The ancient primacy of the Greeks and Romans in the educational system of the West was still flourishing, in many elite educational establishments, in the 1950s, and indeed in the 1960s and beyond. It was a special and privileged position, but of a very curious kind. In literature, Greece and Rome provided “the classics,” and in some sense set the standard by which later writers were judged; the brightest boys and, latterly, the brightest girls were expected to study them.

That did not imply, as perhaps might have been expected, a keen interest in the lives of the ancients, their world, its history and environment. What it did mean was scrupulous attention to the grammar of two highly complex languages, and meticulous imitation of the styles in which certain ancient writers wrote. The really serious business of the week, at a traditional school, was the translation of pieces of English poetry into Greek iambic verse, in the manner (we fondly hoped) of Sophocles, or into Latin verse, in the manner of Virgil (hexameters), or of Ovid (elegiac couplets).

The rules governing those meters are so exigent that at first the exercise feels like trying to knit with barbed wire. By contrast with this most arduous activity, such subjects as history or the visual arts were regarded as soft options, not difficult enough, not worthy to be taken with full seriousness. As for geography, the etiquette in the sixth (highest) form at my own school was to reply to our keen young ancient history teacher, when he asked where a place actually was, with some such carefully Philistine formula as “Oh, Sir, it’s about halfway down on the left-hand side.”

The change has been complete. No longer are eighteen-year-olds laboring to turn Shakespeare’s sonnets into the manner of Ovid, or the oratory of Macaulay and Lincoln into the idiom of Demosthenes. There has been a price to pay, and we see it in the uncouthness of much modern academic writing, and in the low standard of most contemporary political rhetoric. No longer have our leaders, or even our critics, been through that exacting stylistic training, which insisted that every word had to be scrupulously weighed and savored. Instead, what is found most interesting about the ancient world tends to be the visual arts and economic and social history. Religion, trade, medicine, town planning, women, slaves: these are the sorts of subjects which nowadays dominate.

It has followed that students and teachers have felt the need for new reference works that are more helpful in these matters than the old stylistic manuals. That need has become especially acute in the case of reliable and accurate maps of the ancient world. In the last hundred years it has been very hard to get hold of any. The labor involved in producing good maps is very great, the expense very great also. Excavation and aerial surveys have enormously increased our knowledge. Some ambitious plans have come to nothing at all, like the Italian Grande Atlante del Mondo Antico, of which in the end nothing was published. Others have been crucially hampered by all sorts of problems. The publication of separate fascicles has tended to be patchy, with different sections appearing in different formats or (often) not at all. Whole enterprises have been damagingly marked by nationalism: by refusals to cooperate with neighboring states, and by quarrels over definitions and frontiers. What has finally been published has usually cost far too much for anyone but an institution or a library to buy.

Such was the somber picture, and such the most likely outcome, when, in 1980, the American Philological Association turned its attention to the subject and pronounced,

We come, finally, to an area of extremely great importance, where the state of our tools is utterly disastrous, cartography. There is hardly anything more important to understanding ancient history than a clear conception of the terrain on which its events took place.

The statement went on to lament the lack of good new maps and the unavailability of the best of the old. It concluded:

A concerted attempt to produce a uniform series of maps which show both the topography—with all the sophistication of modern cartography—and the ancient toponyms—with the accumulated knowledge of classical scholarship—would be immensely valuable.

In 1988 an approach was made to Richard Talbert, of the University of North Carolina, an Ulster man, to head the project of producing an atlas which would satisfy these demands. Fortunately he proved to be a man of great resource and tireless energy; and in 2000 the completed work was published. The work cost some four and a half million dollars (“relatively modest,” says the editor disarmingly); there is a whole page listing the names of donors and supporters. In its published form it takes its name from a leading supporter, the Barrington Foundation. It recruited seventy-three compilers, answering to ten regional editors (rather charmingly called vicars, echoing the title vicarii of administrators in the late Roman Empire), with ninety-five reviewers to revise and criticize their work, and twenty-two cartographers to create the maps from it. The enterprise is international and consists of scholars on both sides of the Atlantic. It is important that the atlas has got away from the rule that the mapping of each region may be undertaken only by the modern national authority there. That rule has been a charter for the obstructive, and it has gravely hampered many recent cartographic ventures.


