Digital visualization as a scholarly activity ...... Martyn Jessop Centre for Computing in the Humanities, King’s College London, UK ...... Abstract Thought processes are enhanced when ways are found to link external perception with internal mental processes by the use of graphic aids. Such aids range from scribbled diagrams to sophisticated linkages between thought, images, and text such as those employed by Leonardo da Vinci. These tools allow visual percep- tion to be harnessed in the dynamic processes associated with the creation or discovery of new . Digital humanists are applying digital versions of these age-old tools in many areas of , from the graphs generated by text analysis applications to virtual reality models of ancient buildings, methods known collectively as ‘digital visualization’. This article begins with a brief review of the current application of visualization in the digital humanities before moving on to establish a context for digital visualization within ‘traditional’ humanities Correspondence: scholarship. This provides a context for an examination of what is required in Martyn Jessop, Centre for order to ensure that digital visualization work is performed with identifiable Computing in the intellectual . The London Charter is used as a case study for a possible Humanities, King’s College framework for the development of appropriate methods and standards. Digital London Strand, London WC2R 2LS, UK. visualization as a scholarly methodology is discussed and demonstrated as being E-mail: part of a continuum of established academic practice rather than something that [email protected] is in some way new, ‘revolutionary’, or lacking in rigorous scholarly value......

1 Introduction Graphic aids to thinking are not new, but the emer- gence of digital technology has created a new The numerous visual metaphors that are used to medium for these tools that provides extended func- describe our cognitive processes hint at the nexus tionality and many new opportunities for develop- of relationships between what we see and what we ment. This in turn offers the potential for new think. We say we ‘see’ when we mean that we ‘under- research methodologies that amplify cognition. stand’; we try to organize and make our ideas These serve two distinct purposes, one of which is ‘clear’ by bringing them into ‘focus’, and so on. sometimes described by the use of the hackneyed When faced with tasks that require substantial phrase ‘A picture is worth ten thousand words’.1 thought or organization of ideas we will often However, this misses the true purpose of visualiza- reach for a pen and paper to ‘sketch out’ (another tion as what is being described by that phrase is visual metaphor) our thoughts. There exists an simply a matter of transmission, i.e. of having high instinctive deep understanding that one can enhance bandwidth to transmit large volumes of information. one’s thought processes by finding ways of linking Arguably such transmission is not even ‘communi- external perception with internal mental processes. cation’ as it is unidirectional, a monologue rather

Literary and Linguistic Computing, Vol. 23, No. 3, 2008. ß The Author 2008. Published by Oxford University Press on 281 behalf of ALLC and ACH. All rights reserved. For Permissions, please email: [email protected] doi:10.1093/llc/fqn016 Advance Access Published on 5 September 2008

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 M. Jessop

than a conversation. Of far greater importance is the a further question will inevitably be raised by ability of these tools to allow visual perception to be our peers, of whether what we are doing is simply used in the creation or discovery of new knowledge. developing a ‘method’5 or contributing to a ‘meth- However, knowledge is not transferred, revealed, odology’?6 The difference is perhaps of being or perceived, but is created through a dynamic pro- regarded either as ‘technicians’ or ‘’. cess. This raises epistemological issues concerning What is ‘visualization’? Some would define it as a visualization and points the way to an intellectual group of techniques (methods) for creating images, approach to the subject. diagrams, or animations to communicate a message. It is necessary at the outset to define what it is This form of visualization has utilized visual imag- that is meant by ‘scholarly activity’. Many of the ery as an effective way to communicate both definitions are surprisingly circular; scholarly activ- abstract and concrete ideas throughout human his- ity being something which is done by scholars who tory. Examples include cave paintings, Egyptian are in turn defined by being those who conduct hieroglyphs, ancient Greek geometry, and the tech- scholarly activity. It is known instinctively that nical drawings of Leonardo da Vinci. The use of scholarly activity is about the discovery, exchange, images for communication during the larger part interpretation, and presentation of knowledge but of our history is reflected in images produced by what is it precisely? While one may not always com- hand or those reproduced in print media. This his- mend Wikipedia to students it does, in this case, torical perspective raises three important questions, provide a useful set of simple statements describing each of which has a bearing upon an examination scholarly method: ‘the body of principles and prac- of digital visualization as a scholarly activity: tices used by scholars to make their claims about the  What role have visualizations played in human- world as valid and trustworthy as possible, and to ities scholarship in the past? make them known to the scholarly public’.2 There  If the majority of images in print are to be are two aspects to this: one governs the conduct of regarded as ‘illustrations’ what is the distinction research, the other the dissemination of the details between ‘visualization’ and ‘illustration’? and outcomes of that research to other scholars, and  How has the emergence of digital media affected maybe the public at large. The Wikipedia entry con- the development of visualization? tinues, ‘This article focuses on scholarship in the narrower sense, covering rational in areas First, the definition of ‘visualization’ can be that are mostly too complex to yet be treated by completed by examining the second question; . These include history as well as the crea- ‘how does visualization differ from the forms of tions of the human mind in the form of art, graphical representation that have been used to pro- music, literature, religion, , and cultural duce illustrations throughout history?’ For example, beliefs.’3 This refers to disciplines that are at the few would regard the devotional art of the late heart of humanities research and helps us draw Middle Ages as mere illustration. St Augustine and a distinction between ‘scientific visualization’ and St Thomas Aquinas argued that physical images ‘humanistic visualization’. The digital humanities could be used by worshippers to envision the are taking what is perceived by the public to be divine through prayer. These were not just illustra- the tool of the scientist and using it to conduct tions of religious figures and events, they were ‘rational enquiry in areas that are mostly too com- images that were intended to assist thought and, plex to yet be treated by science’.4 Ultimately the depending on your personal beliefs, higher mental approaches used in humanities computing will and metaphysical processes. Jean Gerson, a theolo- therefore, in the main, be substantially different gian and Chancellor of the University of Paris in from those used in the and one could jus- the early fifteenth century wrote about learning to tifiably describe them as belonging to a field of transcend with our minds from visible things to the ‘Humanistic Visualization’. While this prospect is invisible, from the corporeal to the spiritual. He saw very gratifying for those of us working in the area this as the purpose of the image (Donovan, 2007).

