Mapping an ancient historian in a digital age

HESTIA (the Herodotus Encoded Space-Text-Imaging Archive) employs the latest digital technology to develop an innovative methodology to the study of spatial data in Herodotus’ Histories. Using a digital text of Herodotus, freely available from the Perseus on-line library, to capture all the place-names mentioned in the narrative, we construct a database to house that information and represent it in a series of mapping applications, such as GIS, GoogleEarth and GoogleMap Timeline. As a collaboration of academics from the disciplines of Classics, Geography, and Archaeological Computing, HESTIA has the twin aim of investigating the ways geography is represented in the Histories and of bringing Herodotus’ world into people’s homes. Digital technology is fast revolutionising the ways in which we are communicating with each other and perceiving the environment around us. With internet access and web-mapping tools installed as standard features on the latest mobile phone technology, the whole world now can appear at our fingertips. In university circles too there can be little doubt of the role ICT has in the development of future academic practice, both in the sciences and arts, whether as a pedagogical tool or as a means of research: certainly, with major humanities funding bodies putting the digital world at the centre of grant applications, the * We are extremely grateful to the AHRC for funding this research (ID. No. AH/F019459/1). We would also like to thank Malcolm Heath and Kathryn Stevens for perceptive remarks and the many audiences before whom this research has been aired, including those at the e-Science Institute in Edinburgh, the TOPOI research cluster on Herodotus (Berlin), and the DFG-Perseus workshop at Tufts University. (For a full listing go to: http://www.open.ac.uk/Arts/hestia/activities/index.html.) All translations are taken from the Loeb edition by A.D. Godley (Cambridge MA 1920), with adaptations. † Stefan Bouzarovski is also a Visiting Professor at the Department of Social Geography and Regional Development, Faculty of Science, Charles University, Albertov 6, 128 43 Prague 2, Czech Republic, and an External Professor at the Department of Economic Geography, University of Gdańsk, Bażyńskiego 4, 80-952 Gdańsk, Poland. His work is partly supported by the Ministry of Education, Youth and Sports of the Czech Republic, under project no. MSM0021620831, titled ‘Geographic Systems and Risk Processes in the Context of Global Change and European Integration’. ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 2 next decade promises a revolution in approaches to the study of antiquity. Yet, what this revolution will look like, or even how revolutionary the digitalisation of Classics will be, is still very much open to question. It is not only an issue of whether more traditional areas of study will be affected or able to take advantage of the latest technology, though many already are as a matter of course. Even those studies that embrace the new ICT face an uncertain future in this brave new world. With software developing so quickly, what is to prevent work done in one format being rendered practically obsolete by future trends or simply left unrecorded in the virtual world of the digital age? What infrastructure is there to support digital projects in the same discipline, let alone those in different disciplines or across institutions and nations, and, in its absence, how can projects speak to each other in the face of the growing multiplicity of media? 4 What single platform can be established for programmers and users alike, and how is a common standard to be found, managed and maintained? And what impact will these new forms of media, these new ways of gathering, representing and 1 ‘Digital Humanities’ is one of only four ‘emerging themes’ in the AHRC’s ‘Future Directions’ consultation (http://www.ahrc.ac.uk/About/Policy/Documents/FD%20emerging%20themes.pdf). See also the JISC (Joint Information Systems Committee) homepage (http://www.jisc.ac.uk/) and their video ‘How digital technologies are creating a new paradigm in research’ (http://www.jisc.ac.uk/whatwedo/campaigns/res3/video). An initial assessment of the role of digital media in the Humanities is attempted in the article by E.G. Toms and H.L. O’Brien, ‘Understanding the information and communication technology needs of the e-humanist’, Journal of Documentation 64 (2008), 102-30. 2 Such as the widespread deployment in philologically oriented research of the Thesaurus Linguae Graecae (TLG: http://stephanus.tlg.uci.edu/), or the use of the on-line database L’Année philologique (http://www.annee-philologique.com/aph/), the most comprehensive index to scholarly work in Classical Studies. 