=Paper=
{{Paper
|id=Vol-1681/Gil_Barleta_Digital_Atlas_Of_Portuguese_America
|storemode=property
|title=Digital Atlas of Portuguese America
|pdfUrl=https://ceur-ws.org/Vol-1681/Gil_Barleta_Digital_Atlas_Of_Portuguese_America.pdf
|volume=Vol-1681
|authors=Tiago Luís Gil,Leonardo Barleta
|dblpUrl=https://dblp.org/rec/conf/dhlu/GilB13
}}
==Digital Atlas of Portuguese America==
https://ceur-ws.org/Vol-1681/Gil_Barleta_Digital_Atlas_Of_Portuguese_America.pdf
Digital Atlas of Portuguese America
Tiago Luís Gil*, Leonardo Barleta‡
*
University of Brasilia, Brazil.
tiagoluisgil@gmail.com
‡ Stanford University, USA
lbarleta@stanford.edu
Abstract
This paper presents the Digital Atlas of Portuguese America (Atlas Digital da América
Portuguesa), a project that produces an interactive map of South America claimed by the Portuguese
between 1500 and 1808, giving a detailed account on the development and structure of this digital
cartographic platform. The paper narrates the stages that the project went through and details the
structure of integrated tools as well as the process of gathering, organization, and production of the
data made available through the Digital Atlas. The main topics include the customizations made in the
GIS web-platform i3geo to include time as a central variable, the implementation of the data gathering
tool Oduduwa, and the main corpuses of primary sources utilized in the project.
1 Introduction
The Digital Atlas of Portuguese America (Atlas Digital da América Portuguesa) is an interactive map of the
American continent which was claimed and occupied by the Portuguese between 1500 and 1808. The Atlas brings
together diverse geographic information about Portuguese America, including cities, towns, roads, urban plants,
economic resources, and political features. It works as an online system, a “Google Maps” of colonial Brazil,
providing important tools for historical research on space and territory. Using and extending the software i3geo
(developed by the geographer Edmar Moretti) to meet the needs of the project, the Digital Atlas allows the user to
select a period of time – a year or a decade, for example – and specific regions to see the shape of Brazilian space in
the past. Visualization can be done in different scales, encompassing different types of geographic data, which range
from streets of specific towns to the entire set of Portuguese settlements in the Americas. Moreover, a selection of
geographic and statistical analyses is available to refine the analysis.
The project follows a long-standing tradition in historical scholarship regarding the use of maps to produce a
narrative, to explain events and structures, and to trace spatial correlations amongst political, social, and cultural
phenomena in the past. This practice, epitomized by a plethora of existing historical atlases (e.g., Duby, 2013),
intends to establish the chronology of events of a historical epoch in a cartographic medium and trace the
connections between time and space, events and geography. Portuguese America, for instance, is the topic of a few
of these atlases, usually under the geographic macro-region of Latin America or the national boundaries of modern
Brazil (Wilgus, 1969; LaRosa and Germán, 2007; Albuquerque and Campos, 1966). Regional atlases and general
historical works, which seldom do not contain a map, also complement a significant corpus of cartographic
information about Portuguese America (Fundação João Pinheiro, 1997; Westphalen and Cardoso, 1982). These
Copyright held by the author(s).
works produced a massive amount of geographic data about Portuguese America although, due to the medium of
publication, the printed page, it remained fragmentary and dispersed.
In recent years, however, this restriction has become less important once the adoption of digital tools in
humanities and social sciences expanded the limits of the physical paper. Atlases are among the types of projects
that most benefited from change, since technology reduced the time and costs of production, substantially increased
the quality of produced materials, and made the skills related to the cartographic practice more accessible.
Consequently, many new projects flourished. A successful initiative, for instance, is the Digital Atlas of Roman and
Medieval Civilizations (DARMC), which creates an interactive interface of the ancient and medieval societies,
highlighting villages, cities, monasteries, cemeteries and other important places in the period under analysis.
