow U.S. and British experts to inspect and eliminate all his WMD in a groundbreaking cooperative program: he began negotiating with CIA and British intelligence officials about this in March 2003, just before the Iraq invasion, and publicly committed to the elimination of these programs before the end of that year. The VCI Bureau coordinated the U.S. interagency effort in 2004 to oversee and verify this elimination.)
The form and structure of the NCR have evolved over time. Modern editions of the report are more detailed than they used to be, back when the NCR was prepared by ADCA. Starting under the Bush Administration – under State Department officials such as Under Secretary John Bolton and Assistant Secretary Paula DeSutter, who were determined to reverse years of Clinton-era soft-pedaling of nonproliferation violations – the report was expanded. (Full disclosure: I was the Principal Deputy Assistant Secretary in DeSutter's bureau at the time.)
During this period, the Report was expanded to cover missile proliferation commitments, as well as to offer more detailed explanations of the standards that U.S. officials applied in making compliance determinations in the nuclear field, and to offer considerably greater detail with regard to the facts at issue in each case. The drafters also took pains to prepare not just an unclassified version of the NCR for public release but also editions at the SECRET and TOP SECRET/SECURE COMPARTMENTED INFORMATION (a.k.a. TS/SCI) levels for use by cleared U.S. decision-makers with a need to know such detail. (More than one noncompliance assessment of a particular country, in fact, was classified in its entirety: some issues could not even be mentioned at the unclassified level without compromising intelligence sources and methods.)
Another improvement to the NCR adopted in the early 2000s was the expansion of each country discussion in order to provide specific information on the "History of Compliance Evaluation," country "Actions," and "Compliance-Related Dialogue and Analysis." It was felt important to provide this additional information, on the grounds that readers in Congress and within the Executive Branch, especially anyone who did not have previous NCRs in front of them, needed some picture of what had come before in order properly to assess what was being assessed now. And so the NCRs expanded, most prominently with the 2005 edition of the Report. (An edition had been issued in 2003, but its drafting occurred too early for most of these Bush Administration reforms to have much effect.)
Especially in the charged compliance-assessment environment of nonproliferation policy after the 2003 invasion of Iraq, these innovations helped make the NCR a very difficult document to produce and clear through the interagency process. In an era when the United States might perhaps even invade a country deemed to be sufficiently threateningly in violation of its WMD obligations, compliance assessments were understood to be of potentially huge importance, and they unsurprisingly became a subject of considerable contestation within the bureaucracy. (For a discussion of some of VCI's travails with politicized reactions within the State Department to compliance-related information, see my NPF posting of January 4, 2010.)
Quite apart from the stakes involved, moreover, the new NCR format was much more comprehensive than earlier versions, and required correspondingly more time and effort both in its preparation and in the long and painful process of interagency clearance. For both of these reasons – the drafters' ambitions for the NCR, expressed in its greatly increased factual and analytical detail, combined with the degree to which interagency counterparts felt it necessary to fight over its contents now that NCRs actually seemed important – the Bush Administration never managed to make the document an "annual" report. As noted, only two were produced during that period: in 2003 and 2005.
Well, the Obama Administration has now produced its own NCR – a modified version of a lengthy draft that had been circulated by DeSutter's VCI late in the Bush presidency, only to run aground in the process of obtaining clearance from interagency bureaucrats all too well aware that George W. Bush would not have a third term in office. This is not the place to discuss the 2010 NCR in much detail. In this first NPF posting about the new NCR, however, I would like to flag one issue that jumped out at me: its treatment of nuclear testing.
II. Compliance with Nuclear Test Moratoria
Anyone comparing the Obama Administration's 2010 Noncompliance Report to its predecessors will notice the addition of a completely new section: the NCR's first-ever treatment of "moratoria on nuclear testing." This new section, however, is remarkably short: a mere paragraph, dropped into the Report as a seeming afterthought at the very end, on page 92. In its entirely, this new section reads as follows:
"By September 1996, each of the nuclear-weapon States (NWS) under the NPT (China, France, the Russian Federation, the United Kingdom, and the United States) had declared a nuclear testing moratorium and had signed the Comprehensive Test Ban Treaty (CTBT), which has not yet entered into force. Although the scope of each moratorium has not been publicly defined, there were no indications during the reporting period that any NWS engaged in activities inconsistent with its declared moratorium."
For such an important – and contentious – topic, this is a notably cursory treatment, and one which begs more questions than it answers.
This appears to be the first time any NCR has ever treated nuclear testing moratoria, but there is no discussion of the history of compliance with these pledges since their adoption in the 1990s. Indeed, the NCR purports only to pronounce upon what occurred "during the reporting period" of this Report, a range of dates which is not specified but presumably means only the last few years – i.e., since the data cutoff point for the 2005 NCR. The phrasing of this section suggests that violations could have occurred earlier, but we have no way of knowing whether that is the case because of the Obama report's decision to limit its inaugural discussion of the subject to the current "reporting period." Countries' behavior between 1996 and c.2004 is not discussed. (One wonders why. If the intention is to demonstrate that these countries take their test-ban promises seriously, and in fact they do, why not describe the whole period?)