3 The classic example of a venture whose format was soon outstripped by technological advances was the BBC Domesday project (http://en.wikipedia.org/wiki/BBC_Domesday_Project). To a certain extent the rise of institutional on-line research archives, such as the University of Oxford’s Research Archive (ORA: http://ora.ouls.ox.ac.uk/) or the Open University’s Research On-line (ORO: http://oro.open.ac.uk/), has come about in an effort to address the problem of preservation. But these initiatives, while being of crucial importance particularly for maintenance of software formats used in traditional scholarship, do not address problems of compatibility between, and issues of sustainability of, current digital projects themselves. 4 Just knowing what digital resources are available to the scholarly community will be a major task in itself. It is true that the AHRC hosts a handy website of humanities projects (http://www.artshumanities.net/), which can be searched according to discipline, method or resource: but these are limited to AHRC-umbrella projects that are currently on-going. 5 Various inter-organisational humanities-based projects are forming to meet the challenge of the rapidly changing digital world, from the large-scale cross-institutional projects such as Bamboo (which tackles the question ‘How can we advance arts and humanities research through the development of shared technology services?’ at http://projectbamboo.org/), Digital Research Infrastructure for the Arts and Humanities (DARIAH: http://www.dariah.eu/) and the Coalition of Humanities and Arts Infrastructures and Networks (CHAIN: http://digitalhumanities.org/ centernet/?page_id=12), to smaller scale bottom-up groups, such as Digital Classicist (http://www.digitalclassicist.org/) or Antiquist (http://www.antiquist.org/blog/). ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 3 interpreting data, have on approaches to Classics or the Humanities more generally? While the general issues outlined above must in the end be addressed by strategy units at the institutional level and fall beyond the scope of an academic paper, they are a major concern for a particular case study with which we are currently engaged. HESTIA—the Herodotus Encoded Space-Text-Imaging Archive funded by the Arts and Humanities Research Council of the United Kingdom—investigates the ways in which space is represented and conceived in the work of the ancient Greek historian, Herodotus, using the latest information computer technology. In his Histories, in the process of conducting his enquiry into war between Greece and Persia, Herodotus has cause to mention a whole range of different places: a digital mark-up of the text has allowed us to feed these locations into a database and reconstruct the world of the sixth-fifth centuries BCE (of Herodotus) using modern mapping systems. In this paper we outline our methodological approach to the examination of spatial ideas in Herodotus, in the belief that there are lessons to be learned, not only for the scholar of the Histories or of ancient historiography more generally, but also for anyone interested in developing ICT in large text-based corpora, particularly with regard to thinking about the way places are represented and conceived. Herodotus’ opening salvo in his investigation—why was it that the Greeks and barbarians came into conflict with each other?—launches a narrative that is fundamentally interested in issues of space—primarily the growing reach of Persia and the places that come under her dominion or try to resist her power. The 6 The growing awareness of the potential impact of digitalisation on mainstream Classics can be evidenced by it being the subject of two special lectures delivered during 2009: Donna Kurtz (University of Oxford) gave the closing lecture of the Fédération internationale des Associations d’études classiques (FIEC, Berlin on 29 August 2009), on ‘The future of the past: CLAROS’ (the Classical Art Online Research Services); and one of the inaugural lectures in King’s College London’s series The Arts, the Academy and the World was on the theme of ‘Decoding Pasts, Building Futures’ (http://kcl.ac.uk/schools/humanities/week/arts/pastfuture.html). We at HESTIA are also trying to build ‘new worlds out of old texts’ (‘Introduction’ by T.J. Barnes and J.S. Duncan, in Barnes and Duncan (ed.), Writing worlds: Discourse, text and metaphor in the representation of landscape (London 1992), 3). 