Sharon K. Weiner - Specialized Research AI at MIT

Sharon K. Weiner

Looking out, looking in: competing
organizational interests & the prolif-
eration of Soviet wmdexpertise

The collapse of the Soviet Union in
the early 1990s raised concerns about
the security of its nuclear weapons.1
In response, the United States joined
forces with countries of the former So-
viet Union, especially Russia, as well as
the European Union and other states,
to create a series of programs aimed
at securing former Soviet weapons of
mass destruction (wmd), weapons-
relevant materials, and scienti½c exper-
tise. Of these efforts, the most troubled
has been the one aimed at containing
wmd skills and knowledge. Former So-
viet weapons experts haven’t sold their
knowledge around the world; indeed,
there have been almost no document-
ed cases of such proliferation (although
concerns remain about what goes unre-
ported). Rather, it is the means chosen
for ½ghting such proliferation–work-
ing with and reemploying wmd experts
–that have proven problematic.

Cooperation with former cold war
enemies certainly created a host of dif½-
culties, as did the secrecy that surrounds
nuclear, biological, and chemical weap-
ons efforts. But there’s a signi½cant bar-
rier to success found much closer to
home. The U.S. bureaucracies tasked

with implementing programs focused
on the proliferation of wmd expertise
came up against a problem common to
all organizations: the need to pursue
and protect their own interests.

The notion that organizations have
their own interests is well-established,
although too often national security is-
sues are assumed to be so important they
trump this self-interested behavior. Or-
ganizations, the literature claims, seek
to manage the environments in which
they act. In particular, organizations
that have to please similar authorities
and that face comparable constraints
exhibit isomorphic behavior. That is,
they often respond to external pressure
by aligning their interests with the in-
terests of powerful external forces. This,
in turn, means that organizations that
face similar environments come to re-
semble one another, either through co-
ordinated duplication or mimicry. But
organizations also need to ½t in at home.
When organizations are given new tasks,
these tasks come to be de½ned and im-
plemented in ways that accommodate
and reinforce the interests of the parent
organization.

U.S. nonproliferation programs tried

both to accommodate their external
environments and to match their goals
with the goals of their parent organiza-

Dædalus Spring 2009

105

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Sharon K.
Weiner

tions. The two largest programs, one
in the State Department and another
in the Department of Energy, adopted
similar narratives about how their ac-
tivities furthered nonproliferation.
Each program, however, also adopted
an implementation strategy that was
heavily influenced by its parent organi-
zation. What resulted were two pro-
grams that accommodated their inter-
nal and external environments but
were unable to achieve their original
nonproliferation goals.

As the Soviet Union entered its ½nal
days, concerns increased about possible
proliferation from its wmd complex,
in part because of the potential for a vio-
lent transition, the accidental or unau-
thorized seizure and use of weapons,
and uncertainty about the future of po-
litical relations with any Soviet succes-
sor state. Concerns also grew because
Soviet security had focused on external
border points, paying less attention to
protective measures at individual facili-
ties. When the Soviet Union collapsed,
many newly independent states inher-
ited weapons facilities that lacked suf-
½cient measures to ensure that the
weapons and material inside were safe
from theft. In many cases, the contents
of the facilities were inadequately inven-
toried, and sometimes little was known
about what had happened inside. More-
over, economic conditions were grim
for most people in the former Soviet
Union, including guards and experts at
the weapons facilities. Salaries were low
–a weapons scientist might make $100
per month or less, for example–and
paychecks were often delayed by several
months. Weapons facilities saw severe-
ly reduced government subsidies, and
goods and services overall cost more.

