Popular Science Monthly/Volume 39/October 1891/Lessons from the Census I
By CARROLL D. WRIGHT, A. M.,
UNITED STATES COMMISSIONER OF LABOR.
WITH the statement of the total population of the country, and of each State distributed as to counties, cities, and towns, the popular interest in the Federal census begins to wane, and, as the results relative to the features other than merely of enumeration are obtained, the scientific interest increases. This interest is entertained by all classes of students: the economist desires immediate results as to production, wealth, debt, taxation, etc.; the social scientist is looking for statements relative to color and race, conjugal condition, the death-rate and health of the people, and facts covering various other relations; and the states-man and politician are anxious to secure comparisons of the growth of population, the changes incident to new productive enterprises, the concentration of wealth, and all the other expansive elements which concern the great discussions in which they are engaged. Under the new census, the eleventh, the interest of other bodies is brought into activity. The question as to whether the homes and farms of the country are owned by the occupants, and the extent to which they are mortgaged, as well as the psychological reasons for incurring mortgage indebtedness, serves to interest, and in a most lively way, the student who is sociologically inclined. The enumeration of the surviving soldiers and the widows of deceased soldiers of the war of the rebellion brings into play not only the interest of the veterans themselves but of the legislators of the country, and, in addition, the sentiment of the whole community. All these various features of our Federal census excite the interest of the people on a broader scale and in more thoroughly scientific directions than would the enumeration of the people alone. Crystallized statements, and a somewhat popular analysis of the results of the census as they come out, must be of more or less value, and from them many lessons may be drawn. A series of articles, therefore, comprehending comparisons and analyses, and bringing out the salient points in all the vast quantity of material digested by the Census Office, can not fail to interest various elements of the population. It seems wise, however, before entering upon a discussion of the statistical features and of the scientific results of the census and the lessons to be drawn therefrom, that these first two chapters should be devoted to the system under which all these various results are obtained.
The United States census finds its organic authorization in the Federal Constitution, in accordance with Article I, section 2, as follows:
"Representatives and direct taxes shall be apportioned among the several States which may be included within this Union, according to their respective numbers, which shall be determined by adding to the whole number of free persons, including those bound to service for a term of years, and excluding Indians not taxed, three fifths of all other persons. The actual enumeration shall be made within three years after the first meeting of the Congress of the United States, and within every subsequent term of ten years, in such manner as they shall by law direct."
The only other reference in the Constitution to a census is in section 9 of Article I, wherein it is provided that "no capitation or other direct tax shall be laid, unless in proportion to the census or enumeration hereinbefore directed to be taken." This organic provision for a periodical census was the first of its kind in any country. It was the result of a good deal of discussion by the framers of the Constitution, and grew out of the difficulties which they experienced in apportioning representatives and taxation. There were wide differences of opinion in the Constitutional Convention; but after much deliberation the majority settled upon the form of language just quoted, and it became a part of the organic law of the land. There had been, prior to the adoption of the Constitution in 1789, various colonial and local censuses, and foreign countries had made enumerations at intervals of time; so that the framers of the Federal Constitution were not particularly unfamiliar with the benefits of census-taking; but the credit of the first regularly organized periodical census is due to the United States, and this country has had, commencing with 1790, regular enumerations of population, and since and including 1850 what might be denominated national censuses, comprehending various features other than the mere enumeration of the inhabitants of the country.