282 Literary and Linguistic Computing, Vol. 23, No. 3, 2008

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 Digital visualization as a scholarly activity

Although the distinction between illustration and This arguably has a justifiable basis in historical visualization is a topic rich with possibilities fact; William Ivins has written about the limitations for discussion, I must skirt around this question of images in scholarship from ancient times to the as this article concerns digital visualization and relatively recent past of our print culture that were I believe it is in the ‘digital’ that digital visualiza- due to the difficulties of producing what he calls tion can go beyond mere method to become ‘exactly repeatable images’ (Ivins, 1969). He argues methodology. that it has been impossible to use images to dissem- A digital visualization differs from a printed illus- inate ideas because of the problems of reproducing tration in two related characteristics: them accurately. This mistrust also has a philoso- phical underpinning by philosophers from Plato to  It is interactive. Descartes (and beyond). Space does not permit this  It allows manipulation of both the graphical past history to be discussed here, but it is inevitably representation and the data it is derived from. reflected in attitudes towards visualization in the However, these criteria apply to virtually all present day. I suspect that part of the problem graphics produced by computers; a scatter graph also lies in the apparent ‘newness’ of digital visual- plotted using a spreadsheet can be part of an inter- ization. To counter this I will seek to establish a active and iterative process of exploration as the context for digital visualization in established schol- user refines the display, selecting different subsets arly method. The discussion will then proceed to the of the data to display, changing scales or line fitting problems that that ‘newness’ presents and, finally, algorithms, or performing statistical operations some guidelines to ensure the intellectual integrity upon the data. In many respects, it is therefore of digital visualization work. correct to describe a spreadsheet as a visualization application, at least in the style of its usage, as it is 2 effectively applying a visualization interface to a Visualization in the method of exploratory data analysis. Clearly this is Digital Humanities not an adequate description of digital visualization. A visualization departs from being an illustration There are many ways of structuring an examination when it becomes the principal medium of commu- of the use of visualization in the humanities; by nication. The term ‘Illustration’ implies an image discipline, by type of information structure, and so which serves only to support written language; on. To set the context for this article, I have chosen thus the main carrier of information is the asso- to look at the type of data that is being visualized. ciated text not the image. Many consider that an I will not discuss them in depth here as the purpose image can only be a true visualization when it is is to indicate the scope of the visualization methods the primary carrier of information, not a supple- that I am referring to. In practice, the boundaries of ment to a piece of text. I shall settle on a distinction each data type are blurred and each has an impor- in which an illustration is intended merely to tance beyond its immediate scope, but a starting support a rhetorical device (usually textual) whereas point could be as follows: a visualization is intended either to be the primary  Space—the study of spatial relationships and a rhetorical device or serve as an alternative but par- sense of place occur in many humanities disci- allel (rather than subordinate) rhetorical device. plines. This area is dominated by Geographical This indicates where attention needs to be focused Information System (GIS) software which was in order to determine what is required to ensure developed for scientific data that exhibits a that digital visualization is seen to have the intellec- high density of precise information, but is not tual rigour expected of a scholarly activity. ideally suited to the more sparse ‘fuzzy’ quanti- There is a broader issue which concerns the tative or qualitative data used in the humanities. emphasis that humanities scholars place on written From a humanist’s viewpoint, it is interesting language and their apparent mistrust of images. that Denis Wood argues that the impulse to