(McCormick et al.) Another original project is Mapping Gothic France, which combines Historical GIS, and
architecture. It displays the spatial distribution and traces the chronology of Gothic buildings in Europe, especially
churches and monasteries, allowing comparisons about the disposition and shape of the main churches and
monasteries (Murray et al.). Finally, another example is a British project called Medieval Chester, which explores
the urban landscape of the town between 1200 and 1500. (Clarke)
Each project, it is worth noting, has its particular scale and approaches the relation between time and space
differently. The DARMC, for instance, underscores structural movement by adopting a small-scale cartography (that
is, viewing from afar) and large geographic units (Roman and Medieval civilizations). On the other hand, Medieval
Chester quite precisely depicts the urban topography of a late medieval town, but neglects the changes over time of
that spatial setting. The detail of the space is deepened in Gothic France, which uses the spatial distribution of
buildings to explore the differences in architecture by transitioning between maps to photographs. In this sense, by
combining the same toolset, each project highlights (as much as hides) and constructs (or allows the user to by
himself) the argument and story of the space under study.
The Digital Atlas of Portuguese America intends to follow and expand the intersection between history,
geography, and computing. It is composed of three main elements that work together as an integrated geographic
information system (GIS). It includes a management tool of geographic data (Oduduwa), a representation and
analysis tool (i3geo – Iroko) and a virtual encyclopedia (BiblioAtlas). Oduduwa was created by our team and can
process geo-referenced historical information (that is, those with coordinates), storing them in a geographic
relational database. The geographic data handled by the database encompasses different types of geometries such as
points, lines, and polygons, as well as a set of attributes and metadata about the feature (e.g., dates, authors, sources,
classifications) composing a repository for the visualization tool and the encyclopedia.
Figure 1: Basic scheme of the “Digital Atlas of Portuguese America”
Copyright held by the author(s).
The digital map – the central part of the project – automatically presents the geographic features stored in the
database, using their coordinates to draw objects on the cartographic plane. The software i3geo, used to plot the
interactive map, was adapted (in partnership with its creator) to include time as a key of analysis, leading to the
creation of an extension named Iroko. The extension adds to the original package a series of tools to fetch the data
based on specific dates or periods of time. For instance, it is possible to define an interval of dates and see the places
that existed during that period, or to observe how a portion of the territory looked in a particular period. Iroko also
extends the original software to include other functions related to the past, such as measurements in contemporary
units (e.g., leagues or days of travel). In short, it works as a module to turn a GIS system into a historical tool.
(Gregory and Ell, 2007)
The third part, BiblioAtlas, is the gateway of the Digital Atlas that leads the users to the other resources of the
project. It consists of an online encyclopedia, based on MediaWiki system (the same as “Wikipedia”), that manages
textual information such as short articles and encyclopedia entries. These entries tell the history of the places
represented on the digital map. While producing the geographic information, the research team collects historical
information about each site plotted on the interactive map, and uses it later to create a written complement to the
spatial data. In this sense, points, lines, and polygons from the interactive map acquire meanings that are close to the
traditional historical narrative. BiblioAtlas also works in an integrated form with i3geo and is a space for the
publication of the metadata of the project.
The Digital Atlas collects, produces, and publishes geo-referenced historical information gathered form several
primary and secondary sources in order to create a digital cartography that locates places, events, and processes
between 1500 and 1808 in the area of the American continent claimed by Portugal in the early modern period. The
project started in 2009, but the first public version was launched online only in 2011. The enterprise was originally
developed by the Social History Lab (“Laboratório de História Social”) at the University of Brasilia, but has become
a collective experience involving researchers from other institutions in Brazil and Portugal. More than an atlas in the
strict meaning of the word (i.e., a set of maps), the main objective of the project is to create a collaborative platform
for the exchange of data and knowledge of the space of Portuguese America. On one hand, it gathers together in a
single platform data produced by several researches on a variety of topics; on the other, it creates a common
environment for a shared production of historical knowledge. So far, the project counts on more than 9,000
geographic features and more than 7,000 encyclopedia entries. Created by different researchers who share the
common objective of including geographic space as a vector of analysis in their work, the Digital Atlas aims to be a
departure point and a platform for many investigations in historical studies about Portuguese America. In this sense,
it is a product of the so-called Web 2.0: interactive and collaborative.
The structure and data presented is this article were the result of a long process of research and implementation.
The version 1.0 of the Digital Atlas, released in August 2011, only showed towns and parishes of Portuguese
America. Since then we have been working to include more specific information, such as districts, farms and
ranches, urban maps, boundaries and frontiers, routes, historical processes and events, and demographic and
statistical data. Similarly, the technological structure presented above is also a result of the development of the
project over the last five years, which is related to issues of availability of sources, resources, and expertise. To
understand the project, therefore, it is important to look at the choices we made since its inception.
Copyright held by the author(s).
