Trust Towards Fellow Human Beings in Hungary According to the Most Commonly Applied Survey Tools

How can historical research help in reconstructing land use at the national level if no maps of the period under study have survived, or if it would take years to analyse the maps using classical geographical and GIS methods /vectorization/? Although military surveys are now readily available online, their mass evaluation is limited by the lack of projection system or different colours used in their legend. The research led by Éva Konkoly-Gyuró, which projected a 2 km grid over the country’s territory, has gone furthest so far in generating a general overview of land use and land use changes between 1850 and 1930. However, they did not attempt to evaluate the first military survey of the late 18th century, whereas the cadastral mapping of Joseph II only survived for a few hundred settlements due to resistance from the nobility. The surviving 1786 chancellery conscription, however, at least allows us to visualise the relative differences in land use between settlements at municipality and national level – it includes the size of serfs’ arable land /at that time allodial land was still subordinate/ and their meadows. Thus not only the per capita area can be given, but the size of the meadow in relation to the arable land too. The ratio of the two landuse categories implicitly referred to the size of the livestock and pastures /communal properties were not included in the census/. And although this conscription is incomplete, the meadow and ploughland is also given in the 1720 census, so even /relative/ changes can be traced. Thus, with the help of our database /GISta Hungarorum/, it is possible to reconstruct, albeit with considerable limitations, the conditions of an era without map representations. Indicated local differences also highlight different paths of local social development. 19th century cadastral land income censuses are much more accurate. The advantage of this method, based on the visualization of historical databases, is that /in addition to its disadvantages, such as neglecting exact land use characteristics within a settlement/ it helps compress information during visualisation, so that spatial patterns of longue durée changes can be illustrated.
A way forward could be the use of machine learning algorithms from different evaluation software to identify and aggregate raster patches representing land use.