The general direction of the census was placed in the hands of the Secretary of State, where it remained until the passage of the census law of May 23, 1850, when all the functions of census-taking were put in charge of the newly created Department of the Interior, and all census laws since and including that have been administered under the direction of the Secretary of the Interior. The first schedule, that for 1790, was a very simple affair, and was as follows:
In 1800 the scope of the population schedule was enlarged somewhat, and it was used in the following form:
The foregoing schedule was used in 1810 without change, but the scope of the census, by act of May 1, 1810, was enlarged. This act required the several marshals, secretaries, and their assistants, "at the time for taking the census or enumeration aforesaid, to take, under the direction of the Secretary of the Treasury, and according to such instructions as he should give, an account of the several manufacturing establishments and manufactories within their several districts, territories, and divisions"; but no schedule was incorporated into the law, the whole matter being left to the discretion of the Secretary of the Treasury as to construction. No very valuable results were obtained under this provision; yet the experiment was repeated in 1820, under the law of March 14th of that year, which made it the duty of the "several marshals and their assistants, at the time for taking the said census, to take, under the direction of the Secretary of State, and according to such instructions as he shall give, and such forms as he shall prescribe, an account of the several manufacturing establishments and their manufactures, within their several districts, territories, and divisions." These attempts to enlarge the census so as to include statistics of manufactures were so barren of valuable results that in taking the census of 1830 the attempt was wholly abandoned. The manufacturers' schedule was, however, introduced again in 1840; but still the effort was of little account, and it was not until the census taken under the broader law of 1850 that any valuable results were reached. So the census grew, from the simple enumeration of the people in 1790, under the very brief schedule which has been given, not only through various additions to the population schedule, but by the introduction of inquiries relating to production, until nearly all the material conditions surrounding the people, their industries, their wealth, their taxation, their carrying trade, fisheries, schools, health—nearly everything, in fact, that the social scientist wishes to know concerning the people and their conditions—are embodied in the census. With the changes in the scope of the census there have been changes in the methods of enumeration. Until 1880 the United States marshals for the several districts were authorized, under the first census law, approved May 1, 1790, to make the enumeration, the marshals having power to appoint as many assistants within their respective districts as should appear necessary. The enumeration was to commence on the first Monday in August, 1790, and it was provided by law that it should close within nine calendar months thereafter. The marshals were required to file the returns with the clerks of their respective district courts for careful preservation, and to forward the aggregate amount of each description of persons within their respective districts to the President of the United States. Each assistant marshal was required, previous to making his return to the marshal, to cause a correct copy of the schedule, signed by himself, to be set up at two of the most public places within his division, there to remain for public inspection. The use of United States marshals in taking the censuses was continued until the tenth census, that of 1880, although an effort was made to provide other machinery for the ninth census, that of 1870. This effort was embodied in a bill which passed the House of Representatives, but which was defeated in the Senate; so the census of 1870 was taken in accordance with the law of 1850. The act for the tenth census, approved March 3, 1879, provided that enumerators, specially appointed for the purpose, should make the canvass, the body of enumerators working under supervisors appointed by the President. Under this law there were one hundred and fifty supervisors, under whom were employed about thirty thousand enumerators, and under the last-named law the scope of the census was greatly expanded, nothing like it ever having been carried out under any government. The enumerators were to make the canvass from house to house, from manufactory to manufactory, from farm to farm, sending their certified returns to their respective supervisors, who, after careful examination, transmitted the originals to the superintendent at Washington. By this method the central office was put in possession of the original data, avoiding thereby all the errors incident to transcription under previous methods. General Francis A. Walker was the superintendent, and he was also the superintendent of the ninth census, that of 1870. He carried through the census of 1880 on the broad basis laid down by him, and the results, in twenty-two folio volumes, secured for him the admiration of statisticians in every part of the world.
Now comes the eleventh census, the centennial of that of 1790. It is being conducted under a law which is practically the re-enactment of that providing for the tenth census. The law was approved March 1, 1889, and under it Robert P. Porter was appointed superintendent. The schedules are practically those of the tenth census, enlarged, amended, and improved as experience under the tenth census indicated the necessity.
The growth of the Federal census, while clearly illustrated by reference to the various laws, is perhaps more strikingly shown by a statement of the various publications which have resulted from the several censuses. General Garfield, chairman of the Committee on the Ninth Census, made in January, 1870, a very elaborate report covering census-taking in different countries in the world, and especially a history of the United States censuses; and from this report the exhibit relative to publications and expenses up to and including 1860 is taken. For 1870 and 1880 recourse has been had to original sources:
1790.—Return of the Whole Number of Persons within the Several Districts of the United States, etc. This first census publication was an octavo pamphlet of 52 pages, published in 1792. The entire cost of this census was $44,377.18.
1800.—Return of the Whole Number of Persons within the Several Districts of the United States, etc. This was a folio of 78 pages, published in 1801. The cost of this census was $66,609.04.
1810.—The report of this census was in two folio volumes: I. Aggregate Amount of each Description of Persons within the United States, etc. This was an oblong folio of 90 pages, but it does not show the date of publication. II. A Series of Tables of the Several Branches of American Manufactures, exhibiting them in every County of the Union, so far as they are returned in the Reports of the Marshals and of the Secretaries of the Territories, and of their Respective Assistants, in the Autumn of the Year 1810; together with Returns of Certain Doubtful Goods, Productions of the Soil and Agricultural Stock, so far as they have been received. Quarto, 170 pages. Edited by Tench Coxe, and published May 30, 1813. The cost of the census of 1810 was $178,444.67.
1820.—I. Census for 1820, etc., a folio of 164 pages, published in 1821. II. Digest of Accounts of Manufacturing Establishments, etc., a folio of 100 pages, printed in 1823. Cost of the census, $208,525.99.