This combination of poor security, un-

certainty, and dire economic prospects

led the United States to fund a variety
of efforts, collectively known as Coop-
erative Threat Reduction (ctr), aimed
at countering proliferation from the for-
mer Soviet wmd complex. Some parts
of ctr focused on securing the weap-
ons themselves; other efforts dealt with
weapons-relevant materials; and several
programs were created to deal with the
possible proliferation of weapons exper-
tise. This last group of programs concen-
trated on providing income to weapons
experts by funding short-term research
contracts and, in the long term, working
to reemploy the experts in commercial
or non-defense, government work. Such
programs were created in the Depart-
ments of Defense, Energy, and State,
and included a variety of private and
quasi-government efforts. The two larg-
est programs were the Science Centers,
managed by the State Department, and
the Initiatives for Proliferation Preven-
tion (ipp) program, in the Department
of Energy.

ipp and the Science Centers date
from the ½rst years of ctr. In early
1992, U.S. Secretary of State James
Baker announced the creation of the
Science Centers during a visit to the
closed nuclear city of Snezhinsk, Rus-
sia. The United States, along with
members of the European Union, Ja-
pan, Canada, Sweden, and later Nor-
way and South Korea, would create
two centers that would fund short-
term research projects involving for-
mer Soviet weapons experts and an
outside research collaborator. The ½rst
center (also the largest) is the Interna-
tional Science and Technology Center
(istc), based in Moscow, which works
with scientists from Russia, Georgia,
Armenia, Belarus, Kazakhstan, Kyr-
gyzstan, and Tajikistan. The smaller
Science and Technology Center of
Ukraine (stcu), based in Kyiv, fo-

106

Dædalus Spring 2009

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

cuses on Ukraine, Azerbaijan, Georgia,
Moldova, and Uzbekistan. Until 2004,
the United States provided at least half
of the funding for both centers; U.S.
influence has dominated center policy
and implementation strategies.

Although the istc was to be up and
running in 1993, domestic political prob-
lems in Russia caused delays. Some legis-
lators questioned the motives behind the
center, wondering whether it was an ex-
cuse to spy on Russia’s national security
complex. Further delays arose from the
wider struggle for power between the
Duma and Russian President Boris Yelt-
sin. In September 1993, Yeltsin dismissed
the Duma, which had not yet approved
the measures needed to start the istc.
By the end of the year, Yeltsin himself
approved the necessary documents, and
the istc was able to begin operations in
March 1994. The stcu was also delayed,
largely due to administrative issues as-
sociated with creating a government in
the newly independent state of Ukraine.
It began operations in November 1995.
Delays with the Science Centers

prompted calls for more immediate ac-
tion from the U.S. nuclear weapons labo-
ratories. U.S. nuclear weapons experts
had a long history of scienti½c collabora-
tion with their Soviet counterparts that
dated from discussions in the early 1960s
about research on large magnetic ½elds.
Later, joint experiments were conduct-
ed for the purposes of arms-control ver-
i½cation. By the early 1990s, some ex-
perts from the U.S. labs were well-aware
of the dire economic and security prob-
lems in Russia, so when the Science Cen-
ters were delayed, U.S. labs at Los Alam-
os and Livermore began to fund research
collaborations with a few key Russian
nuclear weapons facilities. In 1993 fund-
ing was appropriated by Congress, and
the next year ipp became a formal pro-
gram.

Similar to the Science Centers, ipp
funds collaborative, non-defense-relat-
ed research between former Soviet wmd
experts and outside scientists. Unlike the
Science Centers, collaborators in the ipp
program usually come from the U.S. lab
complex. Careful not to be perceived as
duplicating the work of the Science Cen-
ters, ipp created a second type of coop-
erative research that required matching
contributions from U.S. industry. In the-
ory this would help businesses enter the
market in a post-Soviet country and hire
away former weapons experts whose
skills and products had been evaluated
through the small-team research collab-
orations made possible by ipp.