2 The creation of the Digital Atlas: defining scope, sources, and
technology
How does one start building a digital atlas? And how do theories and methods shape the development of the
project? There was no cartography sufficiently complete or regular that we could base the project on. To define the
scope of the project, then, we needed to have a better idea of the whole picture of the spatial configuration of
Portuguese America. The first choice was to look at the basic units of Portuguese settlements: the towns, plotting
them in space and time and tracing their histories1. The alternative path, more usual than the one we took, was to
look at the captaincies, the “provincial” units under the rule of colonial governors. The problem was that they
changed significantly over time. We would run the risk of projecting the current political formation of Brazil to that
period, as many did before. (e.g. Schwartz, 1996: 70) Towns represented as points, on the other hand, seemed to be a
more stable unit of analysis to begin with.
A comprehensive list of the towns, however, was inexistent for the colonial era forcing us to adopt an arbitrary
methodological procedure. As the starting point, we used an 1872 list of locations in Brazil, composed as a sub-
product of the first national census, made available by the Brazilian Institute of Geography and Statistics (IBGE).
This source was a complete list of towns and cities in 1872, with the population and the region of each one. List in
hand, we searched for each location’s history, also compiled by IBGE in a project called The Brazilian Territorial
Division2. Although this procedure had its problems, notably in a retrospective approach that can be potentially
anachronistic, it was very fruitful because it gave us a homogenous picture of the territorial configuration of
Portuguese America. There are no historical maps that covered the entire region in detail, not even chronological or
chorographic information, let alone demographic sources or lists of villages3. With this procedure, then, we could
comprehensively survey the territory in search for clues that would be compared with other sources in the future4.
The work with the list of 1872 resulted in a new list of locations that existed in 1808, the end-point of the project.
The list was geo-referenced and added to a database that classified what each location was at each moment in its
history: a settlement, a parish, a town, a farm. Although this retrospective approach overlooks places that vanished
before 1872 and tends to lose track of those whose names changed, it was possible to produce a comprehensive list
of the locations in Portuguese America. This initial list, however, is constantly updated and corrected by our staff,
by comparing data with several primary sources and secondary literature.
Besides the initial list, we collected a large amount of historical maps that were scattered throughout several
archives in Brazil and Europe – including the National Library in Rio de Janeiro, Army Historical Archive in
Brasilia, and Overseas Historical Archive in Lisbon – composing a collection of more than 1400 maps stored in
1
We noted that many places had a similar history: they started as an unstable settlement (a mining camp, a farm), followed by the creation of
religious institutions (a chapel, and later, a parish) and finally became towns with their own municipal institutions. (Fonseca, 2011) It is
worth noting that there is a long historiographical tradition arguing that the municipality (represented by its town council) is the central
institution of the Portuguese domains. (Boxer, 1991; Boxer, 1965; Bicalho 2001)
2
IBGE Cidades. http://www.ibge.gov.br/cidadesat.
3
Even the first map that presents the Brazilian territory in its entirety, the Mapa das Cortes, which was used in the diplomatic territorial
disputes between Portugal and Spain in the middle of the eighteenth century, neglects many important locations in Portuguese America.
4
It is important to bear in mind that the actual configuration of the Brazilian territory is far from resembling the colonial area. The expansion
into the backlands became consistent only after the mid-seventeenth century and notably the eighteenth century.
Copyright held by the author(s).
digital media. The collection encompasses a wide variety of cartographic material covering specific regions of the
Portuguese America, and using different techniques, scales, and levels of detail. They constitute, thus, a mosaic of
regional and local maps. Each of them was included in a separate database, after a systematical analysis about its
authorship and publication information, visual language, and geographic data. The database consists of the following
field structure:
Title of work
Authorship
Data
Language
Subject
Metadata
Scale
Projection
Size
Colors
Reference
Points, lines, and polygons in map (to list)
Mentioned places (to list all)
Geographic Elements Main space frame of map
Morphology description and contents of map
Descriptive memorial of the analysis
Author (s)
Author’s biography
Birth and death dates
Style and techniques
Authorship and Technical details
Activity
Studies
Licenses
Used sources
Table1: Structure of the database
Copyright held by the author(s).
