CHAPTER 10

THE LATER ROMAN EMPIRE

Rome and China (Map 12)

In AD 97, the commander of the western region of the Chinese empire, Ban Chao, sent an envoy, one Gan (or Kan) Ying, to the west.¹ As early as the second century BC, the Chinese had become aware of Greek remnants in Baktria,² and in the following century the Chinese had been in contact with the Parthians on the Iranian plateau, learning about the lands farther west, including the notable state of Daqin, the westernmost in the world, which could take several months to reach.³ Many toponyms along the route are mentioned in Gan Ying's report, which are almost impossible to equate with known ones, except perhaps the capital of Daqin, Andu, which is remindful of Antioch (on-the-Orontes), where one might expect a traveler from the far east to reach the Mediterranean. The Chinese would have good reason to believe it was a great city: it had been the Seleukid capital for many years before becoming the capital of the Roman province of Syria, and it was where Indian envoys met Augustus in 20 BC.⁴

Gan Ying never made it to Daqin, although he knew something about the regions in the west beyond what he actually covered. The Parthians may have discouraged him from going by land, and the sea route was exceedingly long.⁵ Nevertheless this is the first documented example of someone from China attempting to reach the Mediterranean. The Chinese and the Mediterranean world had had faint knowledge of each other since perhaps the third century BC, but there had been no known previous attempt to make direct contact.⁶ Nevertheless, the epitome of Chinese material goods, silk, may have come to the Mediterranean as early as the first century BC and had become relatively common by the following century: as “Seric cloth” it was mentioned more than once by the author of the Periplous of the Erythraian Sea.⁷ Even as early as the time of Tiberius, silk had become enough of a symbol of ostentatious luxury—at least to that frugal emperor—that legislation was passed forbidding men to wear it, a stricture that the emperor Gaius Caligula violated.⁸

Map 12. The limits of ancient geographical knowledge.

At about the same time as Gan Ying's journey, a Macedonian named Maes Titianus attempted to determine the distance to China from the known world.⁹ As usual, details are sparse, but Maes' agents (he did not make the journey himself) set forth from a place called the Stone Tower (Lithinos Pyrgos), 26,280 stadia east of the Euphrates, north of the Imaos mountains and well east of Baktra.¹⁰ The name Stone Tower is common in this region, and even though none of the places so called can be equated exactly with the point of departure of Maes' expedition, the town of Tashkurgan (also meaning Stone Tower) on the western border of China is an intriguing possibility.¹¹ Stone Tower was obviously a flourishing trading emporium, where merchants from all directions met. Maes' agents eventually ended up among the Silk People, at a location another 24,000 stadia to the east and lying on the parallel of Rhodes. The distance from Stone Tower to the Silk People, essentially the same as from the Euphrates to Stone Tower, is presumably a calculation based on the reported seven months for the journey. Where the Silk People congregated cannot be determined—no place is named—but it may be at the contemporary Chinese capital of Luoyang.¹² The placement of the Silk People on the parallel of Rhodes probably means a member of Maes' expedition determined the latitude of their city, and the parallels of Rhodes and Luoyang are a mere 1.5° apart.

Thus merchants, not diplomats, sustained the contact between the Mediterranean and China. According to Chinese reports envoys from King An-tun reached China and the emperor Huan-ti in AD 166.¹³ An-tun is an Antonine emperor, presumably Marcus Aurelius or Lucius Verus, but the failure of the embassy to be mentioned in Greco-Roman sources suggests that it was less official than its partipants implied—probably merchants who pretended to be official functionaries. Nevertheless the Chinese recorded it as the first direct contact between Rome and China, although they were not particularly impressed with the gifts that they received—ivory, rhinoceros horn, and tortoise shells—which they considered to be tribute.¹⁴ Despite the fact that no official relations were ever established between the two states, the merchants on the Silk Road continued to ply their trade, and brought knowledge of each civilization to the other. It is interesting that the alleged embassy from An-tun to the Chinese is at the very time of the first thorough and well-informed description of silk and silk worms in Greek literature.¹⁵