Both the Science Centers and ipp
were born from the assumption that
economically desperate weapons ex-
perts are potential proliferation prob-
lems. The founding documents and
debate that led to each of the programs
show that they share the same mandate.
In the short term, they are to fund col-
laborative research between teams of
weapons experts and an outside col-
laborator. Besides providing income to
the former Soviet scientists, engineers,
and technicians, such research is sup-
posed to help them become aware of
Western research standards and how
weapons skills can be translated into
private, non-defense-related jobs. Over
the long term, the Science Centers and
ipp are supposed to help these former
Soviet experts transition to reemploy-
ment outside of the weapons complex.
This long-term mandate, designed to
help Russia and other post-Soviet
countries permanently reduce the size
of their weapons complex, uses com-
mercial interests as a way to, over time,
end U.S. government funding of such
efforts. This mandate gave both pro-
grams signi½cant discretion to imple-
ment their programs and to respond

Competing
organiza-
tional in-
terests &
the prolif-
eration of
Soviet
wmd
expertise

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2009

107

Sharon K.
Weiner

to constraints they would encounter in
the United States and the former Soviet
Union. As predicted by the literature,
both programs ultimately sought to ac-
commodate the constraints that arose
from these external environments in
much the same way.

Organizations have no choice but to
accommodate the demands of their ex-
ternal environments.2 These environ-
ments set not just standards for results,
but assumptions about legitimate out-
comes, and about how outcomes are
to be measured and what constitutes
signi½cant progress. When the actors
who control an organization’s funding,
scope, and autonomy also de½ne these
assumptions, the actors can be ignored
only at great cost to an organization’s
budget, mission, and possible future.
Thus organizations come to do and
measure what is valued by their envi-
ronments, regardless of whether it is
a rational or ef½cient means of achiev-
ing their original goals.

Further, the literature on organiza-
tions suggests that when asked to per-
form similar missions in a similar envi-
ronment, organizations will adopt the
same myths as justi½cations for their
tasks. And they do so without conscious
coordination. Over time organizations
tend to look and act alike, not because
there is general agreement that they are
rationally and ef½ciently pursuing their
goals, but because they share the same
myths about what they consider to be
legitimate activity.

The Science Centers and ipp demon-
strate this isomorphism in several sig-
ni½cant ways, including how the two
programs prove that their projects in-
volve genuine weapons experts; their
responses to pressure to develop an
exit strategy; their measurement of job
creation; and their recent decisions to

cease work in Russia in favor of other
countries. In each case, the programs
adopted similar rules and procedures
as legitimate, yet they did so without
signi½cant coordination. This uninten-
tional alignment caused the programs
to stray from their original nonprolif-
eration goals.

Both ipp and the Science Centers are
supposed to focus on engaging people
with wmd credentials, including engi-
neers and technicians, but especially
scientists who have the core skills need-
ed for making weapons. Both programs
needed to demonstrate that they were
dealing with the most dangerous prolif-
eration concerns, but the legacy of mis-
trust and secrecy between the United
States and the Soviet Union often made
it impossible, or at least impolitic, to in-
quire about the skills of speci½c wmd
experts. The programs, therefore, adopt-
ed two strategies for legitimization.
One used the process of selecting proj-
ect teams to make claims about reach-
ing people with critical weapons skills.
The second drew from the notion that
U.S. project managers “just knew” who
was or was not a weapons expert.

Without any obvious coordination,
both ipp and the Science Centers came
to focus on the rules of project selec-
tion as a means of verifying the weap-
ons credentials of the people involved.
At ipp, projects are normally not ap-
proved unless at least 60 percent of the
project members are weapons experts.
Although the Science Centers were
largely independent of each other, they
also adopted similar criteria: for basic
scienti½c collaboration, project teams
need to be at least 50 percent, prefer-
ably 60 percent, wmd experts before
U.S. funding is considered. Both pro-
grams assumed that, as long as they
funded at least some wmd experts,
they addressed proliferation concerns;

108

Dædalus Spring 2009

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

thus the number of experts on each
project team took priority, with little
emphasis on the quality of those weap-
ons skills or whether they were critical
or tangential to weapons development.
Moreover, there is no evidence that ei-
ther program identi½ed the weapons
experts most likely to proliferate, or
considered whether program funding
should be focused according to the pov-
erty level of experts or in terms of the
relative value of their skills to a poten-
tial proliferator.

