Commission to Assess the Ballistic Missile Threat to the United States
Appendix III: Unclassified Working Papers
Henry Sokolski 1 : "Space Technology Transfers and Missile Proliferation"
Space Cooperation for Peace?
If one had to identify the greatest drivers of missile proliferation,
Russia and China would surely lead the list. These two nations have
transferred missile technicians, critical missile parts and production
facilities to Iran, Brazil, North Korea, Syria, Pakistan, India and Iraq.
Russia, moreover, has been in the wholesale business of modernizing China's
missile force for nearly a decade. When one speaks of missile
proliferation, then, it is not surprising that most policy makers focus on
these suppliers, their client states and the physical transfers they make.
Concentrating on these nations' tangible missile proliferation, though,
ignores a much more uncontrollable phenomenon, which is U.S., Chinese, and
Russian civilian space cooperation. Through this commerce, the U.S. itself
encourages and even subsidizes known Russian and Chinese missile
proliferation entities, while Russia has been able to transfer all manner
of military technology to China and India. Unfortunately, this help is all
too often downplayed within the policy and intelligence communities. A key
reason why is that this financial, technical help is difficult to monitor
(photographs are of little help here) and is politically sensitive. Despite
this or perhaps because of it, it is quite significant. It encourages
Russian and Chinese officials to discount our missile nonproliferation
pleadings, fosters a disturbing cynicism regarding aerospace export
controls, and accelerates missile proliferation generally.
What does this missile technology assistance consist of? In the case of the
U.S., a number of things. First, U.S. - NASA space cooperation worth well
over $400 million (from l994 through l997) with The Russian Space
Agency--an entity that the Russian press itself recently confirmed has
funneled Russian missile technology to Iran. 2 Second, the transfer of
U.S.-made satellites to Chinese and Russian space launcher providers, whose
commercial operations would otherwise would be seriously underutilized.
These firms include Salyut/Khrunichev, STC Complex, Polyot (which make
variety of rockets, including SS-25 military derivatives), Great Wall
Industries (a major exporter of Chinese missiles), the Chinese Academy of
Launch Vehicle Technology (the civilian successor to BWYIC, maker of
Chinese missiles including the M-Family systems fired against Taiwan and
exported to Pakistan). Third, the hard currency to pay for the launch of
these satellites--from $25 million to $85 million a launch; approximately
$500 million to date for Russia (10 launches including low earth
orbit--LEO--launches) and China (14 launches including LEO launches)
each--money these firms can use to finance their own missile development
and exports. 3 Fourth, intangible technology--coupling load analysis,
guidance data packages, upper-stage solid rocket propellant certification,
upper stage control design validation, lower stage design validation,
general quality assurance, etc.--critical not only to assure timely,
reliable, accurate placement of satellites in space, but also of missile
warheads against specific enemy targets. 4
How did this commerce get started? In l987, a year after the Challenger
disaster, the U.S. government decided to relieve the demand on limited ESA
and U.S. space launch services by allowing China a chance to bid on the
launch of U.S.-made satellites. The risks of helping validate the design of
the Long-March rocket (which China had designed as a military system) 5 was
well known. I know this because, as a Senate staffer, I had made inquires
at fairly senior levels at the Defense Department about the attendant risks
of letting our satellite companies work with the Chinese on Long March
launches. The argument made at the time was that such cooperation might
well strengthen China militarily and that given our continuing Cold War
rivalry with the Soviet Union, this was a benefit. In short, indirect
missile technology transfers were assumed.
After the Soviet Union's demise, the U.S. continued to approve Chinese
launches largely because U.S. satellite manufacturers and their customers
demanded it. Not only that, but an active effort was made to encourage
Russian launches of U.S. satellites as well. The argument here was not just
driven by commercial demand, but missile nonproliferation: The U.S. had an
obligation, State Department officials insisted, to employ Russia's missile
makers in peaceful pursuits lest they take their expertise and goods to
countries like Iran.
A corollary rationale was, then, used in l993 to support U.S. sponsorship
of Russian membership in the MTCR. To get Russia to pacify its missile
technology and missile technology exports, the U.S. had to give it positive
incentives--a larger quota of satellite launches, increased
government-to-government space cooperation, and membership in the MTCR. Of
course, under the missile technology control provisions of the National
Security Authorization Act of l990, once a nation became a member of the
MTCR it could only be sanctioned if it could be shown that it intentionally
allowed the export a controlled missile item in violation of its own laws
and refused to investigate the matter - a most improbable matter to prove.
Also, if the export was made to another MTCR nation, sanctions against the
exporting nation were prohibited.
