Iraq's Efforts to Acquire Information about Nuclear Weapons and Nuclear-Related Technologies from the United States

Prepared under DOE purchase order number DE-AP01-99NN50490.000

David Albright and Kevin O'Neill
Institute for Science and International Security

November 12, 1999


The scope of work for the above-referenced project asks the Institute for Science and International Security (ISIS) to provide a detailed, unclassified report of the efforts by the government of Iraq to obtain information about nuclear weapons and fissile materials production technologies, describing what information Iraq currently has of the U.S. nuclear weapons technology, what information the Iraqis are trying to discover, and how the U.S. might better protect U.S. nuclear weapons information. This report was to be solely authored by Khidhir Hamza, a senior Iraqi nuclear scientist who held several high-level positions in Iraq's pre-Gulf War nuclear weapons program, and to reflect his knowledge and views of the issues; the report was never intended to be a comprehensive assessment of these issues.

To fulfill this project, Hamza drafted a report and was interviewed by ISIS staff on three separate occasions. Transcripts of these interview sessions were produced.

We reviewed a draft report submitted by Hamza, and found much relevant information and insights. However, we regret that we found his report to be deficient in several ways. We also found several inconsistencies in the interview transcripts. However, Hamza failed to respond to comments by us on his draft report or to address the transcript inconsistencies. Despite these shortcomings, Hamza's report and the interview transcripts do address many of the questions asked in the project's scope of work.

This memorandum draws out the main observations from Hamza's report and interviews relevant to the project's scope of work. References are made to Hamza's report and the interviews to illustrate these observations.

What were the Iraqi Intelligence Efforts to Gather Information?

Observation: Iraq used a broad range of assets to acquire information to support its nuclear weapons effort.

According to Hamza's report, these efforts can be categorized into three broad areas: Iraqi intelligence efforts run by the Mukhabarat, MIMI intelligence agencies, and Iraqi delegations and envoys from other branches of the government.

Observation: Student networks in the United States, run by the Mukhabarat, provided Iraq with access to libraries at universities and laboratories.

Iraq took advantage of student networks in the United States to obtain documents and open-source literature to support its nuclear weapons program. In his report, Hamza discusses the importance of the student network to obtaining information generated by U.S. laboratories:

Sources of information were most available at universities and in national labs for those who manage to get some of their research carried [out] there. Reports mostly from national labs and research centers are not generally available to third world countries. Large segments in the subscribed [IAEA] INIS system are blocked from the third world, especially U.S. lab reports, even though they are not classified. The U.S.-based Dialog data base system reveals a much larger and useful lists of reports than what arrives through various subscriptions (draft report).

Hamza also discussed how students at university libraries obtained information for the Iraqi nuclear weapons program. He notes in his report that "U.S. university libraries ... contain a much wider selection than available to us. Students usually will photocopy those reports and send them [back to Iraq] or facilitate for a government envoy to gain access and do his own photocopying" (draft report).

At the University of Wisconsin, for example, an Iraqi agent posed as a student in 1984 to gain access to library holdings, where he photocopied more than 300 documents. Hamza said in an interview that "the University [of Wisconsin] had a good, out of the way library that would not have too much security, it was less conspicuous than going to MIT or the Library of Congress, where the security was much tighter."1 The agent, Saleh, "had a friend who was a student at the University of Wisconsin, so he [Saleh] used the student's ID, went to the library and started photocopying all kinds of papers" (May 27, 1999 interview).

According to Hamza:

[Saleh] used the ID of some student to get into the library and no one really checks the photo if you are not borrowing materials. So he was able to copy the materials on the list ... He copied a lot of reports on magnet design and technology, ion sources, diffusion barriers, vacuum technology and leak detectors, how to solve contamination problems, things like that. He brought us something like 300 reports and some papers. They were very useful when we got back (May 27, 1999 interview).

One problem of the student networks, however, is that students did not always return to Iraq. This became especially true in the 1980s during Iraq's war with Iran.

Hamza also reports of "another attempt at gaining more information about centrifuges used for uranium enrichment was made by Dr. Makki at the University of Virginia in Charlottesville." (draft report). He did not elaborate on this assertion in his report, however, nor did he discuss it in the interviews.2

To be sure, not all of the information obtained through the student network was reliable. Hamza reports that the "network achieved many breakthroughs and also a lot of bogus information" (draft report).