The work was performed as in an “assembly line,” which main objective is to train the research assistants in the
interpretation of historical maps and produce information that could be useful to find points, lines and polygons. The
team, mostly composed of history undergraduates, was not familiar with this kind of source and, to ensure quality
and methodological rigor, we created a supervised production system. Each member of the team had controlled
access to a “working tray” with maps recommended by the coordinator. They analyzed each map individually,
following a series of pre-defined steps, and submitted a standardized report for each cartographic piece. The steps
include the identification of the places named on the map, the analysis of the cartographic conventions and
techniques (scales, projections, perspective, and symbols), identification of the artistic style, and a brief research
about the author. The supervisor, then, could check each report, pointing out inaccuracies and other issues for
revision, and approve the entry (after corrections, if necessary). Besides providing data about the locations, the
analyses of the historical maps resulted in a reliable source of information about maps, cartography and cartography
techniques.
This database supported (and supports) the next stage of the project, comparing the list of places that we created
based on the 1872 list with the historical cartography. This is an ongoing process, repeated with every new entry in
the Digital Atlas, which intends to improve the accuracy of the information. When we get new information about a
location, we check its accuracy against the data produced by contemporary cartographers. Additionally, this
database helped us to find places that disappeared over time – for example, those that became depopulated or
destroyed to build dams – and those that are not very well-known in the records of current historiography.
Figure 2 – The “Digital Atlas of Portuguese America” operating under the software i3geo
At this point, we had data for cities, towns, and main parishes in Portuguese America; the next problem was how
to make it available to the public. We were considering a tool that, on the technical side, was scalable and flexible
enough to allow customizations and large datasets and, on the user’s side, encompass the main characteristics of the
modern geographic information platforms, notably a high level of interaction (multi-scale, complex hierarchy of
data, measurements, and customization of the visualization) and a potential for visual and statistical analysis. For
this project it was particularly important that the platform be able to deal smoothly with data, such as Google Maps,
in different scales – from the urban plan of a city right up to the continental level – since external sources varied
immensely in this aspect. Furthermore, as a historiographical project, the necessity of including time (or temporal
Copyright held by the author(s).
scales) in the processes of fetching and displaying information was a crucial matter. Although this requirement is
rarely available in standard GIS packages, we needed, at least, to be able to extend and customize the platform to our
own needs.
After surveying the available tools in the market, we opted for the i3geo system, produced by the geographer
Edmar Moretti, which met most of our requirements. Moreover, the direct contact with the developer also favored
our choice and, in the long term, proved to be essential to the success of the project. According to the author:
i3geo is an internet software based on a set of other free softwares, mainly Mapserver. The main objective is to
facilitate the availability of geographic data and a set of navigational tools, the generation of analysis, and the
sharing and generation of maps upon demand (Wendt and Barreto, 2013).
The next step, after choosing a tool to display the data, was to deal with its management – that is, its creation,
continual editing, and storage. As the standard format of the industry, GIS users are usually comfortable with
shapefiles, vector-based geographic files that compose a layer of a specific type of data (point, line, or polygon). At
first, they seemed the natural choice for the project, and some unproductive experiments were done with them. In
general, shapefiles are very static (it is a single file with a delimited amount of data) and unsuited to deal with the
complexity and ambiguity of historical data 5 . They became a problem when we had to divide the geographic
features into different categories in order to organize the visualization of the final digital map. It turned out to be
very difficult to accurately define in which “box” each feature should fit. For example, a mining camp (arraial) was
at the same time a population unit (where people live) and an economic unit (where people work). It could also have
a small chapel (making it a religious unit) or be the stage of an armed conflict (the place of an event). This complex
historical information, therefore, could hardly fit into the shapefile structure of attributed fields. Redundancy of data,
resources, performance, and maintenance issues would become huge hindrances for the development of the project.
We needed an additional solution to the use of shapefiles that would enable us add historical information to the
geographic features plotted in the map, but that was still simple enough to be the subject of an “automatic”
processing. We adopted a “labeling system,” which does not prescribe inherent attributes to a certain geographic
feature. Instead, we could add a handful of tags to the same geographic element, indicating the roles of the object in
different contexts. For instance, the mining camp could be labeled as a “population unit” but also, in economic
terms, as a unit of production or a site of free or slave labor; it could also be a Catholic center of devotion, and a
place where religious festivities take place. All of the categories are attached to a particular point, line, or polygon in
the map, without changing the database entry for the geographic feature or having limitations of quantity.
Consequently, this system allowed a same feature to appear in many different contexts, scales, and filters,
originating diverse visualizations.
