This chapter is extraordinarily voluminous since there is a multitude of estimates the evaluation of which seemed possible and desirable, some reaching back into the 18th century.
Standardized canton level surveys of marriages, births, and deaths began in the year 1867. The Federal Statistical Office, founded in 1860, was responsible for the handling and publication of the initially incomplete surveys. The results were included in the so-called Statistical Deliveries (“Statistische Lieferungen” / “Livraisons statistiques”) and published under the title “The Movements of the Population in the Year ...”. Today, this series is simply called “Population Movements in Switzerland” and is published by the Federal Statistical Office. Since 1876, this statistic also informs of divorces and causes of death at the canton level (also see the elaborations in the table part of the “Diseases and Causes of Death”). Format and content of the tables changed several times over the decades; and the Office appeared to struggle particularly with the proper presentation of deaths by age class at the canton level. Wherever possible, we attempted to fill the gaps caused by those frequent presentational changes. To this end, we used further statistical source books, on the one hand the Swiss Statistical Year Book and on the other the three large special publications issued between 1895 and 1928 by the Federal Statistical Office under the title of “Marriage, Birth, and Death” (“Ehe, Geburt und Tod” / “Mariages, naissances et décès”). The first of these massive volumes, which was preceded by a smaller publication on population movements in the years 1867–1871, informs on the development of marriages, births, and deaths in the era of 1871–1890. It consists of three separate studies, each of which has its own addendum. In those supplements, the Office added whatever it had available on marriages, births, and deaths for the time between 1801 and 1870. We depended almost exclusively on this data when constructing our own national statistical series, with the sole exception of additional materials for the cantons of Zurich and Appenzell Innerrhoden, which had not been considered by the Federal Statistical Office.
The second volume of “Marriage, Birth, and Death” provides information about population movements in the years 1891–1900, with certain series extended backwards to 1876, and at times even to 1871. The third volume, finally, covers the era of 1901–1920. Among others, we used it to learn about population movements in ten large and medium cities between the turn of the century and the beginning of the time between the world wars.
The abovementioned Federal Statistical Office reviews of “Marriage, Birth, and Death” of the years 1871–1890, 1891–1900, and 1901–1920 – which included volumes 103, 112, 128, 158, 170, 192, and 1928/4 of the series – were also incorporated into the “Statistical Deliveries” (later “Statistical Bulletins”, then “Statistical Sources”, and finally “Statistical Results”).
Our method of projecting federal level developments based on individual series at the canton level can be summarized as follows:
- In general, we began by establishing the end point of a given series of estimates. For a particular period of time for which there was data available for all cantons, we determined the portion of cantons whose governments had already performed surveys in the prior time period. We generally set the length of a time period to four years, with some exceptions where it is three or five years. We then computed the contribution for the cantons in question for each of the four years; the four quotients could not be spaced too far apart. Where this condition was not fulfilled, we did not include prior years in our estimates.
- The average of those four years was then applied to the previous period for which there was data only for part of the cantons. For that period we summed up the values of the available canton series and divided it by the average fraction which those cantons constituted of the total for all of Switzerland in the following four years.
- We divided the first four values of the national series of estimates arrived at by this method by the sum of the values of those cantons where data was available to us for the following time period as well. Again, we then determined the average value of those four quotients. By applying the resulting average quotient to the prior time period, we were able to extend the series of estimates backwards yet again. We repeated this procedure until we computed the starting point of the series of estimates. The following tables show the range of the statistics for each canton.
It should be emphasized that our estimates of stillborns and illegitimate childbirth is limited to the official figures. During the period of restoration, the number of unreported cases of stillborns as well as of illegitimate childbirth was probably much higher than in the later decades. It is therefore certain that both series of estimates do not accurately indicate the portion of stillborns and illegitimate births until approximately the middle of the century.
Processing of Canton Series into National Estimates, by Period
For the years 1869 and 1870, the statistics contained the national total of deceased men and women; for the years 1867 and 1868, the total had to be estimated. In order to do so, we first determined the total of deceased men and of all deaths for the 21 cantons in question. We then related the two sums and applied the value of the quotient (deceased males divided by all deaths in the 21 cantons) to the national level by multiplying the number of all deaths, available in the statistics for the years 1867 and 1868, with the value of the quotient. We used the same method to estimate the total of all male deaths in Switzerland for those two years. In order to determine the total number of female deaths, we only had to subtract the number of male deaths from the total of all deaths. Our further procedure can be explained without the need to differentiate between years: after the number of male deaths in each age category had been determined for the 21 and 22–24 cantons, respectively, each male age group was assigned a proportional value, the quotient of total male deaths and male deaths in the respective age group. The next step consisted of multiplying these quotients by the number of all male deaths in Switzerland (which, for the years 1867 and 1868, is an estimated value). The same process was used for female deaths. The addition of these two series yielded our estimate of all deaths by age group for the years 1867–1870.