SIX

Urbs Roma to Orbis Romanus: Roman Mapping on the Grand Scale

From the perspectives of geography, cartography and much else, the city of Rome was a stimulating place to be during the last two centuries BCE. This turbulent period saw extraordinary expansion of Roman power far beyond the Italian peninsula in all directions. One result was that a conceptual and cultural awakening occurred, a transformation of Romans’ worldview. It would be wrong to assume that previously they had been a people uninterested in organizing and recording their surroundings. On the contrary, their practice of surveying to divide up the cultivable land of their territory—explained in the previous chapter—was already long established (fig. 6.1). For defeated communities whose territory was appropriated by Rome, one of the most painful and lasting consequences of their loss must have been the manner in which Roman surveyors proceeded to alter radically, and forever, the very appearance of a once-familiar landscape. Moreover, its new Roman aspect was symbolically encapsulated in a map on stone or bronze set up in the heart of the community for all to see—for Romans a proud affirmation of conquest and ownership, to the dispossessed a bitter testament to their permanent loss (figs. 6.2a and 6.2b). In time, Roman surveyors extended their coverage to cities, making very detailed maps that in some cases were engraved on marble and presumably put on display; 1:240 came to be a standard scale for such surveys. Various fragments survive to convey a sense of how impressive these painstaking city maps must have been—recording owners’ names, for instance, and the length of a property’s street frontage, not to mention individual walls, columns, steps, staircases, doorways, and other such features (figs. 6.3a and 6.3b).

Despite the fundamental importance of Roman land surveying, its spatial range came to seem relatively modest from the second century BCE onward, when for the first time the Romans found themselves embroiled in struggles spanning the entire Mediterranean and even well beyond (figs. 6.4a and 6.4b).¹ An enlarged spatial vision was now essential. Its public representation no doubt stemmed in part from the established familiarity with land survey maps, in part also from the practice which victorious generals had adopted of commissioning huge pictures of their battles and sieges that were then displayed, among other means, by being carried in triumphal processions through Rome.² A third and growing source of inspiration and understanding is likely to have been the Greek tradition of geography and cartography, which—as chapter 3 demonstrated—had been greatly advanced by Eratosthenes at Alexandria during the third century. We hear that as early as 174 a former consul (chief magistrate and general) set up a tablet in the Temple of Mater Matuta in Rome commemorating his subjugation of the island of Sardinia. Accompanying it, we are told, was a forma—literally a “shape” or “outline”—of the island, on which battle scenes were painted.³ It is conceivable that a map is meant here, although there can be no certainty, because the Latin language in fact never develops a term to signify “map” unequivocally. So the unspecificforma in this account could equally well signify a three-dimensional image, such as a statue personifying the island. But forma is also often the noun used for a survey map, and some sort of map is perhaps the more likely meaning for it in the context here. It was a large map of the island, we may imagine, although the character of its appearance is beyond recovery, and there is also no knowing whether it was painted directly onto a wall, or drawn on a freestanding panel, say, or on some other surface.

FIGURE 6.1 The pattern of Roman centuriation (probably second century BCE) still preserved in the fields of the Po valley plain near Forum Cornelii (modern Imola, Italy), seen from the air. Reproduced with permission from the British School at Rome.

FIGURE 6.2a Marble panel from an official plan recording the centuriated subdivisions of part of the territory of Arausio (modern Orange, France). This plan was made around 100 CE at a scale of approximately 1:6,000.

FIGURE 6.2b Drawing of as much of the same plan as can be reconstructed. Reproduced from A. Piganiol, Les documents cadastraux de la colonie romaine d’Orange, coll. Suppl. Gallia, XVI (Paris: Editions du CNRS, 1962), pl. XXI.

FIGURE 6.3a City plan fragment found in the Via della Polveriera, Rome. Note the figures recording a length at top right above the letter P.

FIGURE 6.3b City plan fragment found in the Via Anicia, Rome. Figures recording lengths, owners’ names, and representation of walls by polygons are among the many features visible here. Reproduced with permission from Sovraintendenza ai Beni Culturali, Direzione Musei, Roma, Italy.

A book completed in the 30s BCE,⁴ which records a discussion imagined to have occurred twenty years or so earlier, opens with the speakers gathered at the Temple of Tellus, or Earth, in Rome, all looking at picta Italia on a wall there. Again, “Italy painted” could be a personification or an allegorical image. But references to geography in the discussion which unfolds support the likelihood that here, too, a large map of some kind is meant, even though no further clue to the nature of the object is offered.

A distinctly more ambitious work was commissioned by Agrippa, a close associate of Rome’s first “emperor” Augustus, and was only completed after Agrippa’s death in 12 BCE. It is described as orbis terrarum urbi spectandus—“the lands of the globe for the city to look at”—and was to be seen in a portico in Rome that is now lost without a trace.⁵ For certain, there was text associated with this work, and some scholars have even argued that the work was exclusively text, like Augustus’s record of his own achievements, his Res Gestae, a document prominently inscribed in Rome and elsewhere across the empire after his death.⁶ But the choice of spectandus—“to look at”—in describing Agrippa’s work implies something more visual than simply a text to read, and once again a large map seems the most persuasive inference, even if the nature of its appearance has to remain matter for speculation.