In technical terms, the labeling system was developed using a relational database that supported geographic
features (such as a field) – PostgreSQL with PostGIS extension. This database setting, however, required an
extensive knowledge of SQL, a specific language to manipulate certain types of databases, that was not available in
the toolset of the research assistants of the project. Furthermore, even if we could train the team to use the language,
it direct use is very little practical in most of the cases. As a solution, we conceived and implemented a new software
called Oduduwa.
5 For instance, Anne K. Knowles (Knowles, 2002; Knowles and Hillier, 2008) published two collections of articles written by the historian
using this technology.
Copyright held by the author(s).
Oduduwa was conceived as a practical and versatile solution for most problems of data management in the
Digital Atlas. It is a web-based system developed in PHP that creates a friendly interface to include and manage
geographic data based on coordinates (filled in as a web form) or KML files, transforming every spatial feature into
a standardized database entry. These records, then, could be fetched in many different ways – mainly by region,
time, and classification. This is the main information Oduduwa handles:
Type of data (Point, Line, Polygon)
Geographic Feature (WKT syntax)
Historic-Geographic Data
Creation date of the site (cities, towns, etc.)
(Time-Spatial Data)
Extinction of the site
Classifications (many)
(Historical and Modern) Name of the Place
Attribute Data Precision of the information
Method used to create the data (historical maps, written sources, etc.)
Author of the research
Author of the database entry
Metadata
Date and time of creation and edition
Sources used in the research
Table 2: Types of information collected
Having set up Oduduwa, we revisited the data that we had gathered in previous steps to publish in the Digital
Atlas. The main goal was to establish a methodological procedure to define place, time, and classification of each
piece of geographic information we had. Based on our previous experience, we adopted two different methods. The
“historical-textual method” draws on traditional historical sources like travelogues, official acts, or even visual
representations, from which we gather the basic characteristics (name, date, classifications) of a geographic feature.
We combined this information with that which we had in the list of places and in the historical map databases. Once
we had sufficient information about the location, we used modern satellite imagery (usually Google Earth) to find
the geographic coordinates of the place. For example, we would make a small research about a church (possibly a
parish seat) and look at Google Earth for the geographic coordinates of its location. In the other method, “geo-
referencing,” we associate historical maps to a coordinate system and directly include the data by visually
comparing it to a current map. Some “tricks” help to improve accuracy, like slicing the map into smaller pieces or
correcting the position of the map by connecting to the locations in the map that we already know. This method was
useful, for instance, to draw urban plans and roads. This second method, however, still needed a lot of historical
information, sometimes considering changes in the locations of places or different statuses through time.
We obtained the geographic information mainly through Google Earth, which has proved to be a powerful and
practical tool for several reasons. First, it allows searches by name of place (a non-existent feature in satellite maps
available on other sites, such as Landsat). Second, it has intuitive tools for marking points and drawing lines and
polygons in space. These features could be easily exported as KML (and open XML format for geographic data)
and, using a functionality of Oduduwa, imported into our geographic database. Finally, an interesting advantage of
Copyright held by the author(s).
Google Earth is the possibility of using “crowdsourcing knowledge.” Several people have spatially tagged photos of
places they have already been to and indexed them with keywords, especially the names of places. In this way, as a
result of contributions from anonymous people who sent their data to Google servers, we can obtain useful clues
about places that are barely known to modern audiences, but which are available on the local level. Since there is no
review process nor are the users experts in GIS or history, the degree of reliability of this information is considered
extremely questionable by the team. Even so, when handled cautiously and using them as leads for further
investigation, we have been able to make several findings that would have otherwise been impossible.
The project is presently in the stage of comparing primary sources with data obtained from lists produced based
on the 1872 census. As we have indicated, cartographic sources have been the priority of the project until now. The
following steps include the study of primary narratives (chronicles, serial sources, among others) and the critical
assessment of data gathered from books, articles, and other publications. Similarly, the set of tools and resources of
the project is under constant improvement, and new functionalities are expected to be available as the project
advances. The “Digital Atlas”, therefore, is a work in progress, based on constant updating, and with no end in sight.
References
ALBUQUERQUE, Manoel Maurício de and CAMPOS, Antônio Pedro de Souza, 1966, Atlas histórico e geográfico
brasileiro. Rio de Janeiro.
BICALHO, Maria Fernanda Baptista. As câmaras ultramarinas e o governo do Império. FRAGOSO, João;
BICALHO, Maria Fernanda B.; GOUVÊA, Maria de Fátima S. O Antigo Regime nos Trópicos: A Dinâmica
Imperial Portuguesa (séculos XVI–XVIII). Rio de Janeiro: Civilização Brasileira, 2001. Pp 189–221.
