Heavy Water Cheaters

Foreign Policy
Winter 1987-1988, p. 100-119

Civilian nuclear exports are founded on two assumptions: No country will export a crucial item without requiring a pledge of peaceful use or use an item imported under such a pledge to make bombs. The same nuclear reaction that makes electricity makes weapons; the importer’s pledge is the only barrier between the two. If countries receiving nuclear imports could freely convert them to arms, the nuclear exporting countries simply would be spreading atomic bomb factories across the world. The human race’s prospects for controlling nuclear weapons would fall rapidly, and so would its chance of survival. To prevent proliferation, the supplier countries routinely require two guarantees—a pledge of peaceful use and inspection of exported material, equipment, and technology. The International Atomic Energy Agency (IAEA), which is based in Vienna and comprises more than 100 member states, conducts the inspections.

It now appears—for the first time in history–that a country has broken the peaceful-use pledge. It also appears that a second country may have broken it, that a third is threatening to break it, and that the civilian exports of a fourth—possibly made without the pledge—have gone freely into bombs. The culprits are, respectively, Israel, France, India, and Norway. Israel has been making plutonium with Norwegian heavy water for more than 20 years and, according to recent evidence, putting the plutonium into bombs; India has been making plutonium with U.S. heavy water for about the same period and is threatening to put that plutonium into bombs. Norway has exported heavy water that France has used to build a thermonuclear arsenal. If these countries can undermine the pledge and avoid the consequences, there is no reason why others will not do the same.

Few people know what heavy water is; yet it has been crucial to the spread of nuclear arms. Nazi Germany pinned its hopes on heavy water from Norway and inspired the United States to launch its own nuclear weapons effort—the Manhattan Project—for fear that Germany would make the bomb first. Heavy water enabled France to begin nuclear research and was used to make the first nuclear explosive material France obtained. The same is true of India. Because of heavy water, France, India, and Israel all have been able to make nuclear explosives free of international control. In addition, the People’s Republic of China, the Soviet Union, and the United States have built atomic bombs with heavy water.

Heavy water looks like ordinary water but is rich in deuterium, a heavy isotope of hydrogen. Heavy water’s outstanding advantage is that it slows down or “moderates” neutrons passing through it. At lower speeds the neutrons can fission the unstable uranium isotope U-235. Each fission releases energy and more neutrons, causing a chain reaction. Heavy water moderates neutrons without absorbing many of them. This feature is essential for reactions with natural uranium, which contains only .7 per cent of the U-235. Heavy water must be extracted from ordinary water—in which it occurs in minute concentrations—through an expensive process requiring large amounts of energy.

Ordinary water also can be used to achieve a chain reaction, but its higher absorption of neutrons means that the uranium must be enriched. In practice, the percentage of U-235 atoms in the uranium must be increased to about 3 per cent. The enrichment process is technically difficult, very expensive, and requires enormous amounts of electricity. It is even more difficult and expensive than producing heavy water. So heavy water is the cheaper way to the bomb. A country needs only a reactor and natural uranium, which is plentiful, as reactor fuel. Plutonium is created in the uranium once the chain reaction is achieved. Every country that has made the bomb, except China, used plutonium for its first explosion.

The Israeli nuclear effort began in earnest in 1957. Israel’s alliance with France produced what essentially was a joint venture in nuclear development that gave Israel three of the five ingredients it needed for nuclear weapons: a small research reactor, a factory to extract plutonium from the reactor’s spent fuel, and nuclear weapon design information. The reactor and extraction plant could produce about 8 kilograms of plutonium per year, enough for two modern fission bombs equivalent to the one dropped on Nagasaki in 1945. France built the reactor and extraction plant secretly in the Negev desert, at a site called Dimona. There is no doubt that France meant to allow Israel to make nuclear weapons. Francis Perrin, high commissioner of the French atomic energy agency from 1951 to 1970, so admitted in an interview published on October 12, 1986, in the Sunday Times of London. Perrin said that after World War II, the United States agreed to allow the French scientists who had worked on the Manhattan Project to apply their knowledge at home, provided they kept it secret. “We considered we could give the secrets to Israel provided they kept it to themselves,” Perrin recalled in the interview. He confirmed the construction of the reactor and the extraction plant, and the cooperation on weapons design. The transactions and their related documents are presented in detail in La deux bombes (1982), a book by the French journalist Pierre Pean.

But Israel still needed two more ingredients: natural uranium and heavy water. The uranium was not difficult to obtain. The earliest supplies apparently came from Argentina, South Africa, Israel’s own phosphate deposits, and French-controlled mines in Africa. For subsequent sources, there was the “plumbat affair.”[1] In 1968 Israel skirted nuclear export controls by diverting 200 metric tons (t) of Belgian uranium in barrels marked Plumbat–the Latin word for lead—on the high seas.

