Constructing an Empirical Test

Naturally, constructing an empirical test of the proposition that cathedral building was an entry-limiting device created by the Church is difficult at these distances in time and data availability. We believe nonetheless that some preliminary evidence might be assembled. Sociologist Guy E. Swanson provides the dates and countries of "final settlement"—that is, the date and place where particular locations either converted to Protestantism or remained Roman Catholic.71 We use a simpler form of a dependent variable—whether a country remained Catholic or converted to a Protestant form of religion irrespective of the date that they settled in our empirical test.

In terms of the independent variable, several variables might capture the concept of a cathedral's "capacity"—the ground plan area (measured in square feet) and/or the interior height (in feet). Our source of information is not systematic; it is a glimmering of data presented by Paul Johnson, who reports height and square-footage data on cathedrals in Britain and elsewhere on the continent of Europe. These data are reported in table 8A.2. We do not know of any systematic collection of data on the dimensions of cathedrals.

We note that the height of cathedrals is as critical in measuring "awe and grandeur'' capital as is the actual capacity of a church. We believe that height is relevant, in other words. As Johnson notes:

Table 8A.2

Height and square footage of selected medieval cathedrals

Internal height (from pavement to apex of vault)

Old St. Paul's 103 Cologne 155

Westminster 103 Beauvais 154

York 102 Bologna 150

Gloucester 86 Amiens 144

Salisbury 84 Bourges 117

Lincoln 82 Chartres 106 Ground area (square feet)

Old St. Paul's 72,460 Seville 150,000

York 63,800 Milan 92,000

Lincoln 57,200 Saragossa 80,000

Winchester 53,480 Amiens 70,000

Ely 46,000 Cluny 66,000

Westminster 46,000 Toledo 66,000

Source: Paul Johnson, British Cathedrals (New York: Morrow, 1980), 19.

The cathedrals were almost out of scale with their times; it is impossible for us to conceive the extent to which, in a literal sense, they loomed large in the lives and vision of medieval people, reducing all else (including themselves) to insignificance and providing inescapable and ocular evidence of the power of majesty of God. They were awesome and overwhelming.73

These structures, which gave "ocular evidence'' of the power of God, also provided evidence of the power of God's only recognized representative on earth during medieval times—the Roman Catholic Church. Thus, in addition to square feet as a measure of physical capacity, some "magisterial quotient'' in terms of height and decoration is relevant to perceptions of potential entrants. We use height as a simple measure of this factor.

Some simple regressions can be performed with our data. The dependent variable is CATH, a dummy variable which takes the value of 1 if the country remained Catholic and 0 if the country converted to Protestantism. There are two explanatory variables: HIGHFT is the height in feet of cathedrals in the country and GROUNDFT is the ground plan area in square feet. Using OLS (defined in appendix 4A), we obtain the following results. Both coefficients are positive and significant at better than the .01 level, as shown below in tables 8A.3 and 8A.4.

The coefficients on area and height are significant in both regressions and are consistent with larger cathedrals acting as an entry barrier. Overall, the regressions explain more than 60 percent of the variation in

Table 8A.3


Table 8A.3





P >|t|









R-squared = 0.6843

Number of obs. = 12

Table 8A.4





P >|t|









height and size in this sample. We also ran more sophisticated regressions models, and although we do not have sufficient data for these exercises, we continue to find positive signs on these variables.

Our data for Protestant cathedrals is British and that for Catholic cathedrals is European. Supplementing this data with data for other Protestant cities in Britain and elsewhere with more modest cathedrals would strengthen our results.

There are cases, such as Germany and Switzerland, for example, in which the entire country or area did not change its religion. Parts of these countries clearly remained Catholic. Our theory would predict that those areas or regions that became Protestant would contain smaller and shorter Catholic churches. Unfortunately, our data does not permit such an analysis since we have no Swiss churches and only one German church (Cologne) in our data set.

Finally, we present a visual picture of our results in figure 8A.2. There, the Catholic and Protestant cathedral heights in feet are plotted, clearly

Figure 8A.2

Height of selected British and European cathedrals xo<* J? xo<* ^ ^ #

Figure 8A.2

Height of selected British and European cathedrals showing the effect we postulate. Areas with higher cathedrals, represented by gray columns, all remained Roman Catholic. Thus, we suggest that while medieval cathedrals served many purposes and, indeed, were some of the greatest technical achievements of their time, they served a rational economic purpose as well. Protestant entry into the market for Christian religion finally materialized in the early sixteenth century. The Roman Catholic Church did not make a mistake in failing to forestall entry. Indeed, we argue that the Church made a conscious rational effort to do so. By supplying excess capacity and "awe and grandeur'' capital in the form of medieval cathedrals, the Church attempted to signal potential entrants that entry would be unsuccessful. For a number of reasons, some of them relating to the pricing structure of Roman Catholic Church services, the attempt to forestall entry for longer was unsuccessful. Cathedral building in certain locales was, as we have argued with some preliminary evidence, one reason why some areas of Europe remained Catholic and others were lost to Protestantism, as well as a reason that Protestant entry did not materialize until 1518 and after.

0 0

Post a comment