Observation: Paid informers and operatives also contributed to Iraq's information gathering effort.

Hamza reports that information was often gathered during the process of acquiring equipment or materials. However, his report elaborates little on these networks, and provides few relevant examples.

Observation: MIMI also operated intelligence gathering networks in the United States.

In his report, Hamza notes that MIMI acquired:

Detailed information and files ...on each technology of interest in the US and elsewhere. Conferences and scientific meetings are a special target to get to know and establish contacts. Embassies trade attaches are usually MIMI operatives and they help in reporting back and establishing contacts and facilitating operations (draft report).

One particular example of MIMI operations was Iraq's participation in the 9th Conference on Detonation in Portland, Oregon. Hamza reports that the August 1989 conference "was targeted as a possible source of contacts, information and supplies" (draft report). Details about the type of information sought by Iraq at the 1989 conference are discussed later in this report.

Observation: Other branches of the Iraqi government were used to gather information, provide cover stories, and provide logistical support.

In his report, Hamza notes that other government activities-including agriculture, oil and other trade activities--were "used when needed to smuggle other equipment and obtain information not available otherwise" (draft report). Other government ministries were also used to avoid suspicions by Western suppliers. According to Hamza:

The [Atomic Energy] Chairman would send me a list and ask what I needed. I would tell him that I need EMIS technology [or that] I need diffusion technology ... He would split the list up and send it to various ministries so that it wouldn't be traceable directly back to Atomic Energy. He would split the companies and not order two or three related items from the same company. That was another technique we used for cover (May 27, 1999 interview).

One particular Iraqi government organization-El Hazen Institute-played a major part in acquiring information, particularly in the early years of the program. According to Hamza, El Hazen "started all of the weapons of mass destruction programs" in Iraq and had excellent contacts at the University of Arizona because of its affiliation with Los Alamos (May 27, 1999 interview). Hamza reports that

El Hazen Institute was a major way of getting information out of the United States. The University of Arizona became a major center for El Hazen and a major contact point. El Hazen sent ten students later to study lasers there, but none returned to Iraq, and stayed in the United States instead ... Responsibility for these students were transferred to Atomic Energy later when El Hazen was closed for a while ... [but] we could not get any of them back (May 27, 1999 interview).

Hamza also notes in his report that "Atomic Energy and MIMI took advantage of Iraqi Airways, which "had the largest cargo hangar of any airline at Kennedy Airport" until the Persian Gulf War. Since Iraqi Airways was often watched, Jordanian airlines were sometimes used, with items then transshipped by land to Iraq.

Observation: Acquiring information was part of a strategy to help keep the Iraqi nuclear weapons program secret.

Developing an indigenous capability to produce nuclear explosives with a minimum of outside assistance was important to Iraq in order to keep its program secret. Regarding the development of high explosive lenses-technology which includes many civil applications--Hamza remarked in one interview that

We could not bring in an expert to tell us what to do--not even for the aspect of developing simple lenses and shock wave studies. We were afraid to bring in an expert because that might expose the program. Iraq had a large nuclear program, and if Iraq were going in the direction of lenses it would signal a bomb program. So were very careful not to get outside expertise, but rather tried to develop the expertise ourselves (May 6, 1999 interview).

During a subsequent interview, Hamza returned to the importance of keeping the Iraqi nuclear program secret. Noting that acquiring lists of documents was a relatively simple task, Hamza remarked that "the problem is how to acquire the items [on the list]. If you acquire it (sic) through the IAEA, you might get asked questions" (May 27, 1999 interview).

Of course, Iraq could not completely avoid interactions with foreign companies who might grow suspicious, and much effort was devoted to developing elaborate "cover stories" to keep the program hidden. However, these cover stories were often as good as the Iraqis' ability to stick to them. During the late-1970s Iraq acquired high voltage ion sources, ostensibly for a civil accelerator, from the IONEX corporation. In reality, Iraq sought the ion sources for its EMIS development work. Before the sources were sent to Iraq, the head of IONEX visited Baghdad in the 1970s:

[IONEX] came to [Iraq] because of the French reactor and because of suspicions raised by the Israelis that Iraq had a nuclear weapons program. He actually came to the atomic energy headquarters in Iraq. Reception called the physics department and told them that there is a guy here from IONEX to check on the order, which was a large order of ion sources and accessories. But the physics department wouldn't talk to him. They were scared.