That left only heavy water. It proved to be the only ingredient that Israel could not get without breaking its word. In the early 1960s only the United States and Norway had any heavy water for export, and both restricted exports to peaceful use and opposed the spread of nuclear arms. Israel itself had made small amounts with an indigenous process bur could not make anything near what was needed for Dimona. So Israel had no choice: It had to import U.S. or Norwegian heavy water.

From 1959 to 1963 Israel imported 20t from Norway and 3.9t from the United States. This would supply Dimona indefinitely if the reactor stayed at its rated power of 24 megawatts. Israel pledged to restrict the heavy water to peaceful use—the plutonium made could not be put into bombs—and to allow on-site inspection so the suppliers could ensure that the pledge was kept. However, Israel seems to have taken the water and ignored the promises. It must have broken its pledge to Norway and may have to the United States.

The evidence of Israel’s intentions began to emerge in the 1970s. In a September 4, 1974, CIA memorandum, Prospects for Further Proliferation of Nuclear Weapons, the agency concluded, “We believe that Israel already has produced nuclear weapons.” And then CIA Deputy Director for Science and Technology Carl Duckett told the Nuclear Regulatory Commission in a February 1976 briefing that Israel was making bombs from Dimona’s plutonium. He also disclosed that the CIA had told President Lyndon Johnson as early as 1968 that Israel had produced bombs. French sources, including Perrin, also have reported that Israel has built bombs.

During the 1973 Arab-Israeli war, Israel reportedly assembled 13 bombs and readied them for use.[2] In 1974 then Israeli President Ephraim Katzir said that “it has always been our intention to develop a nuclear potential…. We now have that potential.” And in May 1985 a Los Angeles businessman, Richard Smyth, was indicted for smuggling to Israel 810 high-speed electronic switches, called krytons, which are used as nuclear weapon detonators. Smyth vanished a week before he was to appear for trial. Moreover, Aerospace Daily reported in its May 1, 1985, issue that Israel had deployed Jericho II missiles, with a range of 400 miles, carrying nuclear warheads. In addition, the NBC telecast of “Nightly News” on July 30, 1985, reported that Israel possessed at least 100 nuclear weapons.

Any doubts left by these accounts should be dispelled by the case of Mordechai Vanunu, an Israeli arms technician who worked at Dimona for 8 years. In September 1986 he detailed Israel’s production techniques to the Sunday Times, providing almost 60 color photographs of what he said was Israel’s underground bomb factory and convincing weapons experts on both sides of the Atlantic Ocean that he was telling the truth. According to Vanunu’s data, Israel now has 100 to 200 advanced fission bombs, has mastered a thermonuclear design, and appears to have a number of thermonuclear bombs ready for use. Instead of 8 kilograms of plutonium per year—what a 24-megawatt reactor would normally make—Dimona, according to Vanunu, is producing 40 kilograms per year, which is enough for 10 advanced fission bombs. Dimona is also making tritium and deuterium, the thermonuclear bomb ingredients. After telling his story to the Sunday Times, Vanunu disappeared on September 30. He turned up 2 weeks later in an Israeli jail charged with espionage and other crimes. Vanunu said he was kidnaped in Rome, though Israel refuses to say how he was arrested.

Vanunu’s story is either a spectacular security leak or the political equivalent of a nuclear test. If the story is a plant, Israel could be telling its neighbors once again that the rumors about its bombs are true, as it did after it destroyed Iraq’s Osirak reactor in June 1981, when former Defense Minister and Foreign Minister Moshe Dayan told the New York Times, “We do have the capacity to produce nuclear weapons, and if the Arabs are willing to introduce nuclear weapons into the Middle East, then Israel should not be too late in having nuclear weapons, too.”[3] In short, there is not much doubt that Israel has made nuclear weapons. The remaining question is whether Israel used plutonium made with Norwegian or U.S. heavy water.

