Towards a digital geography ofHispanic Baroque art

Juan Luis Suarez

The CulturePlex Lab University of Western Ontario

Fernando Sancho-Caparrini

Department of Computer Science and Artificial Intelligence

University of Seville

Elika Ortega Javier de la Rosa Natalia Caldas and David Brown

The CulturePlex Lab University of Western Ontario

AbstractIn this article we propose an approach to the study of art history based on geog-raphy of Hispanic Baroque art by digital means that showcase the multiplicity ofpossible places of art Our study advances four elements of a digital geography ofart (communities semantic maps areas and flows)mdasha methodology that can beexpanded in future Digital Humanities research

1 Introduction

In Toward a Geography of Art Thomas DaCostaKaufmann (2004 p 6) stated that his researchwould lsquoinvestigate how notions of place of the geo-graphical have been inflected into writing aboutchange through time as it has been and is stilldiscussed in art historyrsquo He goes back to some ofthese ideas in his contribution to the multi-volumecatalogue of the 2010ndash11 international exhibitionPainting of the Kingdoms There he insists on thefact that political geography and artistic geographydo not coincide as countries viceroyalties nativeareas and notions of centre and periphery super-pose one another in different research works andcataloguing efforts DaCosta Kaufmann also empha-sizes the need for a theory of diffusion that helpsexplain the movements of creators paintings andfeatures from territory to territory and the effectsthese transfers have in the spatial organization of artthat experts carry out

Here we present the results of a multi-disciplin-ary collaboration in Digital Humanities ComputerScience and Art History that focuses in proposing adigital geography of Hispanic Baroque art By digitalgeography we imply the various possible organiza-tions of the place of art by digital means in amanner that connects various types of data aboutauthors and artworks with different notions ofspace This digital geography of art also takesadvantage of recent advances in data mining andvisualization to offer multiple views of the spaceof Hispanic Baroque art as related to geographymovement through territories transfers over timeand cultural borders clusters of artistic centres (asopposed to centres and peripheries) and move-ments of works from their places of origin owingto contemporary practices of collection by museumsand private collectors

The results shed light on the different ways inwhich social practicesmdashfrom creation to circulationto collectionmdashaffect the spatial organization of art

Correspondence

Juan Luis Suarez

Department of Modern

Languages and Literatures

Western University

University College Building

Room 115 London

Ontario Canada N6A 3K7

E-mail

jsuarezuwoca

Literary and Linguistic Computing The Author 2013 Published by Oxford University Press onbehalf of ALLC All rights reserved For Permissions please email journalspermissionsoupcom

1 of 18

doi101093llcfqt050

Literary and Linguistic Computing Advance Access published August 13 2013 at U

niversity of Western O

ntario on September 13 2013

httpllcoxfordjournalsorgD

ownloaded from

beyond political territories The article also showshow culturemdashdefined as information that affectshuman behaviour (Boyd and Richardson 2005)and represented here by the case of HispanicBaroque paintingsmdashorganizes different real andsymbolic lsquoplacesrsquo in different times We argue thatthe study of large-scale cultural systems such as theHispanic Baroque is better understood through acombination of tools and concepts that deal withthe complex and evolving nature of the systemand can study it through multi-scale techniquesthat reduce that complexity to a minimum offeringnew ways of arranging the space in which thatsystem unfolded over time Finally we argue thatthis methodology can be extended to other projectsin Digital Humanities

2 Methodology Data andVisualization

Over the past few years we have collected an onlineBaroqueArt Database (2010) (httpbaroqueartcul-tureplexca) consisting of gt12000 paintings andgt1500 creators associated with the territories ofthe Hispanic Monarchy from the 16th to the begin-ning of the 19th centuries The database also con-tains around 400 series 200 schools and 2500geographical locations1 On top of the data storedunder a traditional entityndashrelation model we imple-mented a system of annotations that would allow towork on the objects stored in the database and pro-vide enough flexibility to describe all aspects of anyartwork thus defining a hierarchy in a structuresimilar to an ontology From a set of gt200 descrip-tors we carried out a manual semantic annotationof all artworks (with an average of 585 descriptorswork and peaks of 14 per work) We have takenartworks with six or more descriptors as we experi-mentally checked that taking works with fewer thansix descriptors would provoke the emergence of aconsiderably larger number of modularity classeswhich would make it even more difficult to drawaccurate conclusions from the data At the sametime the possible mistakes resulting from themanual annotation of the database would have alesser impact in the analysis as the threshold to

filter works by descriptors increases Measuresobtained were as follows in Fig 1

To analyze the resulting data set we representedit as a graph in which artworks are nodes and rela-tions among them are established as a function ofthe descriptors shared by the works For example ifan artwork is described through seven descriptorsand another work is also using the same set ofdescriptors then we say that these artworks are con-nected through a weighted link of seven We limitedour experiment to the period 1550ndash1850 anddivided the global graph into 12 subgraphs tostudy the temporal evolution each of them covering25 years (Fig 2)

