In lieu of an abstract, here is a brief excerpt of the content:

Technology and Culture 42.4 (2001) 822-823



[Access article in PDF]

Book Review

Trust but Verify: Imagery Analysis in the Cold War


Trust but Verify: Imagery Analysis in the Cold War. By David T. Lindgren. Annapolis, Md.: Naval Institute Press, 2000. Pp. xiii+222. $32.95.

For some reason, books on cryptology apparently sell well. Go to your local Borders and look in the espionage section and you will find half a dozen fairly recent books on code breaking. I find this inexplicable, because numbers are (in my opinion) pretty darn boring. Cryptology is all about numbers, and it represents only one aspect of technical intelligence collection. There are lots of other ways to collect information without using spies. One of the most important is photo interpretation, an aspect of intelligence gathering that has been almost totally ignored by historians. It is a far more interesting exercise because photos are, well, fun to look at. David Lindgren's new book, Trust But Verify, attempts to make this point. It is an unfortunate failure.

Lindgren's thesis is that overhead images taken by satellites and aircraft have played a key role in arms control, not exactly new or riveting information. The book is largely a chronology of events in which aerial and satellite images of the Soviet Union played a role in superpower relations. At no point does Lindgren actually delve into the details of how imagery has affected strategic planning or arms control, nor does he provide much insight as to how photo interpreters--now called "imagery analysts"-- [End Page 822] actually get information off a piece of film. How do they tell the difference between one missile or another? How do they measure the size of objects? How do they thwart attempts to camouflage installations? How do they establish order of battle (the number and composition) of military forces? How do they fix the location of targets on the earth? What tools do they use? How have these tools changed over the years? How are advanced pattern-recognition algorithms and three-dimensional visualization changing the nature of imagery analysis?

These are the kinds of questions that it would be fascinating to have answered. The result of Lindgren's focus, however, is that we get little more information than has already been published elsewhere. His book does not tap the declassified literature on the subject in any serious or concerted way. And doubly unfortunate is his reliance on secondary sources that are not the best available: he uses Glenn Infield's erroneous 1970 book on the U-2 and not Chris Pocock's authoritative 1989 book, for example, and he employs neither the Central Intelligence Agency's official history of the program, declassified in 1998, nor Cargill Hall's landmark 1997 MHQ article on pre-U-2 aerial reconnaissance of the Soviet Union. Indeed, Lindgren has committed a cardinal sin of cold war historiography: failing to use secondary sources that benefited from the declassification of documents.

Then there are the errors, an unacceptable number in a work of this sort. Many are minor, but some--such as the tale that the U-2 spent much of its time gliding with the engine shut off (p. 34)--are not. ARGON mapping cameras did not "piggyback" on CORONA satellite missions (p. 103). The SR-71 did not perform secret mapping missions over China (p. 111), nor does Lindgren have the correct year for its retirement, in 1989, not 1992 (pp. 61, 111). William Casey was not the only head of the CIA to hold cabinet rank (p. 154). The KH-11 satellite photographed the Soviet space shuttle in 1983, not 1977 (p. 145). Lindgren averages one mistake every two pages, and on page 34 I counted six. The book falls apart because of all these holes, which will frustrate and anger the informed reader and mislead the uninformed. Imagery analysis is a subject that deserves much better treatment than it gets here.

Dwayne A. Day

 

Dr. Day is the 2001 Verville Fellow at the National Air and Space Museum.

Permission to reprint a...

pdf

Share