The declared aim is to “take in all regions for which penetration by Greeks or Romans can be documented.” That goes far beyond Greece and Rome, far indeed beyond the Roman Empire at its vastest. The geographical scale is such that it touches on the territory of ninety-one modern countries, from Scotland to Mozambique, from Portugal to China, and from Latvia to Sri Lanka. The gazetteer helpfully lists, for each named place, the modern nation-state in which it lies; the editor adds, wryly but also prudently,

Important Note: The sole purpose of this list is identification. It is not to be taken as making any kind of statement about the status of countries mentioned.

There is a quagmire waiting to engulf anyone rash enough to express a view about the borders of Israel, or the status of Macedonia, or the nationality of Kashmir, or any of dozens of questions of conflicting national claims and ethnic pride. The scale of the index can be seen from the fact that it lists some three thousand names of places beginning with the letter A; it ranges from “A(…),” in North Africa, by way of Zulm (Afghanistan) and Zunsweier (Germany), and Zvezdan (Yugoslavia), to the Zygritai and their capital, Zygris, in what is now Tunisia. There are also, of course, indexes of “unlocated toponyms” and of “false toponyms.”

Those categories open up interesting vistas. Clearly there will be no entry for “Atlantis.” It is delightful, though, to find “Thule” listed; Thule turns out to be the Shetland Islands, northwest of Scotland, but the identification is stigmatized with a scholarly question mark as less than certain. What of that other classical Shangri-la, the Garden of the Hesperides? That too is in, and without even a query: it was in what is now Libya, near the real place called Euesperides, “Best Western.” But of course we are not to suppose that the atlas means that in the Garden there were lovely nymphs cultivating the apples of immortal life. Part of the River Styx flows above- ground, and that duly appears on Map 58; but not the infernal part, where Charon, his eyes ablaze, ferries the trembling souls of the dead. And we look in vain for those other rivers of the underworld: Cocytus, the River of Lamentation, and Phlegethon, the River of Fire. They must be sought on maps of a different kind, along with the Elysian Fields. The Phlegraean Fields, on the other hand, are real, and volcanic; they will be found near Naples.

For some of the ninety-one countries, contact with Greeks and Romans was relatively slight. Still, even the bare place names of coastal settlements can be revealing. It is fascinating to observe, for instance, that on the west coast of India there is another, smaller Byzantium: perhaps the settlers were homesick, or perhaps they had a touch of folie de grandeur. There is also a place called Monoglosson Emporion, “Monoglots’ Market,” a reflection (surely) of ethnocentric impatience with foreigners who would not learn Greek. (One is reminded of Kingsley Amis on the French: “Why can’t they talk English, like everyone else?”) And right down in the south of India we know of a temple to Augustus. Was this a result of imperialism? Or was it, perhaps, a bit of judicious flattery of the Big Man back home?

Along the southern shore of the Black Sea in what is now Turkey, we find, touchingly, a tiny place with the name “Athens.” Like the Indian By-zantium, such a name recalls the secondary Birminghams, Bostons, and Toledos of the New World. On the same Turkish coast we can read the intentions of other settlers with vivid clarity. Some settlers piously put their new towns under the patronage of gods: we find Diospolis (“Zeusville”) and Poseidon and Heraclea. Some made hopeful claims for the future: there is Philokaleia, which means “Love of Beauty.” We may think of Concord or Philadelphia. It may be a local feature that caught their fancy: we find Kerasous and Pityous, which mean respectively “Cherry Trees” and “Pine Trees.” Less imaginatively, some settlers called their own new colony simply Naustathmos, which means “Ship Station,” (compare Anchorage in Alaska), and we find also Bathys Limen, which means “Deep Port,” and Psoron Limen, which means, all too poignantly, “Scabby Port.”


Ancient men tried to map their own world. By 500 BCE the Greeks had maps which represented not only the Aegean world but Egypt and Asia, too. These maps tended to be rounded, with the stream of ocean going all around the edge of a land mass divided up into neatly symmetrical continents. In the fifth century BCE Herodotus denied that the encircling ocean existed at all: it was just an invention of poets. But even so, he was not above arguing that the Hyperboreans, a mythical people dwelling at the back of the north wind, did not exist, “for if there are Hyperboreans, then there are Hypernotians in the south, too.” The great geographer and polymath Eratosthenes, who in the third century BCE measured the circumference of the earth with accuracy by trigonometrical reasoning from the length of shadows, tried to get away from symmetry in maps, but it kept coming back: the world, it was felt, ought to be symmetrical, and on medieval maps it generally is, with Jerusalem at the center. Maps of the world as the ancients themselves envisaged it are among the intriguing possibilities that the Barrington Atlas editors did not follow up.