FIGURE 6.4a The range of Roman control around 200 BCE.

In particular, there is no knowing just how extensive Agrippa’s map made the orbis terrarum, another term that Latin usage leaves imprecise. It can signify literally the entire globe, often thought to be divided horizontally by climate into zones (according to the Greek theory explained in chap. 3 and fig. 3.5): freezing Arctic and Antarctic at top and bottom; torrid equatorial zone in the center; and two habitable zones sandwiched between unbearable cold and heat, the oikoumene in the north, and the corresponding antichthon, or antipodes, in the south, which equatorial heat renders unreachable from the oikoumene. Naturally, the latter zone is the world of the Greeks and Romans, and as early as the second century BCE some admiring Greeks began hailing Rome as the world power, ruler of the world, or orbis terrarum, in the more restricted sense of “the part of the world known to us.”⁷ At that period it would hardly have been feasible for Romans to convey such a grandiose claim cartographically except in the sketchiest way. But this hazy vision sharpened during the first century as a result of the extensive conquests by Pompey in the east, Caesar in the west, and above all Augustus himself (figs. 6.5a and 6.5b). His generals completed the process of bringing the entire Mediterranean under Roman control, as well as pressing on further to the Danube and Euphrates, and even into Germany, Arabia, and Ethiopia. By no means all these bold forays proved successful. Moreover, the sway that Augustus also claimed to hold through diplomacy over Britain, Persia, and India had minimal substance to it at best.⁸

FIGURE 6.4b The range of Roman control around 70 BCE. Maps made by the Ancient World Mapping Center, University of North Carolina, Chapel Hill, and reproduced with permission.

Even so, for purposes of image making Augustus’s exaggerations were immaterial. By now a more comprehensive vision of Rome’s empire had been achieved, and it was this that Agrippa’s large world map proudly presented not in mere fuzzy outline but rather with close attention to accuracy. Although its scope and its appearance both remain irrecoverable, I think we may confidently assert that Agrippa’s commission introduced a novel and highly influential level of map. There had been no previous cartographic power statement on this kind of scale by any state in the ancient Mediterranean or Near East. This type now came to be disseminated not only by Romans themselves, but also subsequently by their successors and emulators from Charlemagne, through the Middle Ages and Renaissance, up to Mussolini, who commissioned a series of five maps outlining the spread of the ancient and modern Roman empires to 1936. The first four of these maps are still to be seen on display along the Via dell’ Impero (today Via dei Fori Imperiali), which Mussolini began to ram through from the Piazza Venezia in central Rome to the Mediterranean at Ostia; the fifth map was removed from there in 1945.⁹

FIGURE 6.5a The range of Roman control around 50 BCE.

FIGURE 6.5b The range of Roman control around 14 CE. Maps made by the Ancient World Mapping Center, University of North Carolina, Chapel Hill, and reproduced with permission.

The little recorded about ambitious maps produced after Agrippa’s time has almost nothing to say about their appearance. For example, a high-ranking lady’s presentation of a tapestry world map to an unidentified Roman emperor (no later than around the mid-first century CE) is commemorated in a Greek epigram, but the actual nature of the map does not concern the writer. At the same period, I believe that the scholar-emperor Claudius employed a map which somehow encompassed Britain to Italy as a visual aid for a speech he made to the senate in Rome; but this can only be an inference.¹⁰ There does at least survive a description of one world map, which perhaps had regional maps associated with it too. To be sure, this description dates to considerably later—to the period of the tetrarchy, or four coemperors, at the end of the third century—and it occurs in an unashamedly rhetorical speech. Even so, its testimony is unique and precious. The map in question is termed an orbis depictus, “a picture of the world,” a display piece and instructional tool installed in a rhetorical school at Augustodunum in Gaul (modern Autun in France). This school had suffered damage during the turmoil of the previous half century, and the speaker Eumenius, the highly paid new head, seeks a provincial governor’s permission to rebuild it at his own expense. One feature that is evidently already in place in the school, and has even been seen by the governor, is this map. Eumenius devotes the climax of his speech to explaining its value:

In the porticoes let the young men see and examine daily every land and all the seas and whatever cities, peoples, nations, our most invincible rulers either restore by affection or conquer by valor or restrain by fear. Since for the purpose of instructing the youth, to have them learn more clearly with their eyes what they comprehend less readily by their ears, there are pictured in that spot—as I believe you have seen yourself—the sites of all locations with their names, their extent, and the distance between them, the sources and mouths of rivers everywhere, likewise the curves of the coastline’s indentations, and the Ocean, both where its circuit girds the earth and where its pressure breaks into it.