As a second measure to verify weap-

ons credentials, the Science Centers
adopted a process by which they asked
each weapons expert to identify his or
her particular skills according to a stan-
dard list of specialties and sub-special-
ties. ipp adopted a process and list that
was close to identical, and neither pro-
gram insisted on much external valida-
tion of the Soviet experts’ self-identi-
½ed skills–despite potential incentives
to misrepresent abilities. Having more
team members with weapons creden-
tials increased the likelihood of a proj-
ect being funded, and project funding
meant that weapons institutes, crippled
by economic conditions, stood to gain
updated equipment, spare parts, and
supplies for their operations. Eager for
such enhancements, institute directors
had little incentive to make sure their
employees were forthright about their
skills.

Instead of verifying weapons creden-
tials rigorously, both the Science Cen-
ters and ipp assumed that secrecy pre-
vented this. Over time, trust did devel-
op between the United States and for-
mer Soviet institutes, and veri½cation
became less sensitive, even at places in
Russia still engaged in nuclear weapons
work. Instead of pursuing more rigor-
ous options however, both programs re-
lied increasingly on the “you just know”

method of veri½cation. Program man-
agers claimed that over time they got to
know weapons experts in speci½c facili-
ties and that their own expertise suited
them to judge commensurate ability.
Rather than collect additional informa-
tion or verify what was offered by for-
mer Soviet experts, U.S. programs re-
lied on an ambiguously described and
informally gained sense of the skills of
former Soviet experts.

In 1997, both the Science Centers and
ipp came under increased pressure in
the United States to show concrete re-
sults and to demonstrate an exit strat-
egy for U.S. government assistance. In
response, both programs reinvigorated
their efforts to move beyond short-term
research collaborations between scien-
tists to more commercial-focused efforts
that involve partial funding from busi-
nesses. At ipp these were referred to as
Thrust II projects. They required that
companies match U.S. government con-
tributions and that research emphasize
a commercial application. (There had
always been a commercial component
of ipp activities, but most funding had
gone toward Thrust I projects, involving
collaborations between scientists over
basic research questions.) Similarly, the
Science Centers also began to emphasize
projects with contributions from third
parties, only some of which were private
companies. Referred to as Partner Proj-
ects, the research focused more on po-
tential commercialization and less on
basic research. Beginning in 1997, the
Science Centers and ipp both placed
more emphasis on these activities, and
by 2000 they constituted at least half
of project funding in both programs.
These new types of projects, with
third-party participants, represented
the means by which both programs
sought to reemploy former Soviet ex-

Competing
organiza-
tional in-
terests &
the prolif-
eration of
Soviet
wmd
expertise

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2009

109

Sharon K.
Weiner

perts, reduce the number of potential
proliferation problems, and, in turn,
provide a strategy for eventually end-
ing U.S.-funded assistance. Economic
conditions in the former Soviet Union,
however, dampened hopes for business
expansion or investment. Repeatedly,
teams of weapons experts found it dif-
½cult to meet international quality stan-
dards, produce on a competitive basis,
or incorporate market demands into
their products. Moreover, the State
and Energy Department personnel who
managed these projects had, on average,
little commercial business experience.
As evidence multiplied that signi½cant
job creation was probably impossible,
some in academic and policy communi-
ties suggested other alternatives, such
as retirement subsidies, increased im-
migration to the West, or subsidies for
U.S. companies to transfer and hire for-
mer Soviet experts in the United States,
although there is no evidence that any
alternative was seriously considered.
As both ipp and the Science Centers
concentrated on creating jobs for for-
mer weapons experts, Congress pres-
sured the programs to demonstrate
their success. Instead of measuring the
number of jobs created, both programs
emphasized the number of scientists
“engaged.” According to the Science
Centers, through 2005 their projects
have involved almost seventy thou-
sand people. ipp claims to have en-
gaged some sixteen thousand, as of
the end of 2006. These ½gures, how-
ever, raise three issues. First, both pro-
grams record the number of project par-
ticipants in general, not the number of
weapons experts in particular. If both
programs adhere to the goal that 60
percent of project teams are weapons
scientists, this makes questionable the
proliferation danger of the remaining
40 percent of people engaged. Further,

neither program established a concrete
estimate of its target population. In the
absence of this context, it is dif½cult to
know whether the reported numbers
represent signi½cant progress.