To understand the transfer of heavy water to Israel, several points must be kept in mind:

  • The reactor at Dimona is Israel’s only means of making plutonium, and plutonium is Israel’s primary nuclear weapon material. Dimona also is the only facility in Israel that uses heavy water in metric ton quantities. When Dimona opened in 1963, the world’s main heavy water suppliers were the United States and Norway. Israel was producing heavy water only in laboratory quantities. Therefore, it was physically impossible to start Dimona without U.S. or Norwegian heavy water.
  • France and Canada also possessed heavy water in the early 1960s, but Canada has never exported any heavy water to Israel, and all of France’s heavy water had been imported previously from the United States and Norway. France pledged not to re-export it without permission. Thus even if Israel had received a secret shipment from France, the water would have been diverted illegally from U.S. or Norwegian stocks, giving Israel no right to use it.
  • Therefore, all the heavy water Israel imported for Dimona must have come initially from the United States and Norway.
  • Reactors like the one at Dimona lose only about .5 per cent of their heavy water each year. So if U.S. or Norwegian heavy water went to Dimona in 1963, more than 85 per cent of it is still there today.

These facts lead to an irresistible conclusion about Norway’s heavy water: It must have gone into Dimona in 1963, must still be there, and must still be making bombs. Norway was the only conceivable source for Dimona’s original charge of heavy water, and only a small amount of it could have been lost. No other possibility exists, given Dimona’s starting date and loss rate. The use of the U.S. heavy water is less clear, but the evidence indicates that the American water may have been shipped to Dimona as well.

The United States sent 3 .9t directly to Israel in 1963, 3 years after Dimona was discovered by a U.S. spy plane.[4] Why the water was sent remains a mystery. According to [AEA officials, Israel pledged to restrict it to peaceful use and to place it under international inspection. The officials say that the heavy water is still in Israel and still being safeguarded by the IAEA. But these same officials admit that the IAEA; did virtually no inspection of heavy water until the late 1970s. So only Israel knows where the heavy water was for the intervening 17 or so years that followed its export.

The earliest U.S. records say that it went to Dimona. In 1977, when the first version of the International Nuclear Facility Codes was published, Dimona was listed as “RTSG.” U.S. records show that the heavy water was shipped to that facility. The records also show the heavy water as still being at facility RTSG in 1972, when the water’s ownership was shifted to Israel through a bookkeeping transaction. In 1979, however, a new version of the facility codes appeared. It still showed Dimona’s code as RTSG, but also assigned Dimona a second code, “RTSL.” Having two codes contradicted the usual practice of keeping the same code for the same facility. Currently, the codes show RTSG designating a facility in Haifa called Technion and RTSL designating Dimona. The records do not show whether the heavy water was moved. Even if the heavy water is now at Technion, it seems to have gone originally to Dimona. The IAEA’s lax safeguards agreement with Israel does not cover Technion or any of the other locations where the export records show the water was kept.

There is another reported source of Israel’s heavy water, but it is not so well established. Les dens bombes contends that France sent Norwegian heavy water to Dimona secretly in 1960. According to Pean, the shipment came from France’s atomic site at Saclay, outside Paris. A high official of the French atomic energy agency went to the atomic research center at Saclay, picked up a few metric tons of heavy water in barrels, hauled them back to his Paris office in three small trucks, and unloaded the barrels in the underground garage. Moments later, agents of the Israeli secret service, the Mossad, came in, loaded them on another truck, drove them to the military zone at Le Bourget Airport, and put them on a French plane to Israel that landed in Sicily on the way. Did all this really happen?

Unnamed French sources familiar with the Dimona export have described Pean’s account as “99 per cent correct.”[5] France imported 20t of U.S. heavy water between 1956 and 1959, all earmarked for Saclay’s EL-3 (for eau lourde, or “heavy water”) reactor. Since EL-3 needed only 4.1t to start up and lost very little in operation, France had U.S. heavy water left over. In 1960 France imported an additional 16.5t for Saclay, earmarked for a facility called Aquilon. So by 1960 France had imported a total of 36.5t from the United States. This exceeded the needs of the French reactors and would have enabled France to supply Dimona. France was importing Norwegian heavy water at the same time. Reliable Norwegian sources report that France obtained more than 140t from 1950 to 1965.[6]

Pean’s account is the only known evidence that France broke its word. U.S. records do not show any retransfer of U.S. water from France, and government sources in Norway say that records there do not reveal any retransfers either. Still, if France were willing to build Israel a secret reactor and a secret plant to extract the plutonium, it would have been a small matter for France to transfer secretly heavy water. Indeed, it seems implausible to build a heavy water reactor with no heavy water to run it. Heavy water reactors are invariably sold “wet”—with the heavy water included. The 20t Norwegian shipment to Israel was approved only months before it was sent in 1959 and could not have been counted on when the reactor was built. Thus France very likely agreed to supply Dimona’s heavy water along with the reactor.

The Norwegian water, in fact, went to Great Britain before it reached Israel. Britain had purchased the water but decided not to use it. Influential supporters of Israel in Norway—including a former defense minister—set up a corporation called Noratom to buy it back. Then they quietly pushed an export license through the Norwegian foreign ministry. The license allowed Noratom to ship the unused water directly from Britain to Israel, which it did.

