7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display

1/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
1
2
DHQ:DigitalHumanitiesQuarterly
2011
Volume5Number1
HumanitiesApproachestoGraphicalDisplay
JohannaDrucker<drucker_at_gseis_dot_ucla_dot_edu>,BreslauerProfessorofBibliographicalStudiesDepartmentof
InformationStudies,UCLA
Abstract
As digital humanists have adopted visualization tools in their work, they have borrowed
methods developed for the graphical display of information in the natural and social
sciences.Thesetoolscarrywiththemassumptionsofknowledgeasobserverindependent
andcertain,ratherthanobservercodependentandinterpretative.Thispaperarguesthat
weneedahumanitiesapproachtothegraphicalexpressionofinterpretation.Tobegin,the
concept of data as a given has to be rethought through a humanistic lens and
characterizedascapta,takenandconstructed.Next,theformsforgraphicalexpressionof
capta need to be more nuanced to show ambiguity and complexity. Finally, the use of a
humanistic approach, rooted in a codependent relation between observer and
experience,needstobeexpressedaccordingtographicsbuiltfrominterpretativemodels.
In summary: all data have to be understood as capta and the conventions created to
express observerindependent models of knowledge need to be radically reworked to
expresshumanisticinterpretation.
Introduction
Asdigitalvisualizationtoolshavebecomemoreubiquitous,humanistshaveadoptedmanyapplicationssuchas
GISmapping,graphs,andchartsforstatisticaldisplaythatweredevelopedinotherdisciplines.But,Iwillargue,
such graphical tools are a kind of intellectual Trojan horse, a vehicle through which assumptions about what
constitutesinformation swarm with potent force. These assumptions are cloaked in a rhetoric taken wholesale
from the techniques of the empirical sciences that conceals their epistemological biases under a guise of
familiarity.SonaturalizedaretheGooglemapsandbarchartsgeneratedfromspreadsheetsthattheypassas
unquestionedrepresentationsof"whatis".Thisisthehallmarkofrealistmodelsofknowledgeandneedstobe
subjectedtoaradicalcritiquetoreturnthehumanistictenetsofconstructednessandinterpretation to the fore.
Realist approaches depend above all upon an idea that phenomena are observerindependent and can be
characterized as data. Data pass themselves off as mere descriptions of a priori conditions. Rendering
observation(theactofcreatingastatistical,empirical,orsubjectiveaccountorimage)asifitwerethesameasthe
phenomena observed collapses the critical distance between the phenomenal world and its interpretation,
undoingthebasisof interpretation on which humanistic knowledge production is based. We know this. But we
seem ready and eager to suspend critical judgment in a rush to visualization. At the very least, humanists
beginning to play at the intersection of statistics and graphics ought to take a detour through the substantial
discussionsofthesociologyofknowledgeanditsdevelopedcritiqueofrealistmodelsofdatagathering
[1]
Atbest,
weneedtotakeonthechallengeofdevelopinggraphicalexpressionsrootedinandappropriatetointerpretative
activity.
Becauserealistapproachestovisualizationassumetransparencyandequivalence,asifthephenomenalworld
were selfevident and the apprehension of it a mere mechanical task, they are fundamentally at odds with
approaches to humanities scholarship premised on constructivist principles. I would argue that even for realist
models,thosethatpresumeanobserverindependentrealityavailabletodescription,themethodsofpresenting
ambiguityanduncertaintyinmorenuancedtermswouldbeuseful.Somesignificantprogressisbeingmadein
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
2/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
3
4
5
6
7
visualizing uncertainty in data models for GIS, decisionmaking, archaeological research and other domains.
[2]
Butanimportantdistinctionneedstobeclearfromtheoutset:thetaskofrepresentingambiguityanduncertainty
hastobedistinguishedfromasecondtaskthatofusinginterpretationsthatariseinobservercodependence,
characterized by ambiguity and uncertainty, as the basis on which a representation is constructed. This is the
difference between putting many kinds of points on a map to show degrees of certainty by shades of color,
degrees of crispness, transparency etc., and creating a map whose basic coordinate grid is constructed as an
effect of these ambiguities. In the first instance, we have a standard map with a nuanced symbol set. In the
second,wecreateanonstandardmapthatexpressestheconstructednessofspace.Bothrelyonrethinkingour
approachtovisualizationandtheassumptionsthatunderpinit.
Tooverturntheassumptionsthatstructureconventionsacquiredfromotherdomainsrequiresthatwereexamine
theintellectualfoundationsofdigitalhumanities,puttingtechniques of graphical display on a foundation that is
humanistic at its base. Thisrequires first and foremost that we reconceive all data as capta. Differences in the
etymologicalrootsofthetermsdataandcaptamakethedistinctionbetweenconstructivistandrealistapproaches
clear.Captais"taken"activelywhiledataisassumedtobea"given"abletoberecordedandobserved.Fromthis
distinction,aworldofdifferencesarises.Humanisticinquiryacknowledgesthesituated,partial, and constitutive
characterofknowledgeproduction,therecognitionthatknowledgeisconstructed,taken, not simply given as a
naturalrepresentationofpreexistingfact.
Mydistinctionbetweendataandcaptaisnotacovertsuggestionthatthehumanitiesandsciencesarelockedinto
intellectualopposition,orthatonlythehumanistshavetheinsightthatintellectualdisciplinescreatetheobjectsof
theirinquiry.Anyselfconscioushistorianofscienceorclinicalresearcherinthenaturalorsocialsciencesinsists
thesameistruefortheirwork.Statisticiansareextremelysavvyabouttheirartifices.Socialscientistsmaydivide
between realist and constructivist foundations for their research, but none are nave when it comes to the
rhetoricalcharacterofstatistics.Thehistoryofknowledgeisthehistoryofformsofexpressionofknowledge,and
thoseformschange.Whatcanbesaid,expressed,representedinanyeraisdistinctfromthatofanyother,withall
the attendant caveats and reservations that attend to the study of the sequence of human intellectual events,
keeping us from any assertion of progress while noting the facts of change and transformation. The historical,
criticalstudyofscienceisasfullofdiscussionsofthismaterialasthehumanities.
Thustherepresentation of knowledge is as crucial to its cultural force as any other facet of its production. The
graphicalformsofdisplaythathavecometotheforeindigitalhumanitiesinthelastdecadeareborrowedfroma
mechanistic approach to realism, and the common conception of data in those forms needs to be completely
rethought for humanistic work. To reiterate what I said above, the sheer power of the graphical display of
"informationvisualization"(anditsnoveltywithinahumanitiescommunitynewlyenthralledwiththetoysofdata
mining and display) seems to have produced a momentary blindness among practitioners who would never
toleratesuchliteralassumptionsintextualwork.
