第一编　丝路综论

Mapping the Multiple Dimensions of Hellenistic Civilization and the Silk Road:Prospects and Methods

Richard Talbert

（University of North Carolina, USA）

Abstract: This paper‒by an expert on cartography broadly conceived rather than on the Silk Road specifically‒identifies maps compiled to illustrate and link the project’s collaborative research findings as a vital missing dimension in its plans to date. Reasons why successive attempts to produce a definitive atlas of the Greek and Roman world failed to encompass the Silk Road and China are outlined. Brief explanation follows of why a longer-range vision is called for now, one that draws together multiple perspectives which cartography can enhance. Scholarly interest in the Silk Road has intensified remarkably (especially from China itself, as in the case of the present project in particular), while digital technology has attained a robustness which now makes it practical to create a seamless, layered map spanning a vast spread of territory at more than a tiny scale‒a basic requirement for this project that no print product can meet adequately. Attention is drawn to the design of the user-friendly digital mapping tool Antiquity-A-La-Carte developed by UNC’s Ancient World Mapping Center [https://www.unc.edu] and made available free of charge together with Map Tiles. In conclusion, pointers are offered to different ways in which Antiquity-A-La-Carte offers the versatility to meet the rich variety of mapping needs that can be expected to emerge from the project’s wide-ranging researches.

It was most kind of Prof. Yang Juping to surprise me only two months ago with an invitation to this international symposium Hellenistic Legacies on the Silk Road, and I am very happy to be participating in it. Even so, you can understand that my first thought was to decline the invitation, because I make no claim to be an expert on the Silk Road. However, the more I learned about the national key project of which the symposium forms an important part, the more I realized that there was the prospect of my being able to make a useful contribution to one dimension where I do have some expertise and resources, namely the provision of maps. In the Research Proposal document shown to me, nowhere do I see reference to the potential value of maps to illustrate and reinforce the multi-volume text and accompanying images planned for publication. This paper, therefore, has three related aims: to highlight the usefulness of maps to many of the approaches being taken for the project; to demonstrate the proven range and versatility of maps that can be created digitally for it; and to offer the help of the Ancient World Mapping Center for the production of such maps.

However, before focusing specifically on these three aims, I should first provide some background context which may or may not be familiar to you already, and then make a humble apology. In Europe during the 19th century scholars were very concerned to map the ancient, classical world of Greek and Roman civilization, and they struggled tirelessly to overcome the formidable difficulties. The work was inevitably slow and expensive, and the technology of map production was similarly challenging. Despite the development of lithography, the conviction persisted that for sharpest presentation a map should be laboriously engraved on copper-plate; and then, once printed, it was colored by hand. Predictably enough, more than one ambitious scheme for a definitive classical atlas was never carried through to completion. In fact, after the 1870s no such classical atlas was actually completed.

During the 20th century, understanding of the earth’s physical landscape greatly improved (thanks especially to aerial photography);further exploration and archaeology kept constantly expanding our knowledge of antiquity; and the technology of map production was significantly advanced by the introduction of a film-based method‒even though this too (like the earlier methods) lacked versatility, as well as being laborious in many respects, and expensive. At the end of the 1920s an international project was established to map the Roman empire, with the intention that each modern country within the empire’s former territory would map its own part of it. But the implementation of this attractive idea in principle increasingly proved to be a failure in practice, and no viable project to make a new definitive atlas of the Greek and Roman world was launched until as late, and as recently, as 1990.

This was a North American initiative, which I was commissioned to lead. What I designed‒a folio volume with 99 maps featuring landscape in particular‒was eventually published by the Princeton University Press as the Barrington Atlas of the Greek and Roman World in 2000, with the devoted commitment of literally hundreds of scholars, cartographers and supporters. This volume and the accompanying Map-by-Map Directory of references and bibliography (1,400 pages in print) are inevitably expensive items; but today the maps are also available as an App for iPad costing only US$20, and the Directory can be consulted online at no charge. A new book, Challenges of Mapping the Classical World (published by Routledge in 2019), includes several of my writings‒some previously published, others not‒about this Barrington Atlas initiative, before, during and after it.