Later, the same guy saw me in Italy. He was a very vigilant guy and he had held up the order because they would not see him in Iraq. But he saw me in Italy when I was at a conference in Trieste. He came to me, and he asked me what we needed the equipment for. I told him it was for ion implantation. He asked if he could quote me. I told him to go ahead. What I told him was true. Ion sources were used for ion implantation ... There was no EMIS then, but that was to gain experience (May 27, 1999 interview).

Ultimately, Iraq devised an elaborate parallel structure to help it solve problems with its military program through civil applications. In 1989-90, Iraq sought to purchase a Russian magnetic accelerator in order to help it to develop EMIS. Hamza describes how Russian scientists, who would work in Iraq, would solve the Iraqis' problems with EMIS:

[The Russians] were desperate to add as many scientists as possible to come to Iraq to install the equipment and they wanted us to pay their salary in U.S. dollars ... They promised that if I took on more scientists, they would even lower the cost of the accelerator.

QUESTION: Were you worried about security -- about bringing Russian scientists and workers to Iraq?

ANSWER: For accelerators? What is the security problem? They weren't going to [work on] EMIS. They would work at the accelerators and we could ask them anything we wanted about EMIS under the cover of accelerators.

QUESTION: So the EMIS construction would take place in secret by Iraqis?

ANSWER: Yes. The Russians would only work on accelerators, but they would help solve our problems indirectly (May 27, 1999 interview).

Exposing the program often carried high personal risks. For example, Hamza traveled in 1987 to Germany to inquire about purchasing a complete foundry from the firms of Degussa and Leybold that would, in actuality, be the core of Iraq's efforts to make uranium nuclear weapons components.3 Hamza reports that the Iraqi agents directly involved in the negotiations-Saleh and Al Shahiri-risked exposing the program by speaking frankly in Arabic during the negotiations. Hamza, who was overseeing the negotiations nearby, was concerned that the Germans might include an Arab speaker on their negotiating team or make audio recordings of the negotiations, and thereby learn about Iraq's intentions. After the Iraqis returned to Baghdad, Hamza reports that

I jailed Saleh when we went back because of it [ie., he spoke Arabic in front of the Germans] ...because he was too dangerous, he was not under control. And eventually he was transferred out of the program (May 12, 1999 interview).4

What Type of Information was Sought/Acquired by Iraq?

Observation: Iraq sought a broad range of open-source literature to support its nuclear weapons program.

The need to "develop the expertise ourselves" forced Iraq to undertake broad searches for information that might assist its nuclear weapons program. According to Hamza, "what we needed was to build the system from the ground up. We wanted to reinvent the technology for the bomb" (May 12, 1999 interview).

During his interviews, Hamza discussed how long lists of documents were acquired. Iraq complied document lists from the IAEA database, the Science Citation Index, and a document database maintained by Lockheed (note: it has been suggested by a technical reviewer that Hamza is referring to the OSTI database).

In many cases, these documents were obtained through commercial purchases by Iraqi agents. In one interview, Hamza described this process:

We made an arrangement to get us all kinds of literature and books. Shockwaves, explosives, fast phenomena, high temperature and high pressure physics, fast electronics. How do you photograph fast phenomena? Literature about frame and streak cameras that we needed.

We gave Mansour5 specific titles. We compiled a list from the Lockheed database, from the IAEA. We asked him to get back issues from different journals -- explosives journals, metals journals, shockwave phenomena journals--there are plenty of those. Some of them we asked him to get from the first issue, since back issues are very expensive and hard to get in Iraq. The order was for several million dollars. He got most of what we asked for.

Mansour ... established a series of front companies all over Europe. They were the end users of the books, journals and reports that he ordered for us. They would receive them and ship them through Iraqi airways as normal cargo to Baghdad. Iraqi Airways were guaranteed to cooperate, but it wasn't always the safest way because it was watched. So sometimes the shipments were sent through other airlines, such as the Jordanian airline (May 12, 1999 interview).

