THE AMERICAN JOURNAL OF SOCIOLOGY
article in the North American Review? Though these conclusions iulhall as to the increase of wealth and the increase of wages in the United States have been widely quoted and almost univer- sally accepted, and though works of reference compiled from our census report containing similarly erroneous deductions, and also newspaper articles of like tenor have been widely circulated, we have heard from the census office no word of warning. 2 3
The error of Mr. Mulhall, and the very many who have arrived at similar conclusions, has been their failure to ascertain and comprehend the radical difference in the methods of compi- lation of the statistics between our earlier and more recent cen- sus reports. As the true character of these statistics may only be ascertained through an extended investigation, this error seems more excusable in others than in officials of our census bureau.
A fair specimen of the contributions of the United States gov- ernment to social science is furnished in a contribution by one of these officials in the January number (1896) of the Bulletin of the Department of Labor: "The Wealth and Receipts and Expenditures of the United States Government," by Wm. M. Steuart. From Colonel Wright's letter of transmittal of the
'June 189.5, "Wealth and Power of the United States."
a ln the Journal of Political Economy (December 1895), the writer of the present article demonstrated the error of Mr. Mulhall's conclusions and the utter worthlessness of these census statistics for about the only purpose for which they are ever used, that of comparison.
3 For the satisfaction of readers who are inclined to accept authorities rather than the investigations of those occupying no official position, the writer presents the fol- lowing letter from one who, besides being an able student of economic questions, occupies the high office of secretary of treasury :
TREASURY DEPARTMENT, OFFICE OF THE SECRETARY,
WASHINGTON, D. C, April 23, 1896. MR. II. L. BLISS, Chicago, 111.,
My dear Sir: I read your article on the " Use of Census Statistics " with a great deal of interest and fully agree with what you say upon the subject. In my opinion Mr. Mulhall is a very unreliable statistician and economist. I at one time commenced the preparation of a paper upon the same subject myself, but it was never completed and of course never published. Very truly yours,
J. G. CARLISLE.
