work 'Six Centuries of Work and Wages' contains several instances of this pernicious use of percentages. So, also, many are constantly using the unscientific conclusions drawn from concomitants—because one thing exists another logically exists contemporaneously.
These illustrations show the necessity of making statistics, the statistical method, thoroughly scientific in all directions, but this scientific conception of statistics does not warrant statisticians in using algebraic formula or in resorting to the calculus to secure results. The results under such methods rarely vary from those secured by the simple common-sense method of statistics; on the other hand, they disturb the reader and the common mind cannot understand them. All statistics should be simple, straightforward and clear, and any confusing element which disturbs this clearness is detrimental to the real purpose of the statistical method. While there are very many illustrations going to show the misuse of the method, nevertheless all right-minded men understand that, in economic questions especially, the comparative and historical method of study is the correct one, and the concrete, historical and comparative method is best carried on through the scientific use of statistics.
By ROLAND P. FALKNER,
THE problem of definitions is a perplexing one, and some of the most distinguished writers on political economy make no attempt to introduce their treatment by a definition. Marshall, for instance, tells us 'Political economy is a study of man's actions in the ordinary business of life; it inquires how he gets his income and how he uses it.' A definition is, at the best, but a guide-post whose accuracy cannot be tested until the path has been tried. Its principal service is as a preliminary line of demarkation in comparison with other things supposed to be more familiar than the field to be entered upon. One definition which has become commonplace is that political economy is the science of wealth, and if we abstain from any discussion of what constitutes a science and what constitutes wealth, it may be deemed measurably satisfactory.
After all, the phraseology of a definition is far less important than the matter which follows it, and while economists, in common with the devotees of other studies, have not been lacking in verbal controversies as to the definition of their subject, the subject matter of their discussions, apart from mere questions of emphasis, has been in the main identical. The question of definition resolves itself, therefore, into an