The topic of structural damage identification is-for good reasons-of topical concern in structural dynamics. Indeed the theme of the 1997 International Modal Analysis Conference (IMAC) is structural damage assessment. The majority of research studies attribute the basic ideas to Cawley and Adams [1].
The mathematical rationale for modal methods for damage detection can be recognized directly from the Rayleigh quotient v = f t Kf/f t Mf. The most common assumption in damage identification is that there is a local degradation in stiffness but no variation in mass. Thus, when utilizing the Rayleigh quotient, the natural frequencies may-in general-be expected to decrease progressively as the structure degrades. When used uncritically this assumption can give rise to absurd results. For example, Williams and Salawu [2] reported a testing programme of a concrete bridge where natural frequencies decreased after repairs. A tentative explanation offered was the possibility that the moisture content of the concrete-and hence its mass-had varied during the course of the test. The dates of testing (autumn/fall-before repair, and spring-after repair) tend to support this conjecture.
The common feature of the overwhelming majority of modal methods known to the author is that the location and severity of damage is identifiable by quantifying changes in frequency and mode shape. The purpose of the current note is to question the whole rationale for this approach.
Aficionados of the Sherlock Holmes stories will know the story of Silver Blaze. Holmes recognized that the reason that a guard dog did not bark in the night was because it recognized the criminal and therefore felt no need to raise the alarm: ''Is there any other incident to which you would wish to draw my attention?'' ''To the curious incident of the dog in the night-time.'' ''The dog did nothing in the night-time.'' ''That was the curious incident '', remarked Sherlock Holmes.