




































Mapping Dante: A Digital Platform for the Study of Places in the Commedia 


Humanist Studies & the Digital Age, 5.1 (2017) 
ISSN: 2158-3846 (online) 
http://journals.oregondigital.org/hsda/ 
DOI: 10.7264/N3MG7MXK 

 

 

 

 

Mapping Dante: A Digital Platform for the Study of Places in the Commedia 

Andrea Gazzoni 

 

Abstract 

This essay presents Mapping Dante, a project for the study of the geography of the Divine Comedy 
through a digital map visualizing all of the place-names mentioned in the text. First, the project 
background is sketched out by a concise overview of the history of the reception and 
visualization of Dante’s geography, of the constellation of digital Dante projects, and of GIS 
literary mapping. Second, specific stages and issues of Mapping Dante are discussed: the making of 
the dataset and its categories, the heterogeneity of medieval geography, the structure of the map 
with layers and pop-up cards. Conceived as a repository of Dante’s encyclopedic use of 
geography in the Commedia, the map is also an experiment in connecting text and cartography 
through the possibilities offered by GIS technology. By exploring different visualizations of a set 
layers based on textual, cultural and rhetorical categories, users can search for patterns in the 
distribution of Dante’s geographical references, and can retrieve information relevant to each 
passage in which a place is mentioned.   

 

Mapping Dante [www.mappingdante.com] is a project in progress for the study of the 
geography of the Divine Comedy. Launched in May 2016, it consists of a digital interactive map 
with all of the place-names mentioned in the poem, and of additional non-cartographic materials: 
the Italian text of the Commedia with links to all of its geographical references, a network 
visualization of the relations between individual cantos and places, and a set of charts with 
quantitative analyses of the same dataset on which the map is based. The constellation of digital 
projects about Dante and his Commedia is remarkably wide, certainly wider than for any other 
Italian author, with some outcomes at the cutting edge of digital literary scholarship. So far, 
however, none of these projects has taken advantage of digital mapping technologies to address 
the full scope of the geographical imagination recalled in the poem by Dante’s “cartographic 
impulse” (Cachey, Jr., “Cartographic Dante”, 325). The purpose of Mapping Dante is to fill this 
gap, providing readers of the Commedia with a map that is complete and reliable, and also capable 
of posing questions about the ways in which its very content has been cartographically generated. 

The major encyclopedic websites on Dante, given their organization of a panoply of 
hyperlinked resources, highlight the value of interconnectivity. The multimodal configuration 
typical of Dante's work and of its reception require links between the modes of space and 
language. The digital artifact must be an extensive and intensive hypertext that points to an 
underlying composite encyclopedia of materials. If Mapping Dante moves away from that 
“panoply-model” that governs websites like Princeton Dante (Hollander), Digital Dante (Barolini) or 
The World of Dante (Parker), it does so out of apparent practical constraints on time and resources, 
as illustrated in section 2 of this essay. Yet, making a virtue of necessity, it might be also said that 
the real raison d'être of Mapping Dante has to do with the evolution of digital scholarship from 
“digitization” of pre-digital materials (Burdick et al. 8-9) to “knowledge design” (Schnapp), that 

http://journals.oregondigital.org/hsda/


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   84 

is, from a use of computational technologies mostly as the infrastructure of data repositories to 
an approach that tries to investigate the forms in which we receive, produce, and circulate 
knowledge. With a number of successful encyclopedic Dante websites already online, it may now 
be time to design not as much big repositories as relatively small-scale experiments characterized 
by a sharper focus on the issues raised by a specific subject and on the possibilities offered by 
specific digital interfaces,1 GIS literary mapping being a case in point. As in Digital Humanities at 
large, the making of new projects goes along with the making of new questions and the re-
examination of old ones. 

 

1. Reading Dante’s Geography: from Glosses to Maps to GIS 

 

Fig. 1. Complete extent of Dante’s geographical references. Source: www.mappingdante.com.  

 

A complete map of the Commedia’s repertoire of places has been long overdue, for cultural 
and material reasons. The interest in the geography of the Commedia dates back to its early 
commentaries (14th century) containing glosses with localization, description, and identification 
of places. The Commedia is the first vernacular work glossed line by line as if equated to the great 
poems of the ancient auctores, Virgil, Ovid, Lucan, and Statius, which were commonly read also as 
encyclopedic repositories of knowledge in a number of fields, including geography. The quantity 
and extent of geographical references in the Commedia covers what was believed to be the 
inhabited world (Moore, 114-119), from Scandinavia to the African desert, from the Pillars of 
Hercules to India (see fig. 1); the invention of the location of Purgatory in the Southern 
hemisphere first visited in Ulysses’s last voyage (Boyde 109-111) is the exception that proves the 