Hamza elaborated further how he ordered and purchased items at Blackwell's, "the equivalent of Barnes and Noble in the United Kingdom:"

I bought a large number of manuals, books, for various things, mostly ... I couldn't buy many books on explosives. It wasn't anything to brag about. I had a list of what I wanted--about 350-400 items. Some textbooks, some handbooks, some conference proceedings. Blackwell's is a very large bookstore. And Mansour ordered what I couldn't get. I placed orders with various agents and got them later. I also had cash on me and bought a few things by cash. Mansour ordered the rest. (May 12, 1999 interview)

In other cases, Iraqi agents, including Hamza himself, photocopied reports from university or government libraries. Hamza remarked to ISIS that he had visited the U.S. Library of Congress, the MIT library, and others to photocopy materials. However, these efforts were often less productive than purchasing books from a retailer:

[When in the United Kingdom] I visited the British Science Library. But actually it was mainly useless. It was too time consuming and difficult to photocopy some materials, such as books on high explosives. I had to identify myself, and I didn't want to do that. We could find a few things there but we found out that to copy things, you need to order them and wait. It was quite a tedious task and I didn't have the time for it. So I just placed an order with Mansour (May 12, 1999 interview).

Nevertheless, libraries played an important role in helping Iraq to fill in the gaps of its program, particularly as it searched broadly for many types of open documents about a variety of topics. Hamza commented on the importance of U.S. libraries in one interview:

A modern library, a U.S. library, is something we didn't have in the third world. It is a great inconvenience. We were used to researching here [in the United States]. Anything you need you could ask for, even Russian reports. When I was a graduate student here, I wanted a report from Russia. There was a translation here at the Library of Congress. They translated it and sent it to me. You don't have these types of services in Iraq (May 27, 1999 interview).

Observation: Iraq learned to discern good information from bad information-and refined its search for information-as the program matured.

Even though Iraq found some useful information in Western literature, much of what was acquired was unhelpful. Moreover, Hamza reports that "disinformation policy in the US in the eighties confused us considerably on what to use and what information was reliable" (draft report). Lack of knowledge led to many false starts that delayed the program.

However, as Iraq learned more about the properties of uranium and became more familiar with nuclear technologies, it was better able to assess the type of information that was important to the success of the program. For example, Hamza reports that

We didn't know, for example, if the uranium compressions and elasticity parameters given in the literature were correct ... it is not a simple matter. It [uranium] could be made in various phases, alpha phase, beta phase, for example. For each of these there are mechanical properties which differ from the others. We found some numbers in the literature but we didn't believe them. Later on, we found out that the different phases didn't matter much for uranium, because unless you are actually crystallizing uranium, you get a mixture anyway.

When you start the explosives, the shock wave will transfer through various media--the explosive itself, the tamper, the reflector, the plating on the uranium, the uranium--so there is an interface between each media. How does this interface? What are the parameters that you would use, especially for uranium? There are several parameters for uranium in the Huguenot equation which we didn't know (May 12, 1999 interview).

Observation: Proprietary information and training was also useful to Iraq.

Open literature-available at bookstores and libraries-was not the only type of information sought by Iraq. During efforts to procure equipment, Iraq frequently sought to acquire or gain access to proprietary information, such as design, operating and training manuals of the equipment it was seeking. Hamza emphasized in one interview that "buying items was no problem for us. How to install it, maintain it, keep it up-that was the problem" (May 27, 1999 interview). Later in the same interview, Hamza frankly stated that "Iraqi technicians were not that good" (May 27, 1999 interview).

From Degussa and Leybold, for example Iraq sought many types of documentation. Hamza reports that Iraq sought to obtain many types of documents, including "manuals, techniques for casting various metals and types of crucibles, company manuals, training manuals, literature surveys. There was some proprietary documents about the iodine process" (May 12, 1999 interview).

Proprietary information from Matrix Churchill was also needed. According to Hamza:

We had very good machines because we bought Matrix Churchill, which were the best in the world. Many of these machines were programmed with U.S. programs. For example, programs of how to develop certain pieces were formed by programs developed in the United States. Matrix Churchill was also a good supplier of reports and documentation.

QUESTION: What kind of reports?

ANSWER: Some proprietary, some general. Some unknown, unless you know the programming system. We needed to know the special programming system for CNC machines. There are some reports on it that they gave us. Some people developed their own machining programs (May 27, 1999 interview).

