Frederic Kaplan (Ecole Polytechnique Fédérale de Lausanne (EPFL))
Napoleonic Cadastre, Digital Humanities, Artificial Intelligence
Rémi Petitpierre and Paul Guhennec’s contribution explains how graphical elements can be annotated and automatically extracted from the corpus of Napoleonic-type historical cadasters maps, generalizing a method that can be applied to several geographically distant and qualitatively different case studies. With this approach, already digitised cartographic corpora of European archives can be used to rapidly produce new historical databases. This volume of information hitherto unthinkable can be studied at different scales and in collaborative manners to develop systematic comparative approaches.
Federica Pardini’s article deals with the historical cadastre, in particular the ‘Napoleonic’ one, from a graphic point of view. The study conducts a wide-ranging analysis on a substantial corpus of maps of the main French cities. The approach is to analyse graphic differences and continuities as signs, not only stylistic but also qualitative, of variations in urban space. The appearance of certain specific graphic elements, or conversely their absence, stands as distinctive markers of the discretisation of urban space introduced by the geometric cadaster. Understanding their genesis allows to formulate a theory of the specificity of the structural elements of the different cities compared, and to deduce a morphological theory of urban space.
Isabella di Lenardo’s contribution illustrates how the automatic extraction of the parcels of land from the Napoleonic cadastre of Venice and the systematic study of the values associated with owners and functions makes it possible to represent social dynamics of inclusion or exclusion from the city’s socio-economic sphere. In this particular case, the city’s first geometric cadastre embodies a paradigmatic image of the city suspended between the continuity of the landed assets of the ancien regime, the fracture with the insertion of new social bodies elected to the rank of owners and the dismantling of the secularisation of land ownership represented by the suppression of ecclesiastical property. These elements emerge from a quantitative analysis enriched by qualitative focuses on single protagonists of this urban dynamic.
Effective Annotation: How to Leverage Manual Work for the Automatic Vectorization of Cadastral Maps
Remi Petitpierre (EPFL) and Paul Guhennec (EPFL)
Historical Cadastral Maps, Annotation, Segmentation
The great potential brought by large-scale data in humanities is still hindered by the time and technicity required for making historical data digitally understandable. Within urban studies, historical cadastres have been hitherto largely under-explored despite their informative value. Powerful and generic technologies to automate their vectorization, based on neural networks, have recently become available. The transfer of these technologies is heavily hampered by a lack of practical literature on the key step of the pipeline: the annotation. In this article, we propose a set of practical recommendations based on empirical findings on document annotation and automatic vectorization, focusing on historical cadastres. Our recommendations are generic and easily applicable, based on a solid experience on concrete and diverse projects.
Cartographier l’Exception. Variations on the Napoleonic Cadastral Space
Federica Pardini (EPFL)
Urban History, Historical Cadastres, Urban Representation
The impressive graphic homogeneity systematised by Napoleonic cadastres provides a clear rendering of the urban situation in a large part of early nineteenth century Europe. As a result of several operations of spatial abstraction, such a standardised language runs the risk of levelling the peculiarities of different city structures: in these land registers, any type of urban object is indeed unequivocally classified to prevent subsequent ambiguous interpretation. Diving into these planimetric plates, however, one can still encounter various representative discontinuities denoting a richness in the urban fabric not a priori classifiable by the centralised French administration. These exceptions appear to be of different natures dictated by a mere logic of taxation in some occasion, by ideological and aesthetic intentions, or by sincere architectural exigencies elsewhere. They thus shed light on the potential visual signature of the city they are portraying, managing to make specific urban characteristics manifest beyond generic standardisation and offering a new glance on the European city system in the moment before the beginning of those grands travaux that would have revolutionised its conformation.
Dynamics of Ownership in the Venetian Cadaster of 1808
Isabella di Lenardo (EPFL)
Historical Cadastral Maps, Computer Vision, Digital Humanities
Since 2016, researchers at the Digital Humanities Laboratory, already active in research into visual computing techniques applied to art history, have begun to explore the possibility of extracting and classifying, automatically, in a supervised manner, the graphic elements of the cartography constituted by the historical land registers of the 19th century.
The first results of automatic analysis on maps and automatic linking with the parcel descriptions contained in the land registers were particularly encouraging, to the point of undertaking several parallel projects of extraction and quantitative and qualitative analysis. The first geometric cadastre of Venice, drawn up around 1808, imposed by the French administration, has been automatically, vectorized, the textual information extracted and realigned with the current plan. The description contained in the land registers were also transcribed and added to each parcel to build a geoweb exploration system.
The analysis of the position of the properties of owners, in particular religious institutions, secular institutions and individuals, allowed us to deduce the existence of two opposite dynamics: the cluster and the dispersion. The properties of the religious institutions that were subject to suppression included almost all the parish churches, but above all the large convents and the 'Schools', those lay organisations with a religious and welfare vocation. It is not to be thought that these organisations were confiscated of the parcel of land constituted by their main headquarters or some adjoining place, but rather of a state of property consisting largely of rented houses and gardens.
This duality of land tenure dynamics is also present among some private individuals. An emblematic case of cluster is that of Almorò Grimani quondam Marcantonio who owned 45 cadastral parcels distributed between Ghetto Vecchio, Nuovo and Nuovissimo, making him the leading owner in terms of the number of possessions in this specific area of the city, owning 'houses' and 'rental warehouses'. The contribution will illustrate with a wealth of examples the problem of textual indicators and the dynamics of land tenure dislocation.