Then for each of the periods of our data wedetermined the clustering classes that can be con-sidered bags of lsquosimilar artworksrsquo and calculatedthe distances between classes by measuring thefrequency of descriptor usage in the artworks con-tained in the cluster We applied our own algorithmto distribute those classes in a 2D space so that theirrelative positions represent the relative distancesamong them (the closer the clusters are the moresimilar descriptors they use) We are aware that100 accuracy is impossible because of the size of

Fig 1 Descriptors and number of artworks sharing themMinimum 100 Median 600 Mean 574 StandardDeviation 154 Maximum 1400 In the extremes ofthe distribution Mean 630 Standard Deviation100

J L Suarez et al

2 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

the descriptors pool we are using which wouldrequire a higher dimensional space

Once these clusters are organized in our 2D spacewith a size proportional to the amount of artworksthey have we went back to the descriptors they con-tain and we generated the areas of influence of eachdescriptor as a potential field As it might be ex-pected owing to the ontological organization ofthe descriptors some of these areas contain otherareas or sections within themselves We also haverepresented the borderlines of the areas to showhow these intersections play out This allowed usto generate different views of the art-space takinginto account elements such as time descriptors bymodularity class or specific descriptors closelyrelated to current discussions by art historians spe-cializing in the period (Fig 3) These different viewsprovide many different facets of Hispanic Baroqueart digital geography

Furthermore we calculated distances betweensimilarity classes in different periods so that wecan infer which class evolved from previous onesand drew the semantic evolution of the artworksThis is fundamental for a better understanding ofthe generation of families of artworks and the vari-ants that this evolution produces which would help

us connect this process with explanations in polit-ical artistic or economic discourses

Finally and from the geographical informationavailable for a subset of artworks (where originallocation and current location has been determined)we were able to make a representation of artworkmovement along time and obtain informationabout how museums (currently the main artworkrepositories) and other collectors have accumulatedartworks from specific areas or those observed inprevious semantic group analysis

This methodology addressed different issuesrelated to the political geographic and cultural as-pects of art production reception and consump-tion Some of these questions are as follows Arepaintings local regional or national How differentvisualizations affect the clustering of art-worksand artists Are there differences between politicaland artistic territories What is the transmission offeatures across time and space Which is the effectof flows of artworks away from their place of origindue to market forces How different clusters of artbehave and what is their effect on centrendashperipherydebates The result is a digital geography ofHispanic Baroque art that will contribute to abetter understanding of art history from a spatial

Fig 2 Graphs showing each one of the 12 periods and the evolution of similarity clusters through three centuries

Hispanic Baroque art

Literary and Linguistic Computing 2013 3 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

point of view and will also shed light on culturaltransfers in complex systems

3 Elements of a Digital Geographyof Art Communities SemanticMaps Areas and Flows

Notions of space in art history have been tradition-ally affected by discourses of nation buildingapproaches focusing on the physicality of the art-work the specific spaces for which paintings werecommissioned or planned or by the places in whichartists lived and worked Art history as it hasgenerally happened in cultural history has had thetendency to emphasize the study of single worksand specific artistsmdashthe artist as a collection of allof their works When dealing with large politicalstructures encompassing many territories or lastingfor long periods as is the case of the HispanicMonarchy the traditional approach to the art-space is not as helpful in revealing that lsquopoliticaland artistic geography do not coincidersquo (DaCostaKaufman 2008 p 99) and that a global vision thattakes into account a universal empire that goesbeyond the notion of Spain is required

Here lsquoglobalrsquo means three interconnected thingsFirst it refers to an initial notion of space that isworld-wide in its scope and that might eventuallyextend to all corners of the world The possibility ofreaching any place in the world does not have to beactualized at every instance of the analysismdashsimply

there are spaces with no artmdashbut it has to show themechanisms susceptible to new connection pointsnot considered thus farmdasha case an event anartist a workmdashto the existing network of artisticnodes At the same time this understanding of aglobal space of art has to make clear how notionsof place are coded into the main network That is ifwe are dealing with lsquoPortuguesersquo or lsquoChristianrsquopainting in Goa we have to semantically load theedges that will connect those paintings amongthemselves and the rest of the network with the ap-propriate notions of geography Are we going to talkabout patterns of artistic diffusion Are we dealingwith centrendashperiphery relations Does our interestlie on local interactions and local transformationof exogenous elements The conclusion is that thepossible space is universal and that this lsquouniversersquohas to remain the geographic framework for specificplaces of art that will emerge through different stu-dies These places of art are the focus of our interest

lsquoGlobalrsquo also means that at least at this pointthere is no predetermined set of valid notions ofan art geography that would exhaust all possibilitiesto find and explore notions of the place of artBecause a geography of art would be connected tospecific cultural constructs and theories and conse-quently to various notions of the place of art thereis no categorical hierarchy that would cover themall In a digital geography of art we try to overcomethis problem both by avoiding the notion of a hier-archy of conceptual categories and by working at thelevel of the raw data to organize information interms of graphs The information about the

