It has come to be considered high praise of any field to call it a science, and conversely, few epithets are stronger than unscientific. Part of this prestige of science is of course due to the great achievements of the natural sciences, and a further explanation is provided by the typical looseness of popular usage of language: unscientific has become a synonym for sloppy. But without attempting to deprecate science in general, it is worth mentioning that the grounds for the prestige of science are nebulous. Surely usefulness is not the criterion, for cooking is more useful than astronomy; nor is aesthetics, for most men prefer music to chemistry. The difficulty of the subject matter of sciences is undoubtedly a factor, but here two things may be said. On the one hand, no subject is difficult to those with aptitudes and training in that direction--a modern Gauss could master higher mathematical analysis much more easily than an ordinary city dweller could learn to farm. And on the other hand, difficulty is a most ambiguous concept--there are many more men who can understand modern physics than can run 100 yards in 9.8 seconds. In sum, to say that economics is a science is a description, not an encomium.

The purpose of this paper is to outline a general method for deriving the survival coefficients for depreciable physical property. By a survival coefficient is meant the percentage of original units of a good acquired in a given period, that may be expected to survive to a given subsequent period. Of course, the difference between a survival coefficient and 100 percent is the mortality coefficient, or the percentage of original units that may be expected to "die" in a given period. The application of the survival or mortality coefficients will result in a table describing the anticipated retirement schedule of a given physical good. This type of information can obviously be of considerable value in constructing depreciation schedules, capital budgets, and similar programs that are of interest to accountants. This paper consists of two parts. First, it illustrates a mechanical procedure for calculating the survival coefficients based on the above two parameters, average service life and maximum service life, both statistics often being obtainable from market research or engineering studies. Second, the mathematical appendix in which the underlying formulas are developed for those who desire them.