Second, even though the goal is job
creation, neither program measures
this result rigorously. The Science Cen-
ters keep no records of job creation,
and although ipp does ask for this in-
formation periodically, it is satis½ed
with gross estimates from project man-
agers or, sometimes, no information
at all. Indeed, repeated investigations
by the U.S. Government Accountabil-
ity Of½ce have faulted ipp for poor rec-
ord keeping. Additionally, neither ipp
nor the Science Centers investigates
the activities or jobs of wmd experts
once they cease working on U.S.-fund-
ed projects.

Finally, ipp and the Science Centers
do little to coordinate their project fund-
ing. Therefore, it is dif½cult to assess re-
liably the degree to which they fund du-
plicate projects or the same former Sovi-
et experts.

In the face of external pressure to
develop concrete indicators of suc-
cess, both the Science Centers and ipp
focused on the myth that the number
of people who participate in their ac-
tivities is a legitimate measure of prog-
ress. The programs ignored measures
that would have more accurately reflect-
ed progress toward the reemployment
of wmd experts or the ultimate goal
of reducing the danger of the prolifera-
tion of wmd expertise.
The response to U.S. concerns about

Russia’s reassertion of power illustrates
further isomorphism between the Sci-
ence Centers and ipp. During the early
2000s, under Putin the internal securi-
ty forces in Russia regained authority.
In turn, they were more likely to use

110

Dædalus Spring 2009

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

security as an excuse for limiting cooper-
ation, making it more dif½cult to inter-
act with former Soviet weapons experts
and gain access to facilities. Russia also
began to resist U.S. priorities for such
cooperation, and broader foreign poli-
cy disputes brought some programs
to temporary halts. The Bush adminis-
tration, in fact, pursued an across-the-
board rollback of cooperative work
with Russia.

The Science Center in Moscow and
ipp both responded by claiming that
Russia could now afford to assume
more of the burden for nonprolifera-
tion efforts, freeing the United States
to focus funding in other countries. In
the early 2000s, both programs began
to reduce funding for projects in Russia
and redirect efforts toward other parts
of the former Soviet Union and points
farther a½eld, such as Iraq and Libya.
Four years later the State Department
proposed including North Korea, South
Asia, and the Middle East. In 2006, ipp,
too, expanded its activities to Libya and
Iraq and changed its name to Global Ini-
tiatives for Proliferation Prevention.

This expansion, done with no obvious

coordination between programs, arose
from the increased external constraints
of working in Russia. Yet as both pro-
grams moved into other countries, they
did so in the knowledge that Russia re-
mained a key proliferation concern. Nei-
ther program had demonstrated signi½-
cant job creation for former weapons
experts, and Russia, while committed
to downsizing its nuclear workforce,
hesitated to do so until jobs were avail-
able for the excess workers. Economic
improvements did not extend to all
areas of the nuclear weapons complex,
and even where they did they were con-
tingent upon oil and gas prices, the
source of much of the Russian govern-
ment’s revenues. Some policy experts

expressed concerns that the Russian
government, despite an increased ability
to do so, had been unwilling to invest in
the utilities, spare parts, and equipment
necessary to keep U.S.-funded coopera-
tive security upgrades in place and func-
tional. They lamented the absence of a
“security culture” in Russia. Thus, in re-
sponse to external constraints–and with
plenty of evidence to the contrary–both
the Science Centers and ipp adopted the
myth that Russia could assume more of
the funding and administrative burden
itself.

Organizations face one set of con-

straints from the environments in
which they operate; they face a differ-
ent set at home. New organizations
created to implement novel tasks are
more likely to be successful if they do
so in ways that reinforce the mission,
interests, and goals of their parent or-
ganization. In other words, new pro-
grams need to ½t in at home.