Marinos and Ptolemy

The final flourishing of the ancient geographical tradition is the Geographical Guide of Ptolemy of Alexandria, written toward the middle of the second century AD. It is an unusual and highly influential work, not truly a geographical treatise but one on cartography. There are certain parallels with the Against the “Geography” of Eratosthenes of Hipparchos from 300 years previously, since the Geographical Guide was also written by a mathematician and astronomer whose primary goal was to use his own disciplines to correct geographical data and the work of a predecessor. In Ptolemy's case this was Marinos of Tyre, one of the most elusive figures in the history of Greek geography, known only from the Geographical Guide. He wrote around AD 110, shortly after the Dacian campaign of the emperor Trajan.¹⁶ About a generation later, Ptolemy took Marinos' work and refined it, with the intent of providing instruction on how to create a map of the world.¹⁷ Ptolemy had already published his Mathematical Syntaxis (popularly known today by the Arabic form of its name, Almagest), which set forth the motions of the heavenly bodies and became the standard work on its topic until the Renaissance. It had a geographical component, since astronomical observations in various places had to be reconciled with one another, and this may have led Ptolemy to a broader work on the location of places in the oikoumene and to create (in theory) a map.¹⁸

The Geographical Guide located more places than any previous author. Ptolemy used Eratosthenes as his basic framework, but created a more extensive oikoumene than had been known, and was perfectly willing to acknowledge that many of his data were taken directly from Marinos.¹⁹ Marinos had access to a variety of information—Ptolemy praised him as a diligent researcher—and his work had already gone through several editions. Marinos seems to have been particularly sensitive to recent data provided by merchants, but in fact few sources of any type are named in the Geographical Guide: in addition to Marinos there are only 11 others, most of which are known nowhere else and are from the generation previous to Marinos (as they were not mentioned by Pliny). Thus they provided new information about the extremities of the inhabited world. Ptolemy and Marinos were not writing geographical works but cartographic manuals, and to them the history of toponyms was less important than determining their location. It is perhaps no accident that of all the luminaries in the history of geography, no one other than Hipparchos is cited in the Geographical Guide, the other mathematician and astronomer to venture into geography, whom Ptolemy rightly described as the only one to provide astronomical data for localities.²⁰ Like Hipparchos, Ptolemy realized that astronomy was the only way to determine an accurate position for places, yet he also fell into the same trap as his predecessor: there simply were not that many locations whose positions had been determined astronomically, and many figures were obtained by a conversion of stadia distances along a route into a longitude and latitude, probably using existing maps.²¹

Ptolemy wrote that 26 maps of the oikoumene were created. Whether this was actually done, or the Geographical Guide was merely an instruction manual, has long been debated.²² The maps in the extant manuscripts show little consistency and are no earlier than AD 1300, and some manuscripts have none. It seems improbable that Ptolemy actually included maps in his treatise, and in all likelihood the extant mapping tradition was developed by the Byzantine scholar, Maximos Planudes, in the late twelfth century, who claimed to have rediscovered the Geographical Guide. But it was certainly possible for any competent reader to create maps based on the data that Ptolemy provided, something that may have happened as early as the later Roman period.

The Geographical Guide consists of 8 books, and thus is second only to Strabo's Geography in length. The first and last are theoretical and technical material, with a discussion of topics such as the discipline of cartography, the greater validity of astronomical data rather than travel records, a critique of Marinos, and the actual instructions on how to make a map. Some of the material is descriptive, especially the information from recent explorers that extended the limits of the oikoumene in Africa and east Asia. There is also a list of parallels: Eratosthenes' eight have become 23, all defined by length of daylight and degrees of latitude, with practically no toponymic material. The middle six books of the Geographical Guide are the detailed lists, with over 8,000 toponyms (Eratosthenes had about 400) each accompanied by a latitude and longitude. Descriptively, the catalogue is sparse (the handful of sources named are all confined to Book 1), and the modern reader can be overwhelmed by the sheer quantity of names. Many of them are new. There are groupings of regional toponyms, with each section introduced by a catalogue of local ethnyms, and an occasional extraneous comment on ethnology, but these are rare.