Literary and Linguistic Computing, Vol. 23, No. 3, 2008 283

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 M. Jessop

write history and to draw maps (geography) such as those used in text analysis. It is gradually sprang from the same source; the need for early permeating into new areas of the humanities civilizations to control their societies (Wood, through work such as that of Franco Moretti 1992). Maps are thought of as spatial devices who has argued for a wider application of quan- but there is also a temporal as well as spatial titative methods in disciplines such as literary dimension to the need for such civilizations to history (Moretti, 2005). record qualitative and quantitative information.  Text—a great many of the visualization methods Digital dynamic maps are one of many alterna- applied to text are derived from analytical quan- tives offering media that are better suited to titative methods that were originally borrowed humanists (Jessop, 2006). In digital visualization from the sciences. This is an interesting area of tools such as GIS, or projects such as the Valley of application because there are also other more the Shadow7 or Salem Witch Trials,8 there exists a imaginative visualization tools that owe more form of visualization that combines time, space, to the arts than the sciences. Projects such as quantitative data, and qualitative data. The TAPoR12 combine these seamlessly in a single Electronic Cultural Atlas Initiative (ECAI)9 with resource for the community. Visualization tech- its utilization of Timemap10 provides an indica- niques using tables and graphs have been com- tion of possible future trends. The study of the monplace in text analysis for many years. These human use of space also finds expression are sometimes variants of statistical visualiza- through the discipline of Space Syntax,11 which tions of numeric data (as in word frequencies), originated in the fields of architecture and urban but in other cases they are more specialized design to assess the social effects of different visual forms of text analysis. Projects such as designs for living spaces (Hillier and Hanson, TAPoR and NORA13 are developing imaginative 1986). In recent years, it has been applied by new visual forms and applications. These new archaeologists and anthropologists, among methods apply not just to the visual representa- others, and is pertinent to another aspect of tion of the results of analysis, but also to the visualization; the application of 3D visualization visualization of the texts themselves. The aim is to studies of cultural heritage in Virtual Realities. to support interpretive scholarship by allowing However, space is important not only because of areas or relationships of interest to be identified the obvious spatial data applications of maps and within large volumes of text. Projects have GIS but also because it is at the heart of all explored specific texts in this way for many visualizations. The process of visualizing infor- years, for example Dante’s Inferno,14 but there mation always involves spatializing data even is substantial scope for a tool that could be when that data may not be intuitively spatial. applied to any text. The use of the word ‘tool’ For example, the data used by John Burrows to here should not be taken to imply that this is a plot graphs of idiolects in the novels of Jane computational problem as the problems being Austin (Burrows, 1987) is not intuitively spatial grappled with are conceptual not technical. For but space must be introduced into it in order to example, what is a text? How should it be dis- explore patterns and relationships within the played visually? Humanities computing provides data through visualizations such as graphs. a myriad of possibilities of representation; uni- Spatialization adds extra dimensionality that is dimensional or two-dimensional physical absent from written language and in doing so objects, abstract objects showing relations facilitates the display of simultaneity. among words or between words and annotations,  Quantitative data—quantitative analysis and and animations. Further questions arise from visualization has been an established tool in this work; is there a need for representational many humanities disciplines for a long time. It as well as interpretive markup? What are the can be found in generic statistical analysis soft- relationships between text visualization and the ware or embedded in specialized applications interpretation of texts? The visualizing of texts is

284 Literary and Linguistic Computing, Vol. 23, No. 3, 2008

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 Digital visualization as a scholarly activity

also an area which links humanities computing a number of successful examples from the past; work to the arts; for example the interactive for example Charles Minard’s 1861 narrative installation Text Rain15 by Camille Utterback graphic of Napoleon’s ill-fated campaign against and Romy Achituv. There is a movement Russia in 1812 where time, geography, and quan- towards tools in which aesthetics are as impor- titative data are presented in such as way that the tant as analysis, which in turn draw the human- scale, location, and timing of the loss of ities closer to the arts. The material culture of Napoleon’s troops during the entire campaign textual objects also offers opportunities for can be assimilated at a glance (Robinson, 1967; visualization, for example in manuscript studies Tufte, 1983, 2006). In many respects, this style of and projects, such as Turning the Pages16 at the graphic has been neglected since the advent of British Library. digital tools because the techniques used to con-  Time. Tools such as timelines allow one to struct diagrams of this type are not easily imple- explore the development of complex historical mented in software. They are, however, of events and the inter-relationships between pre- enormous value to the humanities in the study cursor events. They are of obvious value in the of narratives and the relationships between study of history, where later events build upon people, events, and artefacts. earlier ones but they can also be applied else-  3D Visualization—much of this work has where. and literary criticism are centred upon visualizations of the built environ- both histories of accumulated comments on a ment. This is of interest not only to subject. Matt Jensen has developed a number of and archaeologists but also to anyone who seeks timeline tools which are intended to answer the to find out how the buildings of the past worked styles of questions that are asked by humanists; in human terms. For example, the Theatre of 17 18 for example in the case of political scandals, Pompey and Theatron projects are based on questions of the style ‘who knew what and historical and archaeological data but are pri- when’, or for exploring the response to an marily of interest to scholars of theatre studies. author’s writing over a period of many years The 3D visualization is of special interest in the (Jensen, 2003, 2006). Before Denis Wood wrote context of this article because the sophistication of the links between geography and history, the of its imagery and advanced applications can eighteen century Jacques Barbeau- make it a form of visualization that is vulnerable Dubourg saw history as the study of two ancillary to criticisms of a lack of intellectual rigour (and fields; geography; and chronology (Wainer, where the absence of an established scholarly 2005). At that time geography was a relatively apparatus poses most problems). To counter young and rapidly developing discipline and he these criticisms a great deal of work is currently being focused on defining not only good practice believed that the processes of visualizing geogra- 19 phy through cartography had many parallels (ICT Methods Network ), but also principles for maintaining the intellectual integrity of with the visualization of time which would, he 20 believed, continue to develop in parallel with it. such work, for example the London Charter. Sadly chronology, perhaps better termed chrono- This work, especially documents such as the graphy, did not develop as a distinct discipline as London Charter, may therefore provide pointers both geography and cartography did. However, a for similar work in the development of a broader field of humanistic visualization. study of the visualization processes of map making does allow one to draw out some general Any demarcations between the applications of principles of visualization. The challenges tackled visualization in different disciplines are miscon- by cartographers in the representation of 3D ceived as there is a great deal of common ground. space on flat print media became an order of This offers considerable potential for the digital magnitude greater when a further dimension, humanities and it is here that attention needs to that of time, was added. Despite this there are be focused if digital visualization is to achieve