There let the finest accomplishments of the bravest emperors be recalled through different representations of regions, while the twin rivers of Persia and the thirsty fields of Libya and the convex bends of the Rhine and the fragmented mouths of the Nile are seen again as eager messengers constantly arrive. Meanwhile the minds of those who gaze upon each of these places will imagine Egypt, its madness set aside, peacefully subject to your clemency, Diocletian Augustus, or you, unconquered Maximian, hurling lightning upon the smitten hordes of the Moors, or beneath your right hand, Constantius, Batavia and Britannia raising up their grimy heads from woods and waves, or you, Maximian Caesar [Galerius], trampling upon Persian bows and quivers. For now, now at last it is a delight to examine a picture of the world, since we see nothing in it which is not ours.¹¹

In other words, here is a map, or maps, with ambitious scope and a mass of accurate physical and cultural detail inviting appreciation on more than one level. The map can instruct the school’s pupils in geography. It can also inspire them to celebrate the way in which the four coemperors in Diocletian’s new tetrarchy have suppressed risings by dissident populations in some parts of the empire, and expelled foreign occupiers or invaders (Persians especially) from other parts. In consequence, now at last Rome’s longstanding claim to control the entire world can be reasserted with confidence, and Romans will no longer be distressed when they study this map.

The Marble Plan of Rome (Forma Urbis Romae)

Speculation about the appearance of lost large Roman maps can yield few satisfying results. Ideally, the opportunity to examine surviving examples is wanted, and by good fortune two present themselves, each seemingly very different. That said, each does come with distinct flaws and may prove to be something of a special case rather than typical.

The first survival is fragments of a giant plan of the city of Rome measuring nearly 18 m wide by 12 m high and dating to about 200 CE. This plan was carved onto a set of 150 marble slabs clamped to an end wall of a great chamber in the Temple of Peace complex in the heart of Rome. Remarkably, the wall itself with many of the clamp-holes still survives; today it forms the very visible outside back wall of the church of Sts. Cosmas and Damian on the edge of the forum along the Via dei Fori Imperiali (fig. 6.6). The marble slabs have fared less well. A great hole was punched into the wall sometime early in the medieval period, largely destroying beyond recovery the part of the plan where the forum was represented. Even so, about 1,200 fragments have been recovered to date, constituting perhaps 12% of the entire plan; today some fragments are known only as drawings made of pieces recovered but since lost. As is only to be expected, by no means all the fragments fit together; many are tiny and remain adrift. But plenty have been matched up, and this challenging quest has been advanced most recently by an outstanding Web-based project at Stanford University, which offers a 3-D scanned image of each fragment with ample commentary.¹²

At first glance, this marble plan of Rome might seem to be just an example of the surveyors’ city maps mentioned at the beginning of the chapter. It adopts their standard scale of about 1:240 and covers about 13.5 sq km. Centered upon the Capitoline Hill, and oriented approximately southeast (fig. 6.7),¹³ it too shows every ground floor room in the city as well as individual columns, steps, staircases, doorways, and more (figs. 6.8 and 6.9). On closer inspection, however, the marble plan of Rome does not quite match the surveyors’ city maps. In particular, it omits much important data to be found on those maps: owners’ names and distance figures, for instance, are both missing. At the same time the representation of most walls is by single lines rather than by polygons. No doubt the missing data and fuller linework were already available in maps drawn on more pliable material like papyrus (the ancient equivalent of paper); but the designers of this marble plan of Rome evidently took a deliberate decision to omit such detail. Why should they do that, when the effect would be to reduce the usefulness of their plan? The answer must lie with its presentation, on a wall, with even the bottom of the plan 3 to 4 m up from the floor, and the top a further 12 m up. In other words, it was understood from the outset that a plan installed thus could never be “useful” in any practical way. To see any part of it at all, viewers needed to stand several meters away at least, and to see its entirety they had to step far back into the chamber, which fortunately for the purpose was about 24 m long (fig. 6.10).

FIGURE 6.6 The wall on which the marble plan of Rome was originally mounted. Note the holes for the clamps which kept the slabs in place. Photograph by Elizabeth Robinson, reproduced with permission.

There was no viewing point, however, from which the detail could be appreciated; no one could ever gain more than a general impression of it. The designers duly recognized this limitation and so reduced most walls to single lines in order to prevent the plan from gaining a very cluttered appearance; that decision also meant a huge reduction in the amount of engraving to be done. At the same time the designers strove to help viewers orient themselves by coloring many temples red (reckoning that these were likely to be familiar landmarks). Only structures such as major public buildings, aqueducts, and roads were named; and some of the buildings designed for entertainment were represented in bird’s-eye view rather than showing the ground floor layout in the regular way (fig. 6.11, a). Taken together, these various aids to comprehension suggest a conscious effort to render the map accessible and meaningful to the maximum number of viewers, including ones who were barely literate, let alone accustomed to reading maps.

FIGURE 6.7 Reconstruction of the layout of the slabs forming the marble plan. Rome’s forum area would have appeared more or less at the center. Courtesy of the Digital Forma Urbis Romae Project, Stanford University.