The addition of data from the far east meant that the oikoumene had grown since the time of Eratosthenes,²³ and although Ptolemy reduced some of Marinos' figures in both the east–west and north–south directions, he nevertheless ended up with an inhabited world that extended from the Fortunate Islands (probably the Canaries) in the west, which marked his prime meridian, to the Silk Peoples, 180^o to the east or halfway around the globe, nearly twice the 70,000 stadia that was originally believed to be the east–west extent of the oikoumene. This had a major effect on Renaissance explorers, for these figures (and others supplied by Marco Polo) allowed Columbus to believe that the journey from the west side of the inhabited world around to the east was only about one-quarter of what it was,²⁴ and by 1498, at the time of his third voyage, he thought that he had reached Ptolemy's Golden (modern Malay) Peninsula.²⁵

There are many new toponyms in the Geographical Guide, especially at the extremities of the world. Their density can be astonishing, something that the modern reader can easily see by looking at the maps created from the text. There are dozens in Arabia and nearly 50 on Taprobane, regions hardly known a century before Ptolemy.²⁶ Yet in many other areas there are few that are new: parts of northwestern Europe hardly go beyond what was known in the Augustan period. Moreover, there are serious problems with the orientation of Scotland, which is pulled far to the east. Ever since the rejection of Pytheas' data—a process that began in the second century BC—there had been difficulties in considering the far northwest of the world, and Ptolemy's issues represent their culmination.²⁷ This is a contrast to the large amount of new data in the south and east of the oikoumene, reflecting the greater importance of trade with central Africa and the far east than with the Baltic. Ptolemy was the first in many years to enclose the Caspian Sea—although its status continued to be debated until medieval times—and there are some new rivers that flow into its north end, the Rha, Rhymnos, and Daix (the Rha is the Volga and the Daix is perhaps the Ural), which left little room for a connection to the Ocean.²⁸ On the other hand, Ptolemy continued to believe that the Iaxartes flowed into the Caspian and he had no knowledge of the Aral Sea, so he remained to some extent still dependent on the conception of this region dating back to the time of Alexander and Patrokles.

Yet the most peculiar aspect of Ptolemy's topography was his idea that the Indian Ocean was an enclosed sea. This was a belief that perhaps originated with Hipparchos and was considered plausible by Polybios.²⁹ Hipparchos used Seleukos of Seleukeia and his work on the tides as his authority, but it seems to have been more of a theoretical suggestion than any actual belief—and contradicted the many tales of the circumnavigation of Africa—until Ptolemy placed it in the mainstream of geographical thought. He even suggested that the Aithiopians were not far from Sinai and Kattigara (east of India). With the authority of Ptolemy, the idea of an enclosed Indian Ocean remained a hindrance to exploration until the time of Vasco da Gama.³⁰

Beyond the Oikoumene

Ptolemy established the complete extent of the inhabited world as it was known in antiquity. In fact, he made it too large, because of the repeated extensions due to new data received from the trade network of the Roman empire. He was not a geographical theorist and showed no interest in whether there was anything beyond what was already known, and nevertheless any thought of what lay outside the oikoumene was the concern of allegorists, not geographers, either practical or theoretical. Yet even Ptolemy's expanded inhabited world was still realized to be only a small part of the sphere of the earth. Cicero, who dabbled in geography and even planned to write a treatise on the topic,³¹ had perhaps the best statement on the matter, with Scipio Africanus in the famous dream sequence of the Republic describing the limits of Roman power:

the entire land that you hold is narrow vertically and wider in width, but is a small island surrounded by that sea which you on earth call the Atlantic, the Great, or the Ocean. You can now see how small it is, despite having such a name. Do you believe that your fame, or any of us, could go beyond the settled and known lands, by crossing the Caucasus or swimming across the Ganges? Who among those of the rising or setting sun, or the extreme north or south, will ever hear your name? Cut all these out, and you will not fail to see over what a narrow region you are so eager to spread your glory.³²

To be sure, this is not geography, yet it shows an awareness of the minuteness of the oikoumene compared to the entire sphere of the earth, and thus the insignificance of the known world. In fact, there could be people on the opposite side of the earth, who would be upside down relative to those in the known world, and great expeditions would be needed to reach those lands. There might even be people on the moon, an idea perhaps originating as early as Xenophanes of Kolophon. This raised the question of the position of human beings both on the earth and in the entire cosmos.³³ But this was all speculation, even if some of it were later shown to be true, and there is no evidence that the Greeks or Romans had any specific knowledge of a world beyond the limits defined by Ptolemy.

