The following is the second part in a series exploring themes related to the Christmas Flood of 1717. These maps and reflections are works in progress and reflect the status of the project at the time of the European Society for Environmental History’s Biennial conference in Zagreb, Croatia, 2017.
In the limited time available during conference presentations, it is often difficult to explore the full scope of the subjects discussed in accompanying papers. This was the case with the presentation I briefly outlined in my previous post about the Christmas Flood of 1717. The paper introduced a side-project I have been working on related to a chapter from my manuscript on eighteenth-century Dutch disasters. My motivation to explore this flood in comparative perspective grew from my desire to understand how the Dutch (specifically the Groninger) experience of disaster compared with that of its neighbors. As bad as natural disasters in the early part of the 18th century were for the Dutch Republic, they were far worse in East Frisia. But how to explore and visualize this difference? The experience of the Christmas Flood in East Frisia was a far more devastating, long-lasting affair than in Groningen. Did the divergence begin with the onset of the flood? One of the early strategies I used to explore these questions was to map the flood losses by municipality in Groningen and East Frisia. To my knowledge, this hasn’t been attempted and it revealed some striking geographic differences (and similarities).
Chief among these were the ways that flood victimhood had been distributed across the province and the principality respectively. Much of this information (and the maps) had to be left on the chopping block for the conference, so I will add a few final images for reflection.The first set of maps compare human fatalities. The underlying boundary shapefiles are loose approximations of actual administrative boundaries (less loose in the case of Groningen), but they give a general sense of the geographic distribution of victims. Most are concentrated along the Wadden Sea coast, less along the Dollard. This data is taken from a variety of secondary and primary sources. Areas near dike breaches naturally experienced the greatest fatalities. There is little surprising about this map, although they might be improved with accurate municipal population figures to give a clearer sense of the relative severity of the flood (something I haven’t been able to find so I’m relying on raw data). The takeaway from this map is that the experience of disaster on a provincial level is surprisingly similar between East Frisia and Groningen. In both areas, we see highly variable impacts, though in both cases, they are concentrated along the Wadden Sea coastline, largely centered on sea clay landscapes, populated by (relatively) well-off farmers.
Compare this map to a similar series showing loss of housing and livestock. Subtle differences appear between human fatalities and property loss maps. Although the coasts remain hardest hit, the flood washed away a significant amount of property in the interior of the province as well. In Groningen in particular you see this trend develop. Why the difference? The areas to the north and east of the city of Groningen had been drained and were slowly subsiding, leaving the interior of Groningen bowl-shaped. In addition to the “young” and “old” sea clay polders (blue and green respectively), we see heavy losses in the reclaimed peat bog areas to the north and east of the city of Groningen (burnt orange). A notable exception is in the Western Quarter of the province which experience minimal losses (a subject explored in the previous post and full conference presentation.
The difference between human and property losses was probably partly the result of their location and the nature of the storm surge on Christmas Eve 1717. Drained bog landscapes were slowly subsiding and were significantly lower in elevation than the coastlines. However, the initial impact of the flood along the coasts would have come without warning. Contemporary accounts are unanimous in their description of the early moments of the flood, coming late at night, catching people leaving near the dikes unaware. In contrast, people residing inland would have had more warning. This is described in numerous local sources. The church book from the Groninger town of Woltersum, for instance, describes the event.
“In the day and time of trouble, people saw a white cloth flying on some of the low-lying houses of our polder as a sign of the urgent emergency! The somber sound of the clocks was heard, as distressing as they were insistent!”
“In dien Dagh en tijd der benaeuwtheid sagh men op sommige der lage Huijsen van ons carspel een wit doek waeijen, tot een teeken der dringende nood! Het nare geluijd der aangeroerde klokken, ellende gelijk als uijtroepende, wierde gehoort!”
Source: Doop, Trouw en Begraaf (DTB), Kerkelijke Gemeente Woltersum Kerkeboek 1638-1764 [online] fol. 178, 1717.
Presumably, there would have been less time to protect houses and livestock than human lives. At the same time, the bowl shaped interior would have held water in, preventing easy drainage.
With limited cultural landscape source material for this part of Germany, it is difficult to make an adequate comparison. It would interesting to see if this dynamic played out in both areas. Property damage seems to be more broadly distributed, indicating that other processes were at work. What they are, and how they related (if at all) to land use, is one area of the development this project could take.
One key difference, immediately apparent when we look to livestock losses, however, is their divergent specialization. Both regions had had a long history of intensive cattle production. By the early eighteenth century, cattle rearing and pasturing had become heavily commercialized and connected to the long-distance cattle trades stretching from Denmark to the markets in Amsterdam. Groningen and East Frisia were both centers of transshipment and cattle were raised and fattened on their fertile coastal meadows. The maps of flood losses, however, show a different picture of rural agriculture in 1717. East Frisia shows what one might expect. The majority of livestock lost are cattle, followed by sheep, goats, and horses. In contrast, Groningen appears to have been invaded by sheep! The livestock data reveals much heavier concentration of sheep losses than cattle. Without livestock censuses or other ways to measure this change (short of the tabulations of flood losses), it is difficult to tell when or why this transition occurred, but it cannot be coincidence that a massive cattle plague was ravaging herds between 1713-20 in the region. Is the predominance of sheep in Groningen indicative of an adaptation to rinderpest? In Holland, heavily capitalized farmers tended to restock herds, partly made possible by the increase in the price of beef and milk. Was this also possible in Groningen? If so, this would be particularly valuable evidence of an intersection between multiple disastrous events and confirmation of one of the key theses of my manuscript. It is also possible, however, that this is a dead end. Perhaps more sheep simply died in the flood, and the figures give a false sense of total livestock before the flood. Like the maps on property damage more broadly, these maps open up as many questions as they provide clues to the consequences of the Christmas Flood of 1717.