Fig 3 Twelve art-spaces from the point of view of main descriptors

J L Suarez et al

4 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

paintings authors schools etc is semanticallytagged through an ontology that can be curated bythe researcher as their objectives change from oneproject to another (Suarez et al 2012) Also the in-formation is stored and analyzed in a graph struc-ture A graph is a representation of a set of objectswhere some pairs of objects (nodes or vertices) areconnected by links or edges (Trudeau 1993) Thisallows for a dynamic process in which notions of theplace of art are connected to the specific concepts ofeach analysis notions of space as territory and thepossibilities of data analysis and visualization thatcan be unfolded by mathematical and digitaltechniques

Third lsquoglobalrsquo is still to be populated with con-cepts of cultural theory that helps us understand thelsquoglobalrsquo life of art expressions In the context of thiswork the multi-volume catalogue and the exhib-ition Painting of the Kingdoms2 revolves aroundthe idea of the lsquoshared identitiesrsquo that can be de-tected through the large pool of artistic productioncreated throughout the Hispanic Monarchy For in-stance John H Elliottrsquos chapter is anchored on thenotion of the lsquokingdomsrsquo and the idea of a compos-ite monarchy in which the total was bigger and dif-ferent than the sum of its parts The kingdoms arethe lsquolocirsquo of his analysis in an effort to show both thediversity and the unity of a complex structure(Suarez et al 2007 and Elliott 2008 p 46) ForJuana Gutierrez Haces (2008 p137) the notion oflsquokoinersquo is the thread that allows for a better under-standing of the history of art in the HispanicMonarchy3 Her objective is to explain lsquohow thepainterrsquos mentality in the Spanish realms wasshaped vis-a-vis a process known as koine or level-ing [ ] This process consisted of shedding theunique features of each contributor in favour ofwhat they all shared The purpose was to create anew language and to foster a sense of belonging to agroup as part of adjusting to a new realityrsquo Anotherexample of the different ways of looking at theglobal nature of this type of production is adoptedby Helga von Kugelgen in her systematic study ofthe way Rubensrsquo influence extended across the king-doms in what becomes an incredible source to studythe patterns of cultural diffusion and imitation withreal data in a real case (von Kugelen 2008)

These are just a few examples of what could be apossible geography of art that according to DaCostaKaufmann (2008 p 88) lsquoaddresses questions such ashow is art related to determined by or determines oris affected by or affects the place in which it is madehow art is identified with people culture regionnation or state or combinations of these how artin various places is to be interrelated throughspread contact and circulation and how areas ofstudy are to be definedrsquo4 Given the nature of theseproblems we propose that a digital geography ofHispanic Baroque art would positively influence thedifferent ways in which we perceive relations betweenplace and object and it would shed light on how toarrange those relations through digital means to pro-vide answers to the different issues described above

A digital geography of Hispanic Baroque art isconcerned with the various possible organizations ofthe place of art by digital means in a manner thatconnects various types of data about authors andartworks with different notions of space There aretwo foundations to this approach First we adoptRicherson and Boydrsquos definition of culture as lsquoinfor-mation capable of affecting individualsrsquo behaviourthat they acquire from other members of their spe-cies through teaching imitation and other forms ofsocial transmissionrsquo (2006 p 5) By adopting thisdefinition we are able to deal with information asdata that are encoded that moves and transfersfrom place to place and that are cultural informa-tion because they affect the behaviour of humanindividuals in a way that we can trace and model(Suarez et al 2011)

As DaCosta Kaufmann (2008 p 96) has high-lighted for the case of art views of geographic pro-cesses are entangled with the notion of culturaltransfer5 It is only by unearthing the networksthat allow for information to be transferredamong individuals through time and space(McNeill and McNeill 2003 Castells 2009DiMaggio 2011) that we can rigorously explainhow cultural transfers (Goodenough 2002 andDaCosta Kaufman 2008 p 96) take place andhow they affect artistic production in different set-tings Hence when we talk about cultural transferswe assume that lsquomental representations are non-dis-crete cultural transmission is highly inaccurate and

Hispanic Baroque art

Literary and Linguistic Computing 2013 5 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

mental representations are not replicated but ratherreconstructed through an inferential processrsquo(Heinrich and Boyd 2002 p 87) so that it is likelythat cultural transfers do not happen by exact rep-lication Cultural transmission requires externalstorage where information is ready to be accesseddecoded and replicated in different media and con-texts however inexact this replication might be

In a digital geography of art the cultural objectremains at the centre of all emergent spaces as thesespaces digitally recreate possible arrangements of theplace of artmdashlived spaces in which individuals andgroups experience art according to economic needspressures that require fostering religious prosocialityor aesthetic pleasure The cultural object links theauthor with its audience known or unknown andalso connects the artwork with the means of culturaltransfer Finally the various visualizations of thosespaces allow for new categorizations of the artisticproduction and for emergent meanings of art