In such conditions, therefore, we may fairly ask, why should the designers and engravers of the plan take such pains over the detail that was included, when they already knew that no viewer would ever be able to appreciate it properly? Before attempting an answer, it is worth observing that our modern ability to subject the fragments to close scrutiny enables us to detect that the work was not in fact always as painstaking as it might have been. The proportions of rooms or buildings may appear squashed or stretched, for example (fig. 6.11, b). Such distortion is understandable enough, however, if (as seems most likely) it derives from the delicate challenge of accurately matching up sets of surveyors’ sheets that each covered a very confined area. More puzzling are several instances where archaeology has established that the form in which a building appears on the plan is very outdated. A clear case is the Portico of Octavia, which appears in its original early-first-century CE form, not at all as it would have looked by the end of the second century (fig. 6.11, c). Elsewhere on the plan, by contrast, the effort is made to include the Septizodium, a new monument commissioned by the emperor Septimius Severus, the same ruler who must have commissioned the plan itself.

FIGURE 6.8 Reassembly of marble plan fragments. Reproduced from E. Rodriguez-Almeida, Forma Urbis Marmorea: Aggiornamento generale 1980 (Rome: Quasar, 1981), vol. 2, pl. XXIII, with permission of the publisher.

FIGURE 6.9 Reassembly of marble plan fragments (detail). The V-shaped symbol marks a staircase. Courtesy of the Digital Forma Urbis Romae Project, Stanford University.

I am frankly at a loss for a convincing explanation of such discrepancies. At least it is evident that the makers of the plan felt no special concern to ensure an up-to-date rendering of the city in every detail, and they were shrewd enough to recognize that such perfectionism was uncalled for. To them, the greatest importance of the detail was not so that viewers could examine it minutely, but rather for the overall impression that it created. The sheer capacity to record, organize, engrave, and present it all served as a magnificent tribute to Roman resources, coordination, skill, and control. From the perspective of a viewer standing in the body of the chamber, the countless components of the detail—individually boring and unmemorable—combine to offer an awesome overall vision of the man-made underpinning to this capital of a world empire, by far the largest city in the known world. Such a sense of wonder, variously mixed with pride, or hatred, or fear and other emotions, surely has to be the reaction that the plan was intended to evoke in viewers as they—already themselves present in the heart of Rome—gazed at this gigantic rendering of the entire city looming over them. It may well be that to display Rome or any other city as the marble plan does here constitutes a novel experiment in the tradition of Romans’ large public maps. If so, the creativity of its imaginative, but anonymous, designers merits our admiration all the more.

FIGURE 6.10 The marble plan from its viewers’ perspective (reconstruction). Courtesy of David West Reynolds/Phaeton Group.

The Peutinger Map

CHARACTER, PURPOSE, CONTEXT

The second large Roman map that it is possible for us to examine remains likewise anonymous. Unlike the marble plan, however, it is shorn of a context, and it only survives in the form of a single medieval copy, which is missing the original map’s left-hand end. There is no testimony to where the copy was made or when, let alone the original map. Such depth of ignorance could serve to discourage further enquiry, but there is no cause to be so despairing. The map’s predicament hardly differs from that of countless ancient manuscripts and other material objects which come down to us only as isolated, damaged survivals. It is necessary to seek out clues, hypothesize, and reconstruct; the quest is a challenge with risks, but it can prove instructive especially if a probing range of questions is raised.¹⁴

FIGURE 6.11 Marble plan: a, Bird’s-eye view perspective (Theater of Marcellus). b, Rooms or buildings squashed or squeezed. c, Portico of Octavia. Courtesy of the Digital Forma Urbis Romae Project, Stanford University.

Even incomplete, the map as known to us is a colorful, striking object, especially because of its unusual shape—670 cm long or wide, but only about 13 cm tall. The copy in fact comprises eleven pieces of parchment, each about 60 cm long, that were gummed one to the next to make up the 670-cm-long strip or roll (fig. 6.12 and plate 6). The copying probably occurred around 1200, possibly somewhere in the general area of Germany. At some stage during the three centuries that followed, an institution or individual in possession of the map showed sufficient interest to nail it up for display; several of the nail holes remain visible. Eventually, around 1500, the map was somehow “discovered” at an undisclosed location and removed by the notorious manuscript hunter Konrad Celtis. Soon afterward it came into the possession of the major German collector Konrad Peutinger in Augsburg; hence, it is usually referred to by his name the Peutinger map (latinized as Tabula Peutingeriana). Since 1738 it has been a prized possession of Austria’s national library in Vienna. In 1863 the eleven parchments forming the roll were separated for better conservation.

The map’s design, however, is a single cohesive one which spans all eleven parchments (fig. 6.13). It comprises the world known to the Romans, oriented North, ending with India and Sri Lanka at the right, and no doubt beginning from Britain and the Atlantic coast of mainland Europe and Africa at the lost left-hand end (at the surviving far left, southeast England and southwest France are to be seen). To compress such an extended north-south expanse into a frame no more than 13 cm tall presents a formidable challenge by any standard. In addition, the further, carto-graphically distorting, decision was taken that the map’s central point must be the city of Rome. In other words, there is only the same amount of space for covering the vast arc from Rome to Taprobane (modern Sri Lanka) as there is for covering the notably shorter span from the Atlantic coast of the Iberian peninsula to Rome.