Literary and Linguistic Computing, Vol. 23, No. 3, 2008 285

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 M. Jessop

recognition as a rigorous intellectual activity in surrogates for primary sources and they generally research and teaching. feature a written caption that identifies the source, makes a comment of interpretation, and hints at ways that the visual source relates to the written 3 A Context for Digital account. Images are usually used singularly or if Visualization in Humanities used with other images there may not be a mean- ingful arrangement or juxtaposition between them. Scholarship David Hackett-Fischer’s Liberty and Freedom: A Visual History of America’s Founding Ideas is an elo- Establishing a context for digital visualization and quent and scholarly discussion of the development the use of images in the humanities is important for of the ideas of liberty and freedom in the United two . First, to demonstrate that the approach States (Hackett-Fischer, 2005). The book is heavily is not entirely new and that there are precedents that illustrated and makes extensive use of images to can be drawn upon. Second, as visualizations are a explore the iconography associated with those subset of images as a whole if the acceptance of ideas. It constructs a visual history but only very visualization as a scholarly activity suffers from the rarely does it juxtapose images to impart new mean- attitudes held towards images in general, where can ing. The images are usually presented singularly and one look for more positive examples of the use of isolated within the text, effectively being surrogates visual sources? for primary sources. If primary sources are the arte- David Staley identifies six types of representa- facts themselves and secondary sources are arrange- tional and abstract visual secondary sources that ments of those artefacts, there are occasional are used by historians but his taxonomy secondary sources in Hacket-Fischer’s book when (Staley,2003) applies equally well to all humanists. His categories are as follows: images are placed alongside each other to allow comparisons to be made. Thus because the arrange-  Galleries of images. ment is purposeful they show of the  Museums and collections of objects. author’s syntactic choices, omissions, commissions,  Film, television, and other moving images. arrangements, and analysis, and so become part of a  Dramatic re-creations. visual secondary source. By comparison, John  Maps and atlases. Berger’s (1972) Ways of Seeing contains a number  Pictures of data. of ‘visual essays’. These ‘essays’ have no words, only To these I would add a further category: a collection of paintings arranged in such a manner that the viewer is invited to see patterns for them-  Single images. selves. The skill and knowledge (and scholarship?) His contention is that these visual structures are of the author is expressed through the selection and as useful as verbal structures in organizing our arrangement of the images. Berger confirms the thoughts about the past and communicating these value of these visual essays as scholarly communica- thoughts to others. tion and inquiry when he says they are intended to In this context, galleries of images are not simply raise as many questions as the verbal essays that are collections of images but they are visual displays in also present in the book. So simply displaying visual which the spatial arrangement of visual primary sources is not sufficient for status as a visual sec- sources takes precedence over the linguistic arrange- ondary source; there must be some sort of purpose- ment of any written sources associated with them. ful arrangement or juxtaposition of several images. Information is communicated to the viewer through The origin of museums and collections of objects in the images and their deliberate juxtaposition. The Western culture can be traced to the ‘cabinets of art careful arrangement of these images allows viewers and curiosities’ or Kunstkammern that were popu- to discern patterns and relationships. Photographs lar between the sixteenth and eighteenth centuries. in scholarly publications are frequently used as These were collections of a wide variety of objects

286 Literary and Linguistic Computing, Vol. 23, No. 3, 2008

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 Digital visualization as a scholarly activity

ranging from natural materials to flora and fauna, that takes it beyond mere entertainment. Staley and manmade objects. The intention of the creators describes this as a ‘useful fiction’ (Staley, 2003); was to collect as many objects from the natural the play can be viewed as sharing characteristics world as possible; by displaying natural and man- with a scientific (or humanistic) model in that made objects together, creators of Kunstkammern ‘it describes a much simpler universe by including sought symbolically to possess and understand the some features of the actual universe and neglecting diversity of the outside world. They were intended others. It is an abstraction, a model, of reality not as spaces that would bring order and systematic the reality itself’ (Staley, 2003). It can be said to thought to the collected objects. These were not be ‘useful’ if it generates insights. Film, like theatre, simply jumbled collections of artefacts; each object can also depict vision, sound, speech, body lan- was carefully placed in a conscious manner. As with guage, and movement simultaneously in one modern digital visualizations their creators were medium. In so doing, it approaches what Arnheim seeking patterns and associations between the calls the ‘four-dimensionality of reality’ (Arnheim, objects. The Kunstkammern were primarily tools 1969). In a text-based piece the processes of selec- for enquiry and can thus be considered a type of tion and abstraction applied during the writing of visual secondary source. Many academics found a monograph also make it a useful fiction. Film them useful for theoretical reflection and saw does, however exhibit a characteristic that also them as models of the universe where the patterns afflicts many digital visualizations, in that it is diffi- of the natural world could be visually displayed. cult to display scholarly apparatus, such as bibliog- What is interesting is that the mode of thought raphies, footnotes, and information about context being used was visual and associative not linguistic and content. The space available here does not and logical (or analytical). Few would dispute that permit a discussion of the final four categories (dra- museum displays are not merely artless collections matic re-creations, maps and atlases, pictures of of primary sources, their success relies upon the data, and single images), but each has different con- conscious arrangement of objects into meaningful tributions to make to a discussion of the historical patterns. This is a process very similar to arranging context of visual sources and tools. Their creation the material for a written monograph about the represents valid scholarly activity and forms a con- same sources. In both cases researchers must make tinuum to which digital visualizations are a rela- compositional decisions regarding what to collect tively new addition. Museum displays, films, and and what to ignore. Decisions must be made con- re-enactments are all visualizations that can cap- cerning the placement of the objects, their juxtapo- ture more of the rich multi-dimensionality of sition, how they should be lit, the design of the human experience. They do, however, suffer from setting, and many other issues of compositional the bane of humanities research: the incomplete style. They are an abstraction that has been built record. This suggests a significant weakness of digi- with the application of considerable scholarly, and tal visualization which will have to be addressed if other, thought and knowledge derived from the its scholarly status is to be ensured. Visual sources analysis, interpretation, and synthesis of primary present the viewer with a complete, and convincing, and secondary sources, and can thus be regarded picture that is often derived from an incomplete as secondary sources themselves. record but the nature of the media used often To understand the importance of film it is neces- requires that the gaps be filled during its creation sary to examine another type of experience first. and thus concealed. If the applications of represen- When one attends a theatre one might expect to tational and abstract visual secondary sources are to gain some real understanding about human nature be regarded as anything other than mere entertain- from the people, places, and events that unfold ment it must be ensured that viewers are aware of upon the stage. It is understood that what is being not only what is present but also what is omitted observed is not reality, but that one can learn and the levels of uncertainty of that which is from it all the same if it is meaningful in a way present.

