The function and accuracy of old Dutch urban designs and maps. A computer assisted analysis of the extension of Leiden (1611)

Historical manuscripts and printed maps of the pre-cadastral1
period show enormous differences in scale, precision and
color. Less apparent are differences in reliability between maps,
or between different parts of the same map. True, modern
techniques in computer assisted cartography make it possible
to measure very accurately such differences between maps
and geographic space with very distinct levels of precision and
accuracy. However differences in reliability between maps, or
between different parts of the same map, are not only due to
the accuracy measurement techniques, but also to their original
function and context of (re-)use. Historical information about
the original context of function and context of (re-)use can
give us insight how to measure accuracy, how to choose the
right points for geo-referencing and how to rectify digital maps.
On the other hand computer assisted cartography enables
us to trace and to visualize important information about
mapmaking, especially when further historical evidence is
missing, is hidden or is distorted, consciously or unconsciously.
The proposed paper is embedded in the project: Paper and
Virtual Cities, (subsidized by the Netherlands Organization for
Scientifi c Research) that aims at developing methodologies
that a) permit researchers to use historical maps and related
sources more accurately in creating in digital maps and virtual
reconstructions of cities and b) allow users to recognize
better technical manipulations and distortions of truth used in
the process of mapmaking.2
In this paper we present as one of the outcomes of this project
a method that visualizes different levels of accuracy in and
between designs and maps in relation to their original function
to assess their quality for re-use today. This method is presented
by analyzing different 17th century designs, manuscript and
engraved maps of the city of Leiden, in particular of the land
surveyor, mapmaker Jan Pietersz. Dou. The choice for Leiden
and Dou is no coincidence.
One of the reasons behind differences the accuracy of maps is
the enormous variety in methods and measures used in land
surveying and mapmaking in the Low Countries.3 This variety
was the result of differences in the private training and the backgrounds of the surveyors, in the use of local measures
and in the exam procedures that differed from province to
province.4 These differences would last until the 19th Century.
However, already by the end of the 16th Century we see in and
around Leiden the fi rst signs of standardization in surveying
techniques and the use of measures.5 First of all, the Rhineland
rod (3.767 meter) the standard of the water-administration
body around Leiden is used more and more, along local
measures in the Low Countries and in Dutch expansion
overseas. A second reason to look at Leiden in more detail is
that in 1600 a practical training school in land surveying and
fortifi cation would be founded in the buildings of its university,
the so-called Duytsche Mathematique, that turned out to be
very successful not only in the Low Countries, but also in
other European countries. This not only contributed to the
spread and reception of the Rhineland rod, but also to the
dissemination of more standardized ways of land surveying
and fortifi cation.6 The instructional material of the professors
of this training school and the notes of their pupils are still
preserved, which allows us to study the process of surveying
and mapmaking in more detail.
The reason to look into the work of Jan Pietersz. Dou is his
enormous production of maps. Westra (1994) calculated that
at least 1150 maps still exist.7 Of the object of our case study
alone, the city of Leiden, Dou produced at least 120 maps
between 1600 and 1635, ranging from property maps, designs
for extensions of the city and studies for civil engineering
works etc. We will focus on the maps that Dou made for the
urban extension of the city of Leiden of 1611. Sometimes
these (partial) maps were made for a specifi c purpose; in other
cases Dou tried in comprehensive maps, combining property
estimates and future designs, to tackle problems of illegal
economic activities, pollution, housing and fortifi cation. Since
these measurements were taken in his offi cial role of, sworn-in
land surveyor we can assume that they were supposed to be
accurate. This variety in designs and maps for the same area
allows us to discuss accuracy in relation to function and (re-
)use of maps. We will also explain that the differences between
designs and maps require different methods of geo-referencing
and analysis.
In particular, we will give attention to one design map of Dou
for the northern part of the city of Leiden RAL PV 1002-
06 (Regionaal Archief Leiden) to show how misinterpretation
of features lead to unreliable or biased decisions when the
historical context is not taken into account, even when we
can consider the results, in terms of accuracy, satisfactory.
Since Dou later also made maps for commercial purposes
of the same northern extension of Leiden it is interesting to
compare these maps.