The end of the world in the north was, of course, Thule, which Ptolemy called an island “at the end of the known sea,” implying that there might be something more beyond (Map 12).³⁴ He knew the name Skandiai, an island in the western Baltic, but had no awareness of the Scandinavian peninsula.³⁵ Ptolemy called the Baltic the Sarmatian Ocean, and beyond it was “unknown land.”³⁶ This shows that the exact position of the Ocean was not really understood north and east of the Baltic—in fact all the way to Kattigara—and that there was more land to the northeast, a contrast to the belief of Alexander and the Hellenistic explorers of the region beyond the Caspian, who always thought that the Ocean was just out of sight. The author of the Periplous of the Erythraian Sea had hinted at more land to the north and northeast and knew that it was exceedingly wild.³⁷ Yet in antiquity there was no sense of what was beyond a line roughly from modern St. Petersburg to Beijing.

Nevertheless trade goods could go beyond the oikoumene (Map 12). A few items penetrated into southern Norway and Sweden in the first and second centuries AD, presumably via routes from the islands of Bornholm and Gotland (probably Ptolemy's “Skandiai”) up the east and west coasts of Scandinavia and into southwestern Finland.³⁸ Roman coins have also been found in Iceland as well as in extreme southern Africa and Madagascar.³⁹ Trade goods were discovered on the Mekong River in Vietnam at Oc-éo and P'ong Tuk,⁴⁰ and there is the curious incident of the Roman amphoras near Río de Janeiro.⁴¹ Yet portable objects travel far, and there is no evidence that these demonstrate any direct contact between the Romans and the far reaches of the world.

The End of Ancient Geography

After Ptolemy the creative era of ancient geography was virtually over, although earlier works were still being analyzed. An odd occurence was the emergence of the Geography of Strabo, which, it seems, had never been published after its author's death around AD 24, and which was unknown to Pliny, and probably to Marinos and Ptolemy, although the source citations in the Geographical Guide are so few that this is difficult to prove. Yet by the end of the century Strabo's treatise was known to scholars, since Athenaios quoted it twice, and from that time it was regularly cited.⁴²

The self-image of the Roman world meant that in later times there was an emphasis on map-making, the most obvious way of displaying the extent of Roman power, much more effective than any literary account.⁴³ Presumably the map of Agrippa, created at the end of the first century BC, continued to be on display in the Porticus Vispania in Rome, but the total lack of any evidence beyond the literary, as well as no knowledge whatsoever of the porticus itself⁴⁴ means that any detailed visual sense is almost impossible to obtain. In the first decade of the third century AD a public map of the city of Rome was created by the emperor Septimius Severus and attached to a wall in the Temple of Peace, which had recently been restored.⁴⁵ Fragments of this marble plan (the Forma Urbis Romae) have been recovered since the sixteenth century, and it is the subject of an on-going restoration. It was exceedingly large (60 by 42 feet), and, even though merely a town plan rather than a view of the world, it is the best preserved example of a map actually from antiquity.

The most famous extant world map from the Roman period is the Peutinger Map, which exists today in a parchment copy of the early thirteenth century (or slightly earlier) in the National Library in Vienna.⁴⁶ It was a long strip (now 11 separate sheets), 13 inches high and 22 feet long, the ultimate expression of an oikoumene longer than it was wide. It is based on an original from the pre-Christian era, and the recent suggestion that it might have been a display piece in Diocletian's palace in Spalatum (modern Spalato or Split in Croatia), and thus from the end of the third century AD, is compelling.⁴⁷ Even though the extant copy is a thousand years removed from that era, it nevertheless is a striking visual record of how the Romans conceived their world. The left (western) end is lost (the extant map begins with Gaul and the southeastern British coast), but from this point to India and Taprobane the surviving 11 sections lay out the world of Rome for all to see. Rome itself is exactly at the center (assuming three lost sections in the west), as one might expect. The unusual shape of the map was designed to show the east–west magnitude of the Roman world, presenting an oikoumene more than 20 times as long as it is high, and containing virtually no information about the extremities of north or south. It is also somewhat beholden to a longstanding tradition of depicting routes in straight lines, a geographical rule as early as Eratosthenes' meridians and parallels and still existing today.⁴⁸

The Peutinger Map is geography for political leaders, as Strabo had promoted:⁴⁹ only the important parts of the inhabited world are shown, a smaller oikoumene than that of Ptolemy, limited to those portions that would be important to people in power. Yet it remains the best way to visualize the world as the Roman elite saw it.