The last question about Dimona is its size. Dimona is widely assumed to be both cooled and moderated by heavy water. The heavy water also removes the heat caused by the chain reaction. Under this assumption, Dimona needed about 18t of heavy water to start operation. The 18t can be derived from the scale of India’s new Dhruva reactor, which is cooled and moderated by heavy water (it uses three-fourths of a t per megawatt) or from the scale of India’s older Cirus reactor, which was only moderated by heavy water but which would have used three-fourths of a t per megawatt if it also had been cooled by heavy water.

The figure of 18t, however, assumes that Dimona remained at its original power. Evidently it did not. For the reactor to produce the 40 kilograms of plutonium per year described by Vanunu, it would have had to be scaled up to more than 100 megawatts. The reactor would need more heavy water and a larger cooling circuit. The plutonium extraction plant would have had to be expanded, too. According to Pean, the Israelis convinced their highly placed French supporters to boost Dimona’s size when it was built, quietly making its key components three times as big as called for in the Israeli-French agreement.

These accounts could explain why Israel imported so much heavy water. For Dimona to produce more than its rated power, Israel would need more than 18t. Israel had imported 20t from Norway and 3.9t from the United States, 33 per cent more than the original 18. If France supplied enough to start the reactor, which seems logical, the “few” t Pean reports could have been any amount up to 18. How much plutonium could this heavy water make? Plutonium production depends upon reactor power, and power can be increased by changing the fuel design. Simply by using one-half- rather than one-inch fuel rods, reactor power is increased by a factor of four. But the power increase releases heat, which must be removed by the cooling system, which in turn is limited by the supply of heavy water.

Dimona uses twice as much moderator as coolant, so multiplying its cooling capacity would not mean multiplying its need for heavy water by the same amount. If the amount of coolant were quadrupled, which could allow quadrupled power, Dimona would need about 36t of heavy water—12t of moderator and 24t of coolant. The 36t is slightly less than the total that Israel could have received from Norway, the United States, and France. Thus Vanunu’s reports of plutonium production are consistent with Pean’s reports of additional heavy water exports.

France’s Stunning News

On May 22, 1980, the French national assembly received a remarkable document from the commission on national defense and the armed forces. The commission reported that France was making tritium in two reactors called Celestin I and Celestin II. According to the French government’s nuclear agency, Cogema, Celestin I went critical—that is, achieved a chain reaction—on May 15, 1967, and Celestin II on October 30, 1968. Cogema noted that the reactors were making tritium “used for thermonuclear armaments.” Tritium produces fusion in thermonuclear bombs and is bled into the core of fission bombs to boost or vary their yields. The report disclosed that the reactors were cooled and moderated by heavy water. That was stunning news because when they were started, France had no heavy water of its own to put into them.

France did not produce significant amounts of heavy water until December 1967, when it opened a small production plant at Mazingarbe. The plant’s annual capacity was limited to 20t–25t by the supply of feed gas from an ammonia plant. It operated until January 30, 1972, when it exploded, and reached 90 per cent of capacity only in its second year. Therefore, it could not have produced more than 80t of heavy water in its lifetime. Because Mazingarbe started in December 1967, it could not have furnished the original charge of heavy water for Celestin I in May 1967. Nor could its first 10 months of production, even at full capacity, have furnished Celestin II in October 1968. Both reactors reached full power—and needed their full inventory of heavy water—immediately after going critical. Indeed, Mazingarbe’s entire lifetime production was not sufficient for Celestin even if, by magic, Mazingarbe’s water could have gone into the re-actors before it was made. A March 15, 1969, article in the French daily Le Monde affirmed that Mazingarbe was meant to furnish only “a part” of Celestin’s needs. Therefore, the Celestin reactors could not have been started with French heavy water. They had to use an import, and the only sources were the United States and Norway.

Each Celestin reactor required from 60t to 80t of heavy water to produce its 200 megawatts of power. This estimate is derived from the heavy water requirement of France’s EL-3 reactor, which used about .3t per thermal megawatt, and of its EL-4 reactor, which used about .4t per thermal megawatt. The EL-3 and EL-4 reactors use enriched uranium fuel, as does Celestin, which reduces the need for heavy water. Because the reactors operate at low temperature and pressure, they do not lose much heavy water from leaks—surely less than the 1 per cent per year normally lost by power reactors. That means that more than 80 per cent of the heavy water originally put into the Celestin reactors still should be there, still making bombs.

