Veliky Novgorod is one of Russia’s oldest cities, founded as a small town in the 8th century and located in the far west of the country, close to Estonia and Finland. Novgorod was so important for the trading and warring Swedish Vikings heading towards Ukraine, the Black Sea and Constantinople that they gave it a Swedish name still in use, Holmgard. Novgorod is built on both sides of the wide Volkhov River about 200 km south of Lake Ladoga which is connected to Saint Petersburg and the Baltic Sea via the river Neva. Early Novgorod became a leading Russian-Orthodox Christian spiritual centre, a place of learning where the famous Novgorod Chronicle 1016-1471 was written.

As so common in Russian towns all houses in Medieval Novgorod were built of wood and thus prone to fires. Hence, it is miracle that the chronicle survived several actions of war and more than three scores of major fires. Several years ago I analyzed the history in some detail when I realized that probably no other town in Europe has had so many well recorded fires during the nearly 400 years 1097-1450. The database reveals the fire history as quoted excerpts from the Novgorod Chronicle and a few other chronicles from western Russia and Ukraine.

The animation above was made to give you an impression not only of the number of major fires in Novgorod but to show that fires are more frequent during certain periods. Why? It takes research to find out and it would be interesting to decipher. The simple way is to calculate a mean value. A value used by forest historians is the number of fires per century, the Novgorod figures follow:

1101-­1200 = 12/100 yrs
1201­-1300 = 12/100 yrs
1301­-1400 = 27/100 yrs
1401­-1450 = 24/100 yrs

The result looks even better as a histogram. But what is the value of it? Nature does not keep a record of the shifts of centuries and millennia. The same is true of the “Medieval Climate Optimum”, the “Medieval Warm Period” and the “Little ice age” can be of use some years after their introductions. When we try to apply such concepts in different parts of the world we have to redefine the beginning and end of all these catchall designations. 4 of the 64 fires in Novgorod were caused by war actions, but we can not find out how many were caused by human carelessness with open fires. We could get a better understanding of the causes of Novgorod’s medieval fires if we knew in which part of each year a fire hit. Of the 64 fires, 5 are known to season, 1 to month and 21 to the very day. It can be concluded that fires were most common in these months (in descending order): May, June, September, and August.

The most striking feature of the animation is the concentration of fires in the period 1368-1407. It should be of no surprise, since all of northern Europe was then in the grip of the cold. The summers were cold in the 1370s through 1390s, but in 1398 the winters also started being very harsh through this period. The Baltic Sea was ice covered in its southern part and horse-wagon transports between Germany and Denmark took place over the ice. In the 1491/1492 winter the River Volkhov was covered by ice thick enough to allow both people and horses move freely between the two halves of the town.

All houses in Novgorod were made of wood and they were heated by open fires. The lower the outside air temperature, the more firewood was used inside the houses. After some very cold days everything inside was dehydrated and flammable. A stray spark from an unattended fire could set a house ablaze in seconds. All water in Russia was ice – if the wind was strong, nothing could stop a town fire. The great number of major fires in Novgorod 1368-1407 at the end of the animation show us that fires are possible when organic material is dry and also that drought has two faces – heat and cold. In a world intoxicated by carbon dioxide, many are unaware of the fact that the by far largest deserts on earth are the Antarctic and the Arctic. The database holds information from many scores of old chronicles like the ones from Russia. You will find page references to almost all publications we have quoted. The list of references is 203 pages long.