Literary and Linguistic Computing, Vol. 23, No. 3, 2008 287

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 M. Jessop

The preceding discussion shows that the princi- with the sciences being analytical, the arts being ples behind digital visualization are not new and creative and expressive, but what of the humanities? that there is a precedent for the use of visual sources My research concerning the use of GIS in the and visual tools in humanities scholarship. Digital humanities has shown me something that I believe visualization can be placed in context with estab- to be true of digital visualization as a whole: that lished methodology and, in common with the digi- potentially its greatest value is not for analysis but tal humanities as a whole; it forms part of a for synthesis and modelling. It can be used to draw continuum of scholarly method not a revolutionary material together, compare and combine it, or to upheaval of it. construct models of how, for example, space was Despite its location in the spectrum of accepted used in the public areas of Ancient Rome, or praxis digital visualization, and perhaps all digital develop tools for the more abstract use of space to humanities methodology, does face significant pro- visualize data that is not intuitively spatial. This is blems because of its youthfulness. Among these pro- an optimistic vision of the future, but there is a very blems are: fundamental obstruction and limitation to the pro-  The lack of a historical background involving gress that can be made. lengthy discussions of theoretical issues and The aims and working practices of the artists representing the 3D Visualization in the Arts methodology. 21  Measures to ensure the transparency of method, Network (3DVisA) caused me to rethink the processes of selection and omission, missing nature and potential of digital visualization in the data, etc. humanities. Years of school life, and many of adult-  A lack of in-built scholarly apparatus. hood, are spent mastering written language but the  A small community of users and practice. western system places little effort, or value on, teaching visual literacy. The natural expectation There are other more pragmatic problems but as of an artist who wishes to work in digital media is these have already been discussed elsewhere in a that they will spend many years mastering the different, but highly relevant, context (Jessop, medium. This is time that, in the past, they would 2008) I will focus primarily on the first two issues have spent mastering not just the techniques but the from the above list. However, before discussing ‘language’ of drawing, painting, or sculpture—but these the issue of visual literacy needs to be what is this language? In the nineteen century the addressed. Bauhaus movement attempted to define a language of visual communication and artists such as Paul Klee and Wassily Kandinsky produced detailed 4 Visual Literacy in the Arts, accounts of their efforts (Kandinsky, 1979). Well- Sciences, and Humanities respected guides such as Donis A. Dondis’s A Primer of Visual Literacy (Dondis, 1973) were pub- I recently attended a workshop which bought lished and became standard texts for the education together visualization practitioners from science, of artists. Semioticians have also tried to formalize a engineering, visual arts, performing arts, and the language of visual images. Jacques Bertin proposed humanities. The workshop sought to find intersec- a system similar to that of Dondis in which the visual tions between the visualization methods and appli- marks that make up images are structured by one cations of the diverse disciplines represented there. or a combination of variables of shape, texture, The technology is largely common to all and there is value, colour, size, and orientation (Bertin, 1967/ a great deal of overlap in the methods applied, but it 1983). His study went beyond artistic and graphic was the differences in the aims and views of partici- design images to consider a wide range of visual pants and of their work that I found most illumi- compositions that encompassed two-dimensional nating. There are clearly differences in the visualizations such as maps, diagrams, and charts. methodologies and aims of the three main groups Thus, it must be noted that during their education

288 Literary and Linguistic Computing, Vol. 23, No. 3, 2008

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 Digital visualization as a scholarly activity

