Logo IMG
HOME > PAST ISSUE > Article Detail


Cracking Up

To the Editors:

I found Henry Petroski's essays "Lab Notes" and "Why Things Break" on fractures and failure of metals (Engineering, March-April and May-June) interesting and informative. Such failures, called by Dr. Petroski "the bane of engineering," relate to an important subject long of interest to me: traumatic injuries. Unintended injuries are the most common cause of death among persons 1 to 34 years old, and the fifth most common cause of death among the population as a whole.

One may think about serious injuries in the same way as serious illnesses, except that injuries usually involve speeds of onset and magnitudes of affected tissues exceeding those of the illnesses by orders of magnitude. Injuries as well as illnesses are more easily understood, especially in terms of prevention, if one can identify their major antecedents and determinants. Among the latter are environmental factors, patients' characteristics and exposures, the kind of energy involved and variables related to time. Dr. Petroski's thorough descriptions of two railroad bridge collapses indicate that the catastrophic effects of metal fatigue may have antecedents in engineers' decisions and industrial processes as well as the environmental factor of a severe storm.

Dr. Petroski uses the word "accident" in several ways. A problem with that word is its ambivalence and several meanings, especially the meaning involving chance. I have written in the New England Journal of Medicine and the Journal of the American Medical Association that the term "accident" should not be used in studying injuries, because few if any injuries are due to chance events or random occurrences. I have even threatened quite a few graduate students with dire consequences if they talk about "accidents," unless they are speaking of selecting numbers from a table of random numbers or some other truly randomly-occurring event. After all, if catastrophes and injuries really were "accidents," would there by any point in investigating them?

Theodore C. Doege
University of Illinois

comments powered by Disqus


Other Related Links

Lab Notes

Why Things Break

Subscribe to American Scientist