‘What coverage does the Barrington Atlas offer for China and the Silk Road?’, you may fairly ask. And I have to admit in reply, apologetically,‘None’. The decision not to extend the coverage as far East as this was a conscious one, to be sure. As I mention in my Introduction to the Atlas, during the 1990s I was impressed to learn from the German team working on the Neue Pauly or New Pauly classical encyclopedia that they did intend to include China during the Roman period in the maps they were planning. In fact, however, even the entire atlas that New Pauly eventually issued in 2007 does not do that. In determining the scope of the Barrington Atlas, I concluded reluctantly that there was already so much territory elsewhere that had to be covered (even if at no more than the very small scale of 1:5,000,000), and so much catching up to do. To accommodate China(along with more of Africa also) would just be too much of a stretch.

By now, however, almost 30 years after the Barrington Atlas was planned, there are sound reasons to overturn the decision not to include China within the scope of a definitive classical atlas. The best of these reasons is the marked increase in interest that scholars have been showing in interchange of all kinds between China and the West along the Silk Road, as well as across the Indian Ocean. Needless to say, the national key project Hellenistic Civilization and the Silk Road is advancing and enlarging that interest in important, rewarding ways; and, very notably, it is doing so primarily from China. However, amid this welcome rise in the level of interest and awareness by scholars, one component that ought to be basic which does not seem to have risen commensurately in quality is the provision of maps. I say this with due caution, because I am relying upon no more than impression. If anyone can point us to much better maps than I happen to have found, I shall of course be delighted to know of them and to learn from them. Certainly, the quality of maps found in two recent books by scholars writing in English is well below the quality of the text chapters:Peter Frankopan, The Silk Roads: A New History of the World (2016) [a book with few maps, and no list of them in the Table of Contents]; Nicola Di Cosmo and Michael Maas (eds.), Empires and Exchanges in Eurasian Late Antiquity: Rome. China, Iran, and the Steppe, ca. 250-750 (2018).

Naturally, I have wondered whether a project already exists which is working to fulfill the need I perceive, but again my sense‒stemming from only limited inquiries, I do acknowledge‒is not. To be sure, the freely accessible database of the China Historical Geographic Information Systems project (with its long timeframe, 222 BCE to 1911 CE) has much to contribute for China. But relevant data from this project has not yet been imported to the equivalent database for classical antiquity, Pleiades.stoa.org, and at present (I’m told) there is no plan to do this. The reason(please understand) is not lack of goodwill or of vision on either side, but simply that each feels it necessary to meet other pressing concerns before embarking on such a major exercise. I should explain that Pleiades‒also online and freely accessible‒was begun shortly after the publication of the Barrington Atlas (in 2000) at the Ancient World Mapping Center in Chapel Hill, which was established at the same time. The management of Pleiades subsequently migrated to New York University, but it maintains its vital original purpose of updating, improving and extending the Atlas data. When it is ready (soon perhaps?), a resource very likely to be useful for establishing Chinese equivalent-forms for Greek and Latin names is the dataset being compiled by the team preparing the Chinese translation of the Cambridge Ancient History (second edition, with the timeframe extended to 600 CE).

The maps made for Frankopan and Di Cosmo and Maas obviously suffer, like many others, from having their frame-size limited to what can fit on the pages of their printed books, as well as from a ban against using color because of the substantial increase in costs then required. Certainly, if color can be introduced, it provides an invaluable enhancement for a printed map, in particular for rendering physical landscape meaningfully, something which it is so instructive to do for central Asia. There can be no mistaking that Frankopan’s map fails to convey any sense of the challenging terrain here unfortunately.

Even though the Hellenistic Civilization and the Silk Road project intends to issue printed volumes (which certainly could and should include maps), I would urge that its provision of maps not be limited to printed ones. Above all, what these just can never do‒because page-sizes simply are not big enough‒is to present vast expanses of territory, such as the project covers, at useful scales. During recent years, however, digital technology has become robust enough to meet that goal very efficiently‒a remarkable capacity, and unthinkable when the Barrington Atlas was planned (at a time when there was still no digital cartography; this only became a practical method from the mid-1990s). The Ancient World Mapping Center has in fact developed a digital tool which I regard as ideal for the project’s purpose, called Antiquity-A-La-Carte. It is available free;all its components are standard (so there are no special, custom features to incorporate); and all the materials utilized are open-source (so copyright restrictions are not an issue).