The literature on organizations at-
tests to the problems experienced by
programs that are signi½cantly differ-
ent from their parent organizations.3
Goals and activities that contradict
standard routines, or that require sig-
ni½cant changes to them, tend either
to be marginalized, given away, or ter-
minated, or implemented in such a
way that forces them to line up with
the interests of the parent organiza-
tion. Although both ipp and the Sci-
ence Centers adopted the same strat-
egies for dealing with external con-
straints, they adopted signi½cantly
different implementation strategies,
in an attempt to align their program
goals with the interests of their par-
ent organization.

For the Science Centers, this meant
overcoming several characteristics that
put them out of sync with normal State

Competing
organiza-
tional in-
terests &
the prolif-
eration of
Soviet
wmd
expertise

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2009

111

Sharon K.
Weiner

Department routines. The Centers are
multinational organizations, and U.S.
participation is managed by the State
Department. At ½rst glance, this would
appear to indicate key commonalities:
the Centers require the sort of skills and
processes that make up the core of the
State Department’s diplomatic activi-
ties. However, some in the department
saw the Science Centers as a security
issue that more appropriately belonged
to the Defense Department. To others,
the Science Centers seemed out of place
because they were functional activities,
focusing on a particular type of work,
in contrast to the State Department’s
tendency to organize along country or
geographic-speci½c lines. Further, the
Agency for International Development
(aid), a quasi-autonomous part of the
State Department, oversees most coop-
erative efforts like the Science Centers.
In the past, there have been frequent
conflicts between aid and the State
Department over priorities, strategy,
and program direction. The staff of the
Science Centers also tends to be drawn
from the executive service, rather than
Foreign Service Of½cers (fsos), and so
they often do not enjoy the clout, au-
thority, or career advancement oppor-
tunities that the State Department af-
fords its fsos. Finally, the Science Cen-
ters were out of step with normal State
Department routines because they in-
herited especially strict oversight and
accounting rules from the Defense
Department, which helped lay the
groundwork for the Centers in the
early 1990s.

Conscious that they did not share
many of the features of usual State
Department programs, the Centers at-
tempted to ½t in by adopting the State
Department’s focus on process. The
State Department negotiates agree-
ments and implements decisions that

112

Dædalus Spring 2009

are made elsewhere, and is widely con-
sider to be a “process organization,” fo-
cused on the routines needed to ensure
implementation, not on the decision-
making needed to determine policy it-
self. Similarly, the Science Centers made
administering the Science Centers their
main focus. They established and came
to value a set of detailed rules and pro-
cedures for project development, selec-
tion, and management, thereby helping
to facilitate decision-making between
the different member nations, all of
whom retained the right to determine
their own funding priorities while em-
phasizing decision-making by consen-
sus. For these reasons, the Science Cen-
ters emphasized the process of project
administration, reasoning that nonpro-
liferation goals were met if due process
was respected.

Unfortunately, this led the Centers
to neglect the periodic reexamination
needed to ensure that program activities
still furthered nonproliferation goals.
Take the Partner Projects, for example.
The Science Centers, through U.S. lead-
ership, considered these projects a suc-
cess because they involved external fun-
ders and met established project crite-
ria. However, they neglected to consid-
er whether they resulted in job crea-
tion–a serious problem given that the
vast majority of partners are now U.S.
government agencies and not commer-
cial entities.

At ipp, the pressure to ½t in was differ-

ent. ipp was conceived and established
by experts from the U.S. nuclear weap-
ons labs. These labs are part of the De-
partment of Energy, which has tradition-
ally allowed them considerable autono-
my. As a result, ipp remained a largely
decentralized program in which most
signi½cant decisions, including those
about strategic direction and program
funding, were made by the individual

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

labs. There was a small ipp staff, but
it had the power only to persuade, not
command.