France had imported more than 140t from Norway by the time the Celestin reactors started. Norway got neither inspection rights nor, ac-cording to Per Paust, a spokesman for the Norwegian foreign ministry, an overall pledge of peaceful use. Thus Norway committed a deliberate act of proliferation.

This was a departure from Norway’s stated policy of always requiring peaceful-use guarantees for its heavy water exports. Paust said that instead, France promised not to re-export the heavy water without Norwegian permission and gave Norway a string of certificates stating the “end use” of each shipment. Norway now says that it can restrict France to the uses listed in the certificates. However, as this article went to press, Norway declined to reveal what the certificates require. Do they prohibit use in building nuclear weapons? France, too, when asked in September 1987 about the lack of French heavy water to run the Celestin reactors, refused to comment. Thus the questions remain, Did Norway deliberately help France make H-bombs, or did France break a series of promises that Norway has not bothered to enforce?

France needed at most 150t of heavy water for its civilian reactors up to the time the Celestin reactors started. It needed 4.7t for EL-I in 1948, 6t for EL-2 in 1952, not more than 17t for Aquilon in 1956, 4.1 t for EL-3 in 1957, and not more than 115t for EL-4 in 1965—a total of not more than 147t. If 1 per cent is added for losses, which surely overestimates actual losses, the grand total is just under 150t. But France, it turns out, imported more than that for these same reactors from the United States.

France imported 168t from the United States from 1956 to 1967. The water was covered by pledges of peaceful use, no re-export, and inspection by Euratom, the European Common Market nuclear agency. It was earmarked for the EL-3 reactor, the Aquilon critical assembly, and EL-4, for which the United States furnished the entire inventory. France already had imported from Norway the small quantities needed to run EL-1 and EI-2. Thus the U.S. imports alone were enough to fill and run every civilian reactor in France. In all, France imported more than 308t from the United States and Norway, although it did not need more than 150t for its civilian reactors. This left idle by 1967 about 158t of expensive material—imported for no apparent reason. In fact, however, it had to supply the 120t–160t that the Celestin reactors needed in 1967 and 1968.

The conclusion has to be that France imported about twice the heavy water it needed for its civilian program and used the surplus to make bombs. Tritium from Celestin was reportedly destined for France’s first hydrogen bombs—tested in 1968—and its first tritium-boosted missile warhead. These warheads were the first to be installed on French submarines and had a 500-kiloton yield, enough to destroy a city.

In the future, Celestin’s tritium could prove even more important. France has been considering the manufacture of neutron bombs. Celestin’s tritium would be needed to build the warheads and, because the half-life of tritium is only 12 years, to keep them ready for use. Such warheads could help France change the nuclear equation in Europe.

From France and Norway heavy water technology spread to India. Mazingarbe employed a novel process that promised great economy for middle-sized plants—those making 100t or less of heavy water annually. In 1969 and 1971 India ordered two plants similar to Mazingarbe from its builders, a French-Swiss consortium called GEL-PRA. The first was built at Baroda and the second at Tuticorin. Norway sent Jomar Brun, the chief engineer of Norsk Hydro’s heavy water plant, to offer technical assistance. France reportedly refined the uranium for India’s first heavy water reactor, from which India made its first plutonium. The plants, the uranium, and Brun’s advice were not restricted in any way, and India made no pledge of peaceful use or international inspection. Everything received was free for bombs.

India has not missed the opportunity. Heavy water from Baroda and Tuticorin is now going into three new Indian reactors that are free of controls. The reactors are giving India—for the first time—plutonium that can legally go into weapons.[7] The new Indian stockpile is increasing the pressure on neighboring Pakistan to go nuclear and is making that country’s program more difficult to control.

Norsk Hydro also shipped heavy water and deuterium gas, which is used in thermonuclear bombs, to South Africa for research in 1961; applied for permission, which the government refused, to ship heavy water to India in 1976; and inquired how the government would view a heavy water export to Pakistan in 1974 (the government said no). For a 3-year period during the 1960s, when Noratom was advertising plutonium processing devices, it sent Israel laboratory equipment costing then about $215,000.

The 1968 Treaty on the Nonproliferation of Nuclear Weapons, as well as every other effort to combat proliferation since the 1960s, assumes that peaceful-use and inspection pledges will be kept. Nonproliferation efforts make no sense otherwise. The supplier countries’ promises not to export without controls are meaningless if the controls are not enforced. To retain any credibility for their nuclear export policies, Norway and the United States must enforce their rights in Israel, France, and India.

