FRINGE Centre blog series: S for Statistics

By Blog Admin, on 22 January 2016

In the latest entry on the FRINGE Centre blog, Tomáš Cvrček of UCL SSEES considers statistics and their shortcomings. 

Breaking somewhat with the run of blog posts on big intellectual words, here is an entry about statistics, a mundane yawn of a word that starts with S. Where does the letter “S” come in the word “FRINGE”? It does not, although it could perhaps be appended at the end, making it plural. There are many ways in which things can stand on the fringe and one of them is the frontier of measurability. In line with the other themes in the acronym – such as invisibility, elusiveness and grey zones – the letter S can then stand for things that are somewhat in the statistical shadow, out of the gaze of the data collector.

tomas stats


To open with a confession, I think that data, numbers and statistics are a wonderful thing. They can tell us a great deal about lots of things that people are doing. When used properly, they can help one distinguish what is random and what is systematic. At the same time, as primary sources on various social phenomena, data have their limitations but that does not make them useless – rather, the limitations themselves are interesting.

Consider this. In the Austrian province of Carinthia (or Kärnten in German), there is a small town called St Veit an der Glan that is also the seat of a district of the same name. There is nothing particularly special about it except the fact that, in the mid-to-late 19th century, it arguably held the European record for the rate of out-of-wedlock childbirth. In the late 1860s, the district reported that more than two thirds (two thirds!) of its 1500 newly born babies were born out of wedlock. The neighbouring district of Klagenfurt (which also contained the provincial capital) reported almost 60% of illegitimate children. In fact, a stripe of high-illegitimacy districts and towns stretches from Styria through Carinthia and Salzburg into Bavaria. German demographers have spent some time analyzing the Bavarian illegitimacy rates which in some cities, such as Munich, reached 25% – but Munich is really the boring end of the issue: the epicenter of this phenomenon was Carinthia, and specifically St Veit an der Glan.

And here is where we come to the notion of data limitations, for every social scientist’s first reaction will be disbelief: surely, a 66% illegitimacy rate is a typo? That statistic is just way too high for the staid and prudish 19th century – for even nowadays, after the sexual revolution, it is not common (possible – yes, but not common) to get such high readings on illegitimacy. After all, social phenomena get mismeasured and misreported all the time. It is nice to have things recorded in a statistical fashion but when the statistics start throwing up figures that are clearly beyond belief, some skepticism is in order.

It turns out, however, that these high figures are unlikely to be typos/mismeasures. Various districts in the province of Carinthia kept reporting elevated illegitimacy rates for the better part of the 19th century. The peak came in the three decades from mid-1860s to mid-1890s, but even before, the province stood out. A typo that lasts across several generations is even less believable than high rate of out-of-wedlock childbearing.

But perhaps, the numbers misrepresent the actual operation of the local marriage market. Perhaps the local culture considered betrothal almost as good as the actual marriage, and so some conceptions and births occurred before the actual wedding even as the whole community considered the couple as good as wed?

This argument runs into some other data that we have from the province: the Austrian government recorded what were called legitimations, i.e. number of children who came to be legitimate through the eventual marriage of their parents. Some Habsburg provinces, such as Bohemia, legitimized illegitimate children very avidly, so that of the roughly 15% of children born out of wedlock every year, about a third were legitimated through a shot-gun marriage, another third died before age 5, with the remaining third living out their lives as “bastards”. In Carinthia, however, we do not see much effort towards ex-post legitimation, which is what we should see, if the whole conception-during-betrothal story were to pan out. By age 5, only about one sixth of the illegitimate children were legitimated while more than a half of these children remained illegitimate for life.

But perhaps the local culture offered some as yet unknown close substitute to marriage – an institution or an arrangement that was similar to it but not recognized as such by the law.

Again, records say otherwise. Carinthians did get married (if reluctantly) and that included even those who already had several children out of wedlock. By the testimonies of local priests, some brides were accompanied by as many as three children to the altar (which, by the way, roughly corresponds to the estimated out-of-wedlock fertility rate of Carinthian women). If there were a close (though legally unrecognized) substitute for marriage that Carinthians were happy with, why would they then make the step of actually getting married?

From the statistics alone, it is unclear what really went down in Carinthia. Using the data available, we can observe the end-result of some complex decision-making but we are left in the dark as to what that decision-making process involved. It goes to show that even when a practice is a majority practice, and a very visible one at that, it can still somehow be on the fringe: the fringe of respectability, the fringe of scientific attention, the fringe of what statistical information can capture.

Note: This article gives the views of the author(s), and not the position of the SSEES Research blog, nor of the School of Slavonic and East European Studies, nor of UCL.