                                                           
1 See for example Intertextual Dante (Van Peteghem), a sub-project added to Digital Dante 

(Barolini), DanteSources (Bartalesi et al.), or the Twitter version of Petrarch’s Rerum Vulgarium 

Fragmenta (Lollini and Rosenberg).  

http://www.mappingdante.com/


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   85 

rule. The Middle Ages, however, did not visualize space as modern maps do, that is, as a 
dimension regulated by the uniformity of linear standard measurements (Farinelli, Geografia 15-
16): the world addressed by medieval geography, in fact, remained an ensemble of ontologically 
different places that did not sit on a uniform space (Rouse 18), whether written as a descriptio orbis 
or visualized as a mappa mundi (Harvey 19-38).2 The standards of uniformity, exactness and 
proportion typical of modern cartography are partially anticipated in the late Middle Ages by 
some of the extant regional maps from the late Middle Ages (Harvey 82-93) and by nautical 
charts (Harvey 39-50). As suggested by Cachey, Jr. (“Cosmology” 238-239), Dante may have 
integrated these two cartographic forms with the more traditional view of the descriptio or mappa 
mundi. 

 

Fig. 2. An example of diegetic map: the cross-section of Dante’s hell from Lo’nferno el Purgatorio e’l 
paradiso di Dante Alaghieri, the 1515 Aldine edition of the Commedia. The calculations of infernal 
mileage are derived from the study of Antonio Manetti (1423-1497). Source: 
www.italnet.nd.edu/Dante/text/Hell.html.  

 

It was only in the age of the rise of modern cartography (between the 15th and 16th 
centuries) that the first maps of the Commedia were produced, with representations of the realms 
of afterlife, particularly hell (Kleiner 23-34), rather than of the “real” place-names in the poem: 
diegetic maps of the spatial architecture of the narrative (see fig. 2) instead of non-diegetic maps 
of its geographical encyclopedia. It could be argued that while the subject of diegetic maps (the 
immutable spaces of afterlife) was held to be stable, uniform, regular, and measurable despite 
some inconsistencies on Dante’s part (Kleiner 34-56), the potential subject of non-diegetic maps 
was considered imprecise, erratic, and impervious to the homogeneity of visualized space. In any 
event, diegetic cartography soon became a standard companion to the text of the Commedia; its 
format has remained unchanged even after the switch to the digital.  

                                                           
2 See http://www.themappamundi.co.uk/ for an interactive visualization of one of the most 
important extant mappae mundi, the Hereford map, rich with elements and places drawn from 
mythology, biblical stories, bestiaries, etc.  

http://www.italnet.nd.edu/Dante/text/Hell.html
http://www.themappamundi.co.uk/


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   86 

 

Fig. 3. Non-diegetic maps of Italy and Tuscany from Mary Hensman, Dante Map (David Nutt: 
London, 1982). Source: www.worldofdante.org/dantemap_detail.html. 

  

 

It is only in the second half of the 19th century that the Commedia provides the basis for 
non-diegetic maps, when for the sake of erudition and cultural tourism some literati in Italy and 
Europe compiled lists and descriptions of the places Dante mentioned and visited (Cavalieri 111-
122). The approach was innovative, but such illustrations could not render the full scope of 
Dante’s geography, which reaches well beyond Italy, although the Italo-centric focus was 
certainly reasonable, for the self-evident reason that the majority of the Commedia’s place-names 
are Italian (fig. 3). This is the kind of map that we still find in the apparatus of recent editions of 
the poem.3 Conversely, the most authoritative scholarship on the Dante’s geography from one 
end to the other of the 20th century (Moore; Boyde 96-111; Cachey, Jr., “Cosmology”) has been 
interested more in analyzing the overall outlines of Dante’s world and cosmos than in studying 
how the mappa mundi virtually present in his Commedia becomes concrete through the distribution 
of local references in the text.  

A complete non-diegetic map would be a support meant to make visible the geography of 
the Commedia both in its wider cultural understanding and its particular instantiation.  
Nevertheless, with cartographic techniques based on static printed surfaces the resulting format 
would simply be too large and unwieldy: materially, it should be at the same time large-scaled, to 

                                                           
3 See the maps in Singleton’s commentary on the Inferno: the course of the Po River, the region of 
Virgil’s birth, Italy, and Tuscany. The Princeton Dante has two maps of Italy and Tuscany, while 
The World of Dante reproduces the maps of Italy and Tuscany originally in Mary Hensman’s 1892 
compilation Dante Map. 
 

http://www.worldofdante.org/dantemap_detail.html


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   87 

legibly visualize even the most densely referenced areas like Tuscany, and wide enough to extend 
into the distant expanses of Africa and Asia. This is the first limitation inherent in modern 
cartography, insofar as it is static: on a fairly small-sized map with a fixed scale only a certain 
amount of legible information can be graphically inscribed; the second limitation is that mapped 
surfaces do not allow the inscription of much more than name and location of a geographical 
item, thus cutting off most of the relations between text and place-names.  