digital artists are not spending those years learning knowledge (which I believe it is clear they do), how to use the technology, but rather the time is the problem becomes not one of ‘is visualization spent learning how to express oneself and commu- a scholarly activity?’ but ‘how can the integrity of nicate through digital visual media. They are taking its scholarship be ensured?’ that time to learn the ‘language’ of creative digital One of the most visually exciting applications of visualization. Taking a parallel example from the digital visualization occurs in the research and com- sciences, how many of us would be able understand munication of cultural heritage. It is an area in a blackboard covered with notation written in one of which three-dimensional visualization is being the chosen ‘languages’ of the scientist, advanced employed in a wide range of contexts and provides mathematical formulae? Scholars in each area have a useful case study. If the full potential of the meth- spent many years mastering their respective commu- odology is to be realized it is essential that ways are nication media, using ‘language’ which can only be found to ensure that such work is intellectually and understood by others who share the same knowledge technically rigorous. In this instance, the need is to and experience of that medium. Humanists have establish standards that are responsive to the partic- similar knowledge and skills for the use of written ular properties of 3D visualization; however, many language but little in the way of visual literacy, but of these standards would be more widely applicable still expect to be able to make effective use of images, to other branches of visualization, and indeed the in the form of visualizations, to explore and com- application of digital tools to cultural heritage and municate ideas with very little (if any) additional the digital humanities in general. study. That people so frequently succeed in doing The cultural heritage community has been apply- so is commendable but how much more could ing 3D modelling and Virtual Reality for a number they do if they had higher levels of visual literacy? of years. Many publications have arisen from this Our education system equips our students, and our- work and an active community of practice has devel- selves, with a worryingly low level of visual literacy to oped. All recognize the importance of ensuring that cope with an environment that is increasingly visualization methods are applied with the same dependent on visual media. I have observed a low intellectual rigour that guides traditional, and other level of awareness of the meaning of even simple digital, scholarship. At the core of any discussion is graphics among my students. There is a strong argu- the need for visualizations to accurately convey to ment for increased provision of visual literacy users the distinctions between evidence and hypoth- education alongside any drive to exploit digital esis, and between different levels of probability. The visualization. At the very least this situation demon- London Charter is a ‘work in progress’ established by strates the necessity of expanding the multi- members of the digital cultural heritage community disciplinary teams involved in humanities digital to tackle these and many other issues. The Charter visualization projects to include professional artists. ‘aims to define the basic objectives and principles of Their contribution cannot be dismissed as being the use of 3D visualization methods in relation to merely ‘aesthetic’ for those aesthetics are deeply intellectual integrity, reliability, transparency, docu- embedded in the effective use of the medium of mentation, standards, sustainability, and access’.22 digital visualization. The field is changing rapidly and encompasses an expanding range of research aims so the Charter sets out not to prescribe specific aims or methods 5 A Framework for Scholarly but rather to establish a set of ‘broad principles for Standards in Digital Visualization: the use, in research and communication of cultural heritage, of 3D visualization upon which the intel- The London Charter lectual integrity of such methods and outcomes depend’. Its aims are to: If the characteristics and research outcomes of digital visualization satisfy the roles of the discovery,  Provide a benchmark having widespread recog- exchange, interpretation, and presentation of nition among stakeholders.

Literary and Linguistic Computing, Vol. 23, No. 3, 2008 289

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 M. Jessop

 Promote intellectual and technical rigour in such made carefully, based on the best available knowl- uses. edge and experience, and be reviewed periodically,  Enable appropriate evaluative criteria and meth- resources permitting, as the research process pro- ods to be determined and applied. gresses’.23 One further point is that it is important  Stimulate debate on methodological issues. that the project documentation records why the  Offer a robust foundation upon which specialist choice of a particular method was made thus subject communities can build detailed standards making it easier to defend or review the research and guides. process. This is echoed in the design of projects,  Ensure appropriate access and sustainability such as TAPoR,24 which include mechanisms to strategies are determined and applied. record process and intermediate results as well as  Enable 3D visualization authoritatively to final outcomes (rather like the paradigm of scientific contribute to the study, interpretation, and man- research, or indeed the assessment of artists in agement of cultural heritage assets. training). The Charter addresses these objectives through eight 5.2 principles: Sources Scholarly practice stresses the importance of identi- (1) Subject communities fying and evaluating sources in a structured way. (2) Aims and methods This is equally important when ensuring the intel- (3) Sources lectual integrity of digital visualization methods and (4) Transparency requirements outcomes. The sources must be evaluated as rigor- (5) Documentation ously as they would be for the production of any (6) Standards secondary or tertiary source. One of the items in this (7) Sustainability section of the Charter leads on to the next point (8) Access ‘Careful consideration should be given to the aims All of these are relevant to the maintenance of high and contexts for both visualization creation and dis- standards of scholarship in any digital humanities semination in order to determine whether, or to activity, but four are particularly pertinent to this what extent, the sources considered and the ratio- article: aims and methods; sources; transparency nale for their interpretation, should be published requirements; and documentation. with the visualization outcomes.’25

5.1 Aims and methods 5.3 Transparency requirements 3D visualization methods are used to achieve a Images are seductive and there is a natural tendency wide range of research and communication aims to instinctively believe whatever one sees with one’s in the cultural heritage sector. The visual impact own eyes but in the case of digital visualizations and appeal of the methods must not be abused or what is been seen is entirely a constructed object allowed to seduce the viewer, rather a 3D visualiza- and it is essential to also know the evidence upon tion should only be used to address an aim when which it is based. The Charter observes that ‘Opacity it is the most appropriate and effective available regarding the relationship of sources to outcomes method for that purpose. This is true of many makes 3D visualization anomalous among research visualization methods, right down to the level of methods, and may help to account for the lack of producing a graph with a small data set when in recognition of 3d visualization as a valid research fact a table would be a better device to use. What process or outcome in certain subject commu- is required is a systematic evaluation of the suitabil- nities.’26 It should therefore always be possible to ity of the method. A great deal of the work in the evaluate a visualization independently of the digital humanities is innovative so there are often authority claims of its creators. If this is not possible no precedents on which to base such decisions. its significance as a research method or outcome In these cases ‘the choice of method should be remains indeterminable. Therefore, visualization