Iraq sought proprietary information in other areas. For example, in its EMIS program, Iraq had difficulty generating low pressures. Hamza described how Iraq sought information to overcome this difficulty:

You get it [information] from an accelerator center. These is a lot of development there on high vacuum. Pumps, connections, flanges -- Pulsionics is one of them that were developing flanges. Leak detectors, helium detectors and other types were also developed and we could buy them. Companies were a good supply of reports (May 27, 1999 interview).

Information from companies and high tech institutes was often obtained by sending Iraqi students to the West for training. According to Hamza:

Buying an accelerator would have given us access to any accelerator in the United States. We'd be able to have exchange agreements ... and send people for training ... When you go there, you learn whatever technology you need there, in that center, and then you go back (May 27, 1999 interview).

Later in the same interview, he remarked: "after the IONEX sources were obtained, most of the work was indigenous, but information and reports were needed from the United States. We always used some contacts--people traveling, students, people who came to Iraq on sabbaticals" (May 27, 1999 interview).

Observation: As part of its EMIS program, Iraq targeted Argonne National Laboratory for information and technology.

During one interview, Hamza discussed how Argonne National Laboratory was targeted by Iraq for information and technology related to accelerators in order to support Iraq's EMIS development program. During the mid-1970s, Hamza visited Argonne National Laboratory under the auspices of the U.S. high tech corporation NEC. When asked if his nationality prevented access to Argonne, Hamza replied that "I could only go in under the auspices of some American company ... once you are in, you could talk to anybody" (May 27, 1999 interview).

Hamza states in the same interview that "the visit was useful." He remarked:

The Argonne experience told me many things. First, which types of ion sources would give us the high currency we wanted. Our main problem was high current, because current is equivalent to product. If you double the current, you double the product. We wanted high intensity ion sources and we wanted ion sources that could handle heavy ions like uranium. I talked to several scientists about that and it was general information that you can get at any scientific conference. It was not classified information. I managed to talk at length on a more technical sense of what to look for and what ion sources were helpful. We learned about places to go and where to buy various things (May 27, 1999 interview).

At about the same time as Hamza's visit to Argonne, Iraq was seeking to purchase an accelerator from NEC. The visit to Argonne helped Iraq to "learn which areas of major research were current, so that we could put together a good cover story" (May 27, 1999 interview). In this fashion, Iraq could avoid suspicions about its intentions, particularly as it sought ion sources that could handle heavy elements like uranium:

By the mid 1970s, all the low ion work was done. Nuclear energy levels of the low mass region was covered by then. The 1960's work took care of it, so by the mid-1970s people were switching into the other mass regions--medium and heavy ions--so it was not unusual to do research using ion sources in that region because it was what everybody was doing ...we were buying a simple, basic accelerator and it is not abnormal. Jordan had one. Many countries had accelerators (May 27, 1999 interview).

Ultimately, however, the effort to acquire an accelerator was downgraded because Iraq decided to proceed with the purchase of a French research reactor. According to Hamza, the head of the Iraqi program believed that it was too difficult to successfully manage both the reactor program and the accelerator program simultaneously. However, when Iraq decided to purchase an accelerator from Russia in 1989-1990, Hamza reports that the information gathered from Argonne and U.S. corporations was "helpful" (May 27, 1999 interview).

Observation: Iraq may have obtained some classified U.S. information.

To support its gaseous diffusion program, Hamza states in his report that:

Many ways were used to obtain classified reports on barrier technology. Many bogus reports were brought by Mukhabarat and MIMI intelligence. However a handful of reports that looked useful were also obtained. Some looked at other technologies that looked more complex and demanding and were ignored. Some of these reports were US classified reports or at least stamped that way. (draft report).

However, Hamza's report does not specify how this information was obtained, the type of classification, or more specific details about the information itself.

Observation: Iraq acquired information about sensors to be used in its high explosives program from open-source literature.

Among the information gathered by Iraq was an article by an Israeli scientist about pressure sensors, which Iraq used to inform its high explosive-lens testing program. Hamza said in one interview:

We didn't know what pressure sensors to use. But we got pressure sensor information from the Israelis, who published an extensive paper in The Physical Review on pressure sensors and their calibration. The paper was published in 1983 or 1984, and we became aware of it in 1986. Afterwards, obtaining pressure sensors was just a matter or ordering materials with certain specifications (May 6, 1999 interview).