A possibility of reconnection has emerged with digital mapping platforms known as GIS 
(Geographic Information Systems). If decades ago GIS was a notion familiar only to specialists 
such as geographers and map-makers, today this technology pervades our lives (e.g. Google 
Maps) by virtue of its versatility and inclusiveness, with a range of purposes that goes well 
beyond its original geo-coding function and that makes it integral to the present and the future of 
the “spatial turn” in the humanities (Bodenhamer). GIS puts control of the digital representation 
in the hands of a mapmaker who is capable of manipulating the data tables from which a map is 
generated and the way they are visually translated into a set of cartographic layers. By creating and 
editing connections between the “spatialized variables” in the map, a GIS operator is able to 
produce a range of “composite mappings, visual representations, and spatial models for analysis” 
(Travis 5). Three points are most relevant to the study of how texts negotiate their references to 
the external world (Piatti 89): a GIS map is – in form and content – built up by the arrangement 
of multiple layers, each of them being a particular take on the dataset; it is open to variation and 
manipulation; it can integrate, on a cartographical basis, a huge volume of non-cartographic data. 
Along these lines, GIS mapping proves to be structurally closer to the way literary texts work 
(multiple perspectives, deformation/construction of realities, multimodality); even more 
significantly, they end up resembling the ways medieval geography worked. As scholars involved 
in mapping projects clearly pointed out, what we can reap from maps are patterns and 
conjectures rather than full-fledged interpretations, the elaboration of which will be the task of a 
reader/viewer and will require careful cross-analysis of text and map. (Moretti 53-54, Piatti 91, 
Travis 19-20).   

The development of an increasing number of GIS-based mapping projects attests to both 
the standards imposed by new technological platforms and the variations that digital humanists can 
elaborate in step with their field and object of study. One major distinction that emerges against 
this general background is one addressed before between diegetic and non-diegetic maps. 
Diegetic maps are much used for studies in modern novels, usually qualified by realistic spatial 
configurations within which the narrative unfolds as if on a modern map, but they cannot 
properly visualize the composite ensemble of places that is frequently found, for example, in 
medieval literary texts; nor can diegetic maps properly visualize any geographical imagination that 
is not directly related to the architecture of a narrative. Although not able to solve certain issues 
like the positioning of “imaginary” places, non-diegetic maps illustrate the geographical 
encyclopedia of a given corpus without trying to mirror the spatiality of a narrative that unfolds 
in a real (or realistic) world. Obvious as this characteristic can be, it implicitly underscores the 
non-correspondence between page and place: how do we mentally and materially connect two 
dimensions apparently so heterogeneous? This may not be immediately manifest in non-diegetic 
mapping projects either outside the realm of literature or dealing with a corpus too large to allow 
in-detail examination.4 Yet the question arises from the very process of map-making, and 
becomes central to projects based on medieval culture (Wrisley, “Spatial Humanities”). 
Conversant with those approaches but with a substantially smaller corpus, Mapping Dante 
investigates the connection of text and place both extensively and intensively: in an encyclopedic 
way, but studying only one source text.   

                                                           
4 See for instance the Digital Atlas of Roman and Medieval Civilizations (McCormick) and Visualizing 
Medieval Places (Wrisley). 



Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   88 

I have to note beforehand that a GIS map of the places of the Commedia did already exist 
(Soars), though quite raw and hardly known, probably because it appeared on an information 
design blog rather than on a dedicated scholarly site: its dataset was taken from Wikipedia, and 
the visualization consists of only three layers (places mentioned in Inferno, Purgatorio, and Paradiso 
respectively), with no additional information incorporated: in this first attempt the scholarly and 
didactic potential of GIS technology lies almost completely untapped. The only other GIS Dante 
project which I have known was part of the site Mapping the Medieval, and was limited to Florence, 
being the outcome of a collaborative classroom work on Florentine places referred to by Dante 
and his fellow citizen Dino Compagni (Hamilton). None of these GIS projects was known to me 
until months after the launch of Mapping Dante, as they have not been included in any online list 
of digital Dante resources, and are no longer functioning.5 

 

2. Project Description 

The delimitation of the corpus of Mapping Dante and the definition of its purpose were 
first determined by the time and resources available. The plan was to have the project completed 
and usable in a year of intermittent work supported by an Early Incubation Grant from the Price 
Lab for Digital Humanities of the University of Pennsylvania. At the same time, I intended to be 
able to complete, manage, and update both map and website without extra funding and specialist 
assistance. By envisioning a low-cost and medium-scale project realistically manageable, I 
intended to demonstrate that even without considerable financial and technological resources, a 
satisfactory outcome should be possible to a scholar with average computer skills and limited 
access to funding. 