The Russia-ISRO Case
Before U.S. officials could make Russia an MTCR member, however, they had
to resolve an existing sanctions action taken by the previous
administration against Glavcosmos, a Russian missile export firm that
contracted with the Indian Scientific Research Organization (ISRO) to sell
it several cryogenic upper stages along with the production technology
necessary to produce them. 6 Under the MTCR, the transfer of missile
production technology--hardware or know-how--is strictly forbidden. More
important, the Russian upper stage technology could be used to help India
extend the range of its rockets to reach Beijing and to improve the
accuracy of its missiles. 7 Initially, the Russians refused to terminate
the deal but after U.S. officials promised Russia $400 to cooperate in
developing the space station and to expand the number of U.S.-made
satellite launches Russia could bid on, Moscow relented. In July of l993,
Russian President Yeltsin promised President Clinton to reconfigure
Russia's contract with ISRO by November 1 of l993 so it would exclude any
transfers of production technology.
Between July and October, however, ISRO personnel were found on site in the
Russian rocket manufacturing plants that made the cryogenic stages and
Indian officials claimed that Russia transferred more than "4/5ths" of the
sanctioned production technology. 8 India Minister of State in the
Department of Atomic Energy and Space also announced that, contrary to the
Russians' pledge not to transfer "technical manuals" that would permit
India to produce its own cryogenic rocket engines, Russia sent it "drawings
of the engine" in September of l993 that would enable India to produce the
engines within a few years. 9
Even though all of these missile technology transfers clearly violated the
MTCR and the spirit of the Yeltsin-Clinton agreement, which was
subsequently finalized in September of l993, they were forgiven on the
legal technicality that the agreement with the U.S. allowed Russia until
November 1 of l993 to reconfigure its contract with ISRO. That assessments
of deal should still differ so much (one former high-level official
recently dismissed the deal as a clear failure while Deputy Secretary of
State Strobe Talbott recently cited it as one of the most significant
nonproliferation successes of the Clinton Administration) is itself a
indication of the inherently ambiguous character of such intangible
technology controls. As one Hill staffer complained early on about the
deal, "It's not clear about what's allowed and not allowed." 10
Worse, yet, even where the deal was clear and clearly violated, the l993
understanding was not enforced. This was driven home in June of l994 when
Motorola officials visited the Russian aerospace firm of Salyut/Khrunichev
and discovered a six-foot tall, high-fidelity scale mock-up of the Indian
Geostationary Launch Vehicle (GSLV). The mock-up was extremely detailed and
depicted all the interfaces for the entire launch system--i.e., launch
gantry, pad fittings, fueling tubes, and cryogenic umbilicals--and the
Russian cryogenic upper stage. When asked what the purpose of the model
was, the Russians explained that they were teaching the Indians how to
integrate the launch of their GSLV with the cryogenic upper stage. 11
Transfers of such launch integration technology is clearly prohibited by
paragraphs 2 and 4 of the MTCR guidelines and paragraphs 1 (b); 2 (b)(2) of
the MTCR Annex. Yet, the Chairman of the House Science, Space and
Technology Committee, who had requested to be kept thoroughly briefed on
intelligence relating to Russia's compliance with the July l993 deal, knew
nothing about this violation until his committee staff learned of it
through unclassified channels from me. This revelation promoted a series of
memos and a request from the Chairman for a meeting with Vice President
Gore, who negotiated the final l993 space cooperation-missile
nonproliferation deal with Russia. 12 At this meeting, which was held in
September, Vice President Gore professed no knowledge of the Motorola
meeting but promised to have the intelligence community "look into it."
U.S. Satellite Transfers to China
The factors that doomed the U.S.-Russian understanding of l993 to
failure--the financial pressures of U.S. satellite producers to find
alternative launch services, the political pressures to maintain good
relations with a former foe, the inherent difficulties of controlling
intangible missile technology, and the interchangeability between civilian
and military space technology--have also come into play with recent U.S.
satellite transfers to China. As noted earlier, the U.S. allowed China to
bid on the launch of U.S.-made satellites in l987. Since then the firm,
which sells Chinese space launch services, Great Wall Industries, has been
sanctioned under U.S. missile sanctions twice. These sanctions, imposed in
l99l and l993 for Great Wall Industries' export of M-ll tactical ballistic
missile technology exports to Pakistan, however, were lifted in no small
part because of U.S. satellite manufacturers' desires to unfettered access
to Chinese launch services. More important, the need U.S. satellite makers
have to assure successful launches have tempted them all too frequently to
share some of America's most sensitive missile technology.
The need for such transfers became clear in December of l992 when the
Chinese launch of a Hughes satellite, Optus B2 for Australia failed. Hughes
had plans to launch many more of its satellites with the Chinese. With
satellites costing hundreds of millions of dollars (vice the launch
services, which cost between $25 million to $85 million) , Hughes and; its
satellite customers, and, later, other satellite makers, such as Loral,
could hardly afford additional Chinese launch failures. These would not
only increase the cost of insuring the launches, but delay the operation of
satellites (and, hence, the revenues they might generate) by as much as 18
to 24 months. Since the satellites' purchase were heavily financed, this
time literally cost money.