The first method of a digital geography of artproduces cultural communities as a result of theclusterization and visualization of the data fromthe Baroque Art Database around modularityclasses One of the more usual forms of graphanalysis searches for modularity classes or how anetwork decomposes into modular communitiesor subnetworks with actual meaning in the realworld that they represent These data communitiesrespond to fundamental questions about the forma-tion and maintenance of cultural communities ForSperber and Hirschfeld (2004 p 40) a lsquoculturalgroup is held together by a constant flow of infor-mation most of which is about local transient cir-cumstances and not transmitted much beyond them[ ] Culture refers to this widely distributed infor-mation its representations in peoplersquos minds andits expressions in their behaviors and interactionsrsquoThese communities of data resulting from the ana-lysis of the graph show how the flows of informationof the paintings from the database get reorganizedover time They also show how these flows give riseto other communities that emerge as the effect ofthe information shared by the artworks and used byindividuals and groups in different contexts

The fact that we can demonstrate the existence ofa constant flow of shared information over a long

period leads to the issue of the sustainability ofpolitical and cultural communities across time andspace Our analysis of the religious informationcontent carried out within the network of baroquepaintings in the Hispanic Monarchy proves that aglobal communitymdashhowever fragilemdashwas formedas a result of the European expansion into theAmericas and that it was possible thanks to thecommon religiousmdashCatholicmdashcontent carriedwithin the paintings in the network (Suarez et al2012) The graphs also show that the cultural com-munity was not homogeneous as the different datacommunities change their shape and get trans-formed over time owing to specific artistic politicaland socioeconomic circumstances (Suarez et al2011)

It is interesting to note that the communities arenot necessarily political and that the concept of geo-graphic space does not apply in many of the cases asit is the change of semantic descriptors over timeand the different periods that show how the com-munities are formed and reorganized according to amultiplicity of factors and the combination of se-mantic tags that describe the paintings In thisregard we borrow Gutierrez de Hacesrsquo (2008) con-cept of lsquokoinersquo or process of leveling in New Spainrsquospainting and retool it to express the many differentprocesses of leveling that actually take place not onlybetween Spain and New Spain but also with regionsand periodsmdashcultural areas as territorial inser-tionsmdashin New Spain

This takes us to the second element of a digitalgeography of Hispanic Baroque art that of semanticmaps Semantic maps are strategies to represent themultiple relations of concepts or in our case thesemantic descriptors that make up the ontology weuse to classify the paintings in our data set They areespecially useful when there are many possible rela-tions and also many items to be compared with oneanother They also have the advantage of showingthe knowledge associated with the descriptors

By using semantic maps instead of a traditionalcategorization around genres we exploited thepower of the graph structure the multitude of con-nections that paintings have in the real place(s) ofart and the temporality inscribed in the data struc-ture for our artworks In Graphs Maps Trees

J L Suarez et al

6 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

Franco Moretti (2005 p 14) explains that genres arelsquotemporary structures [ ] morphological arrange-ments that last in time but always only for sometimersquo However the problem with genres is thatthey become closed structures with definite bound-aries that end up forcing live works into dead cate-gories Also genres tend to become permanentboxes as they encapsulate forms to make them notchange over time However paintings change overtime as a result of cultural transfer both within thesame cultural areasmdashcultural replication is notexactmdashand across cultural bordersmdashwhen localtraditions and external forces clash and strive fornew synthesis

To reflect these changes we propose that a mostefficient way to represent the changing descriptorsthat artworks share is through a semantic organiza-tion that includes the features in such a dynamicway that gets the best of an ontology structuringour graph By using semantic maps we capturedthe two dimensions of genres as Moretti refers tothem we represented lsquoformrsquo but we did it by look-ing at History at the changes these forms undergoover time and space

These changing relationships within the semanticcontent of our data set are shown in the evolution ofthe main descriptors that result from our analysisOf special interest are the relationships betweenpaintings with the descriptor lsquoreligiousrsquo and thosethat are described as lsquocivilrsquo Although lsquoreligiousrsquo isover time the most abundant descriptor as it relatesto the large majority of works in the data set we seethat as the 18th century advances the relation isinverted and lsquocivilrsquo starts to take over in absolutenumbers to the point that it becomes the mostused descriptor of the database at the beginning ofthe 19th century Also noticeable is the constantincrease of works related to lsquoportraitrsquo one of themost understudied themes in Hispanic Americanpainting Although we have explained in politicaland historical terms the causes of these variations(Suarez et al 2011a) what we want to highlight isthat these relations do not imprison the works intosingle and exclusive categories On the contrarycomparison through semantic maps showcases therichness of the information contained in the paint-ings the multiple approaches available for its

analysis and the mechanisms by which culturalinformation gives life to different communities indifferent or the same cultural areas