Most maps were produced for definite practical purposes: as navigational guides for ships sailing along a stretch of coast, or to help tax collectors estimate and gather land tax, or, sometimes, for propaganda, like the famous map which Augustus’ friend Agrippa put up in the center of Rome, to illustrate the greatness of the Roman Empire. Mussolini did the same thing, for essentially the same reason.

We know the shape of places much more accurately than the ancients did, especially now that we can use aerial studies. The orthodox view in antiquity, for instance, accepted a miscalculation which made the Pyrenees run north–south, not east–west. That messed things up badly in the west. The Barrington Atlas has been able to use the Operational Navigational Charts produced by the US Defense Mapping Agency, with some input also from the British Directorate-General of Military Survey. The maps thus produced are created by digital means. It turned out providentially that the great advance which this new technology makes possible in cartography coincided with the eleven years that The Barrington Atlas took to produce. The creation of the maps involved six fonts of type, while “a few fonts had to be customized using Fontographer 4.13 software.”

The number of different scales used is small; that is a great convenience to the user. In general terms, the core of the classical world (the Mediterranean, Mare nostrum, Our Sea, as opposed to the outlying areas, deserts, etc.) is shown at 1:500,000, with the environs of the cities of Athens, Rome, and Constantinople shown at 1:150,000. Outlying places (Britain, Germany, Afghanistan, and so on) appear at a scale of 1:1,000,000; and some sparsely occupied areas appear at the scale of 1:5,000,000. The maps themselves are a pleasure to use. They are very large, most maps being in fact a double-page spread of folio size, and they are handsome as well as clear.

A difficult problem is that of representing the physical world, not as it is but as it was then. It is simple enough to think away the Corinth Canal, but other changes are harder to grasp. Sometimes it is possible to chart changes in a coastline, or in the erosion of a range of hills, with pretty fair confidence; at other times the matter is more one of guesswork, and on the map there appear dotted lines. That is a reminder of the fact that the ancient world was not one of stagnation but one of change, and that the chronological range of this book, like its range in space, is very large. The atlas covers the period between circa 1000 BCE and circa 640 CE. In that great spread of time, rivers changed their courses; towns that had been on the seacoast were left high and dry in the middle of the land; cities lost importance or went out of existence, and others grew up and flourished in their stead. That of course poses a problem.

It would be possible to attempt a representation of the more interesting areas (not the deserts of Bactria or the Sahara, or the mountains of Afghanistan) with a whole series of maps, showing their physical and human shape in successive periods. Particular maps can show, for instance, the sites occupied in Greece in the Mycenaean period, or the stages in the conquest of Italy by Rome. Such maps do exist. The much smaller atlas of N.G.L. Hammond, Atlas of the Greek and Roman World in Antiquity,* includes a good many. But on this scale they would make the atlas enormous; and in this book the device is used of underlining place names in five different colors, to indicate attested occupation specific to only one of the five periods: Archaic, Classical, Hellenistic, Roman, and Late Antique (the Mycenaean period is not covered, which one can’t help feeling is a pity, and on the map of Crete the places which were Minoan sites are not identified as such). The device works pretty well, but readers whose historical imagination needs more visual support would still be grateful for publications that met the need more explicitly.

The Barrington Atlas has consciously limited its goals. It does not contain detailed plans of the great cities. The map that includes Athens, for instance, though printed on the largest scale (1:150,000), can indicate the Acropolis only by a single letter, with no attempt to plot its constituent temples and shrines, and the scale is too small to show the Pnyx, the hill on which the popular assembly met. Some readers, and not only armchair strategists, will regret the absence of the sites of battles: Marathon and Salamis are in, but with no indication of why they are famous, and with none of the familiar rectangles with different shading to denote opposing armies.

The choice of features that are included is not wholly self-explanatory. That is not surprising; the atlas has a character of its own. The editors have made a determined and successful attempt to show the main roads that we know of, but there is no attempt to indicate sea routes, or to represent the depth of the sea at different levels. It is nice, but a little inconsequential, that the maps show lighthouses—the only acknowledgment of the importance of sea travel. Lighthouses were not numerous in the ancient world. The “physical features” listed include “Bog, Swamp, Marsh,” “Cataracts,” and “Intermittent Drainage,” but not forest areas, the most important of natural features and of natural resources; with the rather surprising exception that “uniquely the boundary lines of imperial forest holdings in Lebanon, demarcated by inscribed markers during Hadrian’s reign, are shown on Maps 68 and 69.”