Another document obtained by Iraq was a U.S. handbook on high explosives that helped Iraq to design a bunker at Al Atheer. The bunker was constructed to enable Iraq to test high explosive lenses. The handbook came to nuclear weapons program from the Iraqi military, and not directly from the United States. Hamza notes that "there was a huge American handbook on explosives that we got from the Iraqi military that had a chapter on lenses ... there were descriptions of high explosives testing bunkers that were important for us when we built our bunker at Al Atheer" (May 6, 1999 interview).

Hamza later reported to ISIS staff that he believed that this handbook was written at the Lawrence Livermore laboratory, however he was unable to identify the title or edition of the handbook. ISIS later obtained what was thought to be the book Hamza was referring to, the book did not contain this information. However, that Iraq was able to reference descriptions of high explosives testing bunkers in U.S.-origin literature is significant.

Observation: The 9th Detonation Conference, held in August-September 1989 in Portland, Oregon, was very helpful to Iraq.

Hamza reports that:

The U.S. conference was an eye opener for the novice Iraqis. Techniques of shock wave generation, measurement and associated technologies were mostly new to them. The openness and papers delivered showed them the types of experimentation and sources of the various equipment they can get. They also were introduced to experts and the labs they worked at. (draft report).

In one interview, Hamza elaborated:

The Portland conference came at the right time. The two guys came to Portland--they were MIMI guys. They got anything that was current in the field about how to get top notch lenses. Where to go? What materials to use? What is a better design? How much pressure to expect? They were also interested in making contact with some explosive suppliers, especially explosive cap suppliers. By that time, they knew what questions to ask and what information to seek out (May 6, 1999 interview).

Elsewhere during the same interview, Hamza remarked that:

For Atomic Energy, testing and experimentation was a whole new technology. There were several areas that we didn't know about ... in developing explosives you need to work through several problems, including which materials to use, the design itself, the tests you are going to conduct, etc. Most of these issues were solved by the American conference. Many papers were delivered there that gave us many indications on which direction to go (May 6, 1999 interview).

During a subsequent interview, Hamza gave an example of how the 1989 Conference helped them to identify suppliers of sensitive technology:

The Portland conference had an advertisement by an explosives company for detonator caps of 25 nanoseconds. So we expected probably, if we had a collection of these, probably jitter would be less then this. So they would be within space [ie., the tolerance sought by Iraq]. Because the advertisement was placed at an open conference, it must have been cleared; the technology could not have been controlled (May 12, 1999 interview).

Observation: Manhattan Project documents and information were useful to Iraq in overcoming technical obstacles, particularly in its EMIS program.

In his draft report, Hamza minimized the importance of Manhattan Project documentation to the EMIS program by stating "the Manhattan project reports on EMIS were generally useful but not detailed enough to solve some of these bottlenecks."

However, Hamza suggested during his interviews that Manhattan Project documents were extremely useful to Iraq's efforts to overcome particular technical obstacles in the EMIS program. For example, Iraqi designers had trouble in adequately grounding EMIS facilities. According to Hamza:

The question was: how do you isolate something working at very high voltage from the grounding without sparking? To address this, we built a cage system, which we also learned about from the Manhattan Project report and other US [Atomic Energy Commission] reports (May 27, 1999 interview).

Elsewhere in the same interview, Hamza elaborated:

We didn't know how to stop it [the sparking]. We looked up the Manhattan Project reports and somebody caught on that they built the grounding nets underneath the building, so you had the grounding right beneath you. When we built Al Tarmiya, that is how we built the grounding system. We built a net underground and put in outlets in the facility so that everything could be grounded (May 27, 1999 interview).

In his report, Hamza concludes that "with the introduction of cages and better shaping of high voltage objects and improved earthing this problem was more or less resolved."

Hamza remarked, however, that Iraq did not completely trust the Manhattan Project data without independent confirmation:

We could not design a bomb based on a number for uranium. We did not believe that the Manhattan Project were so kind as to give us the actual parameters for the various phases of uranium ... when I worked on the cascade, I took a Manhattan Project book by Cohen to calculate the diffusion enrichment cascade. Cohen wrote the book on enrichment cascade calculations. This is the main book published by the Manhattan Project [on this subject]. It is a very professional technical book. But the equations were wrong. We didn't know if it was intentional or if it was a mistake, but the equation was wrong. So we had to derive it ourselves. That made us suspicious about other Manhattan Project documents. So we went about setting up experiments to study those phases mechanically. But we found out later that many of these parameters were correct (May 12, 1999 interview).