7 See our website http://www.open.ac.uk/Arts/hestia/index.html for more information about our research goals. The use of ICT mapping resources in the Humanities was the major theme of the eScience institute workshop, entitled ‘Mapping Information with and without Geography: Approaches to Data Visualisation and Structure in the Arts, Humanities and Social Sciences’, held in Edinburgh in September 2009 (more information at http://www.nesc.ac.uk/esi/events/1005/). 8 While there has been recent interest in space within archaeology and history, as evidenced by a specialist panel at the 2007 Classical Association Conference dedicated to exploring the interactions of monuments, spaces and rituals in Delphi and Athens, there has been little recent work on submitting ancient Greek literary texts to spatial examination. In fact, a recent article by Tom Harrison on Herodotus (‘The place of geography in Herodotus’ Histories,’ in C. Adams and J. Roy (ed.), Travel, Geography and Culture in Ancient Greece, Egypt and the Near East (Oxford 2007), 44-65) assigns the ‘search of geography and its place in the Histories’ (44) the highest priority, yet fails to consider the potential for historical agents to offer differing views of space, the impact on conceptions of space when represented as discourse, or its performance within Herodotus’ text. ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 4 world that Herodotus subsequently depicts is commonly represented in a form like that of Figure 1. As will be immediately evident from such an image, the gap between representation and its objective—the actual topography of the area as we know it—is considerable: as a result, one primary objective of HESTIA is to deploy the latest satellite imaging freely available from NASA to re-present Herodotus’ world in a more accurate fashion in a form most readily accessible to modern-day users (see Figures 4a-c). Yet, for all of its drawbacks, this nonrealistic map does raise several important issues relating to the conception of space in the Histories, most notably the division of the world into three separate units, Europe, Asia and Libya, the importance of water bodies, in particular rivers, for organising that space, and with it the social, political, and cultural construction of representations of space. Indeed, ‘mental maps’ of this sort are growing in popularity in contemporary studies precisely in recognition of their value in bringing to light the ways in which space may be held to mean different things by different groups, which is a concern too for HESTIA, as we explain below. And yet..., still there is something not quite right with the image in Figure 1: this picture captures only a snap-shot of places mentioned in Herodotus, which otherwise presents a blank canvas regarding their flow through time—a point Herodotus himself is alert to. Immediately after his tongue-in-cheek account of the Persian rationalisation of (Greek) mythology, which has the effect of absolving them of blame for the recent conflict with the Greeks, Herodotus explains his own narrative focus (on Croesus and the generations after him) in terms of ‘going through in detail towns of men both small and great alike: for of the places that were once great, most have now become small, while those that were great in my time were small before’ (ὁμοίως σμικρὰ καὶ μεγάλα ἄστεα ἀνθρώπων ἐπεξιών· τὰ γὰρ τὸ πάλαι μεγάλα ἦν, τὰ πολλὰ σμικρὰ αὐτῶν γέγονε· τὰ δὲ ἐπ ̓ ἐμεῦ ἦν μεγάλα, πρότερον ἦν σμικρά, 1.5). Another of HESTIA’s aims, 9 This map has been chosen because, being published under a Wikimedia Commons license, it can be reproduced without any infringement of copyright—as is the case for all maps included here. But this map is typical of other reproductions of Herodotus’ world: see O.A.W. Dilke, Greek and Roman Maps (London 1985), 58. 10 For the importance of natural boundaries to Herodotus’ conception of history and, in particular, rivers as demarcation boundaries and markers of transgression, see esp. H. Immerwahr, Form and Thought in Herodotus (Cleveland 1966), in the index under river motif; cf. D. Braund, ‘River frontiers in the environmental psychology of the Roman world’, in D.L. Kennedy, ed., The Roman Army in the East (JRA Supp 17, 1996), 43-7. This was also the subject of Katherine Clarke’s presentation at the inaugural HESTIA workshop in Oxford on 13 January 2009 in a paper entitled ‘On rivers: being transgressed and assisting the “good”‘. 11 See e.g. N.J.W. Thrower, Maps and Civilization: Cartography in Culture and Society (Chicago 1996); J.B. Harley, ‘Deconstructing the map’, Cartographica: The International Journal for Geographic Information and Geovisualization 26 (1989), 1-20; P. Jackson, Maps of Meaning: An Introduction to Cultural Geography (London 1994); K.T. Jones, ‘Scale as epistemology’, Political Geography 17 (1998), 25-8. 