ThepolemicIsetforthhereoutlinesseveralbasicprinciplesonwhichtoproceeddifferentlybysuggestingthat
whatisneededisnotasetofapplicationstodisplayhumanities"data"butanewapproachthatuseshumanities
principlestoconstitutecaptaanditsdisplay.Atstake,asIhavesaidbeforeandinmanycontexts,istheauthority
of humanistic knowledge in a culture increasingly beset by quantitative approaches that operate on claims of
certainty. Bureaucracies process human activity through statistical means and when the methods grounded in
empiricalsciencesareputattheserviceofthesocialsciencesorhumanitiesinacrudelyreductivemanner,basic
principlesofcriticalthoughtareviolated,orattheveryleast,puttoofartotheside.Tointerveneinthisideological
system,humanists,andthevaluestheyembraceandenact,mustcounterwithconceptualtoolsthatdemonstrate
humanities principles in their operation, execution, and display. The digital humanities can no longer afford to
take its tools and methods from disciplines whose fundamental epistemological assumptions are at odds with
humanisticmethod.
Thispaperisacalltoimaginativeactionandintellectualengagementwiththechallengeofrethinkingdigitaltools
forvisualizationonbasicprinciplesofthehumanities.Itaketheseprinciplestobe,first,thatthehumanitiesare
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
3/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
8
9
10
committedtotheconceptofknowledgeasinterpretation,and,second,thattheapprehensionofthephenomenaof
the physical, social, cultural world is through constructed and constitutive acts, not mechanistic or naturalistic
realistrepresentationsofpreexistingorselfevidentinformation.Nothinginintellectuallifeisselfevidentorself
identical, nothing in cultural life is mere fact, and nothing in the phenomenal world gives rise to a record or
representationexceptthroughconstructedexpressions.Therhetoricalforceofgraphicaldisplayistooimportanta
fieldforitsdesigntobeadoptedwithoutcriticalscrutinyandthefullforceoftheoreticalinsight.Letmesuggest
whatthatmeansforthevisualizationofinformational,temporal,andspatialphenomena.
Dataascapta:frominformationvisualizationtographicalexpressions
ofinterpretation
IfIsetupabarchartorgraph,myfirstactistodrawasetofoneormoreaxesanddividethemintounits.The
conventionalformsofthegraphicaldisplayofinformation,"data",makeuseofaformal,unambiguoussystemof
standard metrics. Charts use simple (if often misleading) geometric forms that lend themselves to legible
comparisonofvalues,proportions,ortheexhibitionofstatechangesacrosstime.Lines,bars,columns,andpie
charts are the common and familiar forms. They render quantitative relations with a transparency that seems
natural, so that, for instance, if we look at the changes in population across a series of years for a particular
location,wecansimplyacceptthatfromoneyeartothenextrisesordropsoccurredinthenumbersofpersons
aliveinXcityinXcountryatXtime.Apiechartshowingpercentageofresourceallocationfromnationalbudgets
seemscompletelytransparent,selfevidenteven.Abarchartcouldcomparedaylighthoursatdifferentlongitudes,
or the average size of men and women in different countries, or the number of hospital beds in different
institutionsinasinglegeographicallocationandnotraiseaskepticaleyebrow,right?Yes,buttherendering of
statisticalinformationintographicalformgivesitasimplicityandlegibilitythathideseveryaspectoftheoriginal
interpretative framework on which the statistical data were constructed. The graphical force conceals what the
statisticianknows very well that no "data" preexist their parameterization. Data are capta, taken not given,
constructedasaninterpretationofthephenomenalworld,notinherentinit.
To expose the constructedness of data as capta a number of systematic changes have to be applied to the
creation of graphical displays. That is the foundation and purpose of a humanistic approach to the qualitative
display of graphical information. Read that last formulation carefully, humanistic approach means that the
premisesarerootedintherecognitionoftheinterpretativenatureofknowledge,thatthedisplayitselfisconceived
to embody qualitative expressions, and that the information is understood as graphically constituted. Each of
these factors contains an explicit critique of assumptions in the conventional "visual display of quantitative
information"thatisthecommoncurrency.
Let me work through a specific case to show how each of these principles humanistic approach, qualitative
display,andgraphicalinformationcanbedemonstrated.Asanexample,wecanusethatbarchartmentioned
above,onethatcomparesthepercentageofmenandwomeninvariousnationalpopulationsatthepresenttime.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
4/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
11
Figure1.
Figure 1 A basic bar chart compares the number of men (top bar) and the number of women
(bottombar)insevendifferentnations,AthroughF,atthepresenttime(2010).Theassumptionsare
that quantities (number), entities (nations), identities (gender) and temporality (now) are all self
evident.GraphiccreditXreneEskandar.
Certain issues immediately arise. A standard critique of data introduces reservations about the appearance of
certaintysuchachartpresents.Whatcountsasanation?Aretransientandimmigrantpopulationsdocumented?
Whatkindoftimespancountsas"atthepresenttime"withinwhichthesepopulationsarecounted?Ifthebasic
barchartwouldhavelookedlikeaseriesofbandsshowingdiscretecategoriesofinformationinfiniteandcertain
numbers(allduestatisticalcaveatsnoted),whataretheproblems?Genderdefinitionassumesasimplebinary
distinctionofmenandwomen,anassumptionmuchdebatedandhighlyproblematic(gendercanbeunderstood
as a factor of behavior, physiological changes, social expectations, dress, etc., and nation as a function of
permeability of borders, citizenship patterns, naturalization rules, immigration regulations, quotas and border
policies).Sothebarchartreifiesseveralcategories,naturalizingthemasdiscreteandfixed:nationalpopulations,
time span, and gender defined as a simple binary. The representation can only be modified by changing the
termsandpremisesonwhichitisconstructed.Whatwouldarepresentationofgenderbyslidingscalelooklike?
How would permeable boundaries to nations whose populations cross each others borders be shown? How
wouldtheydissolvethebarchartsbasicstructure?Howwouldnotionsofthepresentbedefined?
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
5/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
12
13
Figure2.
Figure 2 In this chart gendered identity is modified. In nation A, the top bar contains a
changing gradient, indicating that "man" is a continuum from male enfant to adult, or in
countries E and D, that gender ambiguity is a factor of genetic mutation or adaptation, thus
showingthatbasisonwhichgenderedindividualsareidentifiedandcountediscomplicatedby
manyfactors.IncountryFwomenonlyregisterasindividualsaftercomingofreproductiveage,
thus showing that quantity is a effect of cultural conditions, not a selfevident fact. The
movement of men back and forth across the border of nations B and C makes the "nations"
unstableentities.GraphiccreditXreneEskandar.
The point Im making is that the basic categories of supposedly quantitative information, the fundamental
parameters of chart production, are already interpreted expressions. But they do not present themselves as
categories of interpretation, riven with ambiguity and uncertainty, because of the representational force of the
visualizationasa"picture"of"data".Forinstance,theassumptionthatgenderisabinarycategory,stableacross
allculturalandnationalcommunities,isanassertion,anargument.Genderedidentitydefinedinbinarytermsis
notaselfevidentfact,nomatterhowoftenOlympiccommitteescomeupagainsttheneedforasinglerigidgenital
criterion on which to determine difference. By recognizing the always interpreted character of data we have
shifted from data to capta, acknowledging the constructedness of the categories according to the uses and
expectationsforwhichtheyareputinservice.Nations,genders,populations,andtimespansarenotselfevident,
stableentitiesthatexistapriori.Theyareeachsubjecttoqualificationsandreservationsthatbeardirectlyonand
arise from the reality of lived experience. The presentation of the comparison in the original formulation
grotesquely distorts the complexity but also, the basic ambiguity of the phenomenon under investigation
(gender, nations, populations). If the challenge we are facing were merely to accommodate higher levels of
complexityintoadatarepresentationmodel,thatwouldrequireonesetofconsiderationsandmodifications.But
the more profound challenge we face is to accept the ambiguity of knowledge, the fundamentally interpreted
conditiononwhichdataisconstructed,inotherwords,therealizationofmyrefrainthatalldataiscapta.