Christian Topography

The end of the ancient geographical tradition is marked by the Christianization of the Roman world, and the imposition of a new set of values. The Jewish antecedents of Christianity had long since made their way into Greek thought: Herodotos was perhaps the first Greek to write about the Jews.⁵⁰ In Strabo's Geography there is a lengthy discussion of Moses—probably derived from Poseidonios—who is depicted as an Egyptian priest who came to Judaea with his followers.⁵¹ However tangential this may be to Strabo's main theme—probably a reflection of his deep interest in cults—nevertheless the passage provides a connection between Jewish culture and Greco-Roman geographical scholarship, something that passed to the early Christians.

Some hints of the change toward a theological view of geography were put forth by Augustine in the early fifth century AD. He was not totally certain that the earth was a sphere—although willing to admit to the possibility—and rejected on biblical grounds that there could be any inhabitants on the opposite side of the Ocean, or any communication with such places. Moreover, he believed that Greek intellectuals such as Plato, no stranger to geographical theory, had been influenced by biblical thought.⁵² Ideas such as these marked a major shift in how the world was understood, and thus would come to affect geography.

In the sixth century AD, a merchant from Alexandria named Kosmas traveled throughout Aithiopia and perhaps as far east as Taprobane.⁵³ From these journeys he gained the surname Indikopleustes (Indian Sailor). He then seems to have become a monk and thereafter devoted his life to writing, composing a geography (now lost) and a surviving work titled Christian Topography.⁵⁴ His voyages had taken him on both the Mediterranean Sea and the Indian Ocean, and down the coast of Africa. He had also visited Axum (in modern northern Ethiopia),⁵⁵ which had emerged in the first century AD and eventually replaced Meroë as the most prominent city of the region.⁵⁶ His visit to Axum was in the fifth year of Justinian (AD 522), and he was writing the Christian Topography 25 years later, or around AD 547.

Despite its interesting comments on Aithiopia and the far east in late antiquity, the Christian Topography is primarily a work on theology, not geography. It sought to use biblical evidence both to deny that the earth was a sphere and to understand the cosmos. There is an elaborately constructed Christian universe, based on an understanding of the sacred furniture that Moses made in the wilderness.⁵⁷ Since the Ocean could not be navigated, there was little point in seeking what was beyond it. An inventive concept of the path of the sun fit into this new cosmos, including the suggestion that it disappeared behind a mountain in the far north, and that it was no larger than the distance from Alexandria to Rhodes.⁵⁸

The Christian Topography is not to be categorically dismissed, despite a world view that is difficult to comprehend, to say the least. Some of the details of Kosmas' cosmology were from the ancient Greek tradition: the sun and its mountain in the far north are remindful of Anaximenes and Pytheas, and the rejection of sailing the Ocean recalls the poem of Albinovanus Pedo. Other details are new, such as some of the material on Pytheas.⁵⁹ Kosmas considered some topographical matters peculiar to the Judeo-Christian tradition, including the question of where Moses and the Israelites crossed the Red Sea (at Klysma, formerly Kleopatris, near modern Suez):⁶⁰ there may have been more about such issues in his lost geography. But on the whole the work represents a new outlook, a rejection of a thousand years of Greco-Roman geographical thought. The primacy of classical geographical theory was not to be restored until the publication of the major texts in the Renaissance: Strabo and Pliny in 1469 and Ptolemy in 1475.⁶¹ Although these early editions were generally textually unreliable, and Ptolemy and Strabo were first printed in Latin, they were available to early Renaissance explorers and humanists, and restored ancient views of the world, as well as inspiring the future that would be represented by Columbus and Copernicus.