Moreover, the Obama text expressly notes that "the scope of each moratorium has not been publicly defined" even as it blithely announces that there are "no indications" that any of these moratoria has been broken. It is not at all clear what this means, if anything. If we don't know how the moratoria are to be interpreted, how do we know what an indication of noncompliance would look like?
The new NCR section, therefore, is a very strange one. To summarize, the Obama Administration seeks to reassure us that although we have no idea what anyone else actually means by these no-testing pledges, there is no reason to believe that anyone has broken them – or at least not recently, anyway. Wow.
The reader would do well to compare the empty nothingness of the Obama NCR's gratuitous test-moratorium assessment to the discussion of CTBT in the final report (in 2009) of the bipartisan Strategic Posture Review Commission (SPRC) appointed by Congress to articulate a consensus position within the U.S. policy community on nuclear weapons issues. The commissioners – led by two former Secretaries of Defense – were given access to U.S. Government information on both the classified and unclassified levels, studied these issues extensively, and indeed did issue a unanimous report. Or rather, an almost unanimous report.
Interestingly, the CTBT was the sole topic on which the commissioners did not reach a consensus position. Instead, they split into "pro" and "anti" factions, each offering a brief section in the final report articulating their opposing arguments. This is not the place to get into the merits of CTBT, but the SPRC's treatment of CTBT helps explain both why the Obama Administration felt it necessary to mention the test moratorium issue at all, and why the 2010 NCR's compliance finding was so embarrassingly insubstantial.
To my eye, the most interesting thing about the SPRC commissioners' CTBT debate was not the fact of their disagreement or the substance of their arguments – for those battle lines have been starkly drawn for years – but the new tidbits of information their clash tossed into the public record. Notably, the anti-CTBT commissioners raised the following argument:
"the treaty remarkably does not define a nuclear test. In practice this allows different interpretations of its prohibitions and asymmetrical restrictions. The strict U.S. interpretation precludes tests that produce nuclear yield. However, other countries with different interpretations could conduct tests with hundreds of tons of nuclear yield – allowing them to develop or advance nuclear capabilities with low-yield, enhanced radiation, and electro-magnetic-pulse. Apparently Russia and possibly China are conducting low yield tests. This is quite serious because Russian and Chinese doctrine highlights tactical nuclear warfighting. With no agreed definition, U.S. relative understanding of these capabilities would fall further behind over time and undermine our capability to deter tactical threats against allies."
In response, the pro-CTBT commissioners argued, in effect, that with regard to the merits of CTBT ratification, it did not really matter whether Russia and China were conducting low-yield tests. "Foreign nuclear programs," they said, "would pose far greater threats to U.S. security without a CTBT than with a CTBT," because "[a]bsent the treaty, other states could develop and test new or improved weapons without constraints." Interestingly, however, the pro-CTBT commissioners never contested the point that Russia ("apparently") and China ("possibly") have been conducting secret nuclear testing – an assertion that, to my considerable surprise, survived declassification review. Rather, they merely argued that the situation would be worse without CTBT.
This, then, was the context for the Obama Administration's inclusion of a test-ban moratorium discussion in the NCR. The administration has made much – rhetorically, at least – of its support for CTBT ratification, and has pledged to reverse the U.S. Senate's 1999 rejection of that treaty. How awkward it must have been, therefore, that the SPRC last year drew attention to the issue of secret Russian and Chinese nuclear testing by raising it for the first time in public.
Keen to promote the merits of CTBT, the Obama Administration apparently thus felt it politically necessary to put at least something reassuring on the record about other countries' compliance with no-testing pledges. Don't worry, it wants the reader to conclude: countries that promise to stop testing really do stop, so we should be comfortable with CTBT.
Yet this new compliance finding, if one can call it that, does not reassure. Indeed, the new section in the NCR actually highlights the problem identified by the anti-CTBT SPRC commissioners. By the Obama Administration's own admission, we don't know what definition of "testing" to apply in analyzing compliance with other powers' no-testing pledges. As a result, we might indeed be able to claim that we have no indication of noncompliance, but this means little because we can't define a violation in the first place.
The new NCR discussion is thus a microcosm of the analytical problems presented by CTBT, for this is precisely the point the anti-CTBT commissioners were making about the CTBT itself: because it lacks a definition of nuclear testing, it opens the door for subjective definitional gamesmanship in pursuing certain types of tests without "violating" the treaty. And if Russia and perhaps China are taking advantage of the non-existence of a definition in their policy moratoria, why would they not game the similarly non-existent definition in CTBT? And if they do, does the 2010 NCR now establish a precedent for finding "no indication" of a CTBT violation even if low-yield nuclear testing continues?
The new moratorium discussion in the Obama NCR is thus rather a mess. It seems to be designed to create the impression that all is well, but it leaves the real questions unanswered and the real issues unaddressed, even while setting a worrying precedent for the future. Its inclusion was not a proud day for the venerable Noncompliance Report.