The first step of Mapping Dante was the choice of a map-making tool among many 
options. I opted for ArcGIS, a high-end software produced by Esri, most suitable in terms of 
interface and versatility, with customizable pop-up cards and practically no limits to the number 
of layers in the map. Although the very first steps of the map were created with ArcGIS Desktop 
(the full version of the software, accessible through my university), then exported to ArcGIS 
Online to be published, I soon decided that the job could be done directly on the online platform 
with a free public account: fewer functions, but still enough for the task. Moreover, while the 
transition to the more basic online version placed some constraints on map-making, it gave me 
full control of the process, with a gentler learning curve.     

Dante’s Commedia has 340 geographical items, for a total of 724 mentions. Before going 
digital, however, I had to start off with an analog exercise: close reading of the entire Commedia in 
order to compile a dataset with the help of commentaries, repertoires, and gazetteers.6 No 
machine reading; what was needed, on the contrary, was annotation and description of the places 
and of the passages where they are mentioned. How can information be extracted from the text 
and made into a workable format? The integrity of the source text had to be fragmented and 
translated into another textuality determined by spatial models and computing protocols rather 
than by the sequential reading. In this respect, Harley’s contention that maps are a “cultural text” 
(7) and that all the steps of its making, namely “selection, omission, simplification, classification” 
are “inherently rhetorical” (11), rings quite true, though with a caveat: maps and texts have 
different rules. Since those imposed by GIS are indeed strict, having been originally fashioned for 

                                                           
5 In Soars’ map the base terrain on which places are pinned is not visible; Hamilton’s map is no 
longer online so that my knowledge of it is limited to what can be read in his article.   
6 The two most helpful resources have been the digital Dartmouth Dante Project (Hollander), with 
easy access to more than 70 commentaries, and the entries in the Enciclopedia dantesca (Bosco). 



Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   89 

quantitative approaches, to speak qualitatively in the native language of GIS is a challenge for 
humanists (Bodenhamer), both frustrating and fascinating. GIS functions with spreadsheets 
formatted in rows and columns, where each place is identified by latitude and longitude and 
optionally described by additional data. What elements shall we write into the spreadsheet, and by 
what categories? The cartographic infrastructure, whether analog or digital, tolerates no ambiguity 
(Farinelli, Crisi 21), since the mapping protocol relies on a rigid opposition between 0 and 1, yes 
and no. At this level, tertium non datur – the very opposite of a literary text. 

With medieval texts, the cartographer may come across a typical problem: a location 
cannot be indicated by exact longitude and latitude coordinates. That could be due to destruction 
that left scant or uncertain archeological evidence, as is the case with the pre-Roman Alba Longa 
(fig. 4), the biblical Sodom and Gomorrah, or the castle of Cunio in Romagna (razed to the 
ground at the end of the 13th century).  

 

Fig. 4. Approximate location of Alba Longa. Source: www.mappingdante.com.  

Uncertainty could derive from the irremediably vague nature of a culture’s knowledge of a 
place, in terms sometimes even bordering on legend and mythology, as happens with places such 
as Tartary or the Riphaean Mountains. In a literary map, however, these issues can be faced with 
the help of commentaries and gazetteers, and with an acceptable degree of approximation and 
arbitrariness. Another problem is posed by places of a certain extension like regions, kingdom, 
and the like: drawing neatly-contoured areas (what in the GIS language is a polygon) would have 
been a debatable choice, given the fluid and uncertain configuration of frontiers in Dante’s age as 
well as in the classical and biblical geography to which he often refers. This is why extended areas 
in the map are marked only by a symbol (a radial color-gradient suggesting expansion). It is true 
that, if I were a cartographer this would not be an example of best practice, since on the same 
surface ill-defined places coexists with precise ones. Yet such problems shed light on a distinctive 
quality of Dante’s geographical imagination already stated above: heterogeneity. What are the 
implications of putting all the place markers on one surface, no matter if layered? As one of the 
tenets of Westphal’s géocritique goes, “space cannot be understood except in its heterogeneity” 
(37): ancient and contemporary, small and vast, localized and vague, near and far, imaginary and 
real, experienced and known by reading or hearsay.  