Cultural areas are another important element inthe digital geography of art We define a culturalarea as a virtual or concrete space organized throughthe same information technology and a flow ofcommon culture shared by a population in variousdegrees An interesting thing about cultural areas isthat once a researcher has collected enough infor-mation about a cultural phenomenon the informa-tion itself gets organized in many different waysvis-a-vis the experiences of various groups andeven the standpoint of the researcher This iswhen DaCosta Kaufmannrsquos statement about the dis-agreements of the political and artistic geographiescomes true as there are as many geographies of artas cultural areas Cultural areas are also relevantwhen dealing with the concept and materiality ofcultural transfers This is especially important inthe case of Baroque art As Llewellyn and Snodinhave pointed out (2009 p 20) lsquothe Baroque meansmany different things even across the visual culturesof Western Europe depending on the date and thecharacter of the work of art under considerationThere is no convincing Baroque Zeitgeist in the full-est sense argued by the great cultural historianJakob Burckhardt nor does Wolfflinrsquos model ofthe Baroquemdashas a reaction against Renaissancemdashalways apply We present the Baroque as a complexstage in the development of the post-Renaissanceclassical language of design and we explore itthrough themes such as assemblage and synthesisthe visual exploration of the physical space the il-lusion of movement and naturalistic ornamentCommon to nearly all the works of art discussedis that they result from the transmission of peopleideas motifs or materialsrsquo When we includeHispanic America and Asia to this picture theBaroque becomes many more things than just anartistic language

Cultural areas allow for the study of similaritiesand degrees of difference As the graphs show theinitial differences and gaps between the artistic pro-duction in Spain (blue and dark green clusters)Mexico (purple) and Peru (dark violet) at the be-ginning of the Baroque period 1550ndash1650 (Fig 4)

Hispanic Baroque art

Literary and Linguistic Computing 2013 7 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

get transformed in the second period 1650ndash1750thanks to the homogenization caused by the excessof religious content in the artistic information(Fig 5) As historical and political circumstancesrelated to the independences of Latin American na-tions affect the production of paintings we see thatthe size number and composition of clusterschange and that while there are certain types ofcontent shared by all three political entities (Spain

New Spain and Peru) others diverge to becomerelevant only in certain territories or becomerelated to newly formed cultural areas (Fig 6) Asthe political geography changes so does the artisticgeography even with much more detail when ana-lyzed and represented digitally But the opposite isalso true as we change the focus of the artistic geog-raphy certain concepts of political geography donot hold for this kind of material

Fig 4 1550ndash1650 period

J L Suarez et al

8 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

In cultural areas we find traces of communica-tive exchanges which are also the stage in whichcycles of cultural change take place Cultural areasrespond to the mechanisms that Sassen (2006p 418) has explained for territorial insertionswhich do not necessarily entail subsumption underexclusive state authority because they are predicated

on specific denationalization in laws and policy inthe service of a global regime These processes ofmulti-authorities used by Sassen to describe the cur-rent wave of globalization have also been well stu-died for the case of the first globalization and theHispanic Monarchy John H Elliottrsquos article quotedabove lsquoOne King Many Kingdomsrsquo explains how

Fig 5 1650ndash1750 period

Hispanic Baroque art

Literary and Linguistic Computing 2013 9 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

the political articulation the legal codes that rule therelations between political entities and even thetraditional customs and allegiances would varyfrom territory to territory depending on the agree-ments achieved between the Monarchy and the localelites The complexity of the political structure isonly a reflection of the even more complex weavingand unweaving of culture that results in commu-nities and areas that share common experiencesand lived spaces As Sassen points out (2006 p3)

lsquo[these processes of globalization] are multisidedtransboundary networks and formations whichcan include normative orders they connect subna-tional or national processes institutions and actorsbut not necessarily through the formal interstatesystemrsquo

One can analyze and visualize the data of theHispanic Baroque Database through the lens ofthe histories of the nation states This is whatDaCosta Kaufmann (2004 p 99) has called the

Fig 6 1750ndash1850 period

J L Suarez et al

10 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

national model of art history one in which thegeography of art gets constrained by the politicalborders of the political entity that serves as the con-tainer of the artistic production When theories ofdiffusion are combined with the national modelconceptual variants around the lsquoindigenousrsquo andthe lsquohybridrsquo or lsquomestizorsquo show different aspects ofthe same national production In these cases thenation becomes more inclusive However we cansee that there is much more richness to be exploredif we apply the concept of cultural area to regionssuch as Mexico City (Fig 7) where the weight thatMexico has in the artistic production of a highlypopulated area is evident as well as the links thatconnect Mexico to Puebla in terms of proximityrivalry or themes

In another case Oaxaca during the same periodof 1750ndash75 (Fig 8) we zoomed in the visualizationand observed ways in which paintings from Oaxacacould be connected through geographic means toCentral America and the Andean region as well astheir Mexican siblings There are possibly many cul-tural areas that fitting into the definition of livedspaces of art are not given the same kind of atten-tion or are wrapped up under the same political or

historiographical concepts or standpoints that applyto better established areas

The internal diversity of cultural areas (DaCostaKauffman 2004 p 99) is also an important theme of adigital geography of art If we connect our culturalareas through the creators and we search for themthroughout the whole territory of the HispanicMonarchy we can see that over time there is a greatshift in the most important nodes of the artistsrsquo net-work In the period 1550ndash1650 (Fig 9) VicenteCarducho (mainly in Spain) Peter Paul Rubens6