Initially the independence of the
labs proved very useful, enabling the
flexibility and individual variation nec-
essary to develop trust and cooperation
with former Soviet weapons institutes.
Over time, however, this independence
led to coordination problems as each
lab sought to send their own teams of
experts to Russia. The result was fre-
quent and excessive visits, which raised
concerns in Russia about spying. In the
United States, the labs allocated ipp re-
sources through logrolling rather than
prioritization, so project spending tend-
ed to reflect the need to harmonize
funding between U.S. labs rather than
the allocation of money according to
project merit. Additionally, ipp was un-
able to force the labs to keep accurate or
detailed records of their projects or the
former Soviet experts involved. This led
to a series of problems, including discov-
eries that more U.S. funding was going
to U.S. labs than to Russian ones, and,
most recently, to concerns that the U.S.
labs have funded project teams in Russia
that are also involved in nuclear energy
and, possibly, weapons activities in Iran.

Looking at U.S. nonproliferation activ-

ities through the lens of organizational
interest yields important policy-relevant
conclusions. Both the Science Centers
and ipp pursued inef½cient strategies for
½ghting the proliferation of wmd exper-
tise from the former Soviet Union, large-
ly because they skewed implementation
of their given tasks in an effort to con-
form to the activities valued by their par-
ent organization. For the Science Cen-
ters, this meant emphasizing the process
of project selection rather than consider-
ations of project quality, relevance, and
outcomes. Focusing on rules and proce-

dures over substance helped the Centers
overcome some of the factors that put
them at odds with normal State Depart-
ment activities. ipp, on the other hand,
preserved the structures of authority
that typically characterize relations be-
tween the U.S. labs and the Department
of Energy. As a result, ipp was unable to
enforce overall priorities or standards of
accountability.

But another source of pressure forced
the Science Centers and ipp away from
their original nonproliferation goals:
that of their external environments. Be-
cause of domestic political demands in
the United States and constraints in Rus-
sia, both programs adopted similar nar-
ratives for justifying their actions and
answers to how they verify weapons cre-
dentials, pursue exit strategies, measure
success, as well as why they are pulling
back from work in Russia. These actions,
responses to organizational interests and
external constraints, ultimately caused
both programs to diverge from the suc-
cessful and ef½cient pursuit of their non-
proliferation goals.

These cases show that the literature
on organizational interests has impor-
tant implications for U.S. nonprolifera-
tion policy. However, this study of the
Science Centers and ipp also provides le-
verage for understanding organizational
interest. Although it is well-established
in the literature that organizational in-
terest requires the accommodation of
both external and internal constraints,
there is insuf½cient attention to how or-
ganizations manage these conflicting de-
mands. The cases of ipp and the Science
Centers demonstrate how organizations
become isomorphic as they adopt strate-
gies to legitimize their activities in a way
valued by their external environment,
but also remain different in an effort to
½t in with their parent organization.

Competing
organiza-
tional in-
terests &
the prolif-
eration of
Soviet
wmd
expertise

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2009

113

Sharon K.
Weiner

ENDNOTES
1 The work leading to this essay was made possible in part by support from the American
Academy of Arts and Sciences, the Program on Science and Global Security at Princeton
University, American University, and, as a Carnegie Scholar in 2001, the Carnegie Corpo-
ration of New York. I also thank Rachel Sullivan Robinson and Barry Cohen for their valu-
able comments.
2 For a good summary of this literature, as well as a discussion of isomorphism, see John W.
Meyer and Brian Rowan, “Institutionalized Organizations: Formal Structure as Myth and
Ceremony,” American Journal of Sociology 83 (2) (1977): 340–363; and Paul J. DiMaggio and
Walter W. Powell, “The Iron Cage Revisited: Institutional Isomorphism and Collective Ra-
tionality,” in The New Institutionalism in Organizational Analysis, ed. Walter W. Powell and
Paul J. DiMaggio (Chicago: University of Chicago Press, 1991), 63–82.
3 Classic examples include Anthony Downs, Inside Bureaucracy (Boston: Little Brown, 1967);
Charles Perrow, Complex Organization, 3rd ed. (New York: McGraw-Hill, 1986); James G.
March and Johan P. Olsen, Rediscovering Institutions: The Organizational Basis of Politics
(New York: The Free Press, 1989).

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
8
2
1
0
5
1
8
2
9
6
0
9
d
a
e
d
2
0
0
9
1
3
8
2
1
0
5
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

114

Dædalus Spring 2009
Download pdf