Ambiguity and complexity point to a variety of relationships between text and world, 
place-names and “real” referents. Let us compare, for example, the Acquacheta waterfall 
mentioned in a simile in Inf. XVI (94-102) with the Carthaginian Zama of Inf. XXXI (115-117): 
while the former is a precise description of a place known first-hand by Dante, the latter is a 
faraway place famous because written accounts of Roman history mention it as the location of 
the final battle of the Second Punic War. By a modern cartographical standard, Acquacheta and 

http://www.mappingdante.com/


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   90 

Zama should have the same quantitative and qualitative properties; alternatively, Dante’s 
geography is multi-focal and differential, which is what the dataset records. Visualizing this on 
the map helps scholars to understand the range of effects of the mention of places in the 
Commedia.  

On the one hand, the intimacy of Acquacheta augments its rhetorical function not only 
by providing a sonic and visual equivalent to the falling of the Phlegeton river down a steep cliff, 
but also by reinforcing Dante’s status as direct witness of things in afterlife inaccessible to the 
experience of his readers.7 The assumption is: “I have seen that waterfall in hell as directly and 
vividly as I saw that waterfall in the Apennines”; not by chance, the passage is followed by the 
appearance of the infernal monster Geryon and by a crucial address to the reader in which Dante 
swears on the Commedia itself that he did see this incredible creature.8 On the other hand, the 
mention of Zama in Virgil’s address to the giant Antaeus in Inferno XXXI is certainly a captatio 
benevolentiae that links the glory of Scipio Africanus’s victory with the origins of Antaeus himself, 
but it also serves to distance him from Dante and his readers.9 The giants Dante sees in lower hell 
all belong to a sort of pre-history, either biblical (Nembrot) or classical (Ephialtes, Briareus, 
Antaeus), their hybris having been crushed by the forces of Good; hence the inaccessibility of 
Zama, with its aura of mythical and historical antiquity, helps us to appreciate the full extent of 
Antaeus’ menacing otherness. At the same time, Zama’s status tells us something about the 
speaker in this passage: the fact that Virgil realistically uses references drawn from his pagan 
culture is a minor but significant instance of the divide that irremediably separates Virgil as a 
character from the Christian era. In short, the range of relations between the text and the places 
mentioned calls for an interpretation that can be effectively supported by the versatility of a GIS 
map.   

The underlying data organization facilitates precise indication of spaces while also 
embracing the ambiguities of Dante's use of these geospatial terms in the Commedia. Each row in 
the spreadsheet corresponds to a specific mention, and rows are divided by columns that separate 
and encapsulate pieces of information (this structure is visible in the pop-up cards describing 
each single mention, as in fig. 4). The row begins with the place-name in Italian and English, and 
then continues with the data we need to locate the passage in the poem: cantica, canto, and line. 
The “Notes” field then gives the aliases of a place if the Commedia differs from today’s usage (e.g. 
in Inf. VII, 22, Charybdis instead of Punta del Faro for the Sicilian tip of the Messina Straits); the 
same entry may signal when the location is uncertain or hypothetical, or when Dante uses a 
municipal landmark to recall a town or a city (e.g. Lateran for Rome in Par. XXXI, 35, Ponte 
Vecchio for Florence in Par. XVI, 145-146, and Rue du Fuarre for Paris in Par. X, 137). Next is 
“Place category”, that could be specific place (cities, towns, mountains, castles, and the like), 

                                                           
7 “Come quel fiume c’ha proprio cammino / prima dal Monte Viso ’nver’ levante, / da la sinistra 
costa d’Apennino, // che si chiama Acquacheta suso, avante / che si divalli giù nel basso letto, / 
e a Forlì di quel nome è vacante, // rimbomba là sovra San Benedetto / de l’Alpe per cadere ad 
una scesa / ove dovea per mille esser recetto” (Inf. XVI.94-102). 
8 “Sempre a quell ver ch’ha faccia di menzogna / de’ l’uom chiuder le labbra fin ch’el ouote, però 
che sanza colpa fa vergogna; // ma qui tacer nol posso; e per le note / di questa comedìa, lettor, 
ti giuro, / s’elle non sien di lunga grazia vòte, // ch’i’ vidi per quell’aere grosso e scuro / venir 
notando una figura in suso” (Inf. XVI.124-133). Note also that this is the first time Dante refers 
to the Commedia as such. 
9 O tu che ne la fortunata valle / che fece Scipion di gloria reda, / quand’Anibàl co’ suoi diede le 
spalle, // recasti già mille leon per preda, / e che, se fossi stato a l’alta guerra / de’ tuoi fratelli, 
ancor par che si creda // ch’avrebber vinto i figli de la terra; / mettine giù, e non ten vegna 
schifo, / dove Cocito la freddura serra” (Inf. XXXI.115-123). The reference to Antaeus’s origins 
comes from Lucan’s Pharsalia (IV.589-594) 



Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   91 

region (differentiated as small, medium and large),10 river, sea, and lake. It must be noted that 
while waters are in the dataset, their course and extension are not visualized on the map now 
because of some technical limitations imposed by the free ArcGIS online account (as a 
workaround, they will be soon visualized by specific pins). Subsequently, we have the Italian and 
English text of the stanza where the mention occurs; if you need more context, quotations are 
followed by a link that redirects you to a page on the website where you see the passage 
highlighted within the entire Italian text of its canto.  