(von Kugelen 2008) and the anonymous paintersare the nodes with most connections In the finalperiod of 1750ndash1850 (Fig 10) it is a single Mexicanpainter Miguel Cabrera who gets all the attentionand becomes the most influential at both sides ofthe Atlantic This begs the question of what alsquoHispanicrsquo history of art around the great influencersand diffusers such as Rubens and Cabrera wouldlook like In the period of 1650ndash1750 (Fig 11) a var-iety7 of artists exert their influence around differentcultural areas semantic descriptors and techniques

Jonathan Brown (1999) talked about theHispanic Monarchy as a triptych in which art influ-ences would commence in the Low Countries

Fig 7 Mexico as cultural area in 1750ndash75 in proximity to Puebla as an area of influence Screenshot generated from thesource tool baroqueartcultureplexca

Hispanic Baroque art

Literary and Linguistic Computing 2013 11 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

would move to the Iberian Peninsula and fromthere would sail the Atlantic to reach and impactthe creation of American art (Fig 12) On each partof the triptych we would have schools authors andlocal contexts that would interact with the incomingflows of information Brownrsquos intuition joins a trad-ition of historiography that has tried the boundariesof the national model and has dealt with larger andlarger geographic areas and time periods A well-known and influential case is the Mediterranean his-tory that Ferdinand Braudel delivers in his famousbook on the subject (1972) During the past fewdecades similar and diverse efforts of writing anAtlantic history from Columbus onwards havebeen tried by Elliott (2006) Lucena (2010) andCanizares-Esguerra (2006) just to name a fewand more recently by collective enterprises such asthe Painting of the Kingdoms research project Evenbolder is the intellectual venture that DavidChristian (2011) is developing around the notionof lsquobig historyrsquo8 a history that starts with the BigBang and is yet to be finished

In all these cases the traditional research meth-ods of the humanities clash with the amount ofinformation needed to make sense select and con-textualize the events that will give shape to thosehistories Key to all these efforts is the concept offlows of information that is the streams that con-tinuously carry cultural information from one loca-tion to another either through the movement ofhuman beings or through the movement of culturalitems that at some point will be decoded and used ina different location from the place of creation Flowsof information respond to a general view of the waycurrents of culture cross borders whether this cul-tural information is adopted by locals in its newdestiny or not as in the case of collections inmodern museums As opposed to cultural transfersin which we assume an immediate interaction be-tween local and exogenous agents flows of informa-tion can take many forms and are telling about highand directed volumes of information

As a last example in this introduction to a digitalgeography of Hispanic Baroque Art we show how

Fig 8 Oaxaca as a cultural area in the context not only of Mexico but of Central America and the Andean region1750ndash75 through which influence flows Screenshot generated from the source tool baroqueartcultureplexca

J L Suarez et al

12 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

current digital techniques allowed us to identifyongoing large flows of information and situate thecultural objectsmdashthe paintingsmdashin contexts andhistories different from those in which they origi-nated It is fair to say that these flows show howdifferent the lived places of art can be from oneperiod of human history to another They also con-firm that culture is an ever shape-changing organ-ism that can serve different purposes in differentcultural contexts and that it is better studiedthrough digital tools focusing on complex systemsanalysis

We performed a query of our database takinginto account the place of origin of the paintings

(the first documented location when they were cre-ated or the original place for which they were com-missioned) and also the current place in which theyare held today and then we visualized the results ona map with the origin in red and the current loca-tion in green The result is the map in Fig 13 inwhich we have huge flows of artistic informationtaking place over time

This visualization shows that most of the flowshave happened from America (Mexico and Peru) toEurope and to a lesser extent to North America Ina few cases the transfers have happened betweenPeru and Mexico and more frequently within re-gions of these countries These flows can be

Fig 9 Carducho Rubens and anonymous are most prominent in the period of 1550ndash1650

Hispanic Baroque art

Literary and Linguistic Computing 2013 13 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

interpreted in many different ways such as the oneproposed by Baez (2009) or in colonial and postco-lonial terms and a variety of theories can be used toexplain these economic and cultural phenomenaWhat interests us in this work is to show how theapplication of data analysis and visualization exposethese flows and create another possible chapter in adigital geography of art In this case the flows callfor the study of communities and cultural areas dif-ferent from those we analyzed earlier in the articlewhen talking about the historical Baroque periodThese communities and areas respond to differentcriteria and are now related to the contemporaryhistory of the museum the global art market orthe postmodern geographies of a postcapitalistworld All of them connect to the various storiesthat can be told through a digital geography ofHispanic Baroque art

4 Conclusions The Lived Spacesof the Hispanic Baroque

We argue that the study of large-scale cultural sys-tems such as the Hispanic Baroque is better tackledby a combination of tools and concepts that dealwith the complex and evolving nature of thesystem and can be studied through multi-scaletechniques that reduce that complexity to a min-imum offering new ways of arranging the space inwhich that system unfolded over time