12 So-called ‘cartograms’ are especially eye-catching in the way that they can represent territories in relation to a particular subject. So, for example, the larger a country’s GDP, the more inflated it looks, and vice versa. See, e.g.: http://www.bbc.co.uk/blogs/thereporters/markeaston/2008/10/ map_of_the_week_the_wealth_of.html ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 5 then, is to be more sensitive to the narrative context in which a particular location is mentioned; in other words, we tie Herodotus’ naming of a place to its position in the text. This allows us to draw a series of maps that depict a world in flux, as well as to capture a sense of space as something lived, not abstractly conceived (Figure 5). But even this approach has yet quite to represent space in a way that sufficiently captures its varied manifestations in Herodotus’ narrative: for a further problem with the static scene portrayed by Figure 1 is the absence of any relational indicators—it is, to put it simply, a world without lines of interaction, where regions are separate from each other, and whose boundaries are rigidly policed by bodies of water. To a certain extent, as we have mentioned, this kind of ideologically marked division of the world into discrete units does map onto contemporary Greek thinking regarding self-definition against others, which is present too in Herodotus’ aim. But, as recent scholars have begun to show, the world of the Mediterranean from antiquity onwards may be better thought of as a ‘contact zone’, in which the key theme is connectivity rather than polarity, and where the seas and rivers act as facilitators of movement through and exchange between different places across the region. Now, it is certainly debatable whether Herodotus represents, or is even interested in representing, a network culture that accurately maps onto his historical circumstances—the situation on the ground, as it were—thus making it problematic to use his narrative as evidence for the existence of specific real-life networks. But, at the same time, at every step of the way through the narrative points of contact are made between different places in a variety of ways, either by Herodotus himself or by his historical agents. By seeking to lift these connections out of the text, we hope to counter the conventional emphasis on topography, which appears all too evident from our contemporary viewpoint of the world afforded by satellite imaging, and refocus attention instead on the topological relationships between places—the links that depend on human agency and the associative clusters that certain places form over the course of the narrative (Figures 8-10). It may also represent an 13 E.g. E. Hall, Inventing the Barbarian (Oxford 1989); P.A. Cartledge, The Greeks and Others (Bristol 2004 [1993]). But the picture of the ‘self and other’ in Herodotus’ narrative is rather more complicated than that, and often shifts according to context so that the Persians may appear irrational and barbaric when facing the Greeks, but a good deal more orderly and civilised when fighting the Scythians: so F. Hartog, In the Mirror of Herodotus (Berkeley 1988 [1980]). For a vigorous challenge to a simple ‘east vs. west’ dichotomy in Herodotus, where it is not a total surprise ‘to find Self in Other and Other in Self’, see C.B.R. Pelling, ‘East is east and west is west—or are they?’, Histos 1997 (http://www.dur.ac.uk/Classics/histos/1997/pelling.html). 14 The Mediterranean as a ‘contact zone’: J. Horden and N. Purcell, The Corrupting Sea (Oxford 2000); as a sea of ‘connectivity’: C. Constantakopolou, The Dance of the Islands (Oxford 2005). On networks in antiquity, see esp. I. Malkin, The Returns of Odysseus (Berkeley 1998); I. Malkin, C. Constantakopoulou and K. Panagopoulou (ed.), Greeks and Roman Networks in the Mediterranean (New York 2009). 15 A point made with characteristic incisiveness and vim by Nicholas Purcell at the project’s preliminary workshop (13 January 2009). ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 6 alternative route towards constructing a series of ‘mental maps’ of the world as conceived of by those who lived it. Before we go on to outline and discuss in more detail our application and use of digital resources to address some of these ideas, let us first take a brief example from the text to demonstrate their importance, particularly because close reading of this nature is used to bring to light spatial concepts that underpin our employment of ICT. In the fifth book of his Histories, Herodotus describes a meeting between Cleomenes, king of Sparta, and Aristagoras, tyrant of Miletus, who has come to Lacedaemon to solicit support for a revolt of Ionian Greeks from Persian control. Coming as it does at a critical juncture in the narrative, as Herodotus makes the transition from mapping out Persian power to narrating the subsequent conflict with the Greek world, this episode brings to light a number of important issues for thinking about space in Herodotus. Having been admitted to converse with Cleomenes, and armed with ‘a bronze tablet on which the map of all the earth was engraved, and all the sea and all the rivers’ (ἔχων χάλκεον πίνακα ἐν τῷ γῆς ἁπάσης περίοδος ἐνετέτμητο καὶ θάλασσά τε πᾶσα καὶ ποταμοὶ πάντες, 5.49.1), Aristagoras first appeals to the Spartan’s sense of Hellenic comradeship, before going on to articulate the material benefits of an Asian invasion, using his map as a visual support (5.49.5-7): ‘The lands in which they dwell lie next to each other, as I shall show: next to the Ionians here are the Lydians, who inhabit a good land and have a great store of silver.’ (This he said pointing to the map of the earth which he had brought engraved on the tablet.) ‘Next to the Lydians,’ said Aristagoras, ‘are the Phrygians here to the east, men that of all known to me are the richest in flocks and in the fruits of the earth. [6] Close by them are the Cappadocians, whom we call Syrians, and their neighbours are the Cilicians, whose land reaches to the sea here, in which here the island of Cyprus lies... [7] Adjoining these is the Cissian land here, in which, on the river Choaspes here lies that Susa, where the great king lives and where the storehouses of his wealth are located...’ “κατοίκηνται δὲ ἀλλήλων ἐχόμενοι ὡς ἐγὼ φράσω, Ἰώνων μὲν τῶνδε οἵδε Λυδοί, οἰκέοντές τε χώρην ἀγαθὴν καὶ πολυαργυρώτατοι ἐόντες.” δεικνὺς δὲ ἔλεγε ταῦτα ἐς τῆς γῆς τὴν περίοδον, τὴν ἐφέρετο ἐν τῷ πίνακι ἐντετμημένην. “Λυδῶν δέ” ἔφη λέγων ὁ Ἀρισταγόρης “οἵδε ἔχονται Φρύγες οἱ πρὸς τὴν ἠῶ, πολυπροβατώτατοί τε ἐόντες πάντων τῶν ἐγὼ οἶδα καὶ πολυκαρπότατοι. [6] Φρυγῶν δὲ ἔχονται Καππαδόκαι, τοὺς ἡμεῖς Συρίους καλέομεν. τούτοισι δὲ 16 According to contemporary geographers, the two-dimensional ‘Cartesian’-style map, which has dominated Western horizons since the Enlightenment (A.J. Gurevich, Categories of Medieval Culture (London 1985)), has in effect replaced the ‘discontinuous patchy space of practical paths by the homogeneous, continuous space of geometry’ (D. Harvey, Consciousness and the Urban Experience (Baltimore 1985), 253) and restricted understanding by ‘flattening out’ multiplicity and multi-directionality (e.g. J. Fabian, Time and the Other (New York 1983)). The new media for visualising the world, then, have the potential to revolutionise not only approaches to space in Herodotus but also our own ways of conceptualising space, just as the new prose medium in which Herodotus was working must have influenced the spatial thinking of his contemporaries. See further §4 below. 17 This is one of the primary reasons behind the book by E. Irwin and E. Greenwood (ed.), Reading Herodotus. A Study of the Logoi in Book 5 of Herodotus’ Histories (Cambridge 2007). ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 7 πρόσουροι Κίλικες, κατήκοντες ἐπὶ θάλασσαν τήνδε, ἐν τῇ ἥδε Κύπρος νῆσος κέεται... [7] ἔχεται δὲ τούτων γῆ ἥδε Κισσίη, ἐν τῇ δὴ παρὰ ποταμὸν τόνδε Χοάσπην κείμενα ἐστὶ τὰ Σοῦσα ταῦτα, ἔνθα βασιλεύς τε μέγας δίαιταν ποιέεται, καὶ τῶν χρημάτων οἱ θησαυροὶ ἐνθαῦτα εἰσί... ” Being Spartan, Cleomenes needs time to process so many words: thus he invites Aristagoras to return three days later to reconvene the meeting. When that moment arrives, Cleomenes asks simply how many days’ journey it was from the Ionian Sea to the king (5.50.1). At this point Herodotus observes a critical error on the part of the man from Miletus: he tells the truth (!). It was, Aristagoras concedes, a three months’ journey inland (5.50.2). The consequences of such a reckless action immediately come to light, for, in response to the answer of ‘thirty days’, Cleomenes bade his Milesian guest depart Sparta before sunset (5.50.3). The key issues that we take from this story for our analysis are as follows. First, Herodotus presents a world not only built on networks but also interconnected and decentred: there is little sense of a Greek-barbarian polarity at play here, other than in the rhetoric that Aristagoras employs. This networked world is evident not only from the string of places which Aristagoras mentions, as he hops from one place to another in his attempt to demonstrate to Cleomenes the ease by which conquest would be accomplished; it is evidenced also from the very fact that a tyrant of Miletus goes to Sparta in order to seek