Dr. Rebecca Slayton's treatise Arguments That Count offers an in-depth look at systems-engineering issues that arose early during the Cold War when missile defense became a national imperative: the fact that computers are complex and that it takes time to program and work the kinks out. Dr. Slayton's background began in physical chemistry, but her take on the missile defense debate within the scope of technology that defines the system is well thought out and expertly researched. She does not play the role of defense-policy wonk in her writing. From this reader's point of view, the book does not contain an obvious agenda that either supports or castigates missile defense. The evidentiary support pulled by the author throughout the last five decades, however, shows the requisite level of criticism needed for the amorphous "missile defense" system, ranging from the nuclear-tipped Nike-X missiles and the Safeguard system to the Gulf War Patriots and today's National Missile Defense interceptors. Early in the introduction, Dr. Slayton asks a resounding question: "If complex software poses such obvious risks, why does the United States continue to rely so heavily upon complex weapon systems, rushing untested technology into the field?" (p. 2). Why, indeed?
Slayton provides a compelling argument as a secondary thesis--that computer engineering and practitioners of the "black art" be held accountable to the men and women whom their software attempts to protect. The text cites many examples of failures, the most recent being those of the Patriot missile batteries during Operation Iraqi Freedom. Problems with identification, friend or foe during that operation caused the shoot-down of two aircraft and the death of three aviators. Issues with software complexity in highly automated systems had been a common concern among developers since the early days with the Air Force's semi-automatic ground environment (SAGE).
The author details how designers had virtually ignored the information-processing aspect of the system. Manpower shortages within the halls of IBM caused a horrendous lack of planning forethought in the programming design. Gen Bernard Schriever's industrial-engineering revolution of "concurrency" failed to meet expectations in SAGE's software world where "each computer and system grew more complex than its predecessor, yet there was no time to complete [testing] . . . before beginning work on the next [revision]" (p. 29).
The Presidential Scientific Advisory Council, led by Dr. Jerome Wiesner, gave a wholehearted "thumbs-up" to the Nike-Zeus interceptor system, stating that the "system appears to have been well designed from a data processing point of view" (p. 57). Slayton recounts that no mention of computing aspects in the missile defense architecture occurred for over a decade afterwards. One unintended effect from the institutionalization of software engineering within the Department of Defense, independent of the missile defense threat, is the shift of understanding from decision makers who know little about software and its complex development cycles to "laypersons" who believe they understand the problem set.
Regarding any drawbacks to the book, the subtitle's inclusion of the word physics seems a bit misleading. The text effectively frames the importance of complex software development for critical national defense systems but fails to deliver on criticism of the physical aspect. As a space operator, this reader realizes that they are intertwined. In terms of the physics aspect of the book, perhaps Dr. Slayton followed the advice of Richard Garwin, a critic of the Strategic Defense Initiative and senior scientist at IBM: "The systems would fail for physical reasons, so why should I waste my time on software?" (p. 6).
The shortcoming of mildly ignoring the physics aspect is that billions of dollars of taxpayer money has been spent on missile defense by decision makers who believe that a defense is possible. Full analysis of the physics element, along with the computer engineering, may draw a few more believers toward the reality of missile defense. The limited scope of the argument, perhaps, nullified the need for a discussion of physics.
A short read with 98 pages of references, the book offers a historical view of the computer-engineering problems required by complex systems--and therein lies its greatest value. Space and cyber operators as well as acquisition professionals should read Arguments That Count for the knowledge that all complex systems have inherent problems and that by learning lessons from the past, we can overcome these obstacles.
Maj Joseph T. Page II, USAF
Joint Space Operations Center
Vandenberg AFB, California