290 Literary and Linguistic Computing, Vol. 23, No. 3, 2008

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 Digital visualization as a scholarly activity

outcomes should be ‘disseminated with sufficient ‘process and outcomes of 3D visualization creation information to allow the relevant subject commu- should be sufficiently documented to enable the nities to understand and evaluate the choice and creation of accurate transparency records, potential application of the method in relation to its reuse of the research conducted and its outcomes aims’.27 The Charter suggests that ‘Sufficient infor- in new contexts, enhanced resource discovery and mation should be provided to allow 3D visualiza- access, and to promote understanding beyond the tion methods and outcomes to be understood and original subject community’.29 The Charter contains evaluated appropriately in relation to the contexts in many specific guidelines in this area but the overall which they are used and disseminated.’28 This is aims are to: broken down in to:  Ensure the transparency of methods by explicit  The kind and status of information the visualiza- statements of what was done and why. This tion represents including the nature and degree allows the method to be assessed and facilitates of factual uncertainty of a hypothetical reuse of the work. reconstruction.  Ensure the transparency of the use of sources.  It also recognizes that the type and quantity of  Assist in the dissemination of knowledge derived transparency information will vary between pro- from the project. jects and at different phases of the same project.  To provide some of the scholarly apparatus that is otherwise missing from non-textual media. New scholarly apparatus similar to those used  Open a debate on methodological and theoretical for textual scholarship must be developed and issues to compensate for the lack of a framework applied. The medium employed to implement this similar to that that has evolved over many hun- apparatus need not be textual, what is important is dreds of years for other research methods. This that the method of providing the transparency is complicated further by the collaborative and information should be the most appropriate avail- interdisciplinary nature of research in the digital able whether it is graphical, textual, video, audio, humanities. Co-workers may lack a common numerical, or combinations of these. The varying episteme or set of conventions that generally levels of experience and knowledge in different characterize subject communities. It is therefore parts of the community mean that the level of essential to reflect upon the ‘unspoken assump- documentation required will vary; in particular, tions and different lexica of the different subject new and/or unfamiliar methods will require more communities engaged in the common visualiza- explanation. In addition, different levels of ‘assumed tion process’.30 knowledge’ apply within subject communities. This also implies that transparency information require- ments may change as levels and sophistication of 6 understanding of particular visualization methods Conclusion rise and that these requirements will also vary Digital Visualization has potential applications in between communities. all areas of the humanities and can be applied to explore virtually any form of data. It does not exist 5.4 Documentation in isolation; rather it is part of a continuum of visual The reuse of research outcomes in new contexts, sources and methodologies the majority of which the enhancement of discovery and access mecha- are also relatively young and so lack the long his- nisms, and facilitating interdisciplinarity are at the tories of debate that have shaped other forms of core of most humanities computing work. The scholarship. This is a significant factor impacting Charter deals specifically with 3D visualization but upon its acceptance as a scholarly activity. its recommendations aim to formalize a strategy In the humanities, it is accepted that texts that applies to much of the work performed by and documents are produced as readings resulting digital humanists. The Charter suggests that the from acts of interpretation between the reader and

Literary and Linguistic Computing, Vol. 23, No. 3, 2008 291

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 M. Jessop

the text; the images produced by digital visualiza- Berger, J. (1972). Ways of Seeing. London: BBC and tions now need to be regarded in the same way. Penguin. Every representation, visual, or otherwise, is an Bertin, J. (1967/1983). The Semiology of Graphics: effort to structure an argument and as such it is a Diagrams, Networks and Maps (Berg, W. J., Trans.). rhetorical device. Although there may be a case for Madison, WI: University of Wisconsin Press. striving to define the visual ‘language’ used in this Burrows, J. F. (1987). Computation into Criticism. form of communication, at this stage this is a minor Oxford: Clarendon Press. consideration. Levels of visual literacy are currently Dondis, D. A. (1973). A Primer of Visual Literacy. so low that the effort may be better spent elsewhere, Cambridge, MA: MIT Press. such as raising general understanding of visual Donovan, E. K. (2007). Sacra Imago: Devotional Art of the sources. What needs to be understood (and disse- Middle Ages. Krannert Art Museum Exhibition Guide. minated) at this time is the relationship between Urbana-Champaign, IL: University of Illinois. what is being communicated and how it is being Drucker, J. (2003). Graphesis: Visual Knowledge Produc- communicated through digital visualizations. It is tion and Representation. http://www.noraproject.org/ clear that visualization is a scholarly activity rich reading.php (accessed 20 October 2007). in opportunities to create knowledge. The method- Hackett-Fischer, D. (2005). Liberty and Freedom: A ology itself and the theoretical issues surrounding it Visual History of America’s Founding Ideas. New York, are still young and lack the long history of debate Oxford: Oxford University Press. that has shaped other scholarly activities. It is there- Hillier, W. and Hanson, J. (1986). The Social Logic of fore essential that such a debate is begun and sus- Space. Cambridge: Cambridge University Press. tained; frameworks such as the London Charter Ivins, W. M. (1969). Prints and Visual Communication. will be useful in this process. The Charter and the Cambridge: MIT Press. organizations that are emerging within groups of Jensen, M. (2003). Visualizing Complex Semantic researchers will also assist in the important process Timelines. Newsblip Technical Report 2003001, of building a community of practice. http://www.newsblip.com/tr/nbtr2003001.pdf (accessed Information graphics, and indeed images gener- 27 October 2007). ally, are expressions of procedures for generating Jensen, M. (2006). Conference Presentation Semantic knowledge through the act of visualization and Timeline Tools for History and Criticism. Digital ways of displaying knowledge embodied in visual Humanities Conference 2006, CATI, Universite´ Paris, imagery (Drucker, 2003). Visualization addresses Paris. epistemological and pedagogical issues that are Jessop, M. (2006). Dynamic Maps in Humanities common to the digital humanities and are at the Computing. Human IT, 8.3, 68–82. http://www.hb.se/ forefront of the developing discipline of humanities bhs/ith/3-8/mj.pdf (accessed 26 August 2007). computing. This is a vast topic and the main aim Jessop, M. (2008). The Inhibition of Geographical here has been to argue the validity of digital visual- Information in Digital Humanities Scholarship. ization as a scholarly methodology; demonstrate Literary and Linguistic Computing, 23(1): 39–50. that it is part of a continuum of established use of Kandinsky, W. (1979). Point, Line and Plane. New York: visual methods and sources in the humanities; and Dover Publications. identify some steps that are required to ensure its Moretti, F. (2005). Graphs, Maps, Trees: Abstract Models acceptance as having the intellectual rigour required for a Literary . London: Verso. of a scholarly methodology. Robinson, A. H. (1967). The Thematic Maps of Charles Joseph Minard. Imago Mundi, 21: 95–108. Staley, D. (2003). Computers, Visualization and History. New York: M. E. Sharpe. References Tufte, E. R. (1983). The Visual Display of Quantitative Arnheim, R. (1969). Visual Thinking. Berkeley: University Information. Cheshire, Connecticut: Graphics Press, of California Press. pp. 40–1.