Thehumanisticaspectofthisapproachshouldbeobviousthatknowledgecreatedwiththeacknowledgement
ofthefundamentallyconstructednatureofitspremisesisnotcommensuratewithprinciplesofcertaintyguiding
empiricalorrealistmethods.Humanistic methods are counter to the idea of reliably repeatable experiments or
standard metrics that assume observer independent phenomena. By definition, a humanistic approach is
centered in the experiential, subjective conditions of interpretation. Phenomena and their observers are co
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
6/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
14
15
16
17
18
19
20
dependent,notnecessarilyinequalmeasure.Aviewergazingonasublimelandscapeorrecordingmigrationsat
alargescalemaybemoreaffectedbythephenomena than the phenomena is by the observation. Theoretical
physicistWernerHeisenbergneversuggestedthattherelationofinterveningobserverandeffectonphenomena
weresymmetrical,merelythattheywerecodependent,whenheintroducedtheconceptofuncertaintyintheearly
20thcentury.
Creatingbarchartswithambiguityanddegreesofuncertaintyorothervariablesinthemmightcausechampions
of legibility and transparency some unease, but the shift away from standard metrics to metrics that express
interpretation is an essential move for humanists and/or constructivists across disciplines. To emphasize the
expressivequalityofinterpretation,Imgoingtocharacterizeconstructedinformationassubjectiveexpressing
themarksofitsinflectioninsomeformalway.Theshifttoexpressivemetricsandgraphicsisessentialinchanging
from the expression of subjective information to the subjective expression of perceived phenomena, but
subjectivity and inflection are not the only features of interpretative approaches. Capta is not an expression of
idiosyncracy,emotion,orindividualquirks,butasystematicexpressionofinformationunderstoodasconstructed,
asphenomenaperceivedaccordingtoprinciplesofinterpretation.Todothis,weneedtoconceiveofeverymetric
"as a factor of X", where X is a point of view, agenda, assumption, presumption, or simply a convention. By
qualifyinganymetricasafactorofsomecondition,thecharacterofthe"information"shiftsfromselfevident"fact"
toconstructedinterpretationmotivatedbyahumanagenda.
[3]
The standard elements of graphic display for statistical information are simple and limited: scale divisions,
coordinatelines,scalefigures,circles,rectangles,curves,bars(orcolumnsorpercentagesofpiechartsorother
forms) and labels (numbers and terms), signs of movement, flow, or state change (arrows, vectors, paths). The
orderingand arrangement of elements within a chart create another level of information, relational information.
Relationalinformationisgraphicallyproducedtheorderingofelementsbysize,bycolor,byalphabeticalorder,
bytexture,shapeorotherfeaturehappensingraphicalspace.Theresultingarrangementhasasemanticvalue
producedbyfeaturesofproximity,grouping,orientation,apparentmovement,andothergraphicaleffects.
Nowtakethesebasicelementsofgraphicaldisplayandrethinkthemaccordingtohumanisticprinciples:
Inconventionalstatisticalgraphics,thescaledivisionsareequalunits.Inhumanistic,interpretative,graphics,they
arenot.
In statistical graphics the coordinate lines are always continuous and straight. In humanistic, interpretative,
graphics,theymighthavebreaks,repetitions,andcurvesordips.Interpretationisstochasticandprobabilistic,not
mechanistic,anditsuncertainties require the same mathematical and computational models as other complex
systems.
Thescalefiguresandlabelsinstatisticalgraphicsneedtobeclearandlegibleinallcases,andallthemoresoin
humanistic,interpretative,graphicssincetheywillneedtodoquiteabitofwork.
Perhaps the most striking feature distinguishing humanistic, interpretative, and constructivist graphical
expressionsfromrealiststatisticalgraphicsisthatthecurves,bars,columns,percentagevalueswouldnotalways
berepresentedasdiscreteboundedentities,butasconditionalexpressionsofinterpretativeparametersakindof
visualfuzzy logic or graphical complexity. Thus their edges might be permeable, lines dotted and broken, dots
and points vary in size and scale or degree of ambiguity of placement, and so on. These graphical strategies
express interpreted knowledge, situated and partial, rather than complete. They can be employed as
systematically as other charting elements, though part of my intention is to disturb the grounds of certainty on
whichconventionsofstatisticallegibilityarebased.Pointofviewsystemsintroducedintographsandchartswill
make evident a perspectival position with respect to their information, an inner standing point in the graphical
renderingofspace.Thisistrueofallcartographicprojections.Everymapcontainswithinitscoordinatesystemfor
graphical expression, a set of assumptions about the place from which the map is drawn. Information spaces
drawnfromapointofview,ratherthanasiftheywereobserverindependent,reinsertthesubjectivestandpointof
their creation into the graphical expression. Finally, any point or mark used as a specific node in a humanistic
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
7/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
21
22
graph is assumed to have many dimensions to it each of which complicates its identity by suggesting the
embeddedness of its existence in a system of codependent relations. Information entities, or units, are thus
understood as fictional abstractions serving a purpose. But their potential to be read again in relation to any
numberofotherequallysignificantrelationscanbemadeevident.Thisapproachdestroysthegroundonwhich
standardmetricsareusedtoabstractquantitativeinformationfromhumancircumstances.Humanisticpremises
replacenotionsofstatisticalconceptsofselfidentitywithentangledcodependenceandcontingencies.
Allofthismaysoundundulycomplicatedtosomeonemerelywantingtocountthenumberofpupilsenrolledina
group,calculatethenumberofpencilsneeded,ortoshowbudgetaryexpendituresonapercapitabasisinthe
classroom, for example. But this example an instance of administrative and bureaucratic management
showsthatsuchcrudelyconceivednumericstatisticsareusefulonlyinthemostreductivecircumstances. They
tellusnothingaboutwhetherthepencilscanbeused,whetherthepupilsarepreparedordisposedtotheirwork,
orwhetherthebudgetswillhaveanyeffectonlearningoutcomesoranyofthemanyotherfactorsthatcomeinto
playinassessmentsbasedonmetricsextractedfromlivedexperience.ButeachmetricnumberofXorYis
actuallyanumberasafactorofaparticularintellectualassumptionordecision:pupilsasafactorofseatsina
room, birthdates, population, illness, etc. pencils as a factor of resource allocation, and so on. All metrics are
metricsaboutsomethingforsomepurpose.
Anyhumanisticstudybasedonstatisticalmethods,eventhesimplesttechniquesofcounting,hastoaddressthe
assumption involved in the categories on which such techniques ("how many of X") are based. Take another
examplefromworkindataminingor"distantreading"asitisknowninthedigitalhumanities:countingthenumber
of novels published in a given year. This involves an enormous number of interpretative decisions each of
whichhasmoreintellectualdimensionsthananynumericassessmentcould.
Figure3.