Then comes the speaker who mentions a place: there are 80 of them, and their list 
ordered by frequency can be seen on the chart section of the website: 173 occurrences for Dante-
narrator, 61 for Virgil, 42 for Cacciaguida, and surprisingly only 39 for Dante-character. Most 
notably, Cacciaguida speaks only in 3 cantos, Par. XV-XVII; Beatrice, Dante’s guide from Purg. 
XXX to Par. XXXIII, ranks only tenth with 14 mentions. This is a good example of how a 
geographical approach can enhance our understanding of the Commedia: as eloquently indicated 
by the figures, while Beatrice leads Dante in his cosmico-theological transcending of the 
limitation of the earthly world and its geography, in the middle of Paradiso Cacciaguida brings 
Dante back to the eminenty concrete geographical dimension of his origin, life and mission. 
These two tensions are complementary, and can be further explored by means of quantitative and 
qualitative analysis.   

The following categories in the data sheet are filled by “Yes” or a blank, and can be 
grouped in two sets. One I would call historico-cultural, and signals if a place was visited by 
Dante,11 and if it was mentioned as classical, mythological, and biblical, a tripartition not perfect 
but handy for the description of the diverse status of places.12 The other group describes 
rhetorical features of how a place is referred to in a passage: explicitly or by periphrasis, as an 
adjective, as part of a simile, and by direct speech.  

The map translates these categories into pop-up cards, one for each mention of a place, 
but first and foremost it plots a number of them into an extensive view of superimposed layers. 
The layers’ list, shown in a window on the right side of the map, is as follows: two frequency 
layers (one with scaled pins, the other as a heat map: the warmer the color the more mentioned 
an area is, see fig.5); four layers per cantica, (Inferno, Purgatorio, and Paradiso being subdivided into 
specific places, small regions, medium regions, and large regions); four historico-cultural layers 
(visited by Dante, classical, mythological, biblical), four rhetorical layers (explicit, by periphrasis, 
as an adjective, in a simile, by direct speech). Within this structure, users can toggle layers on and 
off to customize visualization, and can also zoom in and out to analyze place distribution at 
different scales. By default, the map shows only the layers based on Inferno, Purgatorio, and 
Paradiso. At the most global view (see fig. 5), this really highlights how the Commedia is both a 
Mediterranean text and a North-Central Italian one. At a closer view and with the aid of layer 
selection, more specific distributions of place-names can be studied. 

 

                                                           
10 Of course, this partition had to be done with a fair degree of approximation.  
11 To determine this value in the spreadsheet, my main reference was Inglese’s recent biography 

of Dante.   
12 This classification is similar to the categories of the search tool of The World of Dante website 

(Parker), where places are arranged by “Nature” (mythical or historical) and “Source” (biblical, 

classical, medieval). 



Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   92 

 

Fig. 5. The Mediterranean region as a heat map. The distinction between warm and cool areas 
becomes more articulate as we zoom in. Source: www.mappingdante.com.  

 

A dataset constructed by close reading has thus been turned into a geo-visualization that 
has dismembered its source text and placed it on a surface governed by the partially incompatible 
protocol of cartography. Yet through GIS technology we can reconnect the cartographical with 
the textual, insofar as we can write elements of the poem into the map through layers, pop-up 
cards, and links. The result is a map of relations: the places visualized recompose a network that 
branches off in two directions: into both text and the world. The non-cartographic visualizations 
present in the Mapping Dante website have precisely this goal: to underscore that a map is a 
provisional artifact, as the information it stores lies open to further investigations with diverse 
approaches.  

The same bundle of relations inscribed into the dataset can thus be transformed in ways 
that create new points of entry into the Commedia. A visual analysis of the connections between 
places and cantos in an interactive graph links canti according to places in common or types of 
places. These links suggest pathways of analysis through the poem, much like vocabulary, 
metaphor, characters, or historical sources have offered for previous scholarship (see fig. 6).  

http://www.mappingdante.com/


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   93 

 

Fig. 6. Interactive network visualization of the relation between cantos and place-names. The 
network was built with Gephi and then web-exported with SigmaJS. Source: 
www.mappingdante.com/network/index.html.    