The rights are plain. Norway’s foreign ministry says that Israel made three pledges when it imported the 20t in 1959: to restrict the water to peaceful use, to retransfer it only with permission, and to allow inspection of any facility where it was used. These pledges would cover any plutonium made by a reactor using the heavy water. The agreement was secret when made, but Norway released the details in October 1986. Israel imported an additional t in 1970 under the same restrictions. After long silence, Norway asked Israel in February 1987 to let the IAEA inspect the heavy water. The request was compelled by questions from Norway’s parliament and articles in the press showing that Norway had not enforced its rights. It had inspected the water only once, in 1961, 2 years before Dimona started.

In an April 1987 response, Israel did not deny receiving the water, making the pledges, or, apparently, using the water. Instead, Israel said that it did not trust trust the [AEA to be objective.[8] This excuse is ridiculous. The IAEA has inspected U.S. heavy water in Israel since the 1980s without objection. If the [AEA. can inspect U.S. heavy water in Israel without bias, it can do the same for Norway’s water. The IIAEA’s objectivity does not depend upon the heavy water it is inspecting.

Norway finally made a formal demand for international inspection on September 30, 1987, and Israel finally refused. Norway then announced that it would ask to do its own inspection, a right Norway clearly has. In such an inspection Israel will have to show Norway that the heavy water has not been used, or has been used for peaceful purposes. To demonstrate the former, Israel must produce 21t of heavy water that have never entered a reactor. Otherwise, it must show about 18.5t (allowing for operating losses) of heavy water that are in a reactor or have been in one. It then must put the reactor’s plutonium under inspection to honor the peaceful-use pledge. Fortunately, it is easy to tell if heavy water has passed through a reactor. Irradiation forms tritium in heavy water. Even after “detritiation,” irradiated heavy water contains about 1,000 times more tritium than heavy water that has not been used.

If Israel refuses all inspection, the 1959 agreement states that Norway can take back its heavy water. Retrieving the more than 85 per cent of Norway’s water that must still be in Dimona probably would shut down the reactor and sever Israel’s main supply of nuclear weapon material. The United States holds the same inspection rights as Norway. To ensure that Israel produces the full 24.9t for testing, the United States should join Norway’s inspection request and have both countries’ water tested at the same time.

Norway and the United States also should enforce their rights in France. Norway should ask France for assurance that the no-re-export pledge has been kept, and, if necessary, for the data to back up the assurance. Norway also should reveal what France promised in the end-use certificates. If France pledged peaceful use, Norway should ask for records that this pledge, too, was honored.

The United States is entitled to peaceful use of all the plutonium made by the EL-3 and EL-4 reactors and to have the plutonium inspected. EL-3 yielded about 30 bombs’ worth of plutonium from its opening in 1958 to its shutdown in 1982-84. EL-4 made more than 200 bombs’ worth from 1967 to 1985. The United States is also entitled to have segregated and restricted for peaceful use the roughly 130t of heavy water removed from these decommissioned reactors. That France is a declared nuclear weapons state does not change its obligations. The United States cannot ignore peaceful-use pledges by bomb-making countries and expect other states to keep the pledge. Such a policy would reward proliferation.

The United States also carries rights in India that, if pursued, could call India to account for the plutonium it has made with U.S.-supplied heavy water. That plutonium remains India’s largest single source of nuclear weapon material.

Selective Proliferation

If the United States and Norway do nothing, the lesson will be clear. They will be operating under a policy of selective proliferation—winking at proliferation by friends, or even aiding it, while denouncing it by others. Doing nothing also will mean that Washington cannot deal with proliferation in its own back yard, because France, Israel, and Norway are U.S. allies. If the United States cannot stop proliferation by them, how will it stop proliferation in South Asia or South America? Finally, doing nothing will mean that nuclear export controls really can be ignored. Israel, for all appearances, is violating a peaceful-use guarantee to Norway, and may have broken one to the United States. France appears to be making bombs with a “civilian” import and may have broken the promises of no re-export and peaceful use.

The stakes in Israel and in France are not the same. France is a world power at the center of the Atlantic alliance. Israel is a small country surrounded by hostile neighbors. Israel can escalate any conflict in the volatile Middle East to the nuclear level. The United States—as Israel’s backer—and the Soviet Union—as a backer of Israel’s rivals—could be pulled into the fight. Armed with nuclear warheads, Israel’s aircraft could destroy most of North Africa and all of the Middle East and could threaten the Soviet Union. The July 1987 issue of International Defense Review reported that in a May 1987 test over the Mediterranean Sea, Israel’s Jericho II missile flew 510 miles. It probably will eventually attain its design range of 900 miles, enough to reach the Soviet border. Not surprisingly, the Soviets strongly protested the flight.