Observation: Magnet designs were obtained from abroad.

In one interview, Hamza discussed how Iraq sought computer programs in the 1970s to design magnets for the EMIS program. Many of the programs obtained by Iraq were of U.S.-origin. According to Hamza:

We tried to get some standard U.S. magnet design programs. We found two or three at CERN. These were originally American programs, but CERN used them to design their own magnets. Jaffar brought with him the entire CERN library when he came back to Iraq in 1974. We also sent some people back to CERN and they also brought back some tapes.

Another source of programs was a library at Belfast, in Northern Ireland. There is a library there, which has an international repository of [computer] programs. You can join in and gain access for a couple of thousand dollars a year, so we joined that library--I was the corresponding member ... we [also] purchased a professional package from some companies that designed magnets for CERN. One of them was a Swedish company. I forget the name. The other was German. I don't remember its name, either. According to Jaffar, these two contracts would result in an EMIS magnet, but each, by itself, would not (May 27, 1999 interview).

Iraq had difficulty in finding programs compatible to its computer systems. According to Hamza, "the only problem [with the CERN magnets] is that they were installed on CDC computers in Geneva and we had to install them on IBM. We had to do a lot of changes in the command systems in the program, which we did" (May 27, 1999 interview).

Lessons and Conclusions

Several lessons and conclusions can be drawn from Hamza's draft report and interviews. His information provides a window into how one proliferant state sought information useful to making nuclear weapons. It also highlights the importance of ensuring adequate and realistic controls on sensitive nuclear and nuclear-related information.

Hamza has described how a would-be proliferator will look broadly, both in terms of subjects and locations, for information that will help it build nuclear weapons. He also points out that a proliferant government will seek open, classified, and proprietary information in advanced industrialized countries. In its collection of information, a proliferant can be expected to utilize many methods. Iraq used overseas students, embassy employees, and visiting scientists to acquire desired information. Iraq also went to extraordinary lengths to disguise its activities, even when it was just seeking books and reports available in libraries or major bookstores. Although Iraq's open search for information should not by itself raise concern, the fact that this search was part of a planned effort to gain a wide range of sensitive information should motivate more action. The United States and other industrialized countries need to better protect sensitive information and detect would-be proliferants who are seeking it out.

There are many reasons for Iraq's or another proliferant's search for sensitive information. Hamza identified several: the need to build an indigenous capability to design and manufacture nuclear weapons; a requirement to avoid reliance on outside experts; the need to gain a basic, rudimentary knowledge of how to handle and process nuclear materials and high explosives; help in devising credible "cover stories" for the program; and the desire to obtain leads on potential suppliers of information and other items needed for nuclear weapons.

The reason to look in many countries for information made sense because of the differences in laws and regulations of information supplier countries. Classification rules and their enforcement vary enormously from country to country. There are cases of one country undercutting another as it declassifies or releases information to the public. In addition, research results may be published openly in one country even if the results are considered classified in another. Within a country, a laboratory or a company can have significantly different attitudes to obeying laws or regulations or maintaining a "culture" of information protection.

Although Hamza provides no concrete case where he was involved in obtaining classified U.S. documents, he does point out that "rubbing shoulders" with those with access to sensitive information can potentially reap important rewards. Establishing unclassified discussions or cooperative efforts with suppliers or researchers can in some cases yield classified or export-controlled information or at least important insights into sensitive information. Although the revelation of sensitive information is rarely intentional, it can be very difficult to prevent in these circumstances, unless the interchange is carefully regulated. With regard to export-controlled information, most countries have significant loopholes in their laws and regulations concerning the control of information obtained by a foreign person or agent within the supplier country. Information obtained by a foreign national while at a research laboratory or a company is too often viewed as outside the scope of export controls, even when it is known that the person will return to his own country.

Hamza has demonstrated that Iraq was not very efficient in collecting information overseas. This inefficiency was partly off-set, however, by Iraq's willingness to allocate significant resources to its efforts and its long-term commitment to information gathering.

Moreover, we believe that Hamza's story is not the "last word" on Iraq's efforts to acquire nuclear weapons-related information from the United States or other western countries. According to a technical specialist who reviewed Hamza's report and transcripts, Hamza did not discuss many avenues of acquiring information that were available to Iraq in the 1980s. As a result, documents detailing additional technologies with nuclear weapons-related applications were likely sought and obtained by Iraq's nuclear program.