 

Similarly, quantitative analysis (in charts, see fig. 7) brings to light places that are spatially 
intensive while, by the very absence of tall bars in the chart below, point to moments in each 
cantica that are not explicitly connected to diegetical or non-diegetical spaces. These patterns 
offer a rich site of new investigation into Dante's poetics and “cartographic impulse” (Cachey, 
325).  

 

 

http://www.mappingdante.com/


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   94 

Fig. 7. Chart with number of place mentioned in each canto of the Commedia. Built with Tableau 
Public. Source: www.mappingdante.com/charts/#placespercanto.  

 

The heterogeneity embraced by those relations oscillates between effacement, dictated by 
the logic of cartography, and emergence, allowed by the retrieval of textual elements and patterns 
on the map. We cannot do without the interface of modern cartography and its ontologically 
uniform surface; yet we also seek to recover that type of geographical and textual knowledge that 
was first displayed in line-by-line glossing, and that with the advent of modern maps was pushed 
out of view. 

 

Conclusion  

The potential of this type of map can be summarized in two major points: 1) it is a textual 
repertoire cartographically arranged, hence allowing users to find information on passages of the 
poem from a cartographical perspective; 2) it is a tool that, while not providing new 
interpretations by itself, enable users to observe patterns in the interrelation of geography and 
poetry, patterns that might not have been as evident without a cartographic view. Therefore, 
Mapping Dante partakes of both epistemological functions of literary maps: illustration of elements 
already at hand in the text, and interpretation of what would be otherwise invisible or elusive 
(Piatti 99). Such is indeed the double duty that GIS maps can do as mobile and tentative 
interfaces between text and world. By creating maps that make this condition manifest since they 
do not look like copies or mirrors of either world or text, we may get closer to medieval 
geography – not as if we were its contemporaries but from the inevitable vantage point of our 
own sense of geography, partly still subject to the paradigm of modern maps, partly exposed to 
the uncertainty and intricacy of post-modern spaces and places. Literary mapping is a work-in-
progress that paradoxically aspire to break free from the cartographical rules it depends on, either 
analog or digital; so, we must be ready to question technologies and tool, just like medieval 
mapmakers who were aware that maps are plastic artifacts for the visualization and manipulation 
of a culture’s relation with its world. “In depicting the world, medieval mapmakers tried to show 
the significance of places as well as their location” (Edson 164). To the restoration and 
reassessment of this significance, which is the implicit or explicit goal of any undertaking in digital 
literary cartography, Mapping Dante as a prototype tries to give its contribution. New materials and 
features will be added to both map and website in order to expand and intensify the experience 
of Dante’s geography as a field of interconnectedness. 

 

Works Cited 

Alighieri, Dante. The Divine Comedy. Ed. Charles S. Singleton. 6 vols. Princeton: Princeton 
University Press, 1970-1975. Print. 

Barolini, Teodolinda, ed. Digital Dante [new version], Columbia University, 2014. Web 17 Feb. 
2017. <http://digitaldante.columbia.edu>. 

Bartalesi, Valentina, et al. "DanteSources: a Digital Library to Study Dante Alighieri's Primary 
Sources". Umanistica digitale 1 (2017): 119-128. 
<https://umanisticadigitale.unibo.it/article/view/7250/7007>. 

http://www.mappingdante.com/
http://digitaldante.columbia.edu/
https://umanisticadigitale.unibo.it/article/view/7250/7007


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   95 

 

Bodenhamer, David J. “The Potential of Spatial Humanities.” The Spatial Humanities: GIS and the 
Future of Humanities Scholarship. David J. Bodenhamer et al., eds. Bloomington: Indiana 
University Press, 2010. 14-30. Print. 

Bosco, Umberto, ed. Enciclopedia dantesca. 6 vols. Roma: Istituto della Enciclopedia Italiana, 1970-
1978. Print. 

Boyde, Patrick. Dante Philomytes and Philosopher: Man in the Cosmos. Cambridge: Cambridge 
University Press, 1983. Print. 

Burdick, Anne, et al. Digital Humanities. Cambridge: MIT Press, 2012. Print. 

Cachey, Theodore J., Jr. “Cartographic Dante.” Italica 87:3 (2010): 325-354. Print. 

---. “Cosmology, geography, and cartography.” Dante in Context. Eds. Zygmunt G. Barański and 
Lino Pertile. Cambridge: Cambridge University Press, 2015. 221-240. Print. 

Cavalieri, Raffaella. Il viaggio dantesco: viaggiatori dell’Ottocento sulle orme di Dante. Torino: Robin 
Edizioni, 2006. Print. 