Starting in l993, then, Hughes and other satellite makers who wanted China
to launch their payloads made every effort to limit the possibilities for
error. Discussions were joined with the Chinese Academy of Launch Vehicle
Technology (CALT) about the sufficiency of CALT's attitude control system
on China's Long March launcher, a system the Chinese said they had already
proven on their M-family tactical ballistic missiles. Martin-Marietta,
meanwhile, requested an opportunity to certify the propellant and witness
the test firings of a Chinese apogee kick motor, propellant not dissimilar
to what the Chinese are trying to perfect in their development of a
road-mobile solid rocket ICBM.
Martin-Marietta also sought to exchange information concerning the
reliability and integration of the Chinese upper-stage control systems,
which are critical to the separation and ignition of that section, to its
proper attitude control and the spin release of the satellite. These same
control functions are also critical to China's perfection of an accurate
Multiple Independently targeted reentry vehicle (MIRV) for its new
solid-rocket system as is Motorola's interest in having Great Wall launch
small low earth orbit Iridium satellites with a multi-satellite dispenser
that might double as a multiple warhead dispenser. Finally, late in l993, a
number of U.S. satellite makers asked the U.S. Commerce Department if they
could share coupling load analysis critical to assure that China's space
launchers will ignite, separate their stages, and cut off their engines or
motors in a manner that would not shatter the satellites they are carrying.
This would not only help assure the safe, efficient launching of the most
sensitive civilian payloads, but also of complex MIRVed military systems as
well. None of this should be seen as news. In fact, the sensitivity of such
technology transfers was flagged over four years ago in a detailed article
I wrote for the International Defense Review entitled "U.S. Satellites to
China: Unseen Proliferation Concerns." Again, this piece was based entirely
on unclassified information. It received some notice on Capitol Hill: The
House Armed Services used it to block an attempt to move a substantial
amount of satellite-related hardware and technology off the munitions list
and under Commerce Department jurisdiction.
Yet, the most interesting reception came later. First, almost a year after
the article's publication, the CIA asked me to brief my piece to a group of
analysts. I was introduced as the author of the "brilliant piece" published
in International Defense Review. I flatter pretty easily but this seemed a
bit bizarre: The audience consisted of mostly of aerospace engineers, I'm
only a political science major. Apparently, the Agency was just beginning a
project to assess the military impact of U.S. dual-use technology
transfers. My article, which was merely the result of unclassified
discussions with others, was considered "path breaking" I assume the
intelligence community now has such intangible satellite and space-related
technology transfers much more adequately covered. It would be worth
As disconcerting as this reception was, though, the next bounce I received
from the article was even more bizarre. It turns out that The New York
Times ran a story last Saturday concerning a grand jury investigation into
whether or not Hughes and Loral had transferred rocket guidance technology
to the Chinese after the first Long March launch failure in l996 in
contravention of U.S. export control laws. 13 ABC News, The Los Angeles
Times, The New York Times, and several other news organizations all were
calling wanting to know what I knew about the investigation since, as one
reporter put it, "you predicted it". Unfortunately, I couldn't tell them
much since I didn't know anything. What I do know, though, is that this
story was predictable and that as long as the U.S. tolerates Russian space
cooperation with nations like India and China and continues to extend such
cooperation itself, missile proliferation is likely to accelerate, not
1. Henry Sokolski is the Executive Director of the Nonproliferation Policy
Education Center, a Washington-based nonprofit organization founded in 1994
to promote a better understanding of strategic weapons proliferation issues
for academics, policy makers and the media. Also teaches graduate school
courses on proliferation issues at Boston University's Institute of World
Politics in Washington, D.C.
2. See Yevgeniya Albata, Novaya Gazeta Ponedelnik Russian No. 10, "Hiring
of Missile Experts for Iran Described," take 5 of 7, March 16-22, 1998.
3. Data derived from interviews with State Department officials, Scott Pace
of RAND, and Charles Cline.
4. See Henry Sokolski, "US Satellites to China: Unseen Proliferation
Concerns," International Defense Review, April 1994, pp. 23-26.
5. The Long-March rocket doubles as an ICBM and test bed for developing new
6. See U.S. Department of State, Office of the Assistant
Secretary/Spokesman, Margaret Tutwiller, "Russian Sale of Rocket Engine to
India," May 11, 1992.
7. On the military significance of the cryogenic engines, see Henry
Sokolski, "The Military Significance of the Original Russian Rocket
Technology Deal with India," attached as Appendix A.
8. See Vivek Raghuvanishi, "Russia, India Discuss Cryogenic Contract,"
Space News, November 15-28, 1993.
9. See John Wallach, "Clinton-Yeltsin Pledges Unlock Missile Dispute,"
Hearst Publications, July 26-August 6, 1993 and Reuters, "Russia Not to
Give Rocket Technology, India Says," July 25, 1994.
10. Defense Daily, "NASA Downplays Impact of Russia-India Engine Deal,"
July 27, 1994.
11. Interview with a Defense Department official present at the Moscow
12. See Nick Furman to Mr. Sensenbrenner, "Potential Russian Violations of
the Missile Technology Control Regime," August 1, 1994.
13. Jeff Gerth with Raymond Bonner, "Grand Jury Probes 2 Firms' Ties to
China Missile Program," The New York Times, April 4, 1998.