Draconian selectivity is called for, therefore, and it is duly exercised. A single scale for the map (as might be expected today) is unattainable. Rather, the scales—if we can even call them that—vary across it region by region, with Italy (Rome’s heartland) rendered large relative to all others, while Persia and India by contrast both suffer especially severe compression. A related choice is to privilege land over sea by contracting substantial bodies of open water. The Mediterranean Sea in particular becomes little more than a narrow channel, and its islands, peninsulas, and shorelines are manipulated to fit there. Altogether this is shockingly audacious, creative cartography, manifesting even a touch of whimsy to amuse sophisticated viewers intrigued by its more bizarre distortions.

In all likelihood the basis of the Peutinger map is one or more maps that strove for geographic accuracy in the Alexandrian tradition as instituted by Eratosthenes and developed by Marinus, Ptolemy, and others. It is most obviously from such maps too that the designer is able to mark principal settlements, likewise rivers, islands, provinces and regions, peoples, and seas. Unmistakably missing, by contrast, despite the map’s Roman character, are all military installations—legionary camps, garrisons, fortified frontier lines—and indeed all boundary lines, even the outer limits of the Roman empire.

FIGURE 6.12 Peutinger map: furthest left of the eleven surviving parchments. Southeastern England appears top left. Southwestern France appears immediately below it, and then further below (extending across the entire segment), the narrow channel for the Mediterranean Sea. ÖNB/Wien, Cod. 324, segm. 1. Reproduced with permission.

FIGURE 6.13 Top, The spread of the Peutinger map’s eleven surviving parchments (numbered 4–14), with three at the left-hand end restored; shading denotes the extent of Italy. Bottom, The approximate coverage of each surviving parchment is plotted on a modern outline map. Diagram by the author; map courtesy of Christos Nüssli, Euratlas.

A man-made feature that stands out across the entire map is red linework for land routes that fan out from Rome everywhere across the empire and even onward into Persia and India. This is the feature of the map that has repeatedly captured scholars’ attention (fig. 6.14 and plate 7), leading in consequence to relative neglect of other characteristics. The scholarly consensus has always been that the map was designed to serve as a practical tool for reference, an ancient AAA Triptik that travelers could roll up and conveniently carry with them on journeys. This is the view repeated by Oliver Thomson in his introduction to the 1961 revision of Everyman’s Classical Atlas: “As a map, this ribbon is absurd, but its aim is only to give roads with their stations.”¹⁵ The claim has even been advanced that the route network itself forms the basis of the map—making it a forerunner, so to speak, of the famous London Tube diagram developed by Harry Beck in the 1930s—and that all the map’s other physical and cultural components were only added subsequently as mere decorative embellishment.¹⁶

In my view these ideas are long overdue for reappraisal. Anyone with experience of the actual steps required in the design and production of a map will surely find it impossible to credit that the route network can form the basis of the Peutinger map, rather than its shorelines, principal rivers, and principal mountain ranges; these basic elements of the physical landscape would have to be laid out first. This map is not just a decorated diagram, and its designer was plainly much preoccupied with features other than the route network and unrelated to it (islands, for example), conspicuous though it is. Moreover, attempts to use the network as seen here for actual journeys soon expose multiple limitations above and beyond the fundamental distortion created by the extreme shape of the map and Rome’s placement at its center. For example, stopping points that ought to be marked turn out to be omitted; only a roundabout route is liable to be shown, while a direct one for the same journey is ignored; some routes cannot date to later than the first century CE, while others can be no earlier than the second century CE; no indication is given that certain routes drawn on land can in fact only be traversed by ship; and so on.¹⁷ Altogether, therefore, the strong impression emerges that the designer was not seriously concerned to make the route network a practical guide for travelers. Rather, I suggest, he wanted to feature the network for much the same reasons that the designer of the marble plan wanted to include every wall and column and step in Rome—for the cumulative effect of the detail, not for the individual worth of each routine component. The route network was to serve as a distinctive “filler” for the map.

FIGURE 6.14 The city of Rome and routes fanning out from it on the Peutinger map. ÖNB/Wien, Cod. 324, segm. 4+5. Reproduced with permission.

More generally, if the map’s intended function was not to serve as a portable guide for travelers, what purpose did its designer have in mind? Our best clues in my opinion are once again the map’s extreme shape and the further distortion caused by the placement of Rome at its center; these special characteristics need to be pondered, along with the choice of features for inclusion and exclusion. The resulting combination is most convincingly interpreted, I think, as a cartographic celebration of Rome and its empire, indeed, its sway over the entire known world. Moreover, this is a seamlessly unified sway (boundary lines are absent), urbanized and thus civilized, offering ease of movement by land routes in all directions, peaceful, inclusive. To comprehend the map fully and to grasp the significance of Rome’s central placement, viewers must stand well back, just as they needed to in the case of the marble plan.