Conclusions are drawn addressing the question of whether
Dou used the same measurements to produce a commercial
map or that he settled for less accuracy given the different
purpose of the later map compared to his designs and
property maps. To answer this question, we use modern digital
techniques of geo-processing8 to compare the old maps to
modern cartographical resources and to the cadastral map
of the 1800s in order to determine how accurate the various
maps in question are. We do this, by using the American
National Standard for Spatial Data Accuracy (NSSDA) to defi ne
accuracy at a 95% confi dence level. By point-based analysis we
link distributional errors to classifi ed features in order to fi nd
a relationship between accuracy and map function.9
Notes
(1) Cadastral mapping refers to the “mapping of property
boundaries, particularly to record the limitation of title or
for the assessment of taxation”. The term “cadastral” is also
used for referring to surveys and resurveys of public lands
(Neumann, J. Encyclopedic Dictionary of Cartography in 25
Languages 1.16, 1997).
(2) The project Paper and Virtual Cities. New methodologies
for the use of historical sources in computer-assisted urban
cartography (2003-2008) is a collaboration between the
department of Alfa-Informatics of the University Groningen
and the Virtual Knowledge Studio of the Royal Netherlands
Academy of Arts and Sciences subsidized by the Netherlands
Organization for Scientifi c Research (NWO). http://www.
virtualknowledgestudio.nl/projects/paper-virtualcities.php
(3) Meskens, A, Wiskunde tussen Renaissance en Barok.
Aspecten van wiskunde-beoefening te Antwerpen 1550-
1620, [Publikaties SBA/MVC 41-43], [PhD, University of
Antwerp],Antwerp, 1995.
H.C. Pouls, “Landmeetkundige methoden en instrumenten
voor 1800”, in Stad in kaart, Alphen aan den Rijn, pp. 13-28
Winter, P.J. van, Hoger beroepsonderwijs avant-la-lettre:
Bemoeiingen met de vorming van landmeters en ingenieurs
bij de Nederlandse universiteiten van de 17e en 18e eeuw,
Amsterdam/Oxford/New York, 1988.
(4) Muller, E., Zanvliet, K., eds., Admissies als landmeter in
Nederland voor 1811: Bronnen voor de geschiedenis van de
landmeetkunde en haar toepassing in administratie, architectuur,
kartografi e en vesting-en waterbouwkunde, Alphen aan den Rijn
1987
(5) Zandvliet, K., Mapping for Money. Maps, plans and
topographic paintings and their role in Dutch overseas expansion
during the 16th and 17th centuries. (PhD Rijksuniversiteit
Leiden], Amsterdam, 1998 describes this development as
part of a process of institutionalization of mapmaking, esp, pp.
75-81.
(6) Taverne, E.R.M., In ‘t land van belofte: in de nieue stadt. Ideaal
en werkelijkheid van de stadsuitleg in de Republiek 1580-1680,
[PhD, University of Groningen]Maarssen, 1978. Heuvel, C. van den, ‘Le traité incomplet de l’Art Militaire et
l’ínstruction pour une école des ingénieurs de Simon Stevin’,
Simon Stevin (1548-1620) L’emergence de la nouvelle science,
(tentoonstellingscatalogus/catalogue Koninklijk Bibliotheek
Albert I, Brussel/Bibliothèque Royale Albert I, Bruxelles,
17-09-2004-30-10-2004) Brussels 2004, pp. 101-111. idem,
“The training of noblemen in the arts and sciences in the
Low Countries around 1600. Treatises and instructional
materials” in Alessandro Farnese e le Fiandre/Alexander and the
Low Countries (in print)
(7) Frans Westra, Jan Pietersz. Dou (1573-1635). “Invloedrijk
landmeter van Rijnland”, Caert-thresoor, 13e jaargang 1994, nr.
2, pp. 37-48
(8) During geo-referencing a mathematical algorithm is
used for scaling, rotating and translating the old map to give
modern coordinates to it and allow further comparisons
to modern sources. This algorithm is defi ned by a kind of
transformation we decide to use based on the selection of a
number of control points (GCPS). These items are described
in detail later in this paper. All this processing was done by
using different kind of software for digital and geographical
processing and statistics (PCI Geomatics, ARCGIS, Autocad,
MS Excel, among others).
(9) Jakeline Benavides and John Nerbonne. Approaching
Quantitative Accuracy in Early Dutch City Maps. XXIII
International cartographic Conference. ISBN 978-5-9901203-
1-0 (CD-ROM). Moscow, 2007.