Figure3Achartshowsthenumberofnewnovelsputintoprintbyasinglepublisher in the years
18551862.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
8/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
23
24
25
Figure4.
Figure 4 The "appearance" in 1855 of fourteen novels is shown in relation to the time of writing,
acquisition, editing, prepress work, and release thus showing publication date as a factor of many
other processes whose temporal range is very varied. The date of a work, in terms of its cultural
identityandrelevance,canbeconsideredinrelationtoanynumberofvariables,notjustthemoment
ofitspublication.GraphiccreditXreneEskandar.
For instance, what is a novel, what does "published" mean in this context (date of appearance, editing,
composition,acquisition,review,distribution),andhowwasthe"year"determined.Statisticalmethodscomeinto
playafter these decisions have been made, counting objects whose identity was established by interpretative
decisions. Many aspects of constructedness are in play. But the graphical presentation of supposedly self
evidentinformation(again,formulatedinthisexampleas"thenumberofnovelspublishedinayear")conceals
thesecomplexities,andtheinterpretative factors that bring the numerics into being, under a guise of graphical
legibility.Icannotoverstatetheperniciousnessofsuchtechniquesfortheeffectofpassingconstructionoffasreal,
andviolatingtheverypremisesofhumanisticinquiry.
The challenge is to design graphical expressions suited to the display of interpreted phenomena: information
aboutsubjectiveuserdependentmetrics,subjectivedisplaysofinformation,andsubjectivemethodsofgraphical
expression.Thetermsubjectiveisusedasshorthandforinterpretativeconstruction,fortheregistrationofpointof
view,position,theplacefromwhichandagendaaccordingtowhichparameterizationoccurs.Subjectivityisnot
thesameasindividualinflectionormereidiosyncracy,butismeanttoputcodependentrelationsofobserverand
phenomena(incontrasttopresumptionsofobjectivity,orobserverindependentphenomena).
Thedisplayofinformationaboutinflectionofaffectiveexperiencecaneasilyusestandardmetrics.Forexample,a
chart that shows mood changes or degrees of attraction or any other information related to subjectivity can be
createdwithstandardmetricsandvisualconventions.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
9/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
26
27
28
Figure5.
Figure5Achartofdataaboutaffecttherecordofpositiveandnegativefeelings in the course
of an afternoon. Standard metrics are used and a graphical display of the quantized experience
appears.GraphiccreditXreneEskandar.
The next task is more complicated. Subjective information, that is information whose constitution exhibits its
subjectivecharacter,deviatesfromthestandardnormsbyusinggraphicvariablessuchasintensityoftone,size,
color, or other feature to embody its qualities. Subjective information can use graphical means to show its
inflectedcharacter,demonstratingitsdeviationfromstandardnormsinthewaythedisplaylooks,or,indynamic
displays,thewayitacts.Onemightimagineskittishpointsonanunstablegridtodisplaythedegreesofanxiety
aroundaparticulareventortask,forinstance,orpointsthatglowhotorcolddependingontheotherelementsthat
approachthem.Thatwouldbeasubjectiveevenaffectivedisplayofinformation.
Creating a display whose structure arises from subjective methods of graphical expression extends this last
exampletothedesignofthebasicvisualstructure.
Figure6.
Figure 6 A chart in which the subjective information shapes the metric. The activities are given
tonal values, size, and weight in order to create a mass or volume that then determines the
dimensions of the "day" which they constitute. The box "day" does not have an a priori dimension
that is used to contain the elements, it is created as an effect of the elements. This is a distinctly
differentapproachtometrics.Thechartisgeneratedtoexpressthecodependentrelationofviewer
andexperienceratherthantodisplayuserexperienceasifitwereindependentofobservation.The
temporaldimensionalofeachdaydependsupontherelationsamongevents,moods,andactivities,
butnotpredictably.Theshapeofthedaysismadebythecreationofthelist.GraphiccreditXrene
Eskandar.
A subjective grid to show anxiety might have a widely varying set of spacings to show that the information on
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
10/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
29
30
31
32
33
34
35
displayisconstitutedasavariableofsomeotheraspectofexperience(numberoffamilymemberspresentatan
event,forinstance).Recognizingthatsuchsubjectivemethodsareanathematotheempiricallymindedmakesme
even more convinced that they are essential for the generation of graphical displays of interpretative and
interpretedinformation.
The basic principle underlying such graphical displays is that capta marks its interpreted status. Interpreted
knowledgeissituated,observercodependent,andpartial.Itsvariablesare,intheory,infinite,buttheyarealways
present in some degree or measure by virtue of the performative and participatory character of interpretative
information. Interpretation depends upon and is an expression of an individual reading in a particular set of
circumstancesandneverpresumestocompletenessorobserverindependence. The requirements for legibility
increasewiththeseunfamiliargraphics,andtheywillneedlabelingtomakeexplicitthejustificationsfortheirnon
normative seeming appearance. Im not advocating idiosyncracy, or intellectual solipsism, but a systematic
approachtographicsthatisappropriatetoitsprinciples.
Thesehumanisticprinciplescanbereadilyappliedtothegraphicaldisplayoftemporaland spatial information.
SoIwillturnmyattentioninthesenexttwosectionstosomeoftheprinciplesonwhichtemporalityandspatiality
canalsobegivengraphicalexpressionthroughhumanisticapproaches.
TimeasTemporality
Since antiquity, human conceptions of time have divided between those that consider time a given, an a priori
existingcontainerwithinwhicheventsoccur,andthosewhoconsidertimeaneffectofoccurrencesintemporal
relationtoeachother.Itakethelatterview.Therelationalstructureoftemporalityisalwaysconstitutedaccording
toinflectionsandvariables.Notalldaysareequal.Orallminutes.Orallhours.Timeunderstoodastemporality
can be succinctly stated as follows: Temporality = time as a factor of X where X is any variable (fear, speed,
anxiety,foreshadowing,regret,reconsideration,narration,etc.).
Humanists deal with the representation of temporality of documents (when they were created), in documents
(narrated,represented,depictedtemporality),theconstructionoftemporalityacrossdocuments(thetemporalityof
historicalevents),andalsotheshapeoftemporalitythatemergesfromdocumentaryevidence(theshapeofan
era,aseason,aperiodorepoch).Theyneedawaytographandcharttemporalityinanapproachthatsuitsthe
basicprinciplesofinterpretativeknowledge.
Conceptions of temporality in humanities documents do not conform to those used in the social and empirical
sciences.Inempiricalsciences,timeisunderstoodascontinuous,unidirectional,andhomogenous.Itsmetrics
arestandardized,itsdirectionisirreversible,andithasnobreaks,folds,holes,wrinkles,orreworkings.Butinthe
humanities time is frequently understood and represented as discontinuous, multidirectional, and variable.
Temporal dimensions of humanities artifacts are often expressed in relational terms before such and such
happened,orafterasignificantevent.Retrospectionandanticipationfactorheavilyinhumanisticworks,andthe
modelsoftemporalitythatarisefromhistoricalandliterarydocumentsincludemultipleviewpoints.