Would Israel, its back to the sea and its existence at stake, be saved by nuclear deterrence? The Soviets inevitably would come to the aid of Syria or another client state threatened with nuclear weapons. Soviet support would have little credibility otherwise. The United States then would have to come to Israel’s aid because Israel could not confront the Soviets alone. The French foresaw in 1959 that the Israeli bomb was meant . to compel U.S. support in a crisis. Perrin, quoted in the October 12, 1986, Sunday Times, put it bluntly: “We thought the Israeli bomb was aimed at the Americans, not to launch it against America but to say `if you don’t want to help us in a critical situation we will require you to help us, otherwise we will use our nuclear bombs.’ ”

Policymakers should be asking whether it is in America’s interests to guarantee Israel’s security in such an indirect and dangerous way. Is the United States ready to give Israel the same power as Britain and France to drag America into war? What are Israel’s responsibilities in return? Can Israel be trusted in a crisis not to push things over the edge? The same questions apply to Pakistan. Should the American people be hostage to the decisions of a succession of allies who obtain nuclear weapons? In July 1986 the Soviet Union, at India’s request, gave Pakistan a stiff warning to cool off its nuclear weapons effort. The Reagan administration replied that Pakistan was a U.S. ally and would not be bullied. So the prize for a U.S. ally that acquires nuclear weapons exceeds the weapons themselves: It is a promise of U.S. support.

For their part, the Soviets have handled this problem better. Only Moscow controls the East bloc’s nuclear weapons. The United States is not facing a research reactor like Israel’s in Cuba or Nicaragua. Through their irresponsible exports, France and Norway have injected a nuclear arsenal into the Middle East which they do not control, but to which they are hostage. This can only decrease their own security.

Israel’s official nuclear defense policy is one of “ambiguity”—making or being able to make nuclear weapons, but stopping short of open deployment or testing. This is really a policy of regional dominance. Whatever benefit ambiguity has disappears once a rival becomes ambiguous too. Nuclear rivals must each react to what the others do, or appear to do. Freedom of action is inevitably lost, and the rivals slide from a covert to an overt nuclear arms race. Preserving ambiguity may have been the real reason for Israel’s attack on Iraq’s reactor. Iraq was not threatening Israel’s military position nearly as much as it was Israel’s ability to remain “ambiguous.” Ambiguity in Iraq would require Israel to test and deploy in order to maintain Israel’s advantage. To retain its current policy, Israel must remain the only nuclear power in the Middle East.

Israel’s arsenal constantly presses the Arabs to build their own arsenals. It is naive to think that Arab bombs will never be made, or could not be delivered to the United States. If nuclear arms do spread through the Arab world, humanity’s net control over the atom will have approached the vanishing point.

The alternatives are to do nothing—the current posture—or to take proliferation seriously. Israel depends on U.S. economic support, weapons, and, ultimately, U.S. assurance of Israel’s survival. In return, it is reasonable for the United States to ask Israel to keep its pledges of peaceful use of heavy water. An Israeli refusal would mean that Israel wants nuclear weapons more than alliances or that Israel does not believe that Norway and the United States have the will to enforce their rights.

By right, Norway may inspect all the plutonium its heavy water has ever made in Israel. In theory, this means dismantling virtually every nuclear weapon in Israel. The United States has the same right for any plutonium made with its water. Also by right, Norway may ask France for assurances to back up the no-re-export pledge, and, if there is a peaceful-use pledge, for proof that the pledge has been honored. In theory, this means that France may have to present its heavy water records, dismantle a large number of French nuclear weapons, and possibly postpone any French deployment of neutron bombs. Is it feasible to enforce such rights? Will France and Israel give up nuclear firepower, or France embarrass itself, merely to uphold an agreement?

France has changed since the days when it started Celestin. As a first-line nuclear exporter, it now has an interest in preserving controls. France is trying hard to sell reactors and fuel across the world to keep its nuclear industry going. It is demanding pledges of peaceful use, no re-export, and international inspection. It obviously expects these promises to be honored. But how can France ask others not to make bombs with its exports if France itself makes bombs with Norwegian exports? The question in France should be one of precedent and example. To safeguard the system that France now uses as a supplier, France should reveal its heavy water data and account for Celestin.

Israel’s position is more delicate. Israel’s, indispensable protector and benefactor, the United States, has publicly staked its entire nonproliferation policy on the idea that thepeaceful-use pledge should be respected. So has Norway. Indeed, so have all the nuclear supplier countries that subscribe to either the Nonproliferation Treaty or the Nuclear Supplier Guidelines. To save the credibility of the nuclear export trade, Israel must keep the faith—or appear to do so.