Edson, Evelyn. Mapping Time and Place: How Medieval Mapmakers Viewed the World. London: British 
Library, 1997. Print. 

Hollander, Robert, ed. Dartmouth Dante Project, Dartmouth College, 1988. Web. Feb. 17 2017. 
<dante.dartmouth.edu>. 

---. Princeton Dante Project, Princeton University, 1999. Web. 17 Feb. 2017. 
<etcweb.princeton.edu/dante/index.html>. 

Farinelli, Franco. La crisi della ragion cartografica. Torino: Einaudi, 2009. Print. 

---. Geografia. Un’introduzione ai modelli del mondo. Torino: Einaudi, 2003. Print. 

Hamilton, Louis. “Virtual Cities: GIS as a Tool for the Analysis of Dante’s Commedia.” Pedagogy 
13:1 (2013): 115-123. Print. 

Harley, J. B. “Deconstructing the Map.” Cartographica 26:2 (1989): 1-20. Print. 

Harvey, P. D. Medieval Maps. Toronto: University of Toronto Press, 1991. Print. 

Inglese, Giorgio. Vita di Dante. Una biografia possibile. Roma: Carocci, 2015. Print. 

Lollini, Massimo, and Rebecca Rosenberg. “E-philology and Twitterature”. Humanist Studies & the 
Digital Age 4 (2015): 116-163. Web. 20 Sept. 2017. 
http://journals.oregondigital.org/index.php/hsda/article/view/3634. 

Moore, Edward. “The Geography of Dante.” Studies in Dante: Third Series. Oxford: Clarendon 
Press, 1903. 109-143. Print. 

Moretti, Franco. Graphs, Maps, Trees: Abstract Models for Literary History. London: Verso, 2005. 
Print. 

http://dante.dartmouth.edu/
http://etcweb.princeton.edu/dante/index.html
http://journals.oregondigital.org/index.php/hsda/article/view/3634


Humanist Studies & the Digital Age  Gazzoni 

5.1 (2017)   96 

Parker, Deborah, ed. The World of Dante, University of Virginia, 1997. Web. 17 Feb. 2017. 
<www.worldofdante.org>. 

Piatti, Barbara. “Mapping Fiction: The Theories, Tools and Potential of Literary Cartography.” 
Literary Mapping in the Digital Age. Eds. David Cooper, Christopher Donaldson and 
Patricia Murrieta-Flores. New York: Routledge, 2016. 88-101. Print. 

Rouse, Robert Allen. “What Lies Between? Thinking Through Medieval Narrative Spatiality.” 
Literary Cartographies: Spatiality, Representation and Narrative. Ed. Robert T. Tally Jr. New 
York: Palgrave Macmillan, 2014. 13-29. Print. 

Schnapp, Jeffrey T. “Knowledge Design. Incubating new knowledge forms/genres/spaces in the 
laboratory of digital humanities”. Herrenhausen Conference, 5-7 December 2013, 
Hanover. Keynote address. Web. 20 Sept 2017. 
<https://www.volkswagenstiftung.de/uploads/media/HH_lectures_Schnapp_01.pdf>.  

Soars, Marco. “La mappa divina: tutti i luoghi citati da Dante.” Cartesiani, 30 Oct., 2013. Web. 17 
Feb. 2017.  <www.cartesiani.it/mappa-divina>.   

Travis, Charles B. Abstract Machines: Humanities GIS. New York: Esri Press, 2015. Print. 

Van Peteghem, Julie. “Digital Readers of Allusive Texts: Ovidian Intertextuality in the Commedia 
and the Digital Concordance on Intertextual Dante.” Humanist Studies & the Digital Age 4:1 
(2015): 39-59. Web. 17 Feb 2017. 
<http://journals.oregondigital.org/index.php/hsda/article/view/3584/3493>.  

Westphal, Bertrand. Geocriticism: Real and Fictional Spaces. Transl. Robert T. Tally, Jr. New York: 
Palgrave Macmillan, 2011. Print. 

Wrisley, David Joseph. “Spatial Humanities: An Agenda for Pre-Modern Research.” Porphyra 22 
(2014): 96-107. Web. 17 Feb, 2017. <www.porphyra.it/Porphyra22.pdf>. 

Wrisley, David Joseph, ed. Visualizing Medieval Places, 2013. Web. 17 Feb. 2017. 
<vmp.djwrisley.com>. 

http://www.worldofdante.org/
https://www.volkswagenstiftung.de/uploads/media/HH_lectures_Schnapp_01.pdf
http://www.cartesiani.it/mappa-divina
http://journals.oregondigital.org/index.php/hsda/article/view/3584/3493
http://www.porphyra.it/Porphyra22.pdf
http://vmp.djwrisley.com/