ThetemporalmodelingprojectBethanyNowviskieandIdesignedalmosttenyearsagomadeuseofthesebasic
insights in order to create a graphical application that was the working proof of a concept. We were intent on
demonstratingthatagraphicalmodelcouldbecreatedintuitivelyasaninterpretationandthenusedtogenerate
structureddataasaresult.Invertingthesequenceofintellectualeventswasaradicalmovefordigitalhumanities,
especially at the time, suggesting that graphical knowledge could be primary, leading an interpretation, rather
thanalwaysandonlyfunctioningtodisplaywhatwasalreadyknown(orassumedtobeknown).Wewantedto
demonstratethatvisualspacescouldbeaprimarysiteofintellectualwork.Ofcourse,thataddedyetanotherlevel
ofunfamiliaritytoouralreadycomplexprojectandmanyeveninourimmediatecommunitywereunsettledby
elasticorstretchytimelines,multiplepointsofviewfromwithinthesystem, or other novel seeming conventions
meanttoserveforinterpretationofliteraryandhistoricalartifacts.
Brieflysummarized,theoriginalTemporalModellingprojectaimedatcreatingasetofconceptual primitives for
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
11/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
36
the modeling of temporal relations. These included graphical expressions meant to meet the needs of multiple
pointsofview,reworkingeventsaccordingtoachangedpositionwithinatemporalsequence,andasetofwhat
wecalledinflections.Inflections,akindoflegendformarkingpoints,intervals,orevents(ourbasicunits)witha
quality or attribute, were divided into semantic and syntactic types. Semantic inflections were given their
characteristicsindependently,asentities,andthevocabularyofattributesincludeddegreesofintensityandother
qualities.Syntacticinflectionswerecharacterizedasrelational,markingtheeffectofoneevent,point,orintervalor
another.
Methodsforgraphingtheelasticor"rubbersheet"timelinesmeanttoshowthesubjectivevariationsintemporality
can be derived from catastrophe theory, chaos diagrams, and the visualizations of stochastic and complex
systems.
Figure7.
Figure 7 Models of events as temporal folds along a line of crisis. The first is a simple fold,
showing an event as a combination of stresses warping a plane. An upper branch of
consequences peels off towards an abrupt termination while the lower branch curve back to
allowaretrospectiveviewoftheeventsunfoldingbackonto an earlier moment. Graphic credit
XreneEskandar.
Thesevisualizationsexpressthetopologicalandsystemiccomplexitynecessarytomodelthenumberofvariables
(of coordinates, forces, and the changing relations of variables) present in the experience of events, and/or
analysisoftheirrepresentationinhumanisticdocuments(e.g.novels,films,letters,etc.).Someofthefeaturesof
ourearlier design, such as the dynamic behaviors of syntactic relations, could not be expressed in a standard
Cartesian coordinate system (such as the one on which XML output is generated), even though dynamic and
performativesyntacticrelationscanbemadeoperationalbyusingvectorsorforces.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
12/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
Figure8.
Figure8Twomodelsofaneventreachingacrisiswithstressfactorsshown as vectors. The
first shows the event as a fold, the second shows it as a vortex. Graphic credit Xrene
Eskandar.
But even standard coordinate systems, such as the conventions of perspectival drawing, allow for the
interpretativequalitytemporalexperiencetobeexpressedmorefullythanispossiblewithstandardtimelines.A
parallax view, in which prospective anticipation is gradually replaced with retrospective reassessment, can be
generatedwithasliderthatanimatesthedynamictransformationinthevalue,identity,andrelationoftemporal
events.Insuchaview,temporaleventsexpressedasasetofconditions,ratherthangivens.Thesliderindicatesa
pointofview,aperspectivefromwhichtheexperienceoftemporalityoriginatesinanindividual.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
13/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
37
Figure9.
Figure9Alinearmodelofparallaxshowinganticipationandretrospective assessment of an
event. The "event" is the combination of the moods of the "eye" individual, indicated by the
anticipatory arrow and then the retrospective view (lower arrow) across the bar and star that
mark a moment and a duration in the temporal span. The event is warped in the retrospective
view.Themetricmightbealtered as an effect, though it is not in this depiction. The "eye" is a
nowslider, as per the old temporal modeling design, and its position on the bottom line
indicates the position of the observer within the course of even. Graphic credit Xrene
Eskandar.
By breaking the relentlessly regular grid, the potential for graphing temporal modeling as a complex system of
eventsisgreatlyenhanced.Therelational,andcodependent quality of temporal events finds its expression in
thesemoresophisticatedmodels
Several fundamental principles can now guide these designs. These principles of noncontinuous, non
homogenous, and multidirectional temporality, as well as the point of view parallax, refine the reductive
crudenessofmodelslinkedtostandardapriorimetricsofunidirectional,continuous,homogenoustime. In this
refinementtemporalityisconceivedaccordingtothebasicformulationmentionedabove:timeasafunctionofx
(temporality=time
(x)
).Intheseformulations,xisanyofthe(theoreticallyinfinite)variablesthatinflectthemodel
(mood,events,influences,events,constraints,etc.).Becausetemporalityisanactofformmaking(constructivist),
notanactofexpressingpreexistingoraprioriphenomena(realism),thesequenceofintellectualeventsinthis
formulationinsistsontemporality(and,likewise,spatialityastheresultofconstitutiverelationsamongtemporal
and spatial phenomena. The full realization of this approach requires a multidimensional, complex, model of
spaceandtimeandimaginativerealizationsasgraphicalexpression.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
14/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
Figure10.
Figure 10 In the first image, anxiety (measured subjectively but charted on a standard metric) is
charted against time, also depicted with standard intervals. The change from one state to another
(changesindegreesofanxiety)isshowninacontinuousline.GraphiccreditXreneEskandar.
Figure11.
Figure 11 The difference between one state and the next is used to generate a graphical form
thatisexpressesthechangesfromonemomenttoanother.GraphiccreditXreneEskandar.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
15/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
38
Figure12.
Figure13.
Figures 12 and 13 The differences between states are projected onto the anxiety and time axes
tocreateametricthatistheeffectofperception,ratherthananapririgiven.Byrotatingtheangles
that marked changes of levels of anxiety into a position parallel to the time line, the metrics can be
changedasaprojectionoftheselines(whoselengthsweregeneratedbyacombinationofduration
andchangeofintensityofanxiety)ontothetemporalaxis,thusmovingfroma"perceived"timetoa
"projected" time. The result is a set of transformations from an uninflected, supposedly observer
independent "time" and "anxiety" to one created as an effect of the experience of time on its
expression.GraphiccreditXreneEskandar.
SpaceasSpatiality
Thediscussionofspacecorrespondsexactlytothatoftime,andthedistinctionsbetweentheconceptionofspace
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
16/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
39
40
41
42
43
as an a priori given and that of space as relationally constituted marks the same philosophical division of
approachesasthosethatareusedinchartingorunderstandingtimeandtemporality.Likewise,spatialityistobe
understoodasspaceasafunctionofx(spatiality=space
(x)
).