To be sure, the squat height of the map still needs to be accounted for. Such squatness would seem unnecessarily restrictive if the map was a self-contained object, but not perhaps if it originally constituted only part of a larger artwork that is otherwise now lost. Various possibilities can be imagined. I admit to being most attracted by one in which the map is designed to represent the northern habitable zone—the oikoumene—of a globe image divided horizontally by zones according to the Greek scheme outlined above and in chapter 3. In such an instance, depending on the height imagined for each zone (including possible double height for the equatorial zone), the full image could easily attain 2 m.

Who would commission such a tall, wide image, and where would it be placed? The commission surely has to have come from an emperor or the close associate of one; the map in particular is too strong an affirmation of power for anyone outside that circle to risk ordering its production independently. In addition, its sheer size makes it a piece for display rather than an item suitable for private space. Apart from a few Christian notations manifestly added later by well-meaning copyists, there is no hint of original Christian content or a Christian outlook. The map is likely to date to the period before Constantine, therefore, to Diocletian’s tetrarchy at the latest, around 300 CE. It could not have been produced before the early second century CE, however, because it includes routes in Dacia, which only became Roman territory in that period.

While any date from the early second century to the late third seems possible in principle for the map’s production, there can be no question that such an initiative would fit the tetrarchs’ agenda and image uniquely well. As Eumenius recognized in his praise of the world map at Augustodunum quoted above, the tetrarchs were eager to celebrate how they had restored and reunited the empire, how it was now peaceful and stable again at last, and how its four joint rulers were in concord with one another. They took pride, too, in their tireless travels undertaken for the welfare of the regions assigned to them; in consequence, land routes were of the greatest importance to them. Their outlook is solemnly articulated in the climax of the immense sentence that opens their Edict on Maximum Prices, issued in 301 and known from inscriptions:

Public decency and Roman dignity and majesty desire that the fortune of our state be organized in good faith and elegantly adorned, and that it be thanked—beside the immortal gods—as we recall the wars that we have fought successfully, at a time when the world is in tranquility, placed in the lap of a most profound calm, as well as benefiting from a peace which was toiled for with abundant sweat. Let us therefore, we who with the kind favor of the deities crushed the previous seething ravages of barbarian peoples by destroying those very nations, protect the peace established for eternity with the appropriate defences of justice.¹⁸

The tetrarchs deliberately adopted a loftier, more authoritarian style of rule than that of previous emperors. They sat on thrones and received homage amid elaborate, colorful ceremonial. Where better, then, to place a globe image of the type I envisage, oriented north, than in a tetrarch’saula or throne room, a space developed from the traditional basilica, but now with an apse at one end of the hall within which to set a throne? Moreover, if such a globe image occupied the rounded end of the apse, behind the throne, where would the city of Rome appear but at the center of the oikoumene, directly above the head of the tetrarch sitting upon his throne (fig. 6.15)? Viewers’ attention during ceremonies could in any case hardly avoid focusing on the oikoumene part of the image, because the standing figures of the tetrarch’s close associates flanking the throne would hide much of the lower half; in addition, everywhere outside the oikoumene the environment would appear bleak and insecure, lacking cities and routes in particular.

If there were protruding spur walls on either side of the apse (as seems to have been common),¹⁹ then most viewers in the body of the aula gazing forward at the globe image would have been unable see the far ends of the oikoumene on the map. As a result, the traditional notion that Roman sway was “empire without end” (imperium sine fine) would only be reinforced: for these viewers, the map, and the empire, just continued on out of sight to west and east. None of the empire’s provinces would lie beyond their field of vision. By contrast, India to the far right might be more or less hidden, but it was never under Roman rule, of course, and since to most Romans it merely symbolized the distant and exotic, the map’s token representation of it would be quite sufficient.

Rhetoric, it need hardly be added, claimed control of the world for the tetrarchs, and even of the cosmos. On the arch of the tetrarch Galerius at Thessalonica, two of the emperors sit side by side flanked by the other two standing (fig. 6.16). A pair of busts below the seated pair have been plausibly identified as Sky and Earth, while at either side of the same scene appear reclining personifications of Earth and Sea.²⁰ In their Edict on Maximum Prices, already quoted above, the tetrarchs proclaim that it applies “not [just] to individual cities and peoples and provinces, but to the entire world [universus orbis].”²¹ Most remarkably, in an unprecedented find dating to 2005, a set of imperial emblems was recovered during an official excavation on the northeast slope of Rome’s Palatine Hill (not far from the Colosseum).²² These signa imperii emerged from a sturdy poplar-wood box that had been buried with care. In it were wrapped—in silk and linen—three bronze pikes, a scepter, five javelins, and four spheres or globes, three of glass and one of blue chalcedony (quartz). These latter served as orbs that a Roman ruler mounted on a scepter or held in his hand to symbolize world domination (fig. 6.17 and plate 8). The ruler in this instance was most probably the usurper and would-be tetrarch Maxentius, who held Rome but had to defend it against an attack by his rival Constantine during the fall of 312. In the fateful battle at the Milvian Bridge Maxentius was killed, and these insignia, which had been buried for safekeeping, were never retrieved.