Recommendation: The barn door cannot be closed, but there is no reason to swing it wide open

Hamza himself concludes that it would be counterproductive to take a generally "restrictive approach" to protecting all information (draft report). Once information has been declassified or is otherwise in the public domain, it can spread rapidly. Advances in information technology and communication systems now enable an individual to access data electronically from virtually any place in the world. Although special cases may warrant trying to reclassify mistakenly released information, efforts to try to control the spread of open source information is virtually impossible.

In general, Hamza recommends that a more effective approach would include the development of a broader "understanding of the bottlenecks of the would-be proliferator and acting accordingly." (draft report). As potential suppliers of critical information, whether in government or industry, become more knowledgeable about what these bottlenecks are, they can better protect information that would significantly aid efforts to build nuclear weapons. At least, the suppliers could alert authorities that such information was being sought, if a particular request aroused their suspicions. As has been learned in the United States, such an approach would need to be combined with clear laws and regulations and an educational campaign targeted at potential suppliers.

Hamza points out the importance of acquiring sensitive information from companies. This information could be classified or controlled under national laws. Although an approach focusing on bottlenecks is helpful here, more needs to be done. Export control laws must clearly state that the provision of certain types of information to foreign nationals via conversations or meetings at a company is also controlled. Few countries are developing such controls.

Government declassification offices also need to understand better what are the main potential bottlenecks in getting nuclear weapons by proliferant states. More importantly, they need to be adequately informed on a classified level about the methods and activities used by proliferant states to acquire nuclear weapons. Armed with this knowledge, better decisions can be made about what information to declassify, particularly when under pressure to declassify something for commercial or scientific use.

Hamza's observations and statements reinforce the view that countries must ensure that "high walls" surround truly important information. Because pressures to declassify information will continue, the goal should be to determine the key information that requires classification and to punish violators more severely.

Recommendation: International Information Guidelines

We also recommend that more be done internationally to prevent inadvertent or misguided declassification. Iraq sought information from many countries, including the UK, Germany, France, and Russia, looking for clues of how to succeed in making nuclear weapons. As a result, there is a need for the international community to ensure that sensitive information is not inadvertently declassified. One helpful goal is an international agreement on information guidelines that define the specific information that requires protection or continued classification.

These guidelines would help prevent governments from inadvertently releasing sensitive information by creating a uniform list of protected information. Guidelines would also bolster the norm against the spreading of proliferation-sensitive information, reinforcing efforts to control certain types of technology or technical assistance.


Notes

1 Using an "out of the way ... less conspicuous library" would reduce the chances of exposing the program to U.S. government officials. Back

2 Hamza learned of this attempt from David Albright. Back

3 In the fall of 1987, Iraq discussed with these companies the procurement of a $200 million foundry that would be able to melt, purify, cast, and machine refractory metals. Iraq specifically mentioned tungsten, but it actually planned that the foundry would be used to process uranium into nuclear weapon components. The Iraqis planned to locate the foundry at the Al Atheer site, which was to be its secret site to make nuclear weapons. According to a trip report filed by Khidhir Hamza, then the head of Iraq's nuclear weaponization program, that was later turned over to the IAEA Action Team by Iraq in 1995:

During the negotiations, we suddenly realized that the companies were willing to sell these technologies to us. At the same time, they warned us they were complicated, expressed astonishment that we were entering such a field and indicated that the cost would be high. At first, this move on the part of the companies was puzzling for us, since it seemed the companies were prepared even to disregard the requirement for an export license by making special arrangements and packing the equipment under covers, which made the export process seem natural. They even indicated that they knew the equipment was not for peaceful purposes.

In the end, Iraq did not accept the foundry offered by these companies. It feared that the "turn-key" job would have allowed the Germans to learn the location and true purpose of the Al Atheer facility After 1989, however, Iraq used this offer as a guideline to procure equipment for a foundry piecemeal from several countries. It also used the design of the foundry supplied by the Germans to design its uranium processing building at Al Atheer. Back

4 Disciplinary actions taken against Al Shahiri-if any-were not discussed during the interviews. Back

5 Anees Mansour Wadi was co-director of Meed International, based in London. He was a principal contact and link in Iraq's military procurement network. Back