[4]
To give graphical expression to these ideas requires using nonstandard metrics, intuitive and subjective
principlesofdesign.Theyaremeantasprovocationstothelargerprojectofcreatingmoresystematicrenderings
ofhumanisticphenomena,introducingbasictransformationsofthegraphicalfieldswecreatedfortimelinesinto
mappingandGISapplications.Precedentsforsuchrenderingscanbefounde.g.FrancisGaltonsrenderingof
spaceasafunctionoftraveltime.Galtonsproblem,formulatedinthemid19th century, takes into account that
most statistical phenomena are observerdependent and situated, and cant be separated from the various
dependenciesthatbearuponthecreationofdata.Galton,inotherwords,recognizedthatinmanycircumstances,
data were capta. The statistical description of phenomena depend upon the observers circumstances. A more
recentdemonstrationoftheseprinciplesisamapdesignedbyTomCarden.Hisdynamicinterfaceredrawsthe
LondonUndergroundmapasfunctionoftimeoftravelfromanyselectedstationtoanyotherstation.
[5]
Subjective parameters are even more difficult to inscribe, since they cannot, by definition, be based on simple
consensual standards. We can easily understand these distortions space as a result of travel time. But how
could we visualize the spatial distortions introduced by variables such as fearfulness, anxiety, anticipation,
distraction,ordallianceandthusrenderspaceasspatiality,spaceasafactorofx?Somevariableisalwaysin
playintheexperienceofspaceaswellasitsrepresentation,sospaceisalsoalwaysconstructedaccordingtoa
specificagendaandasituatedexperienceetc.Whilethisisthecommonexperienceofthephenomenalworld,
representations of spatiality have lagged behind, dominated by the navigational or descriptive systems of
standard mapping whose conventions are well known and recognized, and which partake of and impose the
dominantrealistmodel.
In proposing a new model for humanities work, I am suggesting that the subjective display of humanistic
phenomena can be applied across the domains with which we are concerned at at least four basic levels of
interpretationorknowledgeproduction.
Letmedescribeaconcreteexampleandseehowitcanbeunderstoodacrossthesefourdifferentmodels.Take
thefirstinstance,themodelingofaphenomenon.Threepeoplearewaitingforabus,howlongdoesittake?One
islateforworkandanxious,oneisindesperateneedofabathroom,andtheotherdoesnotwanttogotothe
afterschoolprogram.Howcanthevariationsinperceptionbeexpressed?Recentexperimentsonthewaytimeis
understood in relation to different circumstances and tasks have made this experiential variable apparent to
psychologists.So,theinitialgraphicalexpressionofthehumanisticphenomenonrequiresavariablemetric,an
elastictimeline,evenafieldthatmightfoldorbreakunderextremecircumstances.
Whenweshiftfrommodelingexperiencetofindgraphicalexpressionsfortherepresentationofexperience,the
complexity of the problem increases. The modeling of time in documents, in relation to the duration of the
documents(timeoftelling)andtheexperiencestheyrecount(thetimeofthetold)aswellastherelationsamong
these and possible external temporal references, forms a subset of linguistic and narrative analyses. The
graphicalformstorepresentthesearegenerallyinadequatetothecomplexityofthetextualorvisual(and/orfilmic
1. Modelling phenomenological experience in the making of humanities (data as capta, primary
modeling,therepresentationoftemporalandspatialexperience)
2. Modelingrelationsamonghumanitiesdocumentsi.e.discoursefields(adifferentmetricisneededto
understand dates on diplomatic documents in the spring of 1944 than one needed to constitute
understandingofthosedatedtothesameperiodofthespringof1950etc.)
3. Modelingtherepresentationsoftemporalityandspatialitythatareinhumanitiesdocuments(narrative
isthemostobvious)
4. Modeling the interpretation of any of the above (depicting or graphing the performative quality of
interpretation).
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
17/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
44
45
46
andaudio)documents.
Modelling the temporal relations among documents about temporal experience (imagine letters, emails, text
messages,ordiaryentriesfromthesevariousbusriders,onlysomeofwhichisdatestamped),givesrisetoyet
furtherambiguitiesandcomplexities.Alettersentthatwasdelayed,emailrerouted,messagesheldinsuspense
onaserverwillchangethetemporaleffect.Forinstance,lettersoremailsarrangingfamilyeventsandtravelsover
theholidayscontainmanytemporalvaluesthatarecontingentoneachotherandofteninconstantfluxasplans
arebeingmade.Thetemporalsequenceandthedatestampsarenotoneandthesame,atemporalrelationof
theexchangesmightincludemessagesthatcrossinmidstream,andwhosetemporalsequencedoesnotmatch
thesimplealignmentwithdatesonaline.
Plans change, travel times are altered, arrivals and departures rearranged, moods shift, frustrations intensify,
disappointments or unexpected surprises arise in relation to the sequence of events. An email recounting
something that occurred "yesterday" in relation to a date stamp might also contain more vaguely identified
"earlier"and"before"statementsthatputeventsintoarelativesequencewithoutexplicitlyidentifyingwhenthese
occurred.Asthetellingunfolds,theserelationsmaychangeinthewritersexpressionandperception,sothatthe
textualdescriptionofarecollectedeventcontinuestoshiftitsplaceinthetemporalorder.Whowassupposedto
dowhatwhenandwhowasdependingonwhichorderofevents?Bythetimeholidaytravelsandexpectations
aresortedout,eachfamilymemberhasaverydistinctviewofwhathappenedwhenandhowthesequenceof
livedeventsoccurredandwhere.Wasthebusstationlargeorsmall,farorneartoanyotherspotintheitinerary,
or located in a familiar landscape. How was the space experienced as a function of time spent in it? These
constructionsoftemporalityandspatialityfromwithindocuments,acrossdocumentsoradiscoursefield,andof
phenomena are all created with time/space as functions of interpretation. The act of interpreting a series of
documentscreatesitsowntemporality,thatoftheproductionofareading,thatisnotthesameasthetellingorthe
told within the documents, but an independent phenomenon. An interpretation has its own temporality ad
spatiality.
Wecanconstructaconcreteexampleofspatialitythatparallelsthisexampleoftemporality,andalsodependson
temporal models. For instance, imagine an open stretch of beach, relatively unconstrained and unconstructed.
Whenasailingshipiswashedupatacertain point on the beach, not only that point, but the space around it,
becomestransformed.Thepresenceofthewreckcreatesahugeimpact,andthespacealmostpalpablybends,
compresses,expands,andwarpsaroundit,withwavesofresonanceripplingoutwardfromthatpoint.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
18/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
47
Figure14.
Figure14Inthisexample,ageographicalspace(astretchofbeach)isaffectedbyachange
in the state or circumstance. First we see the space mapped according to a regular Cartesian
coordinatesystem.Thenthegridisinflectedbythearrivalofabeachedship,aroundwhichthe
beachbendsbecausethesenseofeachspotasrelativeequalisdistortedbytheattentionthat
theshipcommands.Thespaceacquiresoneinflectionafteranotherasgraffitimarkstheship,a
chainlinkfencegoesupwithapolicenotice,footprintscreateapatterninthesand,pathways
for observation reroute pedestrian traffic etc. The "space" of the beach is transformed
physicallyandintermsofattentiongettingandeffectsothatitisnolongerasetof equal and
neutral elements of a rational spatial system, but one that must be expressed with graphical
distortionsthatshowtheseinflections.GraphiccreditXreneEskandar.