The atlas thus does not display much interest in timber, but it is interested in mines, and it has symbols for no fewer than twenty-five different mined materials, including such exotic ones as alabaster, amethyst, cinnabar, porphyry, and zinc. Professor Talbert makes it explicit in his introduction that

In whole or in part the Greek and Roman world could be mapped in a multitude of different ways…. The maps [of this atlas] should form the basis for branching out further in every direction.

It is now possible, in other words, to use these maps as a basis on which to create all sorts of special maps designed to illustrate the history, politics, economics, culture, and religion of the ancient world. The ball is passed to us. Let us see whether we can play with it the games that it deserves. It is not hard to see how these splendid maps can be the basis for maps of the different places at different periods, and of economic links and trade routes, and of the course of military campaigns, and of the patterns of the barbarian invasions, and of the distribution of different religious cults, and of the spread of Christianity, and of the diaspora of the Jews, and of the topography of myth and legend—where heroes came from, and where they went; where legends cluster, and where there is a comparative dearth. And so on. The prospect is exhilarating.

The editor proclaims the need for a continuing center for such cartographic studies, based in a university, to embody the claim of cartography and informative geographic science to be vitally important to the whole field of ancient studies, to develop new maps for all sorts of special purposes, and to ensure that the maps now completed are kept up to date and not regarded as an end, rather than a beginning, to the subject. The editor has expressed the hope that this will be based at the University of North Carolina.

The maps are of course by far the showiest and most attractive part of any atlas, and these, as I have already said, are very handsome. But they are only a part of the package presented by this most ambitious project. In addition to the map volume, there is a gazetteer, available as two substantial volumes and also as a CD-ROM, which for each map gives an alphabetical list of the places marked, and for each name gives its modern name, the modern country in which it now lies, the period of the ancient world in which it occurs, and—an invaluable resource, which will be richly exploited for years to come—a guide to modern scholarship bearing on the place.

The appearance of modern names is welcome, although they are given only within the entries, which are listed under the ancient names, where they are known. It is a slight drawback to the gazetteer that modern names of places are not given separate entries. Before the riches of the gazetteer can be tapped it is thus necessary for the nonexpert, who wants to look up a place while knowing only its modern name, to discover the ancient one. That is all very well with Athenai or Roma, but often there is no way, if you don’t know, to guess. Of cities named after the Emperor Augustus, for instance, Augusta Taurinorum became Torino, while Augusta Praetoria became Aosta; in Switzerland Kaiser-Augst was derived from Caesar Augustus. A very clever user might perhaps guess that Confluentes, the flowing together of two rivers, was the ancient name of Koblenz. But more than cleverness is called for to identify Chester as Deva, or Baalbek as Heliopolis Colonia Iulia Augusta Felix, or Gubbio as Iguvium, or Wallsend (in Northumberland) as Segedunum. Yet Wallsend is a sort of Roman name: it stands at the eastern end of Hadrian’s Wall across Britain. As for Chester, an important place then and now, it is almost provoking that its absence from the gazetteer is underlined by the comparative insignificance of the names that do precede and follow its absent spot: Chessy-les-Mines (France) and Chesterton (UK).

That, however, is a trivial complaint. A blemish of the kind which a reviewer likes to point out, to show his own conscientiousness, is the absence of any mention of the famous site in Afghanistan called Ai Khanoum, excavated by the French between 1965 and 1978, which has attracted much interest as a textbook example, in so remote a spot, of a Hellenistic Greek city, complete with theater, gymnasium, temples, wealthy houses, and massive fortifications.

The main point to make is that The Barrington Atlas is a major contribution to scholarship, extensive in scale, reliable and up to date, and so laid out as to be really helpful to the user. Apart from the maps, what readers will find most valuable is the systematic references in the gazetteer to the modern scholarly literature on each place. It exists on CD-ROM; that should mean that it can readily be kept up to date. Now we need the excellent and selfless Princeton University Press to produce an updated second edition of its invaluable Princeton Encyclopedia of Classical Sites (1976), which gives the detailed site plans and full accounts of every city and shrine. Then we shall be really equipped to tackle the world of antiquity: the ancient world that the ancients did not see.

This Issue

April 26, 2001