1830.—Fifth Census or Enumeration of the Inhabitants of the United States. This volume was a large folio of 163 pages, printed in 1832. This report was so wretchedly printed that Congress required by law a republication, which was made the same year, under the immediate direction of the Secretary of State. The erroneous and corrected editions are bound together. This republication enhanced the cost of this census to $378,543.13.
1840.—I. Compendium of the Enumeration of the Inhabitants and Statistics of the United States, a folio of 379 pages, printed in 1841. II. Sixth Census or Enumeration of the Inhabitants of the United States, a folio of 470 pages, 1841. III. Statistics of the United States, etc., a large, oblong folio of 410 pages, 1841. IV. Census of Pensioners for Revolutionary and Military Service, with their Names, Ages, and Places of Residence, etc. Quarto, 196 pages. The total cost of these censuses was $833,370.95.
1850.—I. The Seventh Census of the United States, quarto of 1022 pages, 1853. II. Statistical View of United States, octavo of 400 pages, 1854. III. Mortality Statistics of the Seventh Census, etc., octavo, pp. 304, 1855. IV. Digest of the Statistics of Manufactures, octavo, pp. 143, published in 1859 as Senate Executive Document, No. 39, second session Thirty-fifth Congress. Cost up to September 30, 1853, $1,318,027.53. There were three or four subsequent appropriations for this census, amounting to about $11,000.
1860.—I. Preliminary Report of the Eighth Census, 1860, octavo, 294 pages, 1862. II. Final Report, in four quarto volumes, as follows: Vol. I. Population, pp. 694. Published in 1864. Vol.
II. Agriculture, pp. 292. Published in 1864. Vol. III. Manufactures, pp. 746. Published in 1865. Vol. IV. Mortality and Miscellaneous Statistics, pp. 548. Published in 1866.
The total cost of the eighth census was, in round numbers, $2,000,000.
1870.—The results of the ninth census were embodied in three quarto volumes, as follows: Vol. I. Population and Social Statistics, pp. 875. Published in 1872. Vol. II. Vital Statistics, pp. 705. Published in 1872. Vol. III. Statistics of Wealth and Industry, pp. 849. Published in 1872.
The cost of the ninth census, including printing, was $3,696,227.37.
1880.—The results of the tenth census made an encyclopedic report of twenty-two quarto volumes and a compendium.
The cost of the tenth census, including printing, was $5,862,750.24.
The eleventh census, that for 1890, is being taken under the act approved March 1, 1889, and comprehends all the features of the tenth, with two great additions—an account of the mortgage indebtedness on homes and farms, and an enumeration of the surviving veterans of the war of the rebellion. The machinery of the census is practically the same as that organized for the tenth. The forces employed, however, are vastly greater, numerically considered. The number of supervisors was increased to 175. The whole body of enumerators constituted an army of 46,546. The largest number of office employés, not including special agents, was on May 9, 1891, 3,142; and the number of special agents, including special agents on manufactures in cities, was 1,938, or a total force June 1, 1891, exclusive of enumerators, of 5,080. The appropriations, up to July 1, 1891, have been $7,400,000. To complete the work there will be required another and quite large appropriation. The organization of the Census Office, January 1, 1891, comprehended twenty-five specific divisions, each division being under the charge of a chief of division or an expert special agent. These twenty-five divisions are charged with the business features of the Census Office, and the collection and tabulation of the facts relating to the subjects indicated in the following list:
The progress of this vast work is probably at the present writing in as forward a state as could be expected, when the volume of data called for, as indicated, is considered. The results showing aggregate population by States, counties, and minor civil divisions, and by sex; condensed classification by ages, showing the school, militia, and voting ages, by native and foreign white persons and colored persons will be put into compendium form and published, without much doubt, before the close of the present calendar year. The classification regarding families and dwellings, the volume of final reports for population, showing the results in detail, by ages, conjugal condition, place of birth, and all the varied distinctions of population, must not be expected until some time in 1892, possibly by the early summer. All this work is enough for the Census Office to handle at one time; but when there is added to it the multitudinous divisions shown in the foregoing list, it is not to be wondered at that progress is slow, that the country criticises, and that increased appropriations are called for. No superintendent, burdened with the present system, can possibly satisfy the country, Congress, or himself. And so the first lesson to be drawn from the census relates to the system rather than to the results under it, and to what changes are needed that the system may be improved.
A recent find of mummies of dogs in Egypt has prompted M. Maspero to suggest that these objects may furnish opportunities for studying the characteristics of the most ancient of the domesticated species. Some efforts have been made to determine these from the wall-paintings, but the data they afford are very uncertain.