Policebarriersaresetupandsuddenlymakethatbitofbeachintoahighlychargedsite.Additionalfencescreate
zonesofpotentialtransgressionandprohibition,linesintheliteralsandthatwhencrossedbygraffitiartistsand
taggers, vandals and looters, introduce a whole set of spatial relations governed by different rules and
expectations.Thespaceofandaroundtheshipwreckbecomesahotpoint,azone,anarenaofcomplexspatial
negotiationsandmarkedcoordinates,eachdifferentlychargeddependingontheplayersandcircumstances(law
enforcement, owners, passersby, taggers at night, in early morning, broad daylight etc.). Even more than the
open,indeterminatespaceofthebeach,thisspotbecomesanareaofshiftingvaluesandinterpretation.Space,
alwaysmarked,hasbecomeexplicitlyso,andthespatialrelationsdemarcateregionsofauthorityandbehavior
whosedimensionsarenotinstrictcorrespondencetophysicalspace.Thesameamountofphysicalspacehalfa
miledownthebeachhasnone(orfew)ofthesedimensions.CanwestilllocatethewreckonaCartesiangrid
availablethroughanyGPSsystem?Ofcourse,thetwoapproaches,constructivistandrealist,dontcanceleach
other out. But they are not equivalent. The GPS standards locate the spot within those coordinates, but say
nothing about the constituted space as a phenomenon created by these many variables. We have many
adequatemodelsforthefirstmodeofvisualization,butveryfewfortheconstructivistapproachgroundedinan
interpretativemodeofexperience.
Take another example, a map tracing a journey between London and Prague in the 1810s.
[6]
How does the
spacechangedimensionstoreflecthazard,delays,dalliances,terrainchanges,interruptionsofwarandpolitical
strife,danger,weather,or illness? A legend or set of labels or markings could indicate these inflections of the
spacesimplybyputtingsymbolsonamap.Thatwouldbetheregistrationofsubjectivedataonaconventional
map.Butmappingconventionsdontmorphthelandscapetoaccommodatetheeffectsoffear,anger,orviolence.
Now change the map, distort its proportions so that it becomes a terrain shaped by fear, by obstacles, by
disruptionsandconfusions.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
19/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
48
Figure15.
Figure 15 Geographical terrain warped by the experience of travel so that the standard
distances are distorted by the effects of difficulty, fear, delays, and other factors. The map
shows the landscape as an effect of experience rather than a standard ground on which to
depict experience. In the second instance, the points on the metric grid are warped by the
impact of an event, or events, that have simply reordered the standard grid. Graphic credit
XreneEskandar.
Thatisasubjectiveexpression.Thetwoapproachesareradicallydifferent.Inthesecondinstance,spaceisan
effectofspatialrelations,spatialityisexpressedasafactorofdisturbance,anditmightbeexpressedasafactorof
manyvariablesoccurringacrossatemporalextension(fear,anxiety,confusion,anger,disorientation).
The challenge of representing large corpora of texts and immense archives also requires attention, in part
becausetheconventionsofwayfindingandnavigationthatarepartofprintmediaanditsinstitutionalstructures
arenotyetreworkedinadigitalenvironmentmeanttoaddresstheshiftsinscaleandexperiencebroughtonby
newmedia.Ontopofthechallengeofrepresentingrepositoriesandtheiruse,wecanpointtoanotherchallenge
thatofgivinggraphicalexpressiontointerpretationsbuiltonandoutofdocuments,orcollectionsofdocuments.
These present different challenges than the humanistic interpretation of temporal, spatial, and informational
phenomena, but depend upon the basic recognition that subjective and codependent principles must govern
theirdesign.Theconventionalgraphicalfeaturesoftextsthatinscribeinterpretationincludeallofthefeaturesof
layout and format, typography, and design that organize and structure its presentation on the page, screen, or
othersurfaceormedium.Thefeaturesthatinscribeinterpretationinarchivesarethosethatembodyorexpress
theimprintofthepointofviewaccordingtowhichthearchivetakesshape.Theseincludeclassificationsystems,
nomenclature, hierarchies and categories of organization and ordering, systems of search and access,
informationarchitecture,theformatofstorageanddisplay,andanyotherfeatureofthearchivethatisintrinsicto
theformsofitsexpression.Whilealloftheseareexpressionsofarguments,andthusinterpretations,theydonot
showormodelinterpretationontheflyasaconstitutiveactofreading,relating,connecting,andsensemaking.In
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
20/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
49
50
51
52
sum,theseactsofinterpretationmakeuseoftheformatfeaturesofgraphicalpresentationaswellasresponding
to and thus producing the "content" of these artifacts. Some combination of usercentered but codependent
systems analysis and critical reading practices as performative acts would have to underpin such graphical
visualizations.Butthatisalsoworkforanothertime.
Conclusion
Myargumentisapolemicalcalltohumaniststothinkdifferentlyaboutthegraphicalexpressionsinuseindigital
environments. A fundamental prejudice, I suggest, is introduced by conceiving of data within any humanistic
interpretativeframeonaconventional,uncritical,statisticalbasis.Fewsocialscientistswouldproceedthisway,
andtheabandonmentofinterpretationinfavorofanaveapproachtostatisticalcertainlyskewsthegamefromthe
outset in favor of a belief that data is intrinsically quantitative selfevident, value neutral, and observer
independent.Thisbeliefexcludesthepossibilitiesofconceivingdataasqualitative,codependentlyconstituted
inotherwords,ofrecognizingthatalldataiscapta.
Again,toreiterate,Iamnotsuggestingthatwesimplyintroduceaquantitativeanalysisofqualitativeexperience
into our data sets. I am suggesting that we rethink the foundation of the way data are conceived as capta by
shiftingitstermsfromcertaintytoambiguityandfindgraphicalmeansofexpressinginterpretativecomplexity.In
some circumstances (the example of the bar chart given earlier that was displaying information about gender,
nations, and populations) ambiguity merely requires a higher order level of complexity in the model, so that
apparent"certainties"arequalifiedbyvariablesandnuancesthatcanbespecifiedinmathematicalterms.Butthe
idea of capta as fundamentally codependent, constituted relationally, between observer and observed
phenomena,isfundamentallydifferentfromtheconceptofdatacreatedasanobserverindependentphenomena.
That realization has to be at the heart of humanistic approaches to the graphical display of interpretative
phenomena, of interpreted artifacts and the acts of interpretation themselves. Because interpretation is
performative,bringingobjectsintoviewthroughareadingorotheractofintervention,itforeclosesthepossibility
that autonomous objects or phenomena exist within the horizon of human experience. Phenomena of human
experienceareconstitutedasinterpretativeacts.
The natural world and its cultural corollary exist, but the humanistic concept of knowledge depends upon the
interplay between a situated and circumstantial viewer and the objects or experiences under examination and
interpretation.Thatisthebasicdefinitionofhumanisticknowledge,anditsgraphicaldisplaymustbespecificto
this definition in its very foundational principles. The challenge is enormous, but essential, if the humanistic
worldview, grounded in the recognition of the interpretative nature of knowledge, is to be part of the graphical
expressions that come into play in the digital environment. If we dont engage with this challenge, we give the
gameawayinadvance,cedingtheterritoryofinterpretationtotherulingauthorityofcertaintyestablishedonthe
falseclaimsofobserverindependentobjectivityinthe"visualdisplayofquantitativeinformation."