A bogus or superficial inspection is an obvious temptation. It would give the appearance of compliance and free both Israel and Norway from obligation. But if Norway settled for looking at heavy water in barrels without checking it for tritium, Norway would surrender all hope of controlling the plutonium made with the water in the past. Even if Norway saw the water in barrels, however, Israel still would have a problem. Once Norway’s heavy water was located, Norway would have the right to have it inspected in the future. Inspectors could check it so often as to prevent its return to a reactor without their knowing. Therefore, after declaring the water to be Norwegian, Israel could not use it in Dimona without putting the reactor’s plutonium under inspection—or at least a fraction of it equal to the fraction of the reactor’s heavy water that came from Norway.

If Israel put the water back into Dimona and elected to run the reactor partly for military and partly for civilian use, it would be crucial to identify the Norwegian water Israel obtained from France. Doing this would show that virtually all of Dimona’s water came from Norway and thus was restricted. Virtually all of the plutonium therefore would be restricted as well. This essentially would remove Dimona as a future source of atomic bombs.

The same result seems likely if Israel refuses inspection. A failure to cooperate would allow Norway to demand the water back. This would deprive Dimona of its use, probably forcing the facility to dose. Could Israel both refuse inspection and refuse to give the water back? Such a renegade act would repudiate a clear obligation. Norway could punish and isolate Israel diplomatically. Not even the United States could oppose Norwegian-sponsored sanctions, such as a motion to expel Israel from the IAEA.

Regardless of how one feels about applying the peaceful-use pledge to past events, there are rights that can and should be enforced in the future. France and Israel easily could stop making bombs with imported material. India could stop threatening to do so. France, by its good example, would bolster the integrity of export controls; Israel, by cutting back bomb production, would ease its nuclear pressure on the Arabs; and India, by keeping its word, would improve its relations with the United States and reduce the threat to Pakistan. All of these outcomes would enhance world security at a fair and reasonable cost.

In the future, controls on heavy water should be improved. No other supplier should duplicate Norway’s reckless exports. Exporters should enforce their rights and demand an inspection system able to determine where their heavy water is and for what it is used. Today neither Euratom nor the IAEA guarantees the whereabouts or use of heavy water. If the recipient country officially adheres to the Nonproliferation Treaty, the inspectors merely keep track of its plutonium. They do not look at the heavy water because the treaty requires that all the recipient’s plutonium—whether made with heavy or light water—be placed in an inspected inventory. The inspectors assume that the inventory eventually will include any plutonium that heavy water makes. But because no one is looking at the water, the recipient can break the no-re-export pledge without being caught. The recipient simply can refrain from using the water and ship it elsewhere.

If the recipient has not joined the treaty, the inspectors are supposed to keep track of the water so that it cannot be put in an uninspected reactor. However, they do not check the water for tritium, so they cannot know whether the water has been in a reactor. This practice allows Israel, for example, to assert that the U.S. water has not been used since 1963. The problem will continue even if Israel agrees to future inspection’ unless Norway either tests the water for tritium or checks it frequently enough to prevent its being shifted around.

Loose controls on heavy water exports now have enabled Israel and possibly France to break the peaceful-use pledge, tempted India to do so, and brought Israel to the verge of breaking the inspection pledge. These developments throw all nuclear trade into question and threaten global security. Norway and the United States have the responsibility and the power to enforce the pledges and should do so now. Regardless of the political discomfort of applying the pledges to the past, they can and clearly should be enforced in the future.


[1] See Elaine Davenport, Paul Eddy, and Peter Gillman, The Plumbat Affair (Philadelphia: Lippincott, 1978); Steve Weissman and Herbert Krosney, The Islamic Bomb (New York: Times Books, 1981), 124.

[2] “How Israel Got the Bomb,” Time, 12 April 1976, 39.

[3] New York Times, 25 June 1981, A1.

[4] New York Times, 10 November 1986, A10.

[5] “Former Official Says France Helped Build Israel’s Dimona Complex,” Nucleonics Week, 16 October 1986, 6-7.

[6] The only published report is by Knut-Einar Norberg, “New Norwegian Attempts at Heavy Water Control” (in Norwegian), Dagen, 5 April 1987, stating that Norwegian sold France more than 100t in the 1950s and 1960s.

[7] See Gary Milhollin, “Dateline New Delhi: India’s Nuclear Cover-up,” FOREIGN POLICY 64 (Fall 1986): 161-162.

[8] New York Times, 26 May 1987, A13.