FIGURE 6.15 The Peutinger map imagined as the northern habitable zone within a globe image decorating an apse behind a Roman ruler’s throne. Sketch by Daniel Talbert, reproduced with permission.

The excellent preservation of the basement level in the palace built at Split (in modern Croatia) by Diocletian for his retirement aptly illustrates the possible placement of a globe image in the apse of the hall there and the range of vision from the body of the hall (fig. 6.18).²³ Two cautionary notes should be struck in this connection, however. First, it is impossible to establish whether or not the architecture of the aula itself on the lost main level above was identical to what survives at basement level, although in all likelihood it was. Second, any temptation to indulge in the further speculation that the Peutinger map was designed for this specific setting should be avoided. There is no justification for it, and no need. Conceivably, the image as I envisage it may have been mounted on panels, thus allowing all of it or parts (the map especially) to be transported and reassembled in whatever settings presented themselves elsewhere. As it happens, King Henry VIII of England, another ruler who was constantly on the move, is known to have had a world map, a mappa mundi, that went with him on his royal progresses and formed part of “the Removing Guarderobe . . . attendaunt at the Courte uppon the kinges most Roiall personne where the same for the tyme shall happen to be,” as an inventory made after his death in 1547 records.²⁴

FIGURE 6.16 Arch built as the entrance to the palace of the tetrarch Galerius at his capital, Thessalonica, Greece (detail from south pier, north face). Digital image by J. Matthew Harrington, reproduced from Wikipedia Commons.

FIGURE 6.17 Imperial orbs excavated in Rome and associated with the self-proclaimed tetrarch Maxentius. Photograph by Clementina Panella, reproduced with permission.

FIGURE 6.18 Basement-level hall in Diocletian’s palace, Split, Croatia, looking toward the apse. Photograph by Greg Aldrete, reproduced with permission.

ORIGINALITY AND IMPACT

Even if some or all these speculations fail to carry conviction, I nonetheless urge support for my concern to reconstruct a meaningful context and function for the Peutinger map. To date, the scholarship devoted to it has largely failed to appreciate its shape, orientation, layout, and choice of data as a set of deliberate, linked choices on the part of the designer. These fundamentals are not to be taken for granted as standard, or even simply ignored; rather, they are all distinctive, and as such offer the best means by which to recover the thinking that underlies the design.

The degree of originality in the Peutinger map’s design is an important issue, but a hard one to resolve with confidence when our knowledge of large Greek and Roman maps remains so defective. It is at least conceivable, however, that no previous mapmaker had dared to take a frame of such extreme dimensions, and then to set the entire orbis terrarum within it, with a central placement for the city of Rome. All of these initiatives require the landscape to be remolded on an epic scale, a transformation that the Peutinger map’s designer deftly accomplished with the intention above all of promoting and celebrating Roman power. As Alexander Jones observes in chapter 4 above, the style of cartography pursued by Ptolemy in the scientific tradition of Hellenistic Alexandria was studiously “unpolitical” by contrast; it barely acknowledged Roman rule, much less sought to celebrate it. To Ptolemy, moreover, it was vital that the world map he prescribed should look like part of a globe; hence arose his intense preoccupation with projection. To be sure, such cartography still offered scope for parading Roman achievements, and the map in the school at Augustodunum evidently used it thus, as no doubt had Agrippa’s map in Rome much earlier. The radical shifts introduced by the designer of the Peutinger map, however, belong to a wholly different league and could well be unprecedented. He may indeed have realized from the example of the marble plan of Rome how effectively it was possible to adapt detailed, official map materials for a quite different fresh purpose if they were fitted together, appropriately simplified and arrestingly presented; even so, the Peutinger map is altogether far more ambitious in the ways that it reflects such thinking.

It may also be unusual, if not unprecedented, for the city of Rome to occupy the center of the map. As Benet Salway explains in the following chapter, once the Romans’ empire dominated the Mediterranean and beyond, this sea evidently became the central feature in their mind’s eye, and the various surviving descriptions of provinces, say, or listings of legionary bases, typically proceed anticlockwise starting from some point at the fringe, not from Rome itself.

An associated feature of the map that stands out for being not merely unusual, but also in all likelihood unprecedented, is its comprehensive featuring of land routes. Testimony confirming their appearance on other Greek or Roman maps is all but nonexistent, and we can readily imagine that cartographers who strove to produce accurate maps of large regions would hesitate to introduce such complex linework. Equally, it would seem that the emperors of the first two and a half centuries CE were unconcerned to make widespread claims about their control of land routes over a vast area.²⁵ In practice, the maintenance of routes was divided among local communities everywhere, and they may hardly have been visualized as an integrated empire-wide “network” even by emperors. It could have been the new ideology of the tetrarchy, however, that inspired the designer of the Peutinger map to think in these terms. The tetrarchs were committed to coordinating tighter control at the local level across the empire, and mobility was vital to them and their troops. Although the routes as represented by the map are barely adequate for making or planning actual journeys, and any specifically military reference is absent, nonetheless, the range and value of the network are brilliantly conveyed. Forging horizontally and purposefully across the landscape, the routes infuse the map with a cohesion and dynamism not to be found in anything compiled from Ptolemy’s coordinates. For all their dedication to accuracy, such products of the Hellenistic tradition remain static in appearance.