[7]
Ill finish with one more concrete example of the shift from observerindependent realism to codependent
constructivism.Snowsjustlyfamouschartofdeathsfromcholeraallowedcityofficialstotrackthesourceofthe
epidemictoasinglewaterpump.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
21/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
Figure16.
Figure 16 Dr. John Snows famous chart tracing the source of an epidemic using graphical
methodsthatplottedfrequencyofoutbreaksandgeographicallocation.Buteachoutbreakwas
an individual, and their degrees of vulnerability, impact of their illness, effect on the family and
lovedones,wasspecificandparticularinwaysthatasingledotcannotexpress.Seenfromthe
point of view of an individual participant in these tragic events, some of these individuals loom
much larger than others when depicted from within the gaze of someone actually seeing them
occur.GraphiccreditXreneEskandar.
Thedistributionofdotsonthestreetmapmakesevidenttheroleofthepumpbythe way they cluster. A useful
map,crucialtoanalysis,itsclarityandsuccinctnessservedanimportantpurpose.Itwassufficienttothatpurpose,
adequate,butwecouldrevisitthatmapanduseittoexpressotherfactors.Whoarethosedots?Eachindividual
hadaprofile,age,size,health,economicpotential,familyandsocialroles.Inshort,eachdotrepresentsalife,and
noneoftheseareidentical.Manydemographicfeaturescouldbelayeredintothismaptocreateamorecomplex
statisticalviewoftheepidemic.Thatisneithersubjectivedatanorasubjectivedisplay.Butwhatifwetaketherate
ofdeaths,theirfrequency,andchartthatonatemporalaxisinflectedbyincreasingpanic.Thengiveagraphical
expressiontotheshapeoftheterrain,thaturbanstreetscape,asitisredrawntoexpresstheemotionallandscape.
Thenimaginedrawingthissamestreetscape from the point of view of a mother of six young children, a recent
widow,asmallchild,oranelderlymanwhosesonhasjustdied.
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
22/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
Figure17.
Figure17Snow'schartaltered.GraphiccreditXreneEskandar.
Theselatterareallinstancesofthegraphicalexpressionofhumanisticinterpretation.Theyareasdifferentfrom
thevisual display of quantitative information as a close reading of a poem is from the chart of an eye tracker
following movements across a printed page. They are fundamentally different in character and in their basic
assumptionsabouttheroleofgraphicalexpressionasanaspectofknowledgeproduction.Wehaveaverylong
way to go in creating graphical expressions that serve humanistic interpretation, but I hope I have suggested
someofthepremisesonwhichthisworkmightbegin.
Notes
[1]
[KnorrCetina&Amann1990][Lynch&Woolgan1988][Anderson2007][Anderson2008][Porter1995][Lochlann
2010][Latour1986].
[2]
[Griethe&Schumann2006][Joneset.al2008][MacEachrenet.al2005][Shneiderman&Pang2005][Skeelset.
al2008][Wells2008]
[3]
Subjectivecarriesstructuralistconnotations,aspositionandenunciatedidentityandismeanttosuggestboth
thecodependentconditionsofconstructionandexpressivegraphicalfeatures.
[4]
IfirstunderstoodthisnotionwhenIreadSigfriedGiedion'sSpaceTimeandArchitectureaboutthirtyyearsago.
Hisargumentsaboutstructureasspacemakingleftadeepimpression.
[5]
See:[Carden]
[6]
AlternateexamplewasastreetmapofTehranonwhichpointsareplottedtoshowadisturbanceinprogress.
Letthepointsexpandtoregisterdegreesofintensityofimpact,disruptionofnormalfunction,fearandanger,
violence.Thatwouldbetheregistrationofsubjectivedataonaconventionalmap.Nowchangethemap,distortits
proportionssothatitbecomesaterrainshapedbyfear,byobstacles,bydisruptionsandconfusions.
[7]
ThiscitationisofcoursethetitleofEdwardTufte'sfirstvolume.
WorksCited
Anderson2007Anderson,Margo."QuantitativeHistory",TheSageHandbookofSocialScienceMethodology,ed.
WilliamOutwaiteandStephenTurner,London:SagePublications,2007,24663.
Anderson2008Anderson,Margo."TheCensus,Audiences,andPublics",PresidentialAddress,SocialScience
History32:1,Spring2008,118.
CardenCarden,Tom.TravelTimeTubeMaphttp://www.tomcarden.co.uk/p5/tube_map_travel_times/applet/
7/31/12 DHQ: Digital Humanities Quarterly: Humanities Approaches to Graphical Display
23/23 www.digitalhumanities.org/dhq/vol/5/1/000091/000091.html#
Griethe&Schumann2006Griethe,HenningandHeidrunSchumann."VisualizingUncertaintyforImproved
DecisionMaking",SimVis2006,143156.
Joneset.al2008Jones,Josh,RemcoChang,ThomasButkiewiczandWilliamRibarsky."Visualizinguncertaintyfor
geographicalinformationintheterrorismdatabase",SPIESecuritySymposium,2008.
dvg.uncc.edu/publications/index.html
KnorrCetina&Amann1990KnorrCentina,KarinandKlausAmann."ImageDissectioninNaturalScientific
Inquiry",Science,Technology,andHumanValues,15:1990,259.
Latour1986Latour,Bruno."VisualizationandCognition:DrawingThingsTogether",KnowledgeandSociety,6:
1986,140.
Lochlann2010Lochlann,Jain."MoralityEffect:CountingtheDeadintheCancerTrail",PublicCulture.2010,89117.
Lynch&Woolgan1988Lynch,Michael,andSteveWoolgar."Introduction:SociologicalOrientationsto
RepresentationalPracticeinScience",HumanStudies,11:1988,99116.
MacEachrenet.al2005MacEachren,AlanM.,AnthonyRobinson,SusanHopper,StevenGardner,RobertMurray,
MarkGaheganandElisabethHetzler."VisualizingGeospatialInformationUncertainty:WhatWeKnowandWhat
WeNeedtoKnow",CartographyandGeographicInformationScience,Vol.32,No.3:2005,139160.
Porter1995Porter,Ted.TrustinNumbers:ThePursuitofObjectivity,Princeton,NJ:PrincetonUniversityPress,
1995.
Shneiderman&Pang2005Shneiderman,BenandAlexPang.VisualizingUncertainty:ComputerScience
Perspective,NationalAcademyofSciencesWorkshop,Washington,DC:March34,2005.
http://www.cs.umd.edu/hcil/pubs/presenations/NASVisUncertainty6_files/frame.htm
Skeelset.al2008Skeels,Meredith,BongshinLee,GregSmithandGeorgeRobertson."RevealingUncertaintyfor
InformationVisualization",MicrosoftPublications,2008.http://research.microsoft.com/pubs/64267/avi2008
uncertainty.pdf
Wells2008Wells,AudreyElizabeth.VirtualReconstructionofaSeventeenthCenturyPortugueseNau,Master's
Thesis,Texas&MUniversity,VisualizationSciences,2008.
repository.tamu.edu/bitstream/handle/1969.1/86071/Wells.pdf