New York Times on the Web Forums Science

Missile Defense

Russian military leaders have expressed concern about US plans for a national missile defense system. Will defense technology be limited by possibilities for a strategic imbalance? Is this just SDI all over again?

wrcooper - 05:56pm Aug 21, 2001 EST (#7958 of 7969)

Mathematics And The National Missile Defense System

As Congress ponders a $3 billion increase in funding for a national missile defense system, University of Illinois professor Julian Palmore is looking at the program's prospects for success from a mathematician’s perspective.

To predict whether deployment of a proposed NMD system against an intercontinental ballistic missile attack makes sense, the UI mathematics professor and a colleague looked at applied basic insights drawn from a mathematical model known as game theory.

Their conclusions are detailed in the August issue of the journal Defense Analysis, in a paper titled "A Game Theory View of Preventive Defense Against Ballistic Missile Attack."

The paper's co-author is Francois Melese, a professor of economics at the Defense Resources Management Institute's Naval Postgraduate School in Monterey, Calif.

At the UI, Palmore is a faculty member in the UI's Program in Arms Control, Disarmament and International Security and teaches a course called "Technology and Security – Preventive Defense Against Weapons of Mass Destruction." He recently was chosen to serve as guest editor of a special issue of Defense Analysis on ballistic missile defense; the tentative publication date is April 2002.

Regarding the feasibility of the proposed NMD, Palmore and Melese write in the current issue that "the underlying assumption is that the objective of the administration is to minimize overall risk to the nation (or to maximize deterrence) at the lowest cost to taxpayers. Game theory asks us to place ourselves in the shoes of our adversaries as we assess alternative measures in light of potential threats, hostile intent and preventive defense."

In one scenario described in the paper, Palmore and Melese consider the outcome of two-player games in which one player is the United States; the other, an adversary. The object of the game, as stated, "is to drive the adversary to use weapons other than ballistic missiles without the U.S. deploying a national missile defense."

The logic is this, Palmore said: "If we build a defense which everybody including ourselves believed to be 100 percent effective against any single or small number of ICBMs launched with any warheads, then obviously one group is not going to spend money trying to launch an ICBM. They’re going to do one of the many other things. That’s the point that we raise in the paper: that protection is a placebo."

Because the proposed defense program is largely unproven and carries such a steep price tag, Palmore favors a go-slow approach over the rush to deployment – one that focuses on research and development and the examination of other credible alternatives.

"Everyone I talk to who thinks about these things is all for research and development," he said. "It's the deployment issue which is the main sticking point." - By Melissa Mitchell

21-Aug-2001

Copyright © 1995-2001 UniSci. All rights reserved.

rshowalter - 06:46pm Aug 21, 2001 EST (#7959 of 7969)
Robert Showalter showalte@macc.wisc.edu

Nice post, Cooper.

wrcooper - 07:33pm Aug 21, 2001 EST (#7960 of 7969)

rshowalter 8/21/01 6:46pm

I think the analysis is helpful. I don't understand why the authors think the government should proceed with R&D of a missile defense, given that they think one shouldn't ever be deployed. Strange logic there. The other consideration is that an adversary wishing to strike at the US would have a number of other less risky and expensive alternatives than launching a nuclear tipped missile. The limited BMD makes no sense, given the publicly available information in support of it. Is there classified information that would support a stronger case for the proposed system? I don't suppose we're ever likely to find out.

rshowalter - 08:14pm Aug 21, 2001 EST (#7961 of 7969)
Robert Showalter showalte@macc.wisc.edu

I think that if there were, there would be ways to find out.

In game theory, things depend crucially on the details of the "game" -- assumptions, etc. And you can "talk about anything."

A super obvious point is the use of nuclear weapons is not a "game" -- but a horror --- something we need to find ways to avoid.

Academics, because of their culture, have a tendency to be "all for R and D" . But the question "R and D for WHAT?" needs to be asked -- couldn't the people involved be doing other things?

There are MANY better things to do.

R and D is a kind of "investment decision" -- in the presence of risk, where one hopes (for enough "plays" ) for a payoff justifying the costs.

It is EASY to find "possible investments" that don't make sense, just as it is easy to come up with an almost limitless number of "experiments" that scientists wouldn't want to waste their time on. Lots of things can't work, in the ways that matter, for reasons that can be understood.

For R and D as an investment to work, the investments chosen have to make sense when they are carefully evaluated.

In the case of the missile defense proposals on the table, they don't make sense.

A key challenge is showing that, and showing it at levels that people can understand, on the basis of procedures they can respect.

We can do BETTER.

(8 following messages)

New York Times on the Web Forums Science Missile Defense