A digital geography of art is a viable way of deal-ing with such complex systems of culture A digitalgeography of art encompasses the various possibleorganizations of the place of art by digital means ina manner that relates different types of connecteddata about authors and artworks to different

Fig 10 Cabrerarsquos production is prominent over other artistsrsquo in the 1750ndash1850 period

J L Suarez et al

14 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

notions of space and to a variety of problems abouthuman culture The places of art become multiple inas much as they are considered in terms of whatSoja called lsquolived spacesrsquo (1996)9 of art thirdspaces of cultural transitions that tell different stor-ies about the human groups that have createdexperienced and lived through that art

We have provided an initial list of elements of adigital geography of art communities areas seman-tic maps and flows This list can be expanded ormodified according to the data set the findings inthe data and the interests of the researcher Theelements of a digital geography of art serve themethodological purpose of showing the multiple

Fig 11 The period (1650ndash1750) in which a larger variety of artists were exerting their influence at the same time

Hispanic Baroque art

Literary and Linguistic Computing 2013 15 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

Fig 12 Brownrsquos Triptych of Hispanic Baroque Painting illustrating the flow of art across the Atlantic Screenshotgenerated from the source tool baroqueartcultureplexca

Fig 13 Flow of artworks from their original production place (in red) to their current holding locations mostlymuseums galleries and private collections (in green)

J L Suarez et al

16 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

power structures acting in a locality at a given timehow human groups activate cultural works in dif-ferent contexts that is how the art space is lived ineach case and how cultural information forms anetwork that reflects the complexity of human lifein the global age

Funding

This work was supported by the Social Sciences andHumanities Research Council of Canada and theCanada Foundation for Innovation

ReferencesBaez F (2009) El Saqueo Cultural de America Latina De

la Conquista a la Globalizacion Barcelona RandomHouse Mondadori

BaroqueArt Database (2010) Cultureplex Universityof Western Ontario httpbaroqueartcultureplexca(accessed 16 November 2012)

Boyd R and Richerson PJ (2005) Solving the Puzzle ofHuman Cooperation In Levinson S (ed) Evolutionand Culture Cambridge MA MIT Press pp 105ndash132

Braudel F (1972) The Mediterranean and theMediterranean World in the Age of Philip IINew York Harper and Row

Brown J (1999) Los Siglos de Oro en los Virreinatos deAmerica 1500-1700 In Catalogo de la exposicion cele-brada en el Museo de America entre el 23 noviembre de1999 y el 12 de febrero de 2000 Madrid Sociedad Estatalpara la Conmemoracion de los Centenarios de FelipeII y Carlos V

Canizares-Esguerra J (2006) Puritan ConquistadorsIberianizing the Atlantic 1550-1700 StanfordStanford University Press

Castells M (2009) Communication Power OxfordOxford University Press

Christian D (2011) Maps of Time An Introduction toBig History Berkeley University of California Press

DaCosta Kaufmann T (2004) Toward a Geography ofArt Chicago University of Chicago Press

DaCosta Kaufmann T (2008) Pinturas de los reinos AGlobal View of the Cultural Field In Gutierrez Haces J(ed) Painting of the Kingdoms Shared IdentitiesTerritories of the Spanish Monarchy vol 1 MexicoDF Fomento Cultural Banamex pp 86ndash136

DaCosta Kauffman T (2010) Interpreting CulturalTransfer and the Consequences of Markets andExchange Reconsidering Fumi-e In Artistic andCultural Exchanges between Europe and Asia1400-1900 Burlington Ashgate pp 134ndash161

DiMaggio P (2011) Cultural Networks In Scott J andCarrington PJ (eds) The SAGE Handbook of SocialNetwork Analysis Los Angeles SAGE pp 286ndash300

Elliott J H (2006) Empires of the Atlantic World Britainand Spain in America 1492-1830 New Haven YaleUniversity Press

Elliott J H (2008) One King Many Kingdoms InGutierrez Haces J (ed) Painting of the KingdomsShared Identities Territories of the Spanish Monarchy16thndash18th Centuries vol 1 Mexico DF FomentoCultural Banamex pp 40ndash83

Goodenough OR (2002) Information Replication inCulture Three Modes for the Transmission ofCulture Elements through Observed Action InNehaniv C L and Dautenhahn K (eds) Imitationin Animals and Artifacts Cambridge MA MIT Pressp 573

Gutierrez Haces J (2008) Painting in New Spain asAmerican Pictorial Koine Progress in OngoingResearch In Gutierrez Haces J (coord) Painting ofthe Kingdoms Shared Identities Territories of theSpanish Monarchy 16thndash18th Centuries vol 1Mexico DF Fomento Cultural Banamex pp 136ndash185

Flores Flores O and Fernandez Flores L (2008)Pictorial Koineization in the Realms of the SpanishMonarchy In Gutierrez Haces J (ed) Painting of theKingdoms Shared Identities Territories of the SpanishMonarchy 16th-18th Centuries vol 1 Mexico DFFomento Cultural Banamex pp 186ndash335

