** today ** Lecture Nov 1 12p: Laura McNamara and Timothy Trucano

[Stephen Guerin]

*** reminder: today ***
*** special time: 12p ***

SPEAKERs: Laura A. McNamara and Timothy G. Trucano
        Sandia National Laboratories

TITLE: Epistemological Issues in Computational Modeling and Simulation and High
Consequence Decision-Making

TIME: Wed Nov 1, 2006 12:00p  <-------- ** note special time **
LOCATION: 624 Agua Fria Conference Room

We will have breakfast burritos with the speakers at Cafe Dominic at 10:30a.
Everyone invited.
No lunch will be provided.

ABSTRACT:
Since the end of the Cold war, the US intelligence community has faced criticism
for repeatedly failing to predict major international events: the end of the
Cold war, India and Pakistan's nuclear tests, terrorist activities within and
outside the United States.  In response, institutions in the IC have been
looking for methodologies and technologies to improve performance in the
collection and analysis of intelligence information.  In particular, the IC's
analytical community is looking to modeling and simulation tools to
revolutionize intelligence analysis, enabling the collective to bridge
information gaps and promote knowledge discovery across (or perhaps despite)
intellectual, political, and organizational boundaries.    

This situation is not dissimilar to the crisis that the nuclear weapons
laboratories faced in the early 1990s, when the Hatfield Amendment killed the
testing program and the DOE introduced Science Based Stockpile Stewardship as
the new paradigm for assessing and certifying the safety, security, and
reliability of the nuclear stockpile.  In particular, both the nuclear weapons
and intelligence communities have invested in modeling and simulation
technologies for their capacity to synthesize large amounts of information in
relatively short periods of time, and for their predictive promise.  However, as
the nuclear weapons laboratories have discovered, predictive capability is a
hard thing to attain, and modeling and simulation tools often raise more
questions than they answer.

In this talk, we argue that the intelligence community and the nuclear weapons
laboratories are facing remarkably similar challenges in developing, assessing,
and integrating modeling and simulation tools into their mission activities.  In
particular, epistemological issues that tend to remain latent in academic
research environments get thrown into high relief when information generated by
modeling and simulation tools contributes to high consequence decisions. We
illustrate this point by reviewing research on modeling and simulation,
knowledge production, and prediction in economics, weather forecasting, climate
modeling.   We then present case studies from the nuclear weapons programs and
the intelligence community, both of which reveal the close coupling between
technology and organizational dynamics that characterizes modeling and
simulation in high-consequence decision making.

This talk is the outcome of two years' worth of discussion and collaboration
between Trucano, a mathematician who has spent his career in computational
physics at Sandia National Laboratories; and McNamara, a cultural anthropologist
who has studied knowledge production in both the nuclear weapons and the
intelligence communities. All topics will be discussed at the OUO level.