Antiquity-A-La-Carte provides a geo-referenced map base, where the modern physical landscape is returned‒so far as is feasible‒to how it is likely to have been in antiquity. The scale at which it may be displayed is of course variable, with a zoom possible up to about 1:50,000‒which is a large scale, after all, and normally more than adequate except for city-plans or detailed coverage of a small region. The landscape can be edited both by removing features and adding them, as well as altering them. Lines and polygons can be drawn, colors chosen and changed, distances on the map calculated. The ancient landscape can be removed entirely, and today’s substituted, in either Open Street Map or satellite rendering. An accompanying database is available, from which place-names may be selected; when that is done, the name will automatically be sited correctly on the map‒or, if required, it can be shifted. New names and sites can be introduced. So, too, distinctive name-formats can be chosen, as well as type-styles (including Greek and other alphabets, and also Chinese), font-sizes, and symbols. The number of layers that can be created (and then switched on or off, as the user chooses) is limitless‒so there can be one for routes, another for coin hoards, another for grottoes with Buddhist art, one marking names in Chinese, another marking them in, say, English, and so on and so on. A marvelous further benefit of this layering is the ability to differentiate by time-period. Of course, too, even once such a digital map is reckoned to be finished, and has been made accessible to users, it can readily be corrected, modified and added to later.

Altogether, the versatility of what can be created using Antiquity-A-La-Carte is remarkable. I do want to stress in particular how this tool can permit a user to pan seamlessly the entire way from China to, say, Rome. This said, I must warn you apologetically that right now the coverage of Antiquity-A-La-Carte itself hardly reaches much further East than the Ganges delta; but the coverage can be extended, and the foundation landscape base for China is already available in the associated Map Tiles tool (which is also free). You can see an excellent demonstration of what can be achieved over a vast expanse if you visit the digital map of this type made by the Ancient World Mapping Center to accompany the English translation by Duane Roller of the full 17 books of Strabo’s Geography published by Cambridge University Press in 2014. This does not have the quantity of layers that I would envisage for a comparable Hellenistic Civilization and the Silk Road map, but it amply illustrates the potential for them. In scope, the map stretches from Ireland to the Ganges delta and all across North Africa. It marks around 3,000 names‒densely clustered in such regions as Western Asia Minor and the Nile Valley‒and provides a link to the Pleiades entry for each name. This map is accessible free, but there is a bonus if you buy the electronic version of the translation: there, in the text, a link is embedded in each mention of a name marked on the map, so that when you click on a name you are taken automatically to its location on the map. At the Mapping Center we have begun a matching map to accompany the English translation of the geographical part of Pliny the Elder’s Natural History (Books 2 to 6); this translation by Brian Turner and me is well advanced, and is also due for publication by Cambridge University Press (2022). The scope of the Pliny map is as extensive as that for Strabo, and the number of names to be marked even greater.

Let me conclude by once again encouraging you to develop for Hellenistic Civilization and the Silk Road the cartographic dimension that I am convinced will demonstrate links and contrasts in your research findings and enhance their value in multiple other ways, both for you and for your intended audience. To achieve this goal successfully will require plenty of work‒no doubt about that‒as well as sustained collaboration between project members. But, I assure you, the necessary materials and tools for the map-making itself are available, they are ‘user friendly’, and it is entirely practical for you to use them. It is not necessary to call upon the Ancient World Mapping Center to make what I envisage. As has already been said, and I gladly repeat, the Center stands ready to help and to advise; and its website offers specifications for such stages as (for instance) how to organize data and names as you prepare to compile a map. But there is no cause for the Center to be more involved than this. Your team should feel empowered to make the maps for its Hellenistic Civilization and the Silk Road.(1)

(1)Since this paper was delivered in 2018, developments beyond the control of the Ancient World Mapping Center have required changes to its digital toolkit. These are still ongoing. To check current resources, visit https://awmc.unc.edu.