Heinrich J and Boyd R (2002) On ModelingCognition and Culture Journal of Cognition andCulture 2(2) 87 httpwwwswetswisecomeAccessviewAbstractdoarticleIDfrac1427381897amptitleIDfrac14111670(accessed 16 November 2012)

Llewellyn N and Snodin M (eds) (2009) Baroque1620-1800 Style in the Age of Magnificence LondonVampA Publishing

Lucena Geraldo M (2010) Naciones de Rebeldes LasRevoluciones de Independencia Latino Americanas ALos Cuatro Vientos Las Cuidades de la AmericaHispanica Madrid Santillana Ediciones Generales SL

McNeill J R and McNeill W H (2003) The HumanWeb A Birdrsquos Eye View of Human History New YorkNorton

Hispanic Baroque art

Literary and Linguistic Computing 2013 17 of 18

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from

Moretti F (2005) Graphs Maps Trees Abstract Modelsfor Literary History New York Verso

Page S (2011) Diversity and Complexity PrincetonPrinceton University Press

Richerson P J and Boyd R (2006) Not By Genes AloneHow Culture Transformed Human Evolution ChicagoUniversity of Chicago Press

Soja E (1996) Thirdspace Journeys to Los Angelesand Other Real-and-Imagined Spaces CambridgeBlackwell

Sperber D and Hirschfeld L (2004) The CognitiveFoundations of Cultural Stability and DiversityTrends in Cognitive Science 8(1) 40ndash46 httpresolverscholarsportalinforesolve13646613v08i000140_tcfocsad (accessed 16 November 2012)

Suarez J L and Olid-Pena E (2007) Hispanic BaroqueA Model for the Study of Cultural Complexity in theAtlantic World South Atlantic Review 72(1) 31ndash47httpwwwjstororgstable27784678 (accessed 15November 2012)

Suarez J L and Sancho F (2011a) A VirtualLaboratory for the Study of History and CulturalDynamics Journal of Artificial Societies and SocialSimulation 14(4) 19 httpjassssocsurreyacuk14419html (accessed 16 November 2012)

Suarez J L and Sancho F (2011b) Toward a DescriptorsBased Ontology of the Baroque System of Culture GeneralConcepts and Its Applications to the Classification ofPainting Submitted Cultureplex University ofWestern Ontario and Universidad de Sevilla

Suarez J L Sancho F and de la Rosa J (2011) TheArt-Space of a Global Community The Network ofBaroque Paintings in Hispanic-America 2011 SecondInternational Conference on Culture and ComputingKyoto Japan October 2011 httpcultureplexcamediapublications5The_art_space_of_a_global_community_the_network_of_Baroque_paintings_in_Hispanic_Americapdf (accessed 16 November 2012)

Suarez J L Sancho F and de la Rosa J (2012)Sustaining a Global Community Art and Religion inthe Network of Baroque Hispanic-American PaintingsLeonardo 45(3) 281 httpwwwmitpressjournals

orgdoiabs101162LEON_a_00374journalCodefrac14

leon (accessed 16 November 2012)

Trudeau R J (1993) Introduction to Graph TheoryCorrected enlarged republication ed New York Dover

von Kugelen Helga (2008) lsquolsquoPainting from the Kingdoms

and Rubensrsquorsquo In Gutierrez Haces J (ed) Painting ofthe Kingdoms Shared Identities Territories of the SpanishMonarchy 16thndash18th Centuries vol 3 Mexico DF

Fomento Cultural Banamex pp 1008ndash1078

Notes1 For a detailed explanation of the methodology please

refer to Juan Luis Suarez Fernando Sancho and Javierde la Rosa (2011b and 2012)

2 The four-volume catalogue Painting of the Kingdoms

Shared Identities were the result of the exhibition ofthe same name held from March 9 to Aug 31 2011 at

Palacio de Cultura Banamex in Mexico City Websitehttpfomentoculturalbanamexorgpinturadelosrei-nosantecedenteshtml

3 In the same catalogue Oscar Flores Flores and LigiaFernandez Flores (2008) apply Gutierrez Hacesrsquo koinemodel to the various kingdoms of the Monarchy by

exploiting the linguistic analogy and using the idea ofidentity along dialectal varieties

4 See also the introductory chapter to his Toward aGeography of Art

5 See also DaCosta Kauffman on cultural transfers

Interpreting Cultural Transfer and the Consequencesof Markets and Exchange (2010)

6 Rubensrsquo work was present practically in all territories of

Europe and America due to the spread of copies andengraving books

7 See Scott Page (2011) on diversity and complexity8 See also bighistoryprojectcom9 For Soja lsquoSpatiality [ie Socially produced space] is

a substantiated and recognizable social product

part of a lsquolsquosecond naturersquorsquo [the transformed andsocially concretized spatiality arising from the applica-

tion of purposeful human labor] which incorporatesas it socializes and transforms both physical and psy-chological spacesrsquo

J L Suarez et al

18 of 18 Literary and Linguistic Computing 2013

at University of W

estern Ontario on Septem

ber 13 2013httpllcoxfordjournalsorg

Dow

nloaded from