BOXER, Charles R. 1965. Portuguese Society in the Tropics: Municipal Councils of Goa, Macao, Bahia and Luanda,
1510-1800. Madison: University of Wisconsin Press.
BOXER, Charles R. 1991. The Portuguese Seaborne Empire 1415-1825. Manchester: Carcanet in association with
the Calouste Gulbenkian Foundation.
CLARKE, Catherine, Mapping Medieval Chester Project [Online] www.medievalchester.ac.uk, (accessed 15 April,
2015).
DUBY, Georges, 2013. Atlas historique Duby. Paris: Larousse.
FONSECA, Cláudia Damasceno, 2011. Arraiais e vilas d’el rei : espaço e poder nas Minas setecentistas. Belo
Horizonte: EDUFMG.
FUNDAÇÃO João Pinheiro, 1997. Panorama de Belo Horizonte: Atlas Histórico. Coleção Centenário. Belo
Horizonte: Sistema Estadual de Planejamento, Fundação João Pinheiro, Centro de Estudos Históricos e Culturais.
GREGORY, Ian and ELL, Paul 2007. Historical GIS: Technologies, Methodologies and Scholarship. Cambridge:
Cambridge University Press.
KNOWLES, Anne Kelly (ed.) 2002. Past Time, Past Place: GIS for History. Redlands, EUA: ESRI Press.
KNOWLES, Anne Kelly, and HILLIER, Amy (eds.) 2008. Placing History: How Maps, Spatial Data, and GIS Are
Changing Historical Scholarship. Redlands, Calif: ESRI Press.
Copyright held by the author(s).
LAROSA, Michael and Mejía, Germán. 2007. An Atlas and Survey of Latin American History. Armonk, N.Y.: M.E.
Sharpe.
MCCORMICK, Michael; GRIGOLI, Leland; ZAMBOTTI, Giovanni et al. Digital Atlas of Roman and Medieval
Civilizations. (http://darmc.harvard.edu/, accessed 15 April, 2015)
MURRAY, Stephen; TALLON, Andrew and O'NEILL, Rory. Mapping Gothic France (http://mappinggothic.org/,
accessed 15 April, 2015)
SCHWARTZ, Stuart B. 1996., Slaves, Peasants, and Rebels: Reconsidering Brazilian Slavery. Urbana: University of
Illinois Press.
WENDT, E., and BARRETO, A. G. 2013. Inteligência Digital. Rio de Janeiro: BRASPORT.
WESTPHALEN, Cecília Maria, and CARDOSO, Jayme 1982. Atlas Histórico Do Paraná. Curitiba: Ind. Gráfica.
WILGUS, A. Curtis. 1969. Historical Atlas of Latin America: Political, Geographic, Economic, Cultural. New York:
Cooper Square.
Biography of the authors
Tiago Gil is Associate Professor of Brazilian History in the Department of History at University of Brasilia, Brazil
and coordinator of the Digital Atlas of Portuguese America. He holds a B.A. in History from Federal University of
Rio Grande do Sul (2000) and M.A. (2003) and Ph.D. (2009) in Social History from the Federal University of Rio
de Janeiro. He is the author of Infiéis transgressores : elites e contrabandistas nas fronteiras do Rio Grande e do
Rio Pardo, 1760-1810 (Arquivo Nacional, Brazil, 2007), which won the 2005 Research Prize of the Brazilian
National Archive. He also authored "Storici e informatica: l’uso dei database (1968-2013)" Memoria e ricerca, n.
50, (Sep./Oct. 2015) and "Redes e camadas de relacionamentos na economia: metodologias para o estudo da
confiança mercantil na América Portuguesa do Antigo Regime" Revista de Indias, vol. 75, n. 264 (2015), among
other articles, chapters and books.
Leonardo Barleta is a doctoral candidate in Latin American History at Stanford University and co-coordinator of
the Digital Atlas of Portuguese America. He holds a B.A. (2010) and M.A. (2013) in History from Federal
University of Parana. His ongoing dissertation investigates the territorial evolution and development of domestic
markets in colonial Brazil.
Acknowledgements
We would like to thank several research assistants that participated and helped to craft the Digital Atlas of
Portuguese America, who are the soul of this project.
Copyright held by the author(s).