an alliance in the first place. Second, just as the story lays emphasis on the role of human agency in the production of networks, so the importance of focalisation—how a particular individual or group sees—comes to prominence in the conceptualisation of space. This idea raises the crucial prospect that different peoples within the narrative conceive of space in alternative ways, a point that Herodotus wonderfully brings out in Cleomenes’ (delayed) response to Aristagoras’ proposal: the arguments for foreign adventure of the worldly Aristagoras (tyrant of a city at the hub of trade routes on the margins of the Persian empire) fall on the deaf ears of the king of a land-bound people (whose very territory, the Peloponnesos, signifies an ‘island’), who finds a ‘thirty day’ journey from the Ionian Sea beyond his ken. 18 As Chris Pelling suggestively puts it, ‘If Aristagoras were a website, he would be full of links’ (C.B.R. Pelling, ‘Aristagoras (5.49-55, 97)’, in Irwin and Greenwood (n.17), 179-201, quotation on p.179). Pelling here is talking about Aristagoras’ dense textual connections, whose personal history crops up at different points in the narrative of book 5. But the metaphor works rather nicely too for pointing to Aristagoras’ worldliness, which is part of that same picture. 19 In fact, just prior to this episode Herodotus has already recounted Aristagoras’ journey to Sardis, where he makes a similar plea for an alliance, only on that occasion it was the invasion of Greece that he was pedalling to a Persian satrap—which further complicates a Greek-barbarian polarity. By juxtaposing the two episodes, Herodotus brings out the picture of an interconnected world. According to Paola Ceccarelli (in an unpublished paper, ‘The Islands of the Aegean: Europe or Asia? Conceptualising the Aegean Space’), Aristagoras’ method of stringing together of places in both of his road maps to war exploits pre-existing network ties—or at least network thinking. She also posits that Herodotus is drawing on models of the world, possibly produced in Miletus by figures such as Anaximander or Hecateus, wherein Sardis and Sparta act as the two poles around which the (Mediterranean) world revolves. 20 E.g. I. de Jong, Narrators and Focalizers in Herodotus (Amsterdam 1987). ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 8 That answer too reveals a third important idea, that of space not as an abstract notion but as a phenomenon which is experienced. This is important because the ‘bronze tablet’ that Aristagoras brings with him, which marks an attempt to replicate the topography of the world in an abstract form, is then exploited by Aristagoras precisely on the basis of that abstraction: he uses the fact of the map’s small size in order to collapse the distance between places, so that, by repeatedly pointing to them, he creates the impression that conquest was going to be that easy. In this context it is important to note too that the features on the map solely depict seas and rivers—waterways that enable movement—thereby lending it a rhetorical force in an argument about conquest. In Aristagoras’ hands, then, the map becomes a tool of persuasion. Yet, this kind of abstract representation ultimately fails when confronted by the ‘reality’ (τὸ ἐόν, 5.50.2) of Cleomenes’ question, ‘just how far is it from the Ionian Sea to the King (i.e. Susa)?’, which leads to Aristagoras’ immediate expulsion from the king’s company. Significantly, Herodotus follows up this story with his own account of the places that Aristagoras covers, ostensibly in order to bear out the accuracy of Aristagoras’ (foolishly) true answer of thirty days, but more importantly in a manner which conceptualises space as being ‘hodological’: that is to say, as a route, a journey, rather than as some kind of visual abstraction of topography etched on a map. This in turn raises the issue of medium, as Herodotus glosses Aristagoras’ visual display of space with his own discursive representation of that space over the course of two whole chapters (5.52-4). Indeed, we find this example a timely 21 Herodotus makes this process explicit with the aside, ‘he said this showing the map of the world which he had brought engraved on the tablet’ (δεικνὺς δὲ ἔλεγε ταῦτα ἐς τῆς γῆς τὴν περίοδον, τὴν ἐφέρετο ἐν τῷ πίνακι ἐντετμημένην, 5.49). For his part, Aristagoras uses deictics throughout, as, for example: ‘next to the Ionians here are the Lydians’ (Ἰώνων μὲν τῶνδε οἵδε Λυδοί); ‘next are the Phrygians here’ (οἵδε ἔχονται Φρύγες); ‘here the island of Cyprus lies’ (ἥδε Κύπρος νῆσος κέεται); and so on. 22 On the rhetorical value of Aristagoras’ map, see A.C. Purves, Telling Space: Topography, Time ad Narrative from Homer to Xenophon (University of Pennsylvania, Ph.D., 2002), 106-17. 