If fitting together the route network for incorporation in the map was indeed an original initiative, it may well have presented the designer with a more formidable challenge than he had anticipated, because for the most part the extensive data required were probably not already assembled and accessible in a convenient form. Instead, in all likelihood, they had to be first painstakingly gathered from a mass of individual, overlapping itineraries, next organized, and then plotted from scratch.²⁶ The result of this labor was dramatic in its effect, however: perhaps for the first time, physical and man-made features recorded by cartographers in the Alexandrian tradition were now integrated on a map with vast numbers of towns and way stations that had never been marked on any previous map except maybe a local one. Because these places lay along routes shown on the map, the lack of coordinates for them was no obstacle to their inclusion, and in any case the map’s distorted shape confirmed a lack of concern to position them with geographic accuracy.

Despite the thinness of the surviving record, we can identify a twofold impact made more or less directly by the cartography of the Peutinger map through the next millennium. First, it is clear that further maps were produced incorporating the type of data from itineraries that appear to us as an innovatory component of the Peutinger map. The testimony is both textual and cartographic. Earliest is that of the Cosmographia, a text compiled in Latin by an unnamed cleric claiming to be from Ravenna, probably around 700.²⁷ His purpose of seeking to preserve a vision of the world once known to the Romans prompts him to list, region by region, the names of settlements, rivers, islands, and peoples, around five thousand in all. Among his various sources were clearly either maps like the Peutinger map or lists of names along roads derived from such maps. Regardless of whether he took over names directly from maps himself, or whether this was done by previous compilers that he then copied from, it is obvious that the work was done in the most slapdash manner. Even so, two related observations seem beyond dispute. On the one hand, the frequency with which the names offered for an area by the cosmographer match those on the Peutinger map is striking. On the other hand, there are unmistakable instances where the listing of names is entirely comparable in character but cannot in fact derive from the Peutinger map. It marks only a handful of settlements on Sardinia and Corsica, for example, whereas the cosmographer lists a total of thirty-three there; elsewhere, there are entire regions which he covers (such as Arabia and Ethiopia), but the map does not. Altogether, there can be no question that his work is only made possible by maps incorporating the level of detail that the Peutinger map offers, as well as by itineraries and other lists that both underlay such maps and were compiled from them. Whether the Peutinger map itself was among the cosmographer’s sources at either first or second hand is impossible to establish, and the issue hardly matters. Far more significant is the realization that well before about 700 CE various further maps of the same type had been produced, even though they may have differed in scope and shape.

A striking later product in the same tradition, dating to around 1050, is the large map produced at the abbey of Saint-Sever in Gascony (southwestern France) to illustrate a manuscript copy of the Commentary on the Apocalypse originally compiled by a Spanish monk, Beatus, in 776. The degree to which the 270 place-names on this map correspond to ones on the Peutinger map is unmistakable; evidently the designer at Saint-Sever is selecting from the kinds of maps and records of names that it represents. Further continuation of the tradition is apparent in the lost map known only from a detailed text about it, the Descriptio Mappe Mundi written in Paris by Hugh of Saint-Victor around 1135. The tradition is also reflected in the surviving Hereford Cathedral mappamundi produced around 1300.²⁸

Second, the related impact of the Peutinger map was to demonstrate—as no map had done before so forcefully, perhaps—that a cartographer could not merely reshape the world drastically, but also do so in pursuit of aims that looked well beyond cartography. In short, this was a map that encouraged and empowered late antique and medieval mapmakers to devise further creative ways of giving their own work its intended impact. Christians in particular seized the opportunity. To suggest that Christian mappaemundi have their origin in the Peutinger map would be too sweeping and simplistic, but the map’s radical cartography can be reckoned to make a formative contribution that has not been appreciated to date. As is well known, mappaemundi employ both texts and images to convey Christian history and belief within a geographic framework that need be no more than schematic and comes to be developed in a rich variety of forms.²⁹

Testimony to the attention that the Peutinger map itself continued to receive is only now coming to be noticed too. The fact that the one copy known to us was nailed up for display sometime after it had been made around 1200, as mentioned above, is a novel observation made by Martin Steinmann. In addition, it has recently come to light that a very similar map (but not identical, and now lost) was displayed along a wall in an anteroom in the bishop’s residence at Padua around 1495. In my estimation there is also reason to believe that the maker of a map of Britain in the late thirteen century (known to us only through a mid-fourteenth-century copy, the so-called Gough map) had seen either the Peutinger map or one very similar, and that several of the remarkable features in his own design may reflect its influence.³⁰ Altogether, the Peutinger map can be considered to mark a pivotal transition from classical cartography to medieval—a creative advance in the former, a stimulus for the latter to develop its own distinctiveness. Its recognition as Roman work of lasting cartographic and cultural inspiration is long overdue.