292 Literary and Linguistic Computing, Vol. 23, No. 3, 2008

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017 Digital visualization as a scholarly activity

Tufte, E. R. (2006). Beautiful Evidence. Cheshire, extensively applied in the fields of architecture, Connecticut: Graphics Press, pp. 122–39. urban design, planning, transportation, and interior Wainer, H (2005). Graphic Discovery: A Trout in the Milk design. Over the past decade, space syntax techniques and other Visual Adventures. Princeton, NJ: Princeton have also been used for research in fields as diverse as University Press. archaeology, information technology, urban and human geography, and anthropology’. See http:// Wood, D. (1992). The Power of Maps. New York: Guilford www.spacesyntax.org/introduction/index.asp Press. (accessed 30 August 2007). 12 Text Analysis Portal for Research (TaPOR) http:// portal.tapor.ca/portal/portal (accessed 10 October Notes 2007). 1 This is commonly believed to be based on a ‘Chinese 13 NORA Project http://www.noraproject.org/ (accessed proverb’ however Paul Martin Lester believes that it 20 September 2007). was in fact made up by the advert writer Frederick R. 14 A Visualization of Dante’s Inferno. http://www3.iath. Barnard. See http://commfaculty.fullerton.edu/lester/ virginia.edu/hell/(accessed 15 October 2007). writings/1000pictures.html (accessed 26 August 2008). 15 Text Rain by Camille Utterback http://www. 2 Wikipedia at http://en.wikipedia.org/wiki/Scholarly_ camilleutterback.com/textrain.html (accessed 11 method (accessed 3 September 2007). October 2007). 3 Ibid. 16 Turning the Pages http://www.bl.uk/onlinegallery/ttp/ 4 Ibid. ttpbooks.html (accessed 27 october 2007). 5 A ‘method’ is taken to be synonymous with a 17 Theatre of Pompey Project http://www.pompey.cch. ‘technique’. kcl.ac.uk/ (accessed 11 October 2007). 6 ‘Methodology’ refers to more than a simple set of 18 The Theatron Project http://www.theatron.org methods; rather it refers to the rationale and the phil- (accessed 26 August 2008). osophical assumptions that underlie a particular 19 ICT Methods Network http://www.methodsnetwork. study. ac.uk/ (accessed 11 October 2007). 7 Valley of the Shadow http://valley.vcdh.virginia.edu/ 20 The London Charter http://www.londoncharter.org/ (accessed 10 October 2007). (accessed 11 October 2007). 8 Salem Witch Trials http://etext.virginia.edu/salem/ 21 3D Visualization in the Arts Network (3DVisA) http:// witchcraft/ (accessed 10 October 2007). 3dvisa.cch.kcl.ac.uk/index.html (accessed 10 October 9 Electronic Cultural Atlas Initiative (ECAI) http:// 2007). ecai.org/ (accessed 20 September 2007). 22 The London Charter http://www.londoncharter.org/ 10 TimeMap http://www.timemap.net/ (accessed 20 (accessed 11 October 2007), p. 1. September 2007). 23 Ibid., p. 5. 11 The space syntax laboratory at University College 24 The TAPoR URL is http://portal.tapor.ca/portal/portal London describes Space syntax as ‘a set of techniques (accessed 15 October 2007). for the analysis of spatial configurations of all kinds, 25 The London Charter http://www.londoncharter.org/ especially where spatial configuration seems to be (accessed 11 October 2007), p. 6. a significant aspect of human affairs, as it is in build- 26 Ibid., p. 7. ings and cities. Originally conceived by Bill 27 Ibid., p. 7. Hillier and his colleagues at The Bartlett, UCL in the 28 Ibid., p. 7. 1980s as a tool to help architects simulate the likely 29 Ibid., p. 8. effects of their designs, it has since grown to become 30 The London Charter http://www.londoncharter.org/ a tool used around the world in a variety of research (accessed 27 October 2007), p. 8. and areas and design applications. It has been

Literary and Linguistic Computing, Vol. 23, No. 3, 2008 293

Downloaded from https://academic.oup.com/dsh/article-abstract/23/3/281/937022 by Washington State University Libraries user on 27 November 2017