23 As Purves (n.22), 111 perceptively points out, ‘Herodotus never offers us the direct, simultaneous, and all-encompassing view’ of the map (or, indeed, of history). Instead, ‘Cleomenes’ question causes the map’s deceptively “easy” illusion of scale to collapse’ (112), while Aristagoras’ reply ‘swiftly undoes the illusion of the 1:1 scale which [his] combination of image and narrative had suggested’ (112-3). 24 ‘Unlike the mapped version, [Herodotus’] territory is marked by regular borders (πυλαί (gates), ποταμοί (rivers) and οὔρα (boundaries) predominate), and highly specific, neutral distances from place to place... [M]easurement through time acts as the key with which to unlock the dazzling, instantaneous effect of... Aristagoras’ marvellous [display]’: Purves (n.22), 116-7. Purves goes on to describe Herodotus’ narrative of that space as ‘hodological’, following the traveller’s experience and limited in view, in contrast to the ‘objective’, all-seeing cartographical representation of Aristagoras (117-21). Cf. P. Janni, La Mappa e il Periplo. Cartografia antica e spazio odologico (Marcerata 1984); A.C. Bertrand, ‘Stumbling through Gaul: maps, intelligence, and Caesar’s Bellum Gallicum’, Ancient History Bulletin 11 (1997), 107-22. For Romm, Herodotus is a key figure in the shift towards informant-led documentation of space: see J.S. Romm, The Edges of the Earth in Ancient Thought: Geography, Exploration, and Fiction (Princeton 1994), 34-44. 25 As recent scholars have pointed out, until well into the fifth century BC there were very few written texts and little prose literature to speak of: R. Thomas, Literacy and Orality in Ancient Greece (Cambridge 1992); Herodotus’ Histories stand at the cutting edge of research and the development of a new medium, prose literature: E. Bakker, ‘The making of history: Herodotus’ ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 9 warning for our own enterprise, as we try to represent Herodotus’ world using modern GIS mapping tools. This example has been used to draw attention to a number of key themes for thinking about space in Herodotus Histories that we hope to address—namely, the types of networks present and their interpretation, the influence of human agency and focalisation, the idea of space as something experienced and lived in, and the role of the medium in the representation of space—and to emphasise that close textual reading underpins our use of ICT throughout. The rest of the paper will set out in more detail that methodology and the resources that we have employed, paying particular attention to the decisions that we have made and the problems that we have encountered, in the hope that our project can contribute not only to offering a more complex picture of space in Herodotus but also to establishing a model for future digital projects in the Humanities which deal with large textbased corpora. Our methodology takes the form of four stages (Figure 2): 1. the digital markup of Herodotus’ text; 2. the compilation of a spatial database; 3. the production of basic GIS maps, as well as GoogleEarth and Timeline maps, using the database; and, finally, 4. the production and analysis of automated network maps. 1. Digital markup of Herodotus’ text Our first task, to obtain a digital copy of Herodotus’ text, was accomplished with the minimum of fuss due to the text being available at the on-line classical library, Perseus. Moreover, because Perseus releases texts under the Creative Commons Attribution Non-Commercial Share Alike license, we were free to use and adapt that text—in this case the 1920 Loeb translation of A.D. Godley—in whatever way we wanted. Historiēs Apodexis,’ in E. Bakker, I. de Jong and H. van Wees (ed.), Brill’s Companion to Herodotus (Leiden 2002), 3-32; S.D. Goldhill, The Invention of Prose (Oxford 2002). 26 Geographical Information Systems refers to the use of digital media in the representation and analysis of geographic data. See http://en.wikipedia.org/wiki/GIS. 27 One of the more intangible questions that this project raises is the extent to which writing down space changes its conceptualisation, or, to put that another way, how the form of medium influences the ways in which the data are received—a concern too from the perspective of contemporary forms of mapping: see n.11 above. Herodotus is particularly important testimony for the different ways in which space may be conceived of and represented because of his narrative strategy of combining ‘stories’ (logoi): his keenness to record the views of historical agents produces a series of micro-narratives that potentially destabilise any one over-arching conception of space. Indeed, by representing space partly through the eyes of his historical agents as he travels around the Mediterranean, Herodotus offers important new stimuli for our own explorations into space as structured by human experience (Y.-F. Tuan, ‘Literature and geography: implications for geographical research,’ in D. Ley and M.S. Samuels (ed.), Humanistic Geography (London 1978), 194-206) or by flow (D. Rodowick, Gilles Deleuze’s Time Machine (Durham NC 1997)). 28 At http://www.perseus.tufts.edu. 29 For more information, go to http://en.wikipedia.org/wiki/Creative_commons_license. ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 10 We further benefited from the fact that the placenames or toponyms—the data which we were interested in extracting from Herodotus—had already been ‘tagged’ in the English mark-up of the text. That is to say, the XML of the English translation of Herodotus, which basically describes the underlying structure of the text that we see on our computer screens, had been written (or ‘marked up’) in such a way as to identify and isolate all the geographical concepts that Herodotus mentions (Figure 3). This had been done using a standard grammar known as TEI, which sets a common schema for the tagging of data in large textbased corpora. Using a TEI-marked-up text in turn allowed us to manipulate and represent the data in different ways. While the capture of the digital text from Perseus gave our project a welcome initial boost, a number of issues resulting from that inheritance were raised, which have had to be overcome before the data could be properly stored in a database and utilised. These may be broadly classified into two groups, procedural conversion and data cleaning. In both cases below we briefly outline the problem and our response to it. i) Procedural conversion • The documents (Herodotus’ Histories in both English and Greek) obtained from Perseus came marked-up in TEI version P4, which represents the version prior to the current (as of 2009) internationally recognised standard (P5). Therefore, both documents had to be converted from TEI P4 to TEI P5, a process which was facilitated by using an automated conversion tool developed by Sebastian Rahtz of OUCS. This has been largely successful with the exception of the ‘reg’ attribute, which was no longer supported on this platform. While this has no effect on processing the documents, it still requires rectifying before the documents can be resubmitted to Perseus for reuse in their own text (as per the Creative Commons license). • The Greek text was transformed from Betacode to Unicode using Hugh Cayless’s Transcoder tool to simplify reading and enhance compatibility. • The decision was made early on in the project to use the English version of Herodotus’ Histories for investigating spatial data, primarily on the basis of practical concerns. Using the Greek text would have severely limited the potential dissemination impact of the project; but, more importantly, it would also have meant going through the text and tagging the toponyms ‘by hand’, when they were already widely available to us from Perseus using the English 30 XML = extensible mark-up language: http://en.wikipedia.org/wiki/XML. 31 According to its website, http://www.tei-c.org/index.xml, ‘the Text Encoding Initiative (TEI) is a consortium which collectively develops and maintains a standard for the representation of texts in digital form. Its chief deliverable is a set of Guidelines which specify encoding methods for machine-readable texts, chiefly in the humanities, social sciences and linguistics.’ 32 OUCS = Oxford University Computing Services. The process describing this conversion, along with a link to the relevant stylesheet, is laid out at http://www.tei-c.org/Guidelines/P5/migrate.xml. 33 For the reference, see: http://epidoc.sourceforge.net/resources.shtml. ELTON BARKER, MAPPING AN ANCIENT HISTORIAN IN A DIGITAL AGE 11 version. Bearing in mind the scholarly value of using the Greek text, however, we devised a process by which each section of the text in both documents was assigned a unique identifier in order to draw an association between the two versions. This has meant that, while we are using the English text to extract spatial data, at every point in the process the Greek text is available for scrutiny alongside it. • The English version of the Loeb text of Herodotus used by Perseus came with footnotes, which were also marked up. All footnotes were stripped out in order to prevent contamination by modern